Pop OS Kernels Linux 6.8

Benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2404296-NE-POPOSKERN75
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Limit displaying results to tests within:

AV1 4 Tests
Timed Code Compilation 5 Tests
C/C++ Compiler Tests 10 Tests
CPU Massive 16 Tests
Creator Workloads 24 Tests
Encoding 9 Tests
Game Development 3 Tests
HPC - High Performance Computing 6 Tests
Imaging 5 Tests
Common Kernel Benchmarks 2 Tests
Machine Learning 5 Tests
Multi-Core 24 Tests
NVIDIA GPU Compute 2 Tests
Intel oneAPI 4 Tests
OpenMPI Tests 2 Tests
Programmer / Developer System Benchmarks 5 Tests
Python Tests 6 Tests
Raytracing 3 Tests
Renderers 5 Tests
Server 2 Tests
Server CPU Tests 10 Tests
Texture Compression 2 Tests
Video Encoding 8 Tests
Common Workstation Benchmarks 3 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Linux 6.6
April 28
  14 Hours, 50 Minutes
Linux 6.8
April 29
  14 Hours, 7 Minutes
Invert Hiding All Results Option
  14 Hours, 28 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


Pop OS Kernels Linux 6.8ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionLinux 6.6Linux 6.8AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads)Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS)AMD Starship/Matisse4 x 32GB DDR4-3000MT/s CMK64GX4M2D3000C16Samsung SSD 970 EVO Plus 500GBAMD Radeon RX 5700 8GB (1750/875MHz)AMD Navi 10 HDMI AudioDELL P2415QIntel I211 + Intel Wi-Fi 6 AX200Pop 22.046.6.6-76060606-generic (x86_64)GNOME Shell 42.5X Server 1.21.1.44.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.54)1.3.267GCC 11.4.0ext43840x2160AMD Radeon RX 5700 8GB6.8.0-76060800daily20240311-generic (x86_64)4.6 Mesa 24.0.3-1pop1~1711635559~22.04~7a9f319 (LLVM 15.0.7 DRM 3.57)1.3.274OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/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 -v Processor Details- Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107aPython Details- Python 3.10.12Security Details- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Linux 6.6 vs. Linux 6.8 ComparisonPhoronix Test SuiteBaseline+8.7%+8.7%+17.4%+17.4%+26.1%+26.1%+34.8%+34.8%34.8%16%16%13.8%13.7%13.7%13.1%11.9%11%11%10.7%10.6%10.5%10.2%10%9.5%9.4%8.8%8.2%7.3%7.2%6.9%6.9%6.8%6.8%6.8%6.4%6.2%6.2%6%5.7%5.6%5.6%5.4%5.4%5.3%5.2%5.2%5%4.8%4.7%4.5%3.8%3.5%3.5%3.3%2.8%2.6%2.6%2.4%2.3%2.3%WAV To WavPackRand Read20%R.R.W.RPNG - 90Bosphorus 4Klibx265 - Uploadlibx265 - Video On Demandlibx265 - PlatformPNG - 80Bosphorus 4K - SlowBosphorus 4K - MediumSpeed 4 Two-Pass - Bosphorus 4KJPEG - 90Bosphorus 4K - Very FastPreset 4 - Bosphorus 1080pBosphorus 1080p - Fastgravity_spheres_volume/dim_512/scivis/real_time9.7%Preset 4 - Bosphorus 4Klibx265 - Livegravity_spheres_volume/dim_512/ao/real_time9.1%Speed 4 Two-Pass - Bosphorus 1080pBosphorus 4K - Ultra FastSpeed 0 Two-Pass - Bosphorus 4KBosphorus 4K - Super FastJPEG - 100PNG - 100Speed 8 Realtime - Bosphorus 1080pSpeed 10 Realtime - Bosphorus 1080pJPEG - 80Speed 6 Realtime - Bosphorus 1080p1 - 1080p - 32 - Path Tracer - CPU6.3%Update RandSpeed 11 Realtime - Bosphorus 1080pSpeed 9 Realtime - Bosphorus 4KPreset 8 - Bosphorus 4KP.P.B.T.TCPU - 1 - ResNet-50Speed 8 Realtime - Bosphorus 4KSpeed 6 Two-Pass - Bosphorus 1080pBosphorus 1080p - FasterBosphorus 4K - FastSpeed 11 Realtime - Bosphorus 4KSpeed 0 Two-Pass - Bosphorus 1080pSpeed 6 Two-Pass - Bosphorus 4KBosphorus 1080pA.G.R.R.0.F.I - CPUChess Benchmark4.5%HWB Color Space3.9%Speed 10 Realtime - Bosphorus 4KAllPreset 13 - Bosphorus 4KP.D.F - CPU3.4%P.D.F - CPU3.3%Speed 6 Realtime - Bosphorus 4KP.D.F - CPU3.2%P.D.F - CPU3.2%Noise-Gaussian3.1%Mount St. Helens3.1%A.G.R.R.0.F.I - CPUP.P.B.T.T2.7%H.H2.6%Bosphorus 4K - FasterPreset 8 - Bosphorus 1080pBosphorus 1080p - Ultra FastAES-256-GCM2.4%F.D.R.F - CPU2.3%CPUallmodconfig2.3%Bosphorus 1080p - Super FastAES-128-GCM2.3%W.l.H2.2%Read While Writing2.2%F.D.R.F - CPU2.2%WavPack Audio EncodingRocksDBRocksDBJPEG-XL libjxlx265FFmpegFFmpegFFmpegJPEG-XL libjxluvg266uvg266AOM AV1JPEG-XL libjxluvg266SVT-AV1VVenCOSPRaySVT-AV1FFmpegOSPRayAOM AV1uvg266AOM AV1uvg266JPEG-XL libjxlJPEG-XL libjxlAOM AV1AOM AV1JPEG-XL libjxlAOM AV1OSPRay StudioRocksDBAOM AV1AOM AV1SVT-AV1srsRAN ProjectPyTorchAOM AV1AOM AV1VVenCVVenCAOM AV1AOM AV1AOM AV1x265OpenVINOStockfishGraphicsMagickAOM AV1JPEG-XL Decoding libjxlSVT-AV1OpenVINOOpenVINOAOM AV1OpenVINOOpenVINOGraphicsMagickSPECFEM3DOpenVINOsrsRAN ProjectSPECFEM3DVVenCSVT-AV1uvg266OpenSSLOpenVINOChaos Group V-RAYTimed Linux Kernel Compilationuvg266OpenSSLSPECFEM3DRocksDBOpenVINOLinux 6.6Linux 6.8

Pop OS Kernels Linux 6.8tensorflow: CPU - 64 - ResNet-50stockfish: Chess Benchmarkgraph500: 26graph500: 26graph500: 26graph500: 26build-linux-kernel: allmodconfigopenssl: RSA4096openssl: RSA4096blender: Barbershop - CPU-Onlyrocksdb: Rand Readffmpeg: libx264 - Uploadffmpeg: libx265 - Video On Demandffmpeg: libx265 - Platformbuild-nodejs: Time To Compileffmpeg: libx265 - Uploadbrl-cad: VGR Performance Metricpytorch: CPU - 1 - ResNet-50jpegxl: PNG - 80jpegxl: JPEG - 80ffmpeg: libx264 - Video On Demandffmpeg: libx264 - Platformospray: particle_volume/pathtracer/real_timeopenssl: SHA256openssl: ChaCha20-Poly1305openssl: SHA512openssl: AES-256-GCMopenssl: ChaCha20openssl: AES-128-GCMospray-studio: 1 - 1080p - 32 - Path Tracer - CPUospray: particle_volume/scivis/real_timespecfem3d: Water-layered Halfspacesrsran: PUSCH Processor Benchmark, Throughput Totaljpegxl: JPEG - 90ospray-studio: 3 - 4K - 32 - Path Tracer - CPUvvenc: Bosphorus 4K - Fastospray-studio: 2 - 4K - 32 - Path Tracer - CPUospray-studio: 1 - 4K - 32 - Path Tracer - CPUc-ray: 5K - 16build-linux-kernel: defconfigospray-studio: 2 - 1080p - 16 - Path Tracer - CPUospray-studio: 3 - 1080p - 16 - Path Tracer - CPUospray-studio: 1 - 1080p - 16 - Path Tracer - CPUospray: particle_volume/ao/real_timeffmpeg: libx265 - Liveblender: Pabellon Barcelona - CPU-Onlyaom-av1: Speed 4 Two-Pass - Bosphorus 4Kblender: Classroom - CPU-Onlyv-ray: CPUonednn: Recurrent Neural Network Training - CPUospray-studio: 3 - 4K - 1 - Path Tracer - CPUonednn: IP Shapes 1D - CPUavifenc: 0ospray-studio: 1 - 4K - 1 - Path Tracer - CPUonednn: Recurrent Neural Network Inference - CPUospray-studio: 2 - 4K - 1 - Path Tracer - CPUospray-studio: 3 - 4K - 16 - Path Tracer - CPUaom-av1: Speed 6 Two-Pass - Bosphorus 4Kopenvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUospray: gravity_spheres_volume/dim_512/scivis/real_timeospray-studio: 3 - 1080p - 1 - Path Tracer - CPUospray-studio: 1 - 1080p - 1 - Path Tracer - CPUospray-studio: 2 - 1080p - 1 - Path Tracer - CPUopenvino: Face Detection FP16 - CPUopenvino: Face Detection FP16 - CPUospray-studio: 2 - 4K - 16 - Path Tracer - CPUospray: gravity_spheres_volume/dim_512/ao/real_timeospray-studio: 1 - 4K - 16 - Path Tracer - CPUopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Noise Suppression Poconet-Like FP16 - CPUopenvino: Noise Suppression Poconet-Like FP16 - CPUopenvino: Person Re-Identification Retail FP16 - CPUopenvino: Person Re-Identification Retail FP16 - CPUopenvino: Handwritten English Recognition FP16 - CPUopenvino: Handwritten English Recognition FP16 - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Vehicle Detection FP16 - CPUopenvino: Vehicle Detection FP16 - CPUopenvino: Face Detection Retail FP16 - CPUopenvino: Face Detection Retail FP16 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUrocksdb: Update Randgraphics-magick: Resizinggraphics-magick: Rotaterocksdb: Read Rand Write Randgraphics-magick: Sharpengraphics-magick: Noise-Gaussiangraphics-magick: HWB Color Spacegraphics-magick: Enhancedrocksdb: Read While Writinggraphics-magick: Swirlospray: gravity_spheres_volume/dim_512/pathtracer/real_timec-ray: 4K - 16uvg266: Bosphorus 4K - Slowastcenc: Very Thoroughastcenc: Exhaustivevvenc: Bosphorus 4K - Fasterspecfem3d: Layered Halfspaceuvg266: Bosphorus 4K - Mediumffmpeg: libx264 - Livevvenc: Bosphorus 1080p - Fastaom-av1: Speed 0 Two-Pass - Bosphorus 4Kjpegxl: PNG - 90ospray-studio: 3 - 1080p - 32 - Path Tracer - CPUprimesieve: 1e13jpegxl-decode: 1blender: Junkshop - CPU-Onlyaom-av1: Speed 6 Realtime - Bosphorus 4Kaom-av1: Speed 9 Realtime - Bosphorus 4Kospray-studio: 2 - 1080p - 32 - Path Tracer - CPUaom-av1: Speed 11 Realtime - Bosphorus 4Kbuild-php: Time To Compileavifenc: 2webp: Quality 100, Lossless, Highest Compressionaom-av1: Speed 4 Two-Pass - Bosphorus 1080pspecfem3d: Mount St. Helenssrsran: PDSCH Processor Benchmark, Throughput Totalblender: Fishy Cat - CPU-Onlysvt-av1: Preset 4 - Bosphorus 4Kaom-av1: Speed 6 Two-Pass - Bosphorus 1080pblender: BMW27 - CPU-Onlyencode-wavpack: WAV To WavPackbuild-ffmpeg: Time To Compilex265: Bosphorus 4Kdav1d: Chimera 1080p 10-bitjpegxl: PNG - 100tensorflow: CPU - 1 - ResNet-50jpegxl: JPEG - 100jpegxl-decode: Alldav1d: Chimera 1080pvvenc: Bosphorus 1080p - Fasterspecfem3d: Tomographic Modelonednn: Deconvolution Batch shapes_1d - CPUsrsran: PUSCH Processor Benchmark, Throughput Threadspecfem3d: Homogeneous Halfspaceaom-av1: Speed 6 Realtime - Bosphorus 1080paom-av1: Speed 8 Realtime - Bosphorus 1080paom-av1: Speed 9 Realtime - Bosphorus 1080paom-av1: Speed 11 Realtime - Bosphorus 1080paom-av1: Speed 10 Realtime - Bosphorus 1080pastcenc: Thoroughuvg266: Bosphorus 4K - Very Fastuvg266: Bosphorus 1080p - Slowdav1d: Summer Nature 4Kuvg266: Bosphorus 4K - Super Fastsrsran: PDSCH Processor Benchmark, Throughput Threadaom-av1: Speed 0 Two-Pass - Bosphorus 1080psvt-av1: Preset 8 - Bosphorus 4Kwebp: Quality 100, Losslessuvg266: Bosphorus 1080p - Mediumuvg266: Bosphorus 4K - Ultra Fastaom-av1: Speed 8 Realtime - Bosphorus 4Kbuild-mesa: Time To Compilec-ray: 1080p - 16svt-av1: Preset 4 - Bosphorus 1080paom-av1: Speed 10 Realtime - Bosphorus 4Kx265: Bosphorus 1080prnnoise: 26 Minute Long Talking Samplecompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressiondraco: Church Facadeastcenc: Fastonednn: IP Shapes 3D - CPUsvt-av1: Preset 8 - Bosphorus 1080ponednn: Convolution Batch Shapes Auto - CPUsvt-av1: Preset 12 - Bosphorus 4Ksvt-av1: Preset 13 - Bosphorus 4Kastcenc: Mediumdav1d: Summer Nature 1080pdraco: Lionwebp: Quality 100, Highest Compressionavifenc: 6, Losslessavifenc: 10, Losslessuvg266: Bosphorus 1080p - Very Fastuvg266: Bosphorus 1080p - Super Fastuvg266: Bosphorus 1080p - Ultra Fastprimesieve: 1e12avifenc: 6onednn: Deconvolution Batch shapes_3d - CPUsvt-av1: Preset 12 - Bosphorus 1080psvt-av1: Preset 13 - Bosphorus 1080pwebp: Quality 100webp: DefaultLinux 6.6Linux 6.811.1275922419216393000161857000539379000533558000401.549948619.414679.9301.8118453298511.7827.2127.28269.27813.0376853319.0428.68027.23243.9044.08127.56462279387347128107845040227665725572074297992971999048562902261806670702833314.290943.9175019702582.625.0181165144.84910148799823106.40641.56011931138501177314.531566.3595.097.8580.98673221325.9834241.3178276.4072916757.06629626152116.791794.2717.736.780168647357432800.2411.30540577.1317153229412.8677.34411.8577.56270.19118.2552.47608.9021.381491.58371.8285.9123.371352.5317.151850.83121.52525.95108.24590.4123.331367.627.364295.9263.75500.8310.053148.2936.071764.441.6532704.421.1544647.9727.641153.593019841181062349337154133188194907336736710.833659.1659.794.68582.874011.57649.26325049210.97179.6912.3190.4123.7283412445.93554.28743.5043.2145.133046846.2141.98641.4090.5916.2813.1393255019560.736.914.61039.0829.827.39327.64722.35351.9119.0785.4519.237328.085375.3228.71814.21982632411.371811217.58420558987.7188.9996.1093.8093.5633.945731.8433.21205.4833.35489.51.1944.3031.4836.7035.9641.3415.08514.76112.27847.7847.0411.6682.1188718545662.73247.7294188.1550.958085102.498102.526262.8047487.0555053.367.1915.667112.86126.57143.343.5583.4532.15953308.169373.78910.4317.1611.1972639152214883000161136000541789000535202000410.952943736.914635.0299.4115378560711.7830.9430.85265.65514.8276942720.1032.08729.08343.9744.08126.44262255778377126331970173224014120772026166743501960960584902210883059733011914.249944.8904015402621.327.6781170825.103101836100225107.52841.62611970139221180614.467572.5995.268.6980.75689031332.3134461.3312376.7372930763.01129746195417.601794.0017.726.182078687397502824.0711.21544616.5358653719426.5674.83425.0475.15270.47118.1352.26611.3621.021516.00373.5485.5223.061365.5517.241838.91121.97523.96108.19590.7023.281369.877.404252.5864.24496.9110.273076.0136.131761.731.6732552.241.145877.9227.671152.643208501161052724328153129181193887881336610.756459.58710.874.68172.869111.87849.36048859912.18182.3013.5560.4427.5143425046.41354.11243.2644.6447.833051548.6141.92541.1900.5917.7213.54316660710095.436.775.04841.1930.025.48627.65125.44352.9920.3935.5120.564339.653375.7630.24014.33888939311.521811218.04497079493.2895.0496.6099.6699.9233.934335.1733.68205.1135.75476.51.2546.8261.4937.2038.9043.5815.01414.87613.52649.6149.2711.6212.1440218641666.84867.7380490.4200.949172104.229106.112262.8280482.9455473.397.2125.610115.03129.52146.793.5473.4412.17554311.067373.12810.4517.32OpenBenchmarking.org

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: ResNet-50Linux 6.8Linux 6.63691215SE +/- 0.01, N = 3SE +/- 0.02, N = 311.1911.12

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 1024 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess BenchmarkLinux 6.8Linux 6.616M32M48M64M80MSE +/- 1376453.74, N = 12SE +/- 1774428.97, N = 1272639152759224191. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -msse -msse3 -mpopcnt -mavx2 -mbmi -msse4.1 -mssse3 -msse2 -flto -flto-partition=one -flto=jobserver

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsssp max_TEPS, More Is BetterGraph500 3.0Scale: 26Linux 6.8Linux 6.650M100M150M200M250M2148830002163930001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

OpenBenchmarking.orgsssp median_TEPS, More Is BetterGraph500 3.0Scale: 26Linux 6.8Linux 6.630M60M90M120M150M1611360001618570001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

OpenBenchmarking.orgbfs max_TEPS, More Is BetterGraph500 3.0Scale: 26Linux 6.8Linux 6.6120M240M360M480M600M5417890005393790001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

OpenBenchmarking.orgbfs median_TEPS, More Is BetterGraph500 3.0Scale: 26Linux 6.8Linux 6.6110M220M330M440M550M5352020005335580001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: allmodconfigLinux 6.8Linux 6.690180270360450SE +/- 0.75, N = 3SE +/- 0.91, N = 3410.95401.55

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.3Algorithm: RSA4096Linux 6.8Linux 6.6200K400K600K800K1000KSE +/- 823.53, N = 3SE +/- 317.47, N = 3943736.9948619.41. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.3Algorithm: RSA4096Linux 6.8Linux 6.63K6K9K12K15KSE +/- 14.78, N = 3SE +/- 15.24, N = 314635.014679.91. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Barbershop - Compute: CPU-OnlyLinux 6.8Linux 6.670140210280350SE +/- 0.97, N = 3SE +/- 0.78, N = 3299.41301.81

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Random ReadLinux 6.8Linux 6.640M80M120M160M200MSE +/- 3162994.92, N = 15SE +/- 1209121.69, N = 141537856071845329851. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx264 - Scenario: UploadLinux 6.8Linux 6.63691215SE +/- 0.01, N = 3SE +/- 0.01, N = 311.7811.781. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: Video On DemandLinux 6.8Linux 6.6714212835SE +/- 0.02, N = 3SE +/- 0.06, N = 330.9427.211. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: PlatformLinux 6.8Linux 6.6714212835SE +/- 0.01, N = 3SE +/- 0.02, N = 330.8527.281. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 21.7.2Time To CompileLinux 6.8Linux 6.660120180240300SE +/- 1.21, N = 3SE +/- 2.20, N = 3265.66269.28

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: UploadLinux 6.8Linux 6.648121620SE +/- 0.04, N = 3SE +/- 0.05, N = 314.8213.031. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

BRL-CAD

BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.38.2VGR Performance MetricLinux 6.8Linux 6.6160K320K480K640K800K7694277685331. (CXX) g++ options: -std=c++17 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lnetpbm -lregex_brl -lz_brl -lassimp -ldl -lm -ltk8.6

PyTorch

This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-50Linux 6.8Linux 6.6510152025SE +/- 0.20, N = 6SE +/- 0.26, N = 1520.1019.04MIN: 18.81 / MAX: 21.87MIN: 16.87 / MAX: 21.85

JPEG-XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 80Linux 6.8Linux 6.6714212835SE +/- 0.21, N = 3SE +/- 0.47, N = 1532.0928.681. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 80Linux 6.8Linux 6.6714212835SE +/- 0.41, N = 3SE +/- 0.38, N = 1429.0827.231. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx264 - Scenario: Video On DemandLinux 6.8Linux 6.61020304050SE +/- 0.10, N = 3SE +/- 0.01, N = 343.9743.901. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx264 - Scenario: PlatformLinux 6.8Linux 6.61020304050SE +/- 0.03, N = 3SE +/- 0.04, N = 344.0844.081. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: particle_volume/pathtracer/real_timeLinux 6.8Linux 6.6306090120150SE +/- 0.27, N = 3SE +/- 0.45, N = 3126.44127.56

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: SHA256Linux 6.8Linux 6.613000M26000M39000M52000M65000MSE +/- 119311058.77, N = 3SE +/- 85875563.68, N = 362255778377622793873471. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20-Poly1305Linux 6.8Linux 6.630000M60000M90000M120000M150000MSE +/- 143226146.09, N = 3SE +/- 175506084.92, N = 31263319701731281078450401. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: SHA512Linux 6.8Linux 6.65000M10000M15000M20000M25000MSE +/- 35361661.51, N = 3SE +/- 10966945.92, N = 322401412077227665725571. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: AES-256-GCMLinux 6.8Linux 6.640000M80000M120000M160000M200000MSE +/- 98406556.77, N = 3SE +/- 284249597.90, N = 32026166743502074297992971. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20Linux 6.8Linux 6.640000M80000M120000M160000M200000MSE +/- 59010640.14, N = 3SE +/- 40917174.40, N = 31960960584901999048562901. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: AES-128-GCMLinux 6.8Linux 6.650000M100000M150000M200000M250000MSE +/- 184834186.21, N = 3SE +/- 69181218.89, N = 32210883059732261806670701. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.66K12K18K24K30KSE +/- 45.04, N = 3SE +/- 777.74, N = 153011928333

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: particle_volume/scivis/real_timeLinux 6.8Linux 6.648121620SE +/- 0.01, N = 3SE +/- 0.02, N = 314.2514.29

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Water-layered HalfspaceLinux 6.8Linux 6.61020304050SE +/- 0.45, N = 15SE +/- 0.21, N = 344.8943.921. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput TotalLinux 6.8Linux 6.66001200180024003000SE +/- 24.53, N = 15SE +/- 21.32, N = 92621.32582.6MIN: 1454.6 / MAX: 2769.4MIN: 1453.8 / MAX: 2649.81. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

JPEG-XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 90Linux 6.8Linux 6.6714212835SE +/- 0.27, N = 3SE +/- 0.27, N = 1527.6825.021. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.630K60K90K120K150KSE +/- 358.28, N = 3SE +/- 95.84, N = 3117082116514

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.11Video Input: Bosphorus 4K - Video Preset: FastLinux 6.8Linux 6.61.14822.29643.44464.59285.741SE +/- 0.064, N = 3SE +/- 0.011, N = 35.1034.8491. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.620K40K60K80K100KSE +/- 150.23, N = 3SE +/- 36.57, N = 3101836101487

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.620K40K60K80K100KSE +/- 131.90, N = 3SE +/- 149.27, N = 310022599823

C-Ray

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 2.0Resolution: 5K - Rays Per Pixel: 16Linux 6.8Linux 6.620406080100SE +/- 0.34, N = 3SE +/- 0.17, N = 3107.53106.411. (CC) gcc options: -lpthread -lm

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: defconfigLinux 6.8Linux 6.6918273645SE +/- 0.37, N = 7SE +/- 0.35, N = 841.6341.56

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.63K6K9K12K15KSE +/- 20.25, N = 3SE +/- 15.32, N = 31197011931

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.63K6K9K12K15KSE +/- 8.19, N = 3SE +/- 6.03, N = 31392213850

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.63K6K9K12K15KSE +/- 22.93, N = 3SE +/- 10.12, N = 31180611773

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: particle_volume/ao/real_timeLinux 6.8Linux 6.648121620SE +/- 0.03, N = 3SE +/- 0.00, N = 314.4714.53

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: LiveLinux 6.8Linux 6.61632486480SE +/- 0.10, N = 3SE +/- 0.10, N = 372.5966.351. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Pabellon Barcelona - Compute: CPU-OnlyLinux 6.8Linux 6.620406080100SE +/- 0.04, N = 3SE +/- 0.27, N = 395.2695.09

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4KLinux 6.8Linux 6.6246810SE +/- 0.01, N = 3SE +/- 0.05, N = 38.697.851. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Classroom - Compute: CPU-OnlyLinux 6.8Linux 6.620406080100SE +/- 0.50, N = 3SE +/- 0.52, N = 380.7580.98

Chaos Group V-RAY

This is a test of Chaos Group's V-RAY benchmark. V-RAY is a commercial renderer that can integrate with various creator software products like SketchUp and 3ds Max. The V-RAY benchmark is standalone and supports CPU and NVIDIA CUDA/RTX based rendering. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgvsamples, More Is BetterChaos Group V-RAY 6.0Mode: CPULinux 6.8Linux 6.615K30K45K60K75KSE +/- 822.51, N = 3SE +/- 282.60, N = 36890367322

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Recurrent Neural Network Training - Engine: CPULinux 6.8Linux 6.630060090012001500SE +/- 1.24, N = 3SE +/- 1.58, N = 31332.311325.98MIN: 1301.27MIN: 1291.171. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.67001400210028003500SE +/- 2.65, N = 3SE +/- 3.51, N = 334463424

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: IP Shapes 1D - Engine: CPULinux 6.8Linux 6.60.29950.5990.89851.1981.4975SE +/- 0.02342, N = 15SE +/- 0.03534, N = 151.331231.31782MIN: 1.03MIN: 1.011. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 0Linux 6.8Linux 6.620406080100SE +/- 0.11, N = 3SE +/- 0.51, N = 376.7476.411. (CXX) g++ options: -O3 -fPIC -lm

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.66001200180024003000SE +/- 2.03, N = 3SE +/- 3.53, N = 329302916

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Recurrent Neural Network Inference - Engine: CPULinux 6.8Linux 6.6160320480640800SE +/- 0.44, N = 3SE +/- 0.64, N = 3763.01757.07MIN: 733.29MIN: 727.641. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.66001200180024003000SE +/- 2.03, N = 3SE +/- 1.76, N = 329742962

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.613K26K39K52K65KSE +/- 77.47, N = 3SE +/- 26.10, N = 36195461521

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4KLinux 6.8Linux 6.648121620SE +/- 0.18, N = 6SE +/- 0.24, N = 317.6016.791. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPULinux 6.8Linux 6.6400800120016002000SE +/- 2.20, N = 3SE +/- 0.28, N = 31794.001794.27MIN: 1670.12 / MAX: 1830.77MIN: 1695.76 / MAX: 1822.061. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPULinux 6.8Linux 6.648121620SE +/- 0.00, N = 3SE +/- 0.00, N = 317.7217.731. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeLinux 6.8Linux 6.6246810SE +/- 0.01553, N = 3SE +/- 0.00782, N = 36.182076.78016

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.62004006008001000SE +/- 1.20, N = 3SE +/- 0.00, N = 3868864

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.6160320480640800SE +/- 0.88, N = 3SE +/- 0.33, N = 3739735

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.6160320480640800SE +/- 0.88, N = 3SE +/- 0.33, N = 3750743

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection FP16 - Device: CPULinux 6.8Linux 6.66001200180024003000SE +/- 5.03, N = 3SE +/- 12.80, N = 32824.072800.24MIN: 2254.93 / MAX: 3165.67MIN: 2144.88 / MAX: 3178.411. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16 - Device: CPULinux 6.8Linux 6.63691215SE +/- 0.02, N = 3SE +/- 0.04, N = 311.2111.301. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.612K24K36K48K60KSE +/- 12.02, N = 3SE +/- 89.80, N = 35446154057

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: gravity_spheres_volume/dim_512/ao/real_timeLinux 6.8Linux 6.6246810SE +/- 0.03882, N = 3SE +/- 0.01048, N = 36.535867.13171

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.612K24K36K48K60KSE +/- 173.06, N = 3SE +/- 43.89, N = 35371953229

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPULinux 6.8Linux 6.690180270360450SE +/- 1.20, N = 3SE +/- 0.49, N = 3426.56412.86MIN: 145.66 / MAX: 516.1MIN: 131.24 / MAX: 511.621. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPULinux 6.8Linux 6.620406080100SE +/- 0.20, N = 3SE +/- 0.10, N = 374.8377.341. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPULinux 6.8Linux 6.690180270360450SE +/- 0.39, N = 3SE +/- 0.29, N = 3425.04411.85MIN: 153.23 / MAX: 530.53MIN: 118.06 / MAX: 514.031. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPULinux 6.8Linux 6.620406080100SE +/- 0.04, N = 3SE +/- 0.07, N = 375.1577.561. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPULinux 6.8Linux 6.660120180240300SE +/- 1.96, N = 3SE +/- 1.55, N = 3270.47270.19MIN: 130.43 / MAX: 364.05MIN: 140.17 / MAX: 358.211. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPULinux 6.8Linux 6.6306090120150SE +/- 0.85, N = 3SE +/- 0.68, N = 3118.13118.251. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPULinux 6.8Linux 6.61224364860SE +/- 0.19, N = 3SE +/- 0.13, N = 352.2652.47MIN: 27.34 / MAX: 96.1MIN: 35 / MAX: 102.541. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPULinux 6.8Linux 6.6130260390520650SE +/- 2.28, N = 3SE +/- 1.47, N = 3611.36608.901. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPULinux 6.8Linux 6.6510152025SE +/- 0.06, N = 3SE +/- 0.01, N = 321.0221.38MIN: 12.34 / MAX: 44.15MIN: 11.76 / MAX: 43.881. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPULinux 6.8Linux 6.630060090012001500SE +/- 4.44, N = 3SE +/- 0.74, N = 31516.001491.581. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPULinux 6.8Linux 6.680160240320400SE +/- 0.65, N = 3SE +/- 0.40, N = 3373.54371.82MIN: 102.04 / MAX: 455.03MIN: 87.53 / MAX: 446.751. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPULinux 6.8Linux 6.620406080100SE +/- 0.15, N = 3SE +/- 0.10, N = 385.5285.911. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPULinux 6.8Linux 6.6612182430SE +/- 0.05, N = 3SE +/- 0.09, N = 323.0623.37MIN: 10.63 / MAX: 43.28MIN: 13.11 / MAX: 45.221. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPULinux 6.8Linux 6.630060090012001500SE +/- 2.39, N = 3SE +/- 5.06, N = 31365.551352.531. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPULinux 6.8Linux 6.648121620SE +/- 0.02, N = 3SE +/- 0.01, N = 317.2417.15MIN: 9.78 / MAX: 37.37MIN: 9.61 / MAX: 36.361. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPULinux 6.8Linux 6.6400800120016002000SE +/- 2.70, N = 3SE +/- 0.95, N = 31838.911850.831. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16 - Device: CPULinux 6.8Linux 6.6306090120150SE +/- 0.13, N = 3SE +/- 0.16, N = 3121.97121.52MIN: 81.68 / MAX: 222.61MIN: 82.39 / MAX: 220.521. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16 - Device: CPULinux 6.8Linux 6.6110220330440550SE +/- 0.54, N = 3SE +/- 0.72, N = 3523.96525.951. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPULinux 6.8Linux 6.620406080100SE +/- 0.26, N = 3SE +/- 0.07, N = 3108.19108.24MIN: 73.68 / MAX: 125.37MIN: 72.91 / MAX: 130.491. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPULinux 6.8Linux 6.6130260390520650SE +/- 1.48, N = 3SE +/- 0.41, N = 3590.70590.411. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPULinux 6.8Linux 6.6612182430SE +/- 0.00, N = 3SE +/- 0.04, N = 323.2823.33MIN: 11.89 / MAX: 46.96MIN: 12 / MAX: 51.521. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPULinux 6.8Linux 6.630060090012001500SE +/- 0.03, N = 3SE +/- 2.14, N = 31369.871367.621. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPULinux 6.8Linux 6.6246810SE +/- 0.03, N = 3SE +/- 0.02, N = 37.407.36MIN: 4.16 / MAX: 21.13MIN: 4.22 / MAX: 251. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPULinux 6.8Linux 6.69001800270036004500SE +/- 13.32, N = 3SE +/- 9.09, N = 34252.584295.921. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPULinux 6.8Linux 6.61428425670SE +/- 0.09, N = 3SE +/- 0.01, N = 364.2463.75MIN: 15.88 / MAX: 98.58MIN: 23.31 / MAX: 101.151. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPULinux 6.8Linux 6.6110220330440550SE +/- 0.69, N = 3SE +/- 0.07, N = 3496.91500.831. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16 - Device: CPULinux 6.8Linux 6.63691215SE +/- 0.11, N = 3SE +/- 0.06, N = 310.2710.05MIN: 4.86 / MAX: 31.64MIN: 4.71 / MAX: 32.451. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16 - Device: CPULinux 6.8Linux 6.67001400210028003500SE +/- 33.28, N = 3SE +/- 19.77, N = 33076.013148.291. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPULinux 6.8Linux 6.6816243240SE +/- 0.06, N = 3SE +/- 0.03, N = 336.1336.07MIN: 18.31 / MAX: 60.59MIN: 18.65 / MAX: 65.821. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPULinux 6.8Linux 6.6400800120016002000SE +/- 2.70, N = 3SE +/- 1.40, N = 31761.731764.441. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16 - Device: CPULinux 6.8Linux 6.60.37580.75161.12741.50321.879SE +/- 0.01, N = 3SE +/- 0.00, N = 31.671.65MIN: 0.83 / MAX: 11.89MIN: 0.8 / MAX: 14.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16 - Device: CPULinux 6.8Linux 6.67K14K21K28K35KSE +/- 175.08, N = 3SE +/- 89.06, N = 332552.2432704.421. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPULinux 6.8Linux 6.60.25880.51760.77641.03521.294SE +/- 0.00, N = 3SE +/- 0.00, N = 31.101.15MIN: 0.51 / MAX: 12.66MIN: 0.57 / MAX: 12.391. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPULinux 6.8Linux 6.610K20K30K40K50KSE +/- 49.88, N = 3SE +/- 18.81, N = 345877.9244647.971. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16 - Device: CPULinux 6.8Linux 6.6714212835SE +/- 0.01, N = 3SE +/- 0.05, N = 327.6727.64MIN: 15.01 / MAX: 52.41MIN: 15.37 / MAX: 52.581. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16 - Device: CPULinux 6.8Linux 6.62004006008001000SE +/- 0.37, N = 3SE +/- 2.37, N = 31152.641153.591. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Update RandomLinux 6.8Linux 6.670K140K210K280K350KSE +/- 346.93, N = 3SE +/- 306.76, N = 33208503019841. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: ResizingLinux 6.8Linux 6.6306090120150SE +/- 0.33, N = 3SE +/- 0.33, N = 31161181. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -llzma -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: RotateLinux 6.8Linux 6.620406080100SE +/- 0.33, N = 3SE +/- 0.00, N = 31051061. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -llzma -lz -lm -lpthread -lgomp

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read Random Write RandomLinux 6.8Linux 6.6600K1200K1800K2400K3000KSE +/- 6021.59, N = 3SE +/- 9938.97, N = 3272432823493371. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: SharpenLinux 6.8Linux 6.6306090120150SE +/- 0.33, N = 3SE +/- 0.67, N = 31531541. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -llzma -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Noise-GaussianLinux 6.8Linux 6.6306090120150SE +/- 0.58, N = 3SE +/- 0.33, N = 31291331. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -llzma -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: HWB Color SpaceLinux 6.8Linux 6.64080120160200SE +/- 0.00, N = 3SE +/- 0.33, N = 31811881. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -llzma -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: EnhancedLinux 6.8Linux 6.64080120160200SE +/- 0.00, N = 3SE +/- 0.33, N = 31931941. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -llzma -lz -lm -lpthread -lgomp

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read While WritingLinux 6.8Linux 6.62M4M6M8M10MSE +/- 79378.85, N = 3SE +/- 9315.20, N = 3887881390733671. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: SwirlLinux 6.8Linux 6.680160240320400SE +/- 0.88, N = 3SE +/- 0.88, N = 33663671. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -llzma -lz -lm -lpthread -lgomp

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeLinux 6.8Linux 6.63691215SE +/- 0.00, N = 3SE +/- 0.01, N = 310.7610.83

C-Ray

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 2.0Resolution: 4K - Rays Per Pixel: 16Linux 6.8Linux 6.61326395265SE +/- 0.24, N = 3SE +/- 0.42, N = 359.5959.171. (CC) gcc options: -lpthread -lm

uvg266

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 4K - Video Preset: SlowLinux 6.8Linux 6.63691215SE +/- 0.01, N = 3SE +/- 0.01, N = 310.879.79

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.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: Very ThoroughLinux 6.8Linux 6.61.05432.10863.16294.21725.2715SE +/- 0.0122, N = 3SE +/- 0.0155, N = 34.68174.68581. (CXX) g++ options: -O3 -flto -pthread

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ExhaustiveLinux 6.8Linux 6.60.64671.29341.94012.58683.2335SE +/- 0.0108, N = 3SE +/- 0.0119, N = 32.86912.87401. (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.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.11Video Input: Bosphorus 4K - Video Preset: FasterLinux 6.8Linux 6.63691215SE +/- 0.04, N = 3SE +/- 0.01, N = 311.8811.581. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Layered HalfspaceLinux 6.8Linux 6.61122334455SE +/- 0.04, N = 3SE +/- 0.47, N = 349.3649.261. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

uvg266

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 4K - Video Preset: MediumLinux 6.8Linux 6.63691215SE +/- 0.02, N = 3SE +/- 0.01, N = 312.1810.97

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx264 - Scenario: LiveLinux 6.8Linux 6.64080120160200SE +/- 0.36, N = 3SE +/- 0.31, N = 3182.30179.691. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.11Video Input: Bosphorus 1080p - Video Preset: FastLinux 6.8Linux 6.63691215SE +/- 0.14, N = 3SE +/- 0.01, N = 313.5612.321. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4KLinux 6.8Linux 6.60.0990.1980.2970.3960.495SE +/- 0.00, N = 3SE +/- 0.00, N = 30.440.411. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

JPEG-XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 90Linux 6.8Linux 6.6612182430SE +/- 0.13, N = 3SE +/- 0.07, N = 327.5123.731. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.67K14K21K28K35KSE +/- 46.53, N = 3SE +/- 48.01, N = 33425034124

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e13Linux 6.8Linux 6.61122334455SE +/- 0.29, N = 3SE +/- 0.29, N = 346.4145.941. (CXX) g++ options: -O3

JPEG-XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL Decoding libjxl 0.10.1CPU Threads: 1Linux 6.8Linux 6.61224364860SE +/- 0.23, N = 3SE +/- 0.21, N = 354.1154.29

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Junkshop - Compute: CPU-OnlyLinux 6.8Linux 6.61020304050SE +/- 0.05, N = 3SE +/- 0.13, N = 343.2643.50

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KLinux 6.8Linux 6.61020304050SE +/- 0.47, N = 15SE +/- 0.18, N = 344.6443.211. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KLinux 6.8Linux 6.61122334455SE +/- 0.31, N = 3SE +/- 0.49, N = 1547.8345.131. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPULinux 6.8Linux 6.67K14K21K28K35KSE +/- 11.79, N = 3SE +/- 32.04, N = 33051530468

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 11 Realtime - Input: Bosphorus 4KLinux 6.8Linux 6.61122334455SE +/- 0.49, N = 3SE +/- 0.51, N = 1548.6146.211. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Timed PHP Compilation

This test times how long it takes to build PHP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 8.3.4Time To CompileLinux 6.8Linux 6.61020304050SE +/- 0.08, N = 3SE +/- 0.02, N = 341.9341.99

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 2Linux 6.8Linux 6.6918273645SE +/- 0.19, N = 3SE +/- 0.09, N = 341.1941.411. (CXX) g++ options: -O3 -fPIC -lm

WebP Image Encode

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: Quality 100, Lossless, Highest CompressionLinux 6.8Linux 6.60.13280.26560.39840.53120.664SE +/- 0.00, N = 3SE +/- 0.00, N = 30.590.591. (CC) gcc options: -fvisibility=hidden -O2 -lm -lpng16 -ljpeg

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080pLinux 6.8Linux 6.648121620SE +/- 0.04, N = 3SE +/- 0.03, N = 317.7216.281. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Mount St. HelensLinux 6.8Linux 6.63691215SE +/- 0.15, N = 12SE +/- 0.15, N = 313.5413.141. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput TotalLinux 6.8Linux 6.62K4K6K8K10KSE +/- 26.02, N = 3SE +/- 20.83, N = 310095.49560.71. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Fishy Cat - Compute: CPU-OnlyLinux 6.8Linux 6.6816243240SE +/- 0.26, N = 3SE +/- 0.19, N = 336.7736.91

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 4 - Input: Bosphorus 4KLinux 6.8Linux 6.61.13582.27163.40744.54325.679SE +/- 0.008, N = 3SE +/- 0.014, N = 35.0484.6101. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080pLinux 6.8Linux 6.6918273645SE +/- 0.58, N = 3SE +/- 0.33, N = 841.1939.081. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: BMW27 - Compute: CPU-OnlyLinux 6.8Linux 6.6714212835SE +/- 0.27, N = 3SE +/- 0.13, N = 330.0229.82

WavPack Audio Encoding

This test times how long it takes to encode a sample WAV file to WavPack format with very high quality settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.7WAV To WavPackLinux 6.8Linux 6.6246810SE +/- 0.024, N = 25SE +/- 0.040, N = 55.4867.393

Timed FFmpeg Compilation

This test times how long it takes to build the FFmpeg multimedia library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 7.0Time To CompileLinux 6.8Linux 6.6714212835SE +/- 0.05, N = 3SE +/- 0.04, N = 327.6527.65

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.6Video Input: Bosphorus 4KLinux 6.8Linux 6.6612182430SE +/- 0.06, N = 3SE +/- 0.06, N = 325.4422.351. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

dav1d

Dav1d is an open-source, speedy AV1 video decoder supporting modern SIMD CPU features. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.4Video Input: Chimera 1080p 10-bitLinux 6.8Linux 6.680160240320400SE +/- 0.34, N = 3SE +/- 0.10, N = 3352.99351.911. (CC) gcc options: -pthread

JPEG-XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 100Linux 6.8Linux 6.6510152025SE +/- 0.09, N = 3SE +/- 0.08, N = 320.3919.081. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 1 - Model: ResNet-50Linux 6.8Linux 6.61.23982.47963.71944.95926.199SE +/- 0.00, N = 3SE +/- 0.01, N = 35.515.45

JPEG-XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 100Linux 6.8Linux 6.6510152025SE +/- 0.07, N = 3SE +/- 0.03, N = 320.5619.241. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

JPEG-XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL Decoding libjxl 0.10.1CPU Threads: AllLinux 6.8Linux 6.670140210280350SE +/- 1.49, N = 3SE +/- 1.39, N = 3339.65328.09

dav1d

Dav1d is an open-source, speedy AV1 video decoder supporting modern SIMD CPU features. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.4Video Input: Chimera 1080pLinux 6.8Linux 6.680160240320400SE +/- 0.07, N = 3SE +/- 0.14, N = 3375.76375.321. (CC) gcc options: -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.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.11Video Input: Bosphorus 1080p - Video Preset: FasterLinux 6.8Linux 6.6714212835SE +/- 0.10, N = 3SE +/- 0.08, N = 330.2428.721. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Tomographic ModelLinux 6.8Linux 6.648121620SE +/- 0.03, N = 3SE +/- 0.15, N = 514.3414.221. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Deconvolution Batch shapes_1d - Engine: CPULinux 6.8Linux 6.63691215SE +/- 0.10, N = 3SE +/- 0.07, N = 311.5211.37MIN: 10.6MIN: 6.491. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput ThreadLinux 6.8Linux 6.6306090120150SE +/- 0.00, N = 3SE +/- 0.00, N = 3112112MIN: 79.2MIN: 79.21. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Homogeneous HalfspaceLinux 6.8Linux 6.648121620SE +/- 0.07, N = 3SE +/- 0.19, N = 318.0417.581. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080pLinux 6.8Linux 6.620406080100SE +/- 0.82, N = 15SE +/- 0.22, N = 393.2887.711. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pLinux 6.8Linux 6.620406080100SE +/- 0.97, N = 15SE +/- 0.88, N = 395.0488.991. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pLinux 6.8Linux 6.620406080100SE +/- 1.26, N = 15SE +/- 0.90, N = 396.6096.101. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 11 Realtime - Input: Bosphorus 1080pLinux 6.8Linux 6.620406080100SE +/- 1.02, N = 15SE +/- 0.27, N = 399.6693.801. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080pLinux 6.8Linux 6.620406080100SE +/- 0.78, N = 15SE +/- 0.69, N = 399.9293.561. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

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.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ThoroughLinux 6.8Linux 6.6816243240SE +/- 0.07, N = 3SE +/- 0.09, N = 333.9333.951. (CXX) g++ options: -O3 -flto -pthread

uvg266

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 4K - Video Preset: Very FastLinux 6.8Linux 6.6816243240SE +/- 0.10, N = 3SE +/- 0.07, N = 335.1731.84

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 1080p - Video Preset: SlowLinux 6.8Linux 6.6816243240SE +/- 0.02, N = 3SE +/- 0.01, N = 333.6833.21

dav1d

Dav1d is an open-source, speedy AV1 video decoder supporting modern SIMD CPU features. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.4Video Input: Summer Nature 4KLinux 6.8Linux 6.650100150200250SE +/- 0.83, N = 3SE +/- 0.10, N = 3205.11205.481. (CC) gcc options: -pthread

uvg266

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 4K - Video Preset: Super FastLinux 6.8Linux 6.6816243240SE +/- 0.12, N = 3SE +/- 0.09, N = 335.7533.35

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput ThreadLinux 6.8Linux 6.6110220330440550SE +/- 9.38, N = 14SE +/- 0.20, N = 3476.5489.51. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080pLinux 6.8Linux 6.60.28130.56260.84391.12521.4065SE +/- 0.00, N = 3SE +/- 0.00, N = 31.251.191. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 8 - Input: Bosphorus 4KLinux 6.8Linux 6.61122334455SE +/- 0.22, N = 3SE +/- 0.13, N = 346.8344.301. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

WebP Image Encode

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: Quality 100, LosslessLinux 6.8Linux 6.60.33530.67061.00591.34121.6765SE +/- 0.02, N = 3SE +/- 0.00, N = 31.491.481. (CC) gcc options: -fvisibility=hidden -O2 -lm -lpng16 -ljpeg

uvg266

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 1080p - Video Preset: MediumLinux 6.8Linux 6.6918273645SE +/- 0.04, N = 3SE +/- 0.13, N = 337.2036.70

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 4K - Video Preset: Ultra FastLinux 6.8Linux 6.6918273645SE +/- 0.11, N = 3SE +/- 0.06, N = 338.9035.96

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KLinux 6.8Linux 6.61020304050SE +/- 0.13, N = 3SE +/- 0.25, N = 343.5841.341. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Timed Mesa Compilation

This test profile times how long it takes to compile Mesa with Meson/Ninja. For minimizing build dependencies and avoid versioning conflicts, test this is just the core Mesa build without LLVM or the extra Gallium3D/Mesa drivers enabled. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 24.0Time To CompileLinux 6.8Linux 6.648121620SE +/- 0.03, N = 3SE +/- 0.04, N = 315.0115.09

C-Ray

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 2.0Resolution: 1080p - Rays Per Pixel: 16Linux 6.8Linux 6.648121620SE +/- 0.05, N = 3SE +/- 0.03, N = 314.8814.761. (CC) gcc options: -lpthread -lm

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 4 - Input: Bosphorus 1080pLinux 6.8Linux 6.63691215SE +/- 0.06, N = 3SE +/- 0.05, N = 313.5312.281. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.9Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4KLinux 6.8Linux 6.61122334455SE +/- 0.27, N = 3SE +/- 0.06, N = 349.6147.781. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.6Video Input: Bosphorus 1080pLinux 6.8Linux 6.61122334455SE +/- 0.03, N = 3SE +/- 0.03, N = 349.2747.041. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

RNNoise

RNNoise is a recurrent neural network for audio noise reduction developed by Mozilla and Xiph.Org. This test profile is a single-threaded test measuring the time to denoise a sample 26 minute long 16-bit RAW audio file using this recurrent neural network noise suppression library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 0.2Input: 26 Minute Long Talking SampleLinux 6.8Linux 6.63691215SE +/- 0.07, N = 3SE +/- 0.05, N = 311.6211.671. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

Parallel BZIP2 Compression

This test measures the time needed to compress a file (FreeBSD-13.0-RELEASE-amd64-memstick.img) using Parallel BZIP2 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img CompressionLinux 6.8Linux 6.60.48240.96481.44721.92962.412SE +/- 0.054103, N = 14SE +/- 0.060783, N = 152.1440212.1188711. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Church FacadeLinux 6.8Linux 6.62K4K6K8K10KSE +/- 70.28, N = 3SE +/- 18.18, N = 3864185451. (CXX) g++ options: -O3

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.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: FastLinux 6.8Linux 6.6140280420560700SE +/- 0.92, N = 3SE +/- 2.79, N = 3666.85662.731. (CXX) g++ options: -O3 -flto -pthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: IP Shapes 3D - Engine: CPULinux 6.8Linux 6.6246810SE +/- 0.00709, N = 3SE +/- 0.00967, N = 37.738047.72941MIN: 7.62MIN: 7.621. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 8 - Input: Bosphorus 1080pLinux 6.8Linux 6.620406080100SE +/- 0.53, N = 3SE +/- 0.15, N = 390.4288.161. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Convolution Batch Shapes Auto - Engine: CPULinux 6.8Linux 6.60.21560.43120.64680.86241.078SE +/- 0.009774, N = 5SE +/- 0.010842, N = 30.9491720.958085MIN: 0.86MIN: 0.871. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 12 - Input: Bosphorus 4KLinux 6.8Linux 6.620406080100SE +/- 0.81, N = 3SE +/- 0.92, N = 3104.23102.501. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 4KLinux 6.8Linux 6.620406080100SE +/- 1.12, N = 3SE +/- 1.04, N = 3106.11102.531. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: MediumLinux 6.8Linux 6.660120180240300SE +/- 0.24, N = 3SE +/- 0.27, N = 3262.83262.801. (CXX) g++ options: -O3 -flto -pthread

dav1d

Dav1d is an open-source, speedy AV1 video decoder supporting modern SIMD CPU features. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.4Video Input: Summer Nature 1080pLinux 6.8Linux 6.6110220330440550SE +/- 0.43, N = 3SE +/- 0.66, N = 3482.94487.051. (CC) gcc options: -pthread

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: LionLinux 6.8Linux 6.612002400360048006000SE +/- 7.26, N = 3SE +/- 9.67, N = 3554755051. (CXX) g++ options: -O3

WebP Image Encode

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: Quality 100, Highest CompressionLinux 6.8Linux 6.60.76281.52562.28843.05123.814SE +/- 0.00, N = 3SE +/- 0.00, N = 33.393.361. (CC) gcc options: -fvisibility=hidden -O2 -lm -lpng16 -ljpeg

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 6, LosslessLinux 6.8Linux 6.6246810SE +/- 0.055, N = 3SE +/- 0.016, N = 37.2127.1911. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 10, LosslessLinux 6.8Linux 6.61.27512.55023.82535.10046.3755SE +/- 0.009, N = 3SE +/- 0.016, N = 35.6105.6671. (CXX) g++ options: -O3 -fPIC -lm

uvg266

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 1080p - Video Preset: Very FastLinux 6.8Linux 6.6306090120150SE +/- 0.50, N = 3SE +/- 0.03, N = 3115.03112.86

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 1080p - Video Preset: Super FastLinux 6.8Linux 6.6306090120150SE +/- 0.78, N = 3SE +/- 0.33, N = 3129.52126.57

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.8.0Video Input: Bosphorus 1080p - Video Preset: Ultra FastLinux 6.8Linux 6.6306090120150SE +/- 0.57, N = 3SE +/- 0.27, N = 3146.79143.34

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e12Linux 6.8Linux 6.60.80061.60122.40183.20244.003SE +/- 0.006, N = 3SE +/- 0.016, N = 33.5473.5581. (CXX) g++ options: -O3

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 6Linux 6.8Linux 6.60.77691.55382.33073.10763.8845SE +/- 0.007, N = 3SE +/- 0.004, N = 33.4413.4531. (CXX) g++ options: -O3 -fPIC -lm

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Deconvolution Batch shapes_3d - Engine: CPULinux 6.8Linux 6.60.48950.9791.46851.9582.4475SE +/- 0.00666, N = 3SE +/- 0.01366, N = 32.175542.15953MIN: 2.09MIN: 2.091. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 12 - Input: Bosphorus 1080pLinux 6.8Linux 6.670140210280350SE +/- 1.38, N = 3SE +/- 2.48, N = 3311.07308.171. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 1080pLinux 6.8Linux 6.680160240320400SE +/- 0.40, N = 3SE +/- 0.46, N = 3373.13373.791. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

WebP Image Encode

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: Quality 100Linux 6.8Linux 6.63691215SE +/- 0.03, N = 3SE +/- 0.04, N = 310.4510.431. (CC) gcc options: -fvisibility=hidden -O2 -lm -lpng16 -ljpeg

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: DefaultLinux 6.8Linux 6.648121620SE +/- 0.04, N = 3SE +/- 0.01, N = 317.3217.161. (CC) gcc options: -fvisibility=hidden -O2 -lm -lpng16 -ljpeg

206 Results Shown

TensorFlow
Stockfish
Graph500:
  26:
    sssp max_TEPS
    sssp median_TEPS
    bfs max_TEPS
    bfs median_TEPS
Timed Linux Kernel Compilation
OpenSSL:
  RSA4096:
    verify/s
    sign/s
Blender
RocksDB
FFmpeg:
  libx264 - Upload
  libx265 - Video On Demand
  libx265 - Platform
Timed Node.js Compilation
FFmpeg
BRL-CAD
PyTorch
JPEG-XL libjxl:
  PNG - 80
  JPEG - 80
FFmpeg:
  libx264 - Video On Demand
  libx264 - Platform
OSPRay
OpenSSL:
  SHA256
  ChaCha20-Poly1305
  SHA512
  AES-256-GCM
  ChaCha20
  AES-128-GCM
OSPRay Studio
OSPRay
SPECFEM3D
srsRAN Project
JPEG-XL libjxl
OSPRay Studio
VVenC
OSPRay Studio:
  2 - 4K - 32 - Path Tracer - CPU
  1 - 4K - 32 - Path Tracer - CPU
C-Ray
Timed Linux Kernel Compilation
OSPRay Studio:
  2 - 1080p - 16 - Path Tracer - CPU
  3 - 1080p - 16 - Path Tracer - CPU
  1 - 1080p - 16 - Path Tracer - CPU
OSPRay
FFmpeg
Blender
AOM AV1
Blender
Chaos Group V-RAY
oneDNN
OSPRay Studio
oneDNN
libavif avifenc
OSPRay Studio
oneDNN
OSPRay Studio:
  2 - 4K - 1 - Path Tracer - CPU
  3 - 4K - 16 - Path Tracer - CPU
AOM AV1
OpenVINO:
  Face Detection FP16-INT8 - CPU:
    ms
    FPS
OSPRay
OSPRay Studio:
  3 - 1080p - 1 - Path Tracer - CPU
  1 - 1080p - 1 - Path Tracer - CPU
  2 - 1080p - 1 - Path Tracer - CPU
OpenVINO:
  Face Detection FP16 - CPU:
    ms
    FPS
OSPRay Studio
OSPRay
OSPRay Studio
OpenVINO:
  Person Detection FP32 - CPU:
    ms
    FPS
  Person Detection FP16 - CPU:
    ms
    FPS
  Machine Translation EN To DE FP16 - CPU:
    ms
    FPS
  Road Segmentation ADAS FP16-INT8 - CPU:
    ms
    FPS
  Person Vehicle Bike Detection FP16 - CPU:
    ms
    FPS
  Road Segmentation ADAS FP16 - CPU:
    ms
    FPS
  Noise Suppression Poconet-Like FP16 - CPU:
    ms
    FPS
  Person Re-Identification Retail FP16 - CPU:
    ms
    FPS
  Handwritten English Recognition FP16 - CPU:
    ms
    FPS
  Handwritten English Recognition FP16-INT8 - CPU:
    ms
    FPS
  Vehicle Detection FP16-INT8 - CPU:
    ms
    FPS
  Face Detection Retail FP16-INT8 - CPU:
    ms
    FPS
  Vehicle Detection FP16 - CPU:
    ms
    FPS
  Face Detection Retail FP16 - CPU:
    ms
    FPS
  Weld Porosity Detection FP16-INT8 - CPU:
    ms
    FPS
  Age Gender Recognition Retail 0013 FP16 - CPU:
    ms
    FPS
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU:
    ms
    FPS
  Weld Porosity Detection FP16 - CPU:
    ms
    FPS
RocksDB
GraphicsMagick:
  Resizing
  Rotate
RocksDB
GraphicsMagick:
  Sharpen
  Noise-Gaussian
  HWB Color Space
  Enhanced
RocksDB
GraphicsMagick
OSPRay
C-Ray
uvg266
ASTC Encoder:
  Very Thorough
  Exhaustive
VVenC
SPECFEM3D
uvg266
FFmpeg
VVenC
AOM AV1
JPEG-XL libjxl
OSPRay Studio
Primesieve
JPEG-XL Decoding libjxl
Blender
AOM AV1:
  Speed 6 Realtime - Bosphorus 4K
  Speed 9 Realtime - Bosphorus 4K
OSPRay Studio
AOM AV1
Timed PHP Compilation
libavif avifenc
WebP Image Encode
AOM AV1
SPECFEM3D
srsRAN Project
Blender
SVT-AV1
AOM AV1
Blender
WavPack Audio Encoding
Timed FFmpeg Compilation
x265
dav1d
JPEG-XL libjxl
TensorFlow
JPEG-XL libjxl
JPEG-XL Decoding libjxl
dav1d
VVenC
SPECFEM3D
oneDNN
srsRAN Project
SPECFEM3D
AOM AV1:
  Speed 6 Realtime - Bosphorus 1080p
  Speed 8 Realtime - Bosphorus 1080p
  Speed 9 Realtime - Bosphorus 1080p
  Speed 11 Realtime - Bosphorus 1080p
  Speed 10 Realtime - Bosphorus 1080p
ASTC Encoder
uvg266:
  Bosphorus 4K - Very Fast
  Bosphorus 1080p - Slow
dav1d
uvg266
srsRAN Project
AOM AV1
SVT-AV1
WebP Image Encode
uvg266:
  Bosphorus 1080p - Medium
  Bosphorus 4K - Ultra Fast
AOM AV1
Timed Mesa Compilation
C-Ray
SVT-AV1
AOM AV1
x265
RNNoise
Parallel BZIP2 Compression
Google Draco
ASTC Encoder
oneDNN
SVT-AV1
oneDNN
SVT-AV1:
  Preset 12 - Bosphorus 4K
  Preset 13 - Bosphorus 4K
ASTC Encoder
dav1d
Google Draco
WebP Image Encode
libavif avifenc:
  6, Lossless
  10, Lossless
uvg266:
  Bosphorus 1080p - Very Fast
  Bosphorus 1080p - Super Fast
  Bosphorus 1080p - Ultra Fast
Primesieve
libavif avifenc
oneDNN
SVT-AV1:
  Preset 12 - Bosphorus 1080p
  Preset 13 - Bosphorus 1080p
WebP Image Encode:
  Quality 100
  Default