AMD Ryzen 9 9950X benchmarks for a future article.
amd-pstate-epp powersave, EPP balance_performance Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
amd-pstate-epp performance, EPP performance Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
amd-pstate-epp powersave, EPP performance Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: performance) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
amd-pstate-epp powersave, EPP power Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: power) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
amd-pstate-epp powersave, EPP balance_power Processor: AMD Ryzen 9 9950X 16-Core @ 5.75GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG STRIX X670E-E GAMING WIFI (2204 BIOS), Chipset: AMD Device 14d8, Memory: 2 x 32GB DDR5-6400MT/s Corsair CMK64GX5M2B6400C32, Disk: 2000GB Corsair MP700 PRO, Graphics: AMD Radeon RX 7900 GRE 16GB, Audio: AMD Navi 31 HDMI/DP, Monitor: DELL U2723QE, Network: Intel I225-V + Intel Wi-Fi 6E
OS: Ubuntu 24.04, Kernel: 6.10.0-phx (x86_64), Desktop: GNOME Shell 46.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 24.2~git2406040600.8112d4~oibaf~n (git-8112d44 2024-06-04 noble-oibaf-ppa) (LLVM 17.0.6 DRM 3.57), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_power) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
acpi-cpufreq schedutil Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
acpi-cpufreq performance Changed Processor to AMD Ryzen 9 9950X 16-Core @ 4.30GHz (16 Cores / 32 Threads) .
Processor Change: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb40401a
AMD Ryzen 9 9950X P-State Driver Linux OpenBenchmarking.org Phoronix Test Suite AMD Ryzen 9 9950X 16-Core @ 5.75GHz (16 Cores / 32 Threads) AMD Ryzen 9 9950X 16-Core @ 4.30GHz (16 Cores / 32 Threads) ASUS ROG STRIX X670E-E GAMING WIFI (2204 BIOS) AMD Device 14d8 2 x 32GB DDR5-6400MT/s Corsair CMK64GX5M2B6400C32 2000GB Corsair MP700 PRO AMD Radeon RX 7900 GRE 16GB AMD Navi 31 HDMI/DP DELL U2723QE Intel I225-V + Intel Wi-Fi 6E Ubuntu 24.04 6.10.0-phx (x86_64) GNOME Shell 46.0 X Server + Wayland 4.6 Mesa 24.2~git2406040600.8112d4~oibaf~n (git-8112d44 2024-06-04 noble-oibaf-ppa) (LLVM 17.0.6 DRM 3.57) GCC 13.2.0 ext4 3840x2160 Processors Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Compiler File-System Screen Resolution AMD Ryzen 9 9950X P-State Driver Linux Benchmarks System Logs - Transparent Huge Pages: madvise - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - amd-pstate-epp powersave, EPP balance_performance: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xb40401a - amd-pstate-epp performance, EPP performance: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xb40401a - amd-pstate-epp powersave, EPP performance: Scaling Governor: amd-pstate-epp powersave (EPP: performance) - CPU Microcode: 0xb40401a - amd-pstate-epp powersave, EPP power: Scaling Governor: amd-pstate-epp powersave (EPP: power) - CPU Microcode: 0xb40401a - amd-pstate-epp powersave, EPP balance_power: Scaling Governor: amd-pstate-epp powersave (EPP: balance_power) - CPU Microcode: 0xb40401a - acpi-cpufreq schedutil: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xb40401a - acpi-cpufreq performance: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb40401a - BAR1 / Visible vRAM Size: 16368 MB - Python 3.12.3 - gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance Result Overview Phoronix Test Suite 100% 104% 108% 113% srsRAN Project Stargate Digital Audio Workstation FLAC Audio Encoding Ogg Audio Encoding WavPack Audio Encoding Selenium Monkey Audio Encoding LAME MP3 Encoding Opus Codec Encoding Radiance Benchmark simdjson VOSK Speech Recognition Toolkit VP9 libvpx Encoding x265 LeelaChessZero SVT-AV1 LuxCoreRender WebP Image Encode WebP2 Image Encode BRL-CAD Timed LLVM Compilation uvg266 Blender Whisperfile RawTherapee VVenC Timed Linux Kernel Compilation GravityMark Xonotic IndigoBench Chaos Group V-RAY GPUScore: Breaking Limit OSPRay Embree ONNX Runtime
amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance Per Watt Result Overview Phoronix Test Suite 100% 112% 124% 136% 148% GravityMark GPUScore: Breaking Limit srsRAN Project Selenium Stargate Digital Audio Workstation simdjson Xonotic WebP Image Encode WebP2 Image Encode LuxCoreRender x265 LeelaChessZero IndigoBench BRL-CAD SVT-AV1 uvg266 VP9 libvpx Encoding Meta Performance Per Watts Embree Chaos Group V-RAY OSPRay VVenC P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M
AMD Ryzen 9 9950X P-State Driver Linux selenium: SilverBench - Firefox svt-av1: Preset 3 - Beauty 4K 10-bit webp2: Quality 100, Lossless Compression lczero: Eigen lczero: BLAS whisperfile: Medium build-llvm: Unix Makefiles build-llvm: Ninja svt-av1: Preset 3 - Bosphorus 4K brl-cad: VGR Performance Metric gravitymark: 3840 x 2160 - Vulkan ospray: particle_volume/pathtracer/real_time simdjson: PartialTweets svt-av1: Preset 5 - Beauty 4K 10-bit blender: Pabellon Barcelona - CPU-Only ospray: particle_volume/scivis/real_time webp2: Quality 95, Compression Effort 7 whisperfile: Small svt-av1: Preset 8 - Beauty 4K 10-bit simdjson: DistinctUserID ospray: particle_volume/ao/real_time radiance: Serial simdjson: Kostya vvenc: Bosphorus 4K - Fast onnx: yolov4 - CPU - Parallel onnx: yolov4 - CPU - Parallel stargate: 44100 - 1024 selenium: Jetstream 2 - Firefox onnx: GPT-2 - CPU - Standard onnx: GPT-2 - CPU - Standard simdjson: LargeRand stargate: 192000 - 512 onnx: T5 Encoder - CPU - Standard onnx: T5 Encoder - CPU - Standard stargate: 192000 - 1024 simdjson: TopTweet v-ray: CPU stargate: 96000 - 512 webp2: Quality 75, Compression Effort 7 blender: Junkshop - CPU-Only onnx: ResNet50 v1-12-int8 - CPU - Parallel onnx: ResNet50 v1-12-int8 - CPU - Parallel svt-av1: Preset 13 - Beauty 4K 10-bit luxcorerender: Danish Mood - CPU indigobench: CPU - Bedroom indigobench: CPU - Supercar onnx: ResNet101_DUC_HDC-12 - CPU - Parallel onnx: ResNet101_DUC_HDC-12 - CPU - Parallel onnx: ResNet101_DUC_HDC-12 - CPU - Standard onnx: ResNet101_DUC_HDC-12 - CPU - Standard onnx: fcn-resnet101-11 - CPU - Parallel onnx: fcn-resnet101-11 - CPU - Parallel onnx: GPT-2 - CPU - Parallel onnx: GPT-2 - CPU - Parallel onnx: bertsquad-12 - CPU - Parallel onnx: bertsquad-12 - CPU - Parallel onnx: fcn-resnet101-11 - CPU - Standard onnx: fcn-resnet101-11 - CPU - Standard onnx: ZFNet-512 - CPU - Parallel onnx: ZFNet-512 - CPU - Parallel onnx: ZFNet-512 - CPU - Standard onnx: ZFNet-512 - CPU - Standard onnx: bertsquad-12 - CPU - Standard onnx: bertsquad-12 - CPU - Standard onnx: ArcFace ResNet-100 - CPU - Parallel onnx: ArcFace ResNet-100 - CPU - Parallel onnx: T5 Encoder - CPU - Parallel onnx: T5 Encoder - CPU - Parallel onnx: yolov4 - CPU - Standard onnx: yolov4 - CPU - Standard onnx: Faster R-CNN R-50-FPN-int8 - CPU - Standard onnx: Faster R-CNN R-50-FPN-int8 - CPU - Standard onnx: ArcFace ResNet-100 - CPU - Standard onnx: ArcFace ResNet-100 - CPU - Standard onnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallel onnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallel onnx: CaffeNet 12-int8 - CPU - Parallel onnx: CaffeNet 12-int8 - CPU - Parallel onnx: CaffeNet 12-int8 - CPU - Standard onnx: CaffeNet 12-int8 - CPU - Standard onnx: ResNet50 v1-12-int8 - CPU - Standard onnx: ResNet50 v1-12-int8 - CPU - Standard onnx: super-resolution-10 - CPU - Parallel onnx: super-resolution-10 - CPU - Parallel onnx: super-resolution-10 - CPU - Standard onnx: super-resolution-10 - CPU - Standard svt-av1: Preset 5 - Bosphorus 4K xonotic: 3840 x 2160 - Ultimate stargate: 96000 - 1024 vpxenc: Speed 0 - Bosphorus 4K build-linux-kernel: defconfig stargate: 44100 - 512 blender: BMW27 - CPU-Only selenium: Speedometer 3.0 - Firefox xonotic: 3840 x 2160 - Ultra uvg266: Bosphorus 4K - Slow stargate: 480000 - 1024 stargate: 480000 - 512 rawtherapee: Total Benchmark Time radiance: SMP Parallel whisperfile: Tiny encode-opus: WAV To Opus Encode vosk: vvenc: Bosphorus 4K - Faster breaking-limit: 3840 x 2160 - Off webp: Quality 100, Lossless, Highest Compression vpxenc: Speed 0 - Bosphorus 1080p svt-av1: Preset 8 - Bosphorus 4K encode-ogg: WAV To Ogg vpxenc: Speed 5 - Bosphorus 4K embree: Pathtracer ISPC - Crown x265: Bosphorus 4K embree: Pathtracer ISPC - Asian Dragon encode-flac: WAV To FLAC uvg266: Bosphorus 4K - Ultra Fast selenium: Kraken - Firefox srsran: PUSCH Processor Benchmark, Throughput Total webp: Quality 100, Lossless svt-av1: Preset 13 - Bosphorus 4K vpxenc: Speed 5 - Bosphorus 1080p selenium: PSPDFKit WASM - Firefox encode-wavpack: WAV To WavPack encode-ape: WAV To APE luxcorerender: Rainbow Colors and Prism - CPU srsran: PDSCH Processor Benchmark, Throughput Total webp: Quality 100, Highest Compression encode-mp3: WAV To MP3 webp2: Quality 100, Compression Effort 5 webp2: Default webp: Quality 100 webp: Default amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 1950 1.787 0.04 236 248 329.42136 319.918 296.684 11.901 491318 186.8 276.342 13.00 7.660 146.55 11.9791 0.17 126.32442 10.753 13.18 11.9650 302.743 7.58 9.526 125.612 7.96401 7.882371 298.309 6.25566 159.880 2.09 3.824622 5.20789 192.090 4.144614 13.56 55074 5.678037 0.37 62.73 6.48991 154.069 19.499 4.65 5.953 13.609 695.631 1.43761 379.346 2.63611 438.488 2.28069 6.54688 152.681 86.4237 11.5735 196.738 5.08286 13.9483 71.6902 7.80322 128.122 39.6454 25.2224 39.3072 25.4398 5.18192 192.957 70.4660 14.1908 16.1255 62.0076 17.3323 57.6908 24.1224 41.4521 3.96846 251.961 1.16412 858.498 1.71427 583.038 4.52501 220.976 4.44876 224.772 44.190 721.3029224 6.147759 12.24 47.512 8.331987 46.71 31.8 946.9529596 13.86 8.265068 8.070138 34.304 92.937 29.10015 17.849 11.622 22.490 86.30 0.92 23.32 99.960 11.291 26.91 36.1532 38.90 41.0622 11.398 44.39 377.5 2080.6 2.33 260.536 48.27 531 4.986 9.713 19.96 16092.7 5.19 4.417 10.18 15.04 16.17 26.11 1966 1.786 0.04 235 252 329.45273 321.525 299.049 11.938 496949 186.7 276.129 12.68 7.666 146.65 11.9466 0.17 125.98586 10.723 13.08 11.9255 302.142 7.65 9.597 126.048 7.93391 8.477485 302.837 6.25429 159.910 2.07 3.950129 5.43900 183.836 4.305339 12.97 55068 5.857247 0.37 63.16 6.36926 156.991 19.579 4.71 5.915 13.590 698.102 1.43253 375.821 2.66083 439.822 2.27372 6.61050 151.194 84.2079 11.8785 193.945 5.15618 13.8677 72.1044 7.82760 127.726 39.8712 25.0790 39.6910 25.1957 5.18066 193.004 70.4532 14.1934 16.0416 62.3318 17.2122 58.0927 24.0891 41.509 4.02330 248.493 1.16437 858.273 1.70042 587.830 4.52082 221.204 4.43288 225.578 44.484 725.0659720 6.350984 12.54 47.659 8.221870 47.03 32.8 943.9539855 13.92 8.201474 8.183842 33.962 94.067 28.92234 17.364 11.348 22.463 86.14 0.93 23.89 100.268 10.919 27.27 36.1655 39.23 40.9643 10.715 44.71 377.6 2080.7 2.38 260.012 49.75 523 4.911 9.425 19.95 16293.8 5.16 4.354 10.24 14.93 16.42 26.56 2007 1.776 0.04 236 253 329.05262 322.272 298.151 11.913 498314 186.3 276.183 12.73 7.634 146.89 11.9688 0.17 126.21258 10.719 13.66 11.9379 297.905 7.70 9.577 126.028 7.93590 8.383640 302.864 6.11717 163.439 2.12 3.803383 5.17267 193.304 4.192825 13.46 55091 5.719418 0.37 62.43 6.46604 154.642 19.539 4.66 5.904 13.620 701.745 1.42504 376.411 2.65667 437.183 2.28817 6.54213 152.782 84.0855 11.8925 194.483 5.14187 13.9818 71.5143 7.80868 128.040 39.7015 25.1869 39.5280 25.2993 5.16594 193.556 70.2877 14.2269 15.9581 62.6570 17.2117 58.0946 24.2043 41.3111 4.03341 247.948 1.16550 857.478 1.70305 586.928 4.55799 219.378 4.46730 223.866 44.287 722.3642903 6.279553 12.53 47.643 8.110456 46.84 32.2 937.2084225 13.91 8.212730 7.865709 34.043 94.629 29.08032 17.250 11.281 22.417 85.97 0.94 23.8 99.978 11.030 26.95 36.2154 39.20 40.9453 10.754 44.47 380.5 2080.7 2.31 259.808 49.54 530 4.906 9.362 20.08 16229.4 5.23 4.382 10.21 15.11 16.38 26.52 1986 1.785 0.04 236 246 330.76639 325.255 298.931 11.852 494060 185.9 276.408 13.13 7.691 146.99 11.9701 0.17 126.34167 10.798 13.59 11.9327 300.02 7.75 9.574 126.557 7.90547 8.593859 299.474 6.30332 158.644 2.13 3.852325 5.19651 192.502 4.157566 13.66 54799 5.836306 0.36 62.53 6.49163 154.022 19.536 4.63 5.924 13.636 701.161 1.42626 379.116 2.63771 433.561 2.30726 6.58721 151.743 84.3201 11.8603 196.170 5.09761 14.0156 71.3399 7.82813 127.716 39.8729 25.0789 39.4717 25.3340 5.17978 193.035 70.7614 14.1315 16.0433 62.3259 17.4346 57.3542 24.1395 41.4215 4.04379 247.248 1.15589 864.749 1.70746 585.413 4.52214 221.111 4.44079 225.177 44.100 720.2286106 6.274104 12.49 47.905 8.427568 48.07 30.4 941.3818119 13.82 8.444777 8.389622 34.182 91.089 29.25195 17.331 11.650 22.421 86.10 0.91 23.40 99.654 10.697 26.89 36.2457 38.76 41.0112 10.816 44.15 383.5 2080.6 2.37 259.560 48.75 533 5.072 9.504 19.76 12186.1 5.18 4.383 10.09 14.61 15.94 25.85 1953 1.788 0.04 238 249 329.78937 323.642 298.165 11.887 497238 186.9 276.283 12.97 7.627 146.83 11.9717 0.17 125.84052 10.674 13.50 11.9585 296.542 7.68 9.542 129.749 7.70720 8.338883 296.738 6.25270 159.948 2.12 3.828578 5.14062 194.507 4.263705 13.64 55057 5.893716 0.36 62.61 6.47352 154.457 19.523 4.66 5.925 13.626 693.553 1.44193 377.539 2.64873 437.511 2.28589 6.55080 152.606 84.1757 11.8833 193.674 5.16325 13.7439 72.7586 7.81452 127.941 39.8712 25.0799 39.5765 25.2664 5.15748 193.868 70.4524 14.1936 15.9829 62.5610 17.3818 57.5273 24.0838 41.5181 4.01785 248.810 1.16319 859.161 1.70884 584.905 4.56251 219.155 4.44344 225.038 44.266 725.1644602 6.329842 12.34 47.837 8.267507 46.77 31.0 944.5349909 13.87 8.564042 7.995940 34.186 94.174 29.02365 17.693 11.547 22.428 86.27 0.92 23.62 100.321 10.956 27.32 36.2764 38.95 41.0992 10.752 44.28 383.1 2080.6 2.35 260.873 49.80 533 5.035 9.732 19.80 14495.3 5.13 4.441 10.15 14.84 16.59 25.99 1912 1.761 0.04 236 252 329.66344 323.633 298.052 11.663 497167 187.2 276.329 12.95 7.585 147.28 12.0029 0.17 126.56108 10.624 13.74 11.9869 302.723 7.75 9.439 126.897 7.88058 9.250160 290.426 6.25902 159.803 2.11 3.932036 5.16644 193.543 4.414173 13.55 55008 6.120461 0.37 62.30 6.49949 153.895 18.760 4.72 5.933 13.668 689.985 1.44931 379.010 2.63844 430.466 2.32311 6.54343 152.748 83.6834 11.9507 191.119 5.23228 13.8674 72.1027 7.8551 127.276 39.7733 25.1409 39.9692 25.0192 5.11863 195.339 70.3174 14.2208 16.1725 61.8276 17.3024 57.7899 24.5567 40.7196 4.01014 249.287 1.16025 861.426 1.70234 587.199 4.55439 219.555 4.44308 225.060 43.176 722.8591335 6.605524 12.34 47.745 8.730047 46.49 31.0 938.1349185 13.87 9.028488 8.491271 33.975 95.54 29.24412 17.393 11.577 22.383 86.38 0.92 23.75 97.816 10.827 27.07 36.2085 38.43 41.0793 11.391 44.50 385.2 2080.1 2.32 258.699 48.99 565 4.892 9.589 20.07 16633.9 5.24 4.350 10.17 15.05 16.39 26.04 1944 1.764 0.04 232 247 330.08294 322.378 298.661 11.759 491302 187.1 276.569 12.75 7.493 146.37 11.9840 0.17 127.55141 10.650 13.26 11.9674 298.838 7.55 9.507 126.688 7.89388 9.332352 295.397 6.20910 161.079 2.10 4.008207 5.15508 193.968 4.396724 13.13 55079 6.132388 0.36 62.37 6.49100 154.067 19.365 4.63 5.903 13.567 691.968 1.44518 377.030 2.65230 432.642 2.31140 6.60006 151.430 86.1573 11.6084 194.717 5.13563 13.9579 71.6359 7.82105 127.832 39.9725 25.0170 39.4222 25.3672 5.17443 193.232 70.0150 14.2826 16.1517 61.9056 17.2224 58.0586 24.5495 40.7314 4.05533 246.557 1.16379 858.735 1.71720 582.129 4.50600 221.897 4.41932 226.272 43.59 717.9473465 6.634571 12.40 47.903 8.918763 46.80 31.9 941.9172222 13.80 9.028189 8.649094 33.943 97.616 29.42886 17.545 11.516 22.399 86.29 0.93 24.25 98.386 10.819 27.22 36.1586 38.78 40.9800 10.760 44.04 383.0 2080.2 2.31 259.399 48.37 536 4.845 9.481 20.05 16591.7 5.14 4.291 10.17 15.00 16.30 26.48 OpenBenchmarking.org
Selenium Result
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: SilverBench - Browser: Firefox acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance 400 800 1200 1600 2000 SE +/- 6.94, N = 3 SE +/- 4.51, N = 3 SE +/- 18.22, N = 3 SE +/- 15.45, N = 3 SE +/- 9.94, N = 3 SE +/- 17.02, N = 8 SE +/- 19.06, N = 3 1912 1944 1950 1953 1966 1986 2007 1. firefox 126.0.1 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Selenium Benchmark: SilverBench - Browser: Firefox acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power 4 8 12 16 20 15.82 16.29 16.35 16.52 16.72 16.87 16.92 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance 1100 2200 3300 4400 5500 Min: 600 / Avg: 5373.02 / Max: 5849 Min: 600 / Avg: 5399.07 / Max: 5772 Min: 600 / Avg: 5454.84 / Max: 6158 Min: 600 / Avg: 5523.29 / Max: 5714 Min: 3000 / Avg: 5529.07 / Max: 5712 Min: 4300 / Avg: 5530.4 / Max: 5747 Min: 600 / Avg: 5530.4 / Max: 5753 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power 30 60 90 120 150 Min: 18.24 / Avg: 121.47 / Max: 134.81 Min: 19.37 / Avg: 120.87 / Max: 133.81 Min: 22.12 / Avg: 120.69 / Max: 132.35 Min: 20.37 / Avg: 118.91 / Max: 134.18 Min: 22.44 / Avg: 117.72 / Max: 134.77 Min: 19.41 / Avg: 116.65 / Max: 130.91 Min: 19.81 / Avg: 115.46 / Max: 133.38 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power 14 28 42 56 70 Min: 41.38 / Avg: 61.8 / Max: 66.75 Min: 41.13 / Avg: 61.59 / Max: 75 Min: 39.5 / Avg: 60.88 / Max: 67.25 Min: 40.63 / Avg: 60.73 / Max: 66.25 Min: 39.75 / Avg: 60.02 / Max: 65.38 Min: 40.5 / Avg: 59.89 / Max: 66.13 Min: 39.63 / Avg: 59.64 / Max: 65.63
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 3 - Input: Beauty 4K 10-bit acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power 0.4023 0.8046 1.2069 1.6092 2.0115 SE +/- 0.004, N = 3 SE +/- 0.004, N = 3 SE +/- 0.005, N = 3 SE +/- 0.002, N = 3 SE +/- 0.006, N = 3 SE +/- 0.003, N = 3 SE +/- 0.004, N = 3 1.761 1.764 1.776 1.785 1.786 1.787 1.788 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 3 - Input: Beauty 4K 10-bit amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.002 0.004 0.006 0.008 0.01 0.009 0.009 0.009 0.009 0.009 0.009 0.009 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance acpi-cpufreq schedutil 1000 2000 3000 4000 5000 Min: 3019 / Avg: 4987.42 / Max: 5563 Min: 600 / Avg: 4988.73 / Max: 5605 Min: 600 / Avg: 4991.02 / Max: 5700 Min: 600 / Avg: 4996.11 / Max: 5632 Min: 600 / Avg: 4996.53 / Max: 5719 Min: 4300 / Avg: 5008.06 / Max: 5569 Min: 4300 / Avg: 5018.45 / Max: 5690 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.2 CPU Power Consumption Monitor amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance acpi-cpufreq schedutil 40 80 120 160 200 Min: 20.39 / Avg: 192.72 / Max: 202.07 Min: 21.36 / Avg: 192.66 / Max: 202.29 Min: 19.88 / Avg: 192.66 / Max: 203.39 Min: 19.37 / Avg: 192.61 / Max: 201.8 Min: 22.4 / Avg: 192.6 / Max: 201.61 Min: 19.4 / Avg: 191.98 / Max: 201.79 Min: 19.2 / Avg: 191.88 / Max: 201.79 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 2.2 CPU Temperature Monitor amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power 20 40 60 80 100 Min: 45.63 / Avg: 86.15 / Max: 91.13 Min: 45.88 / Avg: 85.55 / Max: 91.5 Min: 44.88 / Avg: 85.53 / Max: 90.25 Min: 44.88 / Avg: 85.43 / Max: 92.5 Min: 44.63 / Avg: 85.42 / Max: 90.13 Min: 44.5 / Avg: 85.4 / Max: 89.38 Min: 44.88 / Avg: 84.7 / Max: 92.25
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 100, Lossless Compression amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.009 0.018 0.027 0.036 0.045 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.04 0.04 0.04 0.04 0.04 0.04 0.04 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -lm -ldl MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 100, Lossless Compression amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better WebP2 Image Encode 20220823 CPU Peak Freq (Highest CPU Core Frequency) Monitor amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 1000 2000 3000 4000 5000 Min: 600 / Avg: 4878.93 / Max: 5706 Min: 600 / Avg: 4881.15 / Max: 5616 Min: 4300 / Avg: 4881.18 / Max: 5690 Min: 600 / Avg: 4881.93 / Max: 5618 Min: 600 / Avg: 4882.79 / Max: 5659 Min: 600 / Avg: 4885.97 / Max: 5652 Min: 3610 / Avg: 4896.62 / Max: 5592 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better WebP2 Image Encode 20220823 CPU Power Consumption Monitor amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power 40 80 120 160 200 Min: 18.4 / Avg: 185.83 / Max: 201.6 Min: 20.18 / Avg: 185.55 / Max: 201.79 Min: 18.18 / Avg: 185.46 / Max: 202.2 Min: 19.85 / Avg: 185.41 / Max: 201.59 Min: 19.11 / Avg: 185.4 / Max: 201.98 Min: 19 / Avg: 185.33 / Max: 203.99 Min: 22.1 / Avg: 185.27 / Max: 201.59 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better WebP2 Image Encode 20220823 CPU Temperature Monitor amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 20 40 60 80 100 Min: 44.88 / Avg: 81.67 / Max: 95.13 Min: 44.63 / Avg: 81.24 / Max: 95.13 Min: 43.75 / Avg: 80.8 / Max: 94.88 Min: 43.63 / Avg: 80.62 / Max: 95 Min: 43.5 / Avg: 80.18 / Max: 94.5 Min: 43.5 / Avg: 80.15 / Max: 94.88 Min: 43.25 / Avg: 79.95 / Max: 93.38
LeelaChessZero Result
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.31.1 Backend: Eigen acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power 50 100 150 200 250 SE +/- 1.67, N = 3 SE +/- 1.53, N = 3 SE +/- 3.00, N = 3 SE +/- 2.08, N = 3 SE +/- 2.00, N = 3 SE +/- 2.90, N = 4 SE +/- 2.08, N = 3 232 235 236 236 236 236 238 1. (CXX) g++ options: -flto -pthread Nodes Per Second Per Watt
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better LeelaChessZero 0.31.1 Backend: Eigen acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power 0.2826 0.5652 0.8478 1.1304 1.413 1.225 1.237 1.244 1.245 1.245 1.249 1.256 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp performance, EPP performance 600 5079 5688 amd-pstate-epp powersave, EPP power 600 5080 5488 amd-pstate-epp powersave, EPP performance 600 5082 5451 amd-pstate-epp powersave, EPP balance_power 600 5093 5397 acpi-cpufreq performance 4300 5096 5422 amd-pstate-epp powersave, EPP balance_performance 600 5099 5509 acpi-cpufreq schedutil 3000 5105 5551 OpenBenchmarking.org Megahertz, More Is Better LeelaChessZero 0.31.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 18.3 189.9 202.4 amd-pstate-epp powersave, EPP balance_performance 19.3 189.7 202.8 amd-pstate-epp powersave, EPP performance 18.3 189.6 202.2 amd-pstate-epp powersave, EPP power 22.1 189.6 203.8 amd-pstate-epp powersave, EPP balance_power 20.1 189.5 204.0 acpi-cpufreq performance 19.6 189.3 202.4 acpi-cpufreq schedutil 19.3 188.9 202.7 OpenBenchmarking.org Watts, Fewer Is Better LeelaChessZero 0.31.1 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.6 84.3 92.8 amd-pstate-epp powersave, EPP power 47.3 83.8 89.8 amd-pstate-epp powersave, EPP balance_performance 46.3 82.8 88.9 amd-pstate-epp powersave, EPP performance 45.9 82.7 89.9 acpi-cpufreq performance 45.8 82.2 88.9 amd-pstate-epp powersave, EPP balance_power 45.9 82.2 89.4 acpi-cpufreq schedutil 45.0 82.2 88.5 OpenBenchmarking.org Celsius, Fewer Is Better LeelaChessZero 0.31.1 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.31.1 Backend: BLAS amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance 60 120 180 240 300 SE +/- 2.67, N = 3 SE +/- 0.58, N = 3 SE +/- 3.00, N = 3 SE +/- 1.86, N = 3 SE +/- 1.20, N = 3 SE +/- 2.52, N = 3 SE +/- 2.85, N = 3 246 247 248 249 252 252 253 1. (CXX) g++ options: -flto -pthread Nodes Per Second Per Watt
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better LeelaChessZero 0.31.1 Backend: BLAS amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil 0.315 0.63 0.945 1.26 1.575 1.353 1.361 1.369 1.374 1.384 1.398 1.400 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5107 5676 amd-pstate-epp performance, EPP performance 600 5110 5707 amd-pstate-epp powersave, EPP power 600 5111 5411 amd-pstate-epp powersave, EPP performance 600 5111 5510 acpi-cpufreq performance 4300 5113 5407 acpi-cpufreq schedutil 3000 5116 5384 amd-pstate-epp powersave, EPP balance_performance 600 5121 5519 OpenBenchmarking.org Megahertz, More Is Better LeelaChessZero 0.31.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 11.0 182.1 199.3 amd-pstate-epp powersave, EPP power 22.2 181.8 198.3 acpi-cpufreq performance 19.8 181.5 197.9 amd-pstate-epp powersave, EPP balance_power 19.1 181.3 198.5 amd-pstate-epp powersave, EPP balance_performance 18.9 181.1 198.2 amd-pstate-epp powersave, EPP performance 19.7 181.0 198.3 acpi-cpufreq schedutil 19.1 180.0 196.4 OpenBenchmarking.org Watts, Fewer Is Better LeelaChessZero 0.31.1 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 40.6 82.4 90.9 amd-pstate-epp powersave, EPP power 39.6 81.8 90.0 amd-pstate-epp powersave, EPP performance 39.5 81.2 90.0 amd-pstate-epp powersave, EPP balance_performance 39.4 81.0 89.5 amd-pstate-epp powersave, EPP balance_power 38.4 80.7 89.8 acpi-cpufreq performance 38.9 80.4 87.4 acpi-cpufreq schedutil 37.6 79.9 88.1 OpenBenchmarking.org Celsius, Fewer Is Better LeelaChessZero 0.31.1 CPU Temperature Monitor 20 40 60 80 100
Whisperfile Result
OpenBenchmarking.org Seconds, Fewer Is Better Whisperfile 20Aug24 Model Size: Medium amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance 70 140 210 280 350 SE +/- 2.44, N = 3 SE +/- 1.54, N = 3 SE +/- 1.62, N = 3 SE +/- 2.43, N = 3 SE +/- 2.13, N = 3 SE +/- 1.13, N = 3 SE +/- 2.49, N = 3 330.77 330.08 329.79 329.66 329.45 329.42 329.05 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5066 5702 amd-pstate-epp powersave, EPP power 600 5080 5726 amd-pstate-epp powersave, EPP performance 600 5083 5715 amd-pstate-epp performance, EPP performance 600 5084 5691 amd-pstate-epp powersave, EPP balance_performance 600 5084 5725 acpi-cpufreq performance 4300 5097 5591 acpi-cpufreq schedutil 4135 5117 5704 OpenBenchmarking.org Megahertz, More Is Better Whisperfile 20Aug24 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.3 144.8 160.3 acpi-cpufreq performance 19.8 144.7 161.2 amd-pstate-epp powersave, EPP balance_power 19.5 144.6 160.3 amd-pstate-epp powersave, EPP power 22.1 144.4 160.8 amd-pstate-epp powersave, EPP performance 19.6 144.3 160.0 acpi-cpufreq schedutil 19.3 144.3 160.5 amd-pstate-epp powersave, EPP balance_performance 19.6 143.8 159.0 OpenBenchmarking.org Watts, Fewer Is Better Whisperfile 20Aug24 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 46.6 69.1 76.1 amd-pstate-epp performance, EPP performance 46.5 69.0 76.6 acpi-cpufreq performance 46.0 68.3 76.0 amd-pstate-epp powersave, EPP balance_power 45.6 68.1 75.5 amd-pstate-epp powersave, EPP power 45.8 67.9 75.8 amd-pstate-epp powersave, EPP performance 44.8 67.6 75.9 acpi-cpufreq schedutil 45.1 67.6 74.8 OpenBenchmarking.org Celsius, Fewer Is Better Whisperfile 20Aug24 CPU Temperature Monitor 20 40 60 80 100
Timed LLVM Compilation This test times how long it takes to compile/build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Unix Makefiles amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance 70 140 210 280 350 SE +/- 1.10, N = 3 SE +/- 0.83, N = 3 SE +/- 1.08, N = 3 SE +/- 1.74, N = 3 SE +/- 1.63, N = 3 SE +/- 0.74, N = 3 SE +/- 0.77, N = 3 325.26 323.64 323.63 322.38 322.27 321.53 319.92 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5182 5719 amd-pstate-epp powersave, EPP power 600 5182 5706 amd-pstate-epp performance, EPP performance 600 5185 5695 amd-pstate-epp powersave, EPP balance_power 600 5186 5697 acpi-cpufreq performance 4300 5200 5641 amd-pstate-epp powersave, EPP performance 600 5202 5718 acpi-cpufreq schedutil 4300 5218 5688 OpenBenchmarking.org Megahertz, More Is Better Timed LLVM Compilation 16.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.7 180.9 202.0 amd-pstate-epp powersave, EPP balance_performance 19.4 180.5 201.2 acpi-cpufreq performance 19.8 180.4 202.2 amd-pstate-epp powersave, EPP performance 19.8 180.1 204.8 amd-pstate-epp powersave, EPP balance_power 19.9 180.0 201.8 acpi-cpufreq schedutil 18.5 179.9 202.8 amd-pstate-epp powersave, EPP power 22.5 179.4 201.1 OpenBenchmarking.org Watts, Fewer Is Better Timed LLVM Compilation 16.0 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.3 86.2 93.1 amd-pstate-epp powersave, EPP power 47.0 85.1 93.3 amd-pstate-epp powersave, EPP balance_performance 45.8 84.5 92.9 amd-pstate-epp powersave, EPP performance 45.6 83.5 91.6 acpi-cpufreq schedutil 45.4 83.1 90.6 amd-pstate-epp powersave, EPP balance_power 45.0 82.7 92.3 acpi-cpufreq performance 45.6 82.5 90.8 OpenBenchmarking.org Celsius, Fewer Is Better Timed LLVM Compilation 16.0 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Ninja amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance 70 140 210 280 350 SE +/- 0.24, N = 3 SE +/- 0.37, N = 3 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 SE +/- 0.15, N = 3 SE +/- 0.29, N = 3 SE +/- 0.06, N = 3 299.05 298.93 298.66 298.17 298.15 298.05 296.68 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp performance, EPP performance 600 5108 5692 amd-pstate-epp powersave, EPP balance_power 600 5132 5730 amd-pstate-epp powersave, EPP power 600 5146 5714 amd-pstate-epp powersave, EPP balance_performance 600 5156 5710 amd-pstate-epp powersave, EPP performance 600 5160 5683 acpi-cpufreq performance 4300 5160 5638 acpi-cpufreq schedutil 4300 5170 5688 OpenBenchmarking.org Megahertz, More Is Better Timed LLVM Compilation 16.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.8 189.0 201.2 amd-pstate-epp powersave, EPP power 22.5 188.8 202.6 amd-pstate-epp performance, EPP performance 22.1 188.7 204.0 amd-pstate-epp powersave, EPP balance_power 20.4 188.7 201.6 amd-pstate-epp powersave, EPP performance 20.2 188.7 201.2 acpi-cpufreq schedutil 18.7 188.6 202.2 amd-pstate-epp powersave, EPP balance_performance 18.9 188.6 201.0 OpenBenchmarking.org Watts, Fewer Is Better Timed LLVM Compilation 16.0 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.8 85.8 93.5 amd-pstate-epp powersave, EPP power 45.5 84.7 92.5 amd-pstate-epp powersave, EPP balance_performance 44.4 84.0 91.6 amd-pstate-epp powersave, EPP performance 45.0 83.8 91.5 acpi-cpufreq schedutil 44.8 83.4 91.8 acpi-cpufreq performance 44.8 82.8 91.8 amd-pstate-epp powersave, EPP balance_power 44.4 82.8 91.6 OpenBenchmarking.org Celsius, Fewer Is Better Timed LLVM Compilation 16.0 CPU Temperature Monitor 20 40 60 80 100
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 3 - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 11.66 11.76 11.85 11.89 11.90 11.91 11.94 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 3 - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power 0.0142 0.0284 0.0426 0.0568 0.071 0.062 0.062 0.063 0.063 0.063 0.063 0.063 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4790 5702 amd-pstate-epp powersave, EPP performance 600 4812 5715 amd-pstate-epp powersave, EPP balance_power 600 4813 5651 amd-pstate-epp powersave, EPP balance_performance 600 4816 5669 amd-pstate-epp performance, EPP performance 600 4824 5696 acpi-cpufreq performance 4281 4826 5706 acpi-cpufreq schedutil 3000 4836 5645 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.9 189.4 202.0 amd-pstate-epp powersave, EPP performance 19.6 188.9 201.3 amd-pstate-epp powersave, EPP balance_power 19.8 188.7 201.8 amd-pstate-epp powersave, EPP power 22.5 188.6 202.0 amd-pstate-epp powersave, EPP balance_performance 17.4 188.5 201.6 acpi-cpufreq performance 19.7 188.5 201.6 acpi-cpufreq schedutil 19.3 187.5 201.4 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 46.9 83.6 88.0 acpi-cpufreq performance 45.9 83.6 87.6 amd-pstate-epp performance, EPP performance 46.4 83.5 89.6 acpi-cpufreq schedutil 45.5 83.2 86.5 amd-pstate-epp powersave, EPP balance_performance 45.9 83.1 87.6 amd-pstate-epp powersave, EPP performance 46.0 82.8 89.3 amd-pstate-epp powersave, EPP balance_power 45.8 82.2 90.1 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 2.2 CPU Temperature Monitor 20 40 60 80 100
BRL-CAD BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.38.2 VGR Performance Metric acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance 110K 220K 330K 440K 550K 491302 491318 494060 496949 497167 497238 498314 1. (CXX) g++ options: -std=c++17 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lnetpbm -lregex_brl -lz_brl -lassimp -ldl -lm -ltk8.6 VGR Performance Metric Per Watt
OpenBenchmarking.org VGR Performance Metric Per Watt, More Is Better BRL-CAD 7.38.2 VGR Performance Metric acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance 500 1000 1500 2000 2500 2480.96 2486.07 2497.35 2510.51 2511.89 2517.17 2526.41 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4645 5524 amd-pstate-epp powersave, EPP power 600 4648 5533 amd-pstate-epp performance, EPP performance 600 4656 5686 amd-pstate-epp powersave, EPP performance 600 4656 5704 amd-pstate-epp powersave, EPP balance_power 600 4659 5500 acpi-cpufreq performance 4300 4672 5512 acpi-cpufreq schedutil 4300 4678 5635 OpenBenchmarking.org Megahertz, More Is Better BRL-CAD 7.38.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 20.6 198.0 201.2 amd-pstate-epp powersave, EPP balance_power 18.8 198.0 201.2 amd-pstate-epp performance, EPP performance 21.9 197.9 201.2 amd-pstate-epp powersave, EPP power 22.6 197.8 201.0 amd-pstate-epp powersave, EPP balance_performance 20.1 197.6 201.8 acpi-cpufreq schedutil 19.6 197.5 201.2 amd-pstate-epp powersave, EPP performance 20.1 197.2 201.2 OpenBenchmarking.org Watts, Fewer Is Better BRL-CAD 7.38.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 47.4 84.2 90.1 amd-pstate-epp performance, EPP performance 47.0 83.8 89.3 acpi-cpufreq performance 46.6 83.5 90.5 amd-pstate-epp powersave, EPP balance_power 46.5 83.0 89.6 acpi-cpufreq schedutil 46.1 82.8 91.3 amd-pstate-epp powersave, EPP performance 45.8 82.5 87.9 amd-pstate-epp powersave, EPP power 46.6 82.3 88.5 OpenBenchmarking.org Celsius, Fewer Is Better BRL-CAD 7.38.2 CPU Temperature Monitor 20 40 60 80 100
GravityMark Result
OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.87 Resolution: 3840 x 2160 - Renderer: Vulkan amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 40 80 120 160 200 SE +/- 0.40, N = 3 SE +/- 0.47, N = 3 SE +/- 0.48, N = 3 SE +/- 0.45, N = 3 SE +/- 0.38, N = 3 SE +/- 0.54, N = 3 SE +/- 0.54, N = 3 185.9 186.3 186.7 186.8 186.9 187.1 187.2 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better GravityMark 1.87 Resolution: 3840 x 2160 - Renderer: Vulkan amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power 1.2359 2.4718 3.7077 4.9436 6.1795 3.709 4.200 4.321 4.592 5.232 5.410 5.493 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 3043 5725 amd-pstate-epp powersave, EPP balance_power 600 3726 5737 amd-pstate-epp powersave, EPP balance_performance 600 4381 5722 acpi-cpufreq schedutil 3000 4924 5739 acpi-cpufreq performance 4300 5427 5722 amd-pstate-epp powersave, EPP performance 600 5496 5755 amd-pstate-epp performance, EPP performance 600 5552 5725 OpenBenchmarking.org Megahertz, More Is Better GravityMark 1.87 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 11.3 50.3 63.6 amd-pstate-epp powersave, EPP performance 19.6 44.4 63.7 acpi-cpufreq performance 20.3 43.3 66.8 acpi-cpufreq schedutil 19.4 40.8 63.1 amd-pstate-epp powersave, EPP balance_performance 19.4 35.7 62.4 amd-pstate-epp powersave, EPP balance_power 20.3 34.5 63.4 amd-pstate-epp powersave, EPP power 19.8 33.8 68.2 OpenBenchmarking.org Watts, Fewer Is Better GravityMark 1.87 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 38.0 43.7 61.0 amd-pstate-epp powersave, EPP performance 39.5 43.4 60.9 amd-pstate-epp powersave, EPP power 37.9 41.4 58.9 amd-pstate-epp powersave, EPP balance_power 38.4 41.3 59.5 amd-pstate-epp powersave, EPP balance_performance 37.3 41.3 59.0 acpi-cpufreq performance 36.1 41.2 58.3 acpi-cpufreq schedutil 35.0 40.9 57.5 OpenBenchmarking.org Celsius, Fewer Is Better GravityMark 1.87 CPU Temperature Monitor 20 40 60 80 100
OSPRay Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: particle_volume/pathtracer/real_time amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance 60 120 180 240 300 SE +/- 0.48, N = 3 SE +/- 0.76, N = 3 SE +/- 0.77, N = 3 SE +/- 0.71, N = 3 SE +/- 0.73, N = 3 SE +/- 0.47, N = 3 SE +/- 0.18, N = 3 276.13 276.18 276.28 276.33 276.34 276.41 276.57 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 3.2 Benchmark: particle_volume/pathtracer/real_time amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power 0.3186 0.6372 0.9558 1.2744 1.593 1.410 1.412 1.413 1.413 1.414 1.414 1.416 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4801 5311 amd-pstate-epp powersave, EPP balance_performance 600 4805 5308 amd-pstate-epp powersave, EPP power 600 4824 5476 amd-pstate-epp powersave, EPP balance_power 600 4826 5346 amd-pstate-epp performance, EPP performance 600 4846 5686 acpi-cpufreq performance 4300 4865 5273 acpi-cpufreq schedutil 3621 4881 5509 OpenBenchmarking.org Megahertz, More Is Better OSPRay 3.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.6 195.8 205.2 acpi-cpufreq schedutil 18.3 195.7 204.4 acpi-cpufreq performance 19.7 195.7 201.2 amd-pstate-epp powersave, EPP performance 19.6 195.5 202.2 amd-pstate-epp powersave, EPP balance_performance 19.3 195.5 205.4 amd-pstate-epp powersave, EPP balance_power 20.0 195.4 205.2 amd-pstate-epp powersave, EPP power 22.5 195.2 201.0 OpenBenchmarking.org Watts, Fewer Is Better OSPRay 3.2 CPU Power Consumption Monitor 60 120 180 240 300 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.4 82.9 88.5 amd-pstate-epp powersave, EPP power 47.4 82.5 87.6 amd-pstate-epp powersave, EPP performance 46.1 81.9 86.8 amd-pstate-epp powersave, EPP balance_performance 46.3 81.8 87.0 acpi-cpufreq schedutil 45.8 81.7 86.4 acpi-cpufreq performance 45.8 81.5 86.8 amd-pstate-epp powersave, EPP balance_power 46.0 81.1 86.1 OpenBenchmarking.org Celsius, Fewer Is Better OSPRay 3.2 CPU Temperature Monitor 20 40 60 80 100
simdjson Result
OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: PartialTweets amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.12, N = 15 SE +/- 0.13, N = 6 SE +/- 0.18, N = 3 SE +/- 0.10, N = 15 SE +/- 0.14, N = 3 12.68 12.73 12.75 12.95 12.97 13.00 13.13 1. (CXX) g++ options: -O3 -lrt GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 3.10 Throughput Test: PartialTweets amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance 0.0556 0.1112 0.1668 0.2224 0.278 0.223 0.234 0.236 0.237 0.245 0.246 0.247 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5320 5740 amd-pstate-epp powersave, EPP performance 600 5322 5653 amd-pstate-epp performance, EPP performance 600 5334 5724 amd-pstate-epp powersave, EPP power 600 5360 5722 acpi-cpufreq performance 4300 5402 5727 acpi-cpufreq schedutil 3000 5461 5677 amd-pstate-epp powersave, EPP balance_performance 600 5463 5717 OpenBenchmarking.org Megahertz, More Is Better simdjson 3.10 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.5 56.9 59.5 amd-pstate-epp powersave, EPP balance_performance 19.6 55.6 58.9 acpi-cpufreq schedutil 18.8 54.5 58.8 acpi-cpufreq performance 15.8 54.1 59.5 amd-pstate-epp powersave, EPP power 22.3 53.3 62.7 amd-pstate-epp powersave, EPP balance_power 19.9 52.9 59.3 amd-pstate-epp powersave, EPP performance 19.1 51.6 58.3 OpenBenchmarking.org Watts, Fewer Is Better simdjson 3.10 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 40.1 62.5 64.3 amd-pstate-epp performance, EPP performance 41.3 61.7 62.5 acpi-cpufreq schedutil 38.9 60.6 65.4 amd-pstate-epp powersave, EPP power 41.3 60.1 64.8 acpi-cpufreq performance 39.6 59.0 65.5 amd-pstate-epp powersave, EPP balance_power 39.4 57.1 63.8 amd-pstate-epp powersave, EPP performance 40.1 56.6 63.8 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 3.10 CPU Temperature Monitor 20 40 60 80 100
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 5 - Input: Beauty 4K 10-bit acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power 2 4 6 8 10 SE +/- 0.032, N = 3 SE +/- 0.023, N = 3 SE +/- 0.035, N = 3 SE +/- 0.034, N = 3 SE +/- 0.015, N = 3 SE +/- 0.065, N = 3 SE +/- 0.045, N = 3 7.493 7.585 7.627 7.634 7.660 7.666 7.691 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 5 - Input: Beauty 4K 10-bit amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0101 0.0202 0.0303 0.0404 0.0505 0.045 0.045 0.045 0.045 0.045 0.045 0.045 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5060 5582 amd-pstate-epp performance, EPP performance 600 5080 5698 amd-pstate-epp powersave, EPP power 600 5088 5710 amd-pstate-epp powersave, EPP balance_power 600 5106 5697 amd-pstate-epp powersave, EPP balance_performance 600 5108 5686 acpi-cpufreq schedutil 4300 5140 5568 acpi-cpufreq performance 4300 5145 5683 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.3 169.9 200.8 amd-pstate-epp powersave, EPP power 22.8 169.8 201.0 amd-pstate-epp powersave, EPP performance 18.4 169.5 200.6 amd-pstate-epp powersave, EPP balance_performance 19.6 169.5 200.8 amd-pstate-epp powersave, EPP balance_power 20.0 169.5 200.8 acpi-cpufreq schedutil 19.2 168.2 200.9 acpi-cpufreq performance 19.5 167.5 201.2 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 47.3 83.4 88.4 acpi-cpufreq schedutil 45.4 82.4 86.6 acpi-cpufreq performance 45.8 82.3 86.6 amd-pstate-epp powersave, EPP balance_performance 45.8 82.3 87.0 amd-pstate-epp powersave, EPP performance 45.6 82.2 88.8 amd-pstate-epp performance, EPP performance 46.0 82.1 88.3 amd-pstate-epp powersave, EPP balance_power 45.6 81.4 86.6 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 2.2 CPU Temperature Monitor 20 40 60 80 100
Blender Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Pabellon Barcelona - Compute: CPU-Only acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance 30 60 90 120 150 SE +/- 0.31, N = 3 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.26, N = 3 SE +/- 0.18, N = 3 SE +/- 0.04, N = 3 147.28 146.99 146.89 146.83 146.65 146.55 146.37 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp performance, EPP performance 600 4622 5689 amd-pstate-epp powersave, EPP performance 600 4623 5714 amd-pstate-epp powersave, EPP power 600 4630 5712 amd-pstate-epp powersave, EPP balance_power 600 4636 5522 amd-pstate-epp powersave, EPP balance_performance 600 4644 5668 acpi-cpufreq performance 4300 4662 5615 acpi-cpufreq schedutil 4300 4695 5621 OpenBenchmarking.org Megahertz, More Is Better Blender 4.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_power 19.9 194.6 201.0 acpi-cpufreq performance 20.2 194.6 201.4 amd-pstate-epp powersave, EPP performance 19.9 194.5 201.2 amd-pstate-epp performance, EPP performance 16.5 194.4 201.2 amd-pstate-epp powersave, EPP balance_performance 19.7 194.4 201.0 acpi-cpufreq schedutil 19.0 194.2 201.0 amd-pstate-epp powersave, EPP power 22.4 194.0 201.2 OpenBenchmarking.org Watts, Fewer Is Better Blender 4.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.9 79.9 81.4 amd-pstate-epp powersave, EPP balance_performance 47.0 79.2 81.4 amd-pstate-epp powersave, EPP power 47.3 79.1 82.3 amd-pstate-epp powersave, EPP performance 46.0 78.4 79.9 acpi-cpufreq performance 46.9 78.4 80.1 acpi-cpufreq schedutil 46.6 78.3 80.0 amd-pstate-epp powersave, EPP balance_power 46.3 78.1 79.6 OpenBenchmarking.org Celsius, Fewer Is Better Blender 4.2 CPU Temperature Monitor 20 40 60 80 100
OSPRay Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: particle_volume/scivis/real_time amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance acpi-cpufreq schedutil 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 11.95 11.97 11.97 11.97 11.98 11.98 12.00 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 3.2 Benchmark: particle_volume/scivis/real_time amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 0.014 0.028 0.042 0.056 0.07 0.061 0.061 0.061 0.061 0.061 0.061 0.062 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4988 5506 amd-pstate-epp powersave, EPP balance_performance 600 4995 5316 amd-pstate-epp powersave, EPP balance_power 600 4999 5313 amd-pstate-epp performance, EPP performance 600 5015 5694 acpi-cpufreq schedutil 3000 5031 5320 amd-pstate-epp powersave, EPP performance 3560 5031 5321 acpi-cpufreq performance 4300 5047 5462 OpenBenchmarking.org Megahertz, More Is Better OSPRay 3.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_power 20.1 195.1 201.4 acpi-cpufreq schedutil 14.6 195.1 201.0 acpi-cpufreq performance 20.0 195.0 200.9 amd-pstate-epp powersave, EPP performance 19.9 194.9 203.2 amd-pstate-epp powersave, EPP balance_performance 19.9 194.9 201.2 amd-pstate-epp powersave, EPP power 22.7 194.9 201.3 amd-pstate-epp performance, EPP performance 20.9 194.8 201.0 OpenBenchmarking.org Watts, Fewer Is Better OSPRay 3.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.4 85.7 87.6 amd-pstate-epp powersave, EPP power 46.8 85.5 87.1 amd-pstate-epp powersave, EPP performance 46.4 84.8 86.3 amd-pstate-epp powersave, EPP balance_performance 46.3 84.5 86.0 acpi-cpufreq schedutil 45.6 84.4 86.1 amd-pstate-epp powersave, EPP balance_power 45.6 83.7 85.6 acpi-cpufreq performance 45.8 83.4 85.8 OpenBenchmarking.org Celsius, Fewer Is Better OSPRay 3.2 CPU Temperature Monitor 20 40 60 80 100
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 95, Compression Effort 7 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0383 0.0766 0.1149 0.1532 0.1915 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.17 0.17 0.17 0.17 0.17 0.17 0.17 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -lm -ldl MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 95, Compression Effort 7 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0002 0.0004 0.0006 0.0008 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4663 5531 amd-pstate-epp powersave, EPP balance_performance 600 4673 5591 amd-pstate-epp powersave, EPP balance_power 600 4683 5603 amd-pstate-epp powersave, EPP performance 600 4687 5612 acpi-cpufreq performance 4300 4720 5461 amd-pstate-epp performance, EPP performance 600 4721 5698 acpi-cpufreq schedutil 4300 4730 5516 OpenBenchmarking.org Megahertz, More Is Better WebP2 Image Encode 20220823 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_power 18.5 189.6 201.2 acpi-cpufreq performance 19.6 189.4 201.4 amd-pstate-epp powersave, EPP performance 19.8 189.4 201.4 amd-pstate-epp powersave, EPP balance_performance 20.1 189.4 201.6 amd-pstate-epp powersave, EPP power 22.3 189.3 202.0 acpi-cpufreq schedutil 19.3 189.2 201.6 amd-pstate-epp performance, EPP performance 20.4 189.1 201.6 OpenBenchmarking.org Watts, Fewer Is Better WebP2 Image Encode 20220823 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.8 81.9 94.8 amd-pstate-epp powersave, EPP power 46.0 81.8 94.6 amd-pstate-epp powersave, EPP balance_performance 45.3 81.2 93.5 amd-pstate-epp powersave, EPP performance 44.9 81.1 93.4 amd-pstate-epp powersave, EPP balance_power 45.5 80.5 93.3 acpi-cpufreq performance 44.8 80.4 93.9 acpi-cpufreq schedutil 44.8 80.1 93.1 OpenBenchmarking.org Celsius, Fewer Is Better WebP2 Image Encode 20220823 CPU Temperature Monitor 20 40 60 80 100
Whisperfile Result
OpenBenchmarking.org Seconds, Fewer Is Better Whisperfile 20Aug24 Model Size: Small acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 30 60 90 120 150 SE +/- 0.82, N = 3 SE +/- 0.07, N = 3 SE +/- 0.14, N = 3 SE +/- 0.24, N = 3 SE +/- 0.58, N = 3 SE +/- 0.49, N = 3 SE +/- 0.15, N = 3 127.55 126.56 126.34 126.32 126.21 125.99 125.84 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 5056 5621 amd-pstate-epp powersave, EPP balance_performance 600 5062 5603 amd-pstate-epp powersave, EPP performance 600 5063 5673 amd-pstate-epp powersave, EPP balance_power 600 5082 5693 amd-pstate-epp performance, EPP performance 600 5130 5740 acpi-cpufreq performance 4300 5159 5656 acpi-cpufreq schedutil 3616 5167 5697 OpenBenchmarking.org Megahertz, More Is Better Whisperfile 20Aug24 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 20.5 138.3 153.1 amd-pstate-epp powersave, EPP balance_power 18.9 137.9 153.3 amd-pstate-epp performance, EPP performance 21.6 137.9 152.7 amd-pstate-epp powersave, EPP power 22.0 137.7 153.4 acpi-cpufreq schedutil 19.5 137.5 152.9 amd-pstate-epp powersave, EPP balance_performance 19.7 137.4 153.0 amd-pstate-epp powersave, EPP performance 18.8 137.2 152.9 OpenBenchmarking.org Watts, Fewer Is Better Whisperfile 20Aug24 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 44.3 67.2 74.5 amd-pstate-epp powersave, EPP balance_performance 44.1 67.2 74.4 amd-pstate-epp powersave, EPP balance_power 43.1 66.3 74.3 acpi-cpufreq performance 43.1 66.3 73.3 amd-pstate-epp powersave, EPP power 43.4 66.1 75.0 acpi-cpufreq schedutil 42.9 65.7 72.8 amd-pstate-epp powersave, EPP performance 42.4 65.5 73.8 OpenBenchmarking.org Celsius, Fewer Is Better Whisperfile 20Aug24 CPU Temperature Monitor 20 40 60 80 100
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 8 - Input: Beauty 4K 10-bit acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power 3 6 9 12 15 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 10.62 10.65 10.67 10.72 10.72 10.75 10.80 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 8 - Input: Beauty 4K 10-bit amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power 0.0151 0.0302 0.0453 0.0604 0.0755 0.066 0.066 0.066 0.066 0.066 0.067 0.067 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5070 5600 amd-pstate-epp powersave, EPP performance 600 5110 5675 amd-pstate-epp performance, EPP performance 600 5129 5692 amd-pstate-epp powersave, EPP balance_performance 600 5135 5703 amd-pstate-epp powersave, EPP power 600 5144 5717 acpi-cpufreq schedutil 4300 5166 5614 acpi-cpufreq performance 4300 5177 5684 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.8 162.6 200.0 amd-pstate-epp powersave, EPP power 22.0 161.9 199.9 amd-pstate-epp powersave, EPP performance 19.5 161.4 199.7 acpi-cpufreq performance 19.4 161.3 199.1 amd-pstate-epp powersave, EPP balance_power 19.9 161.1 200.4 amd-pstate-epp powersave, EPP balance_performance 18.7 161.1 199.8 acpi-cpufreq schedutil 18.8 161.0 198.8 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 46.0 82.0 87.8 amd-pstate-epp performance, EPP performance 44.9 81.7 88.1 amd-pstate-epp powersave, EPP balance_performance 44.9 81.2 86.3 amd-pstate-epp powersave, EPP performance 44.6 81.2 86.9 acpi-cpufreq schedutil 44.3 81.0 86.8 acpi-cpufreq performance 44.3 80.7 86.5 amd-pstate-epp powersave, EPP balance_power 44.3 80.5 86.1 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 2.2 CPU Temperature Monitor 20 40 60 80 100
simdjson Result
OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: DistinctUserID amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil 4 8 12 16 20 SE +/- 0.13, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.13, N = 7 SE +/- 0.19, N = 3 SE +/- 0.11, N = 9 SE +/- 0.02, N = 3 13.08 13.18 13.26 13.50 13.59 13.66 13.74 1. (CXX) g++ options: -O3 -lrt GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 3.10 Throughput Test: DistinctUserID amd-pstate-epp performance, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance 0.059 0.118 0.177 0.236 0.295 0.230 0.245 0.246 0.250 0.255 0.258 0.262 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5290 5715 amd-pstate-epp powersave, EPP balance_power 600 5294 5731 amd-pstate-epp powersave, EPP performance 600 5308 5722 acpi-cpufreq performance 4300 5394 5708 amd-pstate-epp powersave, EPP power 600 5407 5733 amd-pstate-epp performance, EPP performance 600 5480 5722 acpi-cpufreq schedutil 3000 5496 5675 OpenBenchmarking.org Megahertz, More Is Better simdjson 3.10 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.2 57.0 59.6 acpi-cpufreq schedutil 19.4 55.8 59.0 amd-pstate-epp powersave, EPP power 22.4 54.3 59.1 acpi-cpufreq performance 19.6 54.1 57.1 amd-pstate-epp powersave, EPP balance_power 20.0 52.9 59.2 amd-pstate-epp powersave, EPP performance 19.6 52.2 58.8 amd-pstate-epp powersave, EPP balance_performance 16.9 51.1 53.6 OpenBenchmarking.org Watts, Fewer Is Better simdjson 3.10 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max acpi-cpufreq schedutil 38.8 61.6 65.1 amd-pstate-epp performance, EPP performance 41.4 61.6 62.5 amd-pstate-epp powersave, EPP power 41.0 60.7 64.3 acpi-cpufreq performance 39.3 58.3 59.8 amd-pstate-epp powersave, EPP performance 39.5 57.5 64.1 amd-pstate-epp powersave, EPP balance_power 39.1 57.1 63.6 amd-pstate-epp powersave, EPP balance_performance 39.5 55.4 56.6 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 3.10 CPU Temperature Monitor 20 40 60 80 100
OSPRay Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: particle_volume/ao/real_time amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance acpi-cpufreq schedutil 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 11.93 11.93 11.94 11.96 11.97 11.97 11.99 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 3.2 Benchmark: particle_volume/ao/real_time amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.014 0.028 0.042 0.056 0.07 0.062 0.062 0.062 0.062 0.062 0.062 0.062 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4866 5116 amd-pstate-epp powersave, EPP power 600 4886 5599 amd-pstate-epp powersave, EPP balance_performance 600 4909 5644 amd-pstate-epp powersave, EPP performance 600 4919 5716 amd-pstate-epp performance, EPP performance 600 4942 5705 acpi-cpufreq schedutil 3000 4946 5529 acpi-cpufreq performance 4300 4966 5438 OpenBenchmarking.org Megahertz, More Is Better OSPRay 3.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.6 193.2 201.2 acpi-cpufreq schedutil 19.1 193.1 210.2 acpi-cpufreq performance 19.9 193.1 201.8 amd-pstate-epp powersave, EPP balance_performance 19.5 193.1 201.0 amd-pstate-epp powersave, EPP balance_power 20.3 192.8 201.0 amd-pstate-epp powersave, EPP performance 19.9 192.8 201.0 amd-pstate-epp powersave, EPP power 22.6 192.7 201.0 OpenBenchmarking.org Watts, Fewer Is Better OSPRay 3.2 CPU Power Consumption Monitor 60 120 180 240 300 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 45.4 84.7 86.6 amd-pstate-epp performance, EPP performance 45.5 84.5 86.9 amd-pstate-epp powersave, EPP performance 44.6 84.0 86.0 amd-pstate-epp powersave, EPP balance_performance 44.5 83.7 85.6 acpi-cpufreq schedutil 44.3 83.2 85.1 acpi-cpufreq performance 44.3 82.8 84.8 amd-pstate-epp powersave, EPP balance_power 44.4 82.8 84.9 OpenBenchmarking.org Celsius, Fewer Is Better OSPRay 3.2 CPU Temperature Monitor 20 40 60 80 100
Radiance Benchmark This is a benchmark of NREL Radiance, a synthetic imaging system that is open-source and developed by the Lawrence Berkeley National Laboratory in California. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Radiance Benchmark 5.0 Test: Serial amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power 70 140 210 280 350 302.74 302.72 302.14 300.02 298.84 297.91 296.54 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5409 5716 acpi-cpufreq schedutil 4300 5429 5697 amd-pstate-epp performance, EPP performance 600 5459 5716 amd-pstate-epp powersave, EPP balance_power 600 5470 5727 amd-pstate-epp powersave, EPP power 600 5474 5709 acpi-cpufreq performance 4300 5498 5667 amd-pstate-epp powersave, EPP performance 600 5513 5731 OpenBenchmarking.org Megahertz, More Is Better Radiance Benchmark 5.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.4 56.0 57.6 acpi-cpufreq performance 20.0 56.0 57.7 acpi-cpufreq schedutil 19.0 54.3 56.5 amd-pstate-epp powersave, EPP performance 19.7 54.2 56.3 amd-pstate-epp powersave, EPP power 22.8 54.2 56.1 amd-pstate-epp powersave, EPP balance_power 20.2 54.2 56.9 amd-pstate-epp powersave, EPP balance_performance 20.4 52.7 55.1 OpenBenchmarking.org Watts, Fewer Is Better Radiance Benchmark 5.0 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max acpi-cpufreq performance 38.8 64.7 67.1 amd-pstate-epp performance, EPP performance 41.9 64.4 65.1 acpi-cpufreq schedutil 39.0 63.5 66.8 amd-pstate-epp powersave, EPP power 40.0 62.8 65.8 amd-pstate-epp powersave, EPP performance 38.6 62.6 65.6 amd-pstate-epp powersave, EPP balance_power 38.4 62.5 65.6 amd-pstate-epp powersave, EPP balance_performance 39.0 60.6 66.1 OpenBenchmarking.org Celsius, Fewer Is Better Radiance Benchmark 5.0 CPU Temperature Monitor 20 40 60 80 100
simdjson Result
OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: Kostya acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 7.55 7.58 7.65 7.68 7.70 7.75 7.75 1. (CXX) g++ options: -O3 -lrt GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 3.10 Throughput Test: Kostya acpi-cpufreq performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power 0.0313 0.0626 0.0939 0.1252 0.1565 0.131 0.133 0.136 0.137 0.138 0.138 0.139 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5352 5697 amd-pstate-epp powersave, EPP balance_power 600 5363 5702 acpi-cpufreq performance 4300 5405 5649 amd-pstate-epp powersave, EPP power 600 5418 5706 amd-pstate-epp powersave, EPP performance 600 5427 5712 acpi-cpufreq schedutil 3031 5466 5719 amd-pstate-epp performance, EPP performance 600 5497 5703 OpenBenchmarking.org Megahertz, More Is Better simdjson 3.10 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.6 57.7 60.2 acpi-cpufreq performance 19.8 57.5 59.8 acpi-cpufreq schedutil 19.1 56.9 60.0 amd-pstate-epp powersave, EPP power 22.0 55.9 58.6 amd-pstate-epp powersave, EPP balance_power 19.5 55.9 58.7 amd-pstate-epp powersave, EPP performance 18.5 55.7 59.2 amd-pstate-epp powersave, EPP balance_performance 19.6 55.0 58.3 OpenBenchmarking.org Watts, Fewer Is Better simdjson 3.10 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.0 67.8 71.6 acpi-cpufreq performance 46.3 65.4 72.5 acpi-cpufreq schedutil 45.8 65.3 71.1 amd-pstate-epp powersave, EPP power 47.8 64.4 67.5 amd-pstate-epp powersave, EPP performance 46.1 64.1 67.1 amd-pstate-epp powersave, EPP balance_performance 46.6 62.7 67.1 amd-pstate-epp powersave, EPP balance_power 46.4 62.6 65.9 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 3.10 CPU Temperature Monitor 20 40 60 80 100
VVenC VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Fast acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 3 6 9 12 15 SE +/- 0.092, N = 6 SE +/- 0.016, N = 3 SE +/- 0.085, N = 7 SE +/- 0.027, N = 3 SE +/- 0.024, N = 3 SE +/- 0.029, N = 3 SE +/- 0.031, N = 3 9.439 9.507 9.526 9.542 9.574 9.577 9.597 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Fast amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0144 0.0288 0.0432 0.0576 0.072 0.064 0.064 0.064 0.064 0.064 0.064 0.064 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp performance, EPP performance 600 4871 5717 amd-pstate-epp powersave, EPP performance 600 4904 5728 amd-pstate-epp powersave, EPP power 600 4916 5609 amd-pstate-epp powersave, EPP balance_performance 600 4917 5660 amd-pstate-epp powersave, EPP balance_power 600 4961 5644 acpi-cpufreq performance 4300 4992 5590 acpi-cpufreq schedutil 4300 5016 5646 OpenBenchmarking.org Megahertz, More Is Better VVenC 1.11 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.9 149.8 197.7 amd-pstate-epp powersave, EPP performance 19.4 149.2 196.5 amd-pstate-epp powersave, EPP balance_power 20.1 149.0 195.8 amd-pstate-epp powersave, EPP power 22.5 149.0 197.6 amd-pstate-epp powersave, EPP balance_performance 18.8 148.5 198.0 acpi-cpufreq performance 19.7 148.4 198.0 acpi-cpufreq schedutil 9.8 148.1 197.3 OpenBenchmarking.org Watts, Fewer Is Better VVenC 1.11 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max acpi-cpufreq schedutil 41.4 77.0 80.3 amd-pstate-epp powersave, EPP balance_performance 41.6 76.7 80.1 amd-pstate-epp performance, EPP performance 42.4 76.7 80.6 amd-pstate-epp powersave, EPP power 42.8 76.7 80.8 acpi-cpufreq performance 41.8 76.1 80.0 amd-pstate-epp powersave, EPP performance 41.5 75.9 79.5 amd-pstate-epp powersave, EPP balance_power 41.4 75.3 79.3 OpenBenchmarking.org Celsius, Fewer Is Better VVenC 1.11 CPU Temperature Monitor 20 40 60 80 100
ONNX Runtime Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 30 60 90 120 150 SE +/- 0.44, N = 3 SE +/- 0.62, N = 3 SE +/- 0.81, N = 3 SE +/- 1.00, N = 9 SE +/- 0.79, N = 3 SE +/- 1.16, N = 3 SE +/- 1.45, N = 4 129.75 126.90 126.69 126.56 126.05 126.03 125.61 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5125 5481 amd-pstate-epp powersave, EPP performance 600 5145 5593 amd-pstate-epp powersave, EPP power 600 5189 5651 amd-pstate-epp performance, EPP performance 600 5198 5668 amd-pstate-epp powersave, EPP balance_performance 600 5206 5484 acpi-cpufreq performance 4300 5281 5570 acpi-cpufreq schedutil 4300 5296 5585 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq schedutil 18.4 163.8 179.6 amd-pstate-epp powersave, EPP balance_power 20.1 163.3 181.1 acpi-cpufreq performance 19.3 163.3 178.7 amd-pstate-epp powersave, EPP performance 18.5 162.8 179.0 amd-pstate-epp powersave, EPP power 22.3 162.6 178.8 amd-pstate-epp performance, EPP performance 21.0 162.1 178.4 amd-pstate-epp powersave, EPP balance_performance 19.5 161.2 177.5 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 45.1 73.5 76.4 amd-pstate-epp performance, EPP performance 46.0 73.1 76.4 amd-pstate-epp powersave, EPP balance_performance 45.1 72.5 75.5 acpi-cpufreq performance 44.6 71.8 74.9 acpi-cpufreq schedutil 43.9 71.6 74.8 amd-pstate-epp powersave, EPP balance_power 44.1 71.6 74.4 amd-pstate-epp powersave, EPP performance 44.0 71.5 74.6 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 2 4 6 8 10 SE +/- 0.02586, N = 3 SE +/- 0.03895, N = 3 SE +/- 0.05048, N = 3 SE +/- 0.06488, N = 9 SE +/- 0.05006, N = 3 SE +/- 0.07359, N = 3 SE +/- 0.09223, N = 4 7.70720 7.88058 7.89388 7.90547 7.93391 7.93590 7.96401 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 30 60 90 120 150 SE +/- 0.44, N = 3 SE +/- 0.62, N = 3 SE +/- 0.81, N = 3 SE +/- 1.00, N = 9 SE +/- 0.79, N = 3 SE +/- 1.16, N = 3 SE +/- 1.45, N = 4 129.75 126.90 126.69 126.56 126.05 126.03 125.61 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 44100 - Buffer Size: 1024 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil acpi-cpufreq performance 3 6 9 12 15 SE +/- 0.103725, N = 15 SE +/- 0.087409, N = 15 SE +/- 0.082682, N = 15 SE +/- 0.076630, N = 15 SE +/- 0.109263, N = 15 SE +/- 0.121780, N = 3 SE +/- 0.045428, N = 3 7.882371 8.338883 8.383640 8.477485 8.593859 9.250160 9.332352 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 44100 - Buffer Size: 1024 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil acpi-cpufreq performance 0.0137 0.0274 0.0411 0.0548 0.0685 0.050 0.054 0.054 0.054 0.055 0.060 0.061 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4658 5716 amd-pstate-epp powersave, EPP power 600 4658 5713 amd-pstate-epp powersave, EPP performance 600 4700 5683 amd-pstate-epp powersave, EPP balance_performance 600 4800 5486 amd-pstate-epp performance, EPP performance 600 4850 5701 acpi-cpufreq schedutil 4300 4889 5585 acpi-cpufreq performance 4300 4894 5636 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP performance 18.9 156.3 200.4 amd-pstate-epp performance, EPP performance 19.8 156.2 200.5 amd-pstate-epp powersave, EPP balance_performance 19.0 156.1 200.4 amd-pstate-epp powersave, EPP balance_power 18.8 155.5 200.5 amd-pstate-epp powersave, EPP power 22.2 154.8 200.4 acpi-cpufreq performance 19.6 153.1 200.2 acpi-cpufreq schedutil 18.5 153.1 200.4 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 45.1 85.2 96.1 amd-pstate-epp performance, EPP performance 47.0 84.6 95.8 amd-pstate-epp powersave, EPP performance 45.5 83.0 94.8 amd-pstate-epp powersave, EPP power 45.9 82.7 95.6 amd-pstate-epp powersave, EPP balance_power 44.8 82.5 95.5 acpi-cpufreq schedutil 44.9 76.5 90.6 acpi-cpufreq performance 45.4 75.1 82.1 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100
Selenium Result
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance 70 140 210 280 350 SE +/- 0.33, N = 3 SE +/- 1.09, N = 3 SE +/- 1.27, N = 3 SE +/- 2.29, N = 3 SE +/- 2.08, N = 3 SE +/- 1.30, N = 3 SE +/- 2.15, N = 3 290.43 295.40 296.74 298.31 299.47 302.84 302.86 1. firefox 126.0.1 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox acpi-cpufreq performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 1.2002 2.4004 3.6006 4.8008 6.001 5.003 5.042 5.083 5.285 5.288 5.331 5.334 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 5179 5728 amd-pstate-epp powersave, EPP balance_performance 600 5251 5731 amd-pstate-epp powersave, EPP balance_power 600 5271 5727 acpi-cpufreq schedutil 3000 5355 5703 amd-pstate-epp powersave, EPP performance 600 5390 5749 acpi-cpufreq performance 4300 5448 5729 amd-pstate-epp performance, EPP performance 600 5496 5714 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.4 60.1 154.7 acpi-cpufreq performance 20.4 59.0 153.4 acpi-cpufreq schedutil 19.3 57.1 153.9 amd-pstate-epp powersave, EPP performance 19.2 56.8 154.2 amd-pstate-epp powersave, EPP power 22.3 56.7 155.4 amd-pstate-epp powersave, EPP balance_power 19.8 56.1 155.1 amd-pstate-epp powersave, EPP balance_performance 19.7 55.9 153.4 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 42.4 65.4 73.6 acpi-cpufreq performance 41.5 63.9 71.6 amd-pstate-epp powersave, EPP balance_performance 42.3 63.5 71.0 acpi-cpufreq schedutil 41.0 63.2 73.0 amd-pstate-epp powersave, EPP power 41.8 62.2 70.5 amd-pstate-epp powersave, EPP balance_power 40.9 61.9 69.3 amd-pstate-epp powersave, EPP performance 40.6 61.5 69.8 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100
ONNX Runtime Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance 2 4 6 8 10 SE +/- 0.07285, N = 3 SE +/- 0.06117, N = 6 SE +/- 0.07213, N = 4 SE +/- 0.08712, N = 3 SE +/- 0.06902, N = 4 SE +/- 0.07092, N = 4 SE +/- 0.02858, N = 3 6.30332 6.25902 6.25566 6.25429 6.25270 6.20910 6.11717 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 5289 5588 amd-pstate-epp powersave, EPP performance 600 5301 5721 amd-pstate-epp powersave, EPP balance_performance 600 5305 5697 amd-pstate-epp powersave, EPP balance_power 600 5351 5675 amd-pstate-epp performance, EPP performance 600 5420 5686 acpi-cpufreq schedutil 4300 5452 5685 acpi-cpufreq performance 4300 5459 5724 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_power 20.0 115.7 125.3 amd-pstate-epp powersave, EPP performance 19.8 115.6 123.8 acpi-cpufreq performance 19.4 115.4 124.2 amd-pstate-epp performance, EPP performance 19.3 115.1 123.4 amd-pstate-epp powersave, EPP power 22.2 115.0 123.5 acpi-cpufreq schedutil 18.7 114.9 124.3 amd-pstate-epp powersave, EPP balance_performance 18.7 114.5 123.3 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.3 66.2 67.8 amd-pstate-epp powersave, EPP balance_performance 46.4 65.1 66.5 amd-pstate-epp powersave, EPP power 46.5 65.0 66.3 acpi-cpufreq performance 45.9 64.8 66.3 amd-pstate-epp powersave, EPP performance 45.3 64.3 65.6 acpi-cpufreq schedutil 45.5 64.3 65.5 amd-pstate-epp powersave, EPP balance_power 45.5 64.2 65.8 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance 40 80 120 160 200 SE +/- 1.85, N = 3 SE +/- 1.59, N = 6 SE +/- 1.88, N = 4 SE +/- 2.26, N = 3 SE +/- 1.80, N = 4 SE +/- 1.84, N = 4 SE +/- 0.76, N = 3 158.64 159.80 159.88 159.91 159.95 161.08 163.44 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance 2 4 6 8 10 SE +/- 0.07285, N = 3 SE +/- 0.06117, N = 6 SE +/- 0.07213, N = 4 SE +/- 0.08712, N = 3 SE +/- 0.06902, N = 4 SE +/- 0.07092, N = 4 SE +/- 0.02858, N = 3 6.30332 6.25902 6.25566 6.25429 6.25270 6.20910 6.11717 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
simdjson Result
OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: LargeRandom amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power 0.4793 0.9586 1.4379 1.9172 2.3965 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 2.07 2.09 2.10 2.11 2.12 2.12 2.13 1. (CXX) g++ options: -O3 -lrt GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 3.10 Throughput Test: LargeRandom amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance 0.0092 0.0184 0.0276 0.0368 0.046 0.037 0.038 0.039 0.039 0.040 0.040 0.041 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5313 5701 amd-pstate-epp powersave, EPP performance 600 5314 5732 amd-pstate-epp powersave, EPP balance_power 600 5315 5679 amd-pstate-epp performance, EPP performance 600 5363 5719 amd-pstate-epp powersave, EPP power 600 5373 5712 acpi-cpufreq schedutil 3027 5428 5668 acpi-cpufreq performance 4300 5461 5713 OpenBenchmarking.org Megahertz, More Is Better simdjson 3.10 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.9 56.4 59.2 acpi-cpufreq performance 20.0 54.9 58.5 amd-pstate-epp powersave, EPP power 22.6 54.0 57.6 acpi-cpufreq schedutil 19.2 53.6 57.7 amd-pstate-epp powersave, EPP balance_power 19.7 52.4 58.1 amd-pstate-epp powersave, EPP performance 19.6 52.3 57.4 amd-pstate-epp powersave, EPP balance_performance 18.8 52.3 57.2 OpenBenchmarking.org Watts, Fewer Is Better simdjson 3.10 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.6 61.9 62.8 acpi-cpufreq performance 39.9 61.3 65.6 amd-pstate-epp powersave, EPP power 41.5 61.3 64.4 acpi-cpufreq schedutil 39.4 60.0 65.5 amd-pstate-epp powersave, EPP balance_performance 39.9 58.4 63.9 amd-pstate-epp powersave, EPP performance 39.9 57.9 63.9 amd-pstate-epp powersave, EPP balance_power 39.8 57.2 63.6 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 3.10 CPU Temperature Monitor 20 40 60 80 100
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 512 amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance acpi-cpufreq performance 0.9018 1.8036 2.7054 3.6072 4.509 SE +/- 0.021926, N = 3 SE +/- 0.032949, N = 15 SE +/- 0.039946, N = 3 SE +/- 0.050910, N = 3 SE +/- 0.052034, N = 3 SE +/- 0.036686, N = 3 SE +/- 0.033624, N = 3 3.803383 3.824622 3.828578 3.852325 3.932036 3.950129 4.008207 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 512 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0056 0.0112 0.0168 0.0224 0.028 0.024 0.024 0.024 0.024 0.024 0.025 0.025 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4936 5629 amd-pstate-epp powersave, EPP performance 600 5022 5694 amd-pstate-epp powersave, EPP power 600 5046 5712 amd-pstate-epp powersave, EPP balance_performance 600 5050 5722 acpi-cpufreq schedutil 4300 5144 5454 acpi-cpufreq performance 4300 5149 5596 amd-pstate-epp performance, EPP performance 600 5149 5689 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.7 161.6 200.3 amd-pstate-epp performance, EPP performance 21.4 161.4 200.3 amd-pstate-epp powersave, EPP balance_performance 10.0 161.2 200.3 amd-pstate-epp powersave, EPP power 22.0 160.8 200.3 amd-pstate-epp powersave, EPP performance 19.5 160.1 200.4 acpi-cpufreq schedutil 18.4 159.6 200.3 amd-pstate-epp powersave, EPP balance_power 20.0 159.5 200.3 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 45.0 81.7 95.3 amd-pstate-epp performance, EPP performance 46.8 81.4 94.5 amd-pstate-epp powersave, EPP power 46.0 80.8 93.1 amd-pstate-epp powersave, EPP performance 45.3 79.9 93.6 amd-pstate-epp powersave, EPP balance_power 44.3 78.8 91.4 acpi-cpufreq performance 44.9 77.8 89.1 acpi-cpufreq schedutil 45.4 77.7 90.4 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100
ONNX Runtime Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Standard amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power 1.2238 2.4476 3.6714 4.8952 6.119 SE +/- 0.00681, N = 3 SE +/- 0.05843, N = 5 SE +/- 0.05210, N = 6 SE +/- 0.00134, N = 3 SE +/- 0.01641, N = 3 SE +/- 0.01959, N = 3 SE +/- 0.01311, N = 3 5.43900 5.20789 5.19651 5.17267 5.16644 5.15508 5.14062 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5275 5604 amd-pstate-epp powersave, EPP balance_power 600 5279 5571 amd-pstate-epp powersave, EPP power 600 5308 5582 amd-pstate-epp performance, EPP performance 600 5380 5688 amd-pstate-epp powersave, EPP performance 600 5404 5680 acpi-cpufreq schedutil 3000 5456 5615 acpi-cpufreq performance 4300 5471 5676 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 20.1 106.9 113.9 amd-pstate-epp powersave, EPP performance 19.7 106.4 113.4 acpi-cpufreq schedutil 19.5 106.4 114.2 amd-pstate-epp performance, EPP performance 20.8 106.0 113.0 amd-pstate-epp powersave, EPP power 22.4 106.0 114.5 amd-pstate-epp powersave, EPP balance_power 19.1 106.0 113.7 amd-pstate-epp powersave, EPP balance_performance 14.9 105.2 113.2 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.0 63.1 64.3 amd-pstate-epp powersave, EPP balance_performance 46.4 62.4 63.5 acpi-cpufreq performance 46.1 62.2 63.3 amd-pstate-epp powersave, EPP power 46.0 61.8 63.1 acpi-cpufreq schedutil 45.1 61.5 62.8 amd-pstate-epp powersave, EPP performance 44.9 61.2 62.4 amd-pstate-epp powersave, EPP balance_power 45.3 61.1 62.4 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Standard amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power 40 80 120 160 200 SE +/- 0.23, N = 3 SE +/- 2.09, N = 5 SE +/- 1.86, N = 6 SE +/- 0.05, N = 3 SE +/- 0.61, N = 3 SE +/- 0.73, N = 3 SE +/- 0.50, N = 3 183.84 192.09 192.50 193.30 193.54 193.97 194.51 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Standard amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power 1.2238 2.4476 3.6714 4.8952 6.119 SE +/- 0.00681, N = 3 SE +/- 0.05843, N = 5 SE +/- 0.05210, N = 6 SE +/- 0.00134, N = 3 SE +/- 0.01641, N = 3 SE +/- 0.01959, N = 3 SE +/- 0.01311, N = 3 5.43900 5.20789 5.19651 5.17267 5.16644 5.15508 5.14062 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 1024 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil 0.9932 1.9864 2.9796 3.9728 4.966 SE +/- 0.034707, N = 15 SE +/- 0.043380, N = 3 SE +/- 0.028124, N = 3 SE +/- 0.038105, N = 3 SE +/- 0.046132, N = 3 SE +/- 0.008849, N = 3 SE +/- 0.008998, N = 3 4.144614 4.157566 4.192825 4.263705 4.305339 4.396724 4.414173 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 1024 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0063 0.0126 0.0189 0.0252 0.0315 0.026 0.026 0.026 0.027 0.027 0.028 0.028 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4955 5682 amd-pstate-epp powersave, EPP balance_performance 600 4973 5713 amd-pstate-epp powersave, EPP performance 600 4975 5710 amd-pstate-epp powersave, EPP power 600 4984 5720 acpi-cpufreq schedutil 4300 5069 5601 acpi-cpufreq performance 4300 5081 5633 amd-pstate-epp performance, EPP performance 600 5086 5696 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 11.1 158.8 200.4 amd-pstate-epp performance, EPP performance 21.5 158.7 200.3 amd-pstate-epp powersave, EPP performance 18.6 158.2 200.3 acpi-cpufreq performance 19.0 158.2 200.3 amd-pstate-epp powersave, EPP balance_power 19.9 157.9 200.4 amd-pstate-epp powersave, EPP power 22.5 157.6 200.4 acpi-cpufreq schedutil 18.5 157.0 200.3 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 44.9 80.4 93.6 amd-pstate-epp performance, EPP performance 47.5 80.3 90.9 amd-pstate-epp powersave, EPP power 45.8 80.0 93.1 amd-pstate-epp powersave, EPP performance 44.5 78.5 91.9 amd-pstate-epp powersave, EPP balance_power 44.5 76.8 90.1 acpi-cpufreq performance 44.5 76.5 86.6 acpi-cpufreq schedutil 44.5 75.9 85.9 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100
simdjson Result
OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: TopTweet amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power 4 8 12 16 20 SE +/- 0.15, N = 3 SE +/- 0.09, N = 3 SE +/- 0.19, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 12.97 13.13 13.46 13.55 13.56 13.64 13.66 1. (CXX) g++ options: -O3 -lrt GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 3.10 Throughput Test: TopTweet amd-pstate-epp performance, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power 0.0574 0.1148 0.1722 0.2296 0.287 0.229 0.235 0.248 0.249 0.252 0.254 0.255 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5275 5731 amd-pstate-epp powersave, EPP power 600 5312 5732 amd-pstate-epp performance, EPP performance 600 5369 5713 amd-pstate-epp powersave, EPP performance 600 5393 5660 acpi-cpufreq performance 4300 5431 5650 amd-pstate-epp powersave, EPP balance_performance 600 5435 5657 acpi-cpufreq schedutil 3000 5438 5664 OpenBenchmarking.org Megahertz, More Is Better simdjson 3.10 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.5 56.6 59.5 acpi-cpufreq performance 19.3 55.8 59.4 acpi-cpufreq schedutil 19.3 54.5 58.6 amd-pstate-epp powersave, EPP balance_performance 19.4 54.5 57.9 amd-pstate-epp powersave, EPP power 22.1 53.9 58.4 amd-pstate-epp powersave, EPP balance_power 19.7 53.4 59.2 amd-pstate-epp powersave, EPP performance 19.8 53.4 59.1 OpenBenchmarking.org Watts, Fewer Is Better simdjson 3.10 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 42.4 63.5 66.0 amd-pstate-epp powersave, EPP balance_performance 40.9 62.7 65.1 acpi-cpufreq performance 40.8 62.2 65.1 acpi-cpufreq schedutil 40.4 61.9 67.3 amd-pstate-epp powersave, EPP power 41.9 61.2 65.3 amd-pstate-epp powersave, EPP performance 41.0 60.4 65.3 amd-pstate-epp powersave, EPP balance_power 40.6 58.7 65.0 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 3.10 CPU Temperature Monitor 20 40 60 80 100
Chaos Group V-RAY This is a test of Chaos Group's V-RAY benchmark. V-RAY is a commercial renderer that can integrate with various creator software products like SketchUp and 3ds Max. The V-RAY benchmark is standalone and supports CPU and NVIDIA CUDA/RTX based rendering. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org vsamples, More Is Better Chaos Group V-RAY 6.0 Mode: CPU amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance 12K 24K 36K 48K 60K SE +/- 129.51, N = 3 SE +/- 9.53, N = 3 SE +/- 28.58, N = 3 SE +/- 11.33, N = 3 SE +/- 9.81, N = 3 SE +/- 22.33, N = 3 SE +/- 0.33, N = 3 54799 55008 55057 55068 55074 55079 55091 vsamples Per Watt
OpenBenchmarking.org vsamples Per Watt, More Is Better Chaos Group V-RAY 6.0 Mode: CPU acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance 70 140 210 280 350 319.27 319.69 319.95 320.27 320.36 320.49 321.14 Peak Freq (Highest Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4778 5725 amd-pstate-epp powersave, EPP balance_power 600 4785 5728 amd-pstate-epp powersave, EPP power 600 4786 5671 amd-pstate-epp powersave, EPP performance 600 4859 5726 amd-pstate-epp performance, EPP performance 600 4887 5691 acpi-cpufreq performance 4300 4953 5684 acpi-cpufreq schedutil 4300 4955 5718 OpenBenchmarking.org Megahertz, More Is Better Chaos Group V-RAY 6.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 Power Consumption
Min Avg Max acpi-cpufreq performance 20.0 172.5 201.4 amd-pstate-epp performance, EPP performance 19.9 172.1 201.2 acpi-cpufreq schedutil 19.1 172.1 201.6 amd-pstate-epp powersave, EPP performance 19.6 172.0 201.0 amd-pstate-epp powersave, EPP balance_power 19.0 171.8 201.0 amd-pstate-epp powersave, EPP balance_performance 19.7 171.5 201.6 amd-pstate-epp powersave, EPP power 22.4 171.1 201.0 OpenBenchmarking.org Watts, Fewer Is Better Chaos Group V-RAY 6.0 CPU Power Consumption Monitor 50 100 150 200 250 Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 48.0 77.1 80.9 amd-pstate-epp performance, EPP performance 47.8 76.7 80.5 acpi-cpufreq performance 47.3 76.2 79.8 amd-pstate-epp powersave, EPP balance_power 47.0 76.1 80.3 amd-pstate-epp powersave, EPP performance 46.1 75.9 80.0 acpi-cpufreq schedutil 46.8 75.9 79.6 amd-pstate-epp powersave, EPP power 46.6 75.7 80.0 OpenBenchmarking.org Celsius, Fewer Is Better Chaos Group V-RAY 6.0 CPU Temperature Monitor 20 40 60 80 100
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 512 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 2 4 6 8 10 SE +/- 0.068118, N = 3 SE +/- 0.045988, N = 15 SE +/- 0.060979, N = 3 SE +/- 0.048475, N = 15 SE +/- 0.053123, N = 3 SE +/- 0.014060, N = 3 SE +/- 0.017666, N = 3 5.678037 5.719418 5.836306 5.857247 5.893716 6.120461 6.132388 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 512 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 0.0088 0.0176 0.0264 0.0352 0.044 0.035 0.036 0.036 0.037 0.037 0.038 0.039 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4853 5707 amd-pstate-epp powersave, EPP power 600 4873 5697 amd-pstate-epp powersave, EPP performance 600 4899 5730 amd-pstate-epp powersave, EPP balance_power 600 4950 5719 acpi-cpufreq schedutil 3000 5039 5554 acpi-cpufreq performance 4300 5057 5596 amd-pstate-epp performance, EPP performance 600 5058 5691 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 19.7 160.3 200.3 amd-pstate-epp performance, EPP performance 20.4 160.2 200.4 amd-pstate-epp powersave, EPP power 22.5 160.1 200.4 amd-pstate-epp powersave, EPP performance 19.5 159.4 200.4 acpi-cpufreq performance 19.8 159.4 200.4 amd-pstate-epp powersave, EPP balance_power 20.2 158.6 200.5 acpi-cpufreq schedutil 18.8 157.3 200.3 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.3 83.1 95.8 amd-pstate-epp powersave, EPP performance 45.3 81.8 93.0 amd-pstate-epp powersave, EPP balance_performance 45.5 81.1 91.4 amd-pstate-epp powersave, EPP power 45.9 80.9 90.8 amd-pstate-epp powersave, EPP balance_power 44.6 78.2 91.6 acpi-cpufreq performance 45.0 77.2 88.4 acpi-cpufreq schedutil 44.8 76.4 86.0 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 75, Compression Effort 7 amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil 0.0833 0.1666 0.2499 0.3332 0.4165 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.36 0.36 0.36 0.37 0.37 0.37 0.37 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -lm -ldl MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 75, Compression Effort 7 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0005 0.001 0.0015 0.002 0.0025 0.002 0.002 0.002 0.002 0.002 0.002 0.002 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4595 5564 amd-pstate-epp powersave, EPP balance_performance 600 4610 5658 amd-pstate-epp powersave, EPP power 600 4612 5572 amd-pstate-epp powersave, EPP performance 600 4656 5555 amd-pstate-epp performance, EPP performance 600 4670 5724 acpi-cpufreq schedutil 3000 4731 5639 acpi-cpufreq performance 4300 4749 5700 OpenBenchmarking.org Megahertz, More Is Better WebP2 Image Encode 20220823 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP performance 20.0 181.1 201.4 amd-pstate-epp performance, EPP performance 20.5 181.1 201.2 acpi-cpufreq schedutil 19.0 181.1 201.6 amd-pstate-epp powersave, EPP balance_performance 18.9 179.5 201.4 amd-pstate-epp powersave, EPP balance_power 19.7 179.2 201.2 amd-pstate-epp powersave, EPP power 22.6 179.2 201.2 acpi-cpufreq performance 19.7 179.0 201.8 OpenBenchmarking.org Watts, Fewer Is Better WebP2 Image Encode 20220823 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.9 79.4 92.5 amd-pstate-epp powersave, EPP power 41.6 78.8 93.1 amd-pstate-epp powersave, EPP performance 40.3 78.1 91.5 amd-pstate-epp powersave, EPP balance_performance 40.3 77.8 90.9 amd-pstate-epp powersave, EPP balance_power 39.9 77.4 92.0 acpi-cpufreq schedutil 39.4 77.1 90.4 acpi-cpufreq performance 40.3 76.9 90.3 OpenBenchmarking.org Celsius, Fewer Is Better WebP2 Image Encode 20220823 CPU Temperature Monitor 20 40 60 80 100
Blender Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Junkshop - Compute: CPU-Only amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil 14 28 42 56 70 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.13, N = 3 SE +/- 0.08, N = 3 63.16 62.73 62.61 62.53 62.43 62.37 62.30 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4680 5485 amd-pstate-epp powersave, EPP performance 600 4733 5648 amd-pstate-epp powersave, EPP balance_power 600 4736 5519 amd-pstate-epp powersave, EPP power 600 4750 5498 acpi-cpufreq schedutil 4300 4863 5622 amd-pstate-epp performance, EPP performance 600 4875 5689 acpi-cpufreq performance 4300 4876 5638 OpenBenchmarking.org Megahertz, More Is Better Blender 4.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP power 22.9 186.9 201.2 acpi-cpufreq schedutil 19.1 186.7 201.2 amd-pstate-epp powersave, EPP performance 19.7 186.7 201.0 acpi-cpufreq performance 20.0 186.6 201.2 amd-pstate-epp performance, EPP performance 21.5 186.4 201.0 amd-pstate-epp powersave, EPP balance_performance 20.2 186.1 201.2 amd-pstate-epp powersave, EPP balance_power 20.6 185.8 201.0 OpenBenchmarking.org Watts, Fewer Is Better Blender 4.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.1 78.6 81.1 amd-pstate-epp powersave, EPP power 46.9 77.5 79.9 amd-pstate-epp powersave, EPP balance_performance 46.3 77.5 80.0 amd-pstate-epp powersave, EPP performance 45.8 76.9 79.5 acpi-cpufreq schedutil 45.8 76.7 79.0 acpi-cpufreq performance 46.5 76.7 80.0 amd-pstate-epp powersave, EPP balance_power 45.8 76.5 79.0 OpenBenchmarking.org Celsius, Fewer Is Better Blender 4.2 CPU Temperature Monitor 20 40 60 80 100
ONNX Runtime Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 2 4 6 8 10 SE +/- 0.07495, N = 4 SE +/- 0.03200, N = 3 SE +/- 0.06647, N = 3 SE +/- 0.03821, N = 3 SE +/- 0.03293, N = 3 SE +/- 0.04038, N = 3 SE +/- 0.03830, N = 3 6.49949 6.49163 6.49100 6.48991 6.47352 6.46604 6.36926 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5113 5468 amd-pstate-epp powersave, EPP power 600 5117 5510 amd-pstate-epp performance, EPP performance 600 5138 5683 amd-pstate-epp powersave, EPP balance_performance 600 5146 5528 amd-pstate-epp powersave, EPP performance 600 5233 5480 acpi-cpufreq performance 4300 5291 5498 acpi-cpufreq schedutil 4300 5306 5503 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP power 22.3 154.3 175.0 amd-pstate-epp powersave, EPP balance_power 20.3 153.4 175.4 amd-pstate-epp performance, EPP performance 20.8 153.2 174.7 acpi-cpufreq schedutil 16.4 152.2 173.8 amd-pstate-epp powersave, EPP performance 19.4 152.2 172.8 amd-pstate-epp powersave, EPP balance_performance 18.3 151.5 172.2 acpi-cpufreq performance 17.8 151.3 173.5 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.4 72.3 75.5 amd-pstate-epp powersave, EPP balance_performance 48.1 71.9 74.5 amd-pstate-epp powersave, EPP power 47.1 71.1 73.5 amd-pstate-epp powersave, EPP performance 46.3 70.6 73.3 amd-pstate-epp powersave, EPP balance_power 46.4 70.4 73.0 acpi-cpufreq schedutil 46.8 70.2 72.6 acpi-cpufreq performance 47.3 70.1 72.8 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 30 60 90 120 150 SE +/- 1.74, N = 4 SE +/- 0.76, N = 3 SE +/- 1.56, N = 3 SE +/- 0.90, N = 3 SE +/- 0.79, N = 3 SE +/- 0.96, N = 3 SE +/- 0.94, N = 3 153.90 154.02 154.07 154.07 154.46 154.64 156.99 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 2 4 6 8 10 SE +/- 0.07495, N = 4 SE +/- 0.03200, N = 3 SE +/- 0.06647, N = 3 SE +/- 0.03821, N = 3 SE +/- 0.03293, N = 3 SE +/- 0.04038, N = 3 SE +/- 0.03830, N = 3 6.49949 6.49163 6.49100 6.48991 6.47352 6.46604 6.36926 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 13 - Input: Beauty 4K 10-bit acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 18.76 19.37 19.50 19.52 19.54 19.54 19.58 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 13 - Input: Beauty 4K 10-bit acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance 0.038 0.076 0.114 0.152 0.19 0.162 0.167 0.168 0.168 0.168 0.168 0.169 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5224 5673 amd-pstate-epp powersave, EPP power 3019 5301 5672 amd-pstate-epp powersave, EPP performance 600 5324 5709 amd-pstate-epp powersave, EPP balance_performance 600 5338 5719 acpi-cpufreq performance 4300 5364 5628 amd-pstate-epp performance, EPP performance 600 5373 5710 acpi-cpufreq schedutil 4300 5388 5610 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.7 116.6 152.2 amd-pstate-epp powersave, EPP balance_power 19.8 116.3 153.3 acpi-cpufreq performance 19.5 116.2 155.8 amd-pstate-epp powersave, EPP power 22.5 115.9 152.5 acpi-cpufreq schedutil 17.4 115.9 153.5 amd-pstate-epp powersave, EPP performance 19.5 115.9 153.0 amd-pstate-epp powersave, EPP balance_performance 19.7 115.8 154.4 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.2 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 45.9 76.8 82.0 amd-pstate-epp performance, EPP performance 45.0 76.3 81.5 amd-pstate-epp powersave, EPP performance 44.4 76.1 81.1 amd-pstate-epp powersave, EPP balance_performance 44.3 75.9 80.9 amd-pstate-epp powersave, EPP balance_power 43.5 75.5 80.6 acpi-cpufreq schedutil 43.9 74.2 81.0 acpi-cpufreq performance 44.3 73.6 79.5 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 2.2 CPU Temperature Monitor 20 40 60 80 100
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil 1.062 2.124 3.186 4.248 5.31 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 4.63 4.63 4.65 4.66 4.66 4.71 4.72 MIN: 2.1 / MAX: 5.24 MIN: 2.05 / MAX: 5.23 MIN: 2.07 / MAX: 5.26 MIN: 2.07 / MAX: 5.27 MIN: 2.08 / MAX: 5.28 MIN: 2.08 / MAX: 5.31 MIN: 2.17 / MAX: 5.29 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0056 0.0112 0.0168 0.0224 0.028 0.025 0.025 0.025 0.025 0.025 0.025 0.025 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4773 5495 amd-pstate-epp powersave, EPP power 600 4795 5568 amd-pstate-epp powersave, EPP balance_power 600 4857 5478 amd-pstate-epp powersave, EPP performance 600 4875 5667 amd-pstate-epp performance, EPP performance 600 4880 5711 acpi-cpufreq schedutil 3000 4917 5575 acpi-cpufreq performance 4300 4955 5579 OpenBenchmarking.org Megahertz, More Is Better LuxCoreRender 2.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.9 187.3 201.2 amd-pstate-epp powersave, EPP performance 16.0 186.4 204.4 acpi-cpufreq performance 19.7 186.4 201.7 acpi-cpufreq schedutil 18.8 186.4 201.4 amd-pstate-epp powersave, EPP balance_performance 19.6 186.3 201.0 amd-pstate-epp powersave, EPP power 15.1 185.7 202.3 amd-pstate-epp powersave, EPP balance_power 19.9 185.6 201.2 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 45.6 79.0 83.0 amd-pstate-epp performance, EPP performance 45.3 78.8 81.9 amd-pstate-epp powersave, EPP performance 44.6 78.6 81.4 amd-pstate-epp powersave, EPP balance_performance 44.6 78.5 82.0 acpi-cpufreq performance 44.5 78.0 81.3 acpi-cpufreq schedutil 43.9 77.8 81.1 amd-pstate-epp powersave, EPP balance_power 44.3 77.6 80.8 OpenBenchmarking.org Celsius, Fewer Is Better LuxCoreRender 2.6 CPU Temperature Monitor 20 40 60 80 100
IndigoBench This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Bedroom acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance 1.3394 2.6788 4.0182 5.3576 6.697 SE +/- 0.015, N = 3 SE +/- 0.012, N = 3 SE +/- 0.009, N = 3 SE +/- 0.005, N = 3 SE +/- 0.005, N = 3 SE +/- 0.010, N = 3 SE +/- 0.000, N = 3 5.903 5.904 5.915 5.924 5.925 5.933 5.953 M samples/s Per Watt
OpenBenchmarking.org M samples/s Per Watt, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Bedroom amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0072 0.0144 0.0216 0.0288 0.036 0.032 0.032 0.032 0.032 0.032 0.032 0.032 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4695 5683 amd-pstate-epp powersave, EPP power 600 4743 5605 amd-pstate-epp powersave, EPP balance_performance 600 4787 5646 amd-pstate-epp powersave, EPP performance 600 4804 5679 amd-pstate-epp performance, EPP performance 600 4806 5679 acpi-cpufreq schedutil 4300 4906 5581 acpi-cpufreq performance 4300 4914 5704 OpenBenchmarking.org Megahertz, More Is Better IndigoBench 4.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.9 185.4 201.0 amd-pstate-epp performance, EPP performance 22.0 184.7 201.2 amd-pstate-epp powersave, EPP power 22.5 184.6 201.2 amd-pstate-epp powersave, EPP performance 19.9 184.5 201.0 acpi-cpufreq schedutil 18.6 184.3 201.0 amd-pstate-epp powersave, EPP balance_performance 19.8 184.0 201.0 amd-pstate-epp powersave, EPP balance_power 19.9 184.0 201.0 OpenBenchmarking.org Watts, Fewer Is Better IndigoBench 4.4 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.3 79.1 81.4 amd-pstate-epp powersave, EPP balance_performance 47.5 78.3 80.8 amd-pstate-epp powersave, EPP power 47.5 78.2 80.8 acpi-cpufreq performance 47.0 77.5 79.9 amd-pstate-epp powersave, EPP performance 46.8 77.5 80.1 acpi-cpufreq schedutil 46.8 77.3 79.6 amd-pstate-epp powersave, EPP balance_power 46.9 77.2 80.1 OpenBenchmarking.org Celsius, Fewer Is Better IndigoBench 4.4 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Supercar acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 13.57 13.59 13.61 13.62 13.63 13.64 13.67 M samples/s Per Watt
OpenBenchmarking.org M samples/s Per Watt, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Supercar amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0167 0.0334 0.0501 0.0668 0.0835 0.073 0.074 0.074 0.074 0.074 0.074 0.074 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4650 5557 amd-pstate-epp performance, EPP performance 600 4696 5688 amd-pstate-epp powersave, EPP balance_performance 600 4696 5613 amd-pstate-epp powersave, EPP power 600 4722 5671 amd-pstate-epp powersave, EPP performance 600 4727 5718 acpi-cpufreq performance 4300 4842 5683 acpi-cpufreq schedutil 4300 4844 5548 OpenBenchmarking.org Megahertz, More Is Better IndigoBench 4.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.7 185.3 201.1 amd-pstate-epp powersave, EPP power 22.1 184.4 201.2 amd-pstate-epp powersave, EPP performance 20.1 184.3 201.8 acpi-cpufreq performance 19.6 184.2 201.2 amd-pstate-epp powersave, EPP balance_power 20.2 184.2 201.0 amd-pstate-epp powersave, EPP balance_performance 20.0 184.1 201.6 acpi-cpufreq schedutil 18.5 184.0 201.2 OpenBenchmarking.org Watts, Fewer Is Better IndigoBench 4.4 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.4 78.9 81.4 amd-pstate-epp powersave, EPP balance_performance 46.8 78.3 80.8 amd-pstate-epp powersave, EPP power 46.9 78.1 80.6 amd-pstate-epp powersave, EPP performance 46.1 77.5 79.9 acpi-cpufreq performance 46.8 77.5 80.0 acpi-cpufreq schedutil 46.1 77.4 79.8 amd-pstate-epp powersave, EPP balance_power 45.9 77.1 79.5 OpenBenchmarking.org Celsius, Fewer Is Better IndigoBench 4.4 CPU Temperature Monitor 20 40 60 80 100
ONNX Runtime Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 150 300 450 600 750 SE +/- 2.30, N = 3 SE +/- 3.19, N = 3 SE +/- 3.73, N = 3 SE +/- 3.62, N = 3 SE +/- 3.64, N = 3 SE +/- 2.27, N = 3 SE +/- 1.12, N = 3 701.75 701.16 698.10 695.63 693.55 691.97 689.99 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4645 5304 amd-pstate-epp powersave, EPP balance_power 600 4647 5044 amd-pstate-epp powersave, EPP balance_performance 600 4677 5409 amd-pstate-epp powersave, EPP power 600 4679 5115 amd-pstate-epp performance, EPP performance 600 4732 5706 acpi-cpufreq performance 4300 4791 5527 acpi-cpufreq schedutil 4300 4794 5576 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq schedutil 19.2 148.8 164.1 amd-pstate-epp performance, EPP performance 21.2 148.6 165.1 acpi-cpufreq performance 19.7 148.3 163.3 amd-pstate-epp powersave, EPP power 22.4 148.0 166.9 amd-pstate-epp powersave, EPP balance_performance 19.5 147.6 164.8 amd-pstate-epp powersave, EPP balance_power 19.5 147.3 163.3 amd-pstate-epp powersave, EPP performance 19.1 147.0 162.6 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 45.9 70.4 73.4 amd-pstate-epp powersave, EPP balance_performance 45.8 70.3 74.1 amd-pstate-epp powersave, EPP power 44.9 69.1 72.6 amd-pstate-epp powersave, EPP performance 44.0 68.9 72.3 acpi-cpufreq schedutil 44.1 68.8 71.9 acpi-cpufreq performance 44.5 68.7 72.0 amd-pstate-epp powersave, EPP balance_power 43.8 68.2 71.8 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 0.3261 0.6522 0.9783 1.3044 1.6305 SE +/- 0.00467, N = 3 SE +/- 0.00646, N = 3 SE +/- 0.00764, N = 3 SE +/- 0.00751, N = 3 SE +/- 0.00759, N = 3 SE +/- 0.00473, N = 3 SE +/- 0.00236, N = 3 1.42504 1.42626 1.43253 1.43761 1.44193 1.44518 1.44931 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 150 300 450 600 750 SE +/- 2.30, N = 3 SE +/- 3.19, N = 3 SE +/- 3.73, N = 3 SE +/- 3.62, N = 3 SE +/- 3.64, N = 3 SE +/- 2.27, N = 3 SE +/- 1.12, N = 3 701.75 701.16 698.10 695.63 693.55 691.97 689.99 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 80 160 240 320 400 SE +/- 0.57, N = 3 SE +/- 0.68, N = 3 SE +/- 0.34, N = 3 SE +/- 0.53, N = 3 SE +/- 0.07, N = 3 SE +/- 0.63, N = 3 SE +/- 0.46, N = 3 379.35 379.12 379.01 377.54 377.03 376.41 375.82 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4398 4758 amd-pstate-epp powersave, EPP performance 600 4435 5723 amd-pstate-epp powersave, EPP balance_performance 600 4458 5624 amd-pstate-epp powersave, EPP power 600 4494 5445 amd-pstate-epp performance, EPP performance 600 4532 5683 acpi-cpufreq schedutil 3000 4547 4757 acpi-cpufreq performance 4300 4565 5451 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_power 19.8 183.1 197.0 amd-pstate-epp performance, EPP performance 20.0 182.7 197.2 amd-pstate-epp powersave, EPP performance 19.5 182.6 196.6 amd-pstate-epp powersave, EPP power 11.7 182.0 196.5 amd-pstate-epp powersave, EPP balance_performance 19.0 181.9 196.2 acpi-cpufreq schedutil 19.4 181.5 195.6 acpi-cpufreq performance 19.4 181.3 195.5 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 46.9 78.6 80.5 amd-pstate-epp performance, EPP performance 46.9 78.5 80.8 amd-pstate-epp powersave, EPP power 45.9 77.6 79.8 amd-pstate-epp powersave, EPP performance 44.9 77.4 79.6 acpi-cpufreq schedutil 45.4 77.1 79.5 amd-pstate-epp powersave, EPP balance_power 45.1 77.0 79.3 acpi-cpufreq performance 45.8 77.0 79.1 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 0.5987 1.1974 1.7961 2.3948 2.9935 SE +/- 0.00397, N = 3 SE +/- 0.00473, N = 3 SE +/- 0.00234, N = 3 SE +/- 0.00369, N = 3 SE +/- 0.00049, N = 3 SE +/- 0.00447, N = 3 SE +/- 0.00328, N = 3 2.63611 2.63771 2.63844 2.64873 2.65230 2.65667 2.66083 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 80 160 240 320 400 SE +/- 0.57, N = 3 SE +/- 0.68, N = 3 SE +/- 0.34, N = 3 SE +/- 0.53, N = 3 SE +/- 0.07, N = 3 SE +/- 0.63, N = 3 SE +/- 0.46, N = 3 379.35 379.12 379.01 377.54 377.03 376.41 375.82 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance acpi-cpufreq schedutil 100 200 300 400 500 SE +/- 1.96, N = 3 SE +/- 2.41, N = 3 SE +/- 3.24, N = 3 SE +/- 5.78, N = 3 SE +/- 5.71, N = 3 SE +/- 1.14, N = 3 SE +/- 1.63, N = 3 439.82 438.49 437.51 437.18 433.56 432.64 430.47 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4708 5268 amd-pstate-epp powersave, EPP power 600 4737 5341 amd-pstate-epp powersave, EPP balance_performance 600 4754 5342 amd-pstate-epp powersave, EPP performance 600 4808 5693 amd-pstate-epp performance, EPP performance 600 4862 5686 acpi-cpufreq schedutil 4287 4867 5588 acpi-cpufreq performance 4300 4904 5566 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 20.0 138.3 151.6 amd-pstate-epp powersave, EPP power 22.6 136.7 151.2 amd-pstate-epp powersave, EPP balance_power 19.5 136.7 150.4 amd-pstate-epp performance, EPP performance 20.9 136.0 149.8 amd-pstate-epp powersave, EPP performance 18.5 136.0 150.9 acpi-cpufreq schedutil 19.1 136.0 149.2 amd-pstate-epp powersave, EPP balance_performance 19.2 135.7 149.2 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.4 68.7 70.5 amd-pstate-epp powersave, EPP balance_performance 47.8 68.4 70.6 acpi-cpufreq performance 47.4 67.7 69.8 amd-pstate-epp powersave, EPP power 47.3 67.6 70.1 amd-pstate-epp powersave, EPP performance 46.3 67.2 69.4 acpi-cpufreq schedutil 46.5 67.1 69.3 amd-pstate-epp powersave, EPP balance_power 46.5 67.1 69.8 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance acpi-cpufreq schedutil 0.5227 1.0454 1.5681 2.0908 2.6135 SE +/- 0.01020, N = 3 SE +/- 0.01261, N = 3 SE +/- 0.01682, N = 3 SE +/- 0.03063, N = 3 SE +/- 0.03009, N = 3 SE +/- 0.00613, N = 3 SE +/- 0.00877, N = 3 2.27372 2.28069 2.28589 2.28817 2.30726 2.31140 2.32311 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance acpi-cpufreq schedutil 100 200 300 400 500 SE +/- 1.96, N = 3 SE +/- 2.41, N = 3 SE +/- 3.24, N = 3 SE +/- 5.78, N = 3 SE +/- 5.71, N = 3 SE +/- 1.14, N = 3 SE +/- 1.63, N = 3 439.82 438.49 437.51 437.18 433.56 432.64 430.47 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Parallel amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance 2 4 6 8 10 SE +/- 0.00837, N = 3 SE +/- 0.00609, N = 3 SE +/- 0.01316, N = 3 SE +/- 0.05580, N = 3 SE +/- 0.02213, N = 3 SE +/- 0.01392, N = 3 SE +/- 0.00989, N = 3 6.61050 6.60006 6.58721 6.55080 6.54688 6.54343 6.54213 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 5283 5609 amd-pstate-epp powersave, EPP balance_performance 600 5308 5556 amd-pstate-epp powersave, EPP performance 600 5316 5580 amd-pstate-epp performance, EPP performance 600 5318 5688 amd-pstate-epp powersave, EPP balance_power 600 5351 5583 acpi-cpufreq performance 4300 5412 5593 acpi-cpufreq schedutil 4300 5431 5697 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_power 20.1 139.0 149.1 acpi-cpufreq performance 18.1 138.5 149.3 acpi-cpufreq schedutil 18.4 137.7 148.7 amd-pstate-epp performance, EPP performance 20.2 137.5 148.2 amd-pstate-epp powersave, EPP power 22.8 137.3 149.1 amd-pstate-epp powersave, EPP balance_performance 19.5 137.3 148.2 amd-pstate-epp powersave, EPP performance 19.6 137.2 148.8 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.8 65.4 67.3 amd-pstate-epp powersave, EPP balance_performance 46.8 64.5 66.3 amd-pstate-epp powersave, EPP power 46.6 64.4 66.8 acpi-cpufreq performance 46.6 64.2 65.8 acpi-cpufreq schedutil 45.8 63.7 65.6 amd-pstate-epp powersave, EPP performance 46.1 63.6 65.8 amd-pstate-epp powersave, EPP balance_power 45.9 63.5 65.4 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Parallel amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance 30 60 90 120 150 SE +/- 0.19, N = 3 SE +/- 0.14, N = 3 SE +/- 0.30, N = 3 SE +/- 1.30, N = 3 SE +/- 0.52, N = 3 SE +/- 0.33, N = 3 SE +/- 0.23, N = 3 151.19 151.43 151.74 152.61 152.68 152.75 152.78 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Parallel amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance 2 4 6 8 10 SE +/- 0.00837, N = 3 SE +/- 0.00609, N = 3 SE +/- 0.01316, N = 3 SE +/- 0.05580, N = 3 SE +/- 0.02213, N = 3 SE +/- 0.01392, N = 3 SE +/- 0.00989, N = 3 6.61050 6.60006 6.58721 6.55080 6.54688 6.54343 6.54213 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil 20 40 60 80 100 SE +/- 0.96, N = 3 SE +/- 0.83, N = 3 SE +/- 0.58, N = 3 SE +/- 1.02, N = 3 SE +/- 1.06, N = 3 SE +/- 0.31, N = 3 SE +/- 0.61, N = 3 86.42 86.16 84.32 84.21 84.18 84.09 83.68 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5047 5448 amd-pstate-epp powersave, EPP power 600 5061 5618 amd-pstate-epp powersave, EPP balance_power 600 5067 5601 amd-pstate-epp powersave, EPP balance_performance 600 5071 5333 amd-pstate-epp performance, EPP performance 600 5071 5696 acpi-cpufreq performance 4300 5149 5594 acpi-cpufreq schedutil 4300 5161 5630 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.3 146.7 160.9 amd-pstate-epp powersave, EPP balance_power 19.7 145.1 160.8 acpi-cpufreq performance 20.3 145.1 160.8 amd-pstate-epp powersave, EPP power 22.7 145.0 160.0 amd-pstate-epp powersave, EPP performance 19.5 144.3 160.3 amd-pstate-epp powersave, EPP balance_performance 19.0 144.2 159.3 acpi-cpufreq schedutil 19.5 143.0 158.5 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 45.3 70.8 75.0 amd-pstate-epp powersave, EPP balance_performance 44.8 69.3 72.9 acpi-cpufreq performance 44.6 68.4 72.0 amd-pstate-epp powersave, EPP power 44.3 68.2 71.8 amd-pstate-epp powersave, EPP performance 43.5 67.7 72.1 amd-pstate-epp powersave, EPP balance_power 43.6 67.5 70.8 acpi-cpufreq schedutil 43.6 67.3 71.1 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil 3 6 9 12 15 SE +/- 0.13, N = 3 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 SE +/- 0.15, N = 3 SE +/- 0.15, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 11.57 11.61 11.86 11.88 11.88 11.89 11.95 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil 20 40 60 80 100 SE +/- 0.96, N = 3 SE +/- 0.83, N = 3 SE +/- 0.58, N = 3 SE +/- 1.02, N = 3 SE +/- 1.06, N = 3 SE +/- 0.31, N = 3 SE +/- 0.61, N = 3 86.42 86.16 84.32 84.21 84.18 84.09 83.68 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 40 80 120 160 200 SE +/- 0.21, N = 3 SE +/- 0.41, N = 3 SE +/- 0.30, N = 3 SE +/- 0.54, N = 3 SE +/- 0.71, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 196.74 196.17 194.72 194.48 193.95 193.67 191.12 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4444 4970 amd-pstate-epp powersave, EPP power 600 4550 5650 amd-pstate-epp powersave, EPP balance_power 600 4551 4774 amd-pstate-epp powersave, EPP performance 600 4564 5656 acpi-cpufreq schedutil 4244 4651 5454 amd-pstate-epp performance, EPP performance 600 4662 5698 acpi-cpufreq performance 4240 4692 5460 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 20.0 170.3 182.6 amd-pstate-epp performance, EPP performance 21.5 169.5 181.8 acpi-cpufreq schedutil 19.5 169.1 182.5 amd-pstate-epp powersave, EPP balance_power 10.1 169.0 182.8 amd-pstate-epp powersave, EPP performance 19.9 168.5 181.8 amd-pstate-epp powersave, EPP power 22.5 167.9 182.5 amd-pstate-epp powersave, EPP balance_performance 19.2 167.3 180.6 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.1 76.5 78.6 amd-pstate-epp powersave, EPP balance_performance 46.5 76.1 78.3 acpi-cpufreq performance 48.8 75.7 77.8 amd-pstate-epp powersave, EPP power 46.0 75.4 77.5 acpi-cpufreq schedutil 45.4 75.2 77.3 amd-pstate-epp powersave, EPP performance 45.3 75.1 77.3 amd-pstate-epp powersave, EPP balance_power 45.5 75.0 77.3 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 1.1773 2.3546 3.5319 4.7092 5.8865 SE +/- 0.00539, N = 3 SE +/- 0.01075, N = 3 SE +/- 0.00783, N = 3 SE +/- 0.01434, N = 3 SE +/- 0.01895, N = 3 SE +/- 0.00149, N = 3 SE +/- 0.00250, N = 3 5.08286 5.09761 5.13563 5.14187 5.15618 5.16325 5.23228 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 40 80 120 160 200 SE +/- 0.21, N = 3 SE +/- 0.41, N = 3 SE +/- 0.30, N = 3 SE +/- 0.54, N = 3 SE +/- 0.71, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 196.74 196.17 194.72 194.48 193.95 193.67 191.12 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.10, N = 3 14.02 13.98 13.96 13.95 13.87 13.87 13.74 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5244 5701 amd-pstate-epp powersave, EPP balance_power 600 5251 5547 amd-pstate-epp powersave, EPP power 600 5254 5505 amd-pstate-epp powersave, EPP balance_performance 600 5265 5527 amd-pstate-epp performance, EPP performance 600 5335 5687 acpi-cpufreq performance 4300 5357 5577 acpi-cpufreq schedutil 4300 5370 5580 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq schedutil 19.4 159.2 172.9 amd-pstate-epp powersave, EPP balance_power 18.6 158.9 172.7 amd-pstate-epp performance, EPP performance 21.9 158.8 172.9 acpi-cpufreq performance 19.9 158.6 174.1 amd-pstate-epp powersave, EPP performance 18.4 158.0 173.0 amd-pstate-epp powersave, EPP power 22.5 157.9 172.5 amd-pstate-epp powersave, EPP balance_performance 18.7 157.7 172.8 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.4 71.8 74.1 amd-pstate-epp powersave, EPP power 46.5 71.1 73.9 amd-pstate-epp powersave, EPP balance_performance 46.6 71.0 73.8 acpi-cpufreq performance 46.5 70.5 73.1 acpi-cpufreq schedutil 45.6 70.2 72.8 amd-pstate-epp powersave, EPP performance 45.4 70.1 72.6 amd-pstate-epp powersave, EPP balance_power 45.4 69.9 72.5 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 16 32 48 64 80 SE +/- 0.03, N = 3 SE +/- 0.18, N = 3 SE +/- 0.12, N = 3 SE +/- 0.44, N = 3 SE +/- 0.05, N = 3 SE +/- 0.35, N = 3 SE +/- 0.51, N = 3 71.34 71.51 71.64 71.69 72.10 72.10 72.76 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.10, N = 3 14.02 13.98 13.96 13.95 13.87 13.87 13.74 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Standard acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 2 4 6 8 10 SE +/- 0.00560, N = 3 SE +/- 0.00737, N = 3 SE +/- 0.00630, N = 3 SE +/- 0.01824, N = 3 SE +/- 0.02084, N = 3 SE +/- 0.01210, N = 3 SE +/- 0.00980, N = 3 7.85510 7.82813 7.82760 7.82105 7.81452 7.80868 7.80322 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5220 5651 amd-pstate-epp powersave, EPP balance_performance 600 5225 5686 amd-pstate-epp performance, EPP performance 600 5249 5685 amd-pstate-epp powersave, EPP power 3027 5264 5528 amd-pstate-epp powersave, EPP balance_power 600 5271 5504 acpi-cpufreq performance 4300 5352 5671 acpi-cpufreq schedutil 4300 5362 5514 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.7 165.4 178.0 acpi-cpufreq performance 20.1 165.2 179.0 amd-pstate-epp powersave, EPP performance 20.0 164.4 178.6 amd-pstate-epp powersave, EPP balance_power 19.3 164.4 177.4 amd-pstate-epp powersave, EPP balance_performance 19.1 163.9 176.9 amd-pstate-epp powersave, EPP power 22.6 163.7 177.1 acpi-cpufreq schedutil 19.3 163.7 177.1 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.9 72.2 74.3 amd-pstate-epp powersave, EPP balance_performance 47.5 71.7 73.6 acpi-cpufreq performance 46.9 71.2 73.0 amd-pstate-epp powersave, EPP power 47.1 71.2 73.0 amd-pstate-epp powersave, EPP balance_power 46.5 70.5 72.4 amd-pstate-epp powersave, EPP performance 46.0 70.5 72.1 acpi-cpufreq schedutil 46.4 70.2 72.1 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Standard acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 30 60 90 120 150 SE +/- 0.09, N = 3 SE +/- 0.12, N = 3 SE +/- 0.10, N = 3 SE +/- 0.30, N = 3 SE +/- 0.34, N = 3 SE +/- 0.20, N = 3 SE +/- 0.16, N = 3 127.28 127.72 127.73 127.83 127.94 128.04 128.12 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Standard acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 2 4 6 8 10 SE +/- 0.00560, N = 3 SE +/- 0.00737, N = 3 SE +/- 0.00630, N = 3 SE +/- 0.01824, N = 3 SE +/- 0.02084, N = 3 SE +/- 0.01210, N = 3 SE +/- 0.00980, N = 3 7.85510 7.82813 7.82760 7.82105 7.81452 7.80868 7.80322 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Standard acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 9 18 27 36 45 SE +/- 0.22, N = 3 SE +/- 0.17, N = 3 SE +/- 0.16, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 SE +/- 0.12, N = 3 39.97 39.87 39.87 39.87 39.77 39.70 39.65 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4881 5229 amd-pstate-epp powersave, EPP power 600 4902 5201 amd-pstate-epp performance, EPP performance 600 4923 5688 amd-pstate-epp powersave, EPP balance_power 600 4955 5179 amd-pstate-epp powersave, EPP performance 600 4985 5691 acpi-cpufreq performance 4300 5058 5198 acpi-cpufreq schedutil 4300 5066 5579 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.8 171.3 184.0 amd-pstate-epp performance, EPP performance 21.5 171.2 183.5 amd-pstate-epp powersave, EPP balance_power 19.4 170.3 183.0 acpi-cpufreq schedutil 19.7 170.3 182.7 amd-pstate-epp powersave, EPP power 22.1 170.1 182.6 amd-pstate-epp powersave, EPP balance_performance 18.3 170.0 183.4 amd-pstate-epp powersave, EPP performance 19.3 169.1 182.3 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.1 84.4 86.4 amd-pstate-epp powersave, EPP balance_performance 46.6 83.8 86.0 acpi-cpufreq performance 46.3 83.3 85.5 amd-pstate-epp powersave, EPP power 46.0 82.8 85.3 acpi-cpufreq schedutil 45.5 82.2 84.5 amd-pstate-epp powersave, EPP balance_power 45.4 82.1 84.5 amd-pstate-epp powersave, EPP performance 45.0 82.1 84.4 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Standard acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 6 12 18 24 30 SE +/- 0.14, N = 3 SE +/- 0.11, N = 3 SE +/- 0.02, N = 3 SE +/- 0.10, N = 3 SE +/- 0.01, N = 3 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 25.02 25.08 25.08 25.08 25.14 25.19 25.22 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Standard acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 9 18 27 36 45 SE +/- 0.22, N = 3 SE +/- 0.17, N = 3 SE +/- 0.16, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 SE +/- 0.12, N = 3 39.97 39.87 39.87 39.87 39.77 39.70 39.65 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance 9 18 27 36 45 SE +/- 0.19, N = 3 SE +/- 0.28, N = 3 SE +/- 0.07, N = 3 SE +/- 0.25, N = 3 SE +/- 0.15, N = 3 SE +/- 0.25, N = 3 SE +/- 0.10, N = 3 39.97 39.69 39.58 39.53 39.47 39.42 39.31 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5058 5661 amd-pstate-epp powersave, EPP power 600 5086 5574 amd-pstate-epp powersave, EPP balance_performance 600 5103 5417 amd-pstate-epp powersave, EPP balance_power 600 5139 5512 amd-pstate-epp performance, EPP performance 600 5174 5688 acpi-cpufreq performance 4300 5219 5586 acpi-cpufreq schedutil 4300 5254 5564 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP balance_power 19.6 159.1 173.9 amd-pstate-epp performance, EPP performance 18.5 158.7 174.8 amd-pstate-epp powersave, EPP performance 18.0 158.6 173.4 acpi-cpufreq performance 19.9 158.2 175.7 acpi-cpufreq schedutil 19.3 158.1 174.9 amd-pstate-epp powersave, EPP balance_performance 19.1 156.9 173.8 amd-pstate-epp powersave, EPP power 22.7 154.9 175.4 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.8 72.7 76.9 amd-pstate-epp powersave, EPP balance_performance 47.4 72.0 75.9 acpi-cpufreq schedutil 46.1 71.1 74.0 acpi-cpufreq performance 47.1 71.1 74.1 amd-pstate-epp powersave, EPP power 47.1 71.0 74.6 amd-pstate-epp powersave, EPP balance_power 46.3 70.9 74.6 amd-pstate-epp powersave, EPP performance 46.1 70.7 74.1 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance 6 12 18 24 30 SE +/- 0.12, N = 3 SE +/- 0.18, N = 3 SE +/- 0.05, N = 3 SE +/- 0.16, N = 3 SE +/- 0.09, N = 3 SE +/- 0.16, N = 3 SE +/- 0.07, N = 3 25.02 25.20 25.27 25.30 25.33 25.37 25.44 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance 9 18 27 36 45 SE +/- 0.19, N = 3 SE +/- 0.28, N = 3 SE +/- 0.07, N = 3 SE +/- 0.25, N = 3 SE +/- 0.15, N = 3 SE +/- 0.25, N = 3 SE +/- 0.10, N = 3 39.97 39.69 39.58 39.53 39.47 39.42 39.31 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 1.1659 2.3318 3.4977 4.6636 5.8295 SE +/- 0.01860, N = 3 SE +/- 0.02051, N = 3 SE +/- 0.01409, N = 3 SE +/- 0.00173, N = 3 SE +/- 0.02037, N = 3 SE +/- 0.01504, N = 3 SE +/- 0.00744, N = 3 5.18192 5.18066 5.17978 5.17443 5.16594 5.15748 5.11863 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 5257 5542 amd-pstate-epp powersave, EPP balance_power 600 5274 5548 amd-pstate-epp powersave, EPP performance 600 5307 5734 amd-pstate-epp powersave, EPP balance_performance 600 5330 5541 amd-pstate-epp performance, EPP performance 600 5400 5686 acpi-cpufreq performance 4300 5426 5570 acpi-cpufreq schedutil 4300 5436 5606 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 20.0 131.3 141.6 amd-pstate-epp powersave, EPP balance_power 19.6 131.0 141.3 amd-pstate-epp performance, EPP performance 11.2 130.9 140.2 amd-pstate-epp powersave, EPP power 22.7 130.8 140.6 acpi-cpufreq schedutil 19.3 130.8 141.4 amd-pstate-epp powersave, EPP performance 18.5 130.7 140.4 amd-pstate-epp powersave, EPP balance_performance 19.3 130.1 140.2 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.6 64.1 65.6 amd-pstate-epp powersave, EPP balance_performance 47.8 63.3 64.6 amd-pstate-epp powersave, EPP power 47.3 63.1 64.5 acpi-cpufreq performance 47.4 63.0 65.0 amd-pstate-epp powersave, EPP performance 46.5 62.4 64.0 amd-pstate-epp powersave, EPP balance_power 46.5 62.2 63.6 acpi-cpufreq schedutil 46.3 62.1 63.6 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 40 80 120 160 200 SE +/- 0.69, N = 3 SE +/- 0.77, N = 3 SE +/- 0.53, N = 3 SE +/- 0.07, N = 3 SE +/- 0.76, N = 3 SE +/- 0.56, N = 3 SE +/- 0.28, N = 3 192.96 193.00 193.04 193.23 193.56 193.87 195.34 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 1.1659 2.3318 3.4977 4.6636 5.8295 SE +/- 0.01860, N = 3 SE +/- 0.02051, N = 3 SE +/- 0.01409, N = 3 SE +/- 0.00173, N = 3 SE +/- 0.02037, N = 3 SE +/- 0.01504, N = 3 SE +/- 0.00744, N = 3 5.18192 5.18066 5.17978 5.17443 5.16594 5.15748 5.11863 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance acpi-cpufreq performance 16 32 48 64 80 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 SE +/- 0.08, N = 3 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 SE +/- 0.25, N = 3 70.76 70.47 70.45 70.45 70.32 70.29 70.02 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4847 5261 amd-pstate-epp powersave, EPP performance 600 4856 5597 amd-pstate-epp powersave, EPP balance_performance 600 4876 5533 amd-pstate-epp powersave, EPP balance_power 600 4908 5293 amd-pstate-epp performance, EPP performance 600 4914 5686 acpi-cpufreq performance 4300 5000 5574 acpi-cpufreq schedutil 4300 5019 5670 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.3 141.5 157.3 amd-pstate-epp powersave, EPP balance_power 19.1 141.3 154.0 amd-pstate-epp performance, EPP performance 20.6 141.0 158.4 amd-pstate-epp powersave, EPP power 22.3 140.7 158.1 amd-pstate-epp powersave, EPP performance 17.3 140.7 160.9 amd-pstate-epp powersave, EPP balance_performance 19.8 140.2 157.0 acpi-cpufreq schedutil 19.1 140.1 152.1 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.0 68.4 70.0 amd-pstate-epp powersave, EPP balance_performance 47.4 67.5 69.1 amd-pstate-epp powersave, EPP power 47.4 67.4 69.1 acpi-cpufreq performance 46.9 66.9 68.5 amd-pstate-epp powersave, EPP performance 46.1 66.4 68.0 acpi-cpufreq schedutil 46.3 66.3 68.0 amd-pstate-epp powersave, EPP balance_power 46.3 66.3 68.0 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance acpi-cpufreq performance 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 14.13 14.19 14.19 14.19 14.22 14.23 14.28 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance acpi-cpufreq performance 16 32 48 64 80 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 SE +/- 0.08, N = 3 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 SE +/- 0.25, N = 3 70.76 70.47 70.45 70.45 70.32 70.29 70.02 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 16.17 16.15 16.13 16.04 16.04 15.98 15.96 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5189 5533 amd-pstate-epp performance, EPP performance 600 5217 5692 amd-pstate-epp powersave, EPP power 600 5232 5617 amd-pstate-epp powersave, EPP balance_power 600 5235 5507 amd-pstate-epp powersave, EPP performance 600 5243 5727 acpi-cpufreq schedutil 3027 5335 5501 acpi-cpufreq performance 4300 5355 5682 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 20.2 160.8 173.0 acpi-cpufreq schedutil 19.1 160.5 173.2 amd-pstate-epp powersave, EPP power 22.3 160.0 173.6 amd-pstate-epp powersave, EPP balance_power 19.7 159.9 172.4 amd-pstate-epp powersave, EPP performance 18.8 159.7 172.7 amd-pstate-epp performance, EPP performance 18.4 159.7 172.9 amd-pstate-epp powersave, EPP balance_performance 19.4 158.4 170.5 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.4 80.7 82.5 amd-pstate-epp powersave, EPP balance_performance 46.4 80.6 82.6 amd-pstate-epp powersave, EPP power 45.8 79.8 82.1 acpi-cpufreq performance 45.4 79.5 81.6 acpi-cpufreq schedutil 45.3 79.4 81.6 amd-pstate-epp powersave, EPP performance 43.3 79.1 81.1 amd-pstate-epp powersave, EPP balance_power 45.3 79.0 81.1 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance 14 28 42 56 70 SE +/- 0.17, N = 3 SE +/- 0.07, N = 3 SE +/- 0.21, N = 3 SE +/- 0.19, N = 3 SE +/- 0.11, N = 3 SE +/- 0.17, N = 3 SE +/- 0.21, N = 3 61.83 61.91 62.01 62.33 62.33 62.56 62.66 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 16.17 16.15 16.13 16.04 16.04 15.98 15.96 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance 4 8 12 16 20 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 17.43 17.38 17.33 17.30 17.22 17.21 17.21 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4777 5635 amd-pstate-epp powersave, EPP balance_power 600 4806 5328 amd-pstate-epp powersave, EPP balance_performance 600 4859 5121 amd-pstate-epp performance, EPP performance 600 4869 5687 amd-pstate-epp powersave, EPP power 600 4876 5440 acpi-cpufreq performance 4300 4975 5651 acpi-cpufreq schedutil 4300 4986 5672 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.9 165.0 176.1 acpi-cpufreq performance 18.2 164.5 177.5 amd-pstate-epp powersave, EPP balance_power 19.2 164.0 176.2 amd-pstate-epp powersave, EPP balance_performance 19.1 163.5 176.0 amd-pstate-epp powersave, EPP power 22.0 163.4 175.9 acpi-cpufreq schedutil 19.0 163.2 175.8 amd-pstate-epp powersave, EPP performance 17.9 162.8 175.7 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.9 73.5 75.5 amd-pstate-epp powersave, EPP balance_performance 47.5 73.2 75.3 acpi-cpufreq performance 47.0 72.1 74.1 amd-pstate-epp powersave, EPP power 45.6 72.0 74.1 amd-pstate-epp powersave, EPP performance 46.1 71.8 73.8 acpi-cpufreq schedutil 46.4 71.8 73.8 amd-pstate-epp powersave, EPP balance_power 46.4 71.6 73.5 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance 13 26 39 52 65 SE +/- 0.24, N = 3 SE +/- 0.21, N = 3 SE +/- 0.12, N = 3 SE +/- 0.09, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 57.35 57.53 57.69 57.79 58.06 58.09 58.09 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance 4 8 12 16 20 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 17.43 17.38 17.33 17.30 17.22 17.21 17.21 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 24.56 24.55 24.20 24.14 24.12 24.09 24.08 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5210 5677 amd-pstate-epp powersave, EPP balance_performance 600 5258 5535 amd-pstate-epp powersave, EPP power 600 5287 5546 amd-pstate-epp powersave, EPP balance_power 600 5289 5550 amd-pstate-epp performance, EPP performance 600 5357 5695 acpi-cpufreq schedutil 4300 5380 5511 acpi-cpufreq performance 4300 5389 5581 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.0 146.5 160.5 amd-pstate-epp powersave, EPP balance_power 19.6 146.3 159.9 amd-pstate-epp powersave, EPP power 22.3 145.8 158.9 amd-pstate-epp powersave, EPP balance_performance 17.9 145.1 160.1 acpi-cpufreq performance 18.0 144.9 157.3 acpi-cpufreq schedutil 19.3 143.5 158.0 amd-pstate-epp powersave, EPP performance 11.3 141.7 158.5 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 47.3 75.5 78.1 amd-pstate-epp performance, EPP performance 47.5 75.4 78.1 amd-pstate-epp powersave, EPP power 46.8 74.5 77.0 acpi-cpufreq performance 46.8 73.8 76.4 amd-pstate-epp powersave, EPP balance_power 46.0 73.8 76.3 acpi-cpufreq schedutil 46.3 73.8 76.1 amd-pstate-epp powersave, EPP performance 45.9 73.7 76.3 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 9 18 27 36 45 SE +/- 0.11, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 40.72 40.73 41.31 41.42 41.45 41.51 41.52 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 24.56 24.55 24.20 24.14 24.12 24.09 24.08 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance 0.9124 1.8248 2.7372 3.6496 4.562 SE +/- 0.03625, N = 3 SE +/- 0.02051, N = 3 SE +/- 0.05037, N = 3 SE +/- 0.01328, N = 3 SE +/- 0.01431, N = 3 SE +/- 0.01610, N = 3 SE +/- 0.03702, N = 3 4.05533 4.04379 4.03341 4.02330 4.01785 4.01014 3.96846 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5175 5498 amd-pstate-epp powersave, EPP balance_performance 600 5254 5499 amd-pstate-epp powersave, EPP power 600 5274 5515 amd-pstate-epp powersave, EPP balance_power 600 5293 5568 amd-pstate-epp performance, EPP performance 600 5319 5697 acpi-cpufreq performance 4300 5371 5485 acpi-cpufreq schedutil 4300 5375 5497 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.2 154.5 166.8 amd-pstate-epp powersave, EPP performance 19.3 154.2 167.7 acpi-cpufreq performance 19.9 153.7 167.7 acpi-cpufreq schedutil 19.2 153.4 167.2 amd-pstate-epp powersave, EPP balance_power 19.4 153.4 167.9 amd-pstate-epp powersave, EPP balance_performance 19.0 152.7 166.6 amd-pstate-epp powersave, EPP power 22.4 152.0 167.5 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.6 70.2 72.3 amd-pstate-epp powersave, EPP balance_performance 46.4 69.5 71.8 acpi-cpufreq performance 46.1 68.9 71.1 amd-pstate-epp powersave, EPP power 45.5 68.5 71.4 amd-pstate-epp powersave, EPP performance 44.3 68.3 70.5 acpi-cpufreq schedutil 44.6 67.9 70.3 amd-pstate-epp powersave, EPP balance_power 44.8 67.9 70.0 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance 60 120 180 240 300 SE +/- 2.23, N = 3 SE +/- 1.26, N = 3 SE +/- 3.12, N = 3 SE +/- 0.82, N = 3 SE +/- 0.88, N = 3 SE +/- 1.01, N = 3 SE +/- 2.37, N = 3 246.56 247.25 247.95 248.49 248.81 249.29 251.96 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance 0.9124 1.8248 2.7372 3.6496 4.562 SE +/- 0.03625, N = 3 SE +/- 0.02051, N = 3 SE +/- 0.05037, N = 3 SE +/- 0.01328, N = 3 SE +/- 0.01431, N = 3 SE +/- 0.01610, N = 3 SE +/- 0.03702, N = 3 4.05533 4.04379 4.03341 4.02330 4.01785 4.01014 3.96846 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power 0.2622 0.5244 0.7866 1.0488 1.311 SE +/- 0.00424, N = 3 SE +/- 0.00086, N = 3 SE +/- 0.00512, N = 3 SE +/- 0.00197, N = 3 SE +/- 0.00179, N = 3 SE +/- 0.00384, N = 3 SE +/- 0.01213, N = 3 1.16550 1.16437 1.16412 1.16379 1.16319 1.16025 1.15589 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5189 5716 amd-pstate-epp powersave, EPP balance_power 600 5196 5493 amd-pstate-epp powersave, EPP power 600 5199 5451 amd-pstate-epp performance, EPP performance 600 5205 5689 amd-pstate-epp powersave, EPP balance_performance 600 5225 5438 acpi-cpufreq performance 4300 5300 5444 acpi-cpufreq schedutil 4300 5318 5469 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.3 169.4 180.8 amd-pstate-epp powersave, EPP balance_power 19.0 169.2 183.4 amd-pstate-epp powersave, EPP power 22.6 169.1 182.7 acpi-cpufreq performance 19.3 169.0 181.4 acpi-cpufreq schedutil 19.9 168.8 182.7 amd-pstate-epp powersave, EPP performance 19.1 168.6 181.0 amd-pstate-epp powersave, EPP balance_performance 19.2 168.2 180.9 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.1 73.1 75.3 amd-pstate-epp powersave, EPP balance_performance 46.8 72.6 74.6 amd-pstate-epp powersave, EPP power 46.3 72.2 74.8 acpi-cpufreq performance 46.6 71.7 73.9 amd-pstate-epp powersave, EPP performance 45.1 71.4 73.3 acpi-cpufreq schedutil 45.6 71.0 73.4 amd-pstate-epp powersave, EPP balance_power 45.5 71.0 73.3 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power 200 400 600 800 1000 SE +/- 3.11, N = 3 SE +/- 0.63, N = 3 SE +/- 3.76, N = 3 SE +/- 1.44, N = 3 SE +/- 1.35, N = 3 SE +/- 2.88, N = 3 SE +/- 9.16, N = 3 857.48 858.27 858.50 858.74 859.16 861.43 864.75 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power 0.2622 0.5244 0.7866 1.0488 1.311 SE +/- 0.00424, N = 3 SE +/- 0.00086, N = 3 SE +/- 0.00512, N = 3 SE +/- 0.00197, N = 3 SE +/- 0.00179, N = 3 SE +/- 0.00384, N = 3 SE +/- 0.01213, N = 3 1.16550 1.16437 1.16412 1.16379 1.16319 1.16025 1.15589 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance 0.3864 0.7728 1.1592 1.5456 1.932 SE +/- 0.00866, N = 3 SE +/- 0.00550, N = 3 SE +/- 0.00358, N = 3 SE +/- 0.00124, N = 3 SE +/- 0.00579, N = 3 SE +/- 0.00573, N = 3 SE +/- 0.00700, N = 3 1.71720 1.71427 1.70884 1.70746 1.70305 1.70234 1.70042 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4809 5069 amd-pstate-epp powersave, EPP power 600 4826 5300 amd-pstate-epp powersave, EPP balance_performance 600 4870 5070 amd-pstate-epp performance, EPP performance 600 4878 5686 amd-pstate-epp powersave, EPP performance 600 4887 5069 acpi-cpufreq performance 4300 4963 5100 acpi-cpufreq schedutil 4300 4983 5640 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.2 164.7 176.2 amd-pstate-epp powersave, EPP balance_power 19.7 164.0 175.5 acpi-cpufreq schedutil 18.7 163.2 175.5 amd-pstate-epp powersave, EPP performance 19.5 163.1 175.8 amd-pstate-epp performance, EPP performance 20.5 163.0 174.9 amd-pstate-epp powersave, EPP balance_performance 19.2 162.5 174.1 amd-pstate-epp powersave, EPP power 21.8 162.5 175.1 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.6 76.6 78.3 amd-pstate-epp powersave, EPP balance_performance 47.6 76.5 78.1 amd-pstate-epp powersave, EPP power 46.6 75.7 77.6 acpi-cpufreq performance 46.4 75.6 77.5 amd-pstate-epp powersave, EPP performance 46.1 75.6 77.4 acpi-cpufreq schedutil 46.1 75.4 77.4 amd-pstate-epp powersave, EPP balance_power 46.0 75.3 77.4 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance 130 260 390 520 650 SE +/- 2.95, N = 3 SE +/- 1.86, N = 3 SE +/- 1.22, N = 3 SE +/- 0.42, N = 3 SE +/- 2.00, N = 3 SE +/- 1.98, N = 3 SE +/- 2.44, N = 3 582.13 583.04 584.91 585.41 586.93 587.20 587.83 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance 0.3864 0.7728 1.1592 1.5456 1.932 SE +/- 0.00866, N = 3 SE +/- 0.00550, N = 3 SE +/- 0.00358, N = 3 SE +/- 0.00124, N = 3 SE +/- 0.00579, N = 3 SE +/- 0.00573, N = 3 SE +/- 0.00700, N = 3 1.71720 1.71427 1.70884 1.70746 1.70305 1.70234 1.70042 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance 1.0266 2.0532 3.0798 4.1064 5.133 SE +/- 0.01307, N = 3 SE +/- 0.01898, N = 3 SE +/- 0.02587, N = 3 SE +/- 0.01980, N = 3 SE +/- 0.01827, N = 3 SE +/- 0.03884, N = 3 SE +/- 0.01216, N = 3 4.56251 4.55799 4.55439 4.52501 4.52214 4.52082 4.50600 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4820 5460 amd-pstate-epp powersave, EPP performance 600 4847 5632 amd-pstate-epp powersave, EPP balance_performance 600 4859 5084 amd-pstate-epp powersave, EPP power 600 4865 5110 amd-pstate-epp performance, EPP performance 600 4934 5685 acpi-cpufreq performance 4300 4982 5121 acpi-cpufreq schedutil 3000 4986 5347 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP power 22.2 161.0 173.3 amd-pstate-epp performance, EPP performance 20.0 160.9 173.2 acpi-cpufreq performance 19.8 160.7 172.6 acpi-cpufreq schedutil 19.2 159.9 172.2 amd-pstate-epp powersave, EPP balance_power 19.9 159.1 171.6 amd-pstate-epp powersave, EPP balance_performance 19.3 158.8 171.6 amd-pstate-epp powersave, EPP performance 18.8 158.3 170.7 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 47.1 78.6 82.9 amd-pstate-epp performance, EPP performance 47.5 78.1 83.8 amd-pstate-epp powersave, EPP performance 45.8 77.8 83.4 amd-pstate-epp powersave, EPP power 46.4 76.9 83.0 amd-pstate-epp powersave, EPP balance_power 46.3 76.5 81.0 acpi-cpufreq schedutil 46.0 76.0 83.0 acpi-cpufreq performance 46.5 75.3 80.6 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance 50 100 150 200 250 SE +/- 0.63, N = 3 SE +/- 0.91, N = 3 SE +/- 1.25, N = 3 SE +/- 0.97, N = 3 SE +/- 0.89, N = 3 SE +/- 1.89, N = 3 SE +/- 0.60, N = 3 219.16 219.38 219.56 220.98 221.11 221.20 221.90 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Parallel amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance 1.0266 2.0532 3.0798 4.1064 5.133 SE +/- 0.01307, N = 3 SE +/- 0.01898, N = 3 SE +/- 0.02587, N = 3 SE +/- 0.01980, N = 3 SE +/- 0.01827, N = 3 SE +/- 0.03884, N = 3 SE +/- 0.01216, N = 3 4.56251 4.55799 4.55439 4.52501 4.52214 4.52082 4.50600 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance 1.0051 2.0102 3.0153 4.0204 5.0255 SE +/- 0.03439, N = 3 SE +/- 0.00278, N = 3 SE +/- 0.00515, N = 3 SE +/- 0.00738, N = 3 SE +/- 0.00702, N = 3 SE +/- 0.00658, N = 3 SE +/- 0.01137, N = 3 4.46730 4.44876 4.44344 4.44308 4.44079 4.43288 4.41932 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4828 5631 amd-pstate-epp powersave, EPP balance_performance 600 4829 5055 amd-pstate-epp powersave, EPP power 600 4903 5060 amd-pstate-epp powersave, EPP balance_power 600 4906 5058 amd-pstate-epp performance, EPP performance 600 4942 5693 acpi-cpufreq performance 4300 4956 5076 acpi-cpufreq schedutil 4300 4958 5050 OpenBenchmarking.org Megahertz, More Is Better ONNX Runtime 1.19 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.8 148.4 158.1 amd-pstate-epp powersave, EPP power 22.4 148.1 157.7 amd-pstate-epp powersave, EPP balance_power 19.5 147.5 157.2 acpi-cpufreq performance 18.9 147.4 157.6 amd-pstate-epp powersave, EPP balance_performance 18.7 147.0 156.8 acpi-cpufreq schedutil 19.1 146.7 156.4 amd-pstate-epp powersave, EPP performance 19.1 146.4 156.5 OpenBenchmarking.org Watts, Fewer Is Better ONNX Runtime 1.19 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.8 87.7 90.3 amd-pstate-epp powersave, EPP balance_performance 46.5 87.6 89.9 amd-pstate-epp powersave, EPP power 45.6 86.5 88.9 acpi-cpufreq performance 46.4 86.5 89.3 acpi-cpufreq schedutil 45.9 85.7 88.4 amd-pstate-epp powersave, EPP balance_power 45.1 85.4 88.0 amd-pstate-epp powersave, EPP performance 44.9 85.3 87.5 OpenBenchmarking.org Celsius, Fewer Is Better ONNX Runtime 1.19 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance 50 100 150 200 250 SE +/- 1.71, N = 3 SE +/- 0.14, N = 3 SE +/- 0.26, N = 3 SE +/- 0.37, N = 3 SE +/- 0.36, N = 3 SE +/- 0.34, N = 3 SE +/- 0.58, N = 3 223.87 224.77 225.04 225.06 225.18 225.58 226.27 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Standard amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance 1.0051 2.0102 3.0153 4.0204 5.0255 SE +/- 0.03439, N = 3 SE +/- 0.00278, N = 3 SE +/- 0.00515, N = 3 SE +/- 0.00738, N = 3 SE +/- 0.00702, N = 3 SE +/- 0.00658, N = 3 SE +/- 0.01137, N = 3 4.46730 4.44876 4.44344 4.44308 4.44079 4.43288 4.41932 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 5 - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 10 20 30 40 50 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 43.18 43.59 44.10 44.19 44.27 44.29 44.48 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 5 - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power 0.0569 0.1138 0.1707 0.2276 0.2845 0.249 0.250 0.252 0.253 0.253 0.253 0.253 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4772 5601 amd-pstate-epp powersave, EPP power 3016 4795 5681 amd-pstate-epp powersave, EPP balance_power 600 4824 5514 amd-pstate-epp powersave, EPP balance_performance 600 4847 5631 amd-pstate-epp performance, EPP performance 600 4877 5697 acpi-cpufreq schedutil 4209 4945 5583 acpi-cpufreq performance 4225 4957 5703 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.6 175.9 199.5 amd-pstate-epp powersave, EPP performance 19.7 174.8 199.1 amd-pstate-epp powersave, EPP balance_performance 19.5 174.8 199.7 amd-pstate-epp powersave, EPP balance_power 19.0 174.7 199.2 amd-pstate-epp powersave, EPP power 22.4 174.7 200.0 acpi-cpufreq performance 19.8 174.0 199.2 acpi-cpufreq schedutil 19.4 173.6 198.8 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.3 80.3 85.1 amd-pstate-epp powersave, EPP power 46.5 80.3 86.6 acpi-cpufreq performance 45.3 80.1 84.6 acpi-cpufreq schedutil 44.6 79.8 83.8 amd-pstate-epp powersave, EPP balance_performance 45.5 79.6 83.6 amd-pstate-epp powersave, EPP performance 45.1 79.3 83.1 amd-pstate-epp powersave, EPP balance_power 45.4 79.1 84.0 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 2.2 CPU Temperature Monitor 20 40 60 80 100
Xonotic This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game for this open-source first person shooter title. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.6 Resolution: 3840 x 2160 - Effects Quality: Ultimate acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 160 320 480 640 800 SE +/- 3.18, N = 3 SE +/- 0.48, N = 3 SE +/- 2.37, N = 3 SE +/- 1.82, N = 3 SE +/- 0.91, N = 3 SE +/- 2.64, N = 3 SE +/- 1.92, N = 3 717.95 720.23 721.30 722.36 722.86 725.07 725.16 MIN: 177 / MAX: 1171 MIN: 188 / MAX: 1173 MIN: 179 / MAX: 1179 MIN: 168 / MAX: 1171 MIN: 185 / MAX: 1171 MIN: 191 / MAX: 1172 MIN: 173 / MAX: 1166 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Xonotic 0.8.6 Resolution: 3840 x 2160 - Effects Quality: Ultimate amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance 3 6 9 12 15 10.54 10.69 10.76 10.78 10.91 11.25 11.31 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5349 5729 acpi-cpufreq performance 4300 5391 5726 amd-pstate-epp powersave, EPP power 3016 5461 5724 amd-pstate-epp powersave, EPP performance 600 5472 5750 amd-pstate-epp powersave, EPP balance_power 600 5484 5729 acpi-cpufreq schedutil 3032 5491 5690 amd-pstate-epp performance, EPP performance 5448 5596 5723 OpenBenchmarking.org Megahertz, More Is Better Xonotic 0.8.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP power 22.9 68.3 78.1 amd-pstate-epp performance, EPP performance 21.5 67.8 77.3 amd-pstate-epp powersave, EPP balance_power 20.2 67.3 79.2 amd-pstate-epp powersave, EPP performance 19.7 67.1 79.0 acpi-cpufreq schedutil 19.6 66.3 77.9 acpi-cpufreq performance 20.3 63.8 75.3 amd-pstate-epp powersave, EPP balance_performance 19.8 63.8 75.7 OpenBenchmarking.org Watts, Fewer Is Better Xonotic 0.8.6 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.9 62.3 65.8 acpi-cpufreq schedutil 39.8 61.2 65.8 amd-pstate-epp powersave, EPP power 41.4 60.8 64.8 amd-pstate-epp powersave, EPP performance 41.1 60.5 64.6 amd-pstate-epp powersave, EPP balance_power 40.0 60.1 64.4 acpi-cpufreq performance 40.1 59.5 63.6 amd-pstate-epp powersave, EPP balance_performance 40.9 59.2 65.0 OpenBenchmarking.org Celsius, Fewer Is Better Xonotic 0.8.6 CPU Temperature Monitor 20 40 60 80 100
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 1024 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance 2 4 6 8 10 SE +/- 0.074123, N = 14 SE +/- 0.080602, N = 3 SE +/- 0.063622, N = 3 SE +/- 0.060422, N = 3 SE +/- 0.057802, N = 7 SE +/- 0.013290, N = 3 SE +/- 0.004746, N = 3 6.147759 6.274104 6.279553 6.329842 6.350984 6.605524 6.634571 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 1024 amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0097 0.0194 0.0291 0.0388 0.0485 0.039 0.040 0.040 0.040 0.040 0.043 0.043 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4704 5719 amd-pstate-epp powersave, EPP power 600 4731 5701 amd-pstate-epp powersave, EPP performance 600 4784 5714 amd-pstate-epp powersave, EPP balance_performance 600 4870 5685 amd-pstate-epp performance, EPP performance 600 4950 5699 acpi-cpufreq performance 4300 4981 5601 acpi-cpufreq schedutil 4300 4991 5455 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.9 158.3 200.4 amd-pstate-epp powersave, EPP balance_power 20.0 157.3 200.4 amd-pstate-epp powersave, EPP performance 19.9 157.3 200.4 amd-pstate-epp powersave, EPP balance_performance 18.1 157.1 200.4 amd-pstate-epp powersave, EPP power 22.3 157.0 200.3 acpi-cpufreq performance 19.5 155.8 200.3 acpi-cpufreq schedutil 18.6 155.2 200.3 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.8 81.8 92.4 amd-pstate-epp powersave, EPP balance_performance 45.0 81.6 95.3 amd-pstate-epp powersave, EPP performance 44.8 79.6 91.5 amd-pstate-epp powersave, EPP power 46.0 79.6 95.1 amd-pstate-epp powersave, EPP balance_power 44.4 78.3 91.3 acpi-cpufreq schedutil 44.3 75.3 85.0 acpi-cpufreq performance 44.5 74.6 82.4 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100
VP9 libvpx Encoding This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 0 - Input: Bosphorus 4K amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 3 6 9 12 15 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 12.24 12.34 12.34 12.40 12.49 12.53 12.54 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 0 - Input: Bosphorus 4K amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil 0.0209 0.0418 0.0627 0.0836 0.1045 0.090 0.091 0.091 0.091 0.092 0.092 0.093 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5163 5624 amd-pstate-epp powersave, EPP performance 600 5230 5717 amd-pstate-epp powersave, EPP power 600 5241 5594 amd-pstate-epp powersave, EPP balance_power 600 5243 5597 amd-pstate-epp performance, EPP performance 600 5341 5699 acpi-cpufreq performance 4300 5355 5545 acpi-cpufreq schedutil 4300 5389 5590 OpenBenchmarking.org Megahertz, More Is Better VP9 libvpx Encoding 1.13 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.2 137.6 151.8 amd-pstate-epp powersave, EPP balance_power 10.2 136.8 151.1 acpi-cpufreq performance 19.6 136.8 150.7 amd-pstate-epp powersave, EPP power 22.3 136.3 149.4 amd-pstate-epp powersave, EPP performance 19.6 135.7 150.0 amd-pstate-epp powersave, EPP balance_performance 18.7 134.9 147.7 acpi-cpufreq schedutil 19.5 132.0 150.4 OpenBenchmarking.org Watts, Fewer Is Better VP9 libvpx Encoding 1.13 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 46.5 80.0 83.4 amd-pstate-epp performance, EPP performance 45.8 79.6 83.0 amd-pstate-epp powersave, EPP balance_power 45.4 78.6 82.5 amd-pstate-epp powersave, EPP performance 45.5 78.5 81.3 amd-pstate-epp powersave, EPP balance_performance 45.6 78.4 82.0 acpi-cpufreq performance 45.3 76.9 81.3 acpi-cpufreq schedutil 45.0 75.4 80.6 OpenBenchmarking.org Celsius, Fewer Is Better VP9 libvpx Encoding 1.13 CPU Temperature Monitor 20 40 60 80 100
Timed Linux Kernel Compilation This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.8 Build: defconfig amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 11 22 33 44 55 SE +/- 0.36, N = 3 SE +/- 0.34, N = 3 SE +/- 0.35, N = 3 SE +/- 0.33, N = 3 SE +/- 0.28, N = 3 SE +/- 0.33, N = 3 SE +/- 0.30, N = 3 47.91 47.90 47.84 47.75 47.66 47.64 47.51 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4873 5703 amd-pstate-epp powersave, EPP balance_power 600 4916 5712 amd-pstate-epp powersave, EPP performance 600 4918 5719 amd-pstate-epp performance, EPP performance 600 4961 5694 amd-pstate-epp powersave, EPP balance_performance 600 5002 5716 acpi-cpufreq performance 4300 5048 5722 acpi-cpufreq schedutil 4300 5051 5673 OpenBenchmarking.org Megahertz, More Is Better Timed Linux Kernel Compilation 6.8 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.3 155.4 201.2 acpi-cpufreq performance 19.6 155.3 201.0 acpi-cpufreq schedutil 18.7 155.1 201.6 amd-pstate-epp powersave, EPP balance_power 19.4 154.9 201.4 amd-pstate-epp powersave, EPP performance 20.3 154.6 201.0 amd-pstate-epp powersave, EPP power 22.1 154.5 201.2 amd-pstate-epp powersave, EPP balance_performance 19.9 153.3 201.0 OpenBenchmarking.org Watts, Fewer Is Better Timed Linux Kernel Compilation 6.8 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.4 77.7 84.5 amd-pstate-epp powersave, EPP power 46.9 76.4 84.5 amd-pstate-epp powersave, EPP performance 46.6 76.0 84.5 acpi-cpufreq schedutil 45.4 75.8 83.4 amd-pstate-epp powersave, EPP balance_performance 45.9 75.8 84.0 acpi-cpufreq performance 45.3 75.5 83.3 amd-pstate-epp powersave, EPP balance_power 45.4 75.2 84.4 OpenBenchmarking.org Celsius, Fewer Is Better Timed Linux Kernel Compilation 6.8 CPU Temperature Monitor 20 40 60 80 100
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 44100 - Buffer Size: 512 amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil acpi-cpufreq performance 2 4 6 8 10 SE +/- 0.044680, N = 3 SE +/- 0.107307, N = 3 SE +/- 0.084235, N = 6 SE +/- 0.067417, N = 15 SE +/- 0.063960, N = 15 SE +/- 0.034391, N = 3 SE +/- 0.074428, N = 3 8.110456 8.221870 8.267507 8.331987 8.427568 8.730047 8.918763 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 44100 - Buffer Size: 512 amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil acpi-cpufreq performance 0.0131 0.0262 0.0393 0.0524 0.0655 0.052 0.052 0.053 0.053 0.054 0.057 0.058 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4545 5688 amd-pstate-epp performance, EPP performance 600 4634 5660 amd-pstate-epp powersave, EPP balance_power 600 4652 5455 amd-pstate-epp powersave, EPP power 600 4653 5724 amd-pstate-epp powersave, EPP balance_performance 600 4794 5715 acpi-cpufreq performance 4300 4930 5706 acpi-cpufreq schedutil 4300 4983 5603 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 18.7 157.0 200.4 amd-pstate-epp powersave, EPP balance_performance 19.5 156.9 200.4 amd-pstate-epp powersave, EPP balance_power 19.7 156.4 200.4 amd-pstate-epp powersave, EPP performance 18.0 156.2 200.3 amd-pstate-epp powersave, EPP power 22.0 156.2 200.4 acpi-cpufreq performance 16.1 154.5 200.3 acpi-cpufreq schedutil 18.3 154.3 200.4 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 48.1 83.6 95.5 amd-pstate-epp powersave, EPP power 47.3 82.4 95.0 amd-pstate-epp powersave, EPP performance 46.4 82.2 92.3 amd-pstate-epp powersave, EPP balance_performance 46.4 82.2 94.0 amd-pstate-epp powersave, EPP balance_power 45.9 81.1 92.4 acpi-cpufreq schedutil 46.4 78.2 90.1 acpi-cpufreq performance 46.1 77.3 86.3 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100
Blender Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: BMW27 - Compute: CPU-Only amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil 11 22 33 44 55 SE +/- 0.07, N = 3 SE +/- 0.10, N = 3 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.16, N = 3 48.07 47.03 46.84 46.80 46.77 46.71 46.49 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4454 5557 amd-pstate-epp powersave, EPP power 600 4480 5635 amd-pstate-epp powersave, EPP balance_power 600 4494 5508 amd-pstate-epp powersave, EPP performance 600 4541 5652 acpi-cpufreq performance 4300 4647 5726 acpi-cpufreq schedutil 4300 4649 5563 amd-pstate-epp performance, EPP performance 600 4702 5713 OpenBenchmarking.org Megahertz, More Is Better Blender 4.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.4 184.8 201.0 amd-pstate-epp powersave, EPP performance 19.6 184.1 201.0 amd-pstate-epp powersave, EPP power 22.7 183.8 201.0 acpi-cpufreq schedutil 18.5 183.8 201.2 amd-pstate-epp powersave, EPP balance_power 19.7 183.8 201.2 acpi-cpufreq performance 20.2 183.7 201.4 amd-pstate-epp powersave, EPP balance_performance 19.3 183.6 201.2 OpenBenchmarking.org Watts, Fewer Is Better Blender 4.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 43.9 76.4 79.9 amd-pstate-epp powersave, EPP power 42.6 75.2 78.8 amd-pstate-epp powersave, EPP balance_performance 41.8 75.0 78.8 amd-pstate-epp powersave, EPP performance 41.0 74.3 78.3 acpi-cpufreq performance 41.6 74.0 77.9 acpi-cpufreq schedutil 40.9 73.9 77.9 amd-pstate-epp powersave, EPP balance_power 40.9 73.8 78.0 OpenBenchmarking.org Celsius, Fewer Is Better Blender 4.2 CPU Temperature Monitor 20 40 60 80 100
Selenium Result
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Speedometer 3.0 - Browser: Firefox amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 8 16 24 32 40 SE +/- 0.23, N = 3 SE +/- 0.12, N = 3 SE +/- 0.17, N = 3 SE +/- 0.20, N = 3 SE +/- 0.09, N = 3 SE +/- 0.24, N = 3 SE +/- 0.35, N = 3 30.4 31.0 31.0 31.8 31.9 32.2 32.8 1. firefox 126.0.1 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Selenium Benchmark: Speedometer 3.0 - Browser: Firefox amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance 0.1728 0.3456 0.5184 0.6912 0.864 0.565 0.575 0.580 0.587 0.606 0.606 0.768 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4418 5717 amd-pstate-epp powersave, EPP balance_power 600 4527 5747 amd-pstate-epp powersave, EPP balance_performance 600 4811 5728 acpi-cpufreq performance 4300 5011 5731 amd-pstate-epp powersave, EPP performance 600 5018 5724 acpi-cpufreq schedutil 3031 5093 5710 amd-pstate-epp performance, EPP performance 600 5535 5711 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 17.2 58.1 83.3 amd-pstate-epp powersave, EPP performance 18.7 55.5 83.6 amd-pstate-epp powersave, EPP power 22.7 52.9 87.7 acpi-cpufreq schedutil 19.6 52.8 82.3 amd-pstate-epp powersave, EPP balance_performance 19.5 52.5 85.5 amd-pstate-epp powersave, EPP balance_power 19.8 51.1 87.1 acpi-cpufreq performance 12.3 41.5 88.3 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 44.3 56.2 63.0 amd-pstate-epp powersave, EPP performance 42.3 54.5 61.9 amd-pstate-epp powersave, EPP balance_performance 43.6 54.4 62.4 acpi-cpufreq schedutil 42.4 53.5 60.8 amd-pstate-epp powersave, EPP power 42.9 52.9 60.9 amd-pstate-epp powersave, EPP balance_power 42.3 52.5 61.5 acpi-cpufreq performance 38.9 47.7 62.4 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100
Xonotic This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game for this open-source first person shooter title. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.6 Resolution: 3840 x 2160 - Effects Quality: Ultra amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance 200 400 600 800 1000 SE +/- 1.48, N = 3 SE +/- 1.21, N = 3 SE +/- 1.02, N = 3 SE +/- 1.30, N = 3 SE +/- 0.37, N = 3 SE +/- 2.09, N = 3 SE +/- 1.39, N = 3 937.21 938.13 941.38 941.92 943.95 944.53 946.95 MIN: 474 / MAX: 1348 MIN: 467 / MAX: 1358 MIN: 466 / MAX: 1346 MIN: 473 / MAX: 1353 MIN: 470 / MAX: 1360 MIN: 466 / MAX: 1354 MIN: 470 / MAX: 1362 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Xonotic 0.8.6 Resolution: 3840 x 2160 - Effects Quality: Ultra amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance 4 8 12 16 20 13.75 13.76 13.83 13.91 14.35 14.59 14.75 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5327 5721 acpi-cpufreq performance 4300 5366 5722 amd-pstate-epp powersave, EPP power 600 5381 5731 amd-pstate-epp powersave, EPP balance_power 600 5418 5722 acpi-cpufreq schedutil 4300 5419 5690 amd-pstate-epp powersave, EPP performance 600 5432 5727 amd-pstate-epp performance, EPP performance 600 5528 5720 OpenBenchmarking.org Megahertz, More Is Better Xonotic 0.8.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP power 22.9 68.5 80.5 amd-pstate-epp performance, EPP performance 21.6 68.3 79.3 amd-pstate-epp powersave, EPP performance 19.9 68.1 79.0 amd-pstate-epp powersave, EPP balance_power 20.3 67.9 80.8 acpi-cpufreq schedutil 19.5 65.4 75.4 acpi-cpufreq performance 20.1 64.6 76.8 amd-pstate-epp powersave, EPP balance_performance 19.9 64.2 76.9 OpenBenchmarking.org Watts, Fewer Is Better Xonotic 0.8.6 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 40.3 62.5 65.6 amd-pstate-epp powersave, EPP performance 39.1 60.6 64.5 amd-pstate-epp powersave, EPP power 38.8 60.0 65.3 amd-pstate-epp powersave, EPP balance_power 37.6 59.7 64.4 acpi-cpufreq schedutil 36.9 59.4 65.9 acpi-cpufreq performance 38.1 59.3 63.8 amd-pstate-epp powersave, EPP balance_performance 37.4 58.2 64.4 OpenBenchmarking.org Celsius, Fewer Is Better Xonotic 0.8.6 CPU Temperature Monitor 20 40 60 80 100
uvg266 Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.8.0 Video Input: Bosphorus 4K - Video Preset: Slow acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 13.80 13.82 13.86 13.87 13.87 13.91 13.92 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.8.0 Video Input: Bosphorus 4K - Video Preset: Slow amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0171 0.0342 0.0513 0.0684 0.0855 0.076 0.076 0.076 0.076 0.076 0.076 0.076 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4519 5726 amd-pstate-epp powersave, EPP balance_power 600 4520 5464 amd-pstate-epp powersave, EPP balance_performance 600 4539 5490 amd-pstate-epp powersave, EPP performance 600 4557 5379 amd-pstate-epp performance, EPP performance 600 4632 5710 acpi-cpufreq schedutil 4300 4636 5653 acpi-cpufreq performance 4300 4640 5421 OpenBenchmarking.org Megahertz, More Is Better uvg266 0.8.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.7 183.1 200.8 amd-pstate-epp powersave, EPP performance 18.0 183.0 200.6 acpi-cpufreq schedutil 19.2 183.0 201.0 amd-pstate-epp powersave, EPP balance_performance 19.6 183.0 200.7 amd-pstate-epp powersave, EPP power 21.9 182.8 201.4 amd-pstate-epp powersave, EPP balance_power 19.6 182.8 200.6 acpi-cpufreq performance 19.4 182.1 201.0 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.8.0 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 44.4 78.2 81.9 amd-pstate-epp performance, EPP performance 43.9 77.3 81.4 amd-pstate-epp powersave, EPP performance 42.9 77.2 81.0 amd-pstate-epp powersave, EPP balance_performance 43.1 77.0 80.8 acpi-cpufreq performance 43.3 77.0 80.8 acpi-cpufreq schedutil 42.9 76.9 80.9 amd-pstate-epp powersave, EPP balance_power 43.1 76.5 80.3 OpenBenchmarking.org Celsius, Fewer Is Better uvg266 0.8.0 CPU Temperature Monitor 20 40 60 80 100
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 480000 - Buffer Size: 1024 amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 3 6 9 12 15 SE +/- 0.051116, N = 3 SE +/- 0.112940, N = 12 SE +/- 0.080017, N = 15 SE +/- 0.115298, N = 3 SE +/- 0.024025, N = 3 SE +/- 0.056217, N = 3 SE +/- 0.045212, N = 3 8.201474 8.212730 8.265068 8.444777 8.564042 9.028189 9.028488 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 480000 - Buffer Size: 1024 amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 0.0133 0.0266 0.0399 0.0532 0.0665 0.052 0.053 0.053 0.055 0.056 0.058 0.059 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4451 5707 amd-pstate-epp powersave, EPP balance_power 600 4582 5616 amd-pstate-epp powersave, EPP balance_performance 600 4666 5706 amd-pstate-epp performance, EPP performance 600 4687 5696 amd-pstate-epp powersave, EPP performance 600 4807 5718 acpi-cpufreq schedutil 3029 4841 5456 acpi-cpufreq performance 4300 4921 5587 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.1 156.7 200.3 amd-pstate-epp powersave, EPP performance 18.1 156.0 200.4 amd-pstate-epp powersave, EPP balance_performance 19.5 155.6 200.5 acpi-cpufreq performance 19.6 153.9 200.4 amd-pstate-epp powersave, EPP power 22.4 153.6 200.4 acpi-cpufreq schedutil 18.1 153.6 200.3 amd-pstate-epp powersave, EPP balance_power 20.2 153.0 200.3 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.3 82.9 95.6 amd-pstate-epp powersave, EPP performance 44.6 82.4 95.9 amd-pstate-epp powersave, EPP balance_performance 45.1 82.3 95.9 amd-pstate-epp powersave, EPP power 46.1 80.6 90.9 amd-pstate-epp powersave, EPP balance_power 44.4 77.7 89.9 acpi-cpufreq performance 44.5 75.5 87.9 acpi-cpufreq schedutil 44.4 75.5 83.9 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 480000 - Buffer Size: 512 amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil acpi-cpufreq performance 2 4 6 8 10 SE +/- 0.077369, N = 5 SE +/- 0.086873, N = 4 SE +/- 0.061519, N = 15 SE +/- 0.074319, N = 3 SE +/- 0.051547, N = 3 SE +/- 0.068927, N = 3 SE +/- 0.010784, N = 3 7.865709 7.995940 8.070138 8.183842 8.389622 8.491271 8.649094 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 480000 - Buffer Size: 512 amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil acpi-cpufreq performance 0.0124 0.0248 0.0372 0.0496 0.062 0.050 0.051 0.051 0.052 0.054 0.055 0.055 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4542 5707 amd-pstate-epp performance, EPP performance 600 4609 5685 amd-pstate-epp powersave, EPP balance_power 600 4684 5715 amd-pstate-epp powersave, EPP performance 600 4702 5453 amd-pstate-epp powersave, EPP balance_performance 600 4795 5715 acpi-cpufreq performance 4300 4921 5607 acpi-cpufreq schedutil 4300 4987 5692 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.1 158.1 200.5 acpi-cpufreq performance 19.1 157.6 200.3 amd-pstate-epp powersave, EPP balance_performance 18.5 157.3 200.4 amd-pstate-epp powersave, EPP balance_power 19.9 156.8 200.3 amd-pstate-epp powersave, EPP performance 15.1 156.7 200.4 amd-pstate-epp powersave, EPP power 22.6 156.2 200.2 acpi-cpufreq schedutil 18.0 153.9 200.3 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.9 81.4 90.9 amd-pstate-epp powersave, EPP balance_performance 45.3 81.2 94.0 amd-pstate-epp powersave, EPP performance 45.0 81.1 91.1 amd-pstate-epp powersave, EPP balance_power 44.9 79.5 90.4 amd-pstate-epp powersave, EPP power 46.0 78.5 89.5 acpi-cpufreq schedutil 44.3 75.6 83.9 acpi-cpufreq performance 44.5 75.1 82.6 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100
RawTherapee RawTherapee is a cross-platform, open-source multi-threaded RAW image processing program. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better RawTherapee Total Benchmark Time amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance acpi-cpufreq performance 8 16 24 32 40 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 34.30 34.19 34.18 34.04 33.98 33.96 33.94 1. RawTherapee, version 5.10, command line. CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 5138 5723 amd-pstate-epp powersave, EPP balance_performance 600 5164 5693 amd-pstate-epp performance, EPP performance 600 5182 5707 amd-pstate-epp powersave, EPP performance 600 5190 5715 amd-pstate-epp powersave, EPP power 3013 5205 5721 acpi-cpufreq schedutil 3038 5353 5722 acpi-cpufreq performance 4300 5374 5719 OpenBenchmarking.org Megahertz, More Is Better RawTherapee CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.3 98.5 134.7 acpi-cpufreq performance 20.0 97.7 142.8 amd-pstate-epp powersave, EPP balance_power 20.4 97.2 139.7 amd-pstate-epp powersave, EPP power 22.8 97.1 133.8 amd-pstate-epp powersave, EPP performance 20.3 97.0 135.6 acpi-cpufreq schedutil 18.5 96.2 143.0 amd-pstate-epp powersave, EPP balance_performance 19.8 95.5 134.4 OpenBenchmarking.org Watts, Fewer Is Better RawTherapee CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.8 67.8 72.8 acpi-cpufreq performance 39.8 65.0 70.4 amd-pstate-epp powersave, EPP power 40.5 64.5 70.0 acpi-cpufreq schedutil 38.3 64.4 69.4 amd-pstate-epp powersave, EPP balance_performance 39.3 63.8 69.8 amd-pstate-epp powersave, EPP performance 38.8 63.5 68.8 amd-pstate-epp powersave, EPP balance_power 38.8 63.1 68.5 OpenBenchmarking.org Celsius, Fewer Is Better RawTherapee CPU Temperature Monitor 20 40 60 80 100
Radiance Benchmark This is a benchmark of NREL Radiance, a synthetic imaging system that is open-source and developed by the Lawrence Berkeley National Laboratory in California. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Radiance Benchmark 5.0 Test: SMP Parallel acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power 20 40 60 80 100 97.62 95.54 94.63 94.17 94.07 92.94 91.09 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 5320 5624 amd-pstate-epp performance, EPP performance 600 5410 5722 amd-pstate-epp powersave, EPP balance_power 600 5410 5638 amd-pstate-epp powersave, EPP performance 600 5436 5727 amd-pstate-epp powersave, EPP balance_performance 3583 5441 5716 acpi-cpufreq performance 4300 5442 5716 acpi-cpufreq schedutil 4300 5446 5693 OpenBenchmarking.org Megahertz, More Is Better Radiance Benchmark 5.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP power 23.0 97.9 110.5 amd-pstate-epp powersave, EPP balance_power 20.3 97.0 109.7 amd-pstate-epp powersave, EPP balance_performance 19.8 96.5 108.7 amd-pstate-epp performance, EPP performance 21.8 96.4 109.0 acpi-cpufreq performance 19.9 95.8 109.4 amd-pstate-epp powersave, EPP performance 17.8 95.7 109.2 acpi-cpufreq schedutil 18.9 93.9 107.4 OpenBenchmarking.org Watts, Fewer Is Better Radiance Benchmark 5.0 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.5 73.8 77.4 amd-pstate-epp powersave, EPP power 40.1 72.2 75.8 amd-pstate-epp powersave, EPP balance_performance 39.8 71.6 75.5 amd-pstate-epp powersave, EPP performance 38.8 70.8 74.8 amd-pstate-epp powersave, EPP balance_power 37.8 70.5 74.8 acpi-cpufreq schedutil 38.9 70.0 74.3 acpi-cpufreq performance 39.3 68.9 71.8 OpenBenchmarking.org Celsius, Fewer Is Better Radiance Benchmark 5.0 CPU Temperature Monitor 20 40 60 80 100
Whisperfile Result
OpenBenchmarking.org Seconds, Fewer Is Better Whisperfile 20Aug24 Model Size: Tiny acpi-cpufreq performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance 7 14 21 28 35 SE +/- 0.03, N = 3 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 SE +/- 0.02, N = 3 29.43 29.25 29.24 29.10 29.08 29.02 28.92 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5019 5715 amd-pstate-epp powersave, EPP balance_performance 600 5064 5711 amd-pstate-epp powersave, EPP power 600 5090 5691 amd-pstate-epp powersave, EPP balance_power 600 5189 5686 amd-pstate-epp performance, EPP performance 600 5205 5714 acpi-cpufreq performance 4300 5277 5683 acpi-cpufreq schedutil 4300 5281 5651 OpenBenchmarking.org Megahertz, More Is Better Whisperfile 20Aug24 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.4 123.9 160.0 amd-pstate-epp powersave, EPP balance_power 19.5 123.0 160.5 amd-pstate-epp powersave, EPP power 22.2 121.3 160.6 amd-pstate-epp powersave, EPP performance 18.9 120.8 159.4 acpi-cpufreq performance 20.3 120.4 157.1 amd-pstate-epp powersave, EPP balance_performance 19.5 120.2 160.3 acpi-cpufreq schedutil 19.5 117.6 158.7 OpenBenchmarking.org Watts, Fewer Is Better Whisperfile 20Aug24 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.9 65.8 71.6 amd-pstate-epp powersave, EPP balance_performance 42.1 65.7 73.1 amd-pstate-epp powersave, EPP power 41.3 65.0 71.9 amd-pstate-epp powersave, EPP balance_power 41.0 64.7 71.8 acpi-cpufreq performance 40.8 64.5 70.3 amd-pstate-epp powersave, EPP performance 40.9 63.9 70.9 acpi-cpufreq schedutil 41.4 63.3 69.8 OpenBenchmarking.org Celsius, Fewer Is Better Whisperfile 20Aug24 CPU Temperature Monitor 20 40 60 80 100
Opus Codec Encoding Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus five times. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.4 WAV To Opus Encode amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance 4 8 12 16 20 SE +/- 0.03, N = 5 SE +/- 0.03, N = 5 SE +/- 0.06, N = 5 SE +/- 0.03, N = 5 SE +/- 0.11, N = 5 SE +/- 0.12, N = 5 SE +/- 0.13, N = 5 17.85 17.69 17.55 17.39 17.36 17.33 17.25 1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4914 5707 amd-pstate-epp powersave, EPP balance_performance 600 4934 5707 amd-pstate-epp powersave, EPP power 600 4976 5705 amd-pstate-epp powersave, EPP performance 600 5211 5718 amd-pstate-epp performance, EPP performance 600 5226 5717 acpi-cpufreq performance 4300 5281 5760 acpi-cpufreq schedutil 3000 5310 5700 OpenBenchmarking.org Megahertz, More Is Better Opus Codec Encoding 1.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.7 49.7 56.9 amd-pstate-epp powersave, EPP power 22.4 47.9 57.4 acpi-cpufreq performance 18.4 47.2 56.5 acpi-cpufreq schedutil 18.6 46.9 54.5 amd-pstate-epp powersave, EPP performance 19.5 46.7 56.0 amd-pstate-epp powersave, EPP balance_power 20.1 44.7 55.8 amd-pstate-epp powersave, EPP balance_performance 19.7 43.6 55.0 OpenBenchmarking.org Watts, Fewer Is Better Opus Codec Encoding 1.4 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.9 58.4 61.9 amd-pstate-epp powersave, EPP power 41.0 56.5 60.4 acpi-cpufreq schedutil 39.0 55.8 58.6 amd-pstate-epp powersave, EPP performance 38.8 54.8 58.8 acpi-cpufreq performance 39.1 54.7 60.4 amd-pstate-epp powersave, EPP balance_power 38.8 50.8 58.6 amd-pstate-epp powersave, EPP balance_performance 39.5 50.6 59.3 OpenBenchmarking.org Celsius, Fewer Is Better Opus Codec Encoding 1.4 CPU Temperature Monitor 20 40 60 80 100
VOSK Speech Recognition Toolkit VOSK is an open-source offline speech recognition API/toolkit. VOSK supports speech recognition in 17 languages and has a variety of models available and interfaces for different programming languages. This test profile times the speech-to-text process for a roughly three minute audio recording. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better VOSK Speech Recognition Toolkit 0.3.21 amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance 3 6 9 12 15 SE +/- 0.04, N = 4 SE +/- 0.10, N = 14 SE +/- 0.11, N = 15 SE +/- 0.11, N = 4 SE +/- 0.03, N = 4 SE +/- 0.06, N = 5 SE +/- 0.09, N = 5 11.65 11.62 11.58 11.55 11.52 11.35 11.28 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4629 5718 amd-pstate-epp powersave, EPP balance_performance 600 4704 5723 amd-pstate-epp powersave, EPP balance_power 600 4814 5725 amd-pstate-epp powersave, EPP performance 600 4853 5726 amd-pstate-epp performance, EPP performance 600 4862 5714 acpi-cpufreq schedutil 4300 5205 5680 acpi-cpufreq performance 4300 5236 5718 OpenBenchmarking.org Megahertz, More Is Better VOSK Speech Recognition Toolkit 0.3.21 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.0 48.4 60.7 acpi-cpufreq performance 19.3 47.2 58.6 amd-pstate-epp powersave, EPP performance 19.6 46.2 57.6 amd-pstate-epp powersave, EPP power 22.6 46.0 58.5 acpi-cpufreq schedutil 17.9 45.7 58.5 amd-pstate-epp powersave, EPP balance_power 18.5 44.9 57.4 amd-pstate-epp powersave, EPP balance_performance 18.9 44.0 56.4 OpenBenchmarking.org Watts, Fewer Is Better VOSK Speech Recognition Toolkit 0.3.21 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 43.0 60.6 64.5 acpi-cpufreq performance 40.8 58.1 62.8 amd-pstate-epp powersave, EPP performance 40.8 57.9 62.9 acpi-cpufreq schedutil 40.8 57.6 64.6 amd-pstate-epp powersave, EPP power 42.0 56.2 63.4 amd-pstate-epp powersave, EPP balance_performance 41.4 55.2 63.0 amd-pstate-epp powersave, EPP balance_power 40.4 54.3 61.8 OpenBenchmarking.org Celsius, Fewer Is Better VOSK Speech Recognition Toolkit 0.3.21 CPU Temperature Monitor 20 40 60 80 100
VVenC VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Faster acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance 5 10 15 20 25 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 22.38 22.40 22.42 22.42 22.43 22.46 22.49 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Faster amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 0.0331 0.0662 0.0993 0.1324 0.1655 0.146 0.146 0.146 0.147 0.147 0.147 0.147 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4592 5638 amd-pstate-epp powersave, EPP balance_performance 600 4680 5520 amd-pstate-epp powersave, EPP balance_power 600 4709 5634 amd-pstate-epp powersave, EPP power 600 4740 5711 amd-pstate-epp performance, EPP performance 600 4759 5704 acpi-cpufreq performance 4300 4888 5594 acpi-cpufreq schedutil 4300 4926 5638 OpenBenchmarking.org Megahertz, More Is Better VVenC 1.11 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.9 153.5 198.4 amd-pstate-epp powersave, EPP balance_performance 18.2 153.4 198.6 amd-pstate-epp powersave, EPP power 22.3 153.2 196.8 amd-pstate-epp powersave, EPP balance_power 19.3 153.0 198.4 amd-pstate-epp powersave, EPP performance 19.6 153.0 198.1 acpi-cpufreq performance 19.3 152.9 198.2 acpi-cpufreq schedutil 19.3 152.7 199.4 OpenBenchmarking.org Watts, Fewer Is Better VVenC 1.11 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 45.3 75.7 80.6 amd-pstate-epp powersave, EPP power 45.4 75.7 80.4 amd-pstate-epp powersave, EPP balance_performance 44.5 75.0 79.4 acpi-cpufreq schedutil 44.4 74.9 79.9 amd-pstate-epp powersave, EPP performance 44.6 74.9 79.5 acpi-cpufreq performance 44.1 74.7 79.8 amd-pstate-epp powersave, EPP balance_power 44.5 74.6 79.4 OpenBenchmarking.org Celsius, Fewer Is Better VVenC 1.11 CPU Temperature Monitor 20 40 60 80 100
GPUScore: Breaking Limit Result
OpenBenchmarking.org FPS, More Is Better GPUScore: Breaking Limit 1.0 Resolution: 3840 x 2160 - Ray-Tracing: Off amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil 20 40 60 80 100 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 85.97 86.10 86.14 86.27 86.29 86.30 86.38 MIN: 67.72 / MAX: 215.09 MIN: 67.37 / MAX: 222.13 MIN: 67.69 / MAX: 221.08 MIN: 67.72 / MAX: 222.74 MIN: 67.69 / MAX: 226.12 MIN: 68.04 / MAX: 224.43 MIN: 67.58 / MAX: 226.63 FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better GPUScore: Breaking Limit 1.0 Resolution: 3840 x 2160 - Ray-Tracing: Off amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power 0.5416 1.0832 1.6248 2.1664 2.708 1.716 1.898 1.932 2.202 2.336 2.375 2.407 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 3209 5727 amd-pstate-epp powersave, EPP balance_power 600 3603 5725 amd-pstate-epp powersave, EPP balance_performance 600 4187 5723 acpi-cpufreq schedutil 3029 4280 5720 amd-pstate-epp powersave, EPP performance 600 5153 5730 amd-pstate-epp performance, EPP performance 600 5395 5717 acpi-cpufreq performance 4300 5461 5739 OpenBenchmarking.org Megahertz, More Is Better GPUScore: Breaking Limit 1.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.7 50.2 61.6 acpi-cpufreq performance 20.1 45.5 61.9 amd-pstate-epp powersave, EPP performance 20.0 44.5 61.5 acpi-cpufreq schedutil 19.2 39.2 58.9 amd-pstate-epp powersave, EPP balance_performance 18.5 37.0 57.4 amd-pstate-epp powersave, EPP balance_power 17.7 36.3 63.3 amd-pstate-epp powersave, EPP power 11.9 35.8 63.1 OpenBenchmarking.org Watts, Fewer Is Better GPUScore: Breaking Limit 1.0 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 42.0 47.6 54.3 amd-pstate-epp powersave, EPP performance 41.3 45.6 54.0 acpi-cpufreq performance 40.6 45.0 53.3 amd-pstate-epp powersave, EPP power 41.3 44.9 54.8 amd-pstate-epp powersave, EPP balance_power 40.3 44.4 54.0 amd-pstate-epp powersave, EPP balance_performance 41.4 44.2 51.0 acpi-cpufreq schedutil 39.8 43.6 52.0 OpenBenchmarking.org Celsius, Fewer Is Better GPUScore: Breaking Limit 1.0 CPU Temperature Monitor 15 30 45 60 75
WebP Image Encode Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.4 Encode Settings: Quality 100, Lossless, Highest Compression amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance 0.2115 0.423 0.6345 0.846 1.0575 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.91 0.92 0.92 0.92 0.93 0.93 0.94 1. (CC) gcc options: -fvisibility=hidden -O2 -lm MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP Image Encode 1.4 Encode Settings: Quality 100, Lossless, Highest Compression amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance 0.0036 0.0072 0.0108 0.0144 0.018 0.016 0.016 0.016 0.016 0.016 0.016 0.016 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 5151 5726 amd-pstate-epp powersave, EPP performance 600 5169 5729 amd-pstate-epp powersave, EPP balance_power 600 5220 5727 amd-pstate-epp powersave, EPP power 600 5258 5725 acpi-cpufreq schedutil 3000 5339 5722 amd-pstate-epp performance, EPP performance 600 5411 5724 acpi-cpufreq performance 4300 5418 5724 OpenBenchmarking.org Megahertz, More Is Better WebP Image Encode 1.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.8 59.7 70.4 acpi-cpufreq performance 19.9 59.4 72.5 acpi-cpufreq schedutil 18.7 58.9 71.2 amd-pstate-epp powersave, EPP performance 19.2 58.9 69.9 amd-pstate-epp powersave, EPP balance_power 19.7 58.5 70.5 amd-pstate-epp powersave, EPP power 22.4 58.3 70.2 amd-pstate-epp powersave, EPP balance_performance 19.5 58.1 70.2 OpenBenchmarking.org Watts, Fewer Is Better WebP Image Encode 1.4 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 40.3 62.3 69.5 acpi-cpufreq performance 38.1 61.2 69.8 amd-pstate-epp powersave, EPP power 39.4 60.8 67.3 acpi-cpufreq schedutil 37.3 60.7 68.0 amd-pstate-epp powersave, EPP performance 38.6 60.3 66.5 amd-pstate-epp powersave, EPP balance_performance 38.0 60.2 66.3 amd-pstate-epp powersave, EPP balance_power 37.8 59.7 66.3 OpenBenchmarking.org Celsius, Fewer Is Better WebP Image Encode 1.4 CPU Temperature Monitor 20 40 60 80 100
VP9 libvpx Encoding This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 0 - Input: Bosphorus 1080p amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance acpi-cpufreq performance 6 12 18 24 30 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 SE +/- 0.02, N = 3 SE +/- 0.23, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.22, N = 3 23.32 23.40 23.62 23.75 23.80 23.89 24.25 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 0 - Input: Bosphorus 1080p amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance acpi-cpufreq schedutil 0.0446 0.0892 0.1338 0.1784 0.223 0.193 0.194 0.195 0.196 0.196 0.196 0.198 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 5044 5716 amd-pstate-epp powersave, EPP balance_power 600 5068 5623 amd-pstate-epp powersave, EPP balance_performance 600 5076 5722 amd-pstate-epp powersave, EPP power 600 5113 5600 amd-pstate-epp performance, EPP performance 600 5260 5709 acpi-cpufreq performance 4300 5303 5587 acpi-cpufreq schedutil 4300 5310 5557 OpenBenchmarking.org Megahertz, More Is Better VP9 libvpx Encoding 1.13 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.6 123.6 147.0 amd-pstate-epp performance, EPP performance 20.2 123.1 143.9 amd-pstate-epp powersave, EPP performance 19.7 121.3 142.9 amd-pstate-epp powersave, EPP power 22.2 121.2 142.7 amd-pstate-epp powersave, EPP balance_power 19.6 120.8 144.9 acpi-cpufreq schedutil 19.1 120.0 144.7 amd-pstate-epp powersave, EPP balance_performance 19.2 119.4 141.5 OpenBenchmarking.org Watts, Fewer Is Better VP9 libvpx Encoding 1.13 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 43.8 75.6 80.3 amd-pstate-epp powersave, EPP power 44.1 75.5 80.3 amd-pstate-epp powersave, EPP performance 42.9 74.7 79.3 amd-pstate-epp powersave, EPP balance_power 42.5 74.2 79.3 amd-pstate-epp powersave, EPP balance_performance 43.0 74.1 79.0 acpi-cpufreq performance 42.8 72.3 77.9 acpi-cpufreq schedutil 42.9 71.6 76.9 OpenBenchmarking.org Celsius, Fewer Is Better VP9 libvpx Encoding 1.13 CPU Temperature Monitor 20 40 60 80 100
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 8 - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 20 40 60 80 100 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 SE +/- 0.11, N = 3 SE +/- 0.05, N = 3 SE +/- 0.21, N = 3 SE +/- 0.13, N = 3 SE +/- 0.14, N = 3 97.82 98.39 99.65 99.96 99.98 100.27 100.32 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 8 - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power 0.1373 0.2746 0.4119 0.5492 0.6865 0.592 0.595 0.601 0.601 0.602 0.606 0.610 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4642 5477 amd-pstate-epp powersave, EPP balance_performance 600 4663 5724 amd-pstate-epp powersave, EPP balance_power 600 4683 5626 amd-pstate-epp powersave, EPP power 600 4718 5598 amd-pstate-epp performance, EPP performance 600 4891 5695 acpi-cpufreq performance 4273 4896 5680 acpi-cpufreq schedutil 4300 4918 5587 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.3 166.6 200.2 amd-pstate-epp powersave, EPP performance 18.9 166.2 200.0 amd-pstate-epp powersave, EPP power 22.2 165.8 199.9 acpi-cpufreq performance 19.6 165.5 200.4 acpi-cpufreq schedutil 18.9 165.1 201.1 amd-pstate-epp powersave, EPP balance_performance 20.0 164.9 200.2 amd-pstate-epp powersave, EPP balance_power 19.9 164.5 200.4 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 46.3 76.7 82.4 acpi-cpufreq performance 45.3 76.7 82.5 amd-pstate-epp performance, EPP performance 46.3 76.6 82.5 acpi-cpufreq schedutil 44.8 76.4 82.4 amd-pstate-epp powersave, EPP balance_performance 45.6 76.3 84.0 amd-pstate-epp powersave, EPP performance 45.1 76.1 82.4 amd-pstate-epp powersave, EPP balance_power 45.1 75.5 81.6 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 2.2 CPU Temperature Monitor 20 40 60 80 100
Ogg Audio Encoding This test times how long it takes to encode a sample WAV file to Ogg format using the reference Xiph.org tools/libraries. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Ogg Audio Encoding 1.3.4 WAV To Ogg amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power 3 6 9 12 15 SE +/- 0.34, N = 15 SE +/- 0.02, N = 5 SE +/- 0.08, N = 5 SE +/- 0.02, N = 5 SE +/- 0.04, N = 5 SE +/- 0.08, N = 5 SE +/- 0.02, N = 5 11.29 11.03 10.96 10.92 10.83 10.82 10.70 1. (CC) gcc options: -O2 -ffast-math -fsigned-char CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4576 5663 amd-pstate-epp powersave, EPP performance 600 4676 5722 amd-pstate-epp powersave, EPP balance_power 600 4692 5691 amd-pstate-epp powersave, EPP power 3017 4929 5697 amd-pstate-epp performance, EPP performance 600 5138 5716 acpi-cpufreq schedutil 3000 5140 5567 acpi-cpufreq performance 4300 5185 5669 OpenBenchmarking.org Megahertz, More Is Better Ogg Audio Encoding 1.3.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.0 47.7 58.4 amd-pstate-epp powersave, EPP power 22.4 47.0 58.5 acpi-cpufreq performance 19.6 45.6 56.4 acpi-cpufreq schedutil 18.5 44.5 55.6 amd-pstate-epp powersave, EPP balance_power 19.3 43.7 57.0 amd-pstate-epp powersave, EPP performance 18.7 42.7 52.0 amd-pstate-epp powersave, EPP balance_performance 17.9 42.2 56.5 OpenBenchmarking.org Watts, Fewer Is Better Ogg Audio Encoding 1.3.4 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 40.9 60.5 64.3 amd-pstate-epp performance, EPP performance 42.0 58.9 61.9 acpi-cpufreq performance 39.4 58.0 64.4 acpi-cpufreq schedutil 39.6 57.2 62.3 amd-pstate-epp powersave, EPP balance_performance 40.0 53.9 63.8 amd-pstate-epp powersave, EPP balance_power 38.9 53.4 62.3 amd-pstate-epp powersave, EPP performance 39.0 52.6 55.1 OpenBenchmarking.org Celsius, Fewer Is Better Ogg Audio Encoding 1.3.4 CPU Temperature Monitor 20 40 60 80 100
VP9 libvpx Encoding This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 4K amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 6 12 18 24 30 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.23, N = 3 SE +/- 0.11, N = 3 SE +/- 0.05, N = 3 26.89 26.91 26.95 27.07 27.22 27.27 27.32 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 4K amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power 0.0585 0.117 0.1755 0.234 0.2925 0.254 0.255 0.256 0.256 0.256 0.257 0.260 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4976 5643 amd-pstate-epp powersave, EPP balance_power 600 4982 5712 amd-pstate-epp powersave, EPP power 600 5081 5660 amd-pstate-epp powersave, EPP balance_performance 600 5150 5690 acpi-cpufreq schedutil 3091 5289 5609 acpi-cpufreq performance 4300 5313 5574 amd-pstate-epp performance, EPP performance 600 5425 5712 OpenBenchmarking.org Megahertz, More Is Better VP9 libvpx Encoding 1.13 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.6 107.2 126.9 acpi-cpufreq performance 19.2 106.3 126.2 acpi-cpufreq schedutil 19.2 105.8 125.6 amd-pstate-epp powersave, EPP power 22.4 105.2 126.4 amd-pstate-epp powersave, EPP balance_power 19.4 105.0 127.2 amd-pstate-epp powersave, EPP balance_performance 18.4 104.9 125.4 amd-pstate-epp powersave, EPP performance 19.7 104.9 125.7 OpenBenchmarking.org Watts, Fewer Is Better VP9 libvpx Encoding 1.13 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 44.3 70.4 75.5 amd-pstate-epp powersave, EPP power 44.9 70.3 75.4 amd-pstate-epp powersave, EPP performance 43.6 69.1 73.8 amd-pstate-epp powersave, EPP balance_performance 43.8 69.0 74.0 amd-pstate-epp powersave, EPP balance_power 43.3 68.8 74.8 acpi-cpufreq schedutil 43.5 68.7 73.9 acpi-cpufreq performance 43.8 68.2 73.0 OpenBenchmarking.org Celsius, Fewer Is Better VP9 libvpx Encoding 1.13 CPU Temperature Monitor 20 40 60 80 100
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Crown amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power 8 16 24 32 40 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 36.15 36.16 36.17 36.21 36.22 36.25 36.28 MIN: 35.76 / MAX: 36.84 MIN: 35.81 / MAX: 36.9 MIN: 35.82 / MAX: 37.02 MIN: 35.85 / MAX: 36.95 MIN: 35.83 / MAX: 36.95 MIN: 35.88 / MAX: 37.1 MIN: 35.92 / MAX: 37.03 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Crown amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil 0.0486 0.0972 0.1458 0.1944 0.243 0.213 0.214 0.215 0.215 0.215 0.216 0.216 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4595 5151 amd-pstate-epp powersave, EPP balance_performance 600 4646 5128 amd-pstate-epp powersave, EPP power 600 4651 5474 amd-pstate-epp powersave, EPP performance 600 4685 5722 acpi-cpufreq schedutil 3000 4776 5668 amd-pstate-epp performance, EPP performance 600 4869 5697 acpi-cpufreq performance 4300 4912 5149 OpenBenchmarking.org Megahertz, More Is Better Embree 4.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP performance 18.1 170.2 201.0 amd-pstate-epp performance, EPP performance 21.0 169.0 201.0 amd-pstate-epp powersave, EPP power 22.4 168.8 201.0 acpi-cpufreq performance 19.6 168.3 201.0 acpi-cpufreq schedutil 18.9 168.0 201.6 amd-pstate-epp powersave, EPP balance_performance 19.6 167.9 201.2 amd-pstate-epp powersave, EPP balance_power 20.1 167.8 201.4 OpenBenchmarking.org Watts, Fewer Is Better Embree 4.3 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 46.1 74.2 78.6 amd-pstate-epp performance, EPP performance 45.5 74.1 78.5 amd-pstate-epp powersave, EPP performance 44.6 73.7 78.4 acpi-cpufreq performance 44.9 73.6 78.1 amd-pstate-epp powersave, EPP balance_performance 44.5 73.5 78.3 acpi-cpufreq schedutil 44.4 73.0 77.8 amd-pstate-epp powersave, EPP balance_power 44.8 73.0 77.8 OpenBenchmarking.org Celsius, Fewer Is Better Embree 4.3 CPU Temperature Monitor 20 40 60 80 100
x265 This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better x265 3.6 Video Input: Bosphorus 4K acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 9 18 27 36 45 SE +/- 0.26, N = 4 SE +/- 0.08, N = 4 SE +/- 0.07, N = 4 SE +/- 0.22, N = 4 SE +/- 0.15, N = 4 SE +/- 0.06, N = 4 SE +/- 0.10, N = 4 38.43 38.76 38.78 38.90 38.95 39.20 39.23 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better x265 3.6 Video Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 0.0612 0.1224 0.1836 0.2448 0.306 0.265 0.265 0.266 0.267 0.267 0.269 0.272 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 4631 5585 amd-pstate-epp powersave, EPP performance 600 4728 5707 amd-pstate-epp powersave, EPP power 600 4777 5680 amd-pstate-epp powersave, EPP balance_power 600 4805 5536 amd-pstate-epp performance, EPP performance 600 4935 5707 acpi-cpufreq performance 4300 5041 5480 acpi-cpufreq schedutil 4300 5064 5521 OpenBenchmarking.org Megahertz, More Is Better x265 3.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.9 146.7 185.9 acpi-cpufreq performance 20.0 146.2 186.7 amd-pstate-epp powersave, EPP balance_power 20.1 146.1 185.5 amd-pstate-epp powersave, EPP power 22.7 145.6 186.6 amd-pstate-epp powersave, EPP performance 19.7 145.6 184.6 acpi-cpufreq schedutil 19.3 144.8 182.8 amd-pstate-epp powersave, EPP balance_performance 19.6 142.9 185.9 OpenBenchmarking.org Watts, Fewer Is Better x265 3.6 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max acpi-cpufreq schedutil 44.6 75.3 81.8 amd-pstate-epp powersave, EPP power 45.6 75.3 81.0 acpi-cpufreq performance 44.4 75.1 80.8 amd-pstate-epp performance, EPP performance 45.4 74.9 80.3 amd-pstate-epp powersave, EPP performance 44.9 74.5 80.0 amd-pstate-epp powersave, EPP balance_performance 44.6 74.2 80.5 amd-pstate-epp powersave, EPP balance_power 44.5 73.7 79.6 OpenBenchmarking.org Celsius, Fewer Is Better x265 3.6 CPU Temperature Monitor 20 40 60 80 100
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power 9 18 27 36 45 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 40.95 40.96 40.98 41.01 41.06 41.08 41.10 MIN: 40.71 / MAX: 41.55 MIN: 40.76 / MAX: 41.64 MIN: 40.66 / MAX: 41.52 MIN: 40.72 / MAX: 41.61 MIN: 40.79 / MAX: 41.77 MIN: 40.75 / MAX: 41.7 MIN: 40.87 / MAX: 41.66 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance 0.0565 0.113 0.1695 0.226 0.2825 0.247 0.248 0.248 0.248 0.249 0.249 0.251 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4559 5322 amd-pstate-epp powersave, EPP balance_performance 600 4576 5285 amd-pstate-epp powersave, EPP performance 600 4737 5634 amd-pstate-epp powersave, EPP power 600 4771 5527 amd-pstate-epp performance, EPP performance 600 4863 5694 acpi-cpufreq schedutil 3000 4914 5586 acpi-cpufreq performance 4300 5048 5404 OpenBenchmarking.org Megahertz, More Is Better Embree 4.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.4 165.8 201.8 acpi-cpufreq performance 18.8 165.3 201.0 amd-pstate-epp powersave, EPP balance_power 20.0 165.2 206.6 amd-pstate-epp powersave, EPP power 22.6 165.1 201.0 acpi-cpufreq schedutil 19.3 164.8 200.8 amd-pstate-epp powersave, EPP performance 19.8 164.8 201.2 amd-pstate-epp powersave, EPP balance_performance 18.5 163.9 201.0 OpenBenchmarking.org Watts, Fewer Is Better Embree 4.3 CPU Power Consumption Monitor 60 120 180 240 300 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 46.8 73.8 78.5 amd-pstate-epp performance, EPP performance 46.4 73.6 78.1 amd-pstate-epp powersave, EPP performance 46.0 73.0 77.9 amd-pstate-epp powersave, EPP balance_performance 45.6 73.0 78.0 acpi-cpufreq performance 45.9 72.8 77.3 amd-pstate-epp powersave, EPP balance_power 45.6 72.6 77.3 acpi-cpufreq schedutil 45.4 72.6 77.3 OpenBenchmarking.org Celsius, Fewer Is Better Embree 4.3 CPU Temperature Monitor 20 40 60 80 100
FLAC Audio Encoding This test times how long it takes to encode a sample WAV file to FLAC audio format ten times using the --best preset settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.4 WAV To FLAC amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance 3 6 9 12 15 SE +/- 0.03, N = 5 SE +/- 0.02, N = 5 SE +/- 0.04, N = 5 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 SE +/- 0.03, N = 5 SE +/- 0.01, N = 5 11.40 11.39 10.82 10.76 10.75 10.75 10.72 1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4645 5717 amd-pstate-epp powersave, EPP balance_performance 600 4658 5733 amd-pstate-epp powersave, EPP performance 600 4739 5714 amd-pstate-epp powersave, EPP power 600 4861 5706 acpi-cpufreq schedutil 4300 5168 5679 acpi-cpufreq performance 4300 5179 5692 amd-pstate-epp performance, EPP performance 600 5564 5711 OpenBenchmarking.org Megahertz, More Is Better FLAC Audio Encoding 1.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.1 46.4 56.3 acpi-cpufreq performance 18.4 44.9 55.7 amd-pstate-epp powersave, EPP power 21.8 44.3 56.1 amd-pstate-epp powersave, EPP balance_power 19.6 43.9 56.1 amd-pstate-epp powersave, EPP performance 19.3 43.0 55.4 acpi-cpufreq schedutil 18.7 42.2 54.3 amd-pstate-epp powersave, EPP balance_performance 19.5 41.4 54.2 OpenBenchmarking.org Watts, Fewer Is Better FLAC Audio Encoding 1.4 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 44.1 59.8 64.1 amd-pstate-epp powersave, EPP power 42.6 57.9 62.6 acpi-cpufreq performance 40.8 57.0 61.8 amd-pstate-epp powersave, EPP balance_power 40.8 56.0 60.5 amd-pstate-epp powersave, EPP performance 41.5 55.9 61.1 acpi-cpufreq schedutil 41.0 54.4 61.5 amd-pstate-epp powersave, EPP balance_performance 41.8 53.4 60.5 OpenBenchmarking.org Celsius, Fewer Is Better FLAC Audio Encoding 1.4 CPU Temperature Monitor 20 40 60 80 100
uvg266 Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.8.0 Video Input: Bosphorus 4K - Video Preset: Ultra Fast acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance 10 20 30 40 50 SE +/- 0.07, N = 4 SE +/- 0.07, N = 4 SE +/- 0.05, N = 4 SE +/- 0.07, N = 4 SE +/- 0.04, N = 4 SE +/- 0.05, N = 4 SE +/- 0.03, N = 4 44.04 44.15 44.28 44.39 44.47 44.50 44.71 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.8.0 Video Input: Bosphorus 4K - Video Preset: Ultra Fast acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil 0.0646 0.1292 0.1938 0.2584 0.323 0.280 0.282 0.283 0.283 0.285 0.286 0.287 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4181 5541 amd-pstate-epp powersave, EPP balance_performance 600 4278 5423 amd-pstate-epp powersave, EPP balance_power 600 4279 5712 amd-pstate-epp powersave, EPP performance 600 4281 5723 amd-pstate-epp performance, EPP performance 600 4345 5697 acpi-cpufreq schedutil 4300 4595 5441 acpi-cpufreq performance 4300 4608 5683 OpenBenchmarking.org Megahertz, More Is Better uvg266 0.8.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.6 157.6 200.6 amd-pstate-epp powersave, EPP performance 19.7 157.6 200.4 amd-pstate-epp powersave, EPP balance_power 20.0 156.2 201.0 amd-pstate-epp performance, EPP performance 20.2 156.2 200.6 amd-pstate-epp powersave, EPP power 22.3 156.1 201.0 amd-pstate-epp powersave, EPP balance_performance 19.6 155.6 200.8 acpi-cpufreq schedutil 19.6 155.1 200.6 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.8.0 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 46.9 75.0 80.5 amd-pstate-epp performance, EPP performance 46.4 74.0 79.4 amd-pstate-epp powersave, EPP performance 45.5 74.0 79.5 amd-pstate-epp powersave, EPP balance_performance 45.6 73.5 79.4 acpi-cpufreq performance 45.6 73.4 78.9 acpi-cpufreq schedutil 45.4 73.2 79.0 amd-pstate-epp powersave, EPP balance_power 45.9 73.0 78.8 OpenBenchmarking.org Celsius, Fewer Is Better uvg266 0.8.0 CPU Temperature Monitor 20 40 60 80 100
Selenium Result
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Firefox acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance 80 160 240 320 400 SE +/- 2.43, N = 3 SE +/- 1.91, N = 3 SE +/- 2.65, N = 3 SE +/- 2.03, N = 3 SE +/- 3.83, N = 3 SE +/- 2.34, N = 3 SE +/- 0.74, N = 3 385.2 383.5 383.1 383.0 380.5 377.6 377.5 1. firefox 126.0.1 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4014 5720 amd-pstate-epp powersave, EPP balance_power 600 4116 5723 amd-pstate-epp powersave, EPP performance 600 4311 5723 amd-pstate-epp powersave, EPP balance_performance 600 4626 5722 acpi-cpufreq schedutil 3042 4901 5685 amd-pstate-epp performance, EPP performance 600 4951 5710 acpi-cpufreq performance 4300 5031 5709 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.5 45.1 75.2 acpi-cpufreq performance 20.3 43.0 74.7 amd-pstate-epp powersave, EPP power 22.4 41.6 72.8 amd-pstate-epp powersave, EPP performance 19.4 41.3 75.2 acpi-cpufreq schedutil 19.8 41.1 73.1 amd-pstate-epp powersave, EPP balance_performance 19.3 40.5 72.6 amd-pstate-epp powersave, EPP balance_power 19.8 40.4 74.7 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 45.6 57.3 67.8 acpi-cpufreq performance 44.6 56.9 69.4 amd-pstate-epp powersave, EPP balance_performance 45.0 56.6 66.1 amd-pstate-epp powersave, EPP performance 44.1 55.8 65.3 acpi-cpufreq schedutil 44.3 55.6 66.3 amd-pstate-epp powersave, EPP power 44.3 54.9 65.3 amd-pstate-epp powersave, EPP balance_power 43.6 54.6 64.9 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100
srsRAN Project srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mbps, More Is Better srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Total acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance 400 800 1200 1600 2000 SE +/- 0.17, N = 4 SE +/- 0.17, N = 4 SE +/- 0.02, N = 5 SE +/- 0.03, N = 5 SE +/- 0.04, N = 5 SE +/- 0.05, N = 5 SE +/- 0.04, N = 5 2080.1 2080.2 2080.6 2080.6 2080.6 2080.7 2080.7 MIN: 1273.2 / MAX: 2080.3 MIN: 1273.9 / MAX: 2080.6 MIN: 1274.2 / MAX: 2080.7 MIN: 1274.2 / MAX: 2080.7 MIN: 1274.2 / MAX: 2080.7 MIN: 1274.3 / MAX: 2080.9 MIN: 1274.2 / MAX: 2080.8 1. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl Mbps Per Watt
OpenBenchmarking.org Mbps Per Watt, More Is Better srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Total amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance acpi-cpufreq schedutil 4 8 12 16 20 14.46 14.47 14.48 14.67 14.69 15.72 15.92 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4579 5519 amd-pstate-epp powersave, EPP balance_power 600 4627 5515 amd-pstate-epp powersave, EPP power 600 4688 5642 amd-pstate-epp powersave, EPP balance_performance 600 4712 5586 amd-pstate-epp performance, EPP performance 600 4843 5692 acpi-cpufreq schedutil 3014 5021 5468 acpi-cpufreq performance 4300 5091 5463 OpenBenchmarking.org Megahertz, More Is Better srsRAN Project 23.10.1-20240325 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 18.2 143.9 195.2 amd-pstate-epp powersave, EPP performance 19.5 143.8 194.8 amd-pstate-epp powersave, EPP power 22.1 143.7 194.8 amd-pstate-epp powersave, EPP balance_power 19.5 141.8 195.4 amd-pstate-epp powersave, EPP balance_performance 19.1 141.6 194.5 acpi-cpufreq performance 19.8 132.3 180.9 acpi-cpufreq schedutil 19.0 130.6 180.0 OpenBenchmarking.org Watts, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 44.9 79.1 86.5 amd-pstate-epp powersave, EPP power 45.0 79.0 86.9 acpi-cpufreq schedutil 43.5 78.4 87.1 acpi-cpufreq performance 44.0 78.3 87.5 amd-pstate-epp powersave, EPP balance_performance 44.1 78.2 85.9 amd-pstate-epp powersave, EPP performance 44.1 78.2 85.8 amd-pstate-epp powersave, EPP balance_power 44.0 77.6 85.3 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Temperature Monitor 20 40 60 80 100
WebP Image Encode Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.4 Encode Settings: Quality 100, Lossless amd-pstate-epp powersave, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance 0.5355 1.071 1.6065 2.142 2.6775 SE +/- 0.01, N = 5 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 2.31 2.31 2.32 2.33 2.35 2.37 2.38 1. (CC) gcc options: -fvisibility=hidden -O2 -lm MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP Image Encode 1.4 Encode Settings: Quality 100, Lossless amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance 0.0131 0.0262 0.0393 0.0524 0.0655 0.053 0.055 0.055 0.056 0.057 0.058 0.058 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4687 5729 amd-pstate-epp powersave, EPP performance 600 4729 5723 amd-pstate-epp powersave, EPP balance_performance 600 4729 5729 amd-pstate-epp powersave, EPP power 600 4755 5729 amd-pstate-epp performance, EPP performance 600 4931 5724 acpi-cpufreq schedutil 3000 5143 5724 acpi-cpufreq performance 4300 5212 5726 OpenBenchmarking.org Megahertz, More Is Better WebP Image Encode 1.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 18.9 44.8 56.9 amd-pstate-epp powersave, EPP power 22.2 43.1 56.6 acpi-cpufreq performance 18.1 42.0 55.8 amd-pstate-epp powersave, EPP balance_power 19.8 41.3 56.2 acpi-cpufreq schedutil 19.4 41.2 55.9 amd-pstate-epp powersave, EPP balance_performance 18.8 40.1 55.8 amd-pstate-epp powersave, EPP performance 19.6 39.7 55.3 OpenBenchmarking.org Watts, Fewer Is Better WebP Image Encode 1.4 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 40.3 58.2 64.4 amd-pstate-epp powersave, EPP power 39.3 56.3 63.5 acpi-cpufreq performance 37.5 55.4 65.1 acpi-cpufreq schedutil 37.5 54.2 62.0 amd-pstate-epp powersave, EPP balance_power 37.9 53.3 62.5 amd-pstate-epp powersave, EPP balance_performance 37.6 52.7 62.4 amd-pstate-epp powersave, EPP performance 38.4 52.0 62.4 OpenBenchmarking.org Celsius, Fewer Is Better WebP Image Encode 1.4 CPU Temperature Monitor 20 40 60 80 100
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 13 - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power 60 120 180 240 300 SE +/- 0.31, N = 5 SE +/- 0.26, N = 5 SE +/- 0.27, N = 5 SE +/- 0.45, N = 5 SE +/- 0.45, N = 5 SE +/- 0.19, N = 5 SE +/- 0.34, N = 5 258.70 259.40 259.56 259.81 260.01 260.54 260.87 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 13 - Input: Bosphorus 4K acpi-cpufreq performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance 0.4334 0.8668 1.3002 1.7336 2.167 1.886 1.889 1.901 1.905 1.906 1.920 1.926 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4271 5686 amd-pstate-epp powersave, EPP balance_performance 600 4432 5691 amd-pstate-epp powersave, EPP power 600 4519 5683 amd-pstate-epp powersave, EPP performance 600 4712 5724 amd-pstate-epp performance, EPP performance 600 4898 5699 acpi-cpufreq schedutil 4300 5045 5493 acpi-cpufreq performance 4300 5079 5508 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.0 137.6 198.8 acpi-cpufreq performance 19.1 137.5 198.7 amd-pstate-epp powersave, EPP balance_power 20.1 136.9 199.1 amd-pstate-epp powersave, EPP power 11.4 136.3 198.8 acpi-cpufreq schedutil 10.3 136.1 198.5 amd-pstate-epp powersave, EPP balance_performance 18.3 135.7 198.9 amd-pstate-epp powersave, EPP performance 19.5 134.9 198.9 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.2 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.0 72.0 79.3 amd-pstate-epp powersave, EPP power 46.3 72.0 79.0 acpi-cpufreq performance 45.1 71.9 79.6 amd-pstate-epp powersave, EPP balance_performance 45.6 71.7 79.1 acpi-cpufreq schedutil 45.0 71.5 79.1 amd-pstate-epp powersave, EPP performance 45.3 71.4 78.8 amd-pstate-epp powersave, EPP balance_power 45.3 71.3 78.6 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 2.2 CPU Temperature Monitor 20 40 60 80 100
VP9 libvpx Encoding This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 1080p amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 11 22 33 44 55 SE +/- 0.46, N = 4 SE +/- 0.24, N = 4 SE +/- 0.28, N = 4 SE +/- 0.17, N = 4 SE +/- 0.31, N = 4 SE +/- 0.18, N = 4 SE +/- 0.24, N = 4 48.27 48.37 48.75 48.99 49.54 49.75 49.80 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 1080p acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance 0.1532 0.3064 0.4596 0.6128 0.766 0.653 0.663 0.669 0.672 0.673 0.675 0.681 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4711 5690 amd-pstate-epp powersave, EPP balance_performance 600 4834 5679 amd-pstate-epp powersave, EPP performance 600 4876 5707 amd-pstate-epp powersave, EPP balance_power 600 4899 5673 acpi-cpufreq schedutil 3000 5143 5574 acpi-cpufreq performance 4300 5225 5714 amd-pstate-epp performance, EPP performance 600 5225 5714 OpenBenchmarking.org Megahertz, More Is Better VP9 libvpx Encoding 1.13 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 20.0 74.0 94.4 amd-pstate-epp performance, EPP performance 21.5 73.9 94.9 acpi-cpufreq schedutil 19.0 73.9 94.4 amd-pstate-epp powersave, EPP balance_power 19.8 73.7 93.9 amd-pstate-epp powersave, EPP performance 19.7 72.7 93.1 amd-pstate-epp powersave, EPP power 22.5 72.6 92.5 amd-pstate-epp powersave, EPP balance_performance 19.6 72.2 92.6 OpenBenchmarking.org Watts, Fewer Is Better VP9 libvpx Encoding 1.13 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max acpi-cpufreq schedutil 43.0 63.8 70.3 amd-pstate-epp performance, EPP performance 43.6 63.4 69.0 amd-pstate-epp powersave, EPP power 44.0 63.2 67.9 amd-pstate-epp powersave, EPP performance 43.4 62.6 67.4 acpi-cpufreq performance 43.1 62.6 68.3 amd-pstate-epp powersave, EPP balance_power 42.9 62.3 66.4 amd-pstate-epp powersave, EPP balance_performance 43.1 62.2 67.1 OpenBenchmarking.org Celsius, Fewer Is Better VP9 libvpx Encoding 1.13 CPU Temperature Monitor 20 40 60 80 100
Selenium Result
OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Firefox acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 120 240 360 480 600 SE +/- 5.23, N = 5 SE +/- 1.35, N = 4 SE +/- 2.93, N = 5 SE +/- 1.14, N = 5 SE +/- 5.51, N = 5 SE +/- 2.11, N = 5 SE +/- 4.20, N = 5 565 536 533 533 531 530 523 1. firefox 126.0.1 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4198 5692 amd-pstate-epp powersave, EPP balance_performance 600 4539 5725 amd-pstate-epp powersave, EPP balance_power 600 4721 5683 amd-pstate-epp powersave, EPP performance 600 4852 5727 amd-pstate-epp performance, EPP performance 600 4867 5696 acpi-cpufreq schedutil 3042 5069 5670 acpi-cpufreq performance 4300 5250 5723 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 21.8 62.8 117.4 amd-pstate-epp powersave, EPP performance 19.3 59.7 116.8 acpi-cpufreq performance 20.5 58.5 116.6 acpi-cpufreq schedutil 19.5 57.6 114.4 amd-pstate-epp powersave, EPP balance_performance 19.3 57.4 117.2 amd-pstate-epp powersave, EPP balance_power 19.8 57.3 117.8 amd-pstate-epp powersave, EPP power 22.6 56.9 118.3 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 46.3 64.8 73.9 amd-pstate-epp powersave, EPP balance_performance 46.1 63.3 72.4 amd-pstate-epp powersave, EPP performance 44.5 62.2 71.4 amd-pstate-epp powersave, EPP balance_power 45.0 62.1 71.5 acpi-cpufreq schedutil 45.0 61.8 70.6 acpi-cpufreq performance 45.6 61.7 72.0 amd-pstate-epp powersave, EPP power 45.5 61.7 70.6 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100
WavPack Audio Encoding This test times how long it takes to encode a sample WAV file to WavPack format with very high quality settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better WavPack Audio Encoding 5.7 WAV To WavPack amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance 1.1412 2.2824 3.4236 4.5648 5.706 SE +/- 0.023, N = 7 SE +/- 0.015, N = 7 SE +/- 0.037, N = 7 SE +/- 0.022, N = 7 SE +/- 0.013, N = 7 SE +/- 0.017, N = 25 SE +/- 0.009, N = 8 5.072 5.035 4.986 4.911 4.906 4.892 4.845 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 4054 5720 amd-pstate-epp powersave, EPP performance 600 4064 5704 amd-pstate-epp powersave, EPP balance_power 600 4108 5719 amd-pstate-epp powersave, EPP balance_performance 600 4244 5719 amd-pstate-epp performance, EPP performance 600 4497 5713 acpi-cpufreq schedutil 3000 5007 5715 acpi-cpufreq performance 4300 5010 5606 OpenBenchmarking.org Megahertz, More Is Better WavPack Audio Encoding 5.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.6 72.8 120.0 acpi-cpufreq schedutil 19.3 72.5 119.5 amd-pstate-epp performance, EPP performance 21.3 71.9 116.9 amd-pstate-epp powersave, EPP power 22.5 70.7 118.1 amd-pstate-epp powersave, EPP performance 18.6 70.2 114.4 amd-pstate-epp powersave, EPP balance_power 19.6 69.3 117.2 amd-pstate-epp powersave, EPP balance_performance 19.3 68.5 115.9 OpenBenchmarking.org Watts, Fewer Is Better WavPack Audio Encoding 5.7 CPU Power Consumption Monitor 40 80 120 160 200 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP performance 40.6 59.9 67.3 amd-pstate-epp performance, EPP performance 42.1 59.7 66.4 amd-pstate-epp powersave, EPP balance_performance 42.0 58.9 67.5 acpi-cpufreq schedutil 40.9 57.9 68.3 amd-pstate-epp powersave, EPP balance_power 41.0 57.9 65.3 amd-pstate-epp powersave, EPP power 41.3 57.9 65.6 acpi-cpufreq performance 40.1 54.9 61.4 OpenBenchmarking.org Celsius, Fewer Is Better WavPack Audio Encoding 5.7 CPU Temperature Monitor 20 40 60 80 100
Monkey Audio Encoding This test times how long it takes to encode a sample WAV file to Monkey's Audio APE format. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Monkey Audio Encoding 3.99.6 WAV To APE amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance 3 6 9 12 15 SE +/- 0.009, N = 5 SE +/- 0.023, N = 5 SE +/- 0.067, N = 5 SE +/- 0.070, N = 5 SE +/- 0.082, N = 5 SE +/- 0.053, N = 5 SE +/- 0.072, N = 5 9.732 9.713 9.589 9.504 9.481 9.425 9.362 1. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 4493 5698 amd-pstate-epp powersave, EPP balance_power 600 4525 5484 amd-pstate-epp powersave, EPP power 600 4660 5726 amd-pstate-epp performance, EPP performance 600 4845 5693 amd-pstate-epp powersave, EPP balance_performance 600 4877 5724 acpi-cpufreq schedutil 4300 5170 5663 acpi-cpufreq performance 4300 5216 5713 OpenBenchmarking.org Megahertz, More Is Better Monkey Audio Encoding 3.99.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.9 45.1 58.7 amd-pstate-epp powersave, EPP power 22.7 44.0 55.2 amd-pstate-epp powersave, EPP performance 19.7 43.6 54.9 acpi-cpufreq performance 19.8 43.5 54.8 acpi-cpufreq schedutil 18.3 42.1 54.2 amd-pstate-epp powersave, EPP balance_power 19.9 41.7 51.0 amd-pstate-epp powersave, EPP balance_performance 19.6 40.6 49.9 OpenBenchmarking.org Watts, Fewer Is Better Monkey Audio Encoding 3.99.6 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 47.8 61.6 65.8 amd-pstate-epp powersave, EPP power 46.6 59.8 64.3 amd-pstate-epp powersave, EPP performance 44.9 58.6 63.8 acpi-cpufreq performance 44.9 58.1 65.0 acpi-cpufreq schedutil 45.6 56.8 63.1 amd-pstate-epp powersave, EPP balance_performance 46.0 54.4 57.4 amd-pstate-epp powersave, EPP balance_power 45.1 53.4 56.3 OpenBenchmarking.org Celsius, Fewer Is Better Monkey Audio Encoding 3.99.6 CPU Temperature Monitor 20 40 60 80 100
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: CPU amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance 5 10 15 20 25 SE +/- 0.05, N = 5 SE +/- 0.04, N = 5 SE +/- 0.08, N = 5 SE +/- 0.05, N = 5 SE +/- 0.07, N = 5 SE +/- 0.06, N = 5 SE +/- 0.04, N = 5 19.76 19.80 19.95 19.96 20.05 20.07 20.08 MIN: 17.96 / MAX: 20.16 MIN: 17.93 / MAX: 20.17 MIN: 17.97 / MAX: 20.27 MIN: 17.95 / MAX: 20.36 MIN: 17.94 / MAX: 20.39 MIN: 18.04 / MAX: 20.37 MIN: 18.08 / MAX: 20.43 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: CPU amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP performance 0.0342 0.0684 0.1026 0.1368 0.171 0.146 0.147 0.148 0.150 0.150 0.151 0.152 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_power 600 4087 5291 amd-pstate-epp powersave, EPP performance 600 4248 5706 amd-pstate-epp powersave, EPP balance_performance 600 4257 5373 amd-pstate-epp powersave, EPP power 600 4317 5669 acpi-cpufreq schedutil 4300 4785 5386 amd-pstate-epp performance, EPP performance 600 4792 5694 acpi-cpufreq performance 4300 4829 5669 OpenBenchmarking.org Megahertz, More Is Better LuxCoreRender 2.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP power 21.7 135.7 201.6 amd-pstate-epp performance, EPP performance 20.6 135.6 200.8 amd-pstate-epp powersave, EPP balance_power 20.2 134.1 201.2 acpi-cpufreq schedutil 19.2 134.1 201.1 amd-pstate-epp powersave, EPP balance_performance 19.4 132.8 201.0 acpi-cpufreq performance 19.0 132.8 201.0 amd-pstate-epp powersave, EPP performance 19.5 131.9 201.0 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 46.8 73.5 80.8 amd-pstate-epp performance, EPP performance 46.9 73.5 80.8 amd-pstate-epp powersave, EPP balance_performance 46.1 73.1 80.4 amd-pstate-epp powersave, EPP performance 45.9 73.0 80.0 amd-pstate-epp powersave, EPP balance_power 46.0 72.3 79.6 acpi-cpufreq schedutil 45.8 72.0 79.8 acpi-cpufreq performance 45.9 71.8 79.4 OpenBenchmarking.org Celsius, Fewer Is Better LuxCoreRender 2.6 CPU Temperature Monitor 20 40 60 80 100
srsRAN Project srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mbps, More Is Better srsRAN Project 23.10.1-20240325 Test: PDSCH Processor Benchmark, Throughput Total amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil 4K 8K 12K 16K 20K SE +/- 27.95, N = 7 SE +/- 39.24, N = 7 SE +/- 39.08, N = 7 SE +/- 65.96, N = 7 SE +/- 75.87, N = 7 SE +/- 37.33, N = 7 SE +/- 32.29, N = 7 12186.1 14495.3 16092.7 16229.4 16293.8 16591.7 16633.9 1. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl Mbps Per Watt
OpenBenchmarking.org Mbps Per Watt, More Is Better srsRAN Project 23.10.1-20240325 Test: PDSCH Processor Benchmark, Throughput Total amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil 30 60 90 120 150 97.71 117.02 130.29 130.49 130.93 132.66 134.73 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP performance 600 3749 5236 amd-pstate-epp powersave, EPP balance_power 600 3834 5096 amd-pstate-epp powersave, EPP balance_performance 600 3846 5091 amd-pstate-epp powersave, EPP power 600 3923 5442 amd-pstate-epp performance, EPP performance 600 4359 5711 acpi-cpufreq schedutil 3000 4546 5414 acpi-cpufreq performance 4300 4619 5392 OpenBenchmarking.org Megahertz, More Is Better srsRAN Project 23.10.1-20240325 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 19.6 125.1 201.7 amd-pstate-epp performance, EPP performance 20.9 124.9 201.3 amd-pstate-epp powersave, EPP power 22.5 124.7 200.9 amd-pstate-epp powersave, EPP performance 19.0 124.0 201.4 amd-pstate-epp powersave, EPP balance_power 19.5 123.9 201.2 amd-pstate-epp powersave, EPP balance_performance 18.4 123.5 201.4 acpi-cpufreq schedutil 18.7 123.5 201.5 OpenBenchmarking.org Watts, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 43.4 70.8 78.5 amd-pstate-epp performance, EPP performance 43.4 70.6 78.5 amd-pstate-epp powersave, EPP performance 42.6 69.7 77.8 amd-pstate-epp powersave, EPP balance_performance 42.9 69.7 78.1 acpi-cpufreq performance 42.5 69.7 77.6 acpi-cpufreq schedutil 42.5 69.4 77.6 amd-pstate-epp powersave, EPP balance_power 42.5 69.2 77.0 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Temperature Monitor 20 40 60 80 100
WebP Image Encode Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.4 Encode Settings: Quality 100, Highest Compression amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil 1.179 2.358 3.537 4.716 5.895 SE +/- 0.01, N = 8 SE +/- 0.02, N = 8 SE +/- 0.00, N = 8 SE +/- 0.04, N = 8 SE +/- 0.03, N = 8 SE +/- 0.04, N = 8 SE +/- 0.03, N = 8 5.13 5.14 5.16 5.18 5.19 5.23 5.24 1. (CC) gcc options: -fvisibility=hidden -O2 -lm MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP Image Encode 1.4 Encode Settings: Quality 100, Highest Compression amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power 0.0315 0.063 0.0945 0.126 0.1575 0.128 0.134 0.134 0.137 0.138 0.140 0.140 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 3980 5721 amd-pstate-epp powersave, EPP performance 600 3981 5726 amd-pstate-epp powersave, EPP balance_power 600 4147 5682 amd-pstate-epp performance, EPP performance 600 4360 5726 amd-pstate-epp powersave, EPP balance_performance 600 4385 5715 acpi-cpufreq schedutil 4300 4969 5669 acpi-cpufreq performance 4300 5042 5740 OpenBenchmarking.org Megahertz, More Is Better WebP Image Encode 1.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 17.6 40.4 56.3 amd-pstate-epp powersave, EPP power 22.2 38.8 56.6 acpi-cpufreq performance 18.0 38.2 55.9 acpi-cpufreq schedutil 18.6 38.2 55.8 amd-pstate-epp powersave, EPP performance 19.5 37.9 55.9 amd-pstate-epp powersave, EPP balance_performance 19.1 37.0 55.7 amd-pstate-epp powersave, EPP balance_power 17.6 36.8 55.8 OpenBenchmarking.org Watts, Fewer Is Better WebP Image Encode 1.4 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 40.9 56.3 60.8 acpi-cpufreq schedutil 37.5 54.8 62.3 amd-pstate-epp powersave, EPP power 40.1 54.4 63.4 amd-pstate-epp powersave, EPP performance 38.4 54.2 62.8 acpi-cpufreq performance 38.3 54.0 60.3 amd-pstate-epp powersave, EPP balance_performance 38.1 53.2 62.6 amd-pstate-epp powersave, EPP balance_power 38.5 51.1 61.1 OpenBenchmarking.org Celsius, Fewer Is Better WebP Image Encode 1.4 CPU Temperature Monitor 20 40 60 80 100
LAME MP3 Encoding LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance acpi-cpufreq schedutil acpi-cpufreq performance 0.9992 1.9984 2.9976 3.9968 4.996 SE +/- 0.002, N = 8 SE +/- 0.003, N = 8 SE +/- 0.028, N = 8 SE +/- 0.003, N = 8 SE +/- 0.002, N = 8 SE +/- 0.017, N = 8 SE +/- 0.013, N = 8 4.441 4.417 4.383 4.382 4.354 4.350 4.291 1. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lncurses -lm CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 3956 5707 amd-pstate-epp powersave, EPP balance_power 600 4024 5719 amd-pstate-epp powersave, EPP power 600 4121 5693 amd-pstate-epp powersave, EPP performance 600 4213 5457 amd-pstate-epp performance, EPP performance 600 4541 5714 acpi-cpufreq schedutil 3000 4807 5578 acpi-cpufreq performance 4300 4996 5667 OpenBenchmarking.org Megahertz, More Is Better LAME MP3 Encoding 3.100 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.4 39.4 56.2 amd-pstate-epp powersave, EPP power 22.2 38.7 56.5 acpi-cpufreq performance 19.5 38.3 55.9 acpi-cpufreq schedutil 18.4 36.6 54.6 amd-pstate-epp powersave, EPP balance_power 19.1 35.7 51.4 amd-pstate-epp powersave, EPP performance 19.5 35.7 50.4 amd-pstate-epp powersave, EPP balance_performance 19.0 35.5 50.4 OpenBenchmarking.org Watts, Fewer Is Better LAME MP3 Encoding 3.100 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 42.9 56.7 61.1 amd-pstate-epp powersave, EPP power 41.9 56.5 63.6 acpi-cpufreq performance 40.0 56.0 61.4 acpi-cpufreq schedutil 40.1 54.9 63.5 amd-pstate-epp powersave, EPP balance_performance 40.8 51.6 55.1 amd-pstate-epp powersave, EPP performance 40.0 50.8 54.4 amd-pstate-epp powersave, EPP balance_power 40.1 50.5 54.1 OpenBenchmarking.org Celsius, Fewer Is Better LAME MP3 Encoding 3.100 CPU Temperature Monitor 20 40 60 80 100
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 100, Compression Effort 5 amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 3 6 9 12 15 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 10.09 10.15 10.17 10.17 10.18 10.21 10.24 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -lm -ldl MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 100, Compression Effort 5 amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP power acpi-cpufreq schedutil 0.0272 0.0544 0.0816 0.1088 0.136 0.119 0.120 0.120 0.120 0.121 0.121 0.121 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 2994 5670 amd-pstate-epp powersave, EPP performance 600 3366 5700 amd-pstate-epp powersave, EPP power 600 3612 5649 amd-pstate-epp powersave, EPP balance_power 600 3676 5465 acpi-cpufreq schedutil 4300 4538 5485 acpi-cpufreq performance 4300 4548 5729 amd-pstate-epp performance, EPP performance 600 4726 5701 OpenBenchmarking.org Megahertz, More Is Better WebP2 Image Encode 20220823 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 20.9 86.1 201.0 acpi-cpufreq performance 19.6 85.1 200.8 amd-pstate-epp powersave, EPP performance 19.4 84.9 200.6 amd-pstate-epp powersave, EPP balance_power 19.8 84.7 201.6 amd-pstate-epp powersave, EPP balance_performance 18.4 84.4 201.0 acpi-cpufreq schedutil 19.0 83.8 200.8 amd-pstate-epp powersave, EPP power 22.0 83.7 201.0 OpenBenchmarking.org Watts, Fewer Is Better WebP2 Image Encode 20220823 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp powersave, EPP power 47.4 69.7 79.8 amd-pstate-epp performance, EPP performance 45.9 69.1 78.6 amd-pstate-epp powersave, EPP balance_performance 45.5 68.6 77.8 acpi-cpufreq performance 45.6 68.3 78.4 amd-pstate-epp powersave, EPP performance 45.5 68.2 77.6 acpi-cpufreq schedutil 45.4 68.2 78.4 amd-pstate-epp powersave, EPP balance_power 45.4 67.9 76.8 OpenBenchmarking.org Celsius, Fewer Is Better WebP2 Image Encode 20220823 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Default amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance 4 8 12 16 20 SE +/- 0.10, N = 11 SE +/- 0.10, N = 15 SE +/- 0.09, N = 11 SE +/- 0.11, N = 15 SE +/- 0.09, N = 11 SE +/- 0.11, N = 12 SE +/- 0.11, N = 15 14.61 14.84 14.93 15.00 15.04 15.05 15.11 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -lm -ldl MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP2 Image Encode 20220823 Encode Settings: Default amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq schedutil acpi-cpufreq performance 0.0549 0.1098 0.1647 0.2196 0.2745 0.228 0.233 0.241 0.242 0.244 0.244 0.244 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP balance_performance 600 3102 5654 amd-pstate-epp powersave, EPP power 600 3138 5647 amd-pstate-epp powersave, EPP balance_power 600 3677 5722 amd-pstate-epp powersave, EPP performance 600 3886 5724 amd-pstate-epp performance, EPP performance 600 4473 5723 acpi-cpufreq schedutil 3010 4562 5732 acpi-cpufreq performance 4300 4575 5660 OpenBenchmarking.org Megahertz, More Is Better WebP2 Image Encode 20220823 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.5 64.2 200.3 amd-pstate-epp powersave, EPP power 22.3 64.1 199.5 amd-pstate-epp powersave, EPP performance 18.4 62.4 199.4 acpi-cpufreq schedutil 18.8 61.8 199.0 amd-pstate-epp powersave, EPP balance_performance 19.5 61.7 200.2 acpi-cpufreq performance 18.3 61.5 197.4 amd-pstate-epp powersave, EPP balance_power 19.7 61.5 199.8 OpenBenchmarking.org Watts, Fewer Is Better WebP2 Image Encode 20220823 CPU Power Consumption Monitor 50 100 150 200 250 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.5 63.5 74.3 amd-pstate-epp powersave, EPP power 40.5 62.5 73.1 amd-pstate-epp powersave, EPP performance 39.8 61.7 72.8 acpi-cpufreq performance 39.3 61.1 71.9 amd-pstate-epp powersave, EPP balance_power 39.4 61.0 70.6 amd-pstate-epp powersave, EPP balance_performance 39.3 60.8 70.9 acpi-cpufreq schedutil 38.8 60.8 71.6 OpenBenchmarking.org Celsius, Fewer Is Better WebP2 Image Encode 20220823 CPU Temperature Monitor 20 40 60 80 100
WebP Image Encode Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.4 Encode Settings: Quality 100 amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_power 4 8 12 16 20 SE +/- 0.09, N = 11 SE +/- 0.11, N = 12 SE +/- 0.07, N = 12 SE +/- 0.11, N = 12 SE +/- 0.08, N = 12 SE +/- 0.09, N = 12 SE +/- 0.05, N = 15 15.94 16.17 16.30 16.38 16.39 16.42 16.59 1. (CC) gcc options: -fvisibility=hidden -O2 -lm MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP Image Encode 1.4 Encode Settings: Quality 100 amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance 0.1267 0.2534 0.3801 0.5068 0.6335 0.512 0.515 0.547 0.554 0.555 0.560 0.563 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp performance, EPP performance 600 3234 5725 amd-pstate-epp powersave, EPP performance 600 3253 5725 amd-pstate-epp powersave, EPP balance_power 600 3338 5725 amd-pstate-epp powersave, EPP power 600 3418 5688 amd-pstate-epp powersave, EPP balance_performance 600 3624 5706 acpi-cpufreq schedutil 4300 4684 5708 acpi-cpufreq performance 4300 4716 5720 OpenBenchmarking.org Megahertz, More Is Better WebP Image Encode 1.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp performance, EPP performance 19.5 32.1 56.7 amd-pstate-epp powersave, EPP power 22.4 31.0 56.1 amd-pstate-epp powersave, EPP balance_power 19.7 29.9 55.8 acpi-cpufreq performance 19.4 29.8 54.8 amd-pstate-epp powersave, EPP performance 19.8 29.5 55.6 acpi-cpufreq schedutil 19.1 29.2 55.0 amd-pstate-epp powersave, EPP balance_performance 19.3 28.7 55.1 OpenBenchmarking.org Watts, Fewer Is Better WebP Image Encode 1.4 CPU Power Consumption Monitor 16 32 48 64 80 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 40.3 53.7 60.4 amd-pstate-epp powersave, EPP balance_power 38.1 51.8 57.9 acpi-cpufreq performance 38.0 51.1 57.8 acpi-cpufreq schedutil 37.1 50.4 57.1 amd-pstate-epp powersave, EPP performance 38.8 50.0 58.0 amd-pstate-epp powersave, EPP power 40.0 49.7 58.6 amd-pstate-epp powersave, EPP balance_performance 37.8 49.0 58.1 OpenBenchmarking.org Celsius, Fewer Is Better WebP Image Encode 1.4 CPU Temperature Monitor 20 40 60 80 100
Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.4 Encode Settings: Default amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_performance acpi-cpufreq performance amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance 6 12 18 24 30 SE +/- 0.12, N = 13 SE +/- 0.17, N = 13 SE +/- 0.13, N = 13 SE +/- 0.17, N = 13 SE +/- 0.08, N = 13 SE +/- 0.14, N = 13 SE +/- 0.09, N = 13 25.85 25.99 26.04 26.11 26.48 26.52 26.56 1. (CC) gcc options: -fvisibility=hidden -O2 -lm MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP Image Encode 1.4 Encode Settings: Default amd-pstate-epp powersave, EPP power amd-pstate-epp performance, EPP performance acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_performance 0.2221 0.4442 0.6663 0.8884 1.1105 0.867 0.903 0.961 0.968 0.978 0.985 0.987 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate-epp powersave, EPP power 600 3269 5703 amd-pstate-epp powersave, EPP performance 600 3278 5703 amd-pstate-epp powersave, EPP balance_power 600 3322 5720 amd-pstate-epp powersave, EPP balance_performance 600 3363 5727 amd-pstate-epp performance, EPP performance 600 4053 5725 acpi-cpufreq performance 4300 4559 5709 acpi-cpufreq schedutil 4300 4617 5726 OpenBenchmarking.org Megahertz, More Is Better WebP Image Encode 1.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max amd-pstate-epp powersave, EPP power 22.1 29.8 71.1 amd-pstate-epp performance, EPP performance 20.0 29.4 48.3 acpi-cpufreq performance 19.2 27.6 45.3 amd-pstate-epp powersave, EPP performance 10.1 26.9 46.6 amd-pstate-epp powersave, EPP balance_power 19.5 26.8 51.6 acpi-cpufreq schedutil 19.0 26.6 55.6 amd-pstate-epp powersave, EPP balance_performance 19.3 26.5 50.9 OpenBenchmarking.org Watts, Fewer Is Better WebP Image Encode 1.4 CPU Power Consumption Monitor 20 40 60 80 100 CPU Temp
Min Avg Max amd-pstate-epp performance, EPP performance 41.1 52.4 56.4 amd-pstate-epp powersave, EPP power 40.8 50.8 60.1 acpi-cpufreq performance 39.1 50.2 54.9 amd-pstate-epp powersave, EPP performance 39.5 49.8 54.9 amd-pstate-epp powersave, EPP balance_performance 39.4 48.6 55.3 acpi-cpufreq schedutil 39.0 48.3 54.9 amd-pstate-epp powersave, EPP balance_power 39.1 47.6 54.6 OpenBenchmarking.org Celsius, Fewer Is Better WebP Image Encode 1.4 CPU Temperature Monitor 16 32 48 64 80
CPU Temperature Monitor OpenBenchmarking.org Celsius CPU Temperature Monitor Phoronix Test Suite System Monitoring amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP performance acpi-cpufreq schedutil amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP power acpi-cpufreq performance 20 40 60 80 100 Min: 38 / Avg: 74.54 / Max: 95.75 Min: 37.25 / Avg: 73.13 / Max: 96.13 Min: 38.38 / Avg: 72.95 / Max: 95.88 Min: 35 / Avg: 72.41 / Max: 93.38 Min: 37.63 / Avg: 72.28 / Max: 95.5 Min: 37.88 / Avg: 72.26 / Max: 95.63 Min: 36.13 / Avg: 72.14 / Max: 94.5
CPU Power Consumption Monitor OpenBenchmarking.org Watts CPU Power Consumption Monitor Phoronix Test Suite System Monitoring amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP balance_power acpi-cpufreq schedutil amd-pstate-epp powersave, EPP power acpi-cpufreq performance amd-pstate-epp powersave, EPP balance_performance 40 80 120 160 200 Min: 11.03 / Avg: 146.07 / Max: 205.19 Min: 10.13 / Avg: 143.84 / Max: 204.82 Min: 9.89 / Avg: 143.62 / Max: 206.63 Min: 9.4 / Avg: 143.38 / Max: 210.18 Min: 11.3 / Avg: 141.9 / Max: 203.76 Min: 9.47 / Avg: 141.87 / Max: 203.99 Min: 9.89 / Avg: 140.61 / Max: 205.38
CPU Peak Freq (Highest CPU Core Frequency) Monitor OpenBenchmarking.org Megahertz CPU Peak Freq (Highest CPU Core Frequency) Monitor Phoronix Test Suite System Monitoring acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate-epp performance, EPP performance amd-pstate-epp powersave, EPP performance amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power 1100 2200 3300 4400 5500 Min: 4225 / Avg: 5083.97 / Max: 5760 Min: 3000 / Avg: 5063.12 / Max: 5739 Min: 600 / Avg: 5001.29 / Max: 5740 Min: 600 / Avg: 4913.71 / Max: 5755 Min: 600 / Avg: 4902.34 / Max: 5772 Min: 600 / Avg: 4874.3 / Max: 6158 Min: 600 / Avg: 4849.38 / Max: 5849
Meta Performance Per Watts OpenBenchmarking.org Performance Per Watts, More Is Better Meta Performance Per Watts Performance Per Watts amd-pstate-epp powersave, EPP power amd-pstate-epp powersave, EPP balance_performance amd-pstate-epp powersave, EPP balance_power amd-pstate-epp powersave, EPP performance amd-pstate-epp performance, EPP performance acpi-cpufreq performance acpi-cpufreq schedutil 6 12 18 24 30 26.97 27.02 27.08 27.14 27.19 27.24 27.26
amd-pstate-epp powersave, EPP balance_performance Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 22 August 2024 20:49 by user phoronix.
amd-pstate-epp performance, EPP performance Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 23 August 2024 09:45 by user phoronix.
amd-pstate-epp powersave, EPP performance Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: performance) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 23 August 2024 18:31 by user phoronix.
amd-pstate-epp powersave, EPP power Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: power) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 24 August 2024 09:46 by user phoronix.
amd-pstate-epp powersave, EPP balance_power Processor: AMD Ryzen 9 9950X 16-Core @ 5.75GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG STRIX X670E-E GAMING WIFI (2204 BIOS), Chipset: AMD Device 14d8, Memory: 2 x 32GB DDR5-6400MT/s Corsair CMK64GX5M2B6400C32, Disk: 2000GB Corsair MP700 PRO, Graphics: AMD Radeon RX 7900 GRE 16GB, Audio: AMD Navi 31 HDMI/DP, Monitor: DELL U2723QE, Network: Intel I225-V + Intel Wi-Fi 6E
OS: Ubuntu 24.04, Kernel: 6.10.0-phx (x86_64), Desktop: GNOME Shell 46.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 24.2~git2406040600.8112d4~oibaf~n (git-8112d44 2024-06-04 noble-oibaf-ppa) (LLVM 17.0.6 DRM 3.57), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_power) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 24 August 2024 19:47 by user phoronix.
acpi-cpufreq schedutil Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 25 August 2024 13:57 by user phoronix.
acpi-cpufreq performance Processor: AMD Ryzen 9 9950X 16-Core @ 4.30GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG STRIX X670E-E GAMING WIFI (2204 BIOS), Chipset: AMD Device 14d8, Memory: 2 x 32GB DDR5-6400MT/s Corsair CMK64GX5M2B6400C32, Disk: 2000GB Corsair MP700 PRO, Graphics: AMD Radeon RX 7900 GRE 16GB, Audio: AMD Navi 31 HDMI/DP, Monitor: DELL U2723QE, Network: Intel I225-V + Intel Wi-Fi 6E
OS: Ubuntu 24.04, Kernel: 6.10.0-phx (x86_64), Desktop: GNOME Shell 46.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 24.2~git2406040600.8112d4~oibaf~n (git-8112d44 2024-06-04 noble-oibaf-ppa) (LLVM 17.0.6 DRM 3.57), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb40401aGraphics Notes: BAR1 / Visible vRAM Size: 16368 MBPython Notes: Python 3.12.3Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 25 August 2024 23:14 by user phoronix.