AMD Ryzen 7 PRO 7840U testing with a LENOVO 21K5001JUS (R2FET33W_0921_3 1.13 BIOS) and AMD Phoenix1 4GB on Ubuntu 23.10 via the Phoronix Test Suite.
Original Processor: AMD Ryzen 7 PRO 7840U @ 3.30GHz (8 Cores / 16 Threads), Motherboard: LENOVO 21K5001JUS (R2FET31W 1.11 BIOS), Chipset: AMD Device 14e8, Memory: 60GB, Disk: 1024GB Kioxia KXG8AZNV1T02 LA, Graphics: AMD Phoenix1 4GB (2700/400MHz), Audio: AMD Rembrandt Radeon HD Audio, Network: Realtek RTL8111/8168/8411 + Qualcomm QCNFA765
OS: Ubuntu 23.10, Kernel: 6.3.0-7-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 23.1.7-1ubuntu1 (LLVM 15.0.7 DRM 3.52), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 2880x1800
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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) - Platform Profile: low-power - CPU Microcode: 0xa704103 - ACPI Profile: low-powerPython Notes: Python 3.11.5Security 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 Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
low_power balanced 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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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) - Platform Profile: balanced - CPU Microcode: 0xa704103 - ACPI Profile: balancedPython Notes: Python 3.11.5Security 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 Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
performance Processor: AMD Ryzen 7 PRO 7840U @ 3.30GHz (8 Cores / 16 Threads), Motherboard: LENOVO 21K5001JUS (R2FET33W_0921_3 1.13 BIOS)Chipset: AMD Device 14e8, Memory: 60GB, Disk: 1024GB Kioxia KXG8AZNV1T02 LA, Graphics: AMD Phoenix1 4GB (2700/400MHz), Audio: AMD Rembrandt Radeon HD Audio, Network: Realtek RTL8111/8168/8411 + Qualcomm QCNFA765
OS: Ubuntu 23.10, Kernel: 6.3.0-7-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.7 + Wayland, OpenGL: 4.6 Mesa 23.1.7-1ubuntu1 (LLVM 15.0.7 DRM 3.52), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 2880x1800
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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) - Platform Profile: performance - CPU Microcode: 0xa704103 - ACPI Profile: performancePython Notes: Python 3.11.5Security 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 Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Timed Godot Game Engine Compilation This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.
OpenVKL OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Items / Sec, More Is Better OpenVKL 1.3.1 Benchmark: vklBenchmark ISPC Original low_power balanced performance 30 60 90 120 150 SE +/- 1.33, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 SE +/- 0.67, N = 3 127 132 139 139 MIN: 16 / MAX: 1788 MIN: 17 / MAX: 1804 MIN: 19 / MAX: 1839 MIN: 19 / MAX: 1865 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better OpenVKL 1.3.1 CPU Temperature Monitor Original low_power balanced performance 20 40 60 80 100 Min: 50.75 / Avg: 69.95 / Max: 85.63 Min: 48.75 / Avg: 70.98 / Max: 85.63 Min: 52.63 / Avg: 81.2 / Max: 89.38 Min: 53.25 / Avg: 82.41 / Max: 90.63 Items / Sec Per Watt
OpenBenchmarking.org Items / Sec Per Watt, More Is Better OpenVKL 1.3.1 Benchmark: vklBenchmark ISPC low_power balanced performance 1.1729 2.3458 3.5187 4.6916 5.8645 5.213 4.475 4.419 CPU Peak Freq (Highest CPU Core Frequency
OpenBenchmarking.org Megahertz, More Is Better OpenVKL 1.3.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor low_power balanced performance 900 1800 2700 3600 4500 Min: 1600 / Avg: 3926.28 / Max: 5116 Min: 1600 / Avg: 4189.78 / Max: 5115 Min: 1600 / Avg: 4194.45 / Max: 5027 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVKL 1.3.1 CPU Power Consumption Monitor low_power balanced performance 8 16 24 32 40 Min: 2.41 / Avg: 25.32 / Max: 30.26 Min: 2.61 / Avg: 31.06 / Max: 36.33 Min: 2.51 / Avg: 31.45 / Max: 40.32 1. low_power: Approximate power consumption of 10610 Joules per run. 2. balanced: Approximate power consumption of 12768 Joules per run. 3. performance: Approximate power consumption of 12959 Joules per run. Result Confidence
OpenBenchmarking.org Items / Sec, More Is Better OpenVKL 1.3.1 Benchmark: vklBenchmark ISPC Original low_power balanced performance 30 60 90 120 150 Min: 126 / Avg: 127.33 / Max: 130 Min: 131 / Avg: 132.33 / Max: 133 Min: 138 / Avg: 138.67 / Max: 139 Min: 138 / Avg: 139.33 / Max: 140
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 Original low_power balanced performance 5 10 15 20 25 SE +/- 0.11, N = 3 SE +/- 0.18, N = 3 SE +/- 0.18, N = 3 SE +/- 0.02, N = 3 19.75 20.28 22.28 22.64 CPU Temp
Min Avg Max Original 48.4 63.5 79.6 low_power 49.6 63.0 78.8 balanced 54.6 76.3 84.8 performance 53.9 78.8 85.8 OpenBenchmarking.org Celsius, Fewer Is Better TensorFlow 2.12 CPU Temperature Monitor 20 40 60 80 100 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 low_power balanced performance 0.2027 0.4054 0.6081 0.8108 1.0135 0.901 0.733 0.707 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3165 5041 balanced 1600 3654 4966 performance 1600 3744 5115 OpenBenchmarking.org Megahertz, More Is Better TensorFlow 2.12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.5 22.5 27.2 balanced 2.7 30.4 36.4 performance 2.6 32.0 40.5 OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.12 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 7957 Joules per run. 2. balanced: Approximate power consumption of 9794 Joules per run. 3. performance: Approximate power consumption of 10157 Joules per run. Result Confidence
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 Original low_power balanced performance 5 10 15 20 25 Min: 19.63 / Avg: 19.75 / Max: 19.96 Min: 20.1 / Avg: 20.28 / Max: 20.64 Min: 21.98 / Avg: 22.28 / Max: 22.61 Min: 22.62 / Avg: 22.64 / Max: 22.67
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 Original low_power balanced performance 5 10 15 20 25 SE +/- 0.15, N = 12 SE +/- 0.09, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 19.76 21.03 22.26 22.32 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better TensorFlow 2.12 CPU Temperature Monitor Original low_power balanced performance 16 32 48 64 80 Min: 51.25 / Avg: 64.74 / Max: 80 Min: 49.13 / Avg: 67.22 / Max: 80.13 Min: 53.13 / Avg: 77.4 / Max: 85.13 Min: 53 / Avg: 77.81 / Max: 84.13 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 low_power balanced performance 0.1868 0.3736 0.5604 0.7472 0.934 0.830 0.701 0.699 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3454 5041 balanced 1600 3791 5015 performance 1600 3805 5016 OpenBenchmarking.org Megahertz, More Is Better TensorFlow 2.12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.4 30.1 balanced 2.6 31.8 36.4 performance 2.6 31.9 40.4 OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.12 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 4411 Joules per run. 2. balanced: Approximate power consumption of 5250 Joules per run. 3. performance: Approximate power consumption of 5265 Joules per run. Result Confidence
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 Original low_power balanced performance 5 10 15 20 25 Min: 19.4 / Avg: 19.76 / Max: 21.11 Min: 20.86 / Avg: 21.03 / Max: 21.12 Min: 22.25 / Avg: 22.26 / Max: 22.28 Min: 22.3 / Avg: 22.32 / Max: 22.36
SecureMark SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions developed by EEMBC. SecureMark-TLS is benchmarking Transport Layer Security performance with a focus on IoT/edge computing. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org marks, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS Original low_power balanced performance 80K 160K 240K 320K 400K SE +/- 1045.41, N = 3 SE +/- 730.02, N = 3 SE +/- 783.71, N = 3 SE +/- 420.09, N = 3 393682 393372 391760 396194 1. (CC) gcc options: -pedantic -O3 CPU Temp
Min Avg Max Original 44.5 85.5 86.3 low_power 43.4 85.1 86.3 balanced 45.5 86.9 89.8 performance 44.8 86.6 89.6 OpenBenchmarking.org Celsius, Fewer Is Better SecureMark 1.0.4 CPU Temperature Monitor 20 40 60 80 100 marks Per Watt
OpenBenchmarking.org marks Per Watt, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS low_power balanced performance 5K 10K 15K 20K 25K 24780.64 25155.02 24422.35 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4900 4991 balanced 1600 4935 5095 performance 1600 4907 5004 OpenBenchmarking.org Megahertz, More Is Better SecureMark 1.0.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 15.9 24.4 balanced 2.5 15.6 17.7 performance 2.3 16.2 17.4 OpenBenchmarking.org Watts, Fewer Is Better SecureMark 1.0.4 CPU Power Consumption Monitor 8 16 24 32 40 1. low_power: Approximate power consumption of 4519 Joules per run. 2. balanced: Approximate power consumption of 4485 Joules per run. 3. performance: Approximate power consumption of 4618 Joules per run. Result Confidence
OpenBenchmarking.org marks, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS Original low_power balanced performance 70K 140K 210K 280K 350K Min: 392452.66 / Avg: 393681.75 / Max: 395761.09 Min: 391989.09 / Avg: 393372.49 / Max: 394468.47 Min: 390676.09 / Avg: 391760.05 / Max: 393282.53 Min: 395500.91 / Avg: 396194.2 / Max: 396951.88 1. (CC) gcc options: -pedantic -O3
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: LargeRandom Original low_power balanced performance 0.369 0.738 1.107 1.476 1.845 SE +/- 0.00, N = 3 SE +/- 0.02, N = 15 SE +/- 0.02, N = 15 SE +/- 0.02, N = 15 1.64 1.55 1.52 1.56 1. (CXX) g++ options: -O3 CPU Temp
Min Avg Max Original 46.4 81.8 83.6 low_power 46.0 82.2 84.4 balanced 46.9 85.8 88.4 performance 47.0 85.7 88.8 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 2.0 CPU Temperature Monitor 20 40 60 80 100 GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 2.0 Throughput Test: LargeRandom low_power balanced performance 0.0205 0.041 0.0615 0.082 0.1025 0.091 0.091 0.090 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4786 5041 balanced 1600 4760 5040 performance 1600 4771 5094 OpenBenchmarking.org Megahertz, More Is Better simdjson 2.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.3 17.1 19.5 balanced 2.4 16.8 18.5 performance 2.3 17.4 19.6 OpenBenchmarking.org Watts, Fewer Is Better simdjson 2.0 CPU Power Consumption Monitor 6 12 18 24 30 1. low_power: Approximate power consumption of 1236 Joules per run. 2. balanced: Approximate power consumption of 1147 Joules per run. 3. performance: Approximate power consumption of 1298 Joules per run. Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: LargeRandom Original low_power balanced performance 2 4 6 8 10 Min: 1.63 / Avg: 1.64 / Max: 1.64 Min: 1.45 / Avg: 1.55 / Max: 1.61 Min: 1.43 / Avg: 1.52 / Max: 1.61 Min: 1.46 / Avg: 1.56 / Max: 1.62 1. (CXX) g++ options: -O3
libxsmm Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 128 Original low_power balanced performance 60 120 180 240 300 SE +/- 2.38, N = 12 SE +/- 1.86, N = 3 SE +/- 0.10, N = 3 SE +/- 1.34, N = 3 249.1 253.7 287.6 298.3 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2 CPU Temp
OpenBenchmarking.org Celsius, Fewer Is Better libxsmm 2-1.17-3645 CPU Temperature Monitor Original low_power balanced performance 16 32 48 64 80 Min: 47.63 / Avg: 61.9 / Max: 64.13 Min: 47.25 / Avg: 59.81 / Max: 62.5 Min: 53.38 / Avg: 72.73 / Max: 75.38 Min: 55.5 / Avg: 78.76 / Max: 83.38 GFLOPS/s Per Watt
OpenBenchmarking.org GFLOPS/s Per Watt, More Is Better libxsmm 2-1.17-3645 M N K: 128 low_power balanced performance 3 6 9 12 15 12.190 10.153 9.356 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 2699 4971 balanced 1600 3277 3699 performance 1600 3489 3843 OpenBenchmarking.org Megahertz, More Is Better libxsmm 2-1.17-3645 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 20.8 28.5 balanced 2.7 28.3 35.8 performance 2.7 31.9 40.0 OpenBenchmarking.org Watts, Fewer Is Better libxsmm 2-1.17-3645 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 2976 Joules per run. 2. balanced: Approximate power consumption of 3607 Joules per run. 3. performance: Approximate power consumption of 3911 Joules per run. Result Confidence
OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 128 Original low_power balanced performance 50 100 150 200 250 Min: 228.2 / Avg: 249.09 / Max: 254.2 Min: 250.7 / Avg: 253.67 / Max: 257.1 Min: 287.5 / Avg: 287.6 / Max: 287.8 Min: 296.8 / Avg: 298.33 / Max: 301 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU Original low_power balanced performance 0.288 0.576 0.864 1.152 1.44 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 15 SE +/- 0.01, N = 15 1.12 1.12 1.28 1.27 MIN: 0.29 / MAX: 1.44 MIN: 0.29 / MAX: 1.44 MIN: 0.34 / MAX: 1.63 MIN: 0.33 / MAX: 1.65 CPU Temp
Min Avg Max Original 50.0 71.0 75.5 low_power 48.1 67.4 71.9 balanced 52.3 79.8 83.4 performance 53.3 81.2 88.0 OpenBenchmarking.org Celsius, Fewer Is Better LuxCoreRender 2.6 CPU Temperature Monitor 20 40 60 80 100 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU low_power balanced performance 0.0101 0.0202 0.0303 0.0404 0.0505 0.045 0.041 0.040 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3315 5011 balanced 1398 3615 5003 performance 1397 3632 5015 OpenBenchmarking.org Megahertz, More Is Better LuxCoreRender 2.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.2 30.3 balanced 2.8 31.4 36.3 performance 1.6 31.6 40.4 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1685 Joules per run. 2. balanced: Approximate power consumption of 2093 Joules per run. 3. performance: Approximate power consumption of 2108 Joules per run. Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU Original low_power balanced performance 2 4 6 8 10 Min: 1.1 / Avg: 1.12 / Max: 1.14 Min: 1.12 / Avg: 1.12 / Max: 1.13 Min: 1.23 / Avg: 1.28 / Max: 1.33 Min: 1.23 / Avg: 1.27 / Max: 1.37
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/scivis/real_time Original low_power balanced performance 0.6621 1.3242 1.9863 2.6484 3.3105 SE +/- 0.00233, N = 3 SE +/- 0.00234, N = 3 SE +/- 0.00207, N = 3 SE +/- 0.00408, N = 3 2.74661 2.72747 2.92116 2.94274 CPU Temp
Min Avg Max Original 49.9 73.4 78.8 low_power 47.9 71.0 77.6 balanced 52.1 82.0 88.4 performance 52.8 83.1 90.1 OpenBenchmarking.org Celsius, Fewer Is Better OSPRay 2.12 CPU Temperature Monitor 20 40 60 80 100 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 2.12 Benchmark: particle_volume/scivis/real_time low_power balanced performance 0.0239 0.0478 0.0717 0.0956 0.1195 0.106 0.091 0.092 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3770 4257 balanced 1397 3998 4975 performance 1600 4012 4668 OpenBenchmarking.org Megahertz, More Is Better OSPRay 2.12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.2 25.8 30.3 balanced 2.6 32.0 36.3 performance 2.6 32.0 40.4 OpenBenchmarking.org Watts, Fewer Is Better OSPRay 2.12 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 4555 Joules per run. 2. balanced: Approximate power consumption of 5570 Joules per run. 3. performance: Approximate power consumption of 5582 Joules per run. Result Confidence
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/scivis/real_time Original low_power balanced performance 2 4 6 8 10 Min: 2.74 / Avg: 2.75 / Max: 2.75 Min: 2.72 / Avg: 2.73 / Max: 2.73 Min: 2.92 / Avg: 2.92 / Max: 2.92 Min: 2.94 / Avg: 2.94 / Max: 2.95
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/pathtracer/real_time Original low_power balanced performance 30 60 90 120 150 SE +/- 0.13, N = 3 SE +/- 0.13, N = 3 SE +/- 0.07, N = 3 SE +/- 0.14, N = 3 112.53 112.98 118.42 118.08 CPU Temp
Min Avg Max Original 50.1 73.5 79.9 low_power 47.9 70.7 76.1 balanced 51.9 82.1 88.6 performance 52.0 83.3 90.6 OpenBenchmarking.org Celsius, Fewer Is Better OSPRay 2.12 CPU Temperature Monitor 20 40 60 80 100 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 2.12 Benchmark: particle_volume/pathtracer/real_time low_power balanced performance 0.9887 1.9774 2.9661 3.9548 4.9435 4.394 3.706 3.673 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1034 3625 4249 balanced 1600 3858 4428 performance 1600 3879 4434 OpenBenchmarking.org Megahertz, More Is Better OSPRay 2.12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.3 25.7 30.3 balanced 2.6 32.0 36.3 performance 2.5 32.2 40.4 OpenBenchmarking.org Watts, Fewer Is Better OSPRay 2.12 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 4243 Joules per run. 2. balanced: Approximate power consumption of 5155 Joules per run. 3. performance: Approximate power consumption of 5230 Joules per run. Result Confidence
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/pathtracer/real_time Original low_power balanced performance 20 40 60 80 100 Min: 112.38 / Avg: 112.53 / Max: 112.8 Min: 112.73 / Avg: 112.98 / Max: 113.18 Min: 118.29 / Avg: 118.42 / Max: 118.54 Min: 117.81 / Avg: 118.08 / Max: 118.26
VVenC VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 4K - Video Preset: Fast Original low_power balanced performance 0.8951 1.7902 2.6853 3.5804 4.4755 SE +/- 0.009, N = 3 SE +/- 0.010, N = 3 SE +/- 0.018, N = 3 SE +/- 0.010, N = 3 3.659 3.652 3.971 3.978 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects CPU Temp
Min Avg Max Original 48.9 75.6 80.5 low_power 47.6 72.9 79.4 balanced 50.6 83.8 90.0 performance 51.6 84.7 91.3 OpenBenchmarking.org Celsius, Fewer Is Better VVenC 1.9 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VVenC 1.9 Video Input: Bosphorus 4K - Video Preset: Fast low_power balanced performance 0.0322 0.0644 0.0966 0.1288 0.161 0.143 0.126 0.125 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3343 4968 balanced 1600 3705 4966 performance 1600 3684 5008 OpenBenchmarking.org Megahertz, More Is Better VVenC 1.9 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.6 30.2 balanced 2.6 31.6 36.4 performance 2.6 31.9 40.3 OpenBenchmarking.org Watts, Fewer Is Better VVenC 1.9 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 4297 Joules per run. 2. balanced: Approximate power consumption of 4912 Joules per run. 3. performance: Approximate power consumption of 4932 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 4K - Video Preset: Fast Original low_power balanced performance 2 4 6 8 10 Min: 3.64 / Avg: 3.66 / Max: 3.67 Min: 3.63 / Avg: 3.65 / Max: 3.67 Min: 3.94 / Avg: 3.97 / Max: 3.99 Min: 3.96 / Avg: 3.98 / Max: 3.99 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects
High Performance Conjugate Gradient HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 X Y Z: 104 104 104 - RT: 60 Original low_power balanced performance 2 4 6 8 10 SE +/- 0.00265, N = 3 SE +/- 0.00319, N = 3 SE +/- 0.00213, N = 3 SE +/- 0.00212, N = 3 6.48026 6.48526 6.47967 6.48339 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi CPU Temp
Min Avg Max Original 50.3 67.1 71.5 low_power 48.1 65.6 68.3 balanced 51.9 76.4 79.5 performance 51.6 77.2 83.1 OpenBenchmarking.org Celsius, Fewer Is Better High Performance Conjugate Gradient 3.1 CPU Temperature Monitor 20 40 60 80 100 GFLOP/s Per Watt
OpenBenchmarking.org GFLOP/s Per Watt, More Is Better High Performance Conjugate Gradient 3.1 X Y Z: 104 104 104 - RT: 60 low_power balanced performance 0.0594 0.1188 0.1782 0.2376 0.297 0.264 0.209 0.203 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4062 4509 balanced 1600 4350 4738 performance 1600 4380 4894 OpenBenchmarking.org Megahertz, More Is Better High Performance Conjugate Gradient 3.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.4 24.6 30.4 balanced 2.6 30.9 36.6 performance 2.5 32.0 40.6 OpenBenchmarking.org Watts, Fewer Is Better High Performance Conjugate Gradient 3.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 3588 Joules per run. 2. balanced: Approximate power consumption of 4518 Joules per run. 3. performance: Approximate power consumption of 4677 Joules per run. Result Confidence
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 X Y Z: 104 104 104 - RT: 60 Original low_power balanced performance 3 6 9 12 15 Min: 6.48 / Avg: 6.48 / Max: 6.49 Min: 6.48 / Avg: 6.49 / Max: 6.49 Min: 6.48 / Avg: 6.48 / Max: 6.48 Min: 6.48 / Avg: 6.48 / Max: 6.49 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
Selenium Result
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox Original low_power balanced performance 40 80 120 160 200 SE +/- 0.99, N = 3 SE +/- 0.30, N = 3 SE +/- 0.93, N = 3 SE +/- 1.27, N = 3 180.61 175.98 176.89 176.65 1. firefox 117.0 CPU Temp
Min Avg Max Original 42.9 79.5 86.3 low_power 43.8 79.2 86.3 balanced 40.1 73.9 94.8 performance 43.3 82.2 95.3 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox low_power balanced performance 3 6 9 12 15 10.219 11.660 9.797 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1397 4575 5116 balanced 1600 4107 5115 performance 1600 4603 5115 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 17.2 30.2 balanced 2.4 15.2 35.5 performance 2.3 18.0 38.8 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 2216 Joules per run. 2. balanced: Approximate power consumption of 2422 Joules per run. 3. performance: Approximate power consumption of 2308 Joules per run. Result Confidence
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox Original low_power balanced performance 30 60 90 120 150 Min: 178.63 / Avg: 180.61 / Max: 181.75 Min: 175.4 / Avg: 175.98 / Max: 176.37 Min: 175.19 / Avg: 176.89 / Max: 178.39 Min: 174.68 / Avg: 176.65 / Max: 179.04 1. firefox 117.0
WireGuard + Linux Networking Stack Stress Test This is a benchmark of the WireGuard secure VPN tunnel and Linux networking stack stress test. The test runs on the local host but does require root permissions to run. The way it works is it creates three namespaces. ns0 has a loopback device. ns1 and ns2 each have wireguard devices. Those two wireguard devices send traffic through the loopback device of ns0. The end result of this is that tests wind up testing encryption and decryption at the same time -- a pretty CPU and scheduler-heavy workflow. Learn more via the OpenBenchmarking.org test page.
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.
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/ao/real_time Original low_power balanced performance 0.6656 1.3312 1.9968 2.6624 3.328 SE +/- 0.01679, N = 3 SE +/- 0.00372, N = 3 SE +/- 0.00195, N = 3 SE +/- 0.00074, N = 3 2.72310 2.74234 2.92879 2.95816 CPU Temp
Min Avg Max Original 50.1 71.1 72.8 low_power 47.8 69.0 72.9 balanced 52.8 79.3 83.3 performance 53.3 80.8 88.1 OpenBenchmarking.org Celsius, Fewer Is Better OSPRay 2.12 CPU Temperature Monitor 20 40 60 80 100 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 2.12 Benchmark: particle_volume/ao/real_time low_power balanced performance 0.0241 0.0482 0.0723 0.0964 0.1205 0.107 0.092 0.092 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1397 3602 4078 balanced 1600 3829 4907 performance 1600 3875 4341 OpenBenchmarking.org Megahertz, More Is Better OSPRay 2.12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.6 30.3 balanced 2.7 31.9 36.4 performance 2.6 32.1 40.4 OpenBenchmarking.org Watts, Fewer Is Better OSPRay 2.12 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 3213 Joules per run. 2. balanced: Approximate power consumption of 3903 Joules per run. 3. performance: Approximate power consumption of 3937 Joules per run. Result Confidence
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/ao/real_time Original low_power balanced performance 2 4 6 8 10 Min: 2.69 / Avg: 2.72 / Max: 2.74 Min: 2.74 / Avg: 2.74 / Max: 2.75 Min: 2.92 / Avg: 2.93 / Max: 2.93 Min: 2.96 / Avg: 2.96 / Max: 2.96
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 1024 Original low_power balanced performance 0.3374 0.6748 1.0122 1.3496 1.687 SE +/- 0.001250, N = 3 SE +/- 0.000723, N = 3 SE +/- 0.001530, N = 3 SE +/- 0.004599, N = 3 1.384144 1.384806 1.499450 1.497446 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions CPU Temp
Min Avg Max Original 49.5 72.2 84.9 low_power 47.9 69.6 84.1 balanced 51.8 80.2 93.1 performance 52.1 81.1 89.6 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100 Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 1024 low_power balanced performance 0.0124 0.0248 0.0372 0.0496 0.062 0.055 0.048 0.047 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3515 4878 balanced 1600 3819 4912 performance 1600 3825 4891 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.2 30.1 balanced 2.6 31.4 36.2 performance 2.6 31.7 40.1 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 3053 Joules per run. 2. balanced: Approximate power consumption of 3517 Joules per run. 3. performance: Approximate power consumption of 3565 Joules per run. Result Confidence
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 1024 Original low_power balanced performance 2 4 6 8 10 Min: 1.38 / Avg: 1.38 / Max: 1.39 Min: 1.38 / Avg: 1.38 / Max: 1.39 Min: 1.5 / Avg: 1.5 / Max: 1.5 Min: 1.49 / Avg: 1.5 / Max: 1.51 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
Numpy Benchmark This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Score, More Is Better Numpy Benchmark Original low_power balanced performance 160 320 480 640 800 SE +/- 3.63, N = 3 SE +/- 1.59, N = 3 SE +/- 1.57, N = 3 SE +/- 4.82, N = 3 755.31 748.85 741.01 745.52 CPU Temp
Min Avg Max Original 43.5 83.6 86.3 low_power 43.5 83.6 86.4 balanced 44.5 86.0 89.6 performance 44.3 85.5 89.8 OpenBenchmarking.org Celsius, Fewer Is Better Numpy Benchmark CPU Temperature Monitor 20 40 60 80 100 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Numpy Benchmark low_power balanced performance 10 20 30 40 50 43.80 44.26 43.79 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4835 5116 balanced 1600 4816 5065 performance 1600 4864 5115 OpenBenchmarking.org Megahertz, More Is Better Numpy Benchmark CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.2 17.1 22.7 balanced 2.4 16.7 22.6 performance 2.4 17.0 22.8 OpenBenchmarking.org Watts, Fewer Is Better Numpy Benchmark CPU Power Consumption Monitor 6 12 18 24 30 1. low_power: Approximate power consumption of 1869 Joules per run. 2. balanced: Approximate power consumption of 1858 Joules per run. 3. performance: Approximate power consumption of 1856 Joules per run. Result Confidence
OpenBenchmarking.org Score, More Is Better Numpy Benchmark Original low_power balanced performance 130 260 390 520 650 Min: 748.75 / Avg: 755.31 / Max: 761.29 Min: 745.69 / Avg: 748.85 / Max: 750.73 Min: 739.27 / Avg: 741.01 / Max: 744.14 Min: 735.95 / Avg: 745.52 / Max: 751.26
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: Kostya Original low_power balanced performance 1.17 2.34 3.51 4.68 5.85 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 5.20 5.11 5.16 5.20 1. (CXX) g++ options: -O3 CPU Temp
Min Avg Max Original 47.3 81.0 83.1 low_power 45.9 81.3 83.1 balanced 47.9 84.8 87.4 performance 47.6 84.6 87.3 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 2.0 CPU Temperature Monitor 20 40 60 80 100 GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 2.0 Throughput Test: Kostya low_power balanced performance 0.0635 0.127 0.1905 0.254 0.3175 0.276 0.282 0.279 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4868 5041 balanced 1600 4831 5060 performance 1600 4833 5015 OpenBenchmarking.org Megahertz, More Is Better simdjson 2.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 18.5 20.2 balanced 2.5 18.3 19.9 performance 2.4 18.7 20.1 OpenBenchmarking.org Watts, Fewer Is Better simdjson 2.0 CPU Power Consumption Monitor 6 12 18 24 30 1. low_power: Approximate power consumption of 2003 Joules per run. 2. balanced: Approximate power consumption of 1981 Joules per run. 3. performance: Approximate power consumption of 1996 Joules per run. Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: Kostya Original low_power balanced performance 2 4 6 8 10 Min: 5.2 / Avg: 5.2 / Max: 5.21 Min: 5.05 / Avg: 5.11 / Max: 5.16 Min: 5.15 / Avg: 5.16 / Max: 5.17 Min: 5.13 / Avg: 5.2 / Max: 5.24 1. (CXX) g++ options: -O3
Selenium Result
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Google Chrome Original low_power balanced performance 60 120 180 240 300 SE +/- 2.41, N = 3 SE +/- 2.13, N = 3 SE +/- 0.72, N = 3 SE +/- 2.29, N = 8 280.17 273.15 276.75 274.72 1. chrome 116.0.5845.179 CPU Temp
Min Avg Max Original 45.6 77.4 86.1 low_power 44.4 77.4 85.9 balanced 45.8 81.3 92.5 performance 45.4 81.8 94.3 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Google Chrome low_power balanced performance 4 8 12 16 20 16.02 15.48 15.27 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4386 5091 balanced 1600 4386 5115 performance 1559 4385 5064 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 17.1 30.0 balanced 2.4 17.9 36.1 performance 2.3 18.0 40.1 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1267 Joules per run. 2. balanced: Approximate power consumption of 1329 Joules per run. 3. performance: Approximate power consumption of 1356 Joules per run. Result Confidence
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Google Chrome Original low_power balanced performance 50 100 150 200 250 Min: 277.57 / Avg: 280.17 / Max: 284.99 Min: 269.64 / Avg: 273.15 / Max: 277.01 Min: 275.59 / Avg: 276.75 / Max: 278.07 Min: 259.6 / Avg: 274.72 / Max: 280.99 1. chrome 116.0.5845.179
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 Original low_power balanced performance 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 20.84 20.85 22.03 22.07 CPU Temp
Min Avg Max Original 47.1 69.1 80.3 low_power 45.5 66.1 79.4 balanced 46.1 75.2 82.3 performance 46.3 75.7 84.0 OpenBenchmarking.org Celsius, Fewer Is Better TensorFlow 2.12 CPU Temperature Monitor 20 40 60 80 100 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 low_power balanced performance 0.1861 0.3722 0.5583 0.7444 0.9305 0.827 0.699 0.690 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3512 5116 balanced 1600 3835 5065 performance 1600 3846 5115 OpenBenchmarking.org Megahertz, More Is Better TensorFlow 2.12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.2 30.4 balanced 2.5 31.5 36.5 performance 2.3 32.0 40.5 OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.12 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 2318 Joules per run. 2. balanced: Approximate power consumption of 2762 Joules per run. 3. performance: Approximate power consumption of 2782 Joules per run. Result Confidence
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 Original low_power balanced performance 5 10 15 20 25 Min: 20.83 / Avg: 20.84 / Max: 20.86 Min: 20.83 / Avg: 20.85 / Max: 20.86 Min: 21.99 / Avg: 22.03 / Max: 22.09 Min: 22.02 / Avg: 22.07 / Max: 22.16
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Exhaustive Original low_power balanced performance 0.1482 0.2964 0.4446 0.5928 0.741 SE +/- 0.0030, N = 3 SE +/- 0.0027, N = 3 SE +/- 0.0037, N = 3 SE +/- 0.0076, N = 3 0.5889 0.5874 0.6514 0.6587 1. (CXX) g++ options: -O3 -flto -pthread CPU Temp
Min Avg Max Original 49.3 70.4 83.6 low_power 48.1 69.6 83.6 balanced 50.3 78.3 87.6 performance 50.1 79.1 87.3 OpenBenchmarking.org Celsius, Fewer Is Better ASTC Encoder 4.0 CPU Temperature Monitor 20 40 60 80 100 MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.0 Preset: Exhaustive low_power balanced performance 0.0052 0.0104 0.0156 0.0208 0.026 0.023 0.021 0.021 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3223 5016 balanced 1600 3499 4950 performance 1600 3568 5096 OpenBenchmarking.org Megahertz, More Is Better ASTC Encoder 4.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 25.3 30.3 balanced 2.5 31.4 36.3 performance 2.4 31.8 40.4 OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.0 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 2256 Joules per run. 2. balanced: Approximate power consumption of 2564 Joules per run. 3. performance: Approximate power consumption of 2595 Joules per run. Result Confidence
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Exhaustive Original low_power balanced performance 2 4 6 8 10 Min: 0.58 / Avg: 0.59 / Max: 0.59 Min: 0.58 / Avg: 0.59 / Max: 0.59 Min: 0.65 / Avg: 0.65 / Max: 0.66 Min: 0.65 / Avg: 0.66 / Max: 0.67 1. (CXX) g++ options: -O3 -flto -pthread
VVenC VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 4K - Video Preset: Faster Original low_power balanced performance 2 4 6 8 10 SE +/- 0.036, N = 3 SE +/- 0.014, N = 3 SE +/- 0.031, N = 3 SE +/- 0.037, N = 3 7.481 7.522 8.121 8.163 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects CPU Temp
Min Avg Max Original 49.5 73.3 78.1 low_power 49.1 72.5 77.0 balanced 52.0 83.9 88.3 performance 53.6 84.6 90.1 OpenBenchmarking.org Celsius, Fewer Is Better VVenC 1.9 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VVenC 1.9 Video Input: Bosphorus 4K - Video Preset: Faster low_power balanced performance 0.0668 0.1336 0.2004 0.2672 0.334 0.297 0.258 0.256 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3430 4989 balanced 1600 3753 4951 performance 1600 3720 4946 OpenBenchmarking.org Megahertz, More Is Better VVenC 1.9 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.3 30.3 balanced 2.6 31.5 36.3 performance 2.6 31.9 40.4 OpenBenchmarking.org Watts, Fewer Is Better VVenC 1.9 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 2128 Joules per run. 2. balanced: Approximate power consumption of 2459 Joules per run. 3. performance: Approximate power consumption of 2475 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 4K - Video Preset: Faster Original low_power balanced performance 3 6 9 12 15 Min: 7.41 / Avg: 7.48 / Max: 7.54 Min: 7.5 / Avg: 7.52 / Max: 7.55 Min: 8.07 / Avg: 8.12 / Max: 8.18 Min: 8.13 / Avg: 8.16 / Max: 8.24 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 1024 Original low_power balanced performance 0.5094 1.0188 1.5282 2.0376 2.547 SE +/- 0.001815, N = 3 SE +/- 0.002513, N = 3 SE +/- 0.007131, N = 3 SE +/- 0.011161, N = 3 2.043095 2.047177 2.264202 2.259554 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions CPU Temp
Min Avg Max Original 49.3 71.3 79.5 low_power 47.5 69.4 80.1 balanced 51.5 79.6 84.8 performance 52.0 80.7 87.4 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100 Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 1024 low_power balanced performance 0.0182 0.0364 0.0546 0.0728 0.091 0.081 0.072 0.071 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3526 4908 balanced 1600 3753 4940 performance 1600 3754 4890 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.1 30.1 balanced 2.6 31.4 36.1 performance 2.6 31.9 40.1 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 2070 Joules per run. 2. balanced: Approximate power consumption of 2366 Joules per run. 3. performance: Approximate power consumption of 2401 Joules per run. Result Confidence
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 1024 Original low_power balanced performance 2 4 6 8 10 Min: 2.04 / Avg: 2.04 / Max: 2.05 Min: 2.04 / Avg: 2.05 / Max: 2.05 Min: 2.26 / Avg: 2.26 / Max: 2.28 Min: 2.25 / Avg: 2.26 / Max: 2.28 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium Original low_power balanced performance 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 7.65 7.64 8.48 8.51 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt CPU Temp
Min Avg Max Original 45.5 71.4 76.1 low_power 44.3 69.4 74.8 balanced 46.5 78.9 84.4 performance 47.4 80.4 88.6 OpenBenchmarking.org Celsius, Fewer Is Better Kvazaar 2.2 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium low_power balanced performance 0.0675 0.135 0.2025 0.27 0.3375 0.300 0.266 0.263 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1483 3128 4977 balanced 1600 3437 4934 performance 1600 3481 4952 OpenBenchmarking.org Megahertz, More Is Better Kvazaar 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.4 30.2 balanced 2.5 31.9 36.3 performance 2.4 32.4 40.2 OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 2094 Joules per run. 2. balanced: Approximate power consumption of 2381 Joules per run. 3. performance: Approximate power consumption of 2419 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium Original low_power balanced performance 3 6 9 12 15 Min: 7.64 / Avg: 7.65 / Max: 7.67 Min: 7.63 / Avg: 7.64 / Max: 7.66 Min: 8.43 / Avg: 8.48 / Max: 8.56 Min: 8.42 / Avg: 8.51 / Max: 8.67 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
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.
Zstd Compression This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) 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 1.5.4 Compression Level: 19 - Decompression Speed Original low_power balanced performance 400 800 1200 1600 2000 SE +/- 4.33, N = 3 SE +/- 8.20, N = 3 SE +/- 3.92, N = 3 SE +/- 2.16, N = 3 1685.5 1687.3 1693.8 1685.4 1. (CC) gcc options: -O3 -pthread -lz -llzma CPU Temp
Min Avg Max Original 47.9 80.9 86.3 low_power 47.3 79.9 86.3 balanced 48.3 85.9 93.8 performance 48.5 87.5 96.1 OpenBenchmarking.org Celsius, Fewer Is Better Zstd Compression 1.5.4 CPU Temperature Monitor 20 40 60 80 100 MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed low_power balanced performance 20 40 60 80 100 79.66 71.25 68.28 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4618 5116 balanced 1600 4724 5115 performance 1600 4712 5015 OpenBenchmarking.org Megahertz, More Is Better Zstd Compression 1.5.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 21.2 30.3 balanced 2.5 23.8 36.3 performance 2.4 24.7 40.4 OpenBenchmarking.org Watts, Fewer Is Better Zstd Compression 1.5.4 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1553 Joules per run. 2. balanced: Approximate power consumption of 1735 Joules per run. 3. performance: Approximate power consumption of 1802 Joules per run. Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed Original low_power balanced performance 300 600 900 1200 1500 Min: 1677.3 / Avg: 1685.53 / Max: 1692 Min: 1675.8 / Avg: 1687.33 / Max: 1703.2 Min: 1689.2 / Avg: 1693.8 / Max: 1701.6 Min: 1682 / Avg: 1685.4 / Max: 1689.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
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 3.7 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K Original low_power balanced performance 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 10.81 10.77 11.52 11.66 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 49.8 72.8 76.5 low_power 48.4 70.5 75.9 balanced 51.3 79.4 83.3 performance 51.8 80.8 88.0 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K low_power balanced performance 0.0961 0.1922 0.2883 0.3844 0.4805 0.427 0.371 0.367 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3554 5066 balanced 1600 3946 5065 performance 1600 3947 5067 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.2 25.2 29.2 balanced 2.6 31.1 34.4 performance 2.5 31.7 38.3 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1883 Joules per run. 2. balanced: Approximate power consumption of 2195 Joules per run. 3. performance: Approximate power consumption of 2211 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K Original low_power balanced performance 3 6 9 12 15 Min: 10.77 / Avg: 10.81 / Max: 10.85 Min: 10.73 / Avg: 10.77 / Max: 10.8 Min: 11.47 / Avg: 11.52 / Max: 11.57 Min: 11.61 / Avg: 11.66 / Max: 11.73 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: LuxCore Benchmark - Acceleration: CPU Original low_power balanced performance 0.3465 0.693 1.0395 1.386 1.7325 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 4 1.35 1.36 1.53 1.54 MIN: 0.38 / MAX: 1.68 MIN: 0.4 / MAX: 1.67 MIN: 0.44 / MAX: 1.89 MIN: 0.46 / MAX: 1.9 CPU Temp
Min Avg Max Original 47.6 70.7 76.4 low_power 46.5 68.0 73.5 balanced 50.6 78.6 82.8 performance 52.0 80.3 87.5 OpenBenchmarking.org Celsius, Fewer Is Better LuxCoreRender 2.6 CPU Temperature Monitor 20 40 60 80 100 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: LuxCore Benchmark - Acceleration: CPU low_power balanced performance 0.0122 0.0244 0.0366 0.0488 0.061 0.054 0.048 0.048 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3248 4966 balanced 1600 3525 5036 performance 1600 3570 5086 OpenBenchmarking.org Megahertz, More Is Better LuxCoreRender 2.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.5 25.2 30.3 balanced 2.5 31.8 36.3 performance 1.5 32.0 40.4 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1678 Joules per run. 2. balanced: Approximate power consumption of 2109 Joules per run. 3. performance: Approximate power consumption of 2115 Joules per run. Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: LuxCore Benchmark - Acceleration: CPU Original low_power balanced performance 2 4 6 8 10 Min: 1.34 / Avg: 1.35 / Max: 1.37 Min: 1.35 / Avg: 1.36 / Max: 1.36 Min: 1.52 / Avg: 1.53 / Max: 1.55 Min: 1.51 / Avg: 1.54 / Max: 1.59
Zstd Compression This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) 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 1.5.4 Compression Level: 19, Long Mode - Decompression Speed Original low_power balanced performance 300 600 900 1200 1500 SE +/- 16.54, N = 3 SE +/- 0.52, N = 3 SE +/- 6.79, N = 3 SE +/- 7.01, N = 3 1626.0 1609.9 1618.2 1623.2 1. (CC) gcc options: -O3 -pthread -lz -llzma CPU Temp
Min Avg Max Original 47.9 81.9 86.3 low_power 46.5 81.4 86.4 balanced 48.0 83.3 89.6 performance 47.0 83.9 90.3 OpenBenchmarking.org Celsius, Fewer Is Better Zstd Compression 1.5.4 CPU Temperature Monitor 20 40 60 80 100 MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed low_power balanced performance 20 40 60 80 100 82.17 79.33 79.11 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4738 5041 balanced 1600 4796 5040 performance 1600 4790 5040 OpenBenchmarking.org Megahertz, More Is Better Zstd Compression 1.5.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 19.6 29.8 balanced 2.6 20.4 32.9 performance 2.3 20.5 32.9 OpenBenchmarking.org Watts, Fewer Is Better Zstd Compression 1.5.4 CPU Power Consumption Monitor 10 20 30 40 50 1. low_power: Approximate power consumption of 1326 Joules per run. 2. balanced: Approximate power consumption of 1380 Joules per run. 3. performance: Approximate power consumption of 1382 Joules per run. Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed Original low_power balanced performance 300 600 900 1200 1500 Min: 1606.1 / Avg: 1625.97 / Max: 1658.8 Min: 1609.1 / Avg: 1609.93 / Max: 1610.9 Min: 1607.9 / Avg: 1618.17 / Max: 1631 Min: 1611.6 / Avg: 1623.17 / Max: 1635.8 1. (CC) gcc options: -O3 -pthread -lz -llzma
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.1 Binary: Pathtracer ISPC - Model: Crown Original low_power balanced performance 3 6 9 12 15 SE +/- 0.0180, N = 3 SE +/- 0.0407, N = 3 SE +/- 0.0530, N = 3 SE +/- 0.0971, N = 3 9.8253 9.8514 10.4397 10.5212 MIN: 9.68 / MAX: 10.19 MIN: 9.65 / MAX: 10.41 MIN: 10.2 / MAX: 10.89 MIN: 10.23 / MAX: 11.29 CPU Temp
Min Avg Max Original 51.3 70.2 72.3 low_power 48.8 66.5 70.0 balanced 53.4 78.0 82.0 performance 54.3 79.6 86.0 OpenBenchmarking.org Celsius, Fewer Is Better Embree 4.1 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 4.1 Binary: Pathtracer ISPC - Model: Crown low_power balanced performance 0.0875 0.175 0.2625 0.35 0.4375 0.389 0.328 0.328 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3425 4997 balanced 1600 3695 5015 performance 1600 3754 5009 OpenBenchmarking.org Megahertz, More Is Better Embree 4.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.3 30.3 balanced 2.6 31.8 36.3 performance 2.7 32.1 40.4 OpenBenchmarking.org Watts, Fewer Is Better Embree 4.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1762 Joules per run. 2. balanced: Approximate power consumption of 2089 Joules per run. 3. performance: Approximate power consumption of 2095 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.1 Binary: Pathtracer ISPC - Model: Crown Original low_power balanced performance 3 6 9 12 15 Min: 9.79 / Avg: 9.83 / Max: 9.85 Min: 9.78 / Avg: 9.85 / Max: 9.92 Min: 10.35 / Avg: 10.44 / Max: 10.54 Min: 10.39 / Avg: 10.52 / Max: 10.71
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU Original low_power balanced performance 0.7088 1.4176 2.1264 2.8352 3.544 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 2.83 2.84 3.10 3.15 MIN: 2.77 / MAX: 2.97 MIN: 2.77 / MAX: 2.97 MIN: 3.04 / MAX: 3.16 MIN: 3.05 / MAX: 3.28 CPU Temp
Min Avg Max Original 49.9 70.8 75.8 low_power 48.4 67.5 72.9 balanced 52.0 78.5 82.4 performance 53.0 80.3 87.0 OpenBenchmarking.org Celsius, Fewer Is Better LuxCoreRender 2.6 CPU Temperature Monitor 20 40 60 80 100 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU low_power balanced performance 0.0254 0.0508 0.0762 0.1016 0.127 0.113 0.098 0.098 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3224 5016 balanced 1397 3515 5112 performance 1600 3552 4943 OpenBenchmarking.org Megahertz, More Is Better LuxCoreRender 2.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.2 30.3 balanced 2.6 31.5 36.3 performance 2.5 32.1 40.4 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1694 Joules per run. 2. balanced: Approximate power consumption of 2124 Joules per run. 3. performance: Approximate power consumption of 2161 Joules per run. Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU Original low_power balanced performance 2 4 6 8 10 Min: 2.83 / Avg: 2.83 / Max: 2.84 Min: 2.83 / Avg: 2.84 / Max: 2.86 Min: 3.09 / Avg: 3.1 / Max: 3.12 Min: 3.12 / Avg: 3.15 / Max: 3.21
Zstd Compression This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) 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 1.5.4 Compression Level: 8, Long Mode - Decompression Speed Original low_power balanced performance 400 800 1200 1600 2000 SE +/- 12.89, N = 3 SE +/- 3.44, N = 3 SE +/- 5.23, N = 3 SE +/- 1.53, N = 3 2076.8 2047.3 2042.2 2037.8 1. (CC) gcc options: -O3 -pthread -lz -llzma CPU Temp
Min Avg Max Original 48.0 78.9 86.0 low_power 48.0 78.5 86.1 balanced 52.8 85.2 94.1 performance 53.1 87.7 94.8 OpenBenchmarking.org Celsius, Fewer Is Better Zstd Compression 1.5.4 CPU Temperature Monitor 20 40 60 80 100 MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.5.4 Compression Level: 8, Long Mode - Decompression Speed low_power balanced performance 20 40 60 80 100 95.82 83.28 77.67 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4376 5085 balanced 1600 4453 5065 performance 1600 4529 5019 OpenBenchmarking.org Megahertz, More Is Better Zstd Compression 1.5.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.1 21.4 30.3 balanced 2.7 24.5 36.1 performance 2.5 26.2 40.1 OpenBenchmarking.org Watts, Fewer Is Better Zstd Compression 1.5.4 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1403 Joules per run. 2. balanced: Approximate power consumption of 1627 Joules per run. 3. performance: Approximate power consumption of 1740 Joules per run. Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8, Long Mode - Decompression Speed Original low_power balanced performance 400 800 1200 1600 2000 Min: 2058.6 / Avg: 2076.8 / Max: 2101.7 Min: 2041.3 / Avg: 2047.3 / Max: 2053.2 Min: 2034.6 / Avg: 2042.17 / Max: 2052.2 Min: 2035.9 / Avg: 2037.77 / Max: 2040.8 1. (CC) gcc options: -O3 -pthread -lz -llzma
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 12 - Decompression Speed Original low_power balanced performance 400 800 1200 1600 2000 SE +/- 8.36, N = 3 SE +/- 87.22, N = 3 SE +/- 2.15, N = 3 SE +/- 3.72, N = 3 2081.5 1984.5 2064.0 2064.7 1. (CC) gcc options: -O3 -pthread -lz -llzma CPU Temp
Min Avg Max Original 47.8 79.1 86.3 low_power 46.3 78.3 86.3 balanced 49.5 83.9 93.0 performance 50.5 85.3 93.9 OpenBenchmarking.org Celsius, Fewer Is Better Zstd Compression 1.5.4 CPU Temperature Monitor 20 40 60 80 100 MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.5.4 Compression Level: 12 - Decompression Speed low_power balanced performance 20 40 60 80 100 94.14 87.21 81.73 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4543 5091 balanced 1600 4589 5090 performance 1600 4653 5065 OpenBenchmarking.org Megahertz, More Is Better Zstd Compression 1.5.4 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 21.1 30.3 balanced 2.7 23.7 36.4 performance 2.0 25.3 40.4 OpenBenchmarking.org Watts, Fewer Is Better Zstd Compression 1.5.4 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1391 Joules per run. 2. balanced: Approximate power consumption of 1609 Joules per run. 3. performance: Approximate power consumption of 1667 Joules per run. Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 12 - Decompression Speed Original low_power balanced performance 400 800 1200 1600 2000 Min: 2065.9 / Avg: 2081.53 / Max: 2094.5 Min: 1810.2 / Avg: 1984.5 / Max: 2077.5 Min: 2059.8 / Avg: 2064.03 / Max: 2066.8 Min: 2060.9 / Avg: 2064.67 / Max: 2072.1 1. (CC) gcc options: -O3 -pthread -lz -llzma
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU Original low_power balanced performance 0.4455 0.891 1.3365 1.782 2.2275 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 1.70 1.77 1.94 1.98 MIN: 1.63 / MAX: 1.82 MIN: 1.68 / MAX: 1.88 MIN: 1.85 / MAX: 2.03 MIN: 1.86 / MAX: 2.11 CPU Temp
Min Avg Max Original 49.3 68.5 71.1 low_power 48.6 67.8 71.9 balanced 53.0 79.0 83.3 performance 53.6 80.7 87.9 OpenBenchmarking.org Celsius, Fewer Is Better LuxCoreRender 2.6 CPU Temperature Monitor 20 40 60 80 100 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU low_power balanced performance 0.0158 0.0316 0.0474 0.0632 0.079 0.070 0.061 0.062 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3176 3933 balanced 1600 3451 4155 performance 1600 3481 4841 OpenBenchmarking.org Megahertz, More Is Better LuxCoreRender 2.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.3 25.4 30.3 balanced 2.7 31.8 36.3 performance 2.6 32.2 40.4 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1640 Joules per run. 2. balanced: Approximate power consumption of 2068 Joules per run. 3. performance: Approximate power consumption of 2080 Joules per run. Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU Original low_power balanced performance 2 4 6 8 10 Min: 1.69 / Avg: 1.7 / Max: 1.7 Min: 1.76 / Avg: 1.77 / Max: 1.78 Min: 1.93 / Avg: 1.94 / Max: 1.96 Min: 1.96 / Avg: 1.98 / Max: 2.01
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K Original low_power balanced performance 0.6536 1.3072 1.9608 2.6144 3.268 SE +/- 0.007, N = 3 SE +/- 0.007, N = 3 SE +/- 0.020, N = 3 SE +/- 0.021, N = 3 2.629 2.639 2.900 2.905 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq CPU Temp
Min Avg Max Original 49.0 72.8 78.8 low_power 47.6 70.5 76.4 balanced 50.0 80.5 88.1 performance 51.0 82.1 89.3 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K low_power balanced performance 0.0236 0.0472 0.0708 0.0944 0.118 0.105 0.092 0.091 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3347 5063 balanced 1600 3701 5115 performance 1600 3679 5047 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.2 30.2 balanced 2.5 31.4 36.8 performance 2.6 32.0 40.1 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1645 Joules per run. 2. balanced: Approximate power consumption of 1884 Joules per run. 3. performance: Approximate power consumption of 1911 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K Original low_power balanced performance 2 4 6 8 10 Min: 2.62 / Avg: 2.63 / Max: 2.64 Min: 2.63 / Avg: 2.64 / Max: 2.65 Min: 2.88 / Avg: 2.9 / Max: 2.94 Min: 2.88 / Avg: 2.91 / Max: 2.95 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
libxsmm Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 64 Original low_power balanced performance 50 100 150 200 250 SE +/- 0.68, N = 3 SE +/- 0.37, N = 3 SE +/- 0.15, N = 3 SE +/- 0.28, N = 3 188.0 194.8 205.5 207.3 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2 CPU Temp
Min Avg Max Original 48.5 58.6 59.9 low_power 48.1 58.5 59.9 balanced 53.8 70.2 73.4 performance 55.5 77.4 81.5 OpenBenchmarking.org Celsius, Fewer Is Better libxsmm 2-1.17-3645 CPU Temperature Monitor 20 40 60 80 100 GFLOPS/s Per Watt
OpenBenchmarking.org GFLOPS/s Per Watt, More Is Better libxsmm 2-1.17-3645 M N K: 64 low_power balanced performance 3 6 9 12 15 9.938 7.567 6.474 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 2712 3693 balanced 1600 3494 4951 performance 1600 3735 4142 OpenBenchmarking.org Megahertz, More Is Better libxsmm 2-1.17-3645 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 19.6 26.6 balanced 2.7 27.2 36.0 performance 2.6 32.0 40.0 OpenBenchmarking.org Watts, Fewer Is Better libxsmm 2-1.17-3645 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1241 Joules per run. 2. balanced: Approximate power consumption of 1638 Joules per run. 3. performance: Approximate power consumption of 1910 Joules per run. Result Confidence
OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 64 Original low_power balanced performance 40 80 120 160 200 Min: 186.7 / Avg: 188 / Max: 189 Min: 194.1 / Avg: 194.83 / Max: 195.3 Min: 205.2 / Avg: 205.47 / Max: 205.7 Min: 207 / Avg: 207.33 / Max: 207.9 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2
Intel Open Image Denoise Open Image Denoise is a denoising library for ray-tracing and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.0 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only Original low_power balanced performance 0.081 0.162 0.243 0.324 0.405 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.33 0.33 0.36 0.36 CPU Temp
Min Avg Max Original 50.1 68.9 71.8 low_power 47.5 65.7 68.9 balanced 52.4 75.0 79.1 performance 52.6 76.4 82.0 OpenBenchmarking.org Celsius, Fewer Is Better Intel Open Image Denoise 2.0 CPU Temperature Monitor 20 40 60 80 100 Images / Sec Per Watt
OpenBenchmarking.org Images / Sec Per Watt, More Is Better Intel Open Image Denoise 2.0 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only low_power balanced performance 0.0029 0.0058 0.0087 0.0116 0.0145 0.013 0.011 0.011 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3424 4632 balanced 1600 3752 5065 performance 1600 3730 4784 OpenBenchmarking.org Megahertz, More Is Better Intel Open Image Denoise 2.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.3 30.3 balanced 2.7 31.7 36.4 performance 2.6 32.1 40.4 OpenBenchmarking.org Watts, Fewer Is Better Intel Open Image Denoise 2.0 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1617 Joules per run. 2. balanced: Approximate power consumption of 1904 Joules per run. 3. performance: Approximate power consumption of 1915 Joules per run. Result Confidence
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.0 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only Original low_power balanced performance 1 2 3 4 5 Min: 0.33 / Avg: 0.33 / Max: 0.33 Min: 0.33 / Avg: 0.33 / Max: 0.33 Min: 0.36 / Avg: 0.36 / Max: 0.36 Min: 0.36 / Avg: 0.36 / Max: 0.37
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 3.7 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K Original low_power balanced performance 13 26 39 52 65 SE +/- 0.47, N = 15 SE +/- 0.52, N = 15 SE +/- 0.57, N = 15 SE +/- 0.63, N = 15 56.11 55.38 58.39 58.40 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 46.0 67.6 75.1 low_power 44.6 66.4 73.8 balanced 46.0 71.0 79.0 performance 48.3 74.0 81.4 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K low_power balanced performance 0.6093 1.2186 1.8279 2.4372 3.0465 2.708 2.538 2.391 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3850 5067 balanced 1397 3953 5094 performance 1600 4096 5059 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 20.5 27.4 balanced 2.4 23.0 32.9 performance 2.3 24.4 33.6 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 10 20 30 40 50 1. low_power: Approximate power consumption of 327 Joules per run. 2. balanced: Approximate power consumption of 354 Joules per run. 3. performance: Approximate power consumption of 374 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K Original low_power balanced performance 12 24 36 48 60 Min: 52.78 / Avg: 56.11 / Max: 58.31 Min: 52.62 / Avg: 55.38 / Max: 58.02 Min: 54.1 / Avg: 58.39 / Max: 60.52 Min: 55.3 / Avg: 58.4 / Max: 61.45 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
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 Original low_power balanced performance 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 7.58 7.46 7.46 7.48 1. (CXX) g++ options: -O3 CPU Temp
Min Avg Max Original 43.3 83.5 86.3 low_power 42.6 83.6 86.3 balanced 44.5 85.7 88.9 performance 44.8 85.7 88.5 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 2.0 CPU Temperature Monitor 20 40 60 80 100 GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 2.0 Throughput Test: PartialTweets low_power balanced performance 0.0979 0.1958 0.2937 0.3916 0.4895 0.418 0.435 0.428 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4808 5016 balanced 1600 4721 4991 performance 1600 4750 4990 OpenBenchmarking.org Megahertz, More Is Better simdjson 2.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.2 17.9 21.3 balanced 2.5 17.2 20.3 performance 2.3 17.5 20.4 OpenBenchmarking.org Watts, Fewer Is Better simdjson 2.0 CPU Power Consumption Monitor 6 12 18 24 30 1. low_power: Approximate power consumption of 1102 Joules per run. 2. balanced: Approximate power consumption of 1064 Joules per run. 3. performance: Approximate power consumption of 1073 Joules per run. Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: PartialTweets Original low_power balanced performance 3 6 9 12 15 Min: 7.55 / Avg: 7.58 / Max: 7.62 Min: 7.44 / Avg: 7.46 / Max: 7.47 Min: 7.41 / Avg: 7.46 / Max: 7.5 Min: 7.45 / Avg: 7.48 / Max: 7.53 1. (CXX) g++ options: -O3
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: DistinctUserID Original low_power balanced performance 2 4 6 8 10 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 8.78 8.71 8.71 8.88 1. (CXX) g++ options: -O3 CPU Temp
Min Avg Max Original 48.4 83.8 86.3 low_power 47.5 83.6 86.1 balanced 48.4 86.1 88.8 performance 48.0 85.9 88.6 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 2.0 CPU Temperature Monitor 20 40 60 80 100 GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 2.0 Throughput Test: DistinctUserID low_power balanced performance 0.1148 0.2296 0.3444 0.4592 0.574 0.486 0.510 0.492 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4800 5003 balanced 1600 4732 4979 performance 1600 4752 4990 OpenBenchmarking.org Megahertz, More Is Better simdjson 2.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 17.9 20.8 balanced 2.5 17.1 19.1 performance 2.3 18.0 20.5 OpenBenchmarking.org Watts, Fewer Is Better simdjson 2.0 CPU Power Consumption Monitor 6 12 18 24 30 1. low_power: Approximate power consumption of 1105 Joules per run. 2. balanced: Approximate power consumption of 1060 Joules per run. 3. performance: Approximate power consumption of 1100 Joules per run. Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: DistinctUserID Original low_power balanced performance 3 6 9 12 15 Min: 8.65 / Avg: 8.78 / Max: 8.9 Min: 8.7 / Avg: 8.71 / Max: 8.72 Min: 8.66 / Avg: 8.71 / Max: 8.76 Min: 8.73 / Avg: 8.88 / Max: 8.96 1. (CXX) g++ options: -O3
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 480000 - Buffer Size: 1024 Original low_power balanced performance 0.6548 1.3096 1.9644 2.6192 3.274 SE +/- 0.013165, N = 3 SE +/- 0.009816, N = 3 SE +/- 0.007869, N = 3 SE +/- 0.028987, N = 3 2.632120 2.656405 2.910128 2.904394 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions CPU Temp
Min Avg Max Original 49.1 71.0 74.9 low_power 48.4 68.7 72.8 balanced 51.8 79.8 85.3 performance 52.0 81.4 89.0 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100 Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 480000 - Buffer Size: 1024 low_power balanced performance 0.0239 0.0478 0.0717 0.0956 0.1195 0.106 0.092 0.091 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3572 5087 balanced 1600 3852 4896 performance 1600 3857 4920 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 25.1 30.2 balanced 2.7 31.5 36.3 performance 2.5 32.0 40.1 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1605 Joules per run. 2. balanced: Approximate power consumption of 1857 Joules per run. 3. performance: Approximate power consumption of 1890 Joules per run. Result Confidence
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 480000 - Buffer Size: 1024 Original low_power balanced performance 2 4 6 8 10 Min: 2.61 / Avg: 2.63 / Max: 2.65 Min: 2.65 / Avg: 2.66 / Max: 2.68 Min: 2.9 / Avg: 2.91 / Max: 2.92 Min: 2.87 / Avg: 2.9 / Max: 2.96 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
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 Original low_power balanced performance 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 8.93 8.81 8.82 8.87 1. (CXX) g++ options: -O3 CPU Temp
Min Avg Max Original 46.9 83.7 86.3 low_power 45.8 83.6 86.1 balanced 46.3 86.1 89.0 performance 46.3 85.9 88.9 OpenBenchmarking.org Celsius, Fewer Is Better simdjson 2.0 CPU Temperature Monitor 20 40 60 80 100 GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 2.0 Throughput Test: TopTweet low_power balanced performance 0.117 0.234 0.351 0.468 0.585 0.498 0.520 0.502 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4791 5045 balanced 1600 4739 5013 performance 1600 4758 4974 OpenBenchmarking.org Megahertz, More Is Better simdjson 2.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 17.7 20.5 balanced 2.4 17.0 19.6 performance 2.0 17.7 20.4 OpenBenchmarking.org Watts, Fewer Is Better simdjson 2.0 CPU Power Consumption Monitor 6 12 18 24 30 1. low_power: Approximate power consumption of 1080 Joules per run. 2. balanced: Approximate power consumption of 1034 Joules per run. 3. performance: Approximate power consumption of 1078 Joules per run. Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: TopTweet Original low_power balanced performance 3 6 9 12 15 Min: 8.9 / Avg: 8.93 / Max: 8.96 Min: 8.75 / Avg: 8.81 / Max: 8.9 Min: 8.81 / Avg: 8.82 / Max: 8.82 Min: 8.83 / Avg: 8.87 / Max: 8.92 1. (CXX) g++ options: -O3
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 44100 - Buffer Size: 1024 Original low_power balanced performance 0.6721 1.3442 2.0163 2.6884 3.3605 SE +/- 0.008957, N = 3 SE +/- 0.005353, N = 3 SE +/- 0.011252, N = 3 SE +/- 0.011725, N = 3 2.730711 2.734923 2.987250 2.986766 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions CPU Temp
Min Avg Max Original 49.1 71.1 75.4 low_power 47.5 69.9 74.5 balanced 51.9 80.1 85.3 performance 52.6 81.6 89.3 OpenBenchmarking.org Celsius, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Temperature Monitor 20 40 60 80 100 Render Ratio Per Watt
OpenBenchmarking.org Render Ratio Per Watt, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 44100 - Buffer Size: 1024 low_power balanced performance 0.0245 0.049 0.0735 0.098 0.1225 0.109 0.095 0.094 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3644 4935 balanced 1600 3809 4895 performance 1600 3870 4841 OpenBenchmarking.org Megahertz, More Is Better Stargate Digital Audio Workstation 22.11.5 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 25.0 30.3 balanced 2.6 31.5 36.4 performance 2.6 31.8 40.3 OpenBenchmarking.org Watts, Fewer Is Better Stargate Digital Audio Workstation 22.11.5 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1560 Joules per run. 2. balanced: Approximate power consumption of 1807 Joules per run. 3. performance: Approximate power consumption of 1836 Joules per run. Result Confidence
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 44100 - Buffer Size: 1024 Original low_power balanced performance 2 4 6 8 10 Min: 2.71 / Avg: 2.73 / Max: 2.74 Min: 2.72 / Avg: 2.73 / Max: 2.74 Min: 2.97 / Avg: 2.99 / Max: 3.01 Min: 2.97 / Avg: 2.99 / Max: 3.01 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
Selenium Result
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Firefox Original low_power balanced performance 60 120 180 240 300 SE +/- 1.53, N = 3 SE +/- 2.40, N = 3 SE +/- 1.53, N = 3 SE +/- 0.58, N = 3 255 247 244 246 1. firefox 117.0 CPU Temp
Min Avg Max Original 42.6 56.2 69.1 low_power 41.9 53.6 63.1 balanced 42.1 54.1 67.8 performance 43.8 53.3 64.3 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Runs Per Minute Per Watt
OpenBenchmarking.org Runs Per Minute Per Watt, More Is Better Selenium Benchmark: Speedometer - Browser: Firefox low_power balanced performance 6 12 18 24 30 25.39 25.30 24.98 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1591 3328 5041 balanced 1600 3541 5062 performance 1597 3475 5073 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.2 9.7 18.4 balanced 2.3 9.6 18.8 performance 2.2 9.8 18.2 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 6 12 18 24 30 1. low_power: Approximate power consumption of 564 Joules per run. 2. balanced: Approximate power consumption of 579 Joules per run. 3. performance: Approximate power consumption of 558 Joules per run. Result Confidence
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Firefox Original low_power balanced performance 50 100 150 200 250 Min: 253 / Avg: 255 / Max: 258 Min: 242 / Avg: 246.67 / Max: 250 Min: 241 / Avg: 244 / Max: 246 Min: 245 / Avg: 246 / Max: 247 1. firefox 117.0
libxsmm Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 32 Original low_power balanced performance 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 106.9 107.0 107.1 107.0 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2 CPU Temp
Min Avg Max Original 50.4 59.7 61.5 low_power 50.0 60.1 62.4 balanced 54.6 70.9 73.8 performance 55.4 76.9 81.8 OpenBenchmarking.org Celsius, Fewer Is Better libxsmm 2-1.17-3645 CPU Temperature Monitor 20 40 60 80 100 GFLOPS/s Per Watt
OpenBenchmarking.org GFLOPS/s Per Watt, More Is Better libxsmm 2-1.17-3645 M N K: 32 low_power balanced performance 1.1736 2.3472 3.5208 4.6944 5.868 5.216 3.878 3.330 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3521 5016 balanced 1600 4010 4422 performance 1397 4196 4620 OpenBenchmarking.org Megahertz, More Is Better libxsmm 2-1.17-3645 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.5 20.5 29.7 balanced 2.8 27.6 36.1 performance 2.7 32.1 40.0 OpenBenchmarking.org Watts, Fewer Is Better libxsmm 2-1.17-3645 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1183 Joules per run. 2. balanced: Approximate power consumption of 1593 Joules per run. 3. performance: Approximate power consumption of 1853 Joules per run. Result Confidence
OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 32 Original low_power balanced performance 20 40 60 80 100 Min: 106.8 / Avg: 106.87 / Max: 106.9 Min: 106.9 / Avg: 106.97 / Max: 107 Min: 107.1 / Avg: 107.13 / Max: 107.2 Min: 106.9 / Avg: 107 / Max: 107.1 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.1 Binary: Pathtracer ISPC - Model: Asian Dragon Original low_power balanced performance 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 11.90 11.91 12.51 12.58 MIN: 11.73 / MAX: 12.31 MIN: 11.75 / MAX: 12.39 MIN: 12.28 / MAX: 12.82 MIN: 12.32 / MAX: 13.1 CPU Temp
Min Avg Max Original 50.5 69.6 72.6 low_power 48.0 65.6 69.1 balanced 52.9 77.2 81.3 performance 53.6 78.9 85.1 OpenBenchmarking.org Celsius, Fewer Is Better Embree 4.1 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 4.1 Binary: Pathtracer ISPC - Model: Asian Dragon low_power balanced performance 0.106 0.212 0.318 0.424 0.53 0.471 0.393 0.389 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3540 4991 balanced 1600 3835 4451 performance 1600 3838 4237 OpenBenchmarking.org Megahertz, More Is Better Embree 4.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 25.3 30.3 balanced 2.7 31.8 36.3 performance 2.5 32.3 40.4 OpenBenchmarking.org Watts, Fewer Is Better Embree 4.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1475 Joules per run. 2. balanced: Approximate power consumption of 1772 Joules per run. 3. performance: Approximate power consumption of 1798 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.1 Binary: Pathtracer ISPC - Model: Asian Dragon Original low_power balanced performance 4 8 12 16 20 Min: 11.85 / Avg: 11.9 / Max: 11.95 Min: 11.89 / Avg: 11.91 / Max: 11.94 Min: 12.44 / Avg: 12.51 / Max: 12.6 Min: 12.49 / Avg: 12.58 / Max: 12.75
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 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 4K Original low_power balanced performance 15 30 45 60 75 SE +/- 0.75, N = 15 SE +/- 0.75, N = 15 SE +/- 0.60, N = 15 SE +/- 0.60, N = 15 63.45 63.51 66.08 66.75 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 44.8 66.6 74.5 low_power 44.4 65.4 72.4 balanced 45.3 70.4 79.4 performance 46.5 72.4 80.9 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 4K low_power balanced performance 0.7205 1.441 2.1615 2.882 3.6025 3.202 2.997 2.852 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3974 5078 balanced 1600 4059 5082 performance 1600 4134 5073 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 19.8 27.2 balanced 2.0 22.0 30.7 performance 2.3 23.4 32.8 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 10 20 30 40 50 1. low_power: Approximate power consumption of 292 Joules per run. 2. balanced: Approximate power consumption of 316 Joules per run. 3. performance: Approximate power consumption of 332 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 4K Original low_power balanced performance 13 26 39 52 65 Min: 58.62 / Avg: 63.45 / Max: 65.53 Min: 56.45 / Avg: 63.51 / Max: 65.4 Min: 60.39 / Avg: 66.08 / Max: 67.53 Min: 61.22 / Avg: 66.75 / Max: 68.05 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/ao/real_time Original low_power balanced performance 0.5153 1.0306 1.5459 2.0612 2.5765 SE +/- 0.00425, N = 3 SE +/- 0.01311, N = 3 SE +/- 0.00759, N = 3 SE +/- 0.01583, N = 3 2.14908 2.23727 2.28403 2.29018 CPU Temp
Min Avg Max Original 51.4 71.0 80.0 low_power 48.5 68.2 78.6 balanced 53.1 79.8 87.3 performance 53.8 81.4 89.4 OpenBenchmarking.org Celsius, Fewer Is Better OSPRay 2.12 CPU Temperature Monitor 20 40 60 80 100 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/ao/real_time low_power balanced performance 0.0203 0.0406 0.0609 0.0812 0.1015 0.090 0.073 0.072 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3653 5066 balanced 1600 3975 4990 performance 1600 4014 4990 OpenBenchmarking.org Megahertz, More Is Better OSPRay 2.12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 24.8 30.3 balanced 2.7 31.2 36.3 performance 2.6 31.8 40.4 OpenBenchmarking.org Watts, Fewer Is Better OSPRay 2.12 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1388 Joules per run. 2. balanced: Approximate power consumption of 1738 Joules per run. 3. performance: Approximate power consumption of 1793 Joules per run. Result Confidence
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/ao/real_time Original low_power balanced performance 2 4 6 8 10 Min: 2.14 / Avg: 2.15 / Max: 2.15 Min: 2.22 / Avg: 2.24 / Max: 2.26 Min: 2.27 / Avg: 2.28 / Max: 2.29 Min: 2.26 / Avg: 2.29 / Max: 2.32
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time Original low_power balanced performance 0.7309 1.4618 2.1927 2.9236 3.6545 SE +/- 0.00811, N = 3 SE +/- 0.00857, N = 3 SE +/- 0.02101, N = 3 SE +/- 0.02466, N = 3 2.96819 2.98924 3.24858 3.21934 CPU Temp
Min Avg Max Original 50.4 70.7 79.4 low_power 47.8 67.9 77.8 balanced 53.1 79.2 87.0 performance 53.9 81.0 87.6 OpenBenchmarking.org Celsius, Fewer Is Better OSPRay 2.12 CPU Temperature Monitor 20 40 60 80 100 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time low_power balanced performance 0.0272 0.0544 0.0816 0.1088 0.136 0.121 0.105 0.102 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3469 5066 balanced 1600 3786 5015 performance 1600 3840 5015 OpenBenchmarking.org Megahertz, More Is Better OSPRay 2.12 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 24.7 30.3 balanced 2.8 31.0 36.3 performance 2.6 31.7 40.4 OpenBenchmarking.org Watts, Fewer Is Better OSPRay 2.12 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1386 Joules per run. 2. balanced: Approximate power consumption of 1703 Joules per run. 3. performance: Approximate power consumption of 1836 Joules per run. Result Confidence
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time Original low_power balanced performance 2 4 6 8 10 Min: 2.96 / Avg: 2.97 / Max: 2.98 Min: 2.98 / Avg: 2.99 / Max: 3.01 Min: 3.23 / Avg: 3.25 / Max: 3.29 Min: 3.17 / Avg: 3.22 / Max: 3.25
VVenC VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 1080p - Video Preset: Fast Original low_power balanced performance 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 12.06 11.98 13.11 13.20 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects CPU Temp
Min Avg Max Original 48.6 72.6 77.3 low_power 47.1 71.2 76.3 balanced 51.5 81.9 87.0 performance 51.5 82.9 89.6 OpenBenchmarking.org Celsius, Fewer Is Better VVenC 1.9 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VVenC 1.9 Video Input: Bosphorus 1080p - Video Preset: Fast low_power balanced performance 0.1071 0.2142 0.3213 0.4284 0.5355 0.476 0.417 0.411 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3353 4963 balanced 1600 3638 4894 performance 1600 3670 4919 OpenBenchmarking.org Megahertz, More Is Better VVenC 1.9 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.2 30.0 balanced 2.6 31.5 36.3 performance 2.5 32.1 40.0 OpenBenchmarking.org Watts, Fewer Is Better VVenC 1.9 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1359 Joules per run. 2. balanced: Approximate power consumption of 1573 Joules per run. 3. performance: Approximate power consumption of 1583 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 1080p - Video Preset: Fast Original low_power balanced performance 4 8 12 16 20 Min: 12.04 / Avg: 12.06 / Max: 12.07 Min: 11.93 / Avg: 11.98 / Max: 12.01 Min: 13.07 / Avg: 13.11 / Max: 13.15 Min: 13.14 / Avg: 13.2 / Max: 13.31 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Original low_power balanced performance 4 8 12 16 20 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.17, N = 6 15.40 15.38 17.26 17.31 CPU Temp
Min Avg Max Original 47.3 69.4 74.8 low_power 45.4 67.1 72.1 balanced 47.9 76.5 82.6 performance 49.8 79.6 87.3 OpenBenchmarking.org Celsius, Fewer Is Better uvg266 0.4.1 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast low_power balanced performance 0.1384 0.2768 0.4152 0.5536 0.692 0.615 0.544 0.541 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3079 4642 balanced 1600 3462 4906 performance 1600 3413 4906 OpenBenchmarking.org Megahertz, More Is Better uvg266 0.4.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.0 30.3 balanced 2.7 31.7 36.3 performance 1.5 32.0 40.1 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1076 Joules per run. 2. balanced: Approximate power consumption of 1227 Joules per run. 3. performance: Approximate power consumption of 1236 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Original low_power balanced performance 4 8 12 16 20 Min: 15.31 / Avg: 15.4 / Max: 15.53 Min: 15.26 / Avg: 15.38 / Max: 15.53 Min: 17.1 / Avg: 17.26 / Max: 17.41 Min: 17.11 / Avg: 17.31 / Max: 18.15
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Thorough Original low_power balanced performance 2 4 6 8 10 SE +/- 0.0111, N = 3 SE +/- 0.0187, N = 3 SE +/- 0.0200, N = 3 SE +/- 0.0136, N = 3 5.7927 5.7787 6.2011 6.4306 1. (CXX) g++ options: -O3 -flto -pthread CPU Temp
Min Avg Max Original 48.3 70.8 84.1 low_power 47.0 68.3 83.8 balanced 49.3 77.8 87.6 performance 49.9 79.6 87.6 OpenBenchmarking.org Celsius, Fewer Is Better ASTC Encoder 4.0 CPU Temperature Monitor 20 40 60 80 100 MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.0 Preset: Thorough low_power balanced performance 0.0529 0.1058 0.1587 0.2116 0.2645 0.235 0.200 0.202 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3222 5001 balanced 1600 3519 5092 performance 1600 3572 4901 OpenBenchmarking.org Megahertz, More Is Better ASTC Encoder 4.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 24.6 30.3 balanced 2.5 31.0 36.3 performance 2.5 31.9 40.4 OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.0 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1254 Joules per run. 2. balanced: Approximate power consumption of 1425 Joules per run. 3. performance: Approximate power consumption of 1456 Joules per run. Result Confidence
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Thorough Original low_power balanced performance 3 6 9 12 15 Min: 5.77 / Avg: 5.79 / Max: 5.81 Min: 5.76 / Avg: 5.78 / Max: 5.82 Min: 6.17 / Avg: 6.2 / Max: 6.24 Min: 6.41 / Avg: 6.43 / Max: 6.46 1. (CXX) g++ options: -O3 -flto -pthread
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 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K Original low_power balanced performance 15 30 45 60 75 SE +/- 0.72, N = 15 SE +/- 0.09, N = 5 SE +/- 0.66, N = 15 SE +/- 0.84, N = 15 63.23 65.09 65.92 64.83 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 44.4 66.8 75.1 low_power 44.5 63.3 71.9 balanced 47.5 71.1 79.3 performance 46.1 72.3 80.9 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K low_power balanced performance 0.7396 1.4792 2.2188 2.9584 3.698 3.287 2.970 2.775 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1397 3882 5053 balanced 1397 3874 5088 performance 1398 4155 5069 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 19.8 27.1 balanced 2.6 22.2 30.8 performance 1.3 23.4 32.8 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 10 20 30 40 50 1. low_power: Approximate power consumption of 281 Joules per run. 2. balanced: Approximate power consumption of 314 Joules per run. 3. performance: Approximate power consumption of 335 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K Original low_power balanced performance 13 26 39 52 65 Min: 58.46 / Avg: 63.23 / Max: 65.36 Min: 64.85 / Avg: 65.09 / Max: 65.36 Min: 60.62 / Avg: 65.92 / Max: 67.41 Min: 60.68 / Avg: 64.83 / Max: 68.12 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 Original low_power balanced performance 70 140 210 280 350 SE +/- 1.76, N = 3 SE +/- 0.33, N = 3 SE +/- 0.88, N = 3 SE +/- 1.67, N = 3 299 290 287 288 1. chrome 116.0.5845.179 CPU Temp
Min Avg Max Original 41.5 54.9 69.5 low_power 41.1 52.6 65.1 balanced 43.0 52.1 62.4 performance 41.4 53.0 64.1 OpenBenchmarking.org Celsius, Fewer Is Better Selenium CPU Temperature Monitor 20 40 60 80 100 Runs Per Minute Per Watt
OpenBenchmarking.org Runs Per Minute Per Watt, More Is Better Selenium Benchmark: Speedometer - Browser: Google Chrome low_power balanced performance 7 14 21 28 35 31.22 30.62 30.95 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3244 5094 balanced 1600 3165 5012 performance 1437 3286 5062 OpenBenchmarking.org Megahertz, More Is Better Selenium CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.2 9.3 17.1 balanced 2.4 9.4 17.1 performance 2.2 9.3 17.5 OpenBenchmarking.org Watts, Fewer Is Better Selenium CPU Power Consumption Monitor 5 10 15 20 25 1. low_power: Approximate power consumption of 434 Joules per run. 2. balanced: Approximate power consumption of 437 Joules per run. 3. performance: Approximate power consumption of 437 Joules per run. Result Confidence
OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Google Chrome Original low_power balanced performance 50 100 150 200 250 Min: 296 / Avg: 299.33 / Max: 302 Min: 289 / Avg: 289.67 / Max: 290 Min: 286 / Avg: 287.33 / Max: 289 Min: 285 / Avg: 288.33 / Max: 290 1. chrome 116.0.5845.179
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast Original low_power balanced performance 5 10 15 20 25 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.11, N = 3 SE +/- 0.19, N = 5 16.33 16.33 18.26 18.30 CPU Temp
Min Avg Max Original 49.8 69.3 74.1 low_power 49.0 68.2 73.1 balanced 51.3 77.5 83.5 performance 52.6 80.0 86.8 OpenBenchmarking.org Celsius, Fewer Is Better uvg266 0.4.1 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast low_power balanced performance 0.1472 0.2944 0.4416 0.5888 0.736 0.654 0.582 0.575 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3118 4836 balanced 1600 3428 4910 performance 1600 3452 4990 OpenBenchmarking.org Megahertz, More Is Better uvg266 0.4.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 25.0 30.2 balanced 2.5 31.4 36.4 performance 2.5 31.8 40.2 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 1015 Joules per run. 2. balanced: Approximate power consumption of 1161 Joules per run. 3. performance: Approximate power consumption of 1178 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast Original low_power balanced performance 5 10 15 20 25 Min: 16.21 / Avg: 16.33 / Max: 16.49 Min: 16.26 / Avg: 16.33 / Max: 16.46 Min: 18.12 / Avg: 18.26 / Max: 18.48 Min: 18.08 / Avg: 18.3 / Max: 19.05
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 4K Original low_power balanced performance 20 40 60 80 100 SE +/- 1.11, N = 15 SE +/- 1.11, N = 15 SE +/- 1.27, N = 15 SE +/- 1.31, N = 15 84.77 84.78 88.98 90.83 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq CPU Temp
Min Avg Max Original 42.4 63.9 71.8 low_power 41.5 62.9 69.8 balanced 44.6 69.0 77.1 performance 44.6 71.8 81.4 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 4K low_power balanced performance 0.9781 1.9562 2.9343 3.9124 4.8905 4.347 3.923 3.693 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3467 5116 balanced 1600 3572 5114 performance 1600 3753 5079 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.3 19.5 30.5 balanced 2.2 22.6 36.2 performance 2.3 24.6 40.2 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 230 Joules per run. 2. balanced: Approximate power consumption of 262 Joules per run. 3. performance: Approximate power consumption of 279 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 4K Original low_power balanced performance 20 40 60 80 100 Min: 69.23 / Avg: 84.77 / Max: 86.75 Min: 69.24 / Avg: 84.78 / Max: 86.34 Min: 71.33 / Avg: 88.98 / Max: 90.82 Min: 72.62 / Avg: 90.83 / Max: 92.7 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 57 Original low_power balanced performance 120M 240M 360M 480M 600M SE +/- 4496384.97, N = 3 SE +/- 3629951.03, N = 3 SE +/- 2456399.80, N = 3 SE +/- 5787538.91, N = 6 521053333 526603333 580930000 578560000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid CPU Temp
Min Avg Max Original 50.1 70.8 75.4 low_power 48.9 68.6 73.5 balanced 52.1 79.1 85.1 performance 51.5 80.9 88.9 OpenBenchmarking.org Celsius, Fewer Is Better Liquid-DSP 1.6 CPU Temperature Monitor 20 40 60 80 100 samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 57 low_power balanced performance 5M 10M 15M 20M 25M 21249377.90 18688289.01 18205722.11 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1397 3168 3419 balanced 1600 3477 3643 performance 1600 3475 3775 OpenBenchmarking.org Megahertz, More Is Better Liquid-DSP 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1000 2000 3000 4000 5000 CPU Power Consumption
Min Avg Max low_power 2.4 24.8 30.3 balanced 2.8 31.1 36.3 performance 2.5 31.8 40.4 OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 843 Joules per run. 2. balanced: Approximate power consumption of 1067 Joules per run. 3. performance: Approximate power consumption of 1080 Joules per run. Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 57 Original low_power balanced performance 100M 200M 300M 400M 500M Min: 513780000 / Avg: 521053333.33 / Max: 529270000 Min: 522240000 / Avg: 526603333.33 / Max: 533810000 Min: 576850000 / Avg: 580930000 / Max: 585340000 Min: 566380000 / Avg: 578560000 / Max: 605590000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 32 Original low_power balanced performance 70M 140M 210M 280M 350M SE +/- 1646423.73, N = 3 SE +/- 1665422.87, N = 3 SE +/- 3442851.14, N = 5 SE +/- 3635433.26, N = 4 299116667 306910000 328252000 332545000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid CPU Temp
Min Avg Max Original 48.5 62.1 65.8 low_power 47.6 62.0 65.9 balanced 52.8 75.4 81.3 performance 54.6 80.9 86.9 OpenBenchmarking.org Celsius, Fewer Is Better Liquid-DSP 1.6 CPU Temperature Monitor 20 40 60 80 100 samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 32 low_power balanced performance 3M 6M 9M 12M 15M 14262266.85 11398407.16 10468258.07 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3847 4245 balanced 1600 4041 4429 performance 1600 4176 4807 OpenBenchmarking.org Megahertz, More Is Better Liquid-DSP 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.3 21.5 30.0 balanced 2.6 28.8 36.2 performance 2.6 31.8 39.9 OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 732 Joules per run. 2. balanced: Approximate power consumption of 979 Joules per run. 3. performance: Approximate power consumption of 1080 Joules per run. Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 32 Original low_power balanced performance 60M 120M 180M 240M 300M Min: 296210000 / Avg: 299116666.67 / Max: 301910000 Min: 303790000 / Avg: 306910000 / Max: 309480000 Min: 314750000 / Avg: 328252000 / Max: 333630000 Min: 322290000 / Avg: 332545000 / Max: 339460000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 57 Original low_power balanced performance 90M 180M 270M 360M 450M SE +/- 3466142.59, N = 3 SE +/- 4841575.80, N = 4 SE +/- 1819352.75, N = 3 SE +/- 4816545.05, N = 4 389193333 398217500 434333333 439027500 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid CPU Temp
Min Avg Max Original 46.9 65.7 70.5 low_power 46.3 66.5 71.0 balanced 51.6 79.2 84.9 performance 52.9 81.4 89.6 OpenBenchmarking.org Celsius, Fewer Is Better Liquid-DSP 1.6 CPU Temperature Monitor 20 40 60 80 100 samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 57 low_power balanced performance 4M 8M 12M 16M 20M 16810301.43 13798838.82 13722226.44 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3455 4086 balanced 1600 3721 4070 performance 1600 3743 4143 OpenBenchmarking.org Megahertz, More Is Better Liquid-DSP 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1100 2200 3300 4400 5500 CPU Power Consumption
Min Avg Max low_power 1.3 23.7 30.0 balanced 2.8 31.5 36.3 performance 2.7 32.0 40.4 OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 805 Joules per run. 2. balanced: Approximate power consumption of 1070 Joules per run. 3. performance: Approximate power consumption of 1088 Joules per run. Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 57 Original low_power balanced performance 80M 160M 240M 320M 400M Min: 382270000 / Avg: 389193333.33 / Max: 392960000 Min: 383770000 / Avg: 398217500 / Max: 404410000 Min: 431560000 / Avg: 434333333.33 / Max: 437760000 Min: 431210000 / Avg: 439027500 / Max: 453090000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 32 Original low_power balanced performance 120M 240M 360M 480M 600M SE +/- 1584911.49, N = 3 SE +/- 2732788.56, N = 3 SE +/- 4240331.49, N = 3 SE +/- 6635436.90, N = 4 504683333 509400000 548793333 552422500 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid CPU Temp
Min Avg Max Original 50.1 70.0 74.9 low_power 49.3 68.3 73.5 balanced 52.1 79.2 85.5 performance 51.5 80.7 88.6 OpenBenchmarking.org Celsius, Fewer Is Better Liquid-DSP 1.6 CPU Temperature Monitor 20 40 60 80 100 samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 32 low_power balanced performance 4M 8M 12M 16M 20M 20738972.69 17522027.37 17192720.93 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3494 3682 balanced 1600 3561 3885 performance 1600 3676 4026 OpenBenchmarking.org Megahertz, More Is Better Liquid-DSP 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1100 2200 3300 4400 5500 CPU Power Consumption
Min Avg Max low_power 2.4 24.5 30.2 balanced 2.6 31.3 36.3 performance 2.5 32.1 40.4 OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 842 Joules per run. 2. balanced: Approximate power consumption of 1065 Joules per run. 3. performance: Approximate power consumption of 1084 Joules per run. Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 32 Original low_power balanced performance 100M 200M 300M 400M 500M Min: 502150000 / Avg: 504683333.33 / Max: 507600000 Min: 506080000 / Avg: 509400000 / Max: 514820000 Min: 542730000 / Avg: 548793333.33 / Max: 556960000 Min: 543660000 / Avg: 552422500 / Max: 571990000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Original low_power balanced performance 5 10 15 20 25 SE +/- 0.10, N = 3 SE +/- 0.13, N = 3 SE +/- 0.09, N = 3 SE +/- 0.24, N = 4 19.82 19.80 22.06 22.19 CPU Temp
Min Avg Max Original 49.3 68.9 74.0 low_power 48.5 66.5 71.3 balanced 51.8 77.5 83.6 performance 52.4 80.2 87.1 OpenBenchmarking.org Celsius, Fewer Is Better uvg266 0.4.1 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast low_power balanced performance 0.1807 0.3614 0.5421 0.7228 0.9035 0.803 0.714 0.691 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3123 4989 balanced 1600 3361 4850 performance 1600 3514 4995 OpenBenchmarking.org Megahertz, More Is Better uvg266 0.4.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.4 24.7 30.3 balanced 2.6 30.9 36.3 performance 2.5 32.1 40.1 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 847 Joules per run. 2. balanced: Approximate power consumption of 958 Joules per run. 3. performance: Approximate power consumption of 996 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Original low_power balanced performance 5 10 15 20 25 Min: 19.71 / Avg: 19.82 / Max: 20.02 Min: 19.63 / Avg: 19.8 / Max: 20.05 Min: 21.88 / Avg: 22.06 / Max: 22.2 Min: 21.92 / Avg: 22.19 / Max: 22.92
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 512 Original low_power balanced performance 30M 60M 90M 120M 150M SE +/- 694318.21, N = 3 SE +/- 589500.73, N = 3 SE +/- 657072.63, N = 3 SE +/- 2157601.55, N = 3 142886667 142676667 156063333 157446667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid CPU Temp
Min Avg Max Original 49.8 68.9 73.5 low_power 49.1 67.9 72.4 balanced 51.8 78.3 84.3 performance 51.6 79.5 87.4 OpenBenchmarking.org Celsius, Fewer Is Better Liquid-DSP 1.6 CPU Temperature Monitor 20 40 60 80 100 samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 512 low_power balanced performance 1.2M 2.4M 3.6M 4.8M 6M 5786880.24 4952757.56 4852519.85 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3339 3543 balanced 1600 3450 3750 performance 1600 3632 3878 OpenBenchmarking.org Megahertz, More Is Better Liquid-DSP 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1000 2000 3000 4000 5000 CPU Power Consumption
Min Avg Max low_power 2.4 24.7 30.2 balanced 2.6 31.5 36.3 performance 2.5 32.4 40.3 OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 838 Joules per run. 2. balanced: Approximate power consumption of 1071 Joules per run. 3. performance: Approximate power consumption of 1103 Joules per run. Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 512 Original low_power balanced performance 30M 60M 90M 120M 150M Min: 142090000 / Avg: 142886666.67 / Max: 144270000 Min: 141990000 / Avg: 142676666.67 / Max: 143850000 Min: 154880000 / Avg: 156063333.33 / Max: 157150000 Min: 155180000 / Avg: 157446666.67 / Max: 161760000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 512 Original low_power balanced performance 30M 60M 90M 120M 150M SE +/- 41633.32, N = 3 SE +/- 805281.18, N = 3 SE +/- 557265.15, N = 3 SE +/- 438342.84, N = 3 108850000 108646667 117766667 119013333 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid CPU Temp
Min Avg Max Original 49.1 67.5 72.6 low_power 48.6 66.9 71.1 balanced 52.0 78.0 84.3 performance 52.1 79.0 86.6 OpenBenchmarking.org Celsius, Fewer Is Better Liquid-DSP 1.6 CPU Temperature Monitor 20 40 60 80 100 samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 512 low_power balanced performance 1000K 2000K 3000K 4000K 5000K 4461695.80 3754293.53 3688215.19 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3721 4026 balanced 1600 3919 4193 performance 1600 4005 4355 OpenBenchmarking.org Megahertz, More Is Better Liquid-DSP 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1100 2200 3300 4400 5500 CPU Power Consumption
Min Avg Max low_power 2.5 24.4 30.2 balanced 2.6 31.4 36.3 performance 2.5 32.3 40.3 OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 828 Joules per run. 2. balanced: Approximate power consumption of 1067 Joules per run. 3. performance: Approximate power consumption of 1108 Joules per run. Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 512 Original low_power balanced performance 20M 40M 60M 80M 100M Min: 108770000 / Avg: 108850000 / Max: 108910000 Min: 107190000 / Avg: 108646666.67 / Max: 109970000 Min: 116790000 / Avg: 117766666.67 / Max: 118720000 Min: 118570000 / Avg: 119013333.33 / Max: 119890000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Coremark This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second Original low_power balanced performance 80K 160K 240K 320K 400K SE +/- 2000.62, N = 3 SE +/- 1125.85, N = 3 SE +/- 871.46, N = 3 SE +/- 5265.63, N = 3 355143.95 354006.85 385964.61 390347.13 1. (CC) gcc options: -O2 -lrt" -lrt CPU Temp
Min Avg Max Original 44.4 70.5 80.1 low_power 45.0 69.6 80.1 balanced 45.5 78.3 86.5 performance 45.4 80.7 89.8 OpenBenchmarking.org Celsius, Fewer Is Better Coremark 1.0 CPU Temperature Monitor 20 40 60 80 100 Iterations/Sec Per Watt
OpenBenchmarking.org Iterations/Sec Per Watt, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second low_power balanced performance 3K 6K 9K 12K 15K 14748.76 12928.50 12255.63 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3529 5016 balanced 1600 3780 5015 performance 1600 3848 4990 OpenBenchmarking.org Megahertz, More Is Better Coremark 1.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 24.0 30.2 balanced 2.4 29.9 36.6 performance 2.4 31.9 40.4 OpenBenchmarking.org Watts, Fewer Is Better Coremark 1.0 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 808 Joules per run. 2. balanced: Approximate power consumption of 945 Joules per run. 3. performance: Approximate power consumption of 998 Joules per run. Result Confidence
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second Original low_power balanced performance 70K 140K 210K 280K 350K Min: 352552.33 / Avg: 355143.95 / Max: 359079.86 Min: 352578.23 / Avg: 354006.85 / Max: 356228.43 Min: 384892.95 / Avg: 385964.61 / Max: 387690.82 Min: 384276.68 / Avg: 390347.13 / Max: 400835.07 1. (CC) gcc options: -O2 -lrt" -lrt
PHPBench PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite Original low_power balanced performance 200K 400K 600K 800K 1000K SE +/- 8661.15, N = 10 SE +/- 6003.27, N = 3 SE +/- 5093.12, N = 3 SE +/- 13614.93, N = 3 1120443 1147675 1102495 1125118 CPU Temp
Min Avg Max Original 48.9 80.0 86.3 low_power 48.3 79.0 86.3 balanced 49.8 81.9 89.8 performance 49.4 81.4 89.3 OpenBenchmarking.org Celsius, Fewer Is Better PHPBench 0.8.1 CPU Temperature Monitor 20 40 60 80 100 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better PHPBench 0.8.1 PHP Benchmark Suite low_power balanced performance 20K 40K 60K 80K 100K 87203.23 83776.22 82949.78 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 4397 5065 balanced 1600 4333 5065 performance 1600 4545 5070 OpenBenchmarking.org Megahertz, More Is Better PHPBench 0.8.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 13.2 16.9 balanced 2.4 13.2 17.3 performance 2.3 13.6 17.2 OpenBenchmarking.org Watts, Fewer Is Better PHPBench 0.8.1 CPU Power Consumption Monitor 5 10 15 20 25 1. low_power: Approximate power consumption of 285 Joules per run. 2. balanced: Approximate power consumption of 290 Joules per run. 3. performance: Approximate power consumption of 294 Joules per run. Result Confidence
OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite Original low_power balanced performance 200K 400K 600K 800K 1000K Min: 1047829 / Avg: 1120443 / Max: 1148943 Min: 1135669 / Avg: 1147675 / Max: 1153777 Min: 1093913 / Avg: 1102495 / Max: 1111538 Min: 1102999 / Avg: 1125117.67 / Max: 1149931
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Fast Original low_power balanced performance 30 60 90 120 150 SE +/- 0.43, N = 3 SE +/- 0.10, N = 3 SE +/- 0.27, N = 3 SE +/- 0.14, N = 3 129.46 129.89 138.71 143.16 1. (CXX) g++ options: -O3 -flto -pthread CPU Temp
Min Avg Max Original 47.5 69.3 83.1 low_power 46.9 67.6 82.1 balanced 49.4 76.7 88.1 performance 49.0 78.6 87.6 OpenBenchmarking.org Celsius, Fewer Is Better ASTC Encoder 4.0 CPU Temperature Monitor 20 40 60 80 100 MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.0 Preset: Fast low_power balanced performance 1.2101 2.4202 3.6303 4.8404 6.0505 5.378 4.598 4.513 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3329 5062 balanced 1600 3508 5026 performance 1600 3740 4990 OpenBenchmarking.org Megahertz, More Is Better ASTC Encoder 4.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 24.2 30.3 balanced 2.5 30.2 36.3 performance 2.4 31.7 40.4 OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.0 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 805 Joules per run. 2. balanced: Approximate power consumption of 935 Joules per run. 3. performance: Approximate power consumption of 973 Joules per run. Result Confidence
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Fast Original low_power balanced performance 30 60 90 120 150 Min: 128.59 / Avg: 129.46 / Max: 129.96 Min: 129.7 / Avg: 129.89 / Max: 130.04 Min: 138.41 / Avg: 138.71 / Max: 139.25 Min: 142.99 / Avg: 143.16 / Max: 143.43 1. (CXX) g++ options: -O3 -flto -pthread
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 Original low_power balanced performance 14K 28K 42K 56K 70K SE +/- 223.36, N = 3 SE +/- 49.42, N = 3 SE +/- 103.59, N = 3 SE +/- 688.48, N = 3 57250 60566 65997 67194 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC CPU Temp
Min Avg Max Original 48.6 65.1 70.0 low_power 49.5 67.5 71.9 balanced 54.0 79.8 87.8 performance 54.0 82.2 91.6 OpenBenchmarking.org Celsius, Fewer Is Better 7-Zip Compression 22.01 CPU Temperature Monitor 20 40 60 80 100 MIPS Per Watt
OpenBenchmarking.org MIPS Per Watt, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating low_power balanced performance 600 1200 1800 2400 3000 2568.27 2164.66 2118.50 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3494 5066 balanced 1600 3834 5065 performance 1600 3817 5049 OpenBenchmarking.org Megahertz, More Is Better 7-Zip Compression 22.01 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.9 23.6 30.1 balanced 2.8 30.5 36.3 performance 2.6 31.7 40.4 OpenBenchmarking.org Watts, Fewer Is Better 7-Zip Compression 22.01 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 755 Joules per run. 2. balanced: Approximate power consumption of 915 Joules per run. 3. performance: Approximate power consumption of 952 Joules per run. Result Confidence
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating Original low_power balanced performance 12K 24K 36K 48K 60K Min: 56804 / Avg: 57249.67 / Max: 57499 Min: 60470 / Avg: 60565.67 / Max: 60635 Min: 65832 / Avg: 65997 / Max: 66188 Min: 66235 / Avg: 67193.67 / Max: 68529 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
VVenC VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 1080p - Video Preset: Faster Original low_power balanced performance 6 12 18 24 30 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.11, N = 3 25.00 25.02 26.49 26.93 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects CPU Temp
Min Avg Max Original 49.4 71.9 77.9 low_power 48.5 69.7 75.5 balanced 51.4 79.1 86.0 performance 51.8 82.4 89.9 OpenBenchmarking.org Celsius, Fewer Is Better VVenC 1.9 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VVenC 1.9 Video Input: Bosphorus 1080p - Video Preset: Faster low_power balanced performance 0.2336 0.4672 0.7008 0.9344 1.168 1.038 0.900 0.856 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3503 5013 balanced 1600 3685 4971 performance 1600 3801 4857 OpenBenchmarking.org Megahertz, More Is Better VVenC 1.9 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 24.1 29.9 balanced 1.4 29.4 35.4 performance 2.2 31.5 39.2 OpenBenchmarking.org Watts, Fewer Is Better VVenC 1.9 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 683 Joules per run. 2. balanced: Approximate power consumption of 795 Joules per run. 3. performance: Approximate power consumption of 839 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 1080p - Video Preset: Faster Original low_power balanced performance 6 12 18 24 30 Min: 24.89 / Avg: 25 / Max: 25.2 Min: 24.95 / Avg: 25.02 / Max: 25.12 Min: 26.45 / Avg: 26.49 / Max: 26.52 Min: 26.73 / Avg: 26.93 / Max: 27.09 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects
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 3.7 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 40 80 120 160 200 SE +/- 4.43, N = 15 SE +/- 6.37, N = 15 SE +/- 5.90, N = 15 SE +/- 3.54, N = 15 156.02 147.86 158.39 165.59 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 48.0 61.0 73.1 low_power 47.3 60.4 70.6 balanced 50.1 64.6 77.1 performance 50.1 67.8 81.5 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p low_power balanced performance 2 4 6 8 10 8.751 8.160 7.934 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3737 5006 balanced 1397 3620 5115 performance 1397 3788 4956 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 16.9 30.1 balanced 2.5 19.4 36.1 performance 2.5 20.9 40.2 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 144 Joules per run. 2. balanced: Approximate power consumption of 159 Joules per run. 3. performance: Approximate power consumption of 167 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 30 60 90 120 150 Min: 117.76 / Avg: 156.02 / Max: 168.89 Min: 110.17 / Avg: 147.86 / Max: 168.52 Min: 120.97 / Avg: 158.39 / Max: 173.56 Min: 119.05 / Avg: 165.59 / Max: 173.33 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K Original low_power balanced performance 15 30 45 60 75 SE +/- 0.10, N = 5 SE +/- 0.51, N = 9 SE +/- 0.11, N = 5 SE +/- 0.13, N = 5 63.94 63.16 65.95 67.01 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 45.3 63.9 72.4 low_power 45.9 65.5 72.4 balanced 45.9 67.3 76.9 performance 46.0 69.6 79.8 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K low_power balanced performance 0.7081 1.4162 2.1243 2.8324 3.5405 3.147 2.966 2.799 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1397 3992 5070 balanced 1600 3911 5070 performance 1600 3996 5053 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 20.1 27.7 balanced 2.4 22.2 31.4 performance 2.3 23.9 33.9 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 10 20 30 40 50 1. low_power: Approximate power consumption of 294 Joules per run. 2. balanced: Approximate power consumption of 316 Joules per run. 3. performance: Approximate power consumption of 335 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K Original low_power balanced performance 13 26 39 52 65 Min: 63.58 / Avg: 63.94 / Max: 64.19 Min: 59.26 / Avg: 63.16 / Max: 64.04 Min: 65.72 / Avg: 65.95 / Max: 66.24 Min: 66.75 / Avg: 67.01 / Max: 67.4 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 40 80 120 160 200 SE +/- 5.08, N = 15 SE +/- 3.75, N = 15 SE +/- 4.39, N = 15 SE +/- 4.79, N = 15 164.28 148.17 175.76 160.97 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 47.6 61.2 71.5 low_power 47.0 61.6 70.6 balanced 49.6 64.1 74.6 performance 48.5 67.2 78.6 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p low_power balanced performance 3 6 9 12 15 8.877 9.610 7.887 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3832 4970 balanced 1397 3518 4978 performance 1600 3838 4995 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 16.7 30.1 balanced 1.3 18.3 36.1 performance 2.4 20.4 40.0 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 141 Joules per run. 2. balanced: Approximate power consumption of 143 Joules per run. 3. performance: Approximate power consumption of 166 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 30 60 90 120 150 Min: 141.24 / Avg: 164.28 / Max: 191.22 Min: 134.69 / Avg: 148.17 / Max: 172.84 Min: 144.75 / Avg: 175.76 / Max: 198.54 Min: 140.99 / Avg: 160.97 / Max: 186.85 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
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 Original low_power balanced performance 10 20 30 40 50 SE +/- 0.14, N = 4 SE +/- 0.15, N = 4 SE +/- 0.38, N = 4 SE +/- 0.25, N = 4 40.65 40.47 42.96 43.97 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm CPU Temp
Min Avg Max Original 45.1 66.5 73.0 low_power 45.1 64.7 70.6 balanced 47.6 71.4 80.0 performance 48.3 75.6 86.1 OpenBenchmarking.org Celsius, Fewer Is Better SVT-VP9 0.3 CPU Temperature Monitor 20 40 60 80 100 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 low_power balanced performance 0.3942 0.7884 1.1826 1.5768 1.971 1.752 1.541 1.438 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3370 5050 balanced 1600 3555 5102 performance 1600 3737 5097 OpenBenchmarking.org Megahertz, More Is Better SVT-VP9 0.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.3 23.1 30.3 balanced 2.5 27.9 36.3 performance 2.5 30.6 40.2 OpenBenchmarking.org Watts, Fewer Is Better SVT-VP9 0.3 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 456 Joules per run. 2. balanced: Approximate power consumption of 523 Joules per run. 3. performance: Approximate power consumption of 566 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 4K Original low_power balanced performance 9 18 27 36 45 Min: 40.25 / Avg: 40.65 / Max: 40.88 Min: 40.04 / Avg: 40.47 / Max: 40.69 Min: 41.82 / Avg: 42.96 / Max: 43.47 Min: 43.23 / Avg: 43.97 / Max: 44.34 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: CPU Original low_power balanced performance 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 7.80 8.11 8.71 8.99 MIN: 7.34 / MAX: 8.06 MIN: 7.52 / MAX: 8.25 MIN: 8.26 / MAX: 8.89 MIN: 8.35 / MAX: 9.19 CPU Temp
Min Avg Max Original 49.6 66.1 71.8 low_power 48.1 66.0 71.4 balanced 52.5 75.8 83.4 performance 53.6 80.3 89.1 OpenBenchmarking.org Celsius, Fewer Is Better LuxCoreRender 2.6 CPU Temperature Monitor 20 40 60 80 100 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: CPU low_power balanced performance 0.0767 0.1534 0.2301 0.3068 0.3835 0.341 0.303 0.288 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3366 4696 balanced 1600 3363 4381 performance 1600 3584 4475 OpenBenchmarking.org Megahertz, More Is Better LuxCoreRender 2.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.4 23.7 30.2 balanced 2.6 28.7 36.3 performance 2.6 31.2 40.4 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 578 Joules per run. 2. balanced: Approximate power consumption of 680 Joules per run. 3. performance: Approximate power consumption of 717 Joules per run. Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: CPU Original low_power balanced performance 3 6 9 12 15 Min: 7.75 / Avg: 7.8 / Max: 7.84 Min: 8.01 / Avg: 8.11 / Max: 8.18 Min: 8.67 / Avg: 8.71 / Max: 8.77 Min: 8.89 / Avg: 8.99 / Max: 9.15
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 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 40 80 120 160 200 SE +/- 3.32, N = 15 SE +/- 3.99, N = 15 SE +/- 3.55, N = 15 SE +/- 4.31, N = 15 173.62 166.24 176.35 165.13 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 48.6 60.8 69.1 low_power 47.8 60.0 69.9 balanced 50.0 63.8 74.4 performance 50.9 67.8 79.8 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 1080p low_power balanced performance 3 6 9 12 15 10.309 9.632 8.105 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3789 5018 balanced 1397 3524 4966 performance 1398 3882 5011 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.3 16.1 30.1 balanced 2.5 18.3 36.1 performance 1.4 20.4 40.1 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 130 Joules per run. 2. balanced: Approximate power consumption of 142 Joules per run. 3. performance: Approximate power consumption of 164 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 30 60 90 120 150 Min: 138.99 / Avg: 173.62 / Max: 191.21 Min: 138.02 / Avg: 166.24 / Max: 199.07 Min: 142.81 / Avg: 176.35 / Max: 196.56 Min: 138.86 / Avg: 165.13 / Max: 188.89 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p Original low_power balanced performance 10 20 30 40 50 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.10, N = 3 39.83 39.79 42.07 43.35 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 48.6 70.2 81.3 low_power 47.4 68.3 79.6 balanced 50.5 76.2 86.6 performance 49.8 79.7 91.9 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p low_power balanced performance 0.3715 0.743 1.1145 1.486 1.8575 1.651 1.435 1.375 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3584 5114 balanced 1600 3747 5015 performance 1600 3940 5012 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 24.1 30.0 balanced 2.7 29.3 35.7 performance 2.4 31.5 39.4 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 579 Joules per run. 2. balanced: Approximate power consumption of 674 Joules per run. 3. performance: Approximate power consumption of 715 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p Original low_power balanced performance 9 18 27 36 45 Min: 39.65 / Avg: 39.83 / Max: 40.02 Min: 39.69 / Avg: 39.79 / Max: 39.9 Min: 41.9 / Avg: 42.07 / Max: 42.17 Min: 43.16 / Avg: 43.35 / Max: 43.45 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Original low_power balanced performance 8 16 24 32 40 SE +/- 0.29, N = 3 SE +/- 0.38, N = 3 SE +/- 0.04, N = 3 SE +/- 0.22, N = 4 32.92 32.74 35.92 36.81 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt CPU Temp
Min Avg Max Original 49.5 68.1 73.8 low_power 48.6 65.5 70.9 balanced 51.1 74.6 82.9 performance 51.9 78.5 86.3 OpenBenchmarking.org Celsius, Fewer Is Better Kvazaar 2.2 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Ultra Fast low_power balanced performance 0.3098 0.6196 0.9294 1.2392 1.549 1.377 1.221 1.169 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3151 4994 balanced 1600 3385 4754 performance 1600 3595 4925 OpenBenchmarking.org Megahertz, More Is Better Kvazaar 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 23.8 30.2 balanced 2.6 29.4 36.3 performance 2.5 31.5 40.1 OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 539 Joules per run. 2. balanced: Approximate power consumption of 608 Joules per run. 3. performance: Approximate power consumption of 638 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Original low_power balanced performance 8 16 24 32 40 Min: 32.6 / Avg: 32.92 / Max: 33.5 Min: 32.25 / Avg: 32.74 / Max: 33.48 Min: 35.85 / Avg: 35.92 / Max: 35.98 Min: 36.37 / Avg: 36.81 / Max: 37.25 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 1080p Original low_power balanced performance 3 6 9 12 15 SE +/- 0.053, N = 3 SE +/- 0.025, N = 3 SE +/- 0.031, N = 3 SE +/- 0.025, N = 4 9.234 9.156 9.840 10.109 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq CPU Temp
Min Avg Max Original 49.3 70.9 77.6 low_power 48.3 69.1 75.8 balanced 50.6 76.4 86.0 performance 51.6 81.6 91.3 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 1080p low_power balanced performance 0.0878 0.1756 0.2634 0.3512 0.439 0.390 0.356 0.336 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3541 5091 balanced 1600 3644 5015 performance 1600 3921 4997 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 23.5 30.3 balanced 2.5 27.6 36.3 performance 2.4 30.1 40.4 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 516 Joules per run. 2. balanced: Approximate power consumption of 580 Joules per run. 3. performance: Approximate power consumption of 617 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 1080p Original low_power balanced performance 3 6 9 12 15 Min: 9.17 / Avg: 9.23 / Max: 9.34 Min: 9.12 / Avg: 9.16 / Max: 9.2 Min: 9.78 / Avg: 9.84 / Max: 9.89 Min: 10.04 / Avg: 10.11 / Max: 10.14 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
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 3.7 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K Original low_power balanced performance 13 26 39 52 65 SE +/- 0.08, N = 4 SE +/- 0.13, N = 4 SE +/- 0.51, N = 6 SE +/- 0.07, N = 4 53.42 53.18 54.87 56.36 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 45.4 64.4 72.1 low_power 44.4 63.9 72.8 balanced 47.3 70.5 79.9 performance 47.1 71.3 81.6 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K low_power balanced performance 0.5618 1.1236 1.6854 2.2472 2.809 2.497 2.260 2.169 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3993 5060 balanced 1600 3902 5034 performance 1397 4071 5043 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.9 21.3 28.7 balanced 2.5 24.3 33.2 performance 2.1 26.0 35.9 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 10 20 30 40 50 1. low_power: Approximate power consumption of 351 Joules per run. 2. balanced: Approximate power consumption of 392 Joules per run. 3. performance: Approximate power consumption of 409 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K Original low_power balanced performance 11 22 33 44 55 Min: 53.22 / Avg: 53.42 / Max: 53.59 Min: 52.81 / Avg: 53.18 / Max: 53.44 Min: 52.36 / Avg: 54.87 / Max: 55.69 Min: 56.22 / Avg: 56.36 / Max: 56.49 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 40 80 120 160 200 SE +/- 3.62, N = 15 SE +/- 3.24, N = 12 SE +/- 4.86, N = 15 SE +/- 3.75, N = 15 175.93 166.83 176.60 176.61 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 47.9 60.5 70.3 low_power 47.1 59.7 68.3 balanced 50.1 65.3 77.1 performance 48.9 66.3 78.4 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p low_power balanced performance 3 6 9 12 15 10.409 9.556 8.848 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1397 3734 4936 balanced 1600 3531 5011 performance 1600 3940 4989 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.2 16.0 30.0 balanced 2.5 18.5 36.1 performance 2.1 20.0 40.0 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 128 Joules per run. 2. balanced: Approximate power consumption of 144 Joules per run. 3. performance: Approximate power consumption of 156 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 30 60 90 120 150 Min: 140.69 / Avg: 175.93 / Max: 196.88 Min: 140.45 / Avg: 166.83 / Max: 185.1 Min: 139.46 / Avg: 176.6 / Max: 195.47 Min: 146.93 / Avg: 176.61 / Max: 203.01 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Medium Original low_power balanced performance 11 22 33 44 55 SE +/- 0.15, N = 4 SE +/- 0.16, N = 4 SE +/- 0.06, N = 4 SE +/- 0.04, N = 4 42.20 42.18 45.60 47.11 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt CPU Temp
Min Avg Max Original 48.4 67.7 74.5 low_power 48.1 66.1 72.3 balanced 50.5 74.3 83.0 performance 50.5 78.7 89.3 OpenBenchmarking.org Celsius, Fewer Is Better Kvazaar 2.2 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Medium low_power balanced performance 0.4095 0.819 1.2285 1.638 2.0475 1.820 1.634 1.522 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3324 4615 balanced 1600 3481 4743 performance 1600 3655 4888 OpenBenchmarking.org Megahertz, More Is Better Kvazaar 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.3 23.2 30.3 balanced 2.6 27.9 36.3 performance 2.5 31.0 40.3 OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 423 Joules per run. 2. balanced: Approximate power consumption of 481 Joules per run. 3. performance: Approximate power consumption of 519 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Medium Original low_power balanced performance 10 20 30 40 50 Min: 41.95 / Avg: 42.2 / Max: 42.63 Min: 41.95 / Avg: 42.18 / Max: 42.63 Min: 45.49 / Avg: 45.6 / Max: 45.72 Min: 47.02 / Avg: 47.11 / Max: 47.2 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K Original low_power balanced performance 8 16 24 32 40 SE +/- 0.24, N = 3 SE +/- 0.28, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 33.19 33.07 35.54 36.52 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq CPU Temp
Min Avg Max Original 47.5 67.2 73.4 low_power 47.1 65.6 71.1 balanced 49.4 72.9 80.9 performance 50.1 76.3 86.1 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K low_power balanced performance 0.3166 0.6332 0.9498 1.2664 1.583 1.407 1.242 1.170 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3332 5016 balanced 1600 3584 5086 performance 1600 3825 5040 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 23.5 30.9 balanced 2.6 28.6 36.7 performance 2.6 31.2 40.5 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 541 Joules per run. 2. balanced: Approximate power consumption of 630 Joules per run. 3. performance: Approximate power consumption of 666 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K Original low_power balanced performance 8 16 24 32 40 Min: 32.93 / Avg: 33.19 / Max: 33.68 Min: 32.78 / Avg: 33.07 / Max: 33.64 Min: 35.48 / Avg: 35.54 / Max: 35.65 Min: 36.44 / Avg: 36.52 / Max: 36.65 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
libavif avifenc This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6, Lossless Original low_power balanced performance 3 6 9 12 15 SE +/- 0.04, N = 5 SE +/- 0.04, N = 5 SE +/- 0.02, N = 5 SE +/- 0.03, N = 5 10.95 10.94 10.38 10.11 1. (CXX) g++ options: -O3 -fPIC -lm CPU Temp
Min Avg Max Original 47.3 68.9 77.1 low_power 46.1 67.4 76.6 balanced 47.5 72.8 83.3 performance 49.0 75.2 88.0 OpenBenchmarking.org Celsius, Fewer Is Better libavif avifenc 1.0 CPU Temperature Monitor 20 40 60 80 100 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3602 5022 balanced 1600 3476 5097 performance 2200 3829 4999 OpenBenchmarking.org Megahertz, More Is Better libavif avifenc 1.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 20.7 30.2 balanced 2.5 23.6 36.2 performance 2.4 25.6 40.3 OpenBenchmarking.org Watts, Fewer Is Better libavif avifenc 1.0 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 311 Joules per run. 2. balanced: Approximate power consumption of 339 Joules per run. 3. performance: Approximate power consumption of 363 Joules per run. Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6, Lossless Original low_power balanced performance 3 6 9 12 15 Min: 10.84 / Avg: 10.95 / Max: 11.05 Min: 10.8 / Avg: 10.94 / Max: 11 Min: 10.32 / Avg: 10.38 / Max: 10.41 Min: 10.04 / Avg: 10.11 / Max: 10.19 1. (CXX) g++ options: -O3 -fPIC -lm
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Medium Original low_power balanced performance 12 24 36 48 60 SE +/- 0.07, N = 4 SE +/- 0.04, N = 4 SE +/- 0.03, N = 4 SE +/- 0.03, N = 4 45.97 45.91 49.31 51.15 1. (CXX) g++ options: -O3 -flto -pthread CPU Temp
Min Avg Max Original 49.4 68.9 83.4 low_power 48.3 67.2 82.3 balanced 51.5 73.1 86.3 performance 50.6 75.9 87.4 OpenBenchmarking.org Celsius, Fewer Is Better ASTC Encoder 4.0 CPU Temperature Monitor 20 40 60 80 100 MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.0 Preset: Medium low_power balanced performance 0.4741 0.9482 1.4223 1.8964 2.3705 2.107 1.950 1.870 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3430 5080 balanced 1600 3505 5062 performance 1600 3649 4991 OpenBenchmarking.org Megahertz, More Is Better ASTC Encoder 4.0 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 1.3 21.8 30.2 balanced 2.6 25.3 36.3 performance 2.5 27.3 40.3 OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.0 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 381 Joules per run. 2. balanced: Approximate power consumption of 424 Joules per run. 3. performance: Approximate power consumption of 451 Joules per run. Result Confidence
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Medium Original low_power balanced performance 10 20 30 40 50 Min: 45.8 / Avg: 45.97 / Max: 46.12 Min: 45.83 / Avg: 45.91 / Max: 46.02 Min: 49.24 / Avg: 49.31 / Max: 49.36 Min: 51.11 / Avg: 51.15 / Max: 51.23 1. (CXX) g++ options: -O3 -flto -pthread
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.
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 Original low_power balanced performance 12 24 36 48 60 SE +/- 0.10, N = 4 SE +/- 0.08, N = 4 SE +/- 0.05, N = 4 SE +/- 0.10, N = 4 48.99 48.90 50.78 51.55 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm CPU Temp
Min Avg Max Original 45.1 65.5 73.3 low_power 44.5 63.4 70.3 balanced 45.9 69.8 80.4 performance 46.6 73.7 84.5 OpenBenchmarking.org Celsius, Fewer Is Better SVT-VP9 0.3 CPU Temperature Monitor 20 40 60 80 100 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 low_power balanced performance 0.4894 0.9788 1.4682 1.9576 2.447 2.175 1.906 1.763 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3570 5016 balanced 1600 3637 4985 performance 1600 3731 5015 OpenBenchmarking.org Megahertz, More Is Better SVT-VP9 0.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 22.5 30.2 balanced 2.4 26.6 36.3 performance 2.4 29.2 39.9 OpenBenchmarking.org Watts, Fewer Is Better SVT-VP9 0.3 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 388 Joules per run. 2. balanced: Approximate power consumption of 446 Joules per run. 3. performance: Approximate power consumption of 483 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 4K Original low_power balanced performance 10 20 30 40 50 Min: 48.8 / Avg: 48.99 / Max: 49.23 Min: 48.7 / Avg: 48.9 / Max: 49.08 Min: 50.67 / Avg: 50.78 / Max: 50.91 Min: 51.34 / Avg: 51.55 / Max: 51.74 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
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 3.7 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 40 80 120 160 200 SE +/- 3.85, N = 12 SE +/- 1.36, N = 15 SE +/- 4.98, N = 15 SE +/- 1.22, N = 9 167.28 172.71 167.66 179.07 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm CPU Temp
Min Avg Max Original 47.3 59.9 71.0 low_power 47.1 58.8 66.6 balanced 49.3 64.6 76.6 performance 49.0 66.1 78.0 OpenBenchmarking.org Celsius, Fewer Is Better AOM AV1 3.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 3.7 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p low_power balanced performance 3 6 9 12 15 10.911 8.909 8.916 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3694 4979 balanced 1397 3579 5032 performance 1600 3660 4900 OpenBenchmarking.org Megahertz, More Is Better AOM AV1 3.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 15.8 30.1 balanced 2.5 18.8 36.1 performance 2.4 20.1 40.1 OpenBenchmarking.org Watts, Fewer Is Better AOM AV1 3.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 123 Joules per run. 2. balanced: Approximate power consumption of 151 Joules per run. 3. performance: Approximate power consumption of 156 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p Original low_power balanced performance 30 60 90 120 150 Min: 129.04 / Avg: 167.28 / Max: 178.64 Min: 156.85 / Avg: 172.71 / Max: 179.72 Min: 132.05 / Avg: 167.66 / Max: 182.25 Min: 170.23 / Avg: 179.07 / Max: 182.45 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
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: PSNR/SSIM Optimized - Input: Bosphorus 4K Original low_power balanced performance 12 24 36 48 60 SE +/- 0.05, N = 4 SE +/- 0.11, N = 4 SE +/- 0.08, N = 4 SE +/- 0.13, N = 4 52.24 52.14 54.28 55.17 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm CPU Temp
Min Avg Max Original 45.4 64.7 72.9 low_power 44.3 62.9 70.9 balanced 46.1 69.3 80.3 performance 47.0 73.2 84.0 OpenBenchmarking.org Celsius, Fewer Is Better SVT-VP9 0.3 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K low_power balanced performance 0.5265 1.053 1.5795 2.106 2.6325 2.340 2.061 1.902 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3514 5041 balanced 1600 3643 5065 performance 1600 3943 5090 OpenBenchmarking.org Megahertz, More Is Better SVT-VP9 0.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 22.3 30.3 balanced 1.4 26.3 36.2 performance 2.4 29.0 40.2 OpenBenchmarking.org Watts, Fewer Is Better SVT-VP9 0.3 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 368 Joules per run. 2. balanced: Approximate power consumption of 421 Joules per run. 3. performance: Approximate power consumption of 457 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K Original low_power balanced performance 11 22 33 44 55 Min: 52.14 / Avg: 52.24 / Max: 52.36 Min: 51.88 / Avg: 52.14 / Max: 52.34 Min: 54.18 / Avg: 54.28 / Max: 54.5 Min: 54.91 / Avg: 55.17 / Max: 55.53 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast Original low_power balanced performance 20 40 60 80 100 SE +/- 0.29, N = 5 SE +/- 0.06, N = 6 SE +/- 0.12, N = 6 SE +/- 0.10, N = 6 70.37 72.17 78.08 80.83 CPU Temp
Min Avg Max Original 45.1 62.2 70.4 low_power 49.1 63.3 69.5 balanced 52.9 70.4 78.4 performance 53.3 74.5 83.9 OpenBenchmarking.org Celsius, Fewer Is Better uvg266 0.4.1 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast low_power balanced performance 0.7805 1.561 2.3415 3.122 3.9025 3.469 3.207 3.021 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1397 3212 5030 balanced 1600 3289 4735 performance 1600 3421 4531 OpenBenchmarking.org Megahertz, More Is Better uvg266 0.4.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 20.8 30.1 balanced 2.7 24.3 36.3 performance 2.6 26.8 40.2 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 260 Joules per run. 2. balanced: Approximate power consumption of 288 Joules per run. 3. performance: Approximate power consumption of 308 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast Original low_power balanced performance 15 30 45 60 75 Min: 69.22 / Avg: 70.37 / Max: 70.89 Min: 71.95 / Avg: 72.17 / Max: 72.36 Min: 77.5 / Avg: 78.08 / Max: 78.24 Min: 80.46 / Avg: 80.83 / Max: 81.19
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 1080p Original low_power balanced performance 20 40 60 80 100 SE +/- 0.16, N = 6 SE +/- 0.11, N = 6 SE +/- 0.10, N = 6 SE +/- 0.11, N = 6 81.18 80.91 86.13 88.56 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq CPU Temp
Min Avg Max Original 47.5 64.2 72.1 low_power 46.3 62.3 70.0 balanced 48.9 67.9 77.1 performance 49.9 71.4 82.3 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 1080p low_power balanced performance 0.9151 1.8302 2.7453 3.6604 4.5755 4.067 3.727 3.541 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3355 5116 balanced 1600 3420 5096 performance 1600 3620 5028 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 19.9 30.1 balanced 2.5 23.1 35.9 performance 2.4 25.0 40.0 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 235 Joules per run. 2. balanced: Approximate power consumption of 262 Joules per run. 3. performance: Approximate power consumption of 279 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 1080p Original low_power balanced performance 20 40 60 80 100 Min: 80.43 / Avg: 81.18 / Max: 81.49 Min: 80.48 / Avg: 80.91 / Max: 81.23 Min: 85.75 / Avg: 86.13 / Max: 86.51 Min: 88.18 / Avg: 88.56 / Max: 88.94 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 4K Original low_power balanced performance 20 40 60 80 100 SE +/- 0.16, N = 6 SE +/- 0.08, N = 6 SE +/- 0.07, N = 6 SE +/- 0.16, N = 6 86.17 86.09 90.47 92.00 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq CPU Temp
Min Avg Max Original 42.8 61.7 70.6 low_power 41.9 59.7 67.8 balanced 44.0 65.5 76.3 performance 44.6 68.5 79.9 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 4K low_power balanced performance 0.9988 1.9976 2.9964 3.9952 4.994 4.439 3.996 3.743 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3473 5116 balanced 1600 3391 5115 performance 1600 3741 5115 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.1 19.4 30.0 balanced 1.3 22.6 36.0 performance 2.3 24.6 40.3 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 226 Joules per run. 2. balanced: Approximate power consumption of 257 Joules per run. 3. performance: Approximate power consumption of 274 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 4K Original low_power balanced performance 20 40 60 80 100 Min: 85.72 / Avg: 86.17 / Max: 86.63 Min: 85.86 / Avg: 86.09 / Max: 86.41 Min: 90.27 / Avg: 90.47 / Max: 90.67 Min: 91.48 / Avg: 92 / Max: 92.49 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Super Fast Original low_power balanced performance 20 40 60 80 100 SE +/- 0.11, N = 6 SE +/- 0.15, N = 6 SE +/- 0.04, N = 6 SE +/- 0.08, N = 6 77.53 77.72 84.31 87.42 CPU Temp
Min Avg Max Original 46.6 63.6 71.6 low_power 46.1 61.8 69.3 balanced 49.3 68.2 77.4 performance 50.6 72.2 82.9 OpenBenchmarking.org Celsius, Fewer Is Better uvg266 0.4.1 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Super Fast low_power balanced performance 0.8647 1.7294 2.5941 3.4588 4.3235 3.843 3.529 3.375 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3096 4891 balanced 1600 3261 4748 performance 1600 3363 4424 OpenBenchmarking.org Megahertz, More Is Better uvg266 0.4.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.3 20.2 30.3 balanced 1.2 23.9 36.2 performance 2.4 25.9 40.3 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 239 Joules per run. 2. balanced: Approximate power consumption of 267 Joules per run. 3. performance: Approximate power consumption of 281 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Super Fast Original low_power balanced performance 20 40 60 80 100 Min: 77.12 / Avg: 77.53 / Max: 77.81 Min: 77.21 / Avg: 77.72 / Max: 78.24 Min: 84.15 / Avg: 84.31 / Max: 84.45 Min: 87.13 / Avg: 87.42 / Max: 87.7
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
GNU Octave Benchmark This test profile measures how long it takes to complete several reference GNU Octave files via octave-benchmark. GNU Octave is used for numerical computations and is an open-source alternative to MATLAB. Learn more via the OpenBenchmarking.org test page.
Darktable Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Original low_power balanced performance 20 40 60 80 100 SE +/- 0.14, N = 6 SE +/- 0.14, N = 6 SE +/- 0.17, N = 7 SE +/- 0.17, N = 7 93.64 93.61 101.29 104.59 CPU Temp
Min Avg Max Original 47.9 62.6 70.3 low_power 46.5 60.5 67.6 balanced 49.0 66.5 75.5 performance 51.1 70.5 81.4 OpenBenchmarking.org Celsius, Fewer Is Better uvg266 0.4.1 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast low_power balanced performance 1.0915 2.183 3.2745 4.366 5.4575 4.851 4.513 4.304 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1397 3004 3981 balanced 1600 3341 4721 performance 1600 3464 4834 OpenBenchmarking.org Megahertz, More Is Better uvg266 0.4.1 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.4 19.3 30.3 balanced 2.4 22.4 36.3 performance 2.6 24.3 40.4 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 203 Joules per run. 2. balanced: Approximate power consumption of 224 Joules per run. 3. performance: Approximate power consumption of 240 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Original low_power balanced performance 20 40 60 80 100 Min: 93.17 / Avg: 93.64 / Max: 94.15 Min: 93.22 / Avg: 93.61 / Max: 94.25 Min: 100.64 / Avg: 101.29 / Max: 102.09 Min: 104.17 / Avg: 104.59 / Max: 105.42
Darktable Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.
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 1080p Original low_power balanced performance 30 60 90 120 150 SE +/- 0.69, N = 7 SE +/- 0.72, N = 7 SE +/- 0.62, N = 8 SE +/- 0.57, N = 8 128.88 128.67 135.22 137.84 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm CPU Temp
Min Avg Max Original 48.6 61.3 69.3 low_power 48.4 59.6 66.4 balanced 51.8 66.1 74.6 performance 51.8 68.6 79.0 OpenBenchmarking.org Celsius, Fewer Is Better SVT-VP9 0.3 CPU Temperature Monitor 20 40 60 80 100 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 1080p low_power balanced performance 2 4 6 8 10 7.547 6.777 6.325 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3404 5041 balanced 1600 3475 5115 performance 1600 3688 5115 OpenBenchmarking.org Megahertz, More Is Better SVT-VP9 0.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 17.0 30.2 balanced 2.5 20.0 36.1 performance 1.3 21.8 40.1 OpenBenchmarking.org Watts, Fewer Is Better SVT-VP9 0.3 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 153 Joules per run. 2. balanced: Approximate power consumption of 175 Joules per run. 3. performance: Approximate power consumption of 188 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p Original low_power balanced performance 30 60 90 120 150 Min: 124.92 / Avg: 128.88 / Max: 130.4 Min: 124.37 / Avg: 128.67 / Max: 129.65 Min: 130.93 / Avg: 135.22 / Max: 136.16 Min: 133.98 / Avg: 137.84 / Max: 138.96 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 1080p Original low_power balanced performance 40 80 120 160 200 SE +/- 0.19, N = 8 SE +/- 0.16, N = 8 SE +/- 0.12, N = 8 SE +/- 0.14, N = 8 156.02 156.06 162.50 165.13 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm CPU Temp
Min Avg Max Original 48.5 59.8 67.1 low_power 47.6 58.4 65.4 balanced 50.3 63.0 71.1 performance 50.9 66.0 76.1 OpenBenchmarking.org Celsius, Fewer Is Better SVT-VP9 0.3 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 1080p low_power balanced performance 3 6 9 12 15 9.815 8.787 8.230 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3570 5016 balanced 1600 3178 4962 performance 1600 3758 5094 OpenBenchmarking.org Megahertz, More Is Better SVT-VP9 0.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 15.9 30.1 balanced 2.5 18.5 36.3 performance 2.4 20.1 40.2 OpenBenchmarking.org Watts, Fewer Is Better SVT-VP9 0.3 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 129 Joules per run. 2. balanced: Approximate power consumption of 148 Joules per run. 3. performance: Approximate power consumption of 161 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 1080p Original low_power balanced performance 30 60 90 120 150 Min: 155.08 / Avg: 156.02 / Max: 156.78 Min: 155.47 / Avg: 156.06 / Max: 157.05 Min: 162.09 / Avg: 162.5 / Max: 163.15 Min: 164.5 / Avg: 165.13 / Max: 165.73 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Original low_power balanced performance 40 80 120 160 200 SE +/- 0.13, N = 8 SE +/- 0.12, N = 8 SE +/- 0.13, N = 8 SE +/- 0.08, N = 8 144.72 144.52 155.30 160.32 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt CPU Temp
Min Avg Max Original 47.6 59.8 67.4 low_power 46.9 57.9 65.3 balanced 51.0 63.8 72.9 performance 51.9 66.7 77.4 OpenBenchmarking.org Celsius, Fewer Is Better Kvazaar 2.2 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast low_power balanced performance 2 4 6 8 10 8.885 8.278 7.899 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3261 4678 balanced 1600 3138 4668 performance 1600 3398 4609 OpenBenchmarking.org Megahertz, More Is Better Kvazaar 2.2 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max low_power 2.3 16.3 30.1 balanced 2.6 18.8 36.1 performance 2.3 20.3 40.1 OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 134 Joules per run. 2. balanced: Approximate power consumption of 148 Joules per run. 3. performance: Approximate power consumption of 157 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Original low_power balanced performance 30 60 90 120 150 Min: 144.32 / Avg: 144.72 / Max: 145.29 Min: 144.13 / Avg: 144.52 / Max: 145.06 Min: 154.74 / Avg: 155.3 / Max: 155.91 Min: 159.86 / Avg: 160.32 / Max: 160.66 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
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: PSNR/SSIM Optimized - Input: Bosphorus 1080p Original low_power balanced performance 40 80 120 160 200 SE +/- 0.21, N = 8 SE +/- 0.23, N = 8 SE +/- 0.13, N = 8 SE +/- 0.17, N = 9 164.09 163.81 171.28 174.53 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm CPU Temp
Min Avg Max Original 49.1 60.1 67.3 low_power 48.1 58.3 65.0 balanced 50.9 63.1 71.8 performance 51.5 66.0 75.8 OpenBenchmarking.org Celsius, Fewer Is Better SVT-VP9 0.3 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p low_power balanced performance 3 6 9 12 15 10.492 9.485 8.876 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3520 5041 balanced 1600 3438 4998 performance 1600 3617 5115 OpenBenchmarking.org Megahertz, More Is Better SVT-VP9 0.3 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.4 15.6 30.2 balanced 2.5 18.1 36.2 performance 2.4 19.7 40.1 OpenBenchmarking.org Watts, Fewer Is Better SVT-VP9 0.3 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 123 Joules per run. 2. balanced: Approximate power consumption of 142 Joules per run. 3. performance: Approximate power consumption of 153 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p Original low_power balanced performance 30 60 90 120 150 Min: 162.75 / Avg: 164.09 / Max: 164.54 Min: 162.77 / Avg: 163.81 / Max: 164.68 Min: 170.79 / Avg: 171.28 / Max: 171.62 Min: 173.87 / Avg: 174.53 / Max: 175.33 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 1080p Original low_power balanced performance 100 200 300 400 500 SE +/- 7.22, N = 15 SE +/- 7.65, N = 15 SE +/- 7.87, N = 15 SE +/- 8.43, N = 15 414.93 413.82 432.18 440.44 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq CPU Temp
Min Avg Max Original 47.0 52.8 59.1 low_power 46.8 52.0 58.8 balanced 50.6 57.2 66.5 performance 51.6 58.7 67.4 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 1080p low_power balanced performance 10 20 30 40 50 42.19 38.44 36.75 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3288 5116 balanced 1600 3289 5021 performance 1600 3259 5089 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 9.8 30.0 balanced 2.5 11.2 35.9 performance 2.4 12.0 39.8 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 56 Joules per run. 2. balanced: Approximate power consumption of 63 Joules per run. 3. performance: Approximate power consumption of 67 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 1080p Original low_power balanced performance 80 160 240 320 400 Min: 314.12 / Avg: 414.93 / Max: 424.45 Min: 306.85 / Avg: 413.82 / Max: 424.68 Min: 322.06 / Avg: 432.18 / Max: 442.24 Min: 322.72 / Avg: 440.44 / Max: 453.26 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 1080p Original low_power balanced performance 80 160 240 320 400 SE +/- 0.25, N = 11 SE +/- 0.37, N = 11 SE +/- 0.52, N = 11 SE +/- 0.35, N = 11 349.08 348.94 366.09 375.86 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq CPU Temp
Min Avg Max Original 48.0 54.3 61.4 low_power 47.5 53.2 58.3 balanced 48.5 56.5 64.0 performance 50.1 58.3 67.0 OpenBenchmarking.org Celsius, Fewer Is Better SVT-AV1 1.7 CPU Temperature Monitor 20 40 60 80 100 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 1080p low_power balanced performance 7 14 21 28 35 32.17 29.56 28.69 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max low_power 1600 3181 4407 balanced 1600 2980 5115 performance 1600 3464 5115 OpenBenchmarking.org Megahertz, More Is Better SVT-AV1 1.7 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1300 2600 3900 5200 6500 CPU Power Consumption
Min Avg Max low_power 2.3 10.8 30.0 balanced 2.4 12.4 35.9 performance 2.4 13.1 39.8 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 11 22 33 44 55 1. low_power: Approximate power consumption of 65 Joules per run. 2. balanced: Approximate power consumption of 73 Joules per run. 3. performance: Approximate power consumption of 77 Joules per run. Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 1080p Original low_power balanced performance 70 140 210 280 350 Min: 348.07 / Avg: 349.07 / Max: 350.38 Min: 346.18 / Avg: 348.94 / Max: 350.47 Min: 362.56 / Avg: 366.09 / Max: 368.07 Min: 373.91 / Avg: 375.85 / Max: 378.4 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Darktable Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.
CPU Power Consumption Monitor OpenBenchmarking.org Watts CPU Power Consumption Monitor Phoronix Test Suite System Monitoring low_power balanced performance 8 16 24 32 40 Min: 1.19 / Avg: 20.91 / Max: 30.92 Min: 1.18 / Avg: 25.24 / Max: 36.8 Min: 1.12 / Avg: 25.79 / Max: 40.61
CPU Peak Freq (Highest CPU Core Frequency) Monitor OpenBenchmarking.org Megahertz CPU Peak Freq (Highest CPU Core Frequency) Monitor Phoronix Test Suite System Monitoring low_power balanced performance 900 1800 2700 3600 4500 Min: 1034 / Avg: 3642.17 / Max: 5116 Min: 1397 / Avg: 3825.59 / Max: 5123 Min: 1397 / Avg: 3903.12 / Max: 5116
Meta Performance Per Watts OpenBenchmarking.org Performance Per Watts, More Is Better Meta Performance Per Watts Performance Per Watts low_power balanced performance 30 60 90 120 150 120.14 127.20 128.57
CPU Temperature Monitor OpenBenchmarking.org Celsius CPU Temperature Monitor Phoronix Test Suite System Monitoring Original low_power balanced performance 20 40 60 80 100 Min: 35.38 / Avg: 68.88 / Max: 86.38 Min: 38.13 / Avg: 68.62 / Max: 86.5 Min: 35 / Avg: 76.87 / Max: 96 Min: 33.75 / Avg: 78.19 / Max: 96.13
Original Processor: AMD Ryzen 7 PRO 7840U @ 3.30GHz (8 Cores / 16 Threads), Motherboard: LENOVO 21K5001JUS (R2FET31W 1.11 BIOS), Chipset: AMD Device 14e8, Memory: 60GB, Disk: 1024GB Kioxia KXG8AZNV1T02 LA, Graphics: AMD Phoenix1 4GB (2700/400MHz), Audio: AMD Rembrandt Radeon HD Audio, Network: Realtek RTL8111/8168/8411 + Qualcomm QCNFA765
OS: Ubuntu 23.10, Kernel: 6.3.0-7-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 23.1.7-1ubuntu1 (LLVM 15.0.7 DRM 3.52), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 2880x1800
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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) - Platform Profile: low-power - CPU Microcode: 0xa704103 - ACPI Profile: low-powerPython Notes: Python 3.11.5Security 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 Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 10 September 2023 22:46 by user phoronix.
low_power 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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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) - Platform Profile: low-power - CPU Microcode: 0xa704103 - ACPI Profile: low-powerPython Notes: Python 3.11.5Security 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 Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 23 September 2023 22:30 by user phoronix.
balanced 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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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) - Platform Profile: balanced - CPU Microcode: 0xa704103 - ACPI Profile: balancedPython Notes: Python 3.11.5Security 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 Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 24 September 2023 21:23 by user phoronix.
performance Processor: AMD Ryzen 7 PRO 7840U @ 3.30GHz (8 Cores / 16 Threads), Motherboard: LENOVO 21K5001JUS (R2FET33W_0921_3 1.13 BIOS), Chipset: AMD Device 14e8, Memory: 60GB, Disk: 1024GB Kioxia KXG8AZNV1T02 LA, Graphics: AMD Phoenix1 4GB (2700/400MHz), Audio: AMD Rembrandt Radeon HD Audio, Network: Realtek RTL8111/8168/8411 + Qualcomm QCNFA765
OS: Ubuntu 23.10, Kernel: 6.3.0-7-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.7 + Wayland, OpenGL: 4.6 Mesa 23.1.7-1ubuntu1 (LLVM 15.0.7 DRM 3.52), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 2880x1800
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-nEN1TP/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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) - Platform Profile: performance - CPU Microcode: 0xa704103 - ACPI Profile: performancePython Notes: Python 3.11.5Security 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 Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 24 September 2023 10:57 by user phoronix.