EPYC Ondemand vs. Performance vs. Schedutil Linux 5.11

patch testing by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/2102055-HA-EPYC7F52L26&grt.

EPYC Ondemand vs. Performance vs. Schedutil Linux 5.11ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen ResolutionOndemandSchedutilPerformanceAMD EPYC 7F52 16-Core @ 3.91GHz (16 Cores / 32 Threads)Supermicro H11DSi-NT v2.00 (2.1 BIOS)AMD Starship/Matisse8 x 8192 MB DDR4-3200MT/s HMA81GR7CJR8N-XN280GB INTEL SSDPE21D280GAASPEEDVE228Ubuntu 20.045.11.0-rc6-phx (x86_64) 20210203GNOME Shell 3.36.1X Server 1.20.7aspeedGCC 9.3.0ext41920x10801024x768OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-objc-gc=auto --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 Processor Details- Ondemand: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301034- Schedutil: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301034- Performance: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Python Details- Python 2.7.18rc1 + Python 3.8.2Security Details- 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 IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

EPYC Ondemand vs. Performance vs. Schedutil Linux 5.11blender: BMW27 - CPU-Onlyblender: Barbershop - CPU-Onlyclomp: Static OMP Speedupdav1d: Chimera 1080pdav1d: Summer Nature 4Kdav1d: Summer Nature 1080pdav1d: Chimera 1080p 10-bitgraphics-magick: Swirlgraphics-magick: Rotategraphics-magick: Sharpengraphics-magick: Enhancedgraphics-magick: Resizinggraphics-magick: Noise-Gaussiangraphics-magick: HWB Color Spacehugin: Panorama Photo Assistant + Stitching Timeoidn: Memorialjohn-the-ripper: Blowfishjohn-the-ripper: MD5kvazaar: Bosphorus 4K - Very Fastkvazaar: Bosphorus 4K - Ultra Fastkvazaar: Bosphorus 1080p - Very Fastkvazaar: Bosphorus 1080p - Ultra Fastluxcorerender: DLSCluxcorerender: Rainbow Colors and Prismocrmypdf: Processing 60 Page PDF Documentopenvkl: vklBenchmarkospray: San Miguel - SciVisospray: San Miguel - Path Tracerospray: NASA Streamlines - SciVisospray: NASA Streamlines - Path Tracerpennant: sedovbigpennant: leblancbigcompress-rar: Linux Source Tree Archiving To RARredis: GETredis: SETrodinia: OpenMP LavaMDrodinia: OpenMP HotSpot3Drodinia: OpenMP Leukocyterodinia: OpenMP CFD Solverrodinia: OpenMP Streamclustersvt-av1: Enc Mode 4 - 1080psvt-av1: Enc Mode 8 - 1080ptensorflow-lite: SqueezeNettensorflow-lite: Inception V4tensorflow-lite: NASNet Mobiletensorflow-lite: Mobilenet Floattensorflow-lite: Mobilenet Quanttensorflow-lite: Inception ResNet V2build-gdb: Time To Compilebuild-godot: Time To Compilex264: H.264 Video Encodingx265: Bosphorus 4Kx265: Bosphorus 1080pOndemandSchedutilPerformance83.66357.5749.7625.43254.29592.34119.139226382363761582421127150.65414.1926298171966725.3546.8687.16155.973.153.4819.54621924.391.8633.337.0925.6107416.0296774.9831545026.541308179.83126.38990.04289.26913.24313.7695.24441.023107260150441312687869039.370655.1135144095.49284.472172.6421.6565.4783.57357.4950.3536.65246.86488.9592.219236392363761587421112150.46414.2626310172100026.2747.6386.89153.013.193.5719.48321724.391.8733.337.1125.5967015.9934776.9211599356.291313860.03125.97791.31090.43113.24713.7545.37943.040107294150309312598169012.970760.6135065389.60683.313172.6021.0555.0783.73356.9050.7713.18264.98668.50131.749286292373771609420123950.40814.2326323172133326.3947.8092.22161.973.183.5119.28122324.391.8733.337.1125.5425715.9098067.6761604323.001318380.42125.88192.08789.70913.33113.7055.43142.823107390150329712620368976.970739.1135066783.70983.164176.5421.5068.48OpenBenchmarking.org

Blender

Blend File: BMW27 - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: BMW27 - Compute: CPU-OnlyOndemandSchedutilPerformance20406080100SE +/- 0.38, N = 3SE +/- 0.25, N = 3SE +/- 0.50, N = 383.6683.5783.73

Blender

Blend File: Barbershop - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: CPU-OnlyOndemandSchedutilPerformance80160240320400SE +/- 0.51, N = 3SE +/- 0.45, N = 3SE +/- 0.68, N = 3357.57357.49356.90

CLOMP

Static OMP Speedup

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP SpeedupOndemandSchedutilPerformance1122334455SE +/- 0.25, N = 3SE +/- 0.33, N = 3SE +/- 0.43, N = 349.750.350.71. (CC) gcc options: -fopenmp -O3 -lm

dav1d

Video Input: Chimera 1080p

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Chimera 1080pOndemandSchedutilPerformance150300450600750SE +/- 1.15, N = 3SE +/- 0.42, N = 3SE +/- 1.61, N = 3625.43536.65713.18MIN: 491.48 / MAX: 770.18MIN: 417.73 / MAX: 665.43MIN: 554.58 / MAX: 886.761. (CC) gcc options: -pthread

dav1d

Video Input: Summer Nature 4K

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Summer Nature 4KOndemandSchedutilPerformance60120180240300SE +/- 0.06, N = 3SE +/- 0.18, N = 3SE +/- 0.38, N = 3254.29246.86264.98MIN: 211.06 / MAX: 289.79MIN: 205.59 / MAX: 285.26MIN: 222.23 / MAX: 302.431. (CC) gcc options: -pthread

dav1d

Video Input: Summer Nature 1080p

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Summer Nature 1080pOndemandSchedutilPerformance140280420560700SE +/- 2.45, N = 3SE +/- 1.06, N = 3SE +/- 0.74, N = 3592.34488.95668.50MIN: 443.85 / MAX: 644.65MIN: 390.53 / MAX: 532.25MIN: 502.68 / MAX: 730.81. (CC) gcc options: -pthread

dav1d

Video Input: Chimera 1080p 10-bit

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Chimera 1080p 10-bitOndemandSchedutilPerformance306090120150SE +/- 0.19, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3119.1392.21131.74MIN: 78.3 / MAX: 259.88MIN: 59.88 / MAX: 204.44MIN: 85.18 / MAX: 277.621. (CC) gcc options: -pthread

GraphicsMagick

Operation: Swirl

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SwirlOndemandSchedutilPerformance2004006008001000SE +/- 0.88, N = 39229239281. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Rotate

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: RotateOndemandSchedutilPerformance140280420560700SE +/- 4.70, N = 3SE +/- 4.18, N = 3SE +/- 4.18, N = 36386396291. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SharpenOndemandSchedutilPerformance501001502002502362362371. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Enhanced

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: EnhancedOndemandSchedutilPerformance80160240320400SE +/- 0.33, N = 33763763771. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Resizing

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: ResizingOndemandSchedutilPerformance30060090012001500SE +/- 5.49, N = 3SE +/- 18.52, N = 3SE +/- 1.20, N = 31582158716091. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Noise-Gaussian

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Noise-GaussianOndemandSchedutilPerformance90180270360450SE +/- 0.67, N = 3SE +/- 0.58, N = 3SE +/- 0.67, N = 34214214201. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: HWB Color Space

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: HWB Color SpaceOndemandSchedutilPerformance30060090012001500SE +/- 2.52, N = 3SE +/- 2.40, N = 3SE +/- 3.67, N = 31271112112391. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

Hugin

Panorama Photo Assistant + Stitching Time

OpenBenchmarking.orgSeconds, Fewer Is BetterHuginPanorama Photo Assistant + Stitching TimeOndemandSchedutilPerformance1122334455SE +/- 0.33, N = 3SE +/- 0.17, N = 3SE +/- 0.29, N = 350.6550.4650.41

Intel Open Image Denoise

Scene: Memorial

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.2.0Scene: MemorialOndemandSchedutilPerformance48121620SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 314.1914.2614.23

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.9.0-jumbo-1Test: BlowfishOndemandSchedutilPerformance6K12K18K24K30KSE +/- 22.67, N = 3SE +/- 22.45, N = 3SE +/- 23.97, N = 32629826310263231. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2

John The Ripper

Test: MD5

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.9.0-jumbo-1Test: MD5OndemandSchedutilPerformance400K800K1200K1600K2000KSE +/- 3179.80, N = 3SE +/- 3000.00, N = 3SE +/- 2962.73, N = 31719667172100017213331. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Very Fast

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Very FastOndemandSchedutilPerformance612182430SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 325.3526.2726.391. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Kvazaar

Video Input: Bosphorus 4K - Video Preset: Ultra Fast

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra FastOndemandSchedutilPerformance1122334455SE +/- 0.14, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 346.8647.6347.801. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Kvazaar

Video Input: Bosphorus 1080p - Video Preset: Very Fast

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Very FastOndemandSchedutilPerformance20406080100SE +/- 0.29, N = 3SE +/- 0.21, N = 3SE +/- 0.08, N = 387.1686.8992.221. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Kvazaar

Video Input: Bosphorus 1080p - Video Preset: Ultra Fast

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Ultra FastOndemandSchedutilPerformance4080120160200SE +/- 0.36, N = 3SE +/- 0.28, N = 3SE +/- 0.32, N = 3155.97153.01161.971. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

LuxCoreRender

Scene: DLSC

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.3Scene: DLSCOndemandSchedutilPerformance0.71781.43562.15342.87123.589SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 33.153.193.18MIN: 3.06 / MAX: 3.29MIN: 3.04 / MAX: 3.41MIN: 3.05 / MAX: 3.39

LuxCoreRender

Scene: Rainbow Colors and Prism

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.3Scene: Rainbow Colors and PrismOndemandSchedutilPerformance0.80331.60662.40993.21324.0165SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 33.483.573.51MIN: 3.44 / MAX: 3.5MIN: 3.48 / MAX: 3.6MIN: 3.48 / MAX: 3.55

OCRMyPDF

Processing 60 Page PDF Document

OpenBenchmarking.orgSeconds, Fewer Is BetterOCRMyPDF 9.6.0+dfsgProcessing 60 Page PDF DocumentOndemandSchedutilPerformance510152025SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 319.5519.4819.28

OpenVKL

Benchmark: vklBenchmark

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkOndemandSchedutilPerformance50100150200250SE +/- 0.58, N = 3SE +/- 0.88, N = 3219217223MIN: 1 / MAX: 776MIN: 1 / MAX: 786MIN: 1 / MAX: 785

OSPray

Demo: San Miguel - Renderer: SciVis

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisOndemandSchedutilPerformance612182430SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 324.3924.3924.39MIN: 23.26 / MAX: 26.32MIN: 23.26 / MAX: 26.32MIN: 23.26 / MAX: 26.32

OSPray

Demo: San Miguel - Renderer: Path Tracer

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerOndemandSchedutilPerformance0.42080.84161.26241.68322.104SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.861.871.87MIN: 1.85 / MAX: 1.88MIN: 1.85 / MAX: 1.88MIN: 1.86 / MAX: 1.88

OSPray

Demo: NASA Streamlines - Renderer: SciVis

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVisOndemandSchedutilPerformance816243240SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 333.3333.3333.33MIN: 31.25 / MAX: 34.48MIN: 31.25MIN: 31.25 / MAX: 34.48

OSPray

Demo: NASA Streamlines - Renderer: Path Tracer

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerOndemandSchedutilPerformance246810SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 37.097.117.11MIN: 6.99 / MAX: 7.19MIN: 6.99 / MAX: 7.19MIN: 6.99 / MAX: 7.25

Pennant

Test: sedovbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigOndemandSchedutilPerformance612182430SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 325.6125.6025.541. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi

Pennant

Test: leblancbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigOndemandSchedutilPerformance48121620SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 316.0315.9915.911. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi

RAR Compression

Linux Source Tree Archiving To RAR

OpenBenchmarking.orgSeconds, Fewer Is BetterRAR Compression 5.6.1Linux Source Tree Archiving To RAROndemandSchedutilPerformance20406080100SE +/- 0.18, N = 3SE +/- 0.41, N = 3SE +/- 0.24, N = 374.9876.9267.68

Redis

Test: GET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: GETOndemandSchedutilPerformance300K600K900K1200K1500KSE +/- 12901.14, N = 3SE +/- 26190.66, N = 3SE +/- 13020.90, N = 31545026.541599356.291604323.001. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

Redis

Test: SET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SETOndemandSchedutilPerformance300K600K900K1200K1500KSE +/- 3347.62, N = 3SE +/- 16949.35, N = 4SE +/- 17244.90, N = 31308179.831313860.031318380.421. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDOndemandSchedutilPerformance306090120150SE +/- 0.25, N = 3SE +/- 0.27, N = 3SE +/- 0.28, N = 3126.39125.98125.881. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP HotSpot3D

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP HotSpot3DOndemandSchedutilPerformance20406080100SE +/- 0.08, N = 3SE +/- 0.24, N = 3SE +/- 0.46, N = 390.0491.3192.091. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP Leukocyte

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LeukocyteOndemandSchedutilPerformance20406080100SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.29, N = 389.2790.4389.711. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP CFD Solver

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP CFD SolverOndemandSchedutilPerformance3691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 313.2413.2513.331. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP StreamclusterOndemandSchedutilPerformance48121620SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 313.7713.7513.711. (CXX) g++ options: -O2 -lOpenCL

SVT-AV1

Encoder Mode: Enc Mode 4 - Input: 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080pOndemandSchedutilPerformance1.2222.4443.6664.8886.11SE +/- 0.049, N = 3SE +/- 0.025, N = 3SE +/- 0.023, N = 35.2445.3795.4311. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

SVT-AV1

Encoder Mode: Enc Mode 8 - Input: 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080pOndemandSchedutilPerformance1020304050SE +/- 0.11, N = 3SE +/- 0.22, N = 3SE +/- 0.12, N = 341.0243.0442.821. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

TensorFlow Lite

Model: SqueezeNet

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetOndemandSchedutilPerformance20K40K60K80K100KSE +/- 84.90, N = 3SE +/- 80.49, N = 3SE +/- 43.47, N = 3107260107294107390

TensorFlow Lite

Model: Inception V4

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4OndemandSchedutilPerformance300K600K900K1200K1500KSE +/- 385.24, N = 3SE +/- 262.95, N = 3SE +/- 236.24, N = 3150441315030931503297

TensorFlow Lite

Model: NASNet Mobile

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileOndemandSchedutilPerformance30K60K90K120K150KSE +/- 136.25, N = 3SE +/- 40.55, N = 3SE +/- 287.23, N = 3126878125981126203

TensorFlow Lite

Model: Mobilenet Float

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatOndemandSchedutilPerformance15K30K45K60K75KSE +/- 42.21, N = 3SE +/- 21.41, N = 3SE +/- 36.63, N = 369039.369012.968976.9

TensorFlow Lite

Model: Mobilenet Quant

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantOndemandSchedutilPerformance15K30K45K60K75KSE +/- 45.84, N = 3SE +/- 30.64, N = 3SE +/- 30.62, N = 370655.170760.670739.1

TensorFlow Lite

Model: Inception ResNet V2

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2OndemandSchedutilPerformance300K600K900K1200K1500KSE +/- 365.29, N = 3SE +/- 394.05, N = 3SE +/- 483.75, N = 3135144013506531350667

Timed GDB GNU Debugger Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 9.1Time To CompileOndemandSchedutilPerformance20406080100SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.08, N = 395.4989.6183.71

Timed Godot Game Engine Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To CompileOndemandSchedutilPerformance20406080100SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 384.4783.3183.16

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2019-12-17H.264 Video EncodingOndemandSchedutilPerformance4080120160200SE +/- 1.12, N = 3SE +/- 1.39, N = 3SE +/- 1.40, N = 3172.64172.60176.541. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

x265

Video Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KOndemandSchedutilPerformance510152025SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 321.6521.0521.501. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

x265

Video Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pOndemandSchedutilPerformance1530456075SE +/- 0.15, N = 3SE +/- 0.34, N = 3SE +/- 0.12, N = 365.4755.0768.481. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Phoronix Test Suite v10.8.5