AMD Ryzen 9 3900XT - Ubuntu 19.10 vs. Ubuntu 21.10

AMD Ryzen 9 3900XT Zen 2 comparison after two years of Ubuntu Linux... Benchmarks for a future article on Phoronix by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2110109-TJ-UBUNTUZEN05
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

Limit displaying results to tests within:

AV1 2 Tests
Web Browsers 1 Tests
Chess Test Suite 2 Tests
C/C++ Compiler Tests 13 Tests
Compression Tests 3 Tests
CPU Massive 20 Tests
Creator Workloads 15 Tests
Desktop Graphics 2 Tests
Encoding 7 Tests
Fortran Tests 2 Tests
HPC - High Performance Computing 7 Tests
Common Kernel Benchmarks 2 Tests
Machine Learning 2 Tests
Molecular Dynamics 3 Tests
MPI Benchmarks 2 Tests
Multi-Core 19 Tests
NVIDIA GPU Compute 5 Tests
OpenMPI Tests 3 Tests
Programmer / Developer System Benchmarks 6 Tests
Python 2 Tests
Renderers 5 Tests
Scientific Computing 4 Tests
Server 3 Tests
Server CPU Tests 15 Tests
Single-Threaded 4 Tests
Video Encoding 6 Tests
Common Workstation Benchmarks 4 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
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Ubuntu 19.10
October 09 2021
  7 Hours, 37 Minutes
Ubuntu 21.10
October 10 2021
  7 Hours, 2 Minutes
Invert Hiding All Results Option
  7 Hours, 19 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):


AMD Ryzen 9 3900XT - Ubuntu 19.10 vs. Ubuntu 21.10ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionUbuntu 19.10Ubuntu 21.10AMD Ryzen 9 3900XT 12-Core @ 3.80GHz (12 Cores / 24 Threads)MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS)AMD Starship/Matisse16GB500GB Seagate FireCuda 520 SSD ZP500GM30002AMD Radeon RX 56/64 8GB (1630/945MHz)AMD Vega 10 HDMI AudioASUS MG28URealtek Device 2600 + Realtek Device 3000 + Intel Device 2723Ubuntu 19.105.3.0-18-generic (x86_64)GNOME Shell 3.34.1X Server 1.20.54.5 Mesa 19.2.1 (LLVM 9.0.0)1.1.107GCC 9.2.1 20191008ext43840x2160500GB Seagate FireCuda 520 SSD ZP500GM30002 + 0GB Ultra LuxeAMD Radeon RX 56/64 (1630/945MHz)Realtek Device 2600 + Realtek Killer E3000 2.5GbE + Intel Wi-Fi 6 AX200Ubuntu 21.105.13.0-16-generic (x86_64)GNOME Shell 40.5X Server + Wayland1.2.182GCC 11.2.0OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- Ubuntu 19.10: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Ubuntu 21.10: --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-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Ubuntu 19.10: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8701021- Ubuntu 21.10: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021Graphics Details- Ubuntu 19.10: GLAMOR - BAR1 / Visible vRAM Size: 256 MBJava Details- Ubuntu 19.10: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- Ubuntu 21.10: OpenJDK Runtime Environment (build 11.0.12+7-Ubuntu-0ubuntu3)Python Details- Ubuntu 19.10: Python 3.7.5- Ubuntu 21.10: Python 3.9.7Security Details- Ubuntu 19.10: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling - Ubuntu 21.10: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Ubuntu 19.10 vs. Ubuntu 21.10 ComparisonPhoronix Test SuiteBaseline+77%+77%+154%+154%+231%+231%307.8%183.3%182.2%173.2%157.7%147.2%140.8%120.3%111.8%105.9%88.6%86.1%81.6%78.6%73.7%54.8%45.5%34.5%24.9%21.2%21%19.5%19.5%19.5%19.1%18.6%16%14.4%14%10.4%9.2%8.9%8%7.7%7.4%7.4%7.3%7.3%7.3%7.1%7.1%6.9%6.8%6.3%5.9%5.9%5.7%5.5%5.4%5.4%5.3%5.1%4.9%4.8%4.6%4.5%4.2%4.2%4.1%3.7%3.7%3.5%3.4%3.4%3.3%2.6%2.4%2.3%2.3%2.1%2.1%Vulkan GPU - vgg16Vulkan GPU - resnet50Vulkan GPU - googlenetVulkan GPU - resnet18Vulkan GPU - alexnetVulkan GPU - squeezenet_ssdVulkan GPU - mnasnetVulkan GPU-v2-v2 - mobilenet-v2Vulkan GPU-v3-v3 - mobilenet-v3Vulkan GPU - regnety_400mpython_startup1920 x 1080Vulkan GPU - yolov4-tinyVulkan GPU - shufflenet-v2Vulkan GPU - efficientnet-b0Speedometer - FirefoxVulkan GPU - mobilenetOpenGL 1.x - 3840 x 2160OpenGL 1.x - 1920 x 1080django_templateVoiceMark_100Vulkan GPU - blazefaceARES-6 - FirefoxStyleBench - FirefoxSpeed 5 - Bosphorus 1080p1920 x 1080mobilenetV33840 x 2160Scala DottyCPU - regnety_400mJetstream 2 - Firefox9.3%pathlibP.B.SOpenGL 3.x - 1920 x 1080OpenGL 3.x - 3840 x 2160pickle_pure_python3840 x 2160 - HighMobileNetV2_224regex_compileCPU-v3-v3 - mobilenet-v3Speed 6 Realtime - Bosphorus 4K3840 x 2160CPU - efficientnet-b08 - D.SCPU-v2-v2 - mobilenet-v2CPU - mnasnetS.F.P.R3840 x 2160 - Ultra19 - D.SSpeed 5 - Bosphorus 4KWAV To FLACPreset 8 - Bosphorus 4K5.3%CPU - squeezenet_ssdSHA2565.2%PNG - 5squeezenetv1.17 - Bosphorus 1080pRand Read4.8%chaosCPU - yolov4-tiny10 - Bosphorus 1080pCPU - googlenetTime To Compile3840 x 2160 - UltimateT.T.C.C.G.CT.F.A.T.TCPU - resnet50SqueezeNetV1.0CPU - resnet18CPU - shufflenet-v2CPU - alexnetCPU - mobilenet8 - Compression Speed2to3resnet-v2-50NCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNPyPerformanceTesseractNCNNNCNNNCNNSeleniumNCNNyquake2yquake2PyPerformanceGoogle SynthMarkNCNNSeleniumSeleniumVP9 libvpx EncodingET: LegacyMobile Neural NetworkET: LegacyRenaissanceNCNNSeleniumPyPerformancePHPBenchyquake2yquake2PyPerformanceXonoticMobile Neural NetworkPyPerformanceNCNNAOM AV1TesseractNCNNZstd CompressionNCNNNCNNACES DGEMMXonoticZstd CompressionVP9 libvpx EncodingFLAC Audio EncodingSVT-AV1NCNNOpenSSLJPEG XL libjxlMobile Neural NetworkSVT-HEVCFacebook RocksDBPyPerformanceNCNNSVT-HEVCNCNNTimed FFmpeg CompilationXonoticGitPyBenchNCNNMobile Neural NetworkNCNNNCNNNCNNNCNNZstd CompressionPyPerformanceMobile Neural NetworkUbuntu 19.10Ubuntu 21.10

AMD Ryzen 9 3900XT - Ubuntu 19.10 vs. Ubuntu 21.10ncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - regnety_400mpyperformance: python_startuptesseract: 1920 x 1080ncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU - efficientnet-b0selenium: Speedometer - Firefoxncnn: Vulkan GPU - mobilenetyquake2: OpenGL 1.x - 3840 x 2160yquake2: OpenGL 1.x - 1920 x 1080pyperformance: django_templatesynthmark: VoiceMark_100ncnn: Vulkan GPU - blazefaceselenium: ARES-6 - Firefoxselenium: StyleBench - Firefoxvpxenc: Speed 5 - Bosphorus 1080petlegacy: 1920 x 1080mnn: mobilenetV3etlegacy: 3840 x 2160renaissance: Scala Dottyncnn: CPU - regnety_400mselenium: Jetstream 2 - Firefoxpyperformance: pathlibphpbench: PHP Benchmark Suiteyquake2: OpenGL 3.x - 1920 x 1080yquake2: OpenGL 3.x - 3840 x 2160pyperformance: pickle_pure_pythonxonotic: 3840 x 2160 - Highmnn: MobileNetV2_224pyperformance: regex_compilencnn: CPU-v3-v3 - mobilenet-v3aom-av1: Speed 6 Realtime - Bosphorus 4Ktesseract: 3840 x 2160ncnn: CPU - efficientnet-b0compress-zstd: 8 - Decompression Speedncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - mnasnetmt-dgemm: Sustained Floating-Point Ratexonotic: 3840 x 2160 - Ultracompress-zstd: 19 - Decompression Speedvpxenc: Speed 5 - Bosphorus 4Kencode-flac: WAV To FLACsvt-av1: Preset 8 - Bosphorus 4Kncnn: CPU - squeezenet_ssdopenssl: SHA256jpegxl: PNG - 5mnn: squeezenetv1.1svt-hevc: 7 - Bosphorus 1080procksdb: Rand Readpyperformance: chaosncnn: CPU - yolov4-tinysvt-hevc: 10 - Bosphorus 1080pncnn: CPU - googlenetbuild-ffmpeg: Time To Compilexonotic: 3840 x 2160 - Ultimategit: Time To Complete Common Git Commandspybench: Total For Average Test Timesncnn: CPU - resnet50mnn: SqueezeNetV1.0ncnn: CPU - resnet18ncnn: CPU - shufflenet-v2ncnn: CPU - alexnetncnn: CPU - mobilenetcompress-zstd: 8 - Compression Speedpyperformance: 2to3mnn: resnet-v2-50ncnn: CPU - blazefacex265: Bosphorus 4Kmnn: inception-v3aom-av1: Speed 9 Realtime - Bosphorus 4Kmnn: mobilenet-v1-1.0compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9npb: LU.Csvt-av1: Preset 4 - Bosphorus 4Kappleseed: Disney Materialstockfish: Total Timenpb: CG.Cjpegxl: PNG - 8x265: Bosphorus 1080psvt-hevc: 1 - Bosphorus 1080ppyperformance: gobrl-cad: VGR Performance Metricv-ray: CPUcompress-zstd: 19 - Compression Speednpb: SP.Cblender: BMW27 - CPU-Onlyindigobench: CPU - Supercarselenium: PSPDFKit WASM - Firefoxappleseed: Material Testertachyon: Total Timecloverleaf: Lagrangian-Eulerian Hydrodynamicscompress-lz4: 3 - Decompression Speedasmfish: 1024 Hash Memory, 26 Depthrocksdb: Rand Fillgromacs: MPI CPU - water_GMX50_barencnn: CPU - vgg16svt-vp9: PSNR/SSIM Optimized - Bosphorus 1080pnamd: ATPase Simulation - 327,506 Atomscompress-lz4: 3 - Compression Speedopenssl: RSA4096svt-av1: Preset 4 - Bosphorus 1080popenssl: RSA4096npb: SP.Bblender: Classroom - CPU-Onlynpb: MG.Cindigobench: CPU - Bedroomappleseed: Emilysvt-av1: Preset 8 - Bosphorus 1080pUbuntu 19.10Ubuntu 21.1038.9017.4516.176.4210.3612.787.326.397.929.3511.9428.510220.214.5015.79102.111.84644.7753.349.8670.4741.9048.1287.926.65454.52.936444.0805.611.90102.25416.7696173908.9916.8422435.68408824.5861624.5815.43318.17356.863806.15.204.634.186831381.73364973323.213.8213.94018.43318.831792540747046.704.825151.577218434810326.20278.2916.5338.796293.316416045.78992126.496.80316.375.0612.5916.001282.628634.5122.1119.5831.65049.395.49525.42724905.471.648156.85189392643316995.270.8961.5811.712231908981441537.55261.57105.076.1443078158.00832660.1184134.3810543.3397879369737181.13159.82206.451.4980960.88229876.35.2273551.98909.78303.0613431.142.882260.6349764.6759.546.165.732.354.025.173.042.903.744.546.31797.329411.132.529.091588.14867.3941.041.1811.3951.5940.2810531.74539.12.532508.1706.710.7893.58915.3757984981.9987.5393467.78811844.2741514.2716.53340.68716.424066.44.894.374.434919403.41326693507.214.5713.22517.50817.891703657656849.074.600158.836888664098.525.07290.0515.8737.269304.151487844.16189025.616.57815.854.9312.3015.641312.128033.7932.0719.2431.10748.595.41225.79425245.581.669158.820565387888827075.550.960.9211.592211892361429737.25298.41105.766.1823097157.15241359.8075133.7210593.8399477439775271.12859.96206.061.4955160.94230097.15.2233554.48914.81303.2213425.582.881260.55074464.657OpenBenchmarking.org

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: vgg16Ubuntu 19.10Ubuntu 21.10918273645SE +/- 0.03, N = 3SE +/- 0.01, N = 338.909.54-pthread - MIN: 37.85 / MAX: 53.87MIN: 9.44 / MAX: 14.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: vgg16Ubuntu 19.10Ubuntu 21.10816243240Min: 38.84 / Avg: 38.9 / Max: 38.95Min: 9.52 / Avg: 9.54 / Max: 9.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet50Ubuntu 19.10Ubuntu 21.1048121620SE +/- 0.03, N = 3SE +/- 0.01, N = 317.456.16-pthread - MIN: 17.15 / MAX: 28MIN: 6.11 / MAX: 10.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet50Ubuntu 19.10Ubuntu 21.1048121620Min: 17.4 / Avg: 17.45 / Max: 17.51Min: 6.15 / Avg: 6.16 / Max: 6.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: googlenetUbuntu 19.10Ubuntu 21.1048121620SE +/- 0.07, N = 3SE +/- 0.00, N = 316.175.73-pthread - MIN: 15.65 / MAX: 31.04MIN: 5.69 / MAX: 9.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: googlenetUbuntu 19.10Ubuntu 21.1048121620Min: 16.06 / Avg: 16.17 / Max: 16.3Min: 5.73 / Avg: 5.73 / Max: 5.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet18Ubuntu 19.10Ubuntu 21.10246810SE +/- 0.04, N = 2SE +/- 0.01, N = 36.422.35-pthread - MIN: 6.31 / MAX: 11.17MIN: 2.18 / MAX: 3.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet18Ubuntu 19.10Ubuntu 21.103691215Min: 6.37 / Avg: 6.42 / Max: 6.46Min: 2.34 / Avg: 2.35 / Max: 2.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: alexnetUbuntu 19.10Ubuntu 21.103691215SE +/- 0.02, N = 3SE +/- 0.00, N = 310.364.02-pthread - MIN: 10.02 / MAX: 16.1MIN: 3.93 / MAX: 6.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: alexnetUbuntu 19.10Ubuntu 21.103691215Min: 10.33 / Avg: 10.36 / Max: 10.39Min: 4.02 / Avg: 4.02 / Max: 4.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: squeezenet_ssdUbuntu 19.10Ubuntu 21.103691215SE +/- 0.12, N = 3SE +/- 0.01, N = 312.785.17-pthread - MIN: 11.57 / MAX: 30.19MIN: 4.92 / MAX: 8.41. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: squeezenet_ssdUbuntu 19.10Ubuntu 21.1048121620Min: 12.54 / Avg: 12.78 / Max: 12.91Min: 5.16 / Avg: 5.17 / Max: 5.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mnasnetUbuntu 19.10Ubuntu 21.10246810SE +/- 0.08, N = 3SE +/- 0.00, N = 37.323.04-pthread - MIN: 6.89 / MAX: 18.1MIN: 2.96 / MAX: 4.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mnasnetUbuntu 19.10Ubuntu 21.103691215Min: 7.18 / Avg: 7.32 / Max: 7.44Min: 3.04 / Avg: 3.04 / Max: 3.051. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2Ubuntu 19.10Ubuntu 21.10246810SE +/- 0.01, N = 3SE +/- 0.00, N = 36.392.90-pthread - MIN: 6.3 / MAX: 11.06MIN: 2.81 / MAX: 3.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2Ubuntu 19.10Ubuntu 21.103691215Min: 6.37 / Avg: 6.39 / Max: 6.41Min: 2.9 / Avg: 2.9 / Max: 2.911. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3Ubuntu 19.10Ubuntu 21.10246810SE +/- 0.06, N = 3SE +/- 0.00, N = 37.923.74-pthread - MIN: 7.59 / MAX: 19.97MIN: 3.66 / MAX: 4.911. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3Ubuntu 19.10Ubuntu 21.103691215Min: 7.82 / Avg: 7.92 / Max: 8.03Min: 3.74 / Avg: 3.74 / Max: 3.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: regnety_400mUbuntu 19.10Ubuntu 21.103691215SE +/- 0.03, N = 3SE +/- 0.00, N = 39.354.54-pthread - MIN: 8.92 / MAX: 20MIN: 4.49 / MAX: 5.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: regnety_400mUbuntu 19.10Ubuntu 21.103691215Min: 9.3 / Avg: 9.35 / Max: 9.41Min: 4.54 / Avg: 4.54 / Max: 4.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: python_startupUbuntu 19.10Ubuntu 21.103691215SE +/- 0.00, N = 3SE +/- 0.00, N = 311.906.31
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: python_startupUbuntu 19.10Ubuntu 21.103691215Min: 11.9 / Avg: 11.9 / Max: 11.9Min: 6.3 / Avg: 6.31 / Max: 6.31

Tesseract

Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 1920 x 1080Ubuntu 19.10Ubuntu 21.102004006008001000SE +/- 3.53, N = 3SE +/- 9.89, N = 4428.51797.33
OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 1920 x 1080Ubuntu 19.10Ubuntu 21.10140280420560700Min: 421.57 / Avg: 428.51 / Max: 433.11Min: 779.36 / Avg: 797.33 / Max: 820.47

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: yolov4-tinyUbuntu 19.10Ubuntu 21.10510152025SE +/- 0.12, N = 3SE +/- 0.01, N = 320.2111.13-pthread - MIN: 17.6 / MAX: 34.64MIN: 11 / MAX: 11.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: yolov4-tinyUbuntu 19.10Ubuntu 21.10510152025Min: 19.97 / Avg: 20.21 / Max: 20.39Min: 11.12 / Avg: 11.13 / Max: 11.161. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: shufflenet-v2Ubuntu 19.10Ubuntu 21.101.01252.0253.03754.055.0625SE +/- 0.01, N = 3SE +/- 0.00, N = 34.502.52-pthread - MIN: 4.43 / MAX: 9.79MIN: 2.34 / MAX: 3.41. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: shufflenet-v2Ubuntu 19.10Ubuntu 21.10246810Min: 4.48 / Avg: 4.5 / Max: 4.53Min: 2.52 / Avg: 2.52 / Max: 2.531. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: efficientnet-b0Ubuntu 19.10Ubuntu 21.1048121620SE +/- 0.09, N = 3SE +/- 0.02, N = 315.799.09-pthread - MIN: 15.18 / MAX: 30.16MIN: 8.94 / MAX: 13.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: efficientnet-b0Ubuntu 19.10Ubuntu 21.1048121620Min: 15.64 / Avg: 15.79 / Max: 15.94Min: 9.06 / Avg: 9.09 / Max: 9.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers such as Firefox and Google Chrome. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRuns Per Minute, More Is BetterSeleniumBenchmark: Speedometer - Browser: FirefoxUbuntu 19.10Ubuntu 21.10306090120150SE +/- 0.64, N = 3SE +/- 0.88, N = 3102.1158.01. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0
OpenBenchmarking.orgRuns Per Minute, More Is BetterSeleniumBenchmark: Speedometer - Browser: FirefoxUbuntu 19.10Ubuntu 21.10306090120150Min: 101 / Avg: 102.13 / Max: 103.2Min: 156 / Avg: 157.67 / Max: 1591. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mobilenetUbuntu 19.10Ubuntu 21.103691215SE +/- 0.00, N = 3SE +/- 0.07, N = 311.848.14-pthread - MIN: 10.48 / MAX: 24.3MIN: 7.94 / MAX: 15.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mobilenetUbuntu 19.10Ubuntu 21.103691215Min: 11.83 / Avg: 11.84 / Max: 11.84Min: 8.03 / Avg: 8.14 / Max: 8.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 8.0Renderer: OpenGL 1.x - Resolution: 3840 x 2160Ubuntu 19.10Ubuntu 21.102004006008001000SE +/- 1.71, N = 3SE +/- 3.10, N = 3644.7867.31. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 8.0Renderer: OpenGL 1.x - Resolution: 3840 x 2160Ubuntu 19.10Ubuntu 21.10150300450600750Min: 641.6 / Avg: 644.67 / Max: 647.5Min: 862.2 / Avg: 867.33 / Max: 872.91. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 8.0Renderer: OpenGL 1.x - Resolution: 1920 x 1080Ubuntu 19.10Ubuntu 21.102004006008001000SE +/- 3.60, N = 3SE +/- 3.27, N = 3753.3941.01. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 8.0Renderer: OpenGL 1.x - Resolution: 1920 x 1080Ubuntu 19.10Ubuntu 21.10170340510680850Min: 746.2 / Avg: 753.33 / Max: 757.8Min: 937.7 / Avg: 940.97 / Max: 947.51. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: django_templateUbuntu 19.10Ubuntu 21.101122334455SE +/- 0.19, N = 3SE +/- 0.06, N = 349.841.1
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: django_templateUbuntu 19.10Ubuntu 21.101020304050Min: 49.4 / Avg: 49.77 / Max: 50Min: 41 / Avg: 41.1 / Max: 41.2

Google SynthMark

SynthMark is a cross platform tool for benchmarking CPU performance under a variety of real-time audio workloads. It uses a polyphonic synthesizer model to provide standardized tests for latency, jitter and computational throughput. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100Ubuntu 19.10Ubuntu 21.102004006008001000SE +/- 2.13, N = 3SE +/- 10.31, N = 3670.47811.401. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast
OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100Ubuntu 19.10Ubuntu 21.10140280420560700Min: 667.24 / Avg: 670.47 / Max: 674.49Min: 793.28 / Avg: 811.4 / Max: 828.991. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: blazefaceUbuntu 19.10Ubuntu 21.100.42750.8551.28251.712.1375SE +/- 0.03, N = 3SE +/- 0.01, N = 31.901.59-pthread - MIN: 1.67 / MAX: 2.34MIN: 1.5 / MAX: 2.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: blazefaceUbuntu 19.10Ubuntu 21.10246810Min: 1.85 / Avg: 1.9 / Max: 1.94Min: 1.57 / Avg: 1.59 / Max: 1.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers such as Firefox and Google Chrome. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: ARES-6 - Browser: FirefoxUbuntu 19.10Ubuntu 21.101122334455SE +/- 0.06, N = 3SE +/- 0.26, N = 348.1240.281. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0
OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: ARES-6 - Browser: FirefoxUbuntu 19.10Ubuntu 21.101020304050Min: 48.01 / Avg: 48.12 / Max: 48.18Min: 39.85 / Avg: 40.28 / Max: 40.761. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0

OpenBenchmarking.orgRuns / Minute, More Is BetterSeleniumBenchmark: StyleBench - Browser: FirefoxUbuntu 19.10Ubuntu 21.1020406080100SE +/- 0.20, N = 3SE +/- 0.67, N = 387.9105.01. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0
OpenBenchmarking.orgRuns / Minute, More Is BetterSeleniumBenchmark: StyleBench - Browser: FirefoxUbuntu 19.10Ubuntu 21.1020406080100Min: 87.6 / Avg: 87.93 / Max: 88.3Min: 104 / Avg: 104.67 / Max: 1061. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.10714212835SE +/- 0.17, N = 3SE +/- 0.12, N = 326.6531.741. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.10714212835Min: 26.39 / Avg: 26.65 / Max: 26.97Min: 31.57 / Avg: 31.74 / Max: 31.981. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

ET: Legacy

ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.78Resolution: 1920 x 1080Ubuntu 19.10Ubuntu 21.10120240360480600SE +/- 5.72, N = 3SE +/- 3.81, N = 15454.5539.1
OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.78Resolution: 1920 x 1080Ubuntu 19.10Ubuntu 21.10100200300400500Min: 443.2 / Avg: 454.53 / Max: 461.5Min: 510.3 / Avg: 539.11 / Max: 560.3

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenetV3Ubuntu 19.10Ubuntu 21.100.66061.32121.98182.64243.303SE +/- 0.004, N = 3SE +/- 0.024, N = 32.9362.532MIN: 2.89 / MAX: 5.54MIN: 2.45 / MAX: 11.61. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenetV3Ubuntu 19.10Ubuntu 21.10246810Min: 2.93 / Avg: 2.94 / Max: 2.94Min: 2.48 / Avg: 2.53 / Max: 2.561. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

ET: Legacy

ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.78Resolution: 3840 x 2160Ubuntu 19.10Ubuntu 21.10110220330440550SE +/- 5.26, N = 3SE +/- 2.59, N = 3444.0508.1
OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.78Resolution: 3840 x 2160Ubuntu 19.10Ubuntu 21.1090180270360450Min: 433.5 / Avg: 444 / Max: 449.9Min: 504.8 / Avg: 508.1 / Max: 513.2

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Scala DottyUbuntu 19.10Ubuntu 21.102004006008001000SE +/- 5.99, N = 3SE +/- 6.40, N = 3805.6706.7MIN: 592.99 / MAX: 1336.15MIN: 587.36 / MAX: 1274.41
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.12Test: Scala DottyUbuntu 19.10Ubuntu 21.10140280420560700Min: 795.4 / Avg: 805.6 / Max: 816.16Min: 695.47 / Avg: 706.72 / Max: 717.65

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: regnety_400mUbuntu 19.10Ubuntu 21.103691215SE +/- 0.06, N = 3SE +/- 0.02, N = 311.9010.78-pthread - MIN: 11.67 / MAX: 21.85MIN: 10.67 / MAX: 11.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: regnety_400mUbuntu 19.10Ubuntu 21.103691215Min: 11.79 / Avg: 11.9 / Max: 12.01Min: 10.76 / Avg: 10.78 / Max: 10.821. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers such as Firefox and Google Chrome. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream 2 - Browser: FirefoxUbuntu 19.10Ubuntu 21.1020406080100SE +/- 1.06, N = 5SE +/- 0.33, N = 3102.2593.591. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0
OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream 2 - Browser: FirefoxUbuntu 19.10Ubuntu 21.1020406080100Min: 99.16 / Avg: 102.25 / Max: 105.78Min: 93.17 / Avg: 93.59 / Max: 94.241. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pathlibUbuntu 19.10Ubuntu 21.1048121620SE +/- 0.15, N = 3SE +/- 0.00, N = 316.715.3
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pathlibUbuntu 19.10Ubuntu 21.1048121620Min: 16.5 / Avg: 16.73 / Max: 17Min: 15.3 / Avg: 15.3 / Max: 15.3

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteUbuntu 19.10Ubuntu 21.10160K320K480K640K800KSE +/- 4761.83, N = 15SE +/- 10095.58, N = 3696173757984
OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteUbuntu 19.10Ubuntu 21.10130K260K390K520K650KMin: 673467 / Avg: 696172.8 / Max: 722531Min: 747484 / Avg: 757984.33 / Max: 778170

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 8.0Renderer: OpenGL 3.x - Resolution: 1920 x 1080Ubuntu 19.10Ubuntu 21.102004006008001000SE +/- 1.77, N = 3SE +/- 1.03, N = 3908.9981.91. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 8.0Renderer: OpenGL 3.x - Resolution: 1920 x 1080Ubuntu 19.10Ubuntu 21.102004006008001000Min: 905.4 / Avg: 908.93 / Max: 910.7Min: 979.8 / Avg: 981.87 / Max: 982.91. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 8.0Renderer: OpenGL 3.x - Resolution: 3840 x 2160Ubuntu 19.10Ubuntu 21.102004006008001000SE +/- 0.47, N = 3SE +/- 1.53, N = 3916.8987.51. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 8.0Renderer: OpenGL 3.x - Resolution: 3840 x 2160Ubuntu 19.10Ubuntu 21.102004006008001000Min: 915.9 / Avg: 916.83 / Max: 917.3Min: 986 / Avg: 987.53 / Max: 990.61. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pickle_pure_pythonUbuntu 19.10Ubuntu 21.1090180270360450SE +/- 0.58, N = 3SE +/- 0.67, N = 3422393
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pickle_pure_pythonUbuntu 19.10Ubuntu 21.1080160240320400Min: 421 / Avg: 422 / Max: 423Min: 392 / Avg: 393.33 / Max: 394

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: HighUbuntu 19.10Ubuntu 21.10100200300400500SE +/- 2.52, N = 3SE +/- 2.73, N = 3435.68467.79MIN: 240 / MAX: 715MIN: 185 / MAX: 744
OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: HighUbuntu 19.10Ubuntu 21.1080160240320400Min: 430.79 / Avg: 435.68 / Max: 439.19Min: 463.02 / Avg: 467.79 / Max: 472.47

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: MobileNetV2_224Ubuntu 19.10Ubuntu 21.101.03192.06383.09574.12765.1595SE +/- 0.006, N = 3SE +/- 0.036, N = 34.5864.274MIN: 4.54 / MAX: 5.17MIN: 4.18 / MAX: 4.741. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: MobileNetV2_224Ubuntu 19.10Ubuntu 21.10246810Min: 4.58 / Avg: 4.59 / Max: 4.6Min: 4.21 / Avg: 4.27 / Max: 4.331. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compileUbuntu 19.10Ubuntu 21.104080120160200SE +/- 0.00, N = 3162151
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compileUbuntu 19.10Ubuntu 21.10306090120150Min: 151 / Avg: 151 / Max: 151

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU-v3-v3 - Model: mobilenet-v3Ubuntu 19.10Ubuntu 21.101.03052.0613.09154.1225.1525SE +/- 0.02, N = 3SE +/- 0.01, N = 34.584.27-pthread - MIN: 4.48 / MAX: 14.71MIN: 4.2 / MAX: 4.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU-v3-v3 - Model: mobilenet-v3Ubuntu 19.10Ubuntu 21.10246810Min: 4.55 / Avg: 4.58 / Max: 4.62Min: 4.26 / Avg: 4.27 / Max: 4.291. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.1048121620SE +/- 0.07, N = 3SE +/- 0.08, N = 315.4316.53-lpthread1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.1048121620Min: 15.29 / Avg: 15.43 / Max: 15.52Min: 16.38 / Avg: 16.53 / Max: 16.651. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Tesseract

Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 3840 x 2160Ubuntu 19.10Ubuntu 21.1070140210280350SE +/- 3.82, N = 3SE +/- 4.17, N = 4318.17340.69
OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 3840 x 2160Ubuntu 19.10Ubuntu 21.1060120180240300Min: 310.57 / Avg: 318.17 / Max: 322.66Min: 330.79 / Avg: 340.69 / Max: 350.79

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: efficientnet-b0Ubuntu 19.10Ubuntu 21.10246810SE +/- 0.02, N = 3SE +/- 0.02, N = 36.866.42-pthread - MIN: 6.76 / MAX: 7.46MIN: 6.34 / MAX: 6.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: efficientnet-b0Ubuntu 19.10Ubuntu 21.103691215Min: 6.82 / Avg: 6.86 / Max: 6.9Min: 6.39 / Avg: 6.42 / Max: 6.471. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Decompression SpeedUbuntu 19.10Ubuntu 21.109001800270036004500SE +/- 8.17, N = 15SE +/- 3.56, N = 33806.14066.41. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Decompression SpeedUbuntu 19.10Ubuntu 21.107001400210028003500Min: 3737.1 / Avg: 3806.11 / Max: 3841.8Min: 4059.3 / Avg: 4066.4 / Max: 4070.41. (CC) gcc options: -O3 -pthread -lz

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU-v2-v2 - Model: mobilenet-v2Ubuntu 19.10Ubuntu 21.101.172.343.514.685.85SE +/- 0.03, N = 3SE +/- 0.01, N = 35.204.89-pthread - MIN: 5.04 / MAX: 6.67MIN: 4.8 / MAX: 5.451. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU-v2-v2 - Model: mobilenet-v2Ubuntu 19.10Ubuntu 21.10246810Min: 5.16 / Avg: 5.2 / Max: 5.26Min: 4.88 / Avg: 4.89 / Max: 4.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mnasnetUbuntu 19.10Ubuntu 21.101.04182.08363.12544.16725.209SE +/- 0.00, N = 3SE +/- 0.01, N = 34.634.37-pthread - MIN: 4.56 / MAX: 5.87MIN: 4.3 / MAX: 4.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mnasnetUbuntu 19.10Ubuntu 21.10246810Min: 4.63 / Avg: 4.63 / Max: 4.64Min: 4.36 / Avg: 4.37 / Max: 4.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateUbuntu 19.10Ubuntu 21.100.99791.99582.99373.99164.9895SE +/- 0.036921, N = 15SE +/- 0.054020, N = 34.1868314.4349191. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateUbuntu 19.10Ubuntu 21.10246810Min: 3.99 / Avg: 4.19 / Max: 4.51Min: 4.34 / Avg: 4.43 / Max: 4.531. (CC) gcc options: -O3 -march=native -fopenmp

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltraUbuntu 19.10Ubuntu 21.1090180270360450SE +/- 2.19, N = 3SE +/- 2.81, N = 3381.73403.41MIN: 216 / MAX: 616MIN: 207 / MAX: 635
OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltraUbuntu 19.10Ubuntu 21.1070140210280350Min: 377.52 / Avg: 381.73 / Max: 384.9Min: 397.8 / Avg: 403.41 / Max: 406.36

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression SpeedUbuntu 19.10Ubuntu 21.108001600240032004000SE +/- 32.67, N = 3SE +/- 34.79, N = 33323.23507.21. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression SpeedUbuntu 19.10Ubuntu 21.106001200180024003000Min: 3257.9 / Avg: 3323.17 / Max: 3358.5Min: 3446.7 / Avg: 3507.23 / Max: 3567.21. (CC) gcc options: -O3 -pthread -lz

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.1048121620SE +/- 0.11, N = 3SE +/- 0.13, N = 1513.8214.571. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.1048121620Min: 13.61 / Avg: 13.82 / Max: 13.98Min: 13.24 / Avg: 14.57 / Max: 15.021. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

FLAC Audio Encoding

This test times how long it takes to encode a sample WAV file to FLAC format ten times. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.3WAV To FLACUbuntu 19.10Ubuntu 21.1048121620SE +/- 0.04, N = 5SE +/- 0.07, N = 513.9413.231. (CXX) g++ options: -fvisibility=hidden -logg -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.3WAV To FLACUbuntu 19.10Ubuntu 21.1048121620Min: 13.81 / Avg: 13.94 / Max: 14.01Min: 12.97 / Avg: 13.23 / Max: 13.371. (CXX) g++ options: -fvisibility=hidden -logg -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 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.10510152025SE +/- 0.07, N = 3SE +/- 0.08, N = 318.4317.511. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.10510152025Min: 18.32 / Avg: 18.43 / Max: 18.55Min: 17.36 / Avg: 17.51 / Max: 17.641. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: squeezenet_ssdUbuntu 19.10Ubuntu 21.10510152025SE +/- 0.09, N = 3SE +/- 0.02, N = 318.8317.89-pthread - MIN: 18.28 / MAX: 23.01MIN: 17.47 / MAX: 18.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: squeezenet_ssdUbuntu 19.10Ubuntu 21.10510152025Min: 18.66 / Avg: 18.83 / Max: 18.95Min: 17.86 / Avg: 17.89 / Max: 17.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 3.0Algorithm: SHA256Ubuntu 19.10Ubuntu 21.104000M8000M12000M16000M20000MSE +/- 100553740.62, N = 3SE +/- 181854215.96, N = 517925407470170365765681. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.0Algorithm: SHA256Ubuntu 19.10Ubuntu 21.103000M6000M9000M12000M15000MMin: 17781906070 / Avg: 17925407470 / Max: 18119177330Min: 16610864830 / Avg: 17036576568 / Max: 174704503601. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -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 0.5Input: PNG - Encode Speed: 5Ubuntu 19.10Ubuntu 21.101122334455SE +/- 0.12, N = 3SE +/- 0.04, N = 346.7049.07-lbrotlidec -lbrotlicommon -lbrotlienc -pthread-fPIE -pie1. (CXX) g++ options: -funwind-tables -O3 -O2
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.5Input: PNG - Encode Speed: 5Ubuntu 19.10Ubuntu 21.101020304050Min: 46.46 / Avg: 46.7 / Max: 46.87Min: 48.98 / Avg: 49.07 / Max: 49.111. (CXX) g++ options: -funwind-tables -O3 -O2

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: squeezenetv1.1Ubuntu 19.10Ubuntu 21.101.08562.17123.25684.34245.428SE +/- 0.006, N = 3SE +/- 0.037, N = 34.8254.600MIN: 4.75 / MAX: 16.01MIN: 4.48 / MAX: 5.171. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: squeezenetv1.1Ubuntu 19.10Ubuntu 21.10246810Min: 4.82 / Avg: 4.83 / Max: 4.84Min: 4.53 / Avg: 4.6 / Max: 4.651. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.104080120160200SE +/- 0.48, N = 3SE +/- 0.06, N = 3151.57158.831. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.10306090120150Min: 150.83 / Avg: 151.57 / Max: 152.48Min: 158.77 / Avg: 158.83 / Max: 158.941. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

Facebook RocksDB

This is a benchmark of 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 BetterFacebook RocksDB 6.22.1Test: Random ReadUbuntu 19.10Ubuntu 21.1015M30M45M60M75MSE +/- 772623.01, N = 3SE +/- 175480.83, N = 372184348688866401. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Random ReadUbuntu 19.10Ubuntu 21.1013M26M39M52M65MMin: 70639250 / Avg: 72184347.67 / Max: 72975439Min: 68595632 / Avg: 68886640.33 / Max: 692020451. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: chaosUbuntu 19.10Ubuntu 21.1020406080100SE +/- 0.88, N = 3SE +/- 0.06, N = 3103.098.5
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: chaosUbuntu 19.10Ubuntu 21.1020406080100Min: 102 / Avg: 103.33 / Max: 105Min: 98.4 / Avg: 98.5 / Max: 98.6

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: yolov4-tinyUbuntu 19.10Ubuntu 21.10612182430SE +/- 0.40, N = 3SE +/- 0.03, N = 326.2025.07-pthread - MIN: 25.57 / MAX: 35.99MIN: 24.83 / MAX: 28.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: yolov4-tinyUbuntu 19.10Ubuntu 21.10612182430Min: 25.78 / Avg: 26.2 / Max: 27Min: 25.02 / Avg: 25.07 / Max: 25.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.1060120180240300SE +/- 0.08, N = 3SE +/- 0.54, N = 3278.29290.051. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.1050100150200250Min: 278.16 / Avg: 278.29 / Max: 278.42Min: 289.02 / Avg: 290.05 / Max: 290.841. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: googlenetUbuntu 19.10Ubuntu 21.1048121620SE +/- 0.13, N = 3SE +/- 0.07, N = 316.5315.87-pthread - MIN: 15.9 / MAX: 18.56MIN: 15.31 / MAX: 16.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: googlenetUbuntu 19.10Ubuntu 21.1048121620Min: 16.34 / Avg: 16.53 / Max: 16.78Min: 15.79 / Avg: 15.87 / Max: 16.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 4.4Time To CompileUbuntu 19.10Ubuntu 21.10918273645SE +/- 0.15, N = 3SE +/- 0.08, N = 338.8037.27
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.4Time To CompileUbuntu 19.10Ubuntu 21.10816243240Min: 38.51 / Avg: 38.8 / Max: 38.99Min: 37.13 / Avg: 37.27 / Max: 37.38

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltimateUbuntu 19.10Ubuntu 21.1070140210280350SE +/- 1.04, N = 3SE +/- 0.50, N = 3293.32304.15MIN: 53 / MAX: 513MIN: 53 / MAX: 531
OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltimateUbuntu 19.10Ubuntu 21.1050100150200250Min: 291.24 / Avg: 293.32 / Max: 294.44Min: 303.51 / Avg: 304.15 / Max: 305.14

Git

This test measures the time needed to carry out some sample Git operations on an example, static repository that happens to be a copy of the GNOME GTK tool-kit repository. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGitTime To Complete Common Git CommandsUbuntu 19.10Ubuntu 21.101020304050SE +/- 0.09, N = 3SE +/- 0.54, N = 345.7944.161. Ubuntu 19.10: git version 2.20.12. Ubuntu 21.10: git version 2.32.0
OpenBenchmarking.orgSeconds, Fewer Is BetterGitTime To Complete Common Git CommandsUbuntu 19.10Ubuntu 21.10918273645Min: 45.67 / Avg: 45.79 / Max: 45.96Min: 43.13 / Avg: 44.16 / Max: 44.931. Ubuntu 19.10: git version 2.20.12. Ubuntu 21.10: git version 2.32.0

PyBench

This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesUbuntu 19.10Ubuntu 21.102004006008001000SE +/- 3.18, N = 3SE +/- 7.81, N = 3921890
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesUbuntu 19.10Ubuntu 21.10160320480640800Min: 917 / Avg: 920.67 / Max: 927Min: 877 / Avg: 890 / Max: 904

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet50Ubuntu 19.10Ubuntu 21.10612182430SE +/- 0.14, N = 3SE +/- 0.00, N = 326.4925.61-pthread - MIN: 26.02 / MAX: 35.66MIN: 25.47 / MAX: 26.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet50Ubuntu 19.10Ubuntu 21.10612182430Min: 26.21 / Avg: 26.49 / Max: 26.68Min: 25.6 / Avg: 25.61 / Max: 25.611. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: SqueezeNetV1.0Ubuntu 19.10Ubuntu 21.10246810SE +/- 0.016, N = 3SE +/- 0.037, N = 36.8036.578MIN: 6.73 / MAX: 17.71MIN: 6.43 / MAX: 8.611. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: SqueezeNetV1.0Ubuntu 19.10Ubuntu 21.103691215Min: 6.79 / Avg: 6.8 / Max: 6.84Min: 6.51 / Avg: 6.58 / Max: 6.631. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet18Ubuntu 19.10Ubuntu 21.1048121620SE +/- 0.10, N = 3SE +/- 0.01, N = 316.3715.85-pthread - MIN: 16.09 / MAX: 16.75MIN: 15.71 / MAX: 24.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet18Ubuntu 19.10Ubuntu 21.1048121620Min: 16.21 / Avg: 16.37 / Max: 16.54Min: 15.83 / Avg: 15.85 / Max: 15.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: shufflenet-v2Ubuntu 19.10Ubuntu 21.101.13852.2773.41554.5545.6925SE +/- 0.02, N = 3SE +/- 0.02, N = 35.064.93-pthread - MIN: 4.97 / MAX: 6.02MIN: 4.83 / MAX: 5.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: shufflenet-v2Ubuntu 19.10Ubuntu 21.10246810Min: 5.02 / Avg: 5.06 / Max: 5.08Min: 4.91 / Avg: 4.93 / Max: 4.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: alexnetUbuntu 19.10Ubuntu 21.103691215SE +/- 0.07, N = 3SE +/- 0.08, N = 312.5912.30-pthread - MIN: 12.35 / MAX: 22.53MIN: 12.03 / MAX: 12.641. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: alexnetUbuntu 19.10Ubuntu 21.1048121620Min: 12.47 / Avg: 12.59 / Max: 12.71Min: 12.19 / Avg: 12.3 / Max: 12.451. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mobilenetUbuntu 19.10Ubuntu 21.1048121620SE +/- 0.04, N = 3SE +/- 0.02, N = 316.0015.64-pthread - MIN: 15.62 / MAX: 27.97MIN: 15.24 / MAX: 16.381. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mobilenetUbuntu 19.10Ubuntu 21.1048121620Min: 15.93 / Avg: 16 / Max: 16.05Min: 15.62 / Avg: 15.64 / Max: 15.691. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Compression SpeedUbuntu 19.10Ubuntu 21.1030060090012001500SE +/- 11.64, N = 15SE +/- 13.32, N = 31282.61312.11. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Compression SpeedUbuntu 19.10Ubuntu 21.102004006008001000Min: 1169.6 / Avg: 1282.55 / Max: 1333.2Min: 1285.7 / Avg: 1312.13 / Max: 1328.21. (CC) gcc options: -O3 -pthread -lz

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: 2to3Ubuntu 19.10Ubuntu 21.1060120180240300SE +/- 0.00, N = 3286280
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: 2to3Ubuntu 19.10Ubuntu 21.1050100150200250Min: 280 / Avg: 280 / Max: 280

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: resnet-v2-50Ubuntu 19.10Ubuntu 21.10816243240SE +/- 0.22, N = 3SE +/- 0.15, N = 334.5133.79MIN: 33.61 / MAX: 47.01MIN: 32.65 / MAX: 42.761. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: resnet-v2-50Ubuntu 19.10Ubuntu 21.10714212835Min: 34.2 / Avg: 34.51 / Max: 34.93Min: 33.5 / Avg: 33.79 / Max: 33.991. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: blazefaceUbuntu 19.10Ubuntu 21.100.47480.94961.42441.89922.374SE +/- 0.02, N = 3SE +/- 0.01, N = 32.112.07-pthread - MIN: 2.05 / MAX: 4.14MIN: 2.03 / MAX: 2.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: blazefaceUbuntu 19.10Ubuntu 21.10246810Min: 2.08 / Avg: 2.11 / Max: 2.14Min: 2.06 / Avg: 2.07 / Max: 2.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 3.4Video Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.10510152025SE +/- 0.14, N = 3SE +/- 0.03, N = 319.5819.241. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.10510152025Min: 19.34 / Avg: 19.58 / Max: 19.84Min: 19.2 / Avg: 19.24 / Max: 19.31. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: inception-v3Ubuntu 19.10Ubuntu 21.10714212835SE +/- 0.08, N = 3SE +/- 0.26, N = 331.6531.11MIN: 31.15 / MAX: 43.97MIN: 30.19 / MAX: 41.741. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: inception-v3Ubuntu 19.10Ubuntu 21.10714212835Min: 31.49 / Avg: 31.65 / Max: 31.76Min: 30.6 / Avg: 31.11 / Max: 31.411. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.101122334455SE +/- 0.20, N = 3SE +/- 0.11, N = 349.3948.59-lpthread1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.101020304050Min: 49.08 / Avg: 49.39 / Max: 49.76Min: 48.46 / Avg: 48.59 / Max: 48.81. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenet-v1-1.0Ubuntu 19.10Ubuntu 21.101.23642.47283.70924.94566.182SE +/- 0.021, N = 3SE +/- 0.040, N = 35.4955.412MIN: 5.35 / MAX: 16.33MIN: 5.26 / MAX: 5.881. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenet-v1-1.0Ubuntu 19.10Ubuntu 21.10246810Min: 5.47 / Avg: 5.49 / Max: 5.54Min: 5.37 / Avg: 5.41 / Max: 5.491. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

XZ Compression

This test measures the time needed to compress a sample file (an Ubuntu file-system image) using XZ compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXZ Compression 5.2.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9Ubuntu 19.10Ubuntu 21.10612182430SE +/- 0.08, N = 3SE +/- 0.13, N = 325.4325.79-pthread1. (CC) gcc options: -fvisibility=hidden -O2
OpenBenchmarking.orgSeconds, Fewer Is BetterXZ Compression 5.2.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9Ubuntu 19.10Ubuntu 21.10612182430Min: 25.29 / Avg: 25.43 / Max: 25.58Min: 25.56 / Avg: 25.79 / Max: 26.011. (CC) gcc options: -fvisibility=hidden -O2

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CUbuntu 19.10Ubuntu 21.105K10K15K20K25KSE +/- 12.98, N = 3SE +/- 6.18, N = 324905.4725245.58-pthread-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CUbuntu 19.10Ubuntu 21.104K8K12K16K20KMin: 24890.11 / Avg: 24905.47 / Max: 24931.27Min: 25234.1 / Avg: 25245.58 / Max: 25255.271. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0

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 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.100.37550.7511.12651.5021.8775SE +/- 0.002, N = 3SE +/- 0.006, N = 31.6481.6691. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 4KUbuntu 19.10Ubuntu 21.10246810Min: 1.64 / Avg: 1.65 / Max: 1.65Min: 1.66 / Avg: 1.67 / Max: 1.681. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Disney MaterialUbuntu 19.10Ubuntu 21.104080120160200156.85158.82

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 512 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 13Total TimeUbuntu 19.10Ubuntu 21.108M16M24M32M40MSE +/- 344358.26, N = 3SE +/- 395256.04, N = 339264331387888821. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fprofile-use -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -flto -flto=jobserver
OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 13Total TimeUbuntu 19.10Ubuntu 21.107M14M21M28M35MMin: 38740674 / Avg: 39264331 / Max: 39913568Min: 38239643 / Avg: 38788882.33 / Max: 395558781. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fprofile-use -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -flto -flto=jobserver

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CUbuntu 19.10Ubuntu 21.1015003000450060007500SE +/- 21.33, N = 3SE +/- 14.77, N = 36995.277075.55-pthread-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CUbuntu 19.10Ubuntu 21.1012002400360048006000Min: 6954.75 / Avg: 6995.27 / Max: 7027.11Min: 7053.06 / Avg: 7075.55 / Max: 7103.381. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0

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 0.5Input: PNG - Encode Speed: 8Ubuntu 19.10Ubuntu 21.100.20250.4050.60750.811.0125SE +/- 0.01, N = 3SE +/- 0.00, N = 30.890.90-lbrotlidec -lbrotlicommon -lbrotlienc -pthread-fPIE -pie1. (CXX) g++ options: -funwind-tables -O3 -O2
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.5Input: PNG - Encode Speed: 8Ubuntu 19.10Ubuntu 21.10246810Min: 0.87 / Avg: 0.89 / Max: 0.91Min: 0.9 / Avg: 0.9 / Max: 0.91. (CXX) g++ options: -funwind-tables -O3 -O2

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 3.4Video Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.101428425670SE +/- 0.16, N = 3SE +/- 0.46, N = 361.5860.921. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.101224364860Min: 61.26 / Avg: 61.58 / Max: 61.75Min: 60.01 / Avg: 60.92 / Max: 61.491. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.103691215SE +/- 0.02, N = 3SE +/- 0.01, N = 311.7111.591. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.103691215Min: 11.67 / Avg: 11.71 / Max: 11.75Min: 11.58 / Avg: 11.59 / Max: 11.611. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: goUbuntu 19.10Ubuntu 21.1050100150200250SE +/- 1.53, N = 3SE +/- 0.33, N = 3223221
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: goUbuntu 19.10Ubuntu 21.104080120160200Min: 220 / Avg: 223 / Max: 225Min: 221 / Avg: 221.33 / Max: 222

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 7.32.2VGR Performance MetricUbuntu 19.10Ubuntu 21.1040K80K120K160K200K190898189236-pthread1. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -ldl -lm

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 5Mode: CPUUbuntu 19.10Ubuntu 21.103K6K9K12K15KSE +/- 42.93, N = 3SE +/- 55.21, N = 31441514297
OpenBenchmarking.orgvsamples, More Is BetterChaos Group V-RAY 5Mode: CPUUbuntu 19.10Ubuntu 21.103K6K9K12K15KMin: 14330 / Avg: 14415 / Max: 14468Min: 14188 / Avg: 14296.67 / Max: 14368

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression SpeedUbuntu 19.10Ubuntu 21.10918273645SE +/- 0.25, N = 3SE +/- 0.09, N = 337.537.21. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression SpeedUbuntu 19.10Ubuntu 21.10816243240Min: 37 / Avg: 37.5 / Max: 37.8Min: 37.1 / Avg: 37.23 / Max: 37.41. (CC) gcc options: -O3 -pthread -lz

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CUbuntu 19.10Ubuntu 21.1011002200330044005500SE +/- 0.89, N = 3SE +/- 26.57, N = 35261.575298.41-pthread-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CUbuntu 19.10Ubuntu 21.109001800270036004500Min: 5260.34 / Avg: 5261.57 / Max: 5263.3Min: 5258.74 / Avg: 5298.41 / Max: 5348.861. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-OnlyUbuntu 19.10Ubuntu 21.1020406080100SE +/- 0.13, N = 3SE +/- 0.18, N = 3105.07105.76
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-OnlyUbuntu 19.10Ubuntu 21.1020406080100Min: 104.83 / Avg: 105.07 / Max: 105.26Min: 105.55 / Avg: 105.76 / Max: 106.12

IndigoBench

This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: SupercarUbuntu 19.10Ubuntu 21.10246810SE +/- 0.005, N = 3SE +/- 0.019, N = 36.1446.182
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: SupercarUbuntu 19.10Ubuntu 21.10246810Min: 6.14 / Avg: 6.14 / Max: 6.15Min: 6.15 / Avg: 6.18 / Max: 6.21

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers such as Firefox and Google Chrome. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, Fewer Is BetterSeleniumBenchmark: PSPDFKit WASM - Browser: FirefoxUbuntu 19.10Ubuntu 21.107001400210028003500SE +/- 21.63, N = 3SE +/- 20.46, N = 3307830971. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0
OpenBenchmarking.orgScore, Fewer Is BetterSeleniumBenchmark: PSPDFKit WASM - Browser: FirefoxUbuntu 19.10Ubuntu 21.105001000150020002500Min: 3048 / Avg: 3078 / Max: 3120Min: 3073 / Avg: 3097.33 / Max: 31381. Ubuntu 19.10: firefox 69.0.32. Ubuntu 21.10: firefox 93.0

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Material TesterUbuntu 19.10Ubuntu 21.10306090120150158.01157.15

Tachyon

This is a test of the threaded Tachyon, a parallel ray-tracing system, measuring the time to ray-trace a sample scene. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.99b6Total TimeUbuntu 19.10Ubuntu 21.101326395265SE +/- 0.11, N = 3SE +/- 0.11, N = 360.1259.811. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.99b6Total TimeUbuntu 19.10Ubuntu 21.101224364860Min: 59.91 / Avg: 60.12 / Max: 60.28Min: 59.61 / Avg: 59.81 / Max: 601. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm.in input file (Problem 5). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsUbuntu 19.10Ubuntu 21.10306090120150SE +/- 0.12, N = 3SE +/- 0.08, N = 3134.38133.721. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsUbuntu 19.10Ubuntu 21.10306090120150Min: 134.16 / Avg: 134.38 / Max: 134.57Min: 133.59 / Avg: 133.72 / Max: 133.871. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression SpeedUbuntu 19.10Ubuntu 21.102K4K6K8K10KSE +/- 16.81, N = 3SE +/- 25.62, N = 1510543.310593.81. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression SpeedUbuntu 19.10Ubuntu 21.102K4K6K8K10KMin: 10509.8 / Avg: 10543.27 / Max: 10562.7Min: 10445 / Avg: 10593.8 / Max: 10735.61. (CC) gcc options: -O3

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthUbuntu 19.10Ubuntu 21.109M18M27M36M45MSE +/- 286086.06, N = 3SE +/- 268748.21, N = 33978793639947743
OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthUbuntu 19.10Ubuntu 21.107M14M21M28M35MMin: 39230873 / Avg: 39787935.67 / Max: 40179589Min: 39661165 / Avg: 39947743 / Max: 40484836

Facebook RocksDB

This is a benchmark of 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 BetterFacebook RocksDB 6.22.1Test: Random FillUbuntu 19.10Ubuntu 21.10200K400K600K800K1000KSE +/- 5626.77, N = 3SE +/- 11917.32, N = 49737189775271. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.22.1Test: Random FillUbuntu 19.10Ubuntu 21.10200K400K600K800K1000KMin: 962490 / Avg: 973717.67 / Max: 979992Min: 955292 / Avg: 977527 / Max: 10055621. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021.2Implementation: MPI CPU - Input: water_GMX50_bareUbuntu 19.10Ubuntu 21.100.25450.5090.76351.0181.2725SE +/- 0.002, N = 3SE +/- 0.001, N = 31.1311.128-pthread1. (CXX) g++ options: -O3
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021.2Implementation: MPI CPU - Input: water_GMX50_bareUbuntu 19.10Ubuntu 21.10246810Min: 1.13 / Avg: 1.13 / Max: 1.14Min: 1.13 / Avg: 1.13 / Max: 1.131. (CXX) g++ options: -O3

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: vgg16Ubuntu 19.10Ubuntu 21.101326395265SE +/- 0.19, N = 3SE +/- 0.13, N = 359.8259.96-pthread - MIN: 58.72 / MAX: 72.25MIN: 59.16 / MAX: 90.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: vgg16Ubuntu 19.10Ubuntu 21.101224364860Min: 59.45 / Avg: 59.82 / Max: 60.05Min: 59.77 / Avg: 59.96 / Max: 60.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.1050100150200250SE +/- 0.44, N = 3SE +/- 0.17, N = 3206.45206.061. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.104080120160200Min: 205.58 / Avg: 206.45 / Max: 206.97Min: 205.88 / Avg: 206.06 / Max: 206.391. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsUbuntu 19.10Ubuntu 21.100.33710.67421.01131.34841.6855SE +/- 0.00235, N = 3SE +/- 0.00185, N = 31.498091.49551
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsUbuntu 19.10Ubuntu 21.10246810Min: 1.49 / Avg: 1.5 / Max: 1.5Min: 1.49 / Avg: 1.5 / Max: 1.5

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedUbuntu 19.10Ubuntu 21.101428425670SE +/- 0.48, N = 3SE +/- 0.51, N = 1560.8860.941. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedUbuntu 19.10Ubuntu 21.101224364860Min: 59.93 / Avg: 60.88 / Max: 61.4Min: 57.64 / Avg: 60.94 / Max: 64.881. (CC) gcc options: -O3

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 3.0Algorithm: RSA4096Ubuntu 19.10Ubuntu 21.1050K100K150K200K250KSE +/- 63.36, N = 3SE +/- 36.62, N = 3229876.3230097.11. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096Ubuntu 19.10Ubuntu 21.1040K80K120K160K200KMin: 229781.6 / Avg: 229876.33 / Max: 229996.6Min: 230028 / Avg: 230097.07 / Max: 230152.71. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -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 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.101.17612.35223.52834.70445.8805SE +/- 0.004, N = 3SE +/- 0.019, N = 35.2275.2231. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.10246810Min: 5.22 / Avg: 5.23 / Max: 5.23Min: 5.19 / Avg: 5.22 / Max: 5.251. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

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.orgsign/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096Ubuntu 19.10Ubuntu 21.108001600240032004000SE +/- 3.71, N = 3SE +/- 3.20, N = 33551.93554.41. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096Ubuntu 19.10Ubuntu 21.106001200180024003000Min: 3547.8 / Avg: 3551.9 / Max: 3559.3Min: 3551.1 / Avg: 3554.4 / Max: 3560.81. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.BUbuntu 19.10Ubuntu 21.102K4K6K8K10KSE +/- 12.31, N = 3SE +/- 35.42, N = 38909.788914.81-pthread-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.BUbuntu 19.10Ubuntu 21.1015003000450060007500Min: 8886.92 / Avg: 8909.78 / Max: 8929.12Min: 8844.06 / Avg: 8914.81 / Max: 8953.161. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-OnlyUbuntu 19.10Ubuntu 21.1070140210280350SE +/- 0.37, N = 3SE +/- 0.13, N = 3303.06303.22
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-OnlyUbuntu 19.10Ubuntu 21.1050100150200250Min: 302.33 / Avg: 303.06 / Max: 303.56Min: 303.03 / Avg: 303.22 / Max: 303.48

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CUbuntu 19.10Ubuntu 21.103K6K9K12K15KSE +/- 7.22, N = 3SE +/- 3.19, N = 313431.1413425.58-pthread-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CUbuntu 19.10Ubuntu 21.102K4K6K8K10KMin: 13418.83 / Avg: 13431.14 / Max: 13443.82Min: 13419.58 / Avg: 13425.58 / Max: 13430.481. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 19.10: Open MPI 3.1.33. Ubuntu 21.10: Open MPI 4.1.0

IndigoBench

This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: BedroomUbuntu 19.10Ubuntu 21.100.64851.2971.94552.5943.2425SE +/- 0.008, N = 3SE +/- 0.005, N = 32.8822.881
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: BedroomUbuntu 19.10Ubuntu 21.10246810Min: 2.87 / Avg: 2.88 / Max: 2.89Min: 2.87 / Avg: 2.88 / Max: 2.89

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: EmilyUbuntu 19.10Ubuntu 21.1060120180240300260.63260.55

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 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.101428425670SE +/- 0.20, N = 3SE +/- 0.31, N = 364.6864.661. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 1080pUbuntu 19.10Ubuntu 21.101326395265Min: 64.44 / Avg: 64.68 / Max: 65.07Min: 64.05 / Avg: 64.66 / Max: 64.981. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

117 Results Shown

NCNN:
  Vulkan GPU - vgg16
  Vulkan GPU - resnet50
  Vulkan GPU - googlenet
  Vulkan GPU - resnet18
  Vulkan GPU - alexnet
  Vulkan GPU - squeezenet_ssd
  Vulkan GPU - mnasnet
  Vulkan GPU-v2-v2 - mobilenet-v2
  Vulkan GPU-v3-v3 - mobilenet-v3
  Vulkan GPU - regnety_400m
PyPerformance
Tesseract
NCNN:
  Vulkan GPU - yolov4-tiny
  Vulkan GPU - shufflenet-v2
  Vulkan GPU - efficientnet-b0
Selenium
NCNN
yquake2:
  OpenGL 1.x - 3840 x 2160
  OpenGL 1.x - 1920 x 1080
PyPerformance
Google SynthMark
NCNN
Selenium:
  ARES-6 - Firefox
  StyleBench - Firefox
VP9 libvpx Encoding
ET: Legacy
Mobile Neural Network
ET: Legacy
Renaissance
NCNN
Selenium
PyPerformance
PHPBench
yquake2:
  OpenGL 3.x - 1920 x 1080
  OpenGL 3.x - 3840 x 2160
PyPerformance
Xonotic
Mobile Neural Network
PyPerformance
NCNN
AOM AV1
Tesseract
NCNN
Zstd Compression
NCNN:
  CPU-v2-v2 - mobilenet-v2
  CPU - mnasnet
ACES DGEMM
Xonotic
Zstd Compression
VP9 libvpx Encoding
FLAC Audio Encoding
SVT-AV1
NCNN
OpenSSL
JPEG XL libjxl
Mobile Neural Network
SVT-HEVC
Facebook RocksDB
PyPerformance
NCNN
SVT-HEVC
NCNN
Timed FFmpeg Compilation
Xonotic
Git
PyBench
NCNN
Mobile Neural Network
NCNN:
  CPU - resnet18
  CPU - shufflenet-v2
  CPU - alexnet
  CPU - mobilenet
Zstd Compression
PyPerformance
Mobile Neural Network
NCNN
x265
Mobile Neural Network
AOM AV1
Mobile Neural Network
XZ Compression
NAS Parallel Benchmarks
SVT-AV1
Appleseed
Stockfish
NAS Parallel Benchmarks
JPEG XL libjxl
x265
SVT-HEVC
PyPerformance
BRL-CAD
Chaos Group V-RAY
Zstd Compression
NAS Parallel Benchmarks
Blender
IndigoBench
Selenium
Appleseed
Tachyon
CloverLeaf
LZ4 Compression
asmFish
Facebook RocksDB
GROMACS
NCNN
SVT-VP9
NAMD
LZ4 Compression
OpenSSL
SVT-AV1
OpenSSL
NAS Parallel Benchmarks
Blender
NAS Parallel Benchmarks
IndigoBench
Appleseed
SVT-AV1