AMD Radeon September 2023

AMD Ryzen 9 7950X 16-Core testing with a ASUS ROG STRIX X670E-E GAMING WIFI (1416 BIOS) and AMD Radeon RX 7900 XTX 24GB on Ubuntu 23.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2309019-SYST-AMDRADE68
Jump To Table - Results

Statistics

Remove Outliers Before Calculating Averages

Graph Settings

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
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
RX 7900 XTX
August 31 2023
  7 Hours, 33 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 Radeon September 2023OpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads)ASUS ROG STRIX X670E-E GAMING WIFI (1416 BIOS)AMD Device 14d832GBWestern Digital WD_BLACK SN850X 1000GB + 4001GB Western Digital WD_BLACK SN850X 4000GBAMD Radeon RX 7900 XTX 24GB (2304/1249MHz)AMD Device ab30ASUS MG28UIntel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411Ubuntu 23.046.3.13-060313-generic (x86_64)GNOME Shell 44.3X Server 1.21.1.7 + Wayland4.6 Mesa 23.3~git2308310600.c8687a~oibaf~l (git-c8687a4 2023-08-31 lunar-oibaf-ppa) (LLVM 15.0.7 DRM 3.52)GCC 12.3.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionAMD Radeon September 2023 BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-DAPbBt/gcc-12-12.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-DAPbBt/gcc-12-12.3.0/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-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 - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203- BAR1 / Visible vRAM Size: 24560 MB - vBIOS Version: 113-D7020100-102- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

AMD Radeon September 2023csgo: 2560 x 1440 - Vulkanf122: 2560 x 1440 - Lowf122: 2560 x 1440 - Mediumf122: 2560 x 1440 - Highf122: 2560 x 1440 - Ultra High3dmark: 2560 x 1440gravitymark: 2560 x 1440 - OpenGLgravitymark: 2560 x 1440 - Vulkangravitymark: 2560 x 1440 - Vulkan Ray-Tracingunigine-super: 2560 x 1440 - Fullscreen - Low - OpenGLunigine-super: 2560 x 1440 - Fullscreen - Medium - OpenGLunigine-super: 2560 x 1440 - Fullscreen - High - OpenGLunigine-super: 2560 x 1440 - Fullscreen - Ultra - OpenGLunigine-heaven: 2560 x 1440 - Fullscreen - OpenGLquake2rtx: 2560 x 1440 - Off - Off - VK_KHR_ray_queryquake2rtx: 2560 x 1440 - Off - On - VK_KHR_ray_queryquake2rtx: 2560 x 1440 - Medium - Off - VK_KHR_ray_queryquake2rtx: 2560 x 1440 - Medium - On - VK_KHR_ray_queryvkmark: 2560 x 1440 - Mailboxspecviewperf2020: 2560 x 1440 - CATIA-06specviewperf2020: 2560 x 1440 - CREO-03specviewperf2020: 2560 x 1440 - MAYA-06specviewperf2020: 2560 x 1440 - MEDICAL-O3specviewperf2020: 2560 x 1440 - SNX-04specviewperf2020: 2560 x 1440 - SOLIDWORKS-07vkfft: FFT + iFFT C2C 1D batched in single precisionvkfft: FFT + iFFT C2C 1D batched in double precisionvkfft: FFT + iFFT C2C 1D batched in half precisionvkfft: FFT + iFFT C2C multidimensional in single precisionvkfft: FFT + iFFT C2C 1D batched in single precision, no reshufflingvkfft: FFT + iFFT R2C / C2Rvkfft: FFT + iFFT C2C Bluestein in single precisionvkfft: FFT + iFFT C2C Bluestein benchmark in double precisionvkresample: 2x - Singlencnn: Vulkan GPU - mobilenetncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - googlenetncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - vision_transformerncnn: Vulkan GPU - FastestDetvkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp16-scalarvkpeak: fp16-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4vkpeak: int32-scalarvkpeak: int32-vec4vkpeak: int16-scalarvkpeak: int16-vec4paraview: Many Spheres - 2560 x 1440paraview: Many Spheres - 2560 x 1440paraview: Wavelet Volume - 2560 x 1440paraview: Wavelet Volume - 2560 x 1440paraview: Wavelet Contour - 2560 x 1440paraview: Wavelet Contour - 2560 x 1440RX 7900 XTX482.76456426333271111.04226.0325.6244.2380.461.5168.940.597.971027.62616427.07650923.62057523.1452071561466.9615.75673.3247.34302.27452.221193933397021140878022127886880962173066595.0928.063.243.263.403.033.981.398.0723.685.274.3210.2512.897.128.3132.054.0525692.7923932.9827385.4950435.371193.421191.377048.647069.1026906.9849757.5096.329656.906921.7914748.669462.114815.718OpenBenchmarking.org

Counter-Strike: Global Offensive

This is a benchmark of Valve's Counter-Strike: Global Offensive game. The test profile assumes you have a Steam account, have Steam installed for the system, and that Counter-Strike: Global Offensive is already installed. This automates the process of executing the game and using a standardized time demo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterCounter-Strike: Global OffensiveResolution: 2560 x 1440 - Renderer: VulkanRX 7900 XTX100200300400500SE +/- 6.41, N = 2482.76

F1 22

This is a benchmark of F1 22 (2022) running under Steam Play (Proton) on Linux. The test profile assumes you have a Steam account, have Steam installed for the system, and that the game is already purchased/installed. This automates the process of executing the game and using a standardized time demo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterF1 22Resolution: 2560 x 1440 - Graphics Preset: LowRX 7900 XTX100200300400500SE +/- 1.45, N = 3456MIN: 306 / MAX: 582

OpenBenchmarking.orgMilliseconds, Fewer Is BetterF1 22Resolution: 2560 x 1440 - Graphics Preset: Low - Total Frame TimeRX 7900 XTX3691215Min: 0.91 / Avg: 2.19 / Max: 6.69

OpenBenchmarking.orgFrames Per Second, More Is BetterF1 22Resolution: 2560 x 1440 - Graphics Preset: MediumRX 7900 XTX90180270360450SE +/- 1.53, N = 3426MIN: 318 / MAX: 494

OpenBenchmarking.orgMilliseconds, Fewer Is BetterF1 22Resolution: 2560 x 1440 - Graphics Preset: Medium - Total Frame TimeRX 7900 XTX3691215Min: 1.02 / Avg: 2.35 / Max: 8.22

OpenBenchmarking.orgFrames Per Second, More Is BetterF1 22Resolution: 2560 x 1440 - Graphics Preset: HighRX 7900 XTX70140210280350SE +/- 1.67, N = 3333MIN: 224 / MAX: 395

OpenBenchmarking.orgMilliseconds, Fewer Is BetterF1 22Resolution: 2560 x 1440 - Graphics Preset: High - Total Frame TimeRX 7900 XTX3691215Min: 1.31 / Avg: 2.98 / Max: 8.81

OpenBenchmarking.orgFrames Per Second, More Is BetterF1 22Resolution: 2560 x 1440 - Graphics Preset: Ultra HighRX 7900 XTX60120180240300SE +/- 0.33, N = 3271MIN: 223 / MAX: 324

OpenBenchmarking.orgMilliseconds, Fewer Is BetterF1 22Resolution: 2560 x 1440 - Graphics Preset: Ultra High - Total Frame TimeRX 7900 XTX3691215Min: 1.78 / Avg: 3.67 / Max: 9.09

Strange Brigade

This is a benchmark of Strange Brigade running under Steam Play (Proton) on Linux. The test profile assumes you have a Steam account, have Steam installed for the system, and that the game is already purchased/installed. This automates the process of executing the game and using a standardized scene. Learn more via the OpenBenchmarking.org test page.

Resolution: 2560 x 1440 - Graphics Setting: Medium

RX 7900 XTX: The test run did not produce a result. E: sh: 1: exec: ./strange-brigade: not found

Resolution: 2560 x 1440 - Graphics Setting: High

RX 7900 XTX: The test run did not produce a result. E: sh: 1: exec: ./strange-brigade: not found

Resolution: 2560 x 1440 - Graphics Setting: Ultra

RX 7900 XTX: The test run did not produce a result. E: sh: 1: exec: ./strange-brigade: not found

3DMark Wild Life Extreme

This test profile only automates the vendor build of 3DMark with its command-line / JSON support. If you do not have a licensed copy of the necessary 3DMark binaries in your Phoronix Test Suite download cache on your system, this test profile will not do anything and simply fail. You must have already obtained the proper licensed binaries from UL for this test profile to work -- this test profile simply automates the firing of the 3DMark benchmark at your desired resolution and capturing the results within the Phoronix Test Suite while you must already have the necessary 3DMark files on your system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Better3DMark Wild Life Extreme 1.1.2.1Resolution: 2560 x 1440RX 7900 XTX20406080100SE +/- 0.05, N = 3111.04

GravityMark

GravityMark is a cross-API, cross-platform GPU accelerated benchmark developed by Tellusim. GravityMark aims to exploit the performance of modern GPUs and render hundreds of thousands of objects in real-time all using GPU acceleration. GravityMark supports a variety of graphics rendering APIs and optionally supports accelerated ray-tracing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.82Resolution: 2560 x 1440 - Renderer: OpenGLRX 7900 XTX50100150200250SE +/- 0.85, N = 3226.0

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.82Resolution: 2560 x 1440 - Renderer: VulkanRX 7900 XTX70140210280350SE +/- 1.62, N = 3325.6

OpenBenchmarking.orgFrames Per Second, More Is BetterGravityMark 1.82Resolution: 2560 x 1440 - Renderer: Vulkan Ray-TracingRX 7900 XTX50100150200250SE +/- 0.95, N = 3244.2

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: 2560 x 1440 - Mode: Fullscreen - Quality: Low - Renderer: OpenGLRX 7900 XTX80160240320400SE +/- 1.01, N = 3380.4MAX: 558.4

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 2560 x 1440 - Mode: Fullscreen - Quality: Medium - Renderer: OpenGLRX 7900 XTX1428425670SE +/- 0.00, N = 361.5MAX: 112.5

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 2560 x 1440 - Mode: Fullscreen - Quality: High - Renderer: OpenGLRX 7900 XTX4080120160200SE +/- 0.17, N = 3168.9MAX: 209.8

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 2560 x 1440 - Mode: Fullscreen - Quality: Ultra - Renderer: OpenGLRX 7900 XTX918273645SE +/- 0.00, N = 340.5MAX: 66.1

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: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGLRX 7900 XTX20406080100SE +/- 0.05, N = 397.97

Quake II RTX

NVIDIA's Quake II RTX ray-traced (RTX path-traced global illumination) version of the original Quake II game. Quake II RTX makes use of Vulkan ray-tracing extensions on Windows and Linux. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQuake II RTX 1.6Resolution: 2560 x 1440 - Global Illumination: Off - Denoiser: Off - Ray Tracing API: VK_KHR_ray_queryRX 7900 XTX612182430SE +/- 0.00, N = 327.63

OpenBenchmarking.orgFrames Per Second, More Is BetterQuake II RTX 1.6Resolution: 2560 x 1440 - Global Illumination: Off - Denoiser: On - Ray Tracing API: VK_KHR_ray_queryRX 7900 XTX612182430SE +/- 0.00, N = 327.08

OpenBenchmarking.orgFrames Per Second, More Is BetterQuake II RTX 1.6Resolution: 2560 x 1440 - Global Illumination: Medium - Denoiser: Off - Ray Tracing API: VK_KHR_ray_queryRX 7900 XTX612182430SE +/- 0.01, N = 323.62

OpenBenchmarking.orgFrames Per Second, More Is BetterQuake II RTX 1.6Resolution: 2560 x 1440 - Global Illumination: Medium - Denoiser: On - Ray Tracing API: VK_KHR_ray_queryRX 7900 XTX612182430SE +/- 0.01, N = 323.15

VKMark

VKMark is a collection of open-source Vulkan tests / rendering benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2022-05-16Resolution: 2560 x 1440 - Present Mode: MailboxRX 7900 XTX3K6K9K12K15KSE +/- 4.26, N = 3156141. (CXX) g++ options: -pthread -ldl -std=c++14 -O0 -MD -MQ -MF

Resolution: 2560 x 1440 - Present Mode: Immediate

RX 7900 XTX: The test quit with a non-zero exit status. E: Error: Selected present mode Immediate is not supported by the used Vulkan physical device.

Meta Performance Per Watts

OpenBenchmarking.orgPerformance Per Watts, More Is BetterMeta Performance Per WattsPerformance Per WattsRX 7900 XTX12K24K36K48K60K123.85

GPU Power Consumption Monitor

OpenBenchmarking.orgWattsGPU Power Consumption MonitorPhoronix Test Suite System MonitoringRX 7900 XTX60120180240300Min: 15 / Avg: 202.66 / Max: 313

GPU Temperature Monitor

OpenBenchmarking.orgCelsiusGPU Temperature MonitorPhoronix Test Suite System MonitoringRX 7900 XTX1530456075Min: 41 / Avg: 65.28 / Max: 77

SPECViewPerf 2020

This test runs SPECViewPerf 2020 if available on your system. SPECViewPerf is made up of real-world OpenGL workstation tests such as CATIA and SolidWorks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: CATIA-06RX 7900 XTX1530456075SE +/- 0.16, N = 366.96

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: CREO-03RX 7900 XTX48121620SE +/- 0.00, N = 315.75

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: MAYA-06RX 7900 XTX150300450600750SE +/- 3.30, N = 3673.32

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: MEDICAL-O3RX 7900 XTX1122334455SE +/- 0.01, N = 347.34

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: SNX-04RX 7900 XTX70140210280350SE +/- 0.04, N = 3302.27

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: SOLIDWORKS-07RX 7900 XTX100200300400500SE +/- 0.31, N = 3452.22

VkFFT

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in single precisionRX 7900 XTX30K60K90K120K150KSE +/- 121.83, N = 31193931. (CXX) g++ options: -O3

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in double precisionRX 7900 XTX7K14K21K28K35KSE +/- 55.32, N = 3339701. (CXX) g++ options: -O3

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in half precisionRX 7900 XTX50K100K150K200K250KSE +/- 94.57, N = 32114081. (CXX) g++ options: -O3

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C multidimensional in single precisionRX 7900 XTX20K40K60K80K100KSE +/- 662.95, N = 5780221. (CXX) g++ options: -O3

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in single precision, no reshufflingRX 7900 XTX30K60K90K120K150KSE +/- 130.77, N = 31278861. (CXX) g++ options: -O3

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT R2C / C2RRX 7900 XTX20K40K60K80K100KSE +/- 587.21, N = 15880961. (CXX) g++ options: -O3

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C Bluestein in single precisionRX 7900 XTX5K10K15K20K25KSE +/- 169.09, N = 3217301. (CXX) g++ options: -O3

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C Bluestein benchmark in double precisionRX 7900 XTX14002800420056007000SE +/- 47.54, N = 366591. (CXX) g++ options: -O3

VkResample

VkResample is a Vulkan-based image upscaling library based on VkFFT. The sample input file is upscaling a 4K image to 8K using Vulkan-based GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleRX 7900 XTX1.14572.29143.43714.58285.7285SE +/- 0.002, N = 75.0921. (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 20230517Target: Vulkan GPU - Model: mobilenetRX 7900 XTX246810SE +/- 0.00, N = 38.06MIN: 8 / MAX: 11.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2RX 7900 XTX0.7291.4582.1872.9163.645SE +/- 0.07, N = 33.24MIN: 3.11 / MAX: 7.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3RX 7900 XTX0.73351.4672.20052.9343.6675SE +/- 0.09, N = 23.26MIN: 3.13 / MAX: 4.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: shufflenet-v2RX 7900 XTX0.7651.532.2953.063.825SE +/- 0.04, N = 33.40MIN: 3.31 / MAX: 4.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mnasnetRX 7900 XTX0.68181.36362.04542.72723.409SE +/- 0.06, N = 33.03MIN: 2.93 / MAX: 4.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: efficientnet-b0RX 7900 XTX0.89551.7912.68653.5824.4775SE +/- 0.11, N = 33.98MIN: 3.81 / MAX: 5.641. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: blazefaceRX 7900 XTX0.31280.62560.93841.25121.564SE +/- 0.01, N = 31.39MIN: 1.35 / MAX: 2.881. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: googlenetRX 7900 XTX246810SE +/- 0.17, N = 38.07MIN: 7.72 / MAX: 13.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vgg16RX 7900 XTX612182430SE +/- 0.09, N = 323.68MIN: 23.28 / MAX: 33.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet18RX 7900 XTX1.18582.37163.55744.74325.929SE +/- 0.03, N = 35.27MIN: 5.09 / MAX: 9.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: alexnetRX 7900 XTX0.9721.9442.9163.8884.86SE +/- 0.00, N = 34.32MIN: 4.26 / MAX: 5.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet50RX 7900 XTX3691215SE +/- 0.07, N = 310.25MIN: 10.02 / MAX: 12.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: yolov4-tinyRX 7900 XTX3691215SE +/- 0.07, N = 312.89MIN: 12.68 / MAX: 18.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: squeezenet_ssdRX 7900 XTX246810SE +/- 0.02, N = 37.12MIN: 7.02 / MAX: 8.481. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: regnety_400mRX 7900 XTX246810SE +/- 0.07, N = 38.31MIN: 8.1 / MAX: 11.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vision_transformerRX 7900 XTX714212835SE +/- 0.04, N = 332.05MIN: 31.72 / MAX: 45.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: FastestDetRX 7900 XTX0.91131.82262.73393.64524.5565SE +/- 0.05, N = 34.05MIN: 3.91 / MAX: 5.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

vkpeak

Vkpeak is a Vulkan compute benchmark inspired by OpenCL's clpeak. Vkpeak provides Vulkan compute performance measurements for FP16 / FP32 / FP64 / INT16 / INT32 scalar and vec4 performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-scalarRX 7900 XTX6K12K18K24K30KSE +/- 68.88, N = 325692.79

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-vec4RX 7900 XTX5K10K15K20K25KSE +/- 2.96, N = 323932.98

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-scalarRX 7900 XTX6K12K18K24K30KSE +/- 63.34, N = 327385.49

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-vec4RX 7900 XTX11K22K33K44K55KSE +/- 4.13, N = 350435.37

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-scalarRX 7900 XTX30060090012001500SE +/- 0.16, N = 31193.42

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-vec4RX 7900 XTX30060090012001500SE +/- 0.13, N = 31191.37

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-scalarRX 7900 XTX15003000450060007500SE +/- 0.84, N = 37048.64

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-vec4RX 7900 XTX15003000450060007500SE +/- 2.23, N = 37069.10

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-scalarRX 7900 XTX6K12K18K24K30KSE +/- 45.54, N = 326906.98

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-vec4RX 7900 XTX11K22K33K44K55KSE +/- 2.44, N = 349757.50

ParaView

This test runs ParaView benchmarks: an open-source data analytics and visualization application. Paraview describes itself as "an open-source, multi-platform data analysis and visualization application. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.10.1Test: Many Spheres - Resolution: 2560 x 1440RX 7900 XTX20406080100SE +/- 0.06, N = 396.32

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.10.1Test: Wavelet Volume - Resolution: 2560 x 1440RX 7900 XTX2004006008001000SE +/- 6.69, N = 3921.79

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.10.1Test: Wavelet Contour - Resolution: 2560 x 1440RX 7900 XTX100200300400500SE +/- 1.80, N = 3462.11

71 Results Shown

Counter-Strike: Global Offensive
F1 22
F1 22
F1 22
F1 22
F1 22
F1 22
F1 22
F1 22
3DMark Wild Life Extreme
GravityMark:
  2560 x 1440 - OpenGL
  2560 x 1440 - Vulkan
  2560 x 1440 - Vulkan Ray-Tracing
Unigine Superposition:
  2560 x 1440 - Fullscreen - Low - OpenGL
  2560 x 1440 - Fullscreen - Medium - OpenGL
  2560 x 1440 - Fullscreen - High - OpenGL
  2560 x 1440 - Fullscreen - Ultra - OpenGL
Unigine Heaven
Quake II RTX:
  2560 x 1440 - Off - Off - VK_KHR_ray_query
  2560 x 1440 - Off - On - VK_KHR_ray_query
  2560 x 1440 - Medium - Off - VK_KHR_ray_query
  2560 x 1440 - Medium - On - VK_KHR_ray_query
VKMark
Meta Performance Per Watts:
  Performance Per Watts
  Phoronix Test Suite System Monitoring
  Phoronix Test Suite System Monitoring
SPECViewPerf 2020:
  2560 x 1440 - CATIA-06
  2560 x 1440 - CREO-03
  2560 x 1440 - MAYA-06
  2560 x 1440 - MEDICAL-O3
  2560 x 1440 - SNX-04
  2560 x 1440 - SOLIDWORKS-07
VkFFT:
  FFT + iFFT C2C 1D batched in single precision
  FFT + iFFT C2C 1D batched in double precision
  FFT + iFFT C2C 1D batched in half precision
  FFT + iFFT C2C multidimensional in single precision
  FFT + iFFT C2C 1D batched in single precision, no reshuffling
  FFT + iFFT R2C / C2R
  FFT + iFFT C2C Bluestein in single precision
  FFT + iFFT C2C Bluestein benchmark in double precision
VkResample
NCNN:
  Vulkan GPU - mobilenet
  Vulkan GPU-v2-v2 - mobilenet-v2
  Vulkan GPU-v3-v3 - mobilenet-v3
  Vulkan GPU - shufflenet-v2
  Vulkan GPU - mnasnet
  Vulkan GPU - efficientnet-b0
  Vulkan GPU - blazeface
  Vulkan GPU - googlenet
  Vulkan GPU - vgg16
  Vulkan GPU - resnet18
  Vulkan GPU - alexnet
  Vulkan GPU - resnet50
  Vulkan GPU - yolov4-tiny
  Vulkan GPU - squeezenet_ssd
  Vulkan GPU - regnety_400m
  Vulkan GPU - vision_transformer
  Vulkan GPU - FastestDet
vkpeak:
  fp32-scalar
  fp32-vec4
  fp16-scalar
  fp16-vec4
  fp64-scalar
  fp64-vec4
  int32-scalar
  int32-vec4
  int16-scalar
  int16-vec4
ParaView:
  Many Spheres - 2560 x 1440
  Wavelet Volume - 2560 x 1440
  Wavelet Contour - 2560 x 1440