Intel Tiger Lake Driver Comparison

Tiger Lake Gen12 Linux driver comparison 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 2011304-HA-TIGERLAKE60
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
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
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Linux 5.8 + Mesa 20.2
November 29 2020
  4 Hours, 28 Minutes
Linux 5.8 + Mesa 21.0-dev
November 29 2020
  19 Hours, 55 Minutes
Linux 5.9.10 + Mesa 21.0-dev
November 30 2020
  3 Hours, 5 Minutes
Linux 5.10-rc6 + Mesa 21.0-dev
November 30 2020
  3 Hours, 6 Minutes
Invert Behavior (Only Show Selected Data)
  7 Hours, 39 Minutes

Only show results where is faster than
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):


Intel Tiger Lake Driver ComparisonProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen ResolutionLinux 5.8 + Mesa 20.2Linux 5.8 + Mesa 21.0-devLinux 5.9.10 + Mesa 21.0-devLinux 5.10-rc6 + Mesa 21.0-devIntel Core i7-1165G7 @ 4.70GHz (4 Cores / 8 Threads)Dell 0GG9PT (1.0.3 BIOS)Intel Tiger Lake-LP16GBKioxia KBG40ZNS256G NVMe 256GBIntel UHD 3GB (1300MHz)Realtek ALC289Intel Wi-Fi 6 AX201Ubuntu 20.105.8.0-29-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.94.6 Mesa 20.2.11.2.145GCC 10.2.0ext41920x12004.6 Mesa 21.0.0-devel (git-b3504f4 2020-11-29 groovy-oibaf-ppa)5.9.10-050910-generic (x86_64)5.10.0-051000rc6-generic (x86_64) 20201129OpenBenchmarking.orgCompiler 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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,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- Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3Python Details- Python 3.8.6Security 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Linux 5.8 + Mesa 20.2Linux 5.8 + Mesa 21.0-devLinux 5.9.10 + Mesa 21.0-devLinux 5.10-rc6 + Mesa 21.0-devResult OverviewPhoronix Test Suite100%102%104%107%GeekbenchUnigine SuperpositionOpenArenaTesseractyquake2XonoticBasemark GPUGpuTestET: LegacyNCNNGLmark2Unigine HeavenLibplaceboBetsy GPU CompressorRealSR-NCNNWaifu2x-NCNN Vulkan

Intel Tiger Lake Driver Comparisonlibplacebo: deband_heavylibplacebo: polar_nocomputelibplacebo: av1_grain_lapbasemark: OpenGL - 1920 x 1200 - Highbasemark: Vulkan - 1920 x 1200 - Highbasemark: OpenGL - 1920 x 1200 - Mediumbasemark: Vulkan - 1920 x 1200 - Mediumetlegacy: Renderer2 - 1920 x 1200openarena: 1920 x 1080tesseract: 1920 x 1200unigine-heaven: 1920 x 1200 - Fullscreen - OpenGLunigine-super: 1920 x 1200 - Fullscreen - Low - OpenGLxonotic: 1920 x 1200 - Ultimateyquake2: OpenGL 3.x - 1920 x 1080gputest: Plot3D - 1920 x 1200 - Fullscreengputest: Furmark - 1920 x 1200 - Fullscreengputest: TessMark - 1920 x 1200 - Fullscreengputest: Triangle - 1920 x 1200 - Fullscreengputest: Pixmark Piano - 1920 x 1200 - Fullscreengputest: Pixmark Volplosion - 1920 x 1200 - Fullscreenglmark2: 1920 x 1080glmark2: 1920 x 1200geekbench: GPU Vulkanncnn: Vulkan GPU - squeezenetncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - yolov4-tinyrealsr-ncnn: 4x - Norealsr-ncnn: 4x - Yeswaifu2x-ncnn: 2x - 3 - Yesbetsy: ETC1 - Highestbetsy: ETC2 RGB - HighestLinux 5.8 + Mesa 20.2Linux 5.8 + Mesa 21.0-devLinux 5.9.10 + Mesa 21.0-devLinux 5.10-rc6 + Mesa 21.0-dev144.44121.34890.4830.3927.8174.09143.93128.8187.6136.841128.328627.2126.7074579747.61059817184871749215741531896835825511.7710.498.0438.789.5415.4339.1568.017544.05327.32923.73432.312144.48124.05887.6930.9628.1274.28147.06126.8202.4140.792028.504429.3120.8333776747.91103017755047779255801593895833898711.9210.488.1638.779.5515.4839.9268.420543.66227.30823.87432.443144.65124.23879.1431.5128.9374.53154.96131.0191.8146.221828.498029.5121.8912214786.51098217755039790865821595909846871711.659.937.7638.799.5315.4839.4868.200541.47627.31123.75832.333144.36123.19874.6031.3428.3574.37147.36130.5188.8145.053828.251129.2122.7764422781.61096817745047785985831601904836880012.1010.698.3638.859.5815.6038.3768.445541.90327.25023.73832.277OpenBenchmarking.org

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 2.72.2Test: deband_heavyLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev306090120150SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.12, N = 3144.65144.48144.44144.361. (CXX) g++ options: -lm -lglslang -lHLSL -lOGLCompiler -lOSDependent -lSPIRV -lSPVRemapper -lSPIRV-Tools -lSPIRV-Tools-opt -lpthread -pthread -pipe -std=c++11 -fvisibility=hidden -fPIC -MD -MQ -MF

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 2.72.2Test: polar_nocomputeLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev306090120150SE +/- 0.09, N = 3SE +/- 0.20, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 3124.23124.05121.34123.191. (CXX) g++ options: -lm -lglslang -lHLSL -lOGLCompiler -lOSDependent -lSPIRV -lSPVRemapper -lSPIRV-Tools -lSPIRV-Tools-opt -lpthread -pthread -pipe -std=c++11 -fvisibility=hidden -fPIC -MD -MQ -MF

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 2.72.2Test: av1_grain_lapLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev2004006008001000SE +/- 0.48, N = 3SE +/- 4.14, N = 3SE +/- 6.35, N = 3SE +/- 19.20, N = 3879.14887.69890.48874.601. (CXX) g++ options: -lm -lglslang -lHLSL -lOGLCompiler -lOSDependent -lSPIRV -lSPVRemapper -lSPIRV-Tools -lSPIRV-Tools-opt -lpthread -pthread -pipe -std=c++11 -fvisibility=hidden -fPIC -MD -MQ -MF

Basemark GPU

This is a benchmark of Basemark GPU. For this test profile to work, you must have a valid license/copy of BasemarkGPU in your Phoronix Test Suite download cache. This test profile simply automates the execution of BasemarkGPU and you must already have the Windows .zip or Linux .tar.gz in the download cache. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterBasemark GPU 1.2Renderer: OpenGL - Resolution: 1920 x 1200 - Graphics Preset: HighLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev714212835SE +/- 0.44, N = 3SE +/- 0.39, N = 3SE +/- 0.27, N = 3SE +/- 0.45, N = 331.5130.9630.3931.34MIN: 17.16 / MAX: 77.95MIN: 21.21 / MAX: 76.96MIN: 20.13 / MAX: 74.05MIN: 21.73 / MAX: 77.77

OpenBenchmarking.orgFPS, More Is BetterBasemark GPU 1.2Renderer: Vulkan - Resolution: 1920 x 1200 - Graphics Preset: HighLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev714212835SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 328.9328.1227.8128.35MIN: 21.33 / MAX: 60.27MIN: 20.98 / MAX: 58.52MIN: 20.7 / MAX: 59.17MIN: 22.42 / MAX: 48.24

OpenBenchmarking.orgFPS, More Is BetterBasemark GPU 1.2Renderer: OpenGL - Resolution: 1920 x 1200 - Graphics Preset: MediumLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev20406080100SE +/- 0.21, N = 3SE +/- 0.22, N = 3SE +/- 0.29, N = 3SE +/- 0.13, N = 374.5374.2874.0974.37MIN: 23.62 / MAX: 292.85MIN: 20.68 / MAX: 322.43MIN: 21.56 / MAX: 277.33MIN: 21.35 / MAX: 301.37

OpenBenchmarking.orgFPS, More Is BetterBasemark GPU 1.2Renderer: Vulkan - Resolution: 1920 x 1200 - Graphics Preset: MediumLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev306090120150SE +/- 1.51, N = 15SE +/- 1.44, N = 14SE +/- 1.11, N = 15SE +/- 1.37, N = 15154.96147.06143.93147.36MIN: 76.26 / MAX: 295.79MIN: 75.37 / MAX: 275.24MIN: 72.5 / MAX: 278.66MIN: 72.66 / MAX: 284.9

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 2.75Renderer: Renderer2 - Resolution: 1920 x 1200Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev306090120150SE +/- 1.67, N = 4SE +/- 1.30, N = 3SE +/- 1.17, N = 3SE +/- 1.79, N = 4131.0126.8128.8130.5

OpenArena

This is a test of OpenArena, a popular open-source first-person shooter. This game is based upon ioquake3, which in turn uses the GPL version of id Software's Quake 3 engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterOpenArena 0.8.8Resolution: 1920 x 1080Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev4080120160200SE +/- 1.91, N = 15SE +/- 3.97, N = 3SE +/- 1.81, N = 15SE +/- 1.85, N = 15191.8202.4187.6188.8MIN: 1MIN: 1MIN: 2MIN: 1

Tesseract

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

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 1920 x 1200Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev306090120150SE +/- 1.90, N = 4SE +/- 0.80, N = 3SE +/- 0.47, N = 3SE +/- 2.02, N = 3146.22140.79136.84145.05

Unigine Heaven

This test calculates the average frame-rate within the Heaven demo for the Unigine engine. This engine is extremely demanding on the system's graphics card. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Heaven 4.0Resolution: 1920 x 1200 - Mode: Fullscreen - Renderer: OpenGLLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev714212835SE +/- 0.07, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 328.5028.5028.3328.25

Unigine Superposition

This test calculates the average frame-rate within the Superposition demo for the Unigine engine, released in 2017. This engine is extremely demanding on the system's graphics card. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 1920 x 1200 - Mode: Fullscreen - Quality: Low - Renderer: OpenGLLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev714212835SE +/- 0.13, N = 3SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 329.529.327.229.2MAX: 40.1MAX: 39.9MAX: 37MAX: 39.8

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: 1920 x 1200 - Effects Quality: UltimateLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev306090120150SE +/- 0.57, N = 3SE +/- 0.47, N = 3SE +/- 0.50, N = 3SE +/- 0.33, N = 3121.89120.83126.71122.78MIN: 34 / MAX: 210MIN: 34 / MAX: 210MIN: 36 / MAX: 227MIN: 37 / MAX: 211

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 7.45Renderer: OpenGL 3.x - Resolution: 1920 x 1080Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev2004006008001000SE +/- 7.08, N = 3SE +/- 8.92, N = 6SE +/- 10.18, N = 3SE +/- 2.52, N = 3786.5747.9747.6781.61. (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

GpuTest

GpuTest is a cross-platform OpenGL benchmark developed at Geeks3D.com that offers tech demos such as FurMark, TessMark, and other workloads to stress various areas of GPUs and drivers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgPoints, More Is BetterGpuTest 0.7.0Test: Plot3D - Resolution: 1920 x 1200 - Mode: FullscreenLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev2K4K6K8K10KSE +/- 6.17, N = 3SE +/- 0.33, N = 3SE +/- 2.67, N = 310982110301059810968

OpenBenchmarking.orgPoints, More Is BetterGpuTest 0.7.0Test: Furmark - Resolution: 1920 x 1200 - Mode: FullscreenLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev400800120016002000SE +/- 24.06, N = 4SE +/- 30.28, N = 3SE +/- 28.62, N = 3SE +/- 22.16, N = 51775177517181774

OpenBenchmarking.orgPoints, More Is BetterGpuTest 0.7.0Test: TessMark - Resolution: 1920 x 1200 - Mode: FullscreenLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev11002200330044005500SE +/- 14.19, N = 3SE +/- 15.17, N = 3SE +/- 16.07, N = 3SE +/- 19.22, N = 35039504748715047

OpenBenchmarking.orgPoints, More Is BetterGpuTest 0.7.0Test: Triangle - Resolution: 1920 x 1200 - Mode: FullscreenLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev20K40K60K80K100KSE +/- 73.33, N = 3SE +/- 94.41, N = 3SE +/- 258.92, N = 3SE +/- 58.05, N = 379086779257492178598

OpenBenchmarking.orgPoints, More Is BetterGpuTest 0.7.0Test: Pixmark Piano - Resolution: 1920 x 1200 - Mode: FullscreenLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev130260390520650SE +/- 3.33, N = 3SE +/- 3.18, N = 3SE +/- 3.18, N = 3SE +/- 4.51, N = 3582580574583

OpenBenchmarking.orgPoints, More Is BetterGpuTest 0.7.0Test: Pixmark Volplosion - Resolution: 1920 x 1200 - Mode: FullscreenLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev30060090012001500SE +/- 10.54, N = 3SE +/- 9.13, N = 3SE +/- 10.35, N = 3SE +/- 7.51, N = 31595159315311601

GLmark2

This is a test of Linaro's glmark2 port, currently using the X11 OpenGL 2.0 target. GLmark2 is a basic OpenGL benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterGLmark2 2020.04Resolution: 1920 x 1080Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev2004006008001000909895896904

OpenBenchmarking.orgScore, More Is BetterGLmark2 2020.04Resolution: 1920 x 1200Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev2004006008001000846833835836

Geekbench

This is a benchmark of Geekbench 5 Pro. The test profile automates the execution of Geekbench 5 under the Phoronix Test Suite, assuming you have a valid license key for Geekbench 5 Pro. This test will not work without a valid license key for Geekbench Pro. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterGeekbench 5Test: GPU VulkanLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev2K4K6K8K10KSE +/- 139.64, N = 3SE +/- 90.43, N = 15SE +/- 77.21, N = 15SE +/- 110.05, N = 58717898782558800

OpenArena

OpenBenchmarking.orgMilliseconds, Fewer Is BetterOpenArena 0.8.8Resolution: 1920 x 1080 - Total Frame TimeLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev612182430Min: 2 / Avg: 5.26 / Max: 26Min: 2 / Avg: 4.8 / Max: 26Min: 2 / Avg: 5.38 / Max: 25Min: 2 / Avg: 5.33 / Max: 25

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 20200916Target: Vulkan GPU - Model: squeezenetLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev3691215SE +/- 0.13, N = 3SE +/- 0.20, N = 15SE +/- 0.13, N = 15SE +/- 0.16, N = 511.6511.9211.7712.10MIN: 11.14 / MAX: 13.32MIN: 11.16 / MAX: 39.31MIN: 11.18 / MAX: 22.08MIN: 11.4 / MAX: 12.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev3691215SE +/- 0.23, N = 3SE +/- 0.14, N = 15SE +/- 0.06, N = 15SE +/- 0.19, N = 59.9310.4810.4910.69MIN: 9.44 / MAX: 11.86MIN: 8.85 / MAX: 12.17MIN: 10.1 / MAX: 12.17MIN: 10.14 / MAX: 12.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: shufflenet-v2Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev246810SE +/- 0.07, N = 3SE +/- 0.06, N = 15SE +/- 0.06, N = 15SE +/- 0.10, N = 57.768.168.048.36MIN: 7.43 / MAX: 9.33MIN: 5.84 / MAX: 11.31MIN: 6.72 / MAX: 11MIN: 7.49 / MAX: 9.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: vgg16Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev918273645SE +/- 0.01, N = 3SE +/- 0.01, N = 15SE +/- 0.01, N = 15SE +/- 0.02, N = 538.7938.7738.7838.85MIN: 38.46 / MAX: 39.35MIN: 38.34 / MAX: 39.31MIN: 38.36 / MAX: 39.43MIN: 38.5 / MAX: 39.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: alexnetLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev3691215SE +/- 0.02, N = 3SE +/- 0.01, N = 15SE +/- 0.01, N = 15SE +/- 0.02, N = 59.539.559.549.58MIN: 9.09 / MAX: 9.84MIN: 8.86 / MAX: 9.92MIN: 9.21 / MAX: 10.36MIN: 8.92 / MAX: 9.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet50Linux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 15SE +/- 0.01, N = 15SE +/- 0.02, N = 515.4815.4815.4315.60MIN: 15.08 / MAX: 15.94MIN: 15.1 / MAX: 16.07MIN: 15.1 / MAX: 16.01MIN: 15.38 / MAX: 16.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: yolov4-tinyLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev918273645SE +/- 0.54, N = 3SE +/- 0.20, N = 15SE +/- 0.24, N = 15SE +/- 0.83, N = 539.4839.9239.1538.37MIN: 26.29 / MAX: 53.21MIN: 25.8 / MAX: 54.91MIN: 25.22 / MAX: 56.04MIN: 25.74 / MAX: 54.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

RealSR-NCNN

RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev1530456075SE +/- 0.55, N = 3SE +/- 0.55, N = 3SE +/- 0.85, N = 3SE +/- 0.66, N = 368.2068.4268.0268.45

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev120240360480600SE +/- 0.60, N = 3SE +/- 0.94, N = 3SE +/- 1.21, N = 3SE +/- 0.12, N = 3541.48543.66544.05541.90

Waifu2x-NCNN Vulkan

Waifu2x-NCNN is an NCNN neural network implementation of the Waifu2x converter project and accelerated using the Vulkan API. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev612182430SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 327.3127.3127.3327.25

Betsy GPU Compressor

Betsy is an open-source GPU compressor of various GPU compression techniques. Betsy is written in GLSL for Vulkan/OpenGL (compute shader) support for GPU-based texture compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC1 - Quality: HighestLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev612182430SE +/- 0.02, N = 3SE +/- 0.11, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 323.7623.8723.7323.741. (CXX) g++ options: -O3 -O2 -lpthread -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC2 RGB - Quality: HighestLinux 5.9.10 + Mesa 21.0-devLinux 5.8 + Mesa 21.0-devLinux 5.8 + Mesa 20.2Linux 5.10-rc6 + Mesa 21.0-dev816243240SE +/- 0.03, N = 3SE +/- 0.16, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 332.3332.4432.3132.281. (CXX) g++ options: -O3 -O2 -lpthread -ldl