Feb 2021 Vulkan Compute

Vulkan benchmark comparison for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2102067-HA-FEB2021VU83
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:

NVIDIA GPU Compute 4 Tests
Vulkan Compute 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
Disable Color Branding
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Triggered
  Test
  Duration
RX 5700 XT
February 03
  18 Minutes
Radeon VII
February 03
  13 Minutes
RX 6800
February 02
  11 Minutes
RX 6800 XT
February 01
  11 Minutes
RTX 2070 SUPER
February 05
  20 Minutes
RTX 2080 SUPER
February 06
  22 Minutes
RTX 2080 Ti
February 05
  19 Minutes
GTX 3060 Ti
February 04
  18 Minutes
RTX 3080
February 04
  18 Minutes
Invert Hiding All Results Option
  17 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):


Feb 2021 Vulkan ComputeProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen ResolutionOpenCLRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 3080AMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3202 BIOS)AMD Starship/Matisse32GB2000GB Corsair Force MP600 + 2000GBAMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (2100/875MHz)AMD Navi 10 HDMI AudioASUS MG28URealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.105.11.0-051100rc5daily20210129-generic (x86_64) 20210128GNOME Shell 3.38.2X Server 1.20.9amd4.6 Mesa 21.1.0-devel (git-824ae64 2021-02-01 groovy-oibaf-ppa) (LLVM 11.0.1)1.2.145GCC 10.2.0 + Clang 11.0.1-1~oibaf~gext43840x2160AMD Radeon VII 16GB (1801/1000MHz)AMD Vega 20 HDMI AudioAMD SIENNA_CICHLID 16GB (2475/1000MHz)AMD Device ab28AMD SIENNA_CICHLID 16GB (2575/1000MHz)NVIDIA GeForce RTX 2070 SUPER 8GB (1605/7000MHz)NVIDIA TU104 HD Audio5.8.0-41-generic (x86_64)NVIDIA 460.394.6.0OpenCL 1.2 CUDA 11.2.1361.2.155NVIDIA GeForce RTX 2080 SUPER 8GB (1650/7750MHz)NVIDIA GeForce RTX 2080 Ti 11GB (420/405MHz)NVIDIA TU102 HD AudioNVIDIA GeForce RTX 3060 Ti 8GB (1665/7000MHz)NVIDIA Device 228bNVIDIA GeForce RTX 3080 10GB (390/405MHz)NVIDIA Device 1aefOpenBenchmarking.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++,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: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009Graphics Details- RX 5700 XT, Radeon VII, RX 6800, RX 6800 XT: GLAMORSecurity 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: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

RX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 3080Result OverviewPhoronix Test Suite 10.4.0m1100%141%182%222%263%VkResampleVkFFTRealSR-NCNNRealSR-NCNNWaifu2x-NCNN Vulkan2x - Single4x - Yes4x - No2x - 3 - Yes

Feb 2021 Vulkan Computerealsr-ncnn: 4x - Norealsr-ncnn: 4x - Yesvkfft: vkresample: 2x - Singlewaifu2x-ncnn: 2x - 3 - YesRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 30808.91858.0112067632.0645.0666.61239.4653391213.7235.7695.73332.2094480418.3753.4765.55530.5764973017.9473.27310.14964.3183190719.8794.8139.67958.4303250918.6294.6728.54845.7834163515.0134.0219.39858.5513208719.4044.7546.91336.9325203512.1843.981OpenBenchmarking.org

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: NoRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 30803691215SE +/- 0.036, N = 3SE +/- 0.051, N = 3SE +/- 0.015, N = 3SE +/- 0.010, N = 3SE +/- 0.103, N = 3SE +/- 0.094, N = 3SE +/- 0.082, N = 3SE +/- 0.107, N = 3SE +/- 0.077, N = 158.9186.6125.7335.55510.1499.6798.5489.3986.913
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 30803691215Min: 8.87 / Avg: 8.92 / Max: 8.99Min: 6.56 / Avg: 6.61 / Max: 6.71Min: 5.71 / Avg: 5.73 / Max: 5.76Min: 5.55 / Avg: 5.56 / Max: 5.58Min: 10.05 / Avg: 10.15 / Max: 10.36Min: 9.52 / Avg: 9.68 / Max: 9.84Min: 8.46 / Avg: 8.55 / Max: 8.71Min: 9.28 / Avg: 9.4 / Max: 9.61Min: 6.68 / Avg: 6.91 / Max: 7.55

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 30801428425670SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.13, N = 3SE +/- 0.14, N = 3SE +/- 0.16, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 358.0139.4732.2130.5864.3258.4345.7858.5536.93
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 30801326395265Min: 57.87 / Avg: 58.01 / Max: 58.11Min: 39.4 / Avg: 39.46 / Max: 39.51Min: 32.16 / Avg: 32.21 / Max: 32.24Min: 30.51 / Avg: 30.58 / Max: 30.63Min: 64.07 / Avg: 64.32 / Max: 64.51Min: 58.16 / Avg: 58.43 / Max: 58.59Min: 45.48 / Avg: 45.78 / Max: 46.05Min: 58.34 / Avg: 58.55 / Max: 58.71Min: 36.83 / Avg: 36.93 / Max: 37

VkFFT

VkFFT is a Fast Fourier Transform (FFT) Library that is GPU accelerated by means of the Vulkan API. The VkFFT benchmark runs FFT performance differences of many different sizes before returning an overall benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1RX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 308011K22K33K44K55KSE +/- 15.62, N = 3SE +/- 1.53, N = 3SE +/- 137.01, N = 3SE +/- 576.21, N = 3SE +/- 31.18, N = 3SE +/- 76.95, N = 3SE +/- 22.81, N = 3SE +/- 19.65, N = 3SE +/- 193.08, N = 32067633912448044973031907325094163532087520351. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1RX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 30809K18K27K36K45KMin: 20653 / Avg: 20676.33 / Max: 20706Min: 33909 / Avg: 33912 / Max: 33914Min: 44574 / Avg: 44804 / Max: 45048Min: 49141 / Avg: 49729.67 / Max: 50882Min: 31854 / Avg: 31907.33 / Max: 31962Min: 32366 / Avg: 32508.67 / Max: 32630Min: 41590 / Avg: 41635 / Max: 41664Min: 32060 / Avg: 32086.67 / Max: 32125Min: 51767 / Avg: 52035.33 / Max: 524101. (CXX) g++ options: -O3 -pthread

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 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 3080714212835SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 332.0613.7218.3817.9519.8818.6315.0119.4012.181. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 3080714212835Min: 32.06 / Avg: 32.06 / Max: 32.07Min: 13.72 / Avg: 13.72 / Max: 13.73Min: 18.37 / Avg: 18.38 / Max: 18.38Min: 17.94 / Avg: 17.95 / Max: 17.95Min: 19.81 / Avg: 19.88 / Max: 19.98Min: 18.6 / Avg: 18.63 / Max: 18.66Min: 14.93 / Avg: 15.01 / Max: 15.07Min: 19.22 / Avg: 19.4 / Max: 19.55Min: 12.11 / Avg: 12.18 / Max: 12.291. (CXX) g++ options: -O3 -pthread

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: YesRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 30801.2982.5963.8945.1926.49SE +/- 0.014, N = 3SE +/- 0.013, N = 3SE +/- 0.006, N = 3SE +/- 0.011, N = 3SE +/- 0.067, N = 3SE +/- 0.031, N = 14SE +/- 0.046, N = 4SE +/- 0.061, N = 3SE +/- 0.031, N = 35.0665.7693.4763.2734.8134.6724.0214.7543.981
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesRX 5700 XTRadeon VIIRX 6800RX 6800 XTRTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiGTX 3060 TiRTX 3080246810Min: 5.05 / Avg: 5.07 / Max: 5.09Min: 5.75 / Avg: 5.77 / Max: 5.79Min: 3.47 / Avg: 3.48 / Max: 3.48Min: 3.25 / Avg: 3.27 / Max: 3.29Min: 4.73 / Avg: 4.81 / Max: 4.95Min: 4.55 / Avg: 4.67 / Max: 4.97Min: 3.97 / Avg: 4.02 / Max: 4.16Min: 4.69 / Avg: 4.75 / Max: 4.88Min: 3.95 / Avg: 3.98 / Max: 4.04