AMD Ryzen 9 7950X CPUFreq vs. AMD P-State Linux benchmarks by Michael Larabel for a future article.
amd-pstate schedutil Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 XT, Audio: AMD Navi 21 HDMI Audio, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411
OS: Ubuntu 22.04, Kernel: 6.0.0-060000-generic (x86_64), Desktop: GNOME Shell 42.4, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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 schedutil (Boost: Enabled) - CPU Microcode: 0xa601203Graphics Notes: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101Python Notes: Python 3.10.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
amd-pstate performance Changed Graphics to AMD Radeon RX 6800 XT (2575/1000MHz) .
Processor Change: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa601203
acpi-cpufreq schedutil Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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: 0xa601203Graphics Notes: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101Python Notes: Python 3.10.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
acpi-cpufreq ondemand Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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 ondemand (Boost: Enabled) - CPU Microcode: 0xa601203Graphics Notes: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101Python Notes: Python 3.10.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
acpi-cpufreq performance Changed Processor to AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads) .
Processor Change: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203
PyPerformance PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: python_startup amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 2 4 6 8 10 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 6.68 5.13 4.52 4.47 4.36 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4238 5749 acpi-cpufreq ondemand 3000 5067 5750 acpi-cpufreq schedutil 4500 5483 5750 acpi-cpufreq performance 4500 5510 5749 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better PyPerformance 1.0.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 27.7 65.0 68.7 acpi-cpufreq performance 25.8 64.1 68.8 acpi-cpufreq schedutil 25.9 63.9 68.1 acpi-cpufreq ondemand 25.4 58.4 65.7 amd-pstate schedutil 21.5 43.5 61.5 OpenBenchmarking.org Watts, Fewer Is Better PyPerformance 1.0.0 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 4680 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 4722 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 4773 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 4927 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 4614 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.4 60.2 62.5 amd-pstate performance 39.9 59.9 62.3 acpi-cpufreq performance 38.9 58.2 60.3 acpi-cpufreq ondemand 38.6 56.4 60.5 amd-pstate schedutil 37.9 41.4 47.8 OpenBenchmarking.org Celsius, Fewer Is Better PyPerformance 1.0.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: python_startup amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 3 6 9 12 15 Min: 6.63 / Avg: 6.68 / Max: 6.73 Min: 5.1 / Avg: 5.13 / Max: 5.17 Min: 4.52 / Avg: 4.52 / Max: 4.53 Min: 4.47 / Avg: 4.47 / Max: 4.47 Min: 4.35 / Avg: 4.36 / Max: 4.36
AOM AV1 This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 20 40 60 80 100 SE +/- 0.19, N = 4 SE +/- 0.17, N = 4 SE +/- 0.35, N = 4 SE +/- 0.22, N = 5 SE +/- 0.48, N = 5 50.67 63.21 69.52 73.79 74.41 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 0.144 0.288 0.432 0.576 0.72 0.562 0.571 0.612 0.628 0.640 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4593 5499 acpi-cpufreq ondemand 3000 4760 5497 acpi-cpufreq schedutil 3000 5095 5611 acpi-cpufreq performance 4500 5115 5497 amd-pstate performance 5204 5510 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 29.0 118.5 164.5 acpi-cpufreq performance 26.3 115.4 164.2 acpi-cpufreq ondemand 26.2 113.5 161.8 acpi-cpufreq schedutil 24.4 112.4 158.9 amd-pstate schedutil 26.4 88.7 119.2 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 1777 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 1730 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 1788 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1911 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 1708 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 42.6 72.0 81.8 acpi-cpufreq performance 42.0 71.7 80.6 acpi-cpufreq schedutil 42.5 70.7 79.0 acpi-cpufreq ondemand 41.8 70.4 79.5 amd-pstate schedutil 38.3 63.6 69.3 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 14 28 42 56 70 Min: 50.23 / Avg: 50.67 / Max: 51.15 Min: 62.76 / Avg: 63.21 / Max: 63.53 Min: 68.84 / Avg: 69.52 / Max: 70.46 Min: 72.97 / Avg: 73.79 / Max: 74.15 Min: 72.77 / Avg: 74.41 / Max: 75.39 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.29, N = 3 SE +/- 0.06, N = 3 SE +/- 0.17, N = 3 SE +/- 0.09, N = 3 19.52 22.98 23.54 24.94 24.99 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 0.0396 0.0792 0.1188 0.1584 0.198 0.151 0.166 0.167 0.174 0.176 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4957 5751 acpi-cpufreq ondemand 3000 5055 5441 acpi-cpufreq schedutil 3000 5198 5495 acpi-cpufreq performance 4500 5215 5424 amd-pstate performance 5152 5451 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.3 143.8 170.8 acpi-cpufreq performance 26.9 141.8 173.2 acpi-cpufreq ondemand 23.7 140.7 167.8 acpi-cpufreq schedutil 26.6 138.8 168.9 amd-pstate schedutil 26.8 128.9 158.6 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 5032 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 4962 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 5207 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 5413 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 5670 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 43.0 78.9 84.8 amd-pstate performance 43.6 78.8 85.4 acpi-cpufreq performance 43.0 78.3 84.5 acpi-cpufreq ondemand 42.1 77.4 83.9 amd-pstate schedutil 40.5 75.8 81.4 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 6 12 18 24 30 Min: 19.48 / Avg: 19.52 / Max: 19.58 Min: 22.48 / Avg: 22.98 / Max: 23.47 Min: 23.46 / Avg: 23.54 / Max: 23.65 Min: 24.73 / Avg: 24.94 / Max: 25.28 Min: 24.86 / Avg: 24.99 / Max: 25.16 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 30 60 90 120 150 SE +/- 0.84, N = 7 SE +/- 0.62, N = 8 SE +/- 0.58, N = 8 SE +/- 0.89, N = 8 SE +/- 0.81, N = 8 129.74 137.56 138.13 150.60 155.84 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq performance 0.5918 1.1836 1.7754 2.3672 2.959 2.269 2.389 2.412 2.416 2.630 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4100 5702 acpi-cpufreq ondemand 3000 4403 5736 acpi-cpufreq schedutil 3000 4769 5500 acpi-cpufreq performance 4500 4998 5500 amd-pstate performance 5485 5751 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.2 64.5 97.8 amd-pstate schedutil 14.1 60.6 93.1 acpi-cpufreq schedutil 26.2 57.3 89.8 acpi-cpufreq performance 14.9 57.3 94.3 acpi-cpufreq ondemand 26.3 54.3 80.8 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 524 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 523 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 487 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 472 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 481 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 43.1 52.4 58.8 amd-pstate schedutil 42.8 51.2 56.1 acpi-cpufreq schedutil 42.8 50.6 55.8 acpi-cpufreq performance 41.6 50.1 57.1 acpi-cpufreq ondemand 41.4 46.2 49.8 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 16 32 48 64 80 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 30 60 90 120 150 Min: 127.53 / Avg: 129.74 / Max: 134.18 Min: 136.1 / Avg: 137.56 / Max: 141.68 Min: 136.07 / Avg: 138.13 / Max: 141.74 Min: 148.79 / Avg: 150.6 / Max: 156.66 Min: 153.66 / Avg: 155.84 / Max: 159.74 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance amd-pstate performance 30 60 90 120 150 SE +/- 0.29, N = 7 SE +/- 0.21, N = 7 SE +/- 0.35, N = 7 SE +/- 0.17, N = 8 SE +/- 0.48, N = 8 111.92 113.90 115.31 128.73 131.77 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 0.4156 0.8312 1.2468 1.6624 2.078 1.671 1.695 1.738 1.760 1.847 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 3954 5673 acpi-cpufreq ondemand 3000 4455 5500 acpi-cpufreq schedutil 3000 4732 5500 acpi-cpufreq performance 3600 5032 5703 amd-pstate performance 5271 5674 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.8 74.9 113.7 acpi-cpufreq performance 26.8 69.7 107.9 amd-pstate schedutil 27.0 69.0 103.3 acpi-cpufreq schedutil 26.4 67.2 103.0 acpi-cpufreq ondemand 26.0 64.4 97.5 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 655 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 610 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 651 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 634 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 616 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 40.6 53.7 60.6 acpi-cpufreq performance 39.1 52.9 58.3 amd-pstate schedutil 46.5 52.1 59.6 acpi-cpufreq schedutil 39.5 51.8 57.1 acpi-cpufreq ondemand 38.5 48.8 53.3 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance amd-pstate performance 20 40 60 80 100 Min: 110.76 / Avg: 111.92 / Max: 112.88 Min: 112.88 / Avg: 113.9 / Max: 114.47 Min: 113.94 / Avg: 115.31 / Max: 116.45 Min: 128.21 / Avg: 128.73 / Max: 129.61 Min: 129.74 / Avg: 131.77 / Max: 133.3 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Tesseract Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Tesseract 2014-05-12 Resolution: 1920 x 1080 amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 200 400 600 800 1000 SE +/- 7.02, N = 15 SE +/- 0.98, N = 3 SE +/- 2.57, N = 3 SE +/- 0.35, N = 3 SE +/- 0.97, N = 3 849.28 994.79 994.97 996.69 998.32 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Tesseract 2014-05-12 Resolution: 1920 x 1080 amd-pstate schedutil amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 4 8 12 16 20 14.37 14.70 15.00 15.13 15.66 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5183 5750 acpi-cpufreq ondemand 3000 5216 5570 acpi-cpufreq schedutil 3000 5336 5508 acpi-cpufreq performance 4500 5379 5653 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Tesseract 2014-05-12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 37.4 67.8 78.3 acpi-cpufreq performance 26.9 66.6 77.7 acpi-cpufreq schedutil 26.9 65.7 78.9 acpi-cpufreq ondemand 26.7 63.5 74.7 amd-pstate schedutil 26.8 59.1 72.3 OpenBenchmarking.org Watts, Fewer Is Better Tesseract 2014-05-12 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 2395 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 2352 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 2323 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 2244 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 2076 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.6 54.3 62.5 amd-pstate performance 40.0 53.4 61.8 acpi-cpufreq performance 39.1 52.2 59.3 acpi-cpufreq ondemand 38.6 50.7 59.0 amd-pstate schedutil 39.1 46.8 61.4 OpenBenchmarking.org Celsius, Fewer Is Better Tesseract 2014-05-12 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Tesseract 2014-05-12 Resolution: 1920 x 1080 amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 200 400 600 800 1000 Min: 803.69 / Avg: 849.28 / Max: 886.29 Min: 993.31 / Avg: 994.79 / Max: 996.66 Min: 989.93 / Avg: 994.97 / Max: 998.33 Min: 996.1 / Avg: 996.69 / Max: 997.32 Min: 996.64 / Avg: 998.32 / Max: 1000
AOM AV1 This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 20 40 60 80 100 SE +/- 0.57, N = 5 SE +/- 0.61, N = 5 SE +/- 0.57, N = 5 SE +/- 0.64, N = 9 SE +/- 0.67, N = 9 70.05 70.72 75.50 79.13 80.50 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 0.2358 0.4716 0.7074 0.9432 1.179 0.978 1.005 1.035 1.040 1.048 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4382 5737 acpi-cpufreq ondemand 3000 4774 5726 acpi-cpufreq schedutil 3000 5005 5727 acpi-cpufreq performance 4500 5192 5749 amd-pstate performance 4476 5713 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.4 77.8 102.2 acpi-cpufreq performance 27.0 75.5 101.6 acpi-cpufreq ondemand 26.1 72.6 97.3 amd-pstate schedutil 26.5 72.3 96.7 acpi-cpufreq schedutil 26.3 69.7 92.8 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 959 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 940 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 929 Joules per run. 4. amd-pstate schedutil: Approximate power consumption of 969 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 948 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq performance 38.6 54.9 60.0 amd-pstate performance 40.0 54.8 60.6 amd-pstate schedutil 39.5 52.9 58.0 acpi-cpufreq ondemand 37.6 52.5 56.6 acpi-cpufreq schedutil 39.4 52.3 57.1 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 15 30 45 60 75 Min: 68.83 / Avg: 70.05 / Max: 72.19 Min: 69.81 / Avg: 70.72 / Max: 73.14 Min: 73.83 / Avg: 75.5 / Max: 77.31 Min: 78.1 / Avg: 79.13 / Max: 84.16 Min: 77.07 / Avg: 80.5 / Max: 84.55 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Selenium Result
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Google Chrome amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 80 160 240 320 400 SE +/- 2.91, N = 7 SE +/- 2.85, N = 3 SE +/- 1.20, N = 3 SE +/- 2.40, N = 3 SE +/- 2.52, N = 3 332 352 364 377 380 1. chrome 106.0.5249.91 Runs Per Minute Per Watt
OpenBenchmarking.org Runs Per Minute Per Watt, More Is Better Selenium Benchmark: Speedometer - Browser: Google Chrome amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand 2 4 6 8 10 6.504 6.694 7.058 7.282 7.429 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4186 5754 acpi-cpufreq ondemand 3000 4338 5700 acpi-cpufreq schedutil 3000 4771 5664 acpi-cpufreq performance 4500 5027 5639 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.1 56.8 77.9 acpi-cpufreq performance 26.7 51.8 81.8 amd-pstate schedutil 26.8 51.0 75.0 acpi-cpufreq schedutil 26.9 49.9 76.3 acpi-cpufreq ondemand 26.4 49.0 73.8 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 2063 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 1864 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 1962 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1829 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1780 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 38.8 53.9 63.9 amd-pstate performance 38.4 53.8 63.3 acpi-cpufreq performance 37.4 53.4 64.8 acpi-cpufreq ondemand 38.0 52.9 62.5 amd-pstate schedutil 36.9 51.0 62.3 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Google Chrome amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 70 140 210 280 350 Min: 323 / Avg: 332.29 / Max: 347 Min: 346 / Avg: 351.67 / Max: 355 Min: 362 / Avg: 363.67 / Max: 366 Min: 374 / Avg: 377.33 / Max: 382 Min: 377 / Avg: 380 / Max: 385 1. chrome 106.0.5249.91
AOM AV1 This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 0.3015 0.603 0.9045 1.206 1.5075 SE +/- 0.00, N = 3 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.01, N = 4 SE +/- 0.00, N = 4 1.18 1.31 1.31 1.33 1.34 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.002 0.004 0.006 0.008 0.01 0.009 0.009 0.009 0.009 0.009 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4503 5553 acpi-cpufreq schedutil 3000 4950 5449 acpi-cpufreq ondemand 3000 5005 5491 acpi-cpufreq performance 4500 5067 5477 amd-pstate performance 4450 5587 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 19.8 142.2 233.8 acpi-cpufreq performance 14.8 141.8 234.1 acpi-cpufreq ondemand 13.6 140.8 234.0 acpi-cpufreq schedutil 26.6 139.8 232.3 amd-pstate schedutil 27.8 129.1 231.7 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 60 120 180 240 300 1. amd-pstate performance: Approximate power consumption of 2738 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 2730 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 2746 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 2691 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 2754 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 44.3 84.2 92.9 acpi-cpufreq performance 44.5 83.4 92.3 amd-pstate performance 45.0 83.3 92.5 acpi-cpufreq ondemand 43.6 83.1 92.8 amd-pstate schedutil 43.5 78.7 92.5 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 2 4 6 8 10 Min: 1.17 / Avg: 1.18 / Max: 1.18 Min: 1.3 / Avg: 1.31 / Max: 1.31 Min: 1.31 / Avg: 1.31 / Max: 1.32 Min: 1.31 / Avg: 1.33 / Max: 1.34 Min: 1.33 / Avg: 1.34 / Max: 1.34 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Selenium Result
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Firefox amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 60 120 180 240 300 SE +/- 1.86, N = 3 SE +/- 1.53, N = 3 SE +/- 2.19, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 256.0 267.0 271.0 285.9 290.0 1. firefox 105.0.1 Runs Per Minute Per Watt
OpenBenchmarking.org Runs Per Minute Per Watt, More Is Better Selenium Benchmark: Speedometer - Browser: Firefox amd-pstate schedutil amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 1.1572 2.3144 3.4716 4.6288 5.786 4.407 4.748 5.038 5.051 5.143 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 4544 5650 acpi-cpufreq schedutil 3000 4825 5641 amd-pstate schedutil 400 4957 5750 acpi-cpufreq performance 4500 5158 5691 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 35.7 61.1 73.1 amd-pstate schedutil 26.5 58.1 70.7 acpi-cpufreq performance 26.9 56.7 73.0 acpi-cpufreq schedutil 26.9 53.7 71.5 acpi-cpufreq ondemand 26.2 51.9 69.9 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 2687 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 4008 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 2516 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 2468 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 2388 Joules per run. CPU Temp
Min Avg Max amd-pstate schedutil 34.8 54.5 63.8 acpi-cpufreq schedutil 37.1 54.1 63.3 amd-pstate performance 36.4 53.8 61.4 acpi-cpufreq ondemand 38.0 53.3 60.4 acpi-cpufreq performance 35.8 52.9 61.4 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Firefox amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 50 100 150 200 250 Min: 254 / Avg: 256.33 / Max: 260 Min: 264 / Avg: 267 / Max: 269 Min: 267 / Avg: 271.33 / Max: 274 Min: 285 / Avg: 285.93 / Max: 287 Min: 288 / Avg: 289.67 / Max: 291 1. firefox 105.0.1
Result
OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Firefox amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 500 1000 1500 2000 2500 SE +/- 6.23, N = 3 SE +/- 6.89, N = 3 SE +/- 5.86, N = 3 SE +/- 1.45, N = 3 SE +/- 3.79, N = 3 2376 2250 2196 2117 2103 1. firefox 105.0.1 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4374 5748 acpi-cpufreq ondemand 3000 4463 5635 acpi-cpufreq schedutil 3000 4761 5562 acpi-cpufreq performance 4500 5156 5731 amd-pstate performance 5499 5877 5881 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 30.8 58.8 89.7 acpi-cpufreq performance 26.9 54.4 89.8 amd-pstate schedutil 26.7 49.1 74.3 acpi-cpufreq schedutil 26.9 48.3 85.1 acpi-cpufreq ondemand 14.7 47.3 85.4 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1782 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 1687 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 1654 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1579 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1497 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 39.8 55.2 67.9 acpi-cpufreq performance 38.6 53.4 67.4 acpi-cpufreq schedutil 40.6 53.0 66.9 acpi-cpufreq ondemand 38.8 51.9 66.8 amd-pstate schedutil 38.5 46.9 65.6 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Firefox amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 400 800 1200 1600 2000 Min: 2368 / Avg: 2375.67 / Max: 2388 Min: 2236 / Avg: 2249.67 / Max: 2258 Min: 2185 / Avg: 2196 / Max: 2205 Min: 2114 / Avg: 2116.67 / Max: 2119 Min: 2096 / Avg: 2103 / Max: 2109 1. firefox 105.0.1
Tesseract Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Tesseract 2014-05-12 Resolution: 3840 x 2160 amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 200 400 600 800 1000 SE +/- 5.37, N = 3 SE +/- 0.28, N = 3 SE +/- 6.20, N = 3 SE +/- 11.01, N = 4 SE +/- 4.94, N = 3 820.27 882.44 908.49 916.06 919.08 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Tesseract 2014-05-12 Resolution: 3840 x 2160 amd-pstate schedutil amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 4 8 12 16 20 14.29 14.54 14.60 14.73 15.31 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5057 5707 acpi-cpufreq ondemand 3000 5209 5504 acpi-cpufreq schedutil 3000 5353 5505 acpi-cpufreq performance 4500 5389 5581 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Tesseract 2014-05-12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 34.8 63.0 73.7 acpi-cpufreq performance 26.7 62.9 72.4 acpi-cpufreq schedutil 26.7 59.9 69.7 acpi-cpufreq ondemand 26.1 59.3 69.0 amd-pstate schedutil 29.6 57.4 69.1 OpenBenchmarking.org Watts, Fewer Is Better Tesseract 2014-05-12 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 2205 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 2203 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 2097 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 2076 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 2028 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 39.8 48.0 60.8 acpi-cpufreq performance 38.1 47.3 60.0 amd-pstate performance 39.0 46.8 57.6 acpi-cpufreq ondemand 37.9 45.3 57.9 amd-pstate schedutil 37.9 44.1 60.0 OpenBenchmarking.org Celsius, Fewer Is Better Tesseract 2014-05-12 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Tesseract 2014-05-12 Resolution: 3840 x 2160 amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 160 320 480 640 800 Min: 810.4 / Avg: 820.27 / Max: 828.86 Min: 881.99 / Avg: 882.44 / Max: 882.94 Min: 897.99 / Avg: 908.49 / Max: 919.46 Min: 898.66 / Avg: 916.06 / Max: 948.16 Min: 911.07 / Avg: 919.08 / Max: 928.09
Selenium Result
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Google Chrome acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance amd-pstate performance 80 160 240 320 400 SE +/- 1.26, N = 4 SE +/- 1.34, N = 4 SE +/- 1.65, N = 4 SE +/- 0.64, N = 4 SE +/- 0.49, N = 4 383.6 381.2 380.9 357.5 354.3 1. chrome 106.0.5249.91 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 3131 5750 acpi-cpufreq ondemand 3000 3925 5700 acpi-cpufreq schedutil 3000 4370 5724 acpi-cpufreq performance 4500 4904 5749 amd-pstate performance 5499 5871 5881 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.0 49.1 72.8 amd-pstate schedutil 17.6 44.7 71.7 acpi-cpufreq performance 26.9 42.7 72.2 acpi-cpufreq schedutil 26.5 42.1 68.7 acpi-cpufreq ondemand 26.0 40.5 70.1 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 958 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 883 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 843 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 842 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 819 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 38.6 52.4 66.4 amd-pstate performance 38.5 52.2 66.0 acpi-cpufreq performance 38.4 50.7 63.3 amd-pstate schedutil 37.3 50.3 63.9 acpi-cpufreq ondemand 36.9 49.4 62.3 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Google Chrome acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance amd-pstate performance 70 140 210 280 350 Min: 381 / Avg: 383.6 / Max: 386.1 Min: 378.8 / Avg: 381.15 / Max: 385 Min: 376 / Avg: 380.85 / Max: 383.2 Min: 355.8 / Avg: 357.5 / Max: 358.8 Min: 352.9 / Avg: 354.25 / Max: 355.2 1. chrome 106.0.5249.91
Result
OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Google Chrome amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 600 1200 1800 2400 3000 SE +/- 6.33, N = 3 SE +/- 9.77, N = 3 SE +/- 3.84, N = 3 SE +/- 11.68, N = 3 SE +/- 4.67, N = 3 2654 2565 2522 2490 2481 1. chrome 106.0.5249.91 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 4365 5742 amd-pstate schedutil 400 4564 5751 acpi-cpufreq schedutil 3000 4783 5706 acpi-cpufreq performance 4500 5164 5729 amd-pstate performance 5554 5878 5881 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 35.8 55.4 80.4 amd-pstate schedutil 27.4 50.5 71.2 acpi-cpufreq performance 26.8 49.7 78.1 acpi-cpufreq schedutil 26.5 46.0 70.2 acpi-cpufreq ondemand 26.2 45.2 73.5 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1811 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 1786 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 1639 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1563 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1492 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 38.8 54.6 67.5 acpi-cpufreq schedutil 39.3 53.6 66.1 acpi-cpufreq performance 37.8 52.3 66.4 acpi-cpufreq ondemand 37.4 51.6 64.6 amd-pstate schedutil 37.5 47.5 64.5 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Google Chrome amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 500 1000 1500 2000 2500 Min: 2642 / Avg: 2654.33 / Max: 2663 Min: 2551 / Avg: 2565.33 / Max: 2584 Min: 2514 / Avg: 2521.67 / Max: 2526 Min: 2475 / Avg: 2490 / Max: 2513 Min: 2476 / Avg: 2480.67 / Max: 2490 1. chrome 106.0.5249.91
Result
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Google Chrome acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 70 140 210 280 350 SE +/- 1.43, N = 3 SE +/- 2.30, N = 3 SE +/- 1.27, N = 3 SE +/- 0.68, N = 3 SE +/- 1.20, N = 3 312.74 314.05 317.92 333.51 333.52 1. chrome 106.0.5249.91 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Google Chrome acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil 3 6 9 12 15 4.890 5.007 5.063 5.161 10.885 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 1000 2000 3000 4000 5000 Min: 400 / Avg: 2235.59 / Max: 5750 Min: 3000 / Avg: 4986.18 / Max: 5729 Min: 3000 / Avg: 5196.76 / Max: 5625 Min: 4500 / Avg: 5333.06 / Max: 5699 Min: 5399 / Avg: 5861.46 / Max: 5881 CPU Power Consumption
Min Avg Max amd-pstate performance 36.8 66.6 142.3 acpi-cpufreq performance 26.8 64.6 141.9 acpi-cpufreq schedutil 26.8 64.0 140.0 acpi-cpufreq ondemand 26.3 62.8 135.2 amd-pstate schedutil 11.4 28.9 140.7 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 4396 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 4265 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 4285 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 4291 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 42278 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 39.6 65.2 75.4 amd-pstate performance 39.3 64.8 73.0 acpi-cpufreq performance 38.0 63.6 71.8 acpi-cpufreq ondemand 38.1 63.5 72.9 amd-pstate schedutil 34.9 36.5 73.8 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Google Chrome acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 60 120 180 240 300 Min: 309.91 / Avg: 312.74 / Max: 314.48 Min: 311.21 / Avg: 314.05 / Max: 318.61 Min: 315.67 / Avg: 317.92 / Max: 320.07 Min: 332.43 / Avg: 333.51 / Max: 334.76 Min: 331.85 / Avg: 333.52 / Max: 335.85 1. chrome 106.0.5249.91
AOM AV1 This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 0.1013 0.2026 0.3039 0.4052 0.5065 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 0.43 0.44 0.44 0.45 0.45 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.0007 0.0014 0.0021 0.0028 0.0035 0.003 0.003 0.003 0.003 0.003 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4964 5675 acpi-cpufreq ondemand 3000 5076 5479 acpi-cpufreq schedutil 3000 5099 5455 acpi-cpufreq performance 4500 5154 5500 amd-pstate performance 5038 5529 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 26.2 158.8 234.6 amd-pstate performance 35.0 158.7 235.8 acpi-cpufreq schedutil 26.1 158.0 238.0 acpi-cpufreq ondemand 25.4 157.6 239.5 amd-pstate schedutil 26.4 155.8 235.5 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 60 120 180 240 300 1. acpi-cpufreq performance: Approximate power consumption of 7782 Joules per run. 2. amd-pstate performance: Approximate power consumption of 7882 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 7848 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 7775 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 7943 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 46.3 88.4 95.4 acpi-cpufreq schedutil 45.1 88.3 94.6 acpi-cpufreq performance 44.9 88.2 95.1 amd-pstate schedutil 44.3 87.2 94.0 acpi-cpufreq ondemand 44.3 87.2 94.8 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 1 2 3 4 5 Min: 0.43 / Avg: 0.43 / Max: 0.44 Min: 0.44 / Avg: 0.44 / Max: 0.44 Min: 0.44 / Avg: 0.44 / Max: 0.44 Min: 0.44 / Avg: 0.45 / Max: 0.45 Min: 0.44 / Avg: 0.45 / Max: 0.45 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Selenium Result
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Firefox acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance amd-pstate performance 110 220 330 440 550 SE +/- 5.02, N = 3 SE +/- 0.60, N = 3 SE +/- 1.77, N = 3 SE +/- 1.03, N = 3 SE +/- 4.32, N = 3 522.7 515.3 514.4 503.5 500.7 1. firefox 105.0.1 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 4154 5647 amd-pstate schedutil 400 4216 5753 acpi-cpufreq schedutil 3000 4833 5643 acpi-cpufreq performance 4500 5262 5737 amd-pstate performance 5500 5871 5881 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 18.9 58.1 74.4 amd-pstate performance 36.6 55.8 72.7 amd-pstate schedutil 27.3 49.9 71.3 acpi-cpufreq schedutil 26.5 47.5 67.4 acpi-cpufreq ondemand 26.3 44.8 64.8 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 20 40 60 80 100 1. acpi-cpufreq performance: Approximate power consumption of 2672 Joules per run. 2. amd-pstate performance: Approximate power consumption of 1396 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 1280 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1203 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1136 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq performance 37.6 56.9 63.6 amd-pstate performance 39.1 54.9 67.3 acpi-cpufreq schedutil 39.1 54.3 66.0 amd-pstate schedutil 37.9 52.7 66.5 acpi-cpufreq ondemand 37.6 51.7 63.9 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Firefox acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance amd-pstate performance 90 180 270 360 450 Min: 516 / Avg: 522.67 / Max: 532.5 Min: 514.1 / Avg: 515.3 / Max: 515.9 Min: 511 / Avg: 514.43 / Max: 516.9 Min: 501.5 / Avg: 503.53 / Max: 504.8 Min: 492.4 / Avg: 500.73 / Max: 506.9 1. firefox 105.0.1
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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 UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand 60 120 180 240 300 SE +/- 0.24, N = 3 SE +/- 0.30, N = 3 SE +/- 1.76, N = 3 SE +/- 0.07, N = 3 SE +/- 0.84, N = 3 248.0 248.0 253.5 256.5 258.8 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm UE Mb/s Per Watt
OpenBenchmarking.org UE Mb/s Per Watt, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand 0.9745 1.949 2.9235 3.898 4.8725 4.028 4.129 4.224 4.236 4.331 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5091 5676 acpi-cpufreq ondemand 3000 5300 5702 acpi-cpufreq performance 4500 5352 5529 acpi-cpufreq schedutil 3000 5368 5700 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better srsRAN 22.04.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 33.7 61.6 67.7 amd-pstate schedutil 26.5 60.6 68.1 acpi-cpufreq performance 26.6 60.1 68.0 acpi-cpufreq schedutil 25.9 60.0 68.2 acpi-cpufreq ondemand 25.4 59.8 67.3 OpenBenchmarking.org Watts, Fewer Is Better srsRAN 22.04.1 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1744 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 1675 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 1702 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1680 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1633 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.6 63.3 67.3 amd-pstate schedutil 39.5 61.5 63.9 amd-pstate performance 40.8 60.8 63.0 acpi-cpufreq performance 39.5 60.7 64.1 acpi-cpufreq ondemand 38.4 60.5 63.0 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN 22.04.1 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand 50 100 150 200 250 Min: 247.7 / Avg: 248.03 / Max: 248.5 Min: 247.4 / Avg: 248 / Max: 248.3 Min: 251.5 / Avg: 253.5 / Max: 257 Min: 256.4 / Avg: 256.47 / Max: 256.6 Min: 257.1 / Avg: 258.77 / Max: 259.7 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Selenium Result
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 40 80 120 160 200 SE +/- 1.60, N = 3 SE +/- 1.66, N = 3 SE +/- 2.35, N = 3 SE +/- 1.36, N = 3 SE +/- 0.97, N = 3 182.99 184.38 186.91 188.72 190.61 1. firefox 105.0.1 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 1.2148 2.4296 3.6444 4.8592 6.074 2.799 2.812 2.857 2.899 5.399 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance 1000 2000 3000 4000 5000 Min: 4500 / Avg: 4633.62 / Max: 5750 Min: 3000 / Avg: 5244.77 / Max: 5749 Min: 400 / Avg: 5263.69 / Max: 5746 Min: 3000 / Avg: 5325.85 / Max: 5706 Min: 5407 / Avg: 5863.52 / Max: 5881 CPU Power Consumption
Min Avg Max amd-pstate performance 34.9 67.1 124.9 amd-pstate schedutil 26.3 65.4 118.5 acpi-cpufreq schedutil 26.4 64.5 121.4 acpi-cpufreq ondemand 25.9 64.5 121.4 acpi-cpufreq performance 15.3 35.3 122.3 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 8032 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 8042 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 7917 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 7823 Joules per run. 5. acpi-cpufreq performance: Approximate power consumption of 54404 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 39.3 64.7 72.9 amd-pstate performance 38.5 64.6 72.3 acpi-cpufreq ondemand 38.1 63.7 72.5 amd-pstate schedutil 37.5 62.8 71.1 acpi-cpufreq performance 34.9 38.5 71.8 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 30 60 90 120 150 Min: 180.47 / Avg: 182.99 / Max: 185.96 Min: 182.17 / Avg: 184.38 / Max: 187.63 Min: 183.48 / Avg: 186.91 / Max: 191.4 Min: 186.48 / Avg: 188.72 / Max: 191.19 Min: 189 / Avg: 190.61 / Max: 192.36 1. firefox 105.0.1
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: PartialTweets acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.10, N = 3 SE +/- 0.07, N = 3 8.47 8.50 8.56 8.67 8.81 1. (CXX) g++ options: -O3 GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 2.0 Throughput Test: PartialTweets amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand 0.0313 0.0626 0.0939 0.1252 0.1565 0.135 0.135 0.135 0.136 0.139 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5315 5656 acpi-cpufreq schedutil 3000 5372 5475 acpi-cpufreq performance 4500 5373 5450 acpi-cpufreq ondemand 3000 5406 5625 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better simdjson 2.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 31.4 64.4 67.8 amd-pstate schedutil 26.8 63.5 67.7 acpi-cpufreq ondemand 25.8 63.3 68.0 acpi-cpufreq schedutil 24.7 62.9 67.0 acpi-cpufreq performance 22.5 62.6 66.8 OpenBenchmarking.org Watts, Fewer Is Better simdjson 2.0 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 4248 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 4214 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 4197 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 4175 Joules per run. 5. acpi-cpufreq performance: Approximate power consumption of 4133 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq performance 38.8 65.9 67.0 acpi-cpufreq schedutil 40.4 65.7 68.1 acpi-cpufreq ondemand 38.5 65.2 67.8 amd-pstate performance 40.1 64.3 65.9 amd-pstate schedutil 39.0 63.8 66.4 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 2.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: PartialTweets acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand 3 6 9 12 15 Min: 8.45 / Avg: 8.47 / Max: 8.5 Min: 8.47 / Avg: 8.5 / Max: 8.56 Min: 8.46 / Avg: 8.56 / Max: 8.7 Min: 8.55 / Avg: 8.67 / Max: 8.87 Min: 8.67 / Avg: 8.81 / Max: 8.91 1. (CXX) g++ options: -O3
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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 UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM amd-pstate performance acpi-cpufreq performance acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil 60 120 180 240 300 SE +/- 0.84, N = 4 SE +/- 0.58, N = 4 SE +/- 0.35, N = 4 SE +/- 1.69, N = 4 SE +/- 0.54, N = 5 256.8 258.1 258.7 263.3 266.9 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm UE Mb/s Per Watt
OpenBenchmarking.org UE Mb/s Per Watt, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate schedutil 1.0665 2.133 3.1995 4.266 5.3325 4.323 4.573 4.648 4.681 4.740 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4684 5724 acpi-cpufreq ondemand 3000 4899 5507 acpi-cpufreq performance 4500 5228 5500 acpi-cpufreq schedutil 3000 5249 5699 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better srsRAN 22.04.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.4 59.4 68.4 acpi-cpufreq ondemand 26.1 56.6 68.7 amd-pstate schedutil 13.5 56.3 68.2 acpi-cpufreq schedutil 25.9 56.2 69.0 acpi-cpufreq performance 26.3 55.5 67.3 OpenBenchmarking.org Watts, Fewer Is Better srsRAN 22.04.1 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 950 Joules per run. 2. acpi-cpufreq ondemand: Approximate power consumption of 891 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 856 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 886 Joules per run. 5. acpi-cpufreq performance: Approximate power consumption of 875 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 41.4 61.7 67.3 amd-pstate schedutil 40.1 61.0 65.9 amd-pstate performance 42.4 60.9 64.4 acpi-cpufreq performance 40.3 60.7 64.3 acpi-cpufreq ondemand 39.1 58.2 61.4 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN 22.04.1 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM amd-pstate performance acpi-cpufreq performance acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil 50 100 150 200 250 Min: 255.4 / Avg: 256.83 / Max: 258.6 Min: 256.5 / Avg: 258.05 / Max: 259.3 Min: 257.8 / Avg: 258.73 / Max: 259.5 Min: 260.5 / Avg: 263.25 / Max: 268.1 Min: 265.9 / Avg: 266.94 / Max: 268.8 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 1.0 Test: tConvolve OpenMP - Degridding acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil 1300 2600 3900 5200 6500 SE +/- 125.89, N = 2 SE +/- 64.45, N = 5 SE +/- 131.55, N = 2 SE +/- 100.46, N = 4 SE +/- 33.62, N = 4 5790.91 5893.41 5919.72 5954.90 6017.65 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp Million Grid Points Per Second Per Watt
OpenBenchmarking.org Million Grid Points Per Second Per Watt, More Is Better ASKAP 1.0 Test: tConvolve OpenMP - Degridding amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil 20 40 60 80 100 95.11 97.99 101.07 103.05 103.78 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4285 5604 acpi-cpufreq ondemand 3000 4722 5750 acpi-cpufreq schedutil 3000 4926 5750 acpi-cpufreq performance 4500 5222 5750 amd-pstate performance 5424 5842 5881 OpenBenchmarking.org Megahertz, More Is Better ASKAP 1.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 35.7 62.2 101.4 amd-pstate schedutil 26.9 60.1 101.7 acpi-cpufreq schedutil 26.3 58.0 99.5 acpi-cpufreq performance 26.6 57.8 100.4 acpi-cpufreq ondemand 26.1 57.3 100.5 OpenBenchmarking.org Watts, Fewer Is Better ASKAP 1.0 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1961 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 782 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 913 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 910 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1805 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 43.5 56.1 61.9 amd-pstate performance 44.1 54.5 59.4 acpi-cpufreq ondemand 42.6 54.3 60.5 acpi-cpufreq performance 43.0 53.9 58.6 amd-pstate schedutil 43.4 53.8 57.8 OpenBenchmarking.org Celsius, Fewer Is Better ASKAP 1.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 1.0 Test: tConvolve OpenMP - Degridding acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil 1000 2000 3000 4000 5000 Min: 5665.02 / Avg: 5790.91 / Max: 5916.8 Min: 5788.17 / Avg: 5893.41 / Max: 6051.27 Min: 5788.17 / Avg: 5919.72 / Max: 6051.27 Min: 5788.17 / Avg: 5954.9 / Max: 6192 Min: 5916.8 / Avg: 6017.65 / Max: 6051.27 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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 UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance 50 100 150 200 250 SE +/- 0.35, N = 3 SE +/- 0.74, N = 3 SE +/- 0.61, N = 3 SE +/- 0.50, N = 3 SE +/- 0.50, N = 3 236.4 236.5 236.9 238.0 245.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm UE Mb/s Per Watt
OpenBenchmarking.org UE Mb/s Per Watt, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance 0.9342 1.8684 2.8026 3.7368 4.671 3.780 3.897 3.983 3.993 4.152 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4951 5657 acpi-cpufreq ondemand 3000 5093 5500 acpi-cpufreq schedutil 3000 5315 5625 acpi-cpufreq performance 4500 5491 5749 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better srsRAN 22.04.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.6 62.7 69.4 amd-pstate schedutil 26.4 60.7 69.5 acpi-cpufreq schedutil 26.1 59.6 68.7 acpi-cpufreq ondemand 26.1 59.4 68.2 acpi-cpufreq performance 27.1 59.1 67.5 OpenBenchmarking.org Watts, Fewer Is Better srsRAN 22.04.1 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1504 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 1457 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 1411 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 1425 Joules per run. 5. acpi-cpufreq performance: Approximate power consumption of 1378 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq performance 39.5 63.4 66.8 acpi-cpufreq schedutil 40.6 62.5 66.9 amd-pstate performance 41.4 61.5 64.0 amd-pstate schedutil 39.8 60.6 64.5 acpi-cpufreq ondemand 38.5 59.1 61.6 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN 22.04.1 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance 40 80 120 160 200 Min: 236 / Avg: 236.4 / Max: 237.1 Min: 235.4 / Avg: 236.47 / Max: 237.9 Min: 235.7 / Avg: 236.9 / Max: 237.7 Min: 237.3 / Avg: 238.03 / Max: 239 Min: 244.3 / Avg: 245.3 / Max: 245.8 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
SMHasher SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil 20K 40K 60K 80K 100K SE +/- 552.89, N = 15 SE +/- 111.58, N = 7 SE +/- 606.35, N = 15 SE +/- 175.97, N = 8 SE +/- 222.76, N = 8 80547.87 81464.18 82461.39 83419.97 83465.14 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects cycles/hash
OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil 6 12 18 24 30 SE +/- 0.01, N = 15 SE +/- 0.01, N = 7 SE +/- 0.07, N = 15 SE +/- 0.03, N = 8 SE +/- 0.01, N = 8 25.62 25.57 25.06 24.90 24.86 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects Result Confidence
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil 14K 28K 42K 56K 70K Min: 72852.63 / Avg: 80547.87 / Max: 81636.04 Min: 80920.55 / Avg: 81464.18 / Max: 81663.29 Min: 74429.71 / Avg: 82461.39 / Max: 84238.97 Min: 82718.9 / Avg: 83419.97 / Max: 83930.56 Min: 82780.95 / Avg: 83465.14 / Max: 84232.94 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
Node.js V8 Web Tooling Benchmark Running the V8 project's Web-Tooling-Benchmark under Node.js. The Web-Tooling-Benchmark stresses JavaScript-related workloads common to web developers like Babel and TypeScript and Babylon. This test profile can test the system's JavaScript performance with Node.js. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org runs/s, More Is Better Node.js V8 Web Tooling Benchmark amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 5 10 15 20 25 SE +/- 0.25, N = 3 SE +/- 0.19, N = 3 SE +/- 0.22, N = 4 SE +/- 0.24, N = 4 SE +/- 0.15, N = 3 19.51 19.78 19.86 20.07 20.21 runs/s Per Watt
OpenBenchmarking.org runs/s Per Watt, More Is Better Node.js V8 Web Tooling Benchmark amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil 0.0689 0.1378 0.2067 0.2756 0.3445 0.295 0.299 0.300 0.301 0.306 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5299 5765 acpi-cpufreq ondemand 3000 5351 5744 acpi-cpufreq schedutil 3000 5402 5704 acpi-cpufreq performance 4500 5411 5728 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Node.js V8 Web Tooling Benchmark CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.0 67.5 76.9 acpi-cpufreq performance 25.7 66.6 76.1 acpi-cpufreq ondemand 25.8 66.1 74.9 amd-pstate schedutil 25.4 66.1 75.5 acpi-cpufreq schedutil 25.7 64.7 72.9 OpenBenchmarking.org Watts, Fewer Is Better Node.js V8 Web Tooling Benchmark CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 3827 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 3780 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 3768 Joules per run. 4. amd-pstate schedutil: Approximate power consumption of 3789 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 3729 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.4 64.0 67.8 amd-pstate performance 40.0 62.9 66.5 acpi-cpufreq ondemand 38.5 62.7 66.4 acpi-cpufreq performance 38.9 62.5 67.3 amd-pstate schedutil 39.3 62.5 66.4 OpenBenchmarking.org Celsius, Fewer Is Better Node.js V8 Web Tooling Benchmark CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org runs/s, More Is Better Node.js V8 Web Tooling Benchmark amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 5 10 15 20 25 Min: 19.01 / Avg: 19.51 / Max: 19.81 Min: 19.57 / Avg: 19.78 / Max: 20.15 Min: 19.21 / Avg: 19.86 / Max: 20.18 Min: 19.44 / Avg: 20.07 / Max: 20.59 Min: 19.95 / Avg: 20.21 / Max: 20.46
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: TopTweet acpi-cpufreq performance amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand 3 6 9 12 15 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 4 9.82 9.91 9.93 10.10 10.17 1. (CXX) g++ options: -O3 GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 2.0 Throughput Test: TopTweet amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand 0.0362 0.0724 0.1086 0.1448 0.181 0.154 0.155 0.158 0.160 0.161 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5337 5603 acpi-cpufreq ondemand 3000 5373 5657 acpi-cpufreq schedutil 3000 5384 5487 acpi-cpufreq performance 4500 5387 5475 amd-pstate performance 5746 5880 5881 OpenBenchmarking.org Megahertz, More Is Better simdjson 2.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 33.7 64.3 67.9 amd-pstate schedutil 26.5 63.3 67.3 acpi-cpufreq ondemand 25.4 63.3 67.8 acpi-cpufreq performance 17.7 63.2 67.6 acpi-cpufreq schedutil 25.9 63.0 67.2 OpenBenchmarking.org Watts, Fewer Is Better simdjson 2.0 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 4265 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 4177 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 4193 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 4215 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 4176 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.3 66.4 68.0 amd-pstate schedutil 38.8 65.5 67.6 acpi-cpufreq ondemand 38.6 65.3 68.5 acpi-cpufreq performance 38.8 64.1 66.8 amd-pstate performance 39.6 63.7 65.0 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 2.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: TopTweet acpi-cpufreq performance amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand 3 6 9 12 15 Min: 9.73 / Avg: 9.82 / Max: 9.97 Min: 9.81 / Avg: 9.91 / Max: 10.02 Min: 9.8 / Avg: 9.93 / Max: 10.02 Min: 9.89 / Avg: 10.1 / Max: 10.24 Min: 9.97 / Avg: 10.17 / Max: 10.47 1. (CXX) g++ options: -O3
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq performance 6 12 18 24 30 23.08 22.67 22.66 22.60 22.31 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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.
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. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.5 Resolution: 3840 x 2160 - Effects Quality: Ultimate acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil amd-pstate performance acpi-cpufreq performance 110 220 330 440 550 SE +/- 3.73, N = 3 SE +/- 2.68, N = 3 SE +/- 2.57, N = 3 SE +/- 1.86, N = 3 SE +/- 1.81, N = 3 510.17 510.85 512.40 522.16 524.88 MIN: 104 / MAX: 1075 MIN: 103 / MAX: 1051 MIN: 99 / MAX: 1064 MIN: 95 / MAX: 1085 MIN: 109 / MAX: 1089 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Xonotic 0.8.5 Resolution: 3840 x 2160 - Effects Quality: Ultimate acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance 2 4 6 8 10 7.405 7.435 7.599 7.734 7.839 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5297 5717 acpi-cpufreq ondemand 3000 5339 5725 acpi-cpufreq schedutil 3000 5416 5659 acpi-cpufreq performance 4500 5444 5730 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Xonotic 0.8.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.4 70.2 77.8 acpi-cpufreq schedutil 26.8 69.0 76.6 acpi-cpufreq ondemand 26.7 67.1 75.6 acpi-cpufreq performance 22.6 67.0 74.8 amd-pstate schedutil 29.5 66.3 74.3 OpenBenchmarking.org Watts, Fewer Is Better Xonotic 0.8.5 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 4869 Joules per run. 2. acpi-cpufreq schedutil: Approximate power consumption of 4875 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 4722 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 4598 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 4638 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.9 63.9 67.9 amd-pstate performance 40.0 62.4 66.9 acpi-cpufreq ondemand 38.9 61.0 65.5 amd-pstate schedutil 39.6 60.6 67.3 acpi-cpufreq performance 38.6 59.5 64.5 OpenBenchmarking.org Celsius, Fewer Is Better Xonotic 0.8.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.5 Resolution: 3840 x 2160 - Effects Quality: Ultimate acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil amd-pstate performance acpi-cpufreq performance 90 180 270 360 450 Min: 505.31 / Avg: 510.17 / Max: 517.49 Min: 507.98 / Avg: 510.85 / Max: 516.2 Min: 507.63 / Avg: 512.4 / Max: 516.42 Min: 518.44 / Avg: 522.16 / Max: 524.09 Min: 522.91 / Avg: 524.88 / Max: 528.49
SMHasher SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX acpi-cpufreq performance acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil 6K 12K 18K 24K 30K SE +/- 31.12, N = 7 SE +/- 25.80, N = 7 SE +/- 115.35, N = 7 SE +/- 78.33, N = 7 SE +/- 64.63, N = 7 29452.61 29562.66 29628.31 30287.07 30287.90 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects cycles/hash
OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX acpi-cpufreq performance acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil 8 16 24 32 40 SE +/- 0.03, N = 7 SE +/- 0.09, N = 7 SE +/- 0.17, N = 7 SE +/- 0.33, N = 7 SE +/- 0.05, N = 7 33.08 32.85 32.84 32.36 32.07 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects Result Confidence
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX acpi-cpufreq performance acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil 5K 10K 15K 20K 25K Min: 29364.08 / Avg: 29452.61 / Max: 29524.99 Min: 29505.68 / Avg: 29562.66 / Max: 29638.37 Min: 29345.26 / Avg: 29628.31 / Max: 30219.46 Min: 30077.6 / Avg: 30287.07 / Max: 30596.22 Min: 29992.57 / Avg: 30287.9 / Max: 30426.27 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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.
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Decompression Speed acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 1100 2200 3300 4400 5500 SE +/- 37.74, N = 3 SE +/- 12.90, N = 3 SE +/- 117.54, N = 3 SE +/- 35.91, N = 3 SE +/- 34.05, N = 3 5225.6 5247.5 5250.8 5269.8 5362.3 1. (CC) gcc options: -O3 -pthread -lz -llzma MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Decompression Speed amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance 14 28 42 56 70 62.36 62.94 63.12 63.19 64.52 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5216 5738 acpi-cpufreq ondemand 3000 5270 5749 acpi-cpufreq schedutil 3000 5327 5702 acpi-cpufreq performance 4500 5367 5700 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Zstd Compression 1.5.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 33.9 84.5 123.9 amd-pstate schedutil 26.4 83.1 124.1 acpi-cpufreq performance 26.6 83.1 123.6 acpi-cpufreq ondemand 23.8 83.1 123.4 acpi-cpufreq schedutil 26.6 83.0 122.4 OpenBenchmarking.org Watts, Fewer Is Better Zstd Compression 1.5.0 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 3493 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 3464 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 3435 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 3462 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 3432 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 41.4 64.4 69.4 acpi-cpufreq schedutil 41.1 64.3 68.6 acpi-cpufreq performance 40.3 64.1 69.9 acpi-cpufreq ondemand 39.5 63.9 69.3 amd-pstate schedutil 41.1 63.8 68.5 OpenBenchmarking.org Celsius, Fewer Is Better Zstd Compression 1.5.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Decompression Speed acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 900 1800 2700 3600 4500 Min: 5165.7 / Avg: 5225.57 / Max: 5295.3 Min: 5231.7 / Avg: 5247.53 / Max: 5273.1 Min: 5073.7 / Avg: 5250.77 / Max: 5473.2 Min: 5212.7 / Avg: 5269.83 / Max: 5336.1 Min: 5304.4 / Avg: 5362.3 / Max: 5422.3 1. (CC) gcc options: -O3 -pthread -lz -llzma
SMHasher SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance 9K 18K 27K 36K 45K SE +/- 43.08, N = 5 SE +/- 62.09, N = 5 SE +/- 62.68, N = 5 SE +/- 52.80, N = 5 SE +/- 67.80, N = 5 43121.91 43180.18 43195.03 43310.74 44237.88 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects cycles/hash
OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance 13 26 39 52 65 SE +/- 0.19, N = 5 SE +/- 0.09, N = 5 SE +/- 0.07, N = 5 SE +/- 0.04, N = 5 SE +/- 0.13, N = 5 56.34 56.26 56.07 55.99 54.70 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects Result Confidence
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance 8K 16K 24K 32K 40K Min: 43036.97 / Avg: 43121.91 / Max: 43226.18 Min: 42990.04 / Avg: 43180.18 / Max: 43297.72 Min: 43069.55 / Avg: 43195.03 / Max: 43348.52 Min: 43178.84 / Avg: 43310.74 / Max: 43447 Min: 44065.85 / Avg: 44237.88 / Max: 44366.73 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 1.0 Test: tConvolve MT - Degridding acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance 600 1200 1800 2400 3000 SE +/- 21.21, N = 3 SE +/- 13.26, N = 12 SE +/- 29.93, N = 3 SE +/- 28.60, N = 3 SE +/- 11.50, N = 3 2703.44 2731.61 2748.21 2748.46 2772.99 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp Million Grid Points Per Second Per Watt
OpenBenchmarking.org Million Grid Points Per Second Per Watt, More Is Better ASKAP 1.0 Test: tConvolve MT - Degridding acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil 7 14 21 28 35 28.74 28.88 29.09 29.12 29.14 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5319 5678 acpi-cpufreq schedutil 3000 5450 5750 acpi-cpufreq performance 4500 5464 5750 acpi-cpufreq ondemand 3000 5467 5750 amd-pstate performance 5431 5594 5881 OpenBenchmarking.org Megahertz, More Is Better ASKAP 1.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 30.6 95.2 119.5 acpi-cpufreq performance 14.5 94.6 119.3 amd-pstate schedutil 26.5 94.5 119.3 acpi-cpufreq schedutil 24.7 94.3 119.5 acpi-cpufreq ondemand 26.3 94.1 121.1 OpenBenchmarking.org Watts, Fewer Is Better ASKAP 1.0 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 9521 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 9757 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 9509 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 9462 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 9407 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.5 53.4 67.3 amd-pstate performance 40.8 52.8 63.9 acpi-cpufreq ondemand 39.3 52.6 66.3 amd-pstate schedutil 40.0 52.6 63.9 acpi-cpufreq performance 39.8 52.6 63.8 OpenBenchmarking.org Celsius, Fewer Is Better ASKAP 1.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 1.0 Test: tConvolve MT - Degridding acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance 500 1000 1500 2000 2500 Min: 2680.15 / Avg: 2703.44 / Max: 2745.79 Min: 2663.39 / Avg: 2731.61 / Max: 2794.42 Min: 2696.26 / Avg: 2748.21 / Max: 2799.93 Min: 2695.41 / Avg: 2748.46 / Max: 2793.51 Min: 2750.22 / Avg: 2772.99 / Max: 2787.11 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
SMHasher SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: FarmHash128 acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance 3K 6K 9K 12K 15K SE +/- 11.22, N = 5 SE +/- 18.51, N = 5 SE +/- 93.81, N = 5 SE +/- 68.99, N = 5 SE +/- 39.33, N = 5 14937.81 14943.61 15063.04 15294.65 15316.23 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects cycles/hash
OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: FarmHash128 acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate performance amd-pstate schedutil 13 26 39 52 65 SE +/- 0.00, N = 5 SE +/- 0.08, N = 5 SE +/- 0.31, N = 5 SE +/- 0.42, N = 5 SE +/- 0.06, N = 5 58.61 58.44 58.24 57.56 57.09 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects Result Confidence
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: FarmHash128 acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance 3K 6K 9K 12K 15K Min: 14913.12 / Avg: 14937.81 / Max: 14964.9 Min: 14897.86 / Avg: 14943.61 / Max: 14976.91 Min: 14962.23 / Avg: 15063.04 / Max: 15438.16 Min: 15162.58 / Avg: 15294.65 / Max: 15463.87 Min: 15249.97 / Avg: 15316.23 / Max: 15416.41 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
Intel Open Image Denoise Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RTLightmap.hdr.4096x4096 acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance 0.0968 0.1936 0.2904 0.3872 0.484 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 0.42 0.43 0.43 0.43 0.43 Images / Sec Per Watt
OpenBenchmarking.org Images / Sec Per Watt, More Is Better Intel Open Image Denoise 1.4.0 Run: RTLightmap.hdr.4096x4096 amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.0007 0.0014 0.0021 0.0028 0.0035 0.003 0.003 0.003 0.003 0.003 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 3955 5502 acpi-cpufreq schedutil 3000 4701 5449 acpi-cpufreq ondemand 3000 4920 5369 acpi-cpufreq performance 4500 4972 5725 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Intel Open Image Denoise 1.4.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate schedutil 25.0 125.9 188.0 amd-pstate performance 18.4 125.5 187.6 acpi-cpufreq performance 15.6 124.7 187.9 acpi-cpufreq ondemand 13.6 124.5 187.8 acpi-cpufreq schedutil 13.9 123.5 191.7 OpenBenchmarking.org Watts, Fewer Is Better Intel Open Image Denoise 1.4.0 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate schedutil: Approximate power consumption of 9313 Joules per run. 2. amd-pstate performance: Approximate power consumption of 9289 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 9187 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 9169 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 9094 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 42.1 74.9 89.4 amd-pstate schedutil 41.9 74.7 88.9 acpi-cpufreq performance 41.4 74.5 88.5 amd-pstate performance 41.8 73.7 88.9 acpi-cpufreq ondemand 40.5 73.7 87.9 OpenBenchmarking.org Celsius, Fewer Is Better Intel Open Image Denoise 1.4.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RTLightmap.hdr.4096x4096 acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance 1 2 3 4 5 Min: 0.42 / Avg: 0.42 / Max: 0.42 Min: 0.43 / Avg: 0.43 / Max: 0.43 Min: 0.42 / Avg: 0.43 / Max: 0.43 Min: 0.43 / Avg: 0.43 / Max: 0.43 Min: 0.43 / Avg: 0.43 / Max: 0.43
SMHasher SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: wyhash acpi-cpufreq performance amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil 5K 10K 15K 20K 25K SE +/- 74.95, N = 8 SE +/- 181.42, N = 8 SE +/- 149.26, N = 8 SE +/- 121.62, N = 8 SE +/- 167.25, N = 8 24657.61 24854.28 25089.14 25167.48 25195.96 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects cycles/hash
OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: wyhash acpi-cpufreq performance amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil 5 10 15 20 25 SE +/- 0.05, N = 8 SE +/- 0.14, N = 8 SE +/- 0.18, N = 8 SE +/- 0.15, N = 8 SE +/- 0.12, N = 8 18.31 18.20 18.04 17.90 17.85 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects Result Confidence
OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: wyhash acpi-cpufreq performance amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil 4K 8K 12K 16K 20K Min: 24200.71 / Avg: 24657.61 / Max: 24887.72 Min: 24071.69 / Avg: 24854.28 / Max: 25622.44 Min: 24617.77 / Avg: 25089.14 / Max: 25578.03 Min: 24758.07 / Avg: 25167.48 / Max: 25612.18 Min: 24562.48 / Avg: 25195.96 / Max: 25731.94 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
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 5.18 Build: defconfig acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance 8 16 24 32 40 SE +/- 0.31, N = 3 SE +/- 0.19, N = 3 SE +/- 0.25, N = 3 SE +/- 0.23, N = 3 SE +/- 0.23, N = 3 34.17 33.82 33.58 33.55 33.52 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4976 5750 acpi-cpufreq ondemand 3000 5050 5748 acpi-cpufreq schedutil 4500 5185 5747 acpi-cpufreq performance 4500 5220 5749 amd-pstate performance 5148 5419 5881 OpenBenchmarking.org Megahertz, More Is Better Timed Linux Kernel Compilation 5.18 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.6 158.6 217.3 amd-pstate schedutil 28.3 158.2 218.3 acpi-cpufreq performance 26.8 157.9 218.5 acpi-cpufreq ondemand 26.0 156.8 217.4 acpi-cpufreq schedutil 26.9 156.5 216.3 OpenBenchmarking.org Watts, Fewer Is Better Timed Linux Kernel Compilation 5.18 CPU Power Consumption Monitor 60 120 180 240 300 1. amd-pstate performance: Approximate power consumption of 6448 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 6485 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 6473 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 6482 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 6363 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 41.8 82.5 91.6 amd-pstate performance 41.3 82.3 91.4 amd-pstate schedutil 41.0 81.8 90.9 acpi-cpufreq performance 40.6 81.5 90.5 acpi-cpufreq ondemand 39.6 80.5 89.9 OpenBenchmarking.org Celsius, Fewer Is Better Timed Linux Kernel Compilation 5.18 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.18 Build: defconfig acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance 7 14 21 28 35 Min: 33.79 / Avg: 34.17 / Max: 34.78 Min: 33.6 / Avg: 33.82 / Max: 34.2 Min: 33.26 / Avg: 33.58 / Max: 34.07 Min: 33.28 / Avg: 33.55 / Max: 34.01 Min: 33.24 / Avg: 33.52 / Max: 33.98
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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 Samples / Second, More Is Better srsRAN 22.04.1 Test: OFDM_Test acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq ondemand 50M 100M 150M 200M 250M SE +/- 1411067.37, N = 3 SE +/- 57735.03, N = 3 SE +/- 1722723.94, N = 3 SE +/- 2716820.04, N = 3 SE +/- 1219744.96, N = 3 239466667 242900000 243433333 243433333 243533333 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Samples / Second Per Watt
OpenBenchmarking.org Samples / Second Per Watt, More Is Better srsRAN 22.04.1 Test: OFDM_Test amd-pstate performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand 900K 1800K 2700K 3600K 4500K 3957542.12 4045475.14 4091201.50 4133936.26 4137501.41 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4894 5476 acpi-cpufreq ondemand 3000 5051 5475 acpi-cpufreq schedutil 3000 5193 5475 acpi-cpufreq performance 4500 5332 5744 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better srsRAN 22.04.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 35.8 61.4 68.1 amd-pstate schedutil 25.7 60.2 69.0 acpi-cpufreq performance 26.0 58.9 67.1 acpi-cpufreq ondemand 25.7 58.9 68.9 acpi-cpufreq schedutil 25.4 58.5 67.6 OpenBenchmarking.org Watts, Fewer Is Better srsRAN 22.04.1 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1453 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 1424 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 1374 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 1393 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 1385 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.0 68.2 74.3 acpi-cpufreq performance 38.9 66.9 71.1 amd-pstate performance 39.8 66.9 70.4 amd-pstate schedutil 38.9 66.1 71.4 acpi-cpufreq ondemand 38.1 64.8 69.1 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN 22.04.1 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Samples / Second, More Is Better srsRAN 22.04.1 Test: OFDM_Test acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq ondemand 40M 80M 120M 160M 200M Min: 236800000 / Avg: 239466666.67 / Max: 241600000 Min: 242800000 / Avg: 242900000 / Max: 243000000 Min: 240000000 / Avg: 243433333.33 / Max: 245400000 Min: 238000000 / Avg: 243433333.33 / Max: 246200000 Min: 241100000 / Avg: 243533333.33 / Max: 244900000 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
PyPerformance PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: raytrace amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance 50 100 150 200 250 SE +/- 0.00, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 249 246 246 245 245 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5103 5747 acpi-cpufreq ondemand 3000 5177 5643 acpi-cpufreq schedutil 3000 5290 5659 acpi-cpufreq performance 4500 5344 5499 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better PyPerformance 1.0.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 29.9 61.9 66.1 amd-pstate schedutil 25.5 60.7 66.4 acpi-cpufreq schedutil 25.8 60.3 65.7 acpi-cpufreq performance 18.7 60.2 66.2 acpi-cpufreq ondemand 26.0 60.0 65.6 OpenBenchmarking.org Watts, Fewer Is Better PyPerformance 1.0.0 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 2787 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 2814 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 2752 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 2730 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 2741 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.0 70.6 74.4 amd-pstate performance 39.3 70.3 73.4 acpi-cpufreq performance 38.3 69.4 72.6 amd-pstate schedutil 36.4 69.1 72.6 acpi-cpufreq ondemand 37.9 68.7 72.5 OpenBenchmarking.org Celsius, Fewer Is Better PyPerformance 1.0.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: raytrace amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance 50 100 150 200 250 Min: 249 / Avg: 249 / Max: 249 Min: 245 / Avg: 245.67 / Max: 247 Min: 245 / Avg: 245.67 / Max: 246 Min: 244 / Avg: 244.67 / Max: 245 Min: 245 / Avg: 245 / Max: 245
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Node.js Express HTTP Load Test A Node.js Express server with a Node-based loadtest client for facilitating HTTP benchmarking. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Requests Per Second, More Is Better Node.js Express HTTP Load Test acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil 3K 6K 9K 12K 15K SE +/- 64.20, N = 6 SE +/- 30.85, N = 6 SE +/- 42.33, N = 6 SE +/- 36.09, N = 6 SE +/- 24.76, N = 6 13470 13626 13639 13640 13671 Requests Per Second Per Watt
OpenBenchmarking.org Requests Per Second Per Watt, More Is Better Node.js Express HTTP Load Test amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 50 100 150 200 250 212.46 217.01 221.17 227.33 235.15 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4329 5682 acpi-cpufreq ondemand 3000 4633 5573 acpi-cpufreq schedutil 3000 5046 5569 acpi-cpufreq performance 4500 5156 5688 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Node.js Express HTTP Load Test CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 33.5 64.2 83.9 amd-pstate schedutil 26.1 62.9 83.9 acpi-cpufreq performance 22.1 61.6 87.1 acpi-cpufreq schedutil 26.3 60.1 82.0 acpi-cpufreq ondemand 25.6 57.3 80.6 OpenBenchmarking.org Watts, Fewer Is Better Node.js Express HTTP Load Test CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 728 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 723 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 698 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 682 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 659 Joules per run. CPU Temp
Min Avg Max amd-pstate schedutil 38.9 55.8 60.5 amd-pstate performance 39.6 55.5 60.4 acpi-cpufreq schedutil 40.1 55.4 61.8 acpi-cpufreq performance 38.5 54.9 59.6 acpi-cpufreq ondemand 38.6 52.4 57.6 OpenBenchmarking.org Celsius, Fewer Is Better Node.js Express HTTP Load Test CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Requests Per Second, More Is Better Node.js Express HTTP Load Test acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil 2K 4K 6K 8K 10K Min: 13308 / Avg: 13469.83 / Max: 13716 Min: 13556 / Avg: 13625.5 / Max: 13742 Min: 13473 / Avg: 13638.83 / Max: 13746 Min: 13539 / Avg: 13639.83 / Max: 13772 Min: 13589 / Avg: 13671.17 / Max: 13730
oneDNN Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil 0.347 0.694 1.041 1.388 1.735 SE +/- 0.00771, N = 5 SE +/- 0.00854, N = 5 SE +/- 0.01428, N = 5 SE +/- 0.01264, N = 8 SE +/- 0.00756, N = 5 1.54215 1.53499 1.52516 1.52323 1.51969 MIN: 1.44 MIN: 1.45 MIN: 1.43 MIN: 1.44 MIN: 1.43 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4278 5663 acpi-cpufreq ondemand 3000 4755 5417 acpi-cpufreq schedutil 3000 4865 5411 acpi-cpufreq performance 4500 5075 5444 amd-pstate performance 1550 5381 5881 OpenBenchmarking.org Megahertz, More Is Better oneDNN 2.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate schedutil 26.9 104.4 153.5 acpi-cpufreq performance 27.0 103.4 152.4 acpi-cpufreq ondemand 26.1 102.6 151.6 acpi-cpufreq schedutil 26.4 102.3 153.6 amd-pstate performance 37.3 101.3 151.6 OpenBenchmarking.org Watts, Fewer Is Better oneDNN 2.7 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate schedutil: Approximate power consumption of 1371 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 1344 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 1355 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1350 Joules per run. 5. amd-pstate performance: Approximate power consumption of 1337 Joules per run. CPU Temp
Min Avg Max amd-pstate schedutil 40.6 63.4 75.5 acpi-cpufreq schedutil 41.0 62.8 75.9 acpi-cpufreq performance 40.1 62.0 75.5 acpi-cpufreq ondemand 39.1 60.9 73.8 amd-pstate performance 40.9 60.9 74.9 OpenBenchmarking.org Celsius, Fewer Is Better oneDNN 2.7 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil 2 4 6 8 10 Min: 1.52 / Avg: 1.54 / Max: 1.57 Min: 1.51 / Avg: 1.53 / Max: 1.56 Min: 1.5 / Avg: 1.53 / Max: 1.57 Min: 1.5 / Avg: 1.52 / Max: 1.61 Min: 1.51 / Avg: 1.52 / Max: 1.54 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
Y-Cruncher Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 1B acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 20.19 19.98 19.97 19.92 19.91 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4783 5707 acpi-cpufreq ondemand 3000 4968 5700 acpi-cpufreq schedutil 3000 5055 5565 acpi-cpufreq performance 4500 5145 5666 amd-pstate performance 4984 5514 5881 OpenBenchmarking.org Megahertz, More Is Better Y-Cruncher 0.7.10.9513 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 32.9 156.0 224.6 amd-pstate schedutil 27.3 155.6 233.8 acpi-cpufreq performance 27.3 155.2 234.8 acpi-cpufreq ondemand 26.3 154.9 229.4 acpi-cpufreq schedutil 26.8 154.5 230.3 OpenBenchmarking.org Watts, Fewer Is Better Y-Cruncher 0.7.10.9513 CPU Power Consumption Monitor 60 120 180 240 300 1. amd-pstate performance: Approximate power consumption of 3953 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 3941 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 3932 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 3975 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 3914 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 43.9 79.1 90.5 amd-pstate performance 43.9 78.7 90.5 amd-pstate schedutil 42.9 77.9 88.9 acpi-cpufreq performance 42.5 77.7 88.3 acpi-cpufreq ondemand 41.6 76.6 87.4 OpenBenchmarking.org Celsius, Fewer Is Better Y-Cruncher 0.7.10.9513 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 1B acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance 5 10 15 20 25 Min: 20.1 / Avg: 20.19 / Max: 20.3 Min: 19.96 / Avg: 19.98 / Max: 20.01 Min: 19.93 / Avg: 19.97 / Max: 20.01 Min: 19.91 / Avg: 19.92 / Max: 19.93 Min: 19.85 / Avg: 19.91 / Max: 19.99
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs 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 3.13 Binary: Pathtracer - Model: Asian Dragon acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil 8 16 24 32 40 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.18, N = 3 SE +/- 0.03, N = 3 SE +/- 0.15, N = 3 32.41 32.45 32.69 32.82 32.86 MIN: 32.19 / MAX: 32.85 MIN: 32.12 / MAX: 33.15 MIN: 32.3 / MAX: 33.41 MIN: 32.56 / MAX: 33.32 MIN: 32.49 / MAX: 33.56 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 3.13 Binary: Pathtracer - Model: Asian Dragon amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil 0.0448 0.0896 0.1344 0.1792 0.224 0.194 0.195 0.195 0.196 0.199 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4727 5473 acpi-cpufreq schedutil 3000 5008 5616 acpi-cpufreq ondemand 3000 5022 5395 acpi-cpufreq performance 4500 5136 5396 amd-pstate performance 5247 5392 5881 OpenBenchmarking.org Megahertz, More Is Better Embree 3.13 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.7 168.6 202.6 amd-pstate schedutil 28.3 167.5 203.0 acpi-cpufreq performance 26.6 166.6 203.2 acpi-cpufreq ondemand 26.0 166.5 203.4 acpi-cpufreq schedutil 26.1 164.8 202.4 OpenBenchmarking.org Watts, Fewer Is Better Embree 3.13 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 3934 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 3908 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 3943 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 3885 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 3791 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 40.1 74.1 80.6 acpi-cpufreq schedutil 40.0 74.1 80.9 amd-pstate schedutil 39.9 74.0 80.4 acpi-cpufreq performance 39.3 73.5 79.9 acpi-cpufreq ondemand 39.3 73.2 79.5 OpenBenchmarking.org Celsius, Fewer Is Better Embree 3.13 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Asian Dragon acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil 7 14 21 28 35 Min: 32.36 / Avg: 32.41 / Max: 32.47 Min: 32.31 / Avg: 32.45 / Max: 32.67 Min: 32.46 / Avg: 32.69 / Max: 33.05 Min: 32.76 / Avg: 32.82 / Max: 32.85 Min: 32.71 / Avg: 32.86 / Max: 33.15
ParaView This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Many Spheres - Resolution: 1920 x 1080 acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq ondemand 40 80 120 160 200 SE +/- 0.30, N = 4 SE +/- 0.35, N = 4 SE +/- 0.40, N = 4 SE +/- 0.96, N = 4 SE +/- 0.55, N = 4 159.55 160.16 160.35 161.19 161.75 MiPolys / Sec
OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.10.1 Test: Many Spheres - Resolution: 1920 x 1080 acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq ondemand 3K 6K 9K 12K 15K SE +/- 30.07, N = 4 SE +/- 34.68, N = 4 SE +/- 39.75, N = 4 SE +/- 96.07, N = 4 SE +/- 55.27, N = 4 15995.72 16056.91 16075.93 16160.32 16216.21 Result Confidence
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Many Spheres - Resolution: 1920 x 1080 acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq ondemand 30 60 90 120 150 Min: 159 / Avg: 159.55 / Max: 160.34 Min: 159.36 / Avg: 160.16 / Max: 160.88 Min: 159.69 / Avg: 160.35 / Max: 161.47 Min: 159.23 / Avg: 161.19 / Max: 163.08 Min: 160.48 / Avg: 161.75 / Max: 163.05
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Decompression Speed acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 1100 2200 3300 4400 5500 SE +/- 29.41, N = 3 SE +/- 101.87, N = 3 SE +/- 60.37, N = 3 SE +/- 68.60, N = 3 SE +/- 43.48, N = 3 5283.6 5317.0 5327.4 5352.9 5354.2 1. (CC) gcc options: -O3 -pthread -lz -llzma MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Decompression Speed amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil 13 26 39 52 65 57.09 57.41 57.52 58.09 58.16 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5187 5755 acpi-cpufreq ondemand 3000 5249 5750 acpi-cpufreq schedutil 3000 5323 5625 acpi-cpufreq performance 4500 5379 5725 amd-pstate performance 5385 5868 5881 OpenBenchmarking.org Megahertz, More Is Better Zstd Compression 1.5.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.7 93.8 135.3 amd-pstate schedutil 26.3 92.6 136.0 acpi-cpufreq performance 26.0 92.1 135.2 acpi-cpufreq ondemand 26.1 91.9 136.0 acpi-cpufreq schedutil 26.1 91.6 135.8 OpenBenchmarking.org Watts, Fewer Is Better Zstd Compression 1.5.0 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 4533 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 4476 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 4485 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 4471 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 4458 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 39.0 65.8 73.3 amd-pstate schedutil 39.0 65.4 74.1 amd-pstate performance 39.6 65.4 73.4 acpi-cpufreq performance 38.6 64.2 74.1 acpi-cpufreq ondemand 38.1 63.9 72.3 OpenBenchmarking.org Celsius, Fewer Is Better Zstd Compression 1.5.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Decompression Speed acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 900 1800 2700 3600 4500 Min: 5224.8 / Avg: 5283.6 / Max: 5314.5 Min: 5199.6 / Avg: 5317 / Max: 5519.9 Min: 5206.9 / Avg: 5327.43 / Max: 5393.8 Min: 5241.4 / Avg: 5352.9 / Max: 5477.9 Min: 5301.4 / Avg: 5354.17 / Max: 5440.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
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. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.5 Resolution: 1920 x 1080 - Effects Quality: Ultimate acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance 110 220 330 440 550 SE +/- 5.57, N = 3 SE +/- 3.48, N = 3 SE +/- 2.14, N = 3 SE +/- 1.28, N = 3 SE +/- 2.58, N = 3 523.20 523.64 524.80 527.60 530.00 MIN: 99 / MAX: 1072 MIN: 100 / MAX: 1066 MIN: 94 / MAX: 1048 MIN: 95 / MAX: 1066 MIN: 98 / MAX: 1091 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Xonotic 0.8.5 Resolution: 1920 x 1080 - Effects Quality: Ultimate amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance 2 4 6 8 10 7.496 7.589 7.686 7.709 7.749 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 5315 5698 amd-pstate schedutil 400 5385 5750 acpi-cpufreq schedutil 3000 5415 5693 acpi-cpufreq performance 4500 5450 5738 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Xonotic 0.8.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.0 69.9 76.9 amd-pstate schedutil 26.9 69.5 91.1 acpi-cpufreq performance 26.6 68.4 76.5 acpi-cpufreq ondemand 26.2 68.3 75.6 acpi-cpufreq schedutil 26.9 67.9 77.4 OpenBenchmarking.org Watts, Fewer Is Better Xonotic 0.8.5 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 4797 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 4728 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 4628 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 4666 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 4638 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 38.9 63.0 66.8 acpi-cpufreq schedutil 38.8 62.7 67.4 amd-pstate schedutil 37.4 61.7 66.6 acpi-cpufreq ondemand 37.4 61.5 65.6 acpi-cpufreq performance 37.6 60.7 66.3 OpenBenchmarking.org Celsius, Fewer Is Better Xonotic 0.8.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.5 Resolution: 1920 x 1080 - Effects Quality: Ultimate acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance 90 180 270 360 450 Min: 513.71 / Avg: 523.2 / Max: 532.99 Min: 517.45 / Avg: 523.64 / Max: 529.48 Min: 520.83 / Avg: 524.8 / Max: 528.17 Min: 525.17 / Avg: 527.6 / Max: 529.52 Min: 526.09 / Avg: 530 / Max: 534.86
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations Per Second, More Is Better ASKAP 1.0 Test: Hogbom Clean OpenMP acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand 100 200 300 400 500 SE +/- 1.34, N = 5 SE +/- 1.10, N = 5 SE +/- 2.67, N = 4 SE +/- 0.50, N = 4 SE +/- 0.51, N = 4 445.25 446.44 446.48 446.93 450.96 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp Iterations Per Second Per Watt
OpenBenchmarking.org Iterations Per Second Per Watt, More Is Better ASKAP 1.0 Test: Hogbom Clean OpenMP amd-pstate performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand 2 4 6 8 10 6.770 6.789 6.857 6.936 7.001 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4757 5691 acpi-cpufreq ondemand 3000 5055 5546 acpi-cpufreq schedutil 3000 5298 5742 acpi-cpufreq performance 4500 5303 5699 amd-pstate performance 3505 5777 5881 OpenBenchmarking.org Megahertz, More Is Better ASKAP 1.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 32.8 66.0 123.6 amd-pstate schedutil 26.8 65.8 122.6 acpi-cpufreq schedutil 25.9 64.9 122.9 acpi-cpufreq ondemand 25.3 64.4 122.9 acpi-cpufreq performance 26.1 64.4 124.6 OpenBenchmarking.org Watts, Fewer Is Better ASKAP 1.0 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 1022 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 1037 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 987 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 998 Joules per run. 5. acpi-cpufreq performance: Approximate power consumption of 991 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 41.9 57.5 61.5 acpi-cpufreq schedutil 41.5 56.4 61.8 acpi-cpufreq ondemand 40.1 55.9 60.6 amd-pstate schedutil 41.3 55.6 61.1 acpi-cpufreq performance 41.1 55.1 59.4 OpenBenchmarking.org Celsius, Fewer Is Better ASKAP 1.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Iterations Per Second, More Is Better ASKAP 1.0 Test: Hogbom Clean OpenMP acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand 80 160 240 320 400 Min: 440.53 / Avg: 445.25 / Max: 448.43 Min: 444.44 / Avg: 446.44 / Max: 450.45 Min: 438.6 / Avg: 446.48 / Max: 450.45 Min: 446.43 / Avg: 446.93 / Max: 448.43 Min: 450.45 / Avg: 450.96 / Max: 452.49 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Y-Cruncher Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 500M amd-pstate performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 3 6 9 12 15 SE +/- 0.090, N = 6 SE +/- 0.016, N = 5 SE +/- 0.013, N = 5 SE +/- 0.005, N = 5 SE +/- 0.013, N = 5 9.367 9.313 9.298 9.271 9.255 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4529 5682 acpi-cpufreq ondemand 3000 4639 5750 acpi-cpufreq schedutil 3000 4878 5528 acpi-cpufreq performance 4500 5105 5700 amd-pstate performance 1918 5437 5881 OpenBenchmarking.org Megahertz, More Is Better Y-Cruncher 0.7.10.9513 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max acpi-cpufreq schedutil 26.3 132.1 216.5 acpi-cpufreq performance 27.0 131.6 214.6 amd-pstate schedutil 19.0 130.8 212.6 amd-pstate performance 36.0 130.6 215.8 acpi-cpufreq ondemand 25.9 130.6 217.7 OpenBenchmarking.org Watts, Fewer Is Better Y-Cruncher 0.7.10.9513 CPU Power Consumption Monitor 60 120 180 240 300 1. acpi-cpufreq schedutil: Approximate power consumption of 1876 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 1843 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 1883 Joules per run. 4. amd-pstate performance: Approximate power consumption of 1872 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1854 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 39.3 73.1 84.9 amd-pstate performance 38.8 71.5 84.6 acpi-cpufreq performance 38.1 71.4 83.9 amd-pstate schedutil 38.0 71.1 83.9 acpi-cpufreq ondemand 37.1 70.6 83.8 OpenBenchmarking.org Celsius, Fewer Is Better Y-Cruncher 0.7.10.9513 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 500M amd-pstate performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 3 6 9 12 15 Min: 9.25 / Avg: 9.37 / Max: 9.82 Min: 9.28 / Avg: 9.31 / Max: 9.35 Min: 9.26 / Avg: 9.3 / Max: 9.32 Min: 9.26 / Avg: 9.27 / Max: 9.28 Min: 9.23 / Avg: 9.26 / Max: 9.29
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance 130 260 390 520 650 SE +/- 2.93, N = 4 SE +/- 2.87, N = 4 SE +/- 3.10, N = 4 SE +/- 4.78, N = 4 SE +/- 5.21, N = 7 595 590 590 590 588 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4722 5499 acpi-cpufreq ondemand 3000 4849 5429 acpi-cpufreq schedutil 3000 5066 5425 acpi-cpufreq performance 4500 5176 5439 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better PyBench 2018-02-16 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 26.2 57.9 65.8 amd-pstate schedutil 26.4 55.2 64.6 acpi-cpufreq performance 26.7 54.6 64.3 acpi-cpufreq schedutil 26.6 54.0 63.2 acpi-cpufreq ondemand 26.1 53.9 63.1 OpenBenchmarking.org Watts, Fewer Is Better PyBench 2018-02-16 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1034 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 994 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 982 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 971 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 956 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 38.9 71.2 75.8 acpi-cpufreq ondemand 36.9 69.5 74.3 acpi-cpufreq performance 37.5 69.1 74.1 amd-pstate performance 38.9 68.9 74.0 amd-pstate schedutil 37.6 67.7 72.9 OpenBenchmarking.org Celsius, Fewer Is Better PyBench 2018-02-16 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance 110 220 330 440 550 Min: 589 / Avg: 595.25 / Max: 602 Min: 582 / Avg: 589.75 / Max: 595 Min: 581 / Avg: 589.5 / Max: 595 Min: 579 / Avg: 589.75 / Max: 602 Min: 574 / Avg: 587.57 / Max: 616
ParaView This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Contour - Resolution: 1920 x 1080 amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil 110 220 330 440 550 SE +/- 3.18, N = 6 SE +/- 3.42, N = 6 SE +/- 3.31, N = 6 SE +/- 2.56, N = 6 SE +/- 2.20, N = 6 522.42 525.36 525.78 527.05 527.94 MiPolys / Sec
OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Contour - Resolution: 1920 x 1080 amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil 1200 2400 3600 4800 6000 SE +/- 33.11, N = 6 SE +/- 35.65, N = 6 SE +/- 34.48, N = 6 SE +/- 26.64, N = 6 SE +/- 22.93, N = 6 5444.24 5474.85 5479.22 5492.51 5501.75 Result Confidence
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Contour - Resolution: 1920 x 1080 amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil 90 180 270 360 450 Min: 513.33 / Avg: 522.42 / Max: 530.54 Min: 510.03 / Avg: 525.36 / Max: 531.09 Min: 515.07 / Avg: 525.78 / Max: 532.44 Min: 514.69 / Avg: 527.05 / Max: 531.18 Min: 517.21 / Avg: 527.94 / Max: 532.04
ParaView This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Many Spheres - Resolution: 3840 x 2160 acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance 30 60 90 120 150 SE +/- 0.36, N = 4 SE +/- 0.25, N = 4 SE +/- 0.07, N = 4 SE +/- 0.78, N = 4 SE +/- 0.40, N = 4 132.73 133.00 133.18 133.45 133.99 MiPolys / Sec
OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.10.1 Test: Many Spheres - Resolution: 3840 x 2160 acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance 3K 6K 9K 12K 15K SE +/- 36.06, N = 4 SE +/- 24.72, N = 4 SE +/- 7.56, N = 4 SE +/- 78.51, N = 4 SE +/- 40.17, N = 4 13306.52 13334.09 13351.40 13378.53 13432.89 Result Confidence
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Many Spheres - Resolution: 3840 x 2160 acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq schedutil amd-pstate schedutil acpi-cpufreq performance 30 60 90 120 150 Min: 131.99 / Avg: 132.73 / Max: 133.54 Min: 132.46 / Avg: 133 / Max: 133.58 Min: 132.96 / Avg: 133.18 / Max: 133.3 Min: 131.75 / Avg: 133.45 / Max: 135.04 Min: 133.08 / Avg: 133.99 / Max: 134.89
PyPerformance PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: pickle_pure_python acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq performance amd-pstate schedutil 50 100 150 200 250 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 222 222 222 221 220 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5033 5746 acpi-cpufreq ondemand 3000 5116 5607 acpi-cpufreq schedutil 3000 5258 5546 acpi-cpufreq performance 4500 5304 5603 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better PyPerformance 1.0.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 31.4 60.3 66.2 amd-pstate schedutil 26.6 59.6 66.3 acpi-cpufreq performance 25.6 58.5 65.9 acpi-cpufreq schedutil 26.2 58.0 65.1 acpi-cpufreq ondemand 25.6 57.5 65.3 OpenBenchmarking.org Watts, Fewer Is Better PyPerformance 1.0.0 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1729 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 2046 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 1658 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1661 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1666 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 39.9 69.2 72.9 acpi-cpufreq schedutil 40.4 69.2 72.9 acpi-cpufreq ondemand 38.9 68.0 72.0 acpi-cpufreq performance 38.6 67.6 71.6 amd-pstate schedutil 38.8 66.8 72.3 OpenBenchmarking.org Celsius, Fewer Is Better PyPerformance 1.0.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: pickle_pure_python acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq performance amd-pstate schedutil 40 80 120 160 200 Min: 222 / Avg: 222.33 / Max: 223 Min: 221 / Avg: 221.67 / Max: 222 Min: 221 / Avg: 222 / Max: 223 Min: 220 / Avg: 221.33 / Max: 223 Min: 220 / Avg: 220.33 / Max: 221
oneDNN Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance amd-pstate performance acpi-cpufreq schedutil 130 260 390 520 650 SE +/- 1.34, N = 3 SE +/- 0.90, N = 3 SE +/- 0.75, N = 3 SE +/- 0.46, N = 3 SE +/- 2.16, N = 3 587.34 584.40 583.72 583.45 582.15 MIN: 579.17 MIN: 578.72 MIN: 579.03 MIN: 579.33 MIN: 574.93 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4880 5499 acpi-cpufreq ondemand 3000 4920 5475 acpi-cpufreq schedutil 3000 4961 5325 acpi-cpufreq performance 4500 5001 5499 amd-pstate performance 4452 5068 5881 OpenBenchmarking.org Megahertz, More Is Better oneDNN 2.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.2 196.4 223.0 acpi-cpufreq ondemand 25.8 196.3 220.2 acpi-cpufreq performance 26.9 195.9 220.9 amd-pstate schedutil 26.2 195.7 218.7 acpi-cpufreq schedutil 27.1 194.7 222.5 OpenBenchmarking.org Watts, Fewer Is Better oneDNN 2.7 CPU Power Consumption Monitor 60 120 180 240 300 1. amd-pstate performance: Approximate power consumption of 15255 Joules per run. 2. acpi-cpufreq ondemand: Approximate power consumption of 15315 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 15213 Joules per run. 4. amd-pstate schedutil: Approximate power consumption of 15261 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 15188 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 48.5 87.9 91.9 acpi-cpufreq schedutil 49.0 87.8 91.8 amd-pstate schedutil 48.0 87.4 91.6 acpi-cpufreq performance 48.0 87.1 91.1 acpi-cpufreq ondemand 46.6 86.3 90.4 OpenBenchmarking.org Celsius, Fewer Is Better oneDNN 2.7 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance amd-pstate performance acpi-cpufreq schedutil 100 200 300 400 500 Min: 584.91 / Avg: 587.34 / Max: 589.52 Min: 582.69 / Avg: 584.4 / Max: 585.74 Min: 582.35 / Avg: 583.72 / Max: 584.94 Min: 582.59 / Avg: 583.45 / Max: 584.17 Min: 577.88 / Avg: 582.15 / Max: 584.83 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs 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 3.13 Binary: Pathtracer ISPC - Model: Crown acpi-cpufreq performance acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil 8 16 24 32 40 SE +/- 0.17, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 34.89 34.92 35.04 35.07 35.20 MIN: 34.23 / MAX: 35.97 MIN: 34.46 / MAX: 35.74 MIN: 34.54 / MAX: 35.98 MIN: 34.55 / MAX: 36.13 MIN: 34.8 / MAX: 36.11 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Crown acpi-cpufreq performance amd-pstate performance acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil 0.0434 0.0868 0.1302 0.1736 0.217 0.189 0.190 0.190 0.191 0.193 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4522 5280 acpi-cpufreq ondemand 3000 4805 5545 acpi-cpufreq schedutil 3000 4953 5625 acpi-cpufreq performance 4500 5048 5221 amd-pstate performance 5144 5325 5881 OpenBenchmarking.org Megahertz, More Is Better Embree 3.13 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 27.1 184.4 227.6 amd-pstate performance 38.1 184.3 227.8 amd-pstate schedutil 27.1 184.0 230.4 acpi-cpufreq ondemand 25.2 183.7 227.9 acpi-cpufreq schedutil 26.6 181.6 228.4 OpenBenchmarking.org Watts, Fewer Is Better Embree 3.13 CPU Power Consumption Monitor 60 120 180 240 300 1. acpi-cpufreq performance: Approximate power consumption of 4058 Joules per run. 2. amd-pstate performance: Approximate power consumption of 4115 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 4049 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 4103 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 4057 Joules per run. CPU Temp
Min Avg Max amd-pstate schedutil 45.9 84.6 90.8 acpi-cpufreq schedutil 46.3 84.5 91.1 amd-pstate performance 46.3 84.3 90.5 acpi-cpufreq performance 45.3 84.3 90.0 acpi-cpufreq ondemand 44.6 83.5 89.8 OpenBenchmarking.org Celsius, Fewer Is Better Embree 3.13 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Crown acpi-cpufreq performance acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil 8 16 24 32 40 Min: 34.56 / Avg: 34.89 / Max: 35.13 Min: 34.83 / Avg: 34.92 / Max: 34.98 Min: 34.93 / Avg: 35.04 / Max: 35.11 Min: 34.92 / Avg: 35.07 / Max: 35.19 Min: 35.15 / Avg: 35.2 / Max: 35.25
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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 UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil 50 100 150 200 250 SE +/- 0.23, N = 5 SE +/- 0.62, N = 5 SE +/- 1.18, N = 5 SE +/- 0.71, N = 5 SE +/- 1.14, N = 5 246.9 248.1 248.2 248.3 248.8 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm UE Mb/s Per Watt
OpenBenchmarking.org UE Mb/s Per Watt, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq ondemand acpi-cpufreq schedutil 1.0339 2.0678 3.1017 4.1356 5.1695 4.245 4.410 4.526 4.537 4.595 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4595 5604 acpi-cpufreq ondemand 3000 4826 5575 acpi-cpufreq schedutil 3000 5099 5657 acpi-cpufreq performance 4500 5192 5525 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better srsRAN 22.04.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 37.4 58.4 68.6 amd-pstate schedutil 26.1 56.0 69.2 acpi-cpufreq performance 14.3 54.9 68.4 acpi-cpufreq ondemand 26.1 54.7 68.4 acpi-cpufreq schedutil 26.5 54.1 85.3 OpenBenchmarking.org Watts, Fewer Is Better srsRAN 22.04.1 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 806 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 773 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 746 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 744 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 736 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 46.3 62.4 66.0 acpi-cpufreq schedutil 45.6 62.1 67.4 acpi-cpufreq performance 44.0 60.7 65.9 amd-pstate schedutil 44.4 60.4 64.3 acpi-cpufreq ondemand 43.0 59.9 64.0 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN 22.04.1 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil 50 100 150 200 250 Min: 246.4 / Avg: 246.86 / Max: 247.7 Min: 246.7 / Avg: 248.14 / Max: 250.2 Min: 245.8 / Avg: 248.22 / Max: 252.6 Min: 246.8 / Avg: 248.28 / Max: 250.3 Min: 246.8 / Avg: 248.84 / Max: 253.1 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
PyPerformance PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: regex_compile acpi-cpufreq schedutil amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.10, N = 3 SE +/- 0.15, N = 3 SE +/- 0.06, N = 3 79.0 79.0 78.9 78.7 78.4 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 5078 5541 amd-pstate schedutil 400 5103 5750 acpi-cpufreq schedutil 3000 5278 5547 acpi-cpufreq performance 4500 5307 5519 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better PyPerformance 1.0.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 32.7 60.7 66.6 amd-pstate schedutil 26.5 60.2 66.1 acpi-cpufreq schedutil 25.5 59.1 66.0 acpi-cpufreq performance 25.7 58.9 65.1 acpi-cpufreq ondemand 25.8 58.7 65.3 OpenBenchmarking.org Watts, Fewer Is Better PyPerformance 1.0.0 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 2004 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 2007 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 1950 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 1945 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1958 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.5 69.4 72.9 amd-pstate performance 40.3 68.6 72.4 acpi-cpufreq performance 39.3 68.4 71.6 amd-pstate schedutil 39.4 67.7 71.3 acpi-cpufreq ondemand 39.0 67.5 71.4 OpenBenchmarking.org Celsius, Fewer Is Better PyPerformance 1.0.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: regex_compile acpi-cpufreq schedutil amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 15 30 45 60 75 Min: 79 / Avg: 79.03 / Max: 79.1 Min: 78.8 / Avg: 78.97 / Max: 79.1 Min: 78.7 / Avg: 78.9 / Max: 79 Min: 78.5 / Avg: 78.7 / Max: 79 Min: 78.3 / Avg: 78.4 / Max: 78.5
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 4K acpi-cpufreq schedutil amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 20 40 60 80 100 SE +/- 0.62, N = 7 SE +/- 0.75, N = 7 SE +/- 0.64, N = 7 SE +/- 0.58, N = 7 SE +/- 0.75, N = 7 108.63 108.71 108.96 109.08 109.44 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 4K amd-pstate schedutil amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 0.2061 0.4122 0.6183 0.8244 1.0305 0.882 0.888 0.905 0.908 0.916 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 3884 5750 acpi-cpufreq ondemand 3000 4467 5600 acpi-cpufreq schedutil 3000 4877 5475 acpi-cpufreq performance 4500 4978 5550 amd-pstate performance 4734 5544 5881 OpenBenchmarking.org Megahertz, More Is Better SVT-VP9 0.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate schedutil 26.9 123.3 202.5 amd-pstate performance 35.7 122.7 202.2 acpi-cpufreq performance 26.2 120.9 206.3 acpi-cpufreq schedutil 26.3 119.6 200.8 acpi-cpufreq ondemand 26.2 119.1 203.4 OpenBenchmarking.org Watts, Fewer Is Better SVT-VP9 0.3 CPU Power Consumption Monitor 60 120 180 240 300 1. amd-pstate schedutil: Approximate power consumption of 1233 Joules per run. 2. amd-pstate performance: Approximate power consumption of 1263 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 1226 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1230 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1208 Joules per run. CPU Temp
Min Avg Max amd-pstate schedutil 42.0 74.9 85.3 acpi-cpufreq schedutil 42.8 74.9 85.5 amd-pstate performance 42.9 74.6 85.3 acpi-cpufreq performance 41.5 74.3 84.6 acpi-cpufreq ondemand 41.1 73.8 84.4 OpenBenchmarking.org Celsius, Fewer Is Better SVT-VP9 0.3 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 4K acpi-cpufreq schedutil amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 20 40 60 80 100 Min: 104.97 / Avg: 108.63 / Max: 109.61 Min: 104.24 / Avg: 108.71 / Max: 109.76 Min: 105.87 / Avg: 108.96 / Max: 110.01 Min: 105.65 / Avg: 109.08 / Max: 109.98 Min: 105.05 / Avg: 109.44 / Max: 110.71 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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.
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs 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 3.13 Binary: Pathtracer - Model: Crown acpi-cpufreq performance amd-pstate performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand 8 16 24 32 40 SE +/- 0.06, N = 3 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.08, N = 3 32.41 32.52 32.55 32.57 32.62 MIN: 32.04 / MAX: 33.19 MIN: 32.12 / MAX: 33.25 MIN: 32.18 / MAX: 33.39 MIN: 32.14 / MAX: 33.33 MIN: 32.19 / MAX: 33.45 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 3.13 Binary: Pathtracer - Model: Crown acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq schedutil 0.0403 0.0806 0.1209 0.1612 0.2015 0.175 0.176 0.177 0.178 0.179 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4644 5352 acpi-cpufreq ondemand 3000 4872 5369 acpi-cpufreq schedutil 3000 4915 5499 acpi-cpufreq performance 4500 5062 5500 amd-pstate performance 2853 5273 5881 OpenBenchmarking.org Megahertz, More Is Better Embree 3.13 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max acpi-cpufreq performance 14.0 185.6 227.9 amd-pstate schedutil 26.9 185.3 228.4 acpi-cpufreq ondemand 26.5 184.8 229.0 acpi-cpufreq schedutil 26.5 182.4 224.7 amd-pstate performance 36.9 182.3 225.1 OpenBenchmarking.org Watts, Fewer Is Better Embree 3.13 CPU Power Consumption Monitor 60 120 180 240 300 1. acpi-cpufreq performance: Approximate power consumption of 4330 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 4323 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 4312 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 4317 Joules per run. 5. amd-pstate performance: Approximate power consumption of 4253 Joules per run. CPU Temp
Min Avg Max amd-pstate schedutil 44.4 84.6 90.6 acpi-cpufreq schedutil 44.8 84.5 90.5 acpi-cpufreq performance 43.5 84.3 90.1 acpi-cpufreq ondemand 43.0 84.0 90.0 amd-pstate performance 44.6 83.9 90.5 OpenBenchmarking.org Celsius, Fewer Is Better Embree 3.13 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Crown acpi-cpufreq performance amd-pstate performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand 7 14 21 28 35 Min: 32.3 / Avg: 32.41 / Max: 32.49 Min: 32.37 / Avg: 32.52 / Max: 32.77 Min: 32.43 / Avg: 32.55 / Max: 32.6 Min: 32.39 / Avg: 32.57 / Max: 32.74 Min: 32.45 / Avg: 32.62 / Max: 32.7
ParaView This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Contour - Resolution: 3840 x 2160 amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate schedutil 80 160 240 320 400 SE +/- 1.64, N = 6 SE +/- 1.48, N = 6 SE +/- 1.37, N = 6 SE +/- 1.34, N = 6 SE +/- 0.24, N = 6 381.59 382.02 382.30 382.66 383.82 MiPolys / Sec
OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Contour - Resolution: 3840 x 2160 amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate schedutil 900 1800 2700 3600 4500 SE +/- 17.07, N = 6 SE +/- 15.39, N = 6 SE +/- 14.23, N = 6 SE +/- 13.93, N = 6 SE +/- 2.46, N = 6 3976.65 3981.09 3984.03 3987.74 3999.90 Result Confidence
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Contour - Resolution: 3840 x 2160 amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil amd-pstate schedutil 70 140 210 280 350 Min: 375.71 / Avg: 381.59 / Max: 384.46 Min: 374.67 / Avg: 382.02 / Max: 384.07 Min: 375.55 / Avg: 382.3 / Max: 384.45 Min: 375.99 / Avg: 382.66 / Max: 384.21 Min: 383.18 / Avg: 383.82 / Max: 384.52
oneDNN Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil amd-pstate performance 200 400 600 800 1000 SE +/- 2.67, N = 3 SE +/- 0.51, N = 3 SE +/- 0.58, N = 3 SE +/- 2.11, N = 3 SE +/- 1.79, N = 3 1143.65 1138.48 1138.19 1138.19 1137.70 MIN: 1132.73 MIN: 1134.1 MIN: 1133.01 MIN: 1130.67 MIN: 1131.02 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4983 5425 acpi-cpufreq ondemand 3000 5008 5525 acpi-cpufreq schedutil 3000 5055 5425 acpi-cpufreq performance 4423 5072 5375 amd-pstate performance 4727 5156 5881 OpenBenchmarking.org Megahertz, More Is Better oneDNN 2.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max acpi-cpufreq ondemand 25.7 198.0 222.8 amd-pstate performance 32.5 196.6 222.1 amd-pstate schedutil 29.5 196.3 219.0 acpi-cpufreq performance 24.7 196.2 221.6 acpi-cpufreq schedutil 26.4 196.2 222.1 OpenBenchmarking.org Watts, Fewer Is Better oneDNN 2.7 CPU Power Consumption Monitor 60 120 180 240 300 1. acpi-cpufreq ondemand: Approximate power consumption of 15643 Joules per run. 2. amd-pstate performance: Approximate power consumption of 15598 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 15510 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 15564 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 15496 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 45.0 86.6 90.6 amd-pstate schedutil 44.3 86.2 90.8 amd-pstate performance 44.8 86.0 90.6 acpi-cpufreq performance 44.3 85.9 90.1 acpi-cpufreq ondemand 42.9 85.3 89.6 OpenBenchmarking.org Celsius, Fewer Is Better oneDNN 2.7 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil amd-pstate performance 200 400 600 800 1000 Min: 1140.96 / Avg: 1143.65 / Max: 1149 Min: 1137.85 / Avg: 1138.48 / Max: 1139.49 Min: 1137.03 / Avg: 1138.19 / Max: 1138.78 Min: 1133.99 / Avg: 1138.19 / Max: 1140.64 Min: 1134.87 / Avg: 1137.7 / Max: 1141.01 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
Blender Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: BMW27 - Compute: CPU-Only acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq ondemand amd-pstate performance 11 22 33 44 55 SE +/- 0.12, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 49.14 49.08 49.06 49.01 48.89 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4838 5562 acpi-cpufreq ondemand 3000 4986 5557 acpi-cpufreq schedutil 3000 5036 5639 acpi-cpufreq performance 4500 5089 5599 amd-pstate performance 5096 5198 5881 OpenBenchmarking.org Megahertz, More Is Better Blender 3.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max acpi-cpufreq ondemand 25.9 202.0 228.6 amd-pstate schedutil 27.7 200.9 227.0 amd-pstate performance 36.5 200.9 227.2 acpi-cpufreq performance 26.6 200.8 227.4 acpi-cpufreq schedutil 26.7 199.1 229.0 OpenBenchmarking.org Watts, Fewer Is Better Blender 3.3 CPU Power Consumption Monitor 60 120 180 240 300 1. acpi-cpufreq ondemand: Approximate power consumption of 10706 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 10649 Joules per run. 3. amd-pstate performance: Approximate power consumption of 10645 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 10707 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 10554 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 40.1 91.1 95.3 amd-pstate schedutil 39.9 91.0 95.3 amd-pstate performance 40.3 90.9 95.4 acpi-cpufreq performance 39.4 90.3 94.6 acpi-cpufreq ondemand 39.0 90.1 94.6 OpenBenchmarking.org Celsius, Fewer Is Better Blender 3.3 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: BMW27 - Compute: CPU-Only acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq ondemand amd-pstate performance 10 20 30 40 50 Min: 48.94 / Avg: 49.14 / Max: 49.34 Min: 48.87 / Avg: 49.08 / Max: 49.25 Min: 48.85 / Avg: 49.06 / Max: 49.17 Min: 48.9 / Avg: 49.01 / Max: 49.14 Min: 48.83 / Avg: 48.89 / Max: 48.98
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
oneDNN Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU acpi-cpufreq ondemand amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil 0.0525 0.105 0.1575 0.21 0.2625 SE +/- 0.000316, N = 4 SE +/- 0.000451, N = 4 SE +/- 0.000183, N = 4 SE +/- 0.000092, N = 4 SE +/- 0.000446, N = 4 0.233125 0.232791 0.232555 0.232475 0.232161 MIN: 0.22 MIN: 0.22 MIN: 0.22 MIN: 0.22 MIN: 0.22 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4228 5499 acpi-cpufreq ondemand 3000 4628 5414 acpi-cpufreq schedutil 3000 4835 5404 acpi-cpufreq performance 4500 5014 5428 amd-pstate performance 4376 5346 5881 OpenBenchmarking.org Megahertz, More Is Better oneDNN 2.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate schedutil 29.6 132.2 224.3 amd-pstate performance 19.8 131.5 220.7 acpi-cpufreq ondemand 26.3 130.6 222.7 acpi-cpufreq performance 27.2 129.6 221.5 acpi-cpufreq schedutil 26.5 129.2 224.6 OpenBenchmarking.org Watts, Fewer Is Better oneDNN 2.7 CPU Power Consumption Monitor 60 120 180 240 300 1. amd-pstate schedutil: Approximate power consumption of 2115 Joules per run. 2. amd-pstate performance: Approximate power consumption of 2136 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 2089 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 2074 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 2068 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 43.4 69.4 85.9 amd-pstate schedutil 43.0 69.3 85.9 amd-pstate performance 42.8 69.0 85.5 acpi-cpufreq performance 42.6 68.6 85.4 acpi-cpufreq ondemand 41.4 67.7 84.4 OpenBenchmarking.org Celsius, Fewer Is Better oneDNN 2.7 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU acpi-cpufreq ondemand amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil 1 2 3 4 5 Min: 0.23 / Avg: 0.23 / Max: 0.23 Min: 0.23 / Avg: 0.23 / Max: 0.23 Min: 0.23 / Avg: 0.23 / Max: 0.23 Min: 0.23 / Avg: 0.23 / Max: 0.23 Min: 0.23 / Avg: 0.23 / Max: 0.23 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance amd-pstate performance 30 60 90 120 150 SE +/- 0.14, N = 7 SE +/- 0.12, N = 7 SE +/- 0.05, N = 7 SE +/- 0.09, N = 7 SE +/- 0.11, N = 7 113.35 113.43 113.45 113.68 113.81 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 4K amd-pstate performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand 0.2471 0.4942 0.7413 0.9884 1.2355 1.043 1.059 1.095 1.097 1.098 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4484 5650 acpi-cpufreq ondemand 3000 4627 5575 acpi-cpufreq schedutil 3000 4966 5575 acpi-cpufreq performance 4500 5023 5625 amd-pstate performance 5241 5610 5881 OpenBenchmarking.org Megahertz, More Is Better SVT-VP9 0.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.9 109.1 177.1 amd-pstate schedutil 26.6 107.1 172.9 acpi-cpufreq performance 26.9 103.6 173.6 acpi-cpufreq schedutil 26.4 103.5 175.5 acpi-cpufreq ondemand 26.2 103.3 173.0 OpenBenchmarking.org Watts, Fewer Is Better SVT-VP9 0.3 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 1091 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 1086 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 1036 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1050 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 1047 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 43.6 69.3 77.6 acpi-cpufreq schedutil 43.8 69.1 77.8 amd-pstate schedutil 43.6 68.7 77.0 acpi-cpufreq performance 42.9 68.0 76.4 acpi-cpufreq ondemand 42.6 67.6 76.5 OpenBenchmarking.org Celsius, Fewer Is Better SVT-VP9 0.3 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance amd-pstate performance 20 40 60 80 100 Min: 112.84 / Avg: 113.35 / Max: 113.76 Min: 113.06 / Avg: 113.43 / Max: 113.97 Min: 113.17 / Avg: 113.45 / Max: 113.63 Min: 113.38 / Avg: 113.68 / Max: 114.07 Min: 113.43 / Avg: 113.81 / Max: 114.19 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
7-Zip Compression This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq ondemand amd-pstate performance 40K 80K 120K 160K 200K SE +/- 336.47, N = 3 SE +/- 653.22, N = 3 SE +/- 462.37, N = 3 SE +/- 489.16, N = 3 SE +/- 187.74, N = 3 182652 182708 182787 182919 183185 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC MIPS Per Watt
OpenBenchmarking.org MIPS Per Watt, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance acpi-cpufreq schedutil 300 600 900 1200 1500 1194.61 1211.43 1212.28 1217.43 1227.43 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4875 5625 acpi-cpufreq ondemand 3000 4894 5725 acpi-cpufreq schedutil 3000 4964 5725 acpi-cpufreq performance 4500 5162 5750 amd-pstate performance 5189 5403 5881 OpenBenchmarking.org Megahertz, More Is Better 7-Zip Compression 22.01 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.6 153.3 207.4 acpi-cpufreq ondemand 26.3 150.9 211.0 amd-pstate schedutil 27.5 150.9 204.5 acpi-cpufreq performance 26.0 150.1 208.7 acpi-cpufreq schedutil 26.6 148.8 205.0 OpenBenchmarking.org Watts, Fewer Is Better 7-Zip Compression 22.01 CPU Power Consumption Monitor 60 120 180 240 300 1. amd-pstate performance: Approximate power consumption of 3629 Joules per run. 2. acpi-cpufreq ondemand: Approximate power consumption of 3571 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 3621 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 3502 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 3571 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 41.0 79.9 88.3 acpi-cpufreq schedutil 40.9 79.8 88.0 amd-pstate schedutil 40.5 79.5 87.6 acpi-cpufreq ondemand 39.6 79.1 87.1 acpi-cpufreq performance 39.9 78.8 87.6 OpenBenchmarking.org Celsius, Fewer Is Better 7-Zip Compression 22.01 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq ondemand amd-pstate performance 30K 60K 90K 120K 150K Min: 181998 / Avg: 182651.67 / Max: 183117 Min: 181957 / Avg: 182707.67 / Max: 184009 Min: 181892 / Avg: 182787 / Max: 183436 Min: 181966 / Avg: 182919 / Max: 183587 Min: 182848 / Avg: 183184.67 / Max: 183497 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs 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 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon acpi-cpufreq performance amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil 8 16 24 32 40 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 35.25 35.28 35.31 35.34 35.35 MIN: 34.84 / MAX: 36.48 MIN: 34.91 / MAX: 36.14 MIN: 34.87 / MAX: 36.25 MIN: 34.88 / MAX: 36.23 MIN: 35 / MAX: 36.38 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate schedutil acpi-cpufreq schedutil 0.0484 0.0968 0.1452 0.1936 0.242 0.212 0.213 0.213 0.215 0.215 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4646 5360 acpi-cpufreq ondemand 3000 4938 5650 acpi-cpufreq schedutil 3000 4962 5620 acpi-cpufreq performance 4500 5122 5331 amd-pstate performance 5242 5395 5881 OpenBenchmarking.org Megahertz, More Is Better Embree 3.13 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 37.2 166.5 203.3 acpi-cpufreq ondemand 26.4 166.1 204.4 acpi-cpufreq performance 27.5 165.5 203.6 acpi-cpufreq schedutil 26.0 164.2 202.8 amd-pstate schedutil 26.9 163.8 204.1 OpenBenchmarking.org Watts, Fewer Is Better Embree 3.13 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 3663 Joules per run. 2. acpi-cpufreq ondemand: Approximate power consumption of 3599 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 3586 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 3613 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 3604 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 46.4 76.7 82.1 amd-pstate performance 46.6 76.1 81.3 acpi-cpufreq performance 46.0 76.0 81.1 amd-pstate schedutil 46.8 76.0 81.5 acpi-cpufreq ondemand 45.3 75.3 80.3 OpenBenchmarking.org Celsius, Fewer Is Better Embree 3.13 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon acpi-cpufreq performance amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil 8 16 24 32 40 Min: 35.14 / Avg: 35.25 / Max: 35.36 Min: 35.2 / Avg: 35.28 / Max: 35.38 Min: 35.16 / Avg: 35.31 / Max: 35.41 Min: 35.16 / Avg: 35.34 / Max: 35.48 Min: 35.27 / Avg: 35.35 / Max: 35.39
Blender Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Barbershop - Compute: CPU-Only acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate performance amd-pstate schedutil 100 200 300 400 500 SE +/- 0.54, N = 3 SE +/- 0.80, N = 3 SE +/- 0.81, N = 3 SE +/- 0.53, N = 3 SE +/- 0.16, N = 3 470.73 470.61 470.43 469.66 469.56 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better Blender 3.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 1000 2000 3000 4000 5000 Min: 400 / Avg: 5089.12 / Max: 5575 Min: 3000 / Avg: 5101.6 / Max: 5747 Min: 3000 / Avg: 5109.28 / Max: 5592 Min: 4500 / Avg: 5118.05 / Max: 5719 Min: 5095 / Avg: 5134.49 / Max: 5881 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Blender 3.3 CPU Power Consumption Monitor acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil 40 80 120 160 200 Min: 26.48 / Avg: 208.55 / Max: 221.81 Min: 26.22 / Avg: 208.41 / Max: 221.78 Min: 27.21 / Avg: 208.32 / Max: 221.98 Min: 35.46 / Avg: 207.94 / Max: 221.07 Min: 25.84 / Avg: 206.29 / Max: 218.89 1. acpi-cpufreq performance: Approximate power consumption of 98576 Joules per run. 2. acpi-cpufreq ondemand: Approximate power consumption of 98439 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 98329 Joules per run. 4. amd-pstate performance: Approximate power consumption of 98149 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 97643 Joules per run. CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better Blender 3.3 CPU Temperature Monitor amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand 20 40 60 80 100 Min: 48.38 / Avg: 94.28 / Max: 95.13 Min: 48.38 / Avg: 94.16 / Max: 95.38 Min: 48.75 / Avg: 94.09 / Max: 95 Min: 47.63 / Avg: 93.73 / Max: 94.88 Min: 47.63 / Avg: 92.98 / Max: 94.5 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Barbershop - Compute: CPU-Only acpi-cpufreq schedutil acpi-cpufreq performance acpi-cpufreq ondemand amd-pstate performance amd-pstate schedutil 80 160 240 320 400 Min: 469.7 / Avg: 470.73 / Max: 471.51 Min: 469.16 / Avg: 470.61 / Max: 471.93 Min: 469.06 / Avg: 470.43 / Max: 471.85 Min: 469.09 / Avg: 469.66 / Max: 470.72 Min: 469.28 / Avg: 469.56 / Max: 469.83
GravityMark Result
OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.72 Resolution: 1920 x 1080 - Renderer: Vulkan amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 50 100 150 200 250 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.13, N = 3 SE +/- 0.19, N = 3 SE +/- 0.07, N = 3 220.6 220.6 220.6 220.7 220.8 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better GravityMark 1.72 Resolution: 1920 x 1080 - Renderer: Vulkan amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 2 4 6 8 10 4.078 4.396 4.824 5.691 6.290 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 3056 5750 amd-pstate schedutil 400 4219 5757 acpi-cpufreq schedutil 3000 4265 5750 acpi-cpufreq performance 4500 4858 5754 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better GravityMark 1.72 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 28.1 54.1 69.6 amd-pstate schedutil 24.2 50.2 67.6 acpi-cpufreq performance 24.9 45.8 67.7 acpi-cpufreq schedutil 19.8 38.8 68.7 acpi-cpufreq ondemand 17.8 35.1 66.7 OpenBenchmarking.org Watts, Fewer Is Better GravityMark 1.72 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 9327 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 8631 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 7888 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 6680 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 6044 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 38.5 40.3 65.3 acpi-cpufreq schedutil 38.5 40.2 66.9 amd-pstate schedutil 37.4 39.8 65.3 acpi-cpufreq performance 37.3 38.9 64.1 acpi-cpufreq ondemand 36.1 37.8 63.6 OpenBenchmarking.org Celsius, Fewer Is Better GravityMark 1.72 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.72 Resolution: 1920 x 1080 - Renderer: Vulkan amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 40 80 120 160 200 Min: 220.6 / Avg: 220.63 / Max: 220.7 Min: 220.5 / Avg: 220.57 / Max: 220.7 Min: 220.5 / Avg: 220.63 / Max: 220.9 Min: 220.5 / Avg: 220.73 / Max: 221.1 Min: 220.7 / Avg: 220.77 / Max: 220.9
Intel Open Image Denoise Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.hdr_alb_nrm.3840x2160 amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.2003 0.4006 0.6009 0.8012 1.0015 SE +/- 0.00, N = 3 SE +/- 0.01, N = 12 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.89 0.89 0.89 0.89 0.89 Images / Sec Per Watt
OpenBenchmarking.org Images / Sec Per Watt, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.hdr_alb_nrm.3840x2160 amd-pstate performance amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.0018 0.0036 0.0054 0.0072 0.009 0.007 0.008 0.008 0.008 0.008 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 3674 5634 acpi-cpufreq schedutil 3000 4800 5646 acpi-cpufreq ondemand 3000 4883 5500 acpi-cpufreq performance 4500 4934 5456 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Intel Open Image Denoise 1.4.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 20.2 120.0 189.4 amd-pstate schedutil 25.6 118.6 185.1 acpi-cpufreq performance 14.6 118.2 186.4 acpi-cpufreq schedutil 13.4 118.1 186.1 acpi-cpufreq ondemand 23.4 117.6 192.4 OpenBenchmarking.org Watts, Fewer Is Better Intel Open Image Denoise 1.4.0 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 4560 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 4467 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 4454 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 4410 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 4429 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 47.0 77.5 90.6 acpi-cpufreq performance 46.6 76.8 90.3 acpi-cpufreq ondemand 45.8 76.2 89.5 amd-pstate schedutil 47.3 76.2 90.5 amd-pstate performance 44.6 75.7 90.3 OpenBenchmarking.org Celsius, Fewer Is Better Intel Open Image Denoise 1.4.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.hdr_alb_nrm.3840x2160 amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 2 4 6 8 10 Min: 0.89 / Avg: 0.89 / Max: 0.9 Min: 0.81 / Avg: 0.89 / Max: 0.9 Min: 0.89 / Avg: 0.89 / Max: 0.89 Min: 0.89 / Avg: 0.89 / Max: 0.89 Min: 0.89 / Avg: 0.89 / Max: 0.89
CPU Temperature Monitor OpenBenchmarking.org Celsius CPU Temperature Monitor Phoronix Test Suite System Monitoring amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq performance 20 40 60 80 100 Min: 36.38 / Avg: 65.91 / Max: 95.38 Min: 37.13 / Avg: 65.19 / Max: 95.25 Min: 35.5 / Avg: 64.58 / Max: 94.75 Min: 34.75 / Avg: 56.3 / Max: 95.25 Min: 34.88 / Avg: 55.96 / Max: 95.13
CPU Power Consumption Monitor OpenBenchmarking.org Watts CPU Power Consumption Monitor Phoronix Test Suite System Monitoring amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate schedutil 40 80 120 160 200 Min: 18.39 / Avg: 98.73 / Max: 235.8 Min: 13.29 / Avg: 94.19 / Max: 239.51 Min: 13.37 / Avg: 91.54 / Max: 237.99 Min: 13.21 / Avg: 76.99 / Max: 234.8 Min: 11.36 / Avg: 75.84 / Max: 235.46
CPU Peak Freq (Highest CPU Core Frequency) Monitor OpenBenchmarking.org Megahertz CPU Peak Freq (Highest CPU Core Frequency) Monitor Phoronix Test Suite System Monitoring amd-pstate performance acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate schedutil 1000 2000 3000 4000 5000 Min: 983 / Avg: 5692.09 / Max: 5881 Min: 560 / Avg: 4973.17 / Max: 5754 Min: 3000 / Avg: 4943.95 / Max: 5750 Min: 3000 / Avg: 4632.13 / Max: 5750 Min: 400 / Avg: 3959.27 / Max: 5806
srsRAN Min Avg Max acpi-cpufreq performance 39.3 60.0 64.3 acpi-cpufreq schedutil 40.6 59.7 64.6 amd-pstate performance 40.5 59.0 63.3 amd-pstate schedutil 39.4 58.7 62.6 acpi-cpufreq ondemand 38.5 58.1 61.9 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN 22.04.1 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 36.1 58.1 67.8 amd-pstate schedutil 25.8 56.3 69.1 acpi-cpufreq schedutil 25.9 55.6 68.6 acpi-cpufreq ondemand 26.1 55.1 68.1 acpi-cpufreq performance 26.6 54.8 67.5 OpenBenchmarking.org Watts, Fewer Is Better srsRAN 22.04.1 CPU Power Consumption Monitor 20 40 60 80 100
Min Avg Max amd-pstate schedutil 400 4669 5749 acpi-cpufreq ondemand 3000 4847 5651 acpi-cpufreq schedutil 3000 5148 5700 acpi-cpufreq performance 4500 5305 5700 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better srsRAN 22.04.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org UE Mb/s Per Watt, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.5864 1.1728 1.7592 2.3456 2.932 2.370 2.402 2.505 2.560 2.606
Result
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance 30 60 90 120 150 SE +/- 4.07, N = 5 SE +/- 0.27, N = 5 SE +/- 0.87, N = 5 SE +/- 0.70, N = 5 SE +/- 1.18, N = 5 131.8 139.5 141.1 141.2 142.9 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm UE Mb/s Per Watt
OpenBenchmarking.org UE Mb/s Per Watt, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.5864 1.1728 1.7592 2.3456 2.932 2.370 2.402 2.505 2.560 2.606 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4669 5749 acpi-cpufreq ondemand 3000 4847 5651 acpi-cpufreq schedutil 3000 5148 5700 acpi-cpufreq performance 4500 5305 5700 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better srsRAN 22.04.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.1 58.1 67.8 amd-pstate schedutil 25.8 56.3 69.1 acpi-cpufreq schedutil 25.9 55.6 68.6 acpi-cpufreq ondemand 26.1 55.1 68.1 acpi-cpufreq performance 26.6 54.8 67.5 OpenBenchmarking.org Watts, Fewer Is Better srsRAN 22.04.1 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 848 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 811 Joules per run. 3. acpi-cpufreq schedutil: Approximate power consumption of 823 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 783 Joules per run. 5. acpi-cpufreq performance: Approximate power consumption of 779 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq performance 39.3 60.0 64.3 acpi-cpufreq schedutil 40.6 59.7 64.6 amd-pstate performance 40.5 59.0 63.3 amd-pstate schedutil 39.4 58.7 62.6 acpi-cpufreq ondemand 38.5 58.1 61.9 OpenBenchmarking.org Celsius, Fewer Is Better srsRAN 22.04.1 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM acpi-cpufreq schedutil amd-pstate performance amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq performance 30 60 90 120 150 Min: 124.2 / Avg: 131.76 / Max: 143.2 Min: 138.8 / Avg: 139.48 / Max: 140.4 Min: 138 / Avg: 141.06 / Max: 142.9 Min: 139.4 / Avg: 141.16 / Max: 143.7 Min: 138.3 / Avg: 142.9 / Max: 144.6 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Min Avg Max acpi-cpufreq schedutil 44.6 60.5 80.4 amd-pstate schedutil 44.1 60.4 79.9 amd-pstate performance 45.3 60.1 80.0 acpi-cpufreq ondemand 43.9 59.9 81.3 acpi-cpufreq performance 44.3 59.9 79.5 OpenBenchmarking.org Celsius, Fewer Is Better ASKAP 1.0 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 36.1 107.4 172.4 acpi-cpufreq ondemand 25.9 106.2 174.4 amd-pstate schedutil 26.8 106.0 172.8 acpi-cpufreq performance 26.4 105.9 173.4 acpi-cpufreq schedutil 27.1 105.0 173.5 OpenBenchmarking.org Watts, Fewer Is Better ASKAP 1.0 CPU Power Consumption Monitor 50 100 150 200 250
Min Avg Max amd-pstate schedutil 400 5075 5713 acpi-cpufreq ondemand 3000 5219 5496 acpi-cpufreq schedutil 3000 5253 5482 acpi-cpufreq performance 4500 5321 5502 amd-pstate performance 5446 5877 5881 OpenBenchmarking.org Megahertz, More Is Better ASKAP 1.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Mpix/sec Per Watt, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance 20 40 60 80 100 103.31 105.37 105.52 105.70 106.93
Result
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil amd-pstate performance acpi-cpufreq performance 2K 4K 6K 8K 10K SE +/- 463.30, N = 3 SE +/- 287.45, N = 3 SE +/- 0.00, N = 3 SE +/- 80.90, N = 3 SE +/- 80.90, N = 3 10973.3 11101.7 11165.8 11327.6 11327.6 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp Mpix/sec Per Watt
OpenBenchmarking.org Mpix/sec Per Watt, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance 20 40 60 80 100 103.31 105.37 105.52 105.70 106.93 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 5075 5713 acpi-cpufreq ondemand 3000 5219 5496 acpi-cpufreq schedutil 3000 5253 5482 acpi-cpufreq performance 4500 5321 5502 amd-pstate performance 5446 5877 5881 OpenBenchmarking.org Megahertz, More Is Better ASKAP 1.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.1 107.4 172.4 acpi-cpufreq ondemand 25.9 106.2 174.4 amd-pstate schedutil 26.8 106.0 172.8 acpi-cpufreq performance 26.4 105.9 173.4 acpi-cpufreq schedutil 27.1 105.0 173.5 OpenBenchmarking.org Watts, Fewer Is Better ASKAP 1.0 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 3507 Joules per run. 2. acpi-cpufreq ondemand: Approximate power consumption of 3470 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 3462 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 3460 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 3396 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 44.6 60.5 80.4 amd-pstate schedutil 44.1 60.4 79.9 amd-pstate performance 45.3 60.1 80.0 acpi-cpufreq ondemand 43.9 59.9 81.3 acpi-cpufreq performance 44.3 59.9 79.5 OpenBenchmarking.org Celsius, Fewer Is Better ASKAP 1.0 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding acpi-cpufreq ondemand acpi-cpufreq schedutil amd-pstate schedutil amd-pstate performance acpi-cpufreq performance 2K 4K 6K 8K 10K Min: 10092.1 / Avg: 10973.3 / Max: 11662 Min: 10710 / Avg: 11101.73 / Max: 11662 Min: 11165.8 / Avg: 11165.8 / Max: 11165.8 Min: 11165.8 / Avg: 11327.6 / Max: 11408.5 Min: 11165.8 / Avg: 11327.6 / Max: 11408.5 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
AOM AV1 Min Avg Max acpi-cpufreq performance 41.1 68.5 75.9 acpi-cpufreq ondemand 40.5 68.5 74.6 amd-pstate performance 41.9 66.8 74.0 acpi-cpufreq schedutil 40.9 66.5 75.5 amd-pstate schedutil 37.5 51.2 60.8 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 37.5 108.0 139.3 acpi-cpufreq performance 13.6 108.0 144.7 acpi-cpufreq ondemand 13.9 107.3 144.0 acpi-cpufreq schedutil 14.2 99.9 145.7 amd-pstate schedutil 26.1 72.2 110.8 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 40 80 120 160 200
Min Avg Max amd-pstate schedutil 400 4536 5696 acpi-cpufreq ondemand 3000 4897 5577 acpi-cpufreq schedutil 3000 5148 5499 acpi-cpufreq performance 3610 5171 5749 amd-pstate performance 4870 5535 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance amd-pstate schedutil 0.1253 0.2506 0.3759 0.5012 0.6265 0.426 0.488 0.494 0.507 0.557
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 12 24 36 48 60 SE +/- 0.45, N = 4 SE +/- 1.19, N = 12 SE +/- 1.06, N = 15 SE +/- 0.24, N = 4 SE +/- 0.55, N = 15 40.25 42.58 52.36 53.35 54.69 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance amd-pstate schedutil 0.1253 0.2506 0.3759 0.5012 0.6265 0.426 0.488 0.494 0.507 0.557 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4536 5696 acpi-cpufreq ondemand 3000 4897 5577 acpi-cpufreq schedutil 3000 5148 5499 acpi-cpufreq performance 3610 5171 5749 amd-pstate performance 4870 5535 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 37.5 108.0 139.3 acpi-cpufreq performance 13.6 108.0 144.7 acpi-cpufreq ondemand 13.9 107.3 144.0 acpi-cpufreq schedutil 14.2 99.9 145.7 amd-pstate schedutil 26.1 72.2 110.8 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 1728 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 1684 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 1724 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1874 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 1409 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq performance 41.1 68.5 75.9 acpi-cpufreq ondemand 40.5 68.5 74.6 amd-pstate performance 41.9 66.8 74.0 acpi-cpufreq schedutil 40.9 66.5 75.5 amd-pstate schedutil 37.5 51.2 60.8 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 11 22 33 44 55 Min: 38.95 / Avg: 40.25 / Max: 40.9 Min: 38.29 / Avg: 42.58 / Max: 49.41 Min: 44.5 / Avg: 52.36 / Max: 56.36 Min: 52.8 / Avg: 53.35 / Max: 53.97 Min: 52.63 / Avg: 54.69 / Max: 60.54 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Min Avg Max amd-pstate performance 41.4 64.1 72.3 acpi-cpufreq performance 40.9 63.8 71.9 acpi-cpufreq ondemand 39.5 62.1 70.6 acpi-cpufreq schedutil 40.3 59.4 65.3 amd-pstate schedutil 39.8 43.5 50.6 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 36.2 94.4 149.9 acpi-cpufreq performance 24.1 91.2 148.6 acpi-cpufreq ondemand 26.1 90.8 146.8 acpi-cpufreq schedutil 26.4 84.3 127.4 amd-pstate schedutil 18.3 56.4 89.1 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 40 80 120 160 200
Min Avg Max amd-pstate schedutil 400 3845 5499 acpi-cpufreq ondemand 3000 4678 5500 acpi-cpufreq schedutil 3000 4901 5551 acpi-cpufreq performance 3758 5039 5499 amd-pstate performance 3373 5606 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance amd-pstate schedutil 0.286 0.572 0.858 1.144 1.43 0.874 1.059 1.064 1.179 1.271
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 20 40 60 80 100 SE +/- 1.18, N = 15 SE +/- 1.57, N = 12 SE +/- 2.48, N = 15 SE +/- 2.03, N = 15 SE +/- 2.21, N = 15 71.63 73.66 96.21 100.50 107.53 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance amd-pstate schedutil 0.286 0.572 0.858 1.144 1.43 0.874 1.059 1.064 1.179 1.271 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 3845 5499 acpi-cpufreq ondemand 3000 4678 5500 acpi-cpufreq schedutil 3000 4901 5551 acpi-cpufreq performance 3758 5039 5499 amd-pstate performance 3373 5606 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.2 94.4 149.9 acpi-cpufreq performance 24.1 91.2 148.6 acpi-cpufreq ondemand 26.1 90.8 146.8 acpi-cpufreq schedutil 26.4 84.3 127.4 amd-pstate schedutil 18.3 56.4 89.1 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 963 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 894 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 951 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 1039 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 710 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 41.4 64.1 72.3 acpi-cpufreq performance 40.9 63.8 71.9 acpi-cpufreq ondemand 39.5 62.1 70.6 acpi-cpufreq schedutil 40.3 59.4 65.3 amd-pstate schedutil 39.8 43.5 50.6 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 20 40 60 80 100 Min: 67.62 / Avg: 71.63 / Max: 82.13 Min: 69.74 / Avg: 73.66 / Max: 88.69 Min: 86.2 / Avg: 96.21 / Max: 115.96 Min: 94.79 / Avg: 100.5 / Max: 117.05 Min: 94.93 / Avg: 107.53 / Max: 123.57 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Min Avg Max amd-pstate performance 40.3 61.1 67.1 acpi-cpufreq performance 38.9 60.4 67.0 amd-pstate schedutil 39.0 58.6 64.3 acpi-cpufreq schedutil 39.4 58.0 63.6 acpi-cpufreq ondemand 38.0 58.0 63.6 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 37.5 98.6 126.7 acpi-cpufreq performance 14.6 91.8 126.9 acpi-cpufreq ondemand 26.0 90.0 117.8 amd-pstate schedutil 27.0 89.3 114.0 acpi-cpufreq schedutil 26.2 85.7 114.9 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 40 80 120 160 200
Min Avg Max amd-pstate schedutil 400 4663 5496 acpi-cpufreq ondemand 3000 4884 5674 acpi-cpufreq schedutil 3000 4939 5506 acpi-cpufreq performance 560 5067 5537 amd-pstate performance 5433 5617 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 0.1595 0.319 0.4785 0.638 0.7975 0.600 0.626 0.626 0.683 0.709
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 15 30 45 60 75 SE +/- 0.33, N = 4 SE +/- 0.13, N = 4 SE +/- 0.18, N = 5 SE +/- 1.79, N = 15 SE +/- 0.33, N = 5 53.59 53.63 56.39 65.14 67.31 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 0.1595 0.319 0.4785 0.638 0.7975 0.600 0.626 0.626 0.683 0.709 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4663 5496 acpi-cpufreq ondemand 3000 4884 5674 acpi-cpufreq schedutil 3000 4939 5506 acpi-cpufreq performance 560 5067 5537 amd-pstate performance 5433 5617 5881 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 37.5 98.6 126.7 acpi-cpufreq performance 14.6 91.8 126.9 acpi-cpufreq ondemand 26.0 90.0 117.8 amd-pstate schedutil 27.0 89.3 114.0 acpi-cpufreq schedutil 26.2 85.7 114.9 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.5 CPU Power Consumption Monitor 40 80 120 160 200 1. amd-pstate performance: Approximate power consumption of 1341 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 1279 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 1369 Joules per run. 4. amd-pstate schedutil: Approximate power consumption of 1407 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 1350 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 40.3 61.1 67.1 acpi-cpufreq performance 38.9 60.4 67.0 amd-pstate schedutil 39.0 58.6 64.3 acpi-cpufreq schedutil 39.4 58.0 63.6 acpi-cpufreq ondemand 38.0 58.0 63.6 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.5 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 13 26 39 52 65 Min: 52.63 / Avg: 53.59 / Max: 54.09 Min: 53.34 / Avg: 53.63 / Max: 53.9 Min: 55.86 / Avg: 56.39 / Max: 56.92 Min: 40.14 / Avg: 65.14 / Max: 67.28 Min: 66.42 / Avg: 67.31 / Max: 68.32 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
SVT-AV1 Min Avg Max acpi-cpufreq schedutil 42.4 69.3 79.3 acpi-cpufreq performance 41.3 68.8 78.8 amd-pstate performance 42.5 68.8 79.9 acpi-cpufreq ondemand 40.6 68.5 78.1 amd-pstate schedutil 41.8 68.1 78.1 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.2 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 19.6 98.3 177.4 acpi-cpufreq performance 26.8 97.3 180.6 acpi-cpufreq ondemand 26.2 96.0 178.3 acpi-cpufreq schedutil 26.4 95.6 179.1 amd-pstate schedutil 27.7 95.0 174.9 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.2 CPU Power Consumption Monitor 50 100 150 200 250
Min Avg Max amd-pstate schedutil 400 4023 5750 acpi-cpufreq ondemand 3000 4426 5725 acpi-cpufreq schedutil 3000 4902 5575 acpi-cpufreq performance 4500 4971 5725 amd-pstate performance 983 5554 5881 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.2 Encoder Mode: Preset 10 - Input: Bosphorus 4K amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.3411 0.6822 1.0233 1.3644 1.7055 1.380 1.470 1.490 1.505 1.516
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.2 Encoder Mode: Preset 10 - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 30 60 90 120 150 SE +/- 0.87, N = 15 SE +/- 0.39, N = 8 SE +/- 3.46, N = 15 SE +/- 0.28, N = 8 SE +/- 0.34, N = 8 131.08 142.38 144.46 144.54 147.48 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 1.2 Encoder Mode: Preset 10 - Input: Bosphorus 4K amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance 0.3411 0.6822 1.0233 1.3644 1.7055 1.380 1.470 1.490 1.505 1.516 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 4023 5750 acpi-cpufreq ondemand 3000 4426 5725 acpi-cpufreq schedutil 3000 4902 5575 acpi-cpufreq performance 4500 4971 5725 amd-pstate performance 983 5554 5881 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 19.6 98.3 177.4 acpi-cpufreq performance 26.8 97.3 180.6 acpi-cpufreq ondemand 26.2 96.0 178.3 acpi-cpufreq schedutil 26.4 95.6 179.1 amd-pstate schedutil 27.7 95.0 174.9 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.2 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 858 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 839 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 816 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 836 Joules per run. 5. amd-pstate schedutil: Approximate power consumption of 868 Joules per run. CPU Temp
Min Avg Max acpi-cpufreq schedutil 42.4 69.3 79.3 acpi-cpufreq performance 41.3 68.8 78.8 amd-pstate performance 42.5 68.8 79.9 acpi-cpufreq ondemand 40.6 68.5 78.1 amd-pstate schedutil 41.8 68.1 78.1 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.2 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.2 Encoder Mode: Preset 10 - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 30 60 90 120 150 Min: 121.29 / Avg: 131.08 / Max: 136.03 Min: 140.88 / Avg: 142.38 / Max: 144.32 Min: 96.12 / Avg: 144.46 / Max: 149.7 Min: 143.09 / Avg: 144.53 / Max: 145.31 Min: 145.8 / Avg: 147.48 / Max: 148.98 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Min Avg Max amd-pstate performance 44.6 66.5 75.3 acpi-cpufreq schedutil 44.4 66.2 75.4 acpi-cpufreq performance 43.5 65.4 74.3 acpi-cpufreq ondemand 43.0 65.2 74.0 amd-pstate schedutil 44.1 65.0 74.4 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.2 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 36.4 90.0 173.6 amd-pstate schedutil 27.1 86.6 170.3 acpi-cpufreq ondemand 25.9 84.1 172.1 acpi-cpufreq performance 17.2 83.6 173.8 acpi-cpufreq schedutil 26.6 83.3 173.6 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.2 CPU Power Consumption Monitor 50 100 150 200 250
Min Avg Max amd-pstate schedutil 400 3546 5668 acpi-cpufreq ondemand 3000 4216 5741 acpi-cpufreq schedutil 4500 4902 5550 acpi-cpufreq performance 4500 4953 5750 amd-pstate performance 5285 5706 5881 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.2 Encoder Mode: Preset 12 - Input: Bosphorus 4K amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance 0.5429 1.0858 1.6287 2.1716 2.7145 2.125 2.243 2.342 2.351 2.413
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.2 Encoder Mode: Preset 12 - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 40 80 120 160 200 SE +/- 4.95, N = 15 SE +/- 5.35, N = 15 SE +/- 5.38, N = 15 SE +/- 5.56, N = 15 SE +/- 5.46, N = 15 184.06 195.85 197.07 201.85 201.89 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 1.2 Encoder Mode: Preset 12 - Input: Bosphorus 4K amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance 0.5429 1.0858 1.6287 2.1716 2.7145 2.125 2.243 2.342 2.351 2.413 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max amd-pstate schedutil 400 3546 5668 acpi-cpufreq ondemand 3000 4216 5741 acpi-cpufreq schedutil 4500 4902 5550 acpi-cpufreq performance 4500 4953 5750 amd-pstate performance 5285 5706 5881 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.4 90.0 173.6 amd-pstate schedutil 27.1 86.6 170.3 acpi-cpufreq ondemand 25.9 84.1 172.1 acpi-cpufreq performance 17.2 83.6 173.8 acpi-cpufreq schedutil 26.6 83.3 173.6 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.2 CPU Power Consumption Monitor 50 100 150 200 250 1. amd-pstate performance: Approximate power consumption of 672 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 670 Joules per run. 3. acpi-cpufreq ondemand: Approximate power consumption of 634 Joules per run. 4. acpi-cpufreq performance: Approximate power consumption of 625 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 628 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 44.6 66.5 75.3 acpi-cpufreq schedutil 44.4 66.2 75.4 acpi-cpufreq performance 43.5 65.4 74.3 acpi-cpufreq ondemand 43.0 65.2 74.0 amd-pstate schedutil 44.1 65.0 74.4 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.2 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.2 Encoder Mode: Preset 12 - Input: Bosphorus 4K amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 40 80 120 160 200 Min: 115.35 / Avg: 184.06 / Max: 191.92 Min: 121.1 / Avg: 195.85 / Max: 203.23 Min: 122.52 / Avg: 197.07 / Max: 206.82 Min: 124.1 / Avg: 201.85 / Max: 209.61 Min: 125.57 / Avg: 201.89 / Max: 209.12 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
ParaView OpenBenchmarking.org MiVoxels / Sec Per Watt, More Is Better ParaView 5.10.1 Test: Wavelet Volume - Resolution: 3840 x 2160 amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance 40 80 120 160 200 148.39 148.49 156.62 156.97 160.61
Result
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Volume - Resolution: 3840 x 2160 acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 110 220 330 440 550 SE +/- 2.05, N = 7 SE +/- 8.13, N = 15 SE +/- 4.68, N = 15 SE +/- 0.51, N = 7 SE +/- 1.63, N = 7 443.60 455.64 461.48 479.05 485.42 MiVoxels / Sec
OpenBenchmarking.org MiVoxels / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Volume - Resolution: 3840 x 2160 acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 1700 3400 5100 6800 8500 SE +/- 32.83, N = 7 SE +/- 130.13, N = 15 SE +/- 74.89, N = 15 SE +/- 8.09, N = 7 SE +/- 26.09, N = 7 7097.51 7290.24 7383.70 7664.85 7766.64 Result Confidence
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Volume - Resolution: 3840 x 2160 acpi-cpufreq ondemand amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 90 180 270 360 450 Min: 434.72 / Avg: 443.6 / Max: 448.89 Min: 409.96 / Avg: 455.64 / Max: 487.09 Min: 417.57 / Avg: 461.48 / Max: 480.27 Min: 476.8 / Avg: 479.05 / Max: 480.62 Min: 476.07 / Avg: 485.42 / Max: 488.72
OpenBenchmarking.org MiVoxels / Sec Per Watt, More Is Better ParaView 5.10.1 Test: Wavelet Volume - Resolution: 1920 x 1080 amd-pstate schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance 60 120 180 240 300 244.10 272.32 282.71 291.75 296.58
Result
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Volume - Resolution: 1920 x 1080 amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 200 400 600 800 1000 SE +/- 32.88, N = 15 SE +/- 7.05, N = 15 SE +/- 5.97, N = 15 SE +/- 7.55, N = 15 SE +/- 7.44, N = 7 746.15 804.86 848.16 871.11 886.55 MiVoxels / Sec
OpenBenchmarking.org MiVoxels / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Volume - Resolution: 1920 x 1080 amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 3K 6K 9K 12K 15K SE +/- 526.16, N = 15 SE +/- 112.73, N = 15 SE +/- 95.56, N = 15 SE +/- 120.83, N = 15 SE +/- 118.99, N = 7 11938.40 12877.75 13570.48 13937.78 14184.90 Result Confidence
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.10.1 Test: Wavelet Volume - Resolution: 1920 x 1080 amd-pstate schedutil acpi-cpufreq ondemand acpi-cpufreq schedutil acpi-cpufreq performance amd-pstate performance 160 320 480 640 800 Min: 550.4 / Avg: 746.15 / Max: 887.88 Min: 724.91 / Avg: 804.86 / Max: 828.11 Min: 804.6 / Avg: 848.16 / Max: 877 Min: 804.17 / Avg: 871.11 / Max: 902.39 Min: 847.2 / Avg: 886.55 / Max: 911.01
OpenBenchmarking.org MiPolys / Sec Per Watt, More Is Better ParaView 5.10.1 Test: Wavelet Contour - Resolution: 3840 x 2160 amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 20 40 60 80 100 71.96 76.69 78.74 80.97 82.21
OpenBenchmarking.org MiPolys / Sec Per Watt, More Is Better ParaView 5.10.1 Test: Wavelet Contour - Resolution: 1920 x 1080 amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 30 60 90 120 150 100.10 105.37 109.78 111.42 112.29
OpenBenchmarking.org MiPolys / Sec Per Watt, More Is Better ParaView 5.10.1 Test: Many Spheres - Resolution: 3840 x 2160 amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 60 120 180 240 300 229.38 236.74 250.37 252.83 259.92
OpenBenchmarking.org MiPolys / Sec Per Watt, More Is Better ParaView 5.10.1 Test: Many Spheres - Resolution: 1920 x 1080 amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 70 140 210 280 350 275.46 284.64 299.96 305.67 318.93
GravityMark Min Avg Max amd-pstate performance 37.8 39.9 66.0 acpi-cpufreq schedutil 37.8 39.8 66.8 amd-pstate schedutil 37.4 39.5 65.1 acpi-cpufreq performance 36.5 38.5 65.6 acpi-cpufreq ondemand 35.5 37.5 66.1 OpenBenchmarking.org Celsius, Fewer Is Better GravityMark 1.72 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 26.6 52.1 69.9 amd-pstate schedutil 23.3 47.9 67.3 acpi-cpufreq performance 13.8 42.1 68.3 acpi-cpufreq schedutil 19.2 36.7 65.5 acpi-cpufreq ondemand 16.8 33.8 67.2 OpenBenchmarking.org Watts, Fewer Is Better GravityMark 1.72 CPU Power Consumption Monitor 20 40 60 80 100
Min Avg Max acpi-cpufreq ondemand 3000 3035 5719 acpi-cpufreq schedutil 3000 4479 5700 acpi-cpufreq performance 4500 4708 5751 amd-pstate schedutil 400 4839 5806 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better GravityMark 1.72 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better GravityMark 1.72 Resolution: 3840 x 2160 - Renderer: Vulkan amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 0.9644 1.9288 2.8932 3.8576 4.822 2.787 3.031 3.512 3.957 4.286
Result
OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.72 Resolution: 3840 x 2160 - Renderer: Vulkan acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 2.62, N = 12 145.0 145.1 145.1 145.2 147.7 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better GravityMark 1.72 Resolution: 3840 x 2160 - Renderer: Vulkan amd-pstate performance amd-pstate schedutil acpi-cpufreq performance acpi-cpufreq schedutil acpi-cpufreq ondemand 0.9644 1.9288 2.8932 3.8576 4.822 2.787 3.031 3.512 3.957 4.286 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 3035 5719 acpi-cpufreq schedutil 3000 4479 5700 acpi-cpufreq performance 4500 4708 5751 amd-pstate schedutil 400 4839 5806 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better GravityMark 1.72 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 26.6 52.1 69.9 amd-pstate schedutil 23.3 47.9 67.3 acpi-cpufreq performance 13.8 42.1 68.3 acpi-cpufreq schedutil 19.2 36.7 65.5 acpi-cpufreq ondemand 16.8 33.8 67.2 OpenBenchmarking.org Watts, Fewer Is Better GravityMark 1.72 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 8939 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 8250 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 7228 Joules per run. 4. acpi-cpufreq schedutil: Approximate power consumption of 6311 Joules per run. 5. acpi-cpufreq ondemand: Approximate power consumption of 5819 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 37.8 39.9 66.0 acpi-cpufreq schedutil 37.8 39.8 66.8 amd-pstate schedutil 37.4 39.5 65.1 acpi-cpufreq performance 36.5 38.5 65.6 acpi-cpufreq ondemand 35.5 37.5 66.1 OpenBenchmarking.org Celsius, Fewer Is Better GravityMark 1.72 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.72 Resolution: 3840 x 2160 - Renderer: Vulkan acpi-cpufreq ondemand amd-pstate schedutil amd-pstate performance acpi-cpufreq schedutil acpi-cpufreq performance 30 60 90 120 150 Min: 145 / Avg: 145 / Max: 145 Min: 145.1 / Avg: 145.13 / Max: 145.2 Min: 145.1 / Avg: 145.1 / Max: 145.1 Min: 145.1 / Avg: 145.17 / Max: 145.2 Min: 145 / Avg: 147.72 / Max: 176.5
Unvanquished Min Avg Max amd-pstate performance 38.6 55.0 65.6 acpi-cpufreq performance 37.8 53.1 64.6 acpi-cpufreq ondemand 37.0 51.8 63.4 acpi-cpufreq schedutil 39.6 51.4 65.6 amd-pstate schedutil 37.3 49.8 72.0 OpenBenchmarking.org Celsius, Fewer Is Better Unvanquished 0.53 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 32.7 62.7 73.3 acpi-cpufreq performance 26.9 60.0 74.2 amd-pstate schedutil 27.1 58.1 103.4 acpi-cpufreq ondemand 26.1 57.2 71.6 acpi-cpufreq schedutil 26.9 54.7 73.3 OpenBenchmarking.org Watts, Fewer Is Better Unvanquished 0.53 CPU Power Consumption Monitor 20 40 60 80 100
Min Avg Max acpi-cpufreq ondemand 3000 4927 5712 amd-pstate schedutil 400 5033 5757 acpi-cpufreq schedutil 3000 5084 5724 acpi-cpufreq performance 4500 5318 5727 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Unvanquished 0.53 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Unvanquished 0.53 Resolution: 3840 x 2160 - Effects Quality: Ultra amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 3 6 9 12 15 7.574 8.145 10.951 11.045 11.444
Result
OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.53 Resolution: 3840 x 2160 - Effects Quality: Ultra amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 150 300 450 600 750 SE +/- 16.07, N = 15 SE +/- 12.58, N = 15 SE +/- 2.00, N = 4 SE +/- 1.82, N = 4 SE +/- 0.99, N = 4 440.3 445.7 632.1 686.8 687.1 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Unvanquished 0.53 Resolution: 3840 x 2160 - Effects Quality: Ultra amd-pstate schedutil acpi-cpufreq schedutil amd-pstate performance acpi-cpufreq ondemand acpi-cpufreq performance 3 6 9 12 15 7.574 8.145 10.951 11.045 11.444 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 4927 5712 amd-pstate schedutil 400 5033 5757 acpi-cpufreq schedutil 3000 5084 5724 acpi-cpufreq performance 4500 5318 5727 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Unvanquished 0.53 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 32.7 62.7 73.3 acpi-cpufreq performance 26.9 60.0 74.2 amd-pstate schedutil 27.1 58.1 103.4 acpi-cpufreq ondemand 26.1 57.2 71.6 acpi-cpufreq schedutil 26.9 54.7 73.3 OpenBenchmarking.org Watts, Fewer Is Better Unvanquished 0.53 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1208 Joules per run. 2. acpi-cpufreq performance: Approximate power consumption of 1125 Joules per run. 3. amd-pstate schedutil: Approximate power consumption of 1469 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 1145 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 1328 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 38.6 55.0 65.6 acpi-cpufreq performance 37.8 53.1 64.6 acpi-cpufreq ondemand 37.0 51.8 63.4 acpi-cpufreq schedutil 39.6 51.4 65.6 amd-pstate schedutil 37.3 49.8 72.0 OpenBenchmarking.org Celsius, Fewer Is Better Unvanquished 0.53 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.53 Resolution: 3840 x 2160 - Effects Quality: Ultra amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand acpi-cpufreq performance amd-pstate performance 120 240 360 480 600 Min: 384.7 / Avg: 440.29 / Max: 631.8 Min: 393 / Avg: 445.73 / Max: 548.5 Min: 627.8 / Avg: 632.05 / Max: 635.8 Min: 681.7 / Avg: 686.78 / Max: 689.5 Min: 684.5 / Avg: 687.08 / Max: 689
Min Avg Max amd-pstate performance 39.3 54.7 65.9 acpi-cpufreq performance 38.3 53.1 64.4 acpi-cpufreq schedutil 39.9 52.1 67.4 acpi-cpufreq ondemand 37.8 51.8 64.3 amd-pstate schedutil 38.5 51.8 72.0 OpenBenchmarking.org Celsius, Fewer Is Better Unvanquished 0.53 CPU Temperature Monitor 20 40 60 80 100
Min Avg Max amd-pstate performance 36.3 63.1 73.2 amd-pstate schedutil 26.9 60.6 103.9 acpi-cpufreq performance 27.5 59.1 75.1 acpi-cpufreq ondemand 25.9 57.4 68.7 acpi-cpufreq schedutil 15.5 55.3 72.2 OpenBenchmarking.org Watts, Fewer Is Better Unvanquished 0.53 CPU Power Consumption Monitor 20 40 60 80 100
Min Avg Max acpi-cpufreq ondemand 3000 5006 5748 acpi-cpufreq schedutil 3000 5077 5750 amd-pstate schedutil 400 5135 5746 acpi-cpufreq performance 4500 5297 5750 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Unvanquished 0.53 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Unvanquished 0.53 Resolution: 1920 x 1080 - Effects Quality: Ultra amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 3 6 9 12 15 7.328 9.026 10.921 10.953 11.717
Result
OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.53 Resolution: 1920 x 1080 - Effects Quality: Ultra amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 150 300 450 600 750 SE +/- 1.38, N = 3 SE +/- 22.14, N = 15 SE +/- 2.53, N = 4 SE +/- 3.95, N = 4 SE +/- 0.98, N = 4 444.1 499.2 626.7 690.8 693.0 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Unvanquished 0.53 Resolution: 1920 x 1080 - Effects Quality: Ultra amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 3 6 9 12 15 7.328 9.026 10.921 10.953 11.717 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max acpi-cpufreq ondemand 3000 5006 5748 acpi-cpufreq schedutil 3000 5077 5750 amd-pstate schedutil 400 5135 5746 acpi-cpufreq performance 4500 5297 5750 amd-pstate performance 5881 5881 5881 OpenBenchmarking.org Megahertz, More Is Better Unvanquished 0.53 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1600 3200 4800 6400 8000 CPU Power Consumption
Min Avg Max amd-pstate performance 36.3 63.1 73.2 amd-pstate schedutil 26.9 60.6 103.9 acpi-cpufreq performance 27.5 59.1 75.1 acpi-cpufreq ondemand 25.9 57.4 68.7 acpi-cpufreq schedutil 15.5 55.3 72.2 OpenBenchmarking.org Watts, Fewer Is Better Unvanquished 0.53 CPU Power Consumption Monitor 20 40 60 80 100 1. amd-pstate performance: Approximate power consumption of 1230 Joules per run. 2. amd-pstate schedutil: Approximate power consumption of 1596 Joules per run. 3. acpi-cpufreq performance: Approximate power consumption of 1124 Joules per run. 4. acpi-cpufreq ondemand: Approximate power consumption of 1176 Joules per run. 5. acpi-cpufreq schedutil: Approximate power consumption of 1261 Joules per run. CPU Temp
Min Avg Max amd-pstate performance 39.3 54.7 65.9 acpi-cpufreq performance 38.3 53.1 64.4 acpi-cpufreq schedutil 39.9 52.1 67.4 acpi-cpufreq ondemand 37.8 51.8 64.3 amd-pstate schedutil 38.5 51.8 72.0 OpenBenchmarking.org Celsius, Fewer Is Better Unvanquished 0.53 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.53 Resolution: 1920 x 1080 - Effects Quality: Ultra amd-pstate schedutil acpi-cpufreq schedutil acpi-cpufreq ondemand amd-pstate performance acpi-cpufreq performance 120 240 360 480 600 Min: 442.2 / Avg: 444.13 / Max: 446.8 Min: 414.6 / Avg: 499.18 / Max: 673.6 Min: 621.7 / Avg: 626.65 / Max: 632 Min: 680 / Avg: 690.8 / Max: 697.1 Min: 691.6 / Avg: 693 / Max: 695.8
amd-pstate schedutil Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 XT, Audio: AMD Navi 21 HDMI Audio, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411
OS: Ubuntu 22.04, Kernel: 6.0.0-060000-generic (x86_64), Desktop: GNOME Shell 42.4, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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 schedutil (Boost: Enabled) - CPU Microcode: 0xa601203Graphics Notes: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101Python Notes: Python 3.10.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 6 October 2022 20:40 by user pts.
amd-pstate performance Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 XT (2575/1000MHz), Audio: AMD Navi 21 HDMI Audio, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411
OS: Ubuntu 22.04, Kernel: 6.0.0-060000-generic (x86_64), Desktop: GNOME Shell 42.4, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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 performance (Boost: Enabled) - CPU Microcode: 0xa601203Graphics Notes: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101Python Notes: Python 3.10.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 7 October 2022 05:37 by user pts.
acpi-cpufreq schedutil Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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: 0xa601203Graphics Notes: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101Python Notes: Python 3.10.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 7 October 2022 12:59 by user pts.
acpi-cpufreq ondemand Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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 ondemand (Boost: Enabled) - CPU Microcode: 0xa601203Graphics Notes: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101Python Notes: Python 3.10.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 8 October 2022 05:16 by user pts.
acpi-cpufreq performance Processor: AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 XT (2575/1000MHz), Audio: AMD Navi 21 HDMI Audio, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411
OS: Ubuntu 22.04, Kernel: 6.0.0-060000-generic (x86_64), Desktop: GNOME Shell 42.4, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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: 0xa601203Graphics Notes: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101Python Notes: Python 3.10.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 8 October 2022 08:52 by user pts.