Vulkan Compute

AMD Ryzen 5 3600 6-Core testing with a Gigabyte X570 AORUS PRO (F34 BIOS) and AMD Radeon VII on ManjaroLinux 21.1.0 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 2108171-IB-2107307PT42
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 6 Tests
Vulkan Compute 6 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

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

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
Run
  Test
  Duration
NVIDIA RTX 3060
July 30 2021
  2 Hours, 55 Minutes
Radeon VII
August 17 2021
  2 Hours, 50 Minutes
Invert Hiding All Results Option
  2 Hours, 52 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):


Vulkan ComputeProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen ResolutionOpenCLNVIDIA RTX 3060Radeon VIIAMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads)ASUS ROG CROSSHAIR VIII HERO (3501 BIOS)AMD Starship/Matisse16GB1000GB Sabrent Rocket 4.0 Plus + 2000GBeVGA NVIDIA GeForce RTX 3060 12GBNVIDIA Device 228eASUS VP28URealtek RTL8125 2.5GbE + Intel I211Ubuntu 21.045.11.0-25-generic (x86_64)GNOME Shell 3.38.4X Server 1.20.11NVIDIA 470.57.024.6.01.2.175GCC 10.3.0ext43840x2160AMD Ryzen 5 3600 6-Core @ 3.60GHz (6 Cores / 12 Threads)Gigabyte X570 AORUS PRO (F34 BIOS)32GB1000GB Sabrent Rocket 4.0 1TB + 240GB SanDisk SDSSDA24 + 256GB SanDisk SD8SN8U2 + 0GB Multiple ReaderAMD Radeon VII (1801/1000MHz)AMD Vega 20 HDMI AudioIntel I211 + Intel Wi-Fi 6 AX200ManjaroLinux 21.1.05.14.0-1-MANJARO (x86_64)X Server 1.20.134.6 Mesa 21.1.6 (LLVM 12.0.1)OpenCL 2.0 AMD-APP.dbg (3305.0)1.2.174GCC 11.1.0 + Clang 12.0.1 + CUDA 11.4f2fs2560x1440OpenBenchmarking.orgKernel Details- NVIDIA RTX 3060: Transparent Huge Pages: madvise- Radeon VII: amdgpu.ppfeaturemask=0xffffffff - Transparent Huge Pages: madviseCompiler Details- NVIDIA RTX 3060: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/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-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 - Radeon VII: --disable-libssp --disable-libstdcxx-pch --disable-libunwind-exceptions --disable-werror --enable-__cxa_atexit --enable-cet=auto --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-gnu-indirect-function --enable-gnu-unique-object --enable-install-libiberty --enable-languages=c,c++,ada,fortran,go,lto,objc,obj-c++,d --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-isl --with-linker-hash-style=gnu Processor Details- NVIDIA RTX 3060: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009- Radeon VII: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021Security Details- NVIDIA RTX 3060: 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 - Radeon VII: 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 Graphics Details- Radeon VII: GLAMOR

NVIDIA RTX 3060 vs. Radeon VII ComparisonPhoronix Test SuiteBaseline+363.2%+363.2%+726.4%+726.4%+1089.6%+1089.6%1452.7%1433.3%94.8%87.2%50.9%46.5%34.1%33.1%21.3%Vulkan GPU - resnet18726.9%Vulkan GPU - vgg16658.2%Vulkan GPU - resnet50572.6%Vulkan GPU - alexnet512.9%Vulkan GPU - squeezenet_ssd348.2%Vulkan GPU - regnety_400m298%Vulkan GPU - yolov4-tiny258.1%Vulkan GPU - mobilenet254.5%Vulkan GPU - googlenet249.4%Vulkan GPU - shufflenet-v2162.6%Vulkan GPU-v2-v2 - mobilenet-v2157.4%Vulkan GPU - efficientnet-b0117.2%Vulkan GPU - mnasnet107.8%Vulkan GPU-v3-v3 - mobilenet-v3103.7%Vulkan GPU - blazeface101%fp64-vec4fp64-scalarfp32-scalarint16-vec4int32-vec460.5%int32-scalar56.1%fp16-vec4int16-scalarfp32-vec44x - Yes4x - No2x - 3 - Yes15.2%fp16-scalar3.5%NCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNvkpeakvkpeakvkpeakvkpeakvkpeakvkpeakvkpeakvkpeakvkpeakRealSR-NCNNRealSR-NCNNWaifu2x-NCNN VulkanvkpeakNVIDIA RTX 3060Radeon VII

Vulkan Computencnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - mobilenetncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - blazefacevkpeak: fp64-vec4vkpeak: fp64-scalarvkpeak: fp32-scalarvkpeak: int16-vec4vkpeak: int32-vec4vkpeak: int32-scalarvkpeak: fp16-vec4vkpeak: int16-scalarvkpeak: fp32-vec4realsr-ncnn: 4x - Yesrealsr-ncnn: 4x - Nowaifu2x-ncnn: 2x - 3 - Yesvkpeak: fp16-scalarvkresample: 2x - Singlevkfft: NVIDIA RTX 3060Radeon VII1.937.34.192.104.882.447.214.574.271.741.953.252.042.180.98214.29214.256829.755957.526766.466830.2813242.604480.119079.5067.90310.5054.9746849.6423.1492733715.9655.3528.1812.8721.879.7125.8216.2014.924.575.027.064.244.441.973327.363285.0413306.1311149.714216.744374.3419985.456562.7412171.1151.0338.6585.7316619.51OpenBenchmarking.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: resnet18NVIDIA RTX 3060Radeon VII48121620SE +/- 0.00, N = 2SE +/- 0.08, N = 31.9315.96MIN: 1.91 / MAX: 2.25MIN: 15.56 / MAX: 23.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet18NVIDIA RTX 3060Radeon VII48121620Min: 1.93 / Avg: 1.93 / Max: 1.93Min: 15.81 / Avg: 15.96 / Max: 16.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: vgg16NVIDIA RTX 3060Radeon VII1224364860SE +/- 0.00, N = 3SE +/- 0.11, N = 37.3055.35MIN: 7.17 / MAX: 15.83MIN: 53.94 / MAX: 79.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: vgg16NVIDIA RTX 3060Radeon VII1122334455Min: 7.3 / Avg: 7.3 / Max: 7.3Min: 55.23 / Avg: 55.35 / Max: 55.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet50NVIDIA RTX 3060Radeon VII714212835SE +/- 0.01, N = 3SE +/- 0.65, N = 34.1928.18MIN: 4.17 / MAX: 4.84MIN: 27.11 / MAX: 241.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet50NVIDIA RTX 3060Radeon VII612182430Min: 4.18 / Avg: 4.19 / Max: 4.2Min: 27.5 / Avg: 28.18 / Max: 29.481. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: alexnetNVIDIA RTX 3060Radeon VII3691215SE +/- 0.00, N = 3SE +/- 0.18, N = 32.1012.87MIN: 2.07 / MAX: 4.52MIN: 12.33 / MAX: 18.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: alexnetNVIDIA RTX 3060Radeon VII48121620Min: 2.09 / Avg: 2.1 / Max: 2.1Min: 12.52 / Avg: 12.87 / Max: 13.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: squeezenet_ssdNVIDIA RTX 3060Radeon VII510152025SE +/- 0.07, N = 3SE +/- 0.14, N = 34.8821.87MIN: 4.67 / MAX: 10.31MIN: 21.28 / MAX: 43.291. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: squeezenet_ssdNVIDIA RTX 3060Radeon VII510152025Min: 4.78 / Avg: 4.88 / Max: 5.01Min: 21.59 / Avg: 21.87 / Max: 22.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: regnety_400mNVIDIA RTX 3060Radeon VII3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 32.449.71MIN: 2.41 / MAX: 3.44MIN: 9.48 / MAX: 16.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: regnety_400mNVIDIA RTX 3060Radeon VII3691215Min: 2.43 / Avg: 2.44 / Max: 2.44Min: 9.71 / Avg: 9.71 / Max: 9.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: yolov4-tinyNVIDIA RTX 3060Radeon VII612182430SE +/- 0.01, N = 3SE +/- 0.08, N = 37.2125.82MIN: 7.02 / MAX: 7.52MIN: 25.32 / MAX: 69.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: yolov4-tinyNVIDIA RTX 3060Radeon VII612182430Min: 7.2 / Avg: 7.21 / Max: 7.22Min: 25.68 / Avg: 25.82 / Max: 25.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mobilenetNVIDIA RTX 3060Radeon VII48121620SE +/- 0.01, N = 3SE +/- 0.08, N = 34.5716.20MIN: 4.52 / MAX: 4.92MIN: 15.73 / MAX: 24.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mobilenetNVIDIA RTX 3060Radeon VII48121620Min: 4.56 / Avg: 4.57 / Max: 4.58Min: 16.05 / Avg: 16.2 / Max: 16.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: googlenetNVIDIA RTX 3060Radeon VII48121620SE +/- 0.08, N = 3SE +/- 0.09, N = 34.2714.92MIN: 3.95 / MAX: 15.21MIN: 14.55 / MAX: 20.511. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: googlenetNVIDIA RTX 3060Radeon VII48121620Min: 4.18 / Avg: 4.27 / Max: 4.43Min: 14.8 / Avg: 14.92 / Max: 15.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: shufflenet-v2NVIDIA RTX 3060Radeon VII1.02832.05663.08494.11325.1415SE +/- 0.00, N = 3SE +/- 0.06, N = 31.744.57MIN: 1.71 / MAX: 3.74MIN: 4.38 / MAX: 9.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: shufflenet-v2NVIDIA RTX 3060Radeon VII246810Min: 1.73 / Avg: 1.74 / Max: 1.74Min: 4.49 / Avg: 4.57 / Max: 4.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2NVIDIA RTX 3060Radeon VII1.12952.2593.38854.5185.6475SE +/- 0.00, N = 3SE +/- 0.02, N = 31.955.02MIN: 1.92 / MAX: 2.86MIN: 4.79 / MAX: 14.911. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2NVIDIA RTX 3060Radeon VII246810Min: 1.94 / Avg: 1.95 / Max: 1.95Min: 4.99 / Avg: 5.02 / Max: 5.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: efficientnet-b0NVIDIA RTX 3060Radeon VII246810SE +/- 0.01, N = 3SE +/- 0.11, N = 33.257.06MIN: 3.22 / MAX: 3.7MIN: 6.78 / MAX: 61.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: efficientnet-b0NVIDIA RTX 3060Radeon VII3691215Min: 3.24 / Avg: 3.25 / Max: 3.26Min: 6.93 / Avg: 7.06 / Max: 7.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mnasnetNVIDIA RTX 3060Radeon VII0.9541.9082.8623.8164.77SE +/- 0.01, N = 3SE +/- 0.03, N = 32.044.24MIN: 2.02 / MAX: 4.46MIN: 4.1 / MAX: 9.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mnasnetNVIDIA RTX 3060Radeon VII246810Min: 2.03 / Avg: 2.04 / Max: 2.05Min: 4.21 / Avg: 4.24 / Max: 4.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3NVIDIA RTX 3060Radeon VII0.9991.9982.9973.9964.995SE +/- 0.00, N = 3SE +/- 0.02, N = 32.184.44MIN: 2.16 / MAX: 3.53MIN: 4.27 / MAX: 14.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3NVIDIA RTX 3060Radeon VII246810Min: 2.18 / Avg: 2.18 / Max: 2.19Min: 4.41 / Avg: 4.44 / Max: 4.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: blazefaceNVIDIA RTX 3060Radeon VII0.44330.88661.32991.77322.2165SE +/- 0.00, N = 3SE +/- 0.05, N = 30.981.97MIN: 0.95 / MAX: 2.33MIN: 1.81 / MAX: 49.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: blazefaceNVIDIA RTX 3060Radeon VII246810Min: 0.98 / Avg: 0.98 / Max: 0.99Min: 1.91 / Avg: 1.97 / Max: 2.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

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 20210424fp64-vec4NVIDIA RTX 3060Radeon VII7001400210028003500SE +/- 0.01, N = 3SE +/- 56.99, N = 3214.293327.36
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4NVIDIA RTX 3060Radeon VII6001200180024003000Min: 214.27 / Avg: 214.29 / Max: 214.32Min: 3246.39 / Avg: 3327.36 / Max: 3437.31

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarNVIDIA RTX 3060Radeon VII7001400210028003500SE +/- 0.03, N = 3SE +/- 61.71, N = 3214.253285.04
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarNVIDIA RTX 3060Radeon VII6001200180024003000Min: 214.2 / Avg: 214.25 / Max: 214.31Min: 3196.27 / Avg: 3285.04 / Max: 3403.68

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarNVIDIA RTX 3060Radeon VII3K6K9K12K15KSE +/- 10.62, N = 3SE +/- 139.79, N = 36829.7513306.13
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarNVIDIA RTX 3060Radeon VII2K4K6K8K10KMin: 6808.54 / Avg: 6829.75 / Max: 6841.32Min: 13061.44 / Avg: 13306.13 / Max: 13545.6

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4NVIDIA RTX 3060Radeon VII2K4K6K8K10KSE +/- 0.03, N = 3SE +/- 177.74, N = 35957.5211149.71
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4NVIDIA RTX 3060Radeon VII2K4K6K8K10KMin: 5957.48 / Avg: 5957.52 / Max: 5957.57Min: 10914.69 / Avg: 11149.71 / Max: 11498.18

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4NVIDIA RTX 3060Radeon VII15003000450060007500SE +/- 17.45, N = 3SE +/- 71.24, N = 36766.464216.74
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4NVIDIA RTX 3060Radeon VII12002400360048006000Min: 6748.89 / Avg: 6766.46 / Max: 6801.36Min: 4114.81 / Avg: 4216.74 / Max: 4353.92

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarNVIDIA RTX 3060Radeon VII15003000450060007500SE +/- 0.58, N = 3SE +/- 85.69, N = 36830.284374.34
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarNVIDIA RTX 3060Radeon VII12002400360048006000Min: 6829.54 / Avg: 6830.28 / Max: 6831.42Min: 4264.03 / Avg: 4374.34 / Max: 4543.08

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4NVIDIA RTX 3060Radeon VII4K8K12K16K20KSE +/- 2.31, N = 3SE +/- 540.60, N = 313242.6019985.45
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4NVIDIA RTX 3060Radeon VII3K6K9K12K15KMin: 13239.29 / Avg: 13242.6 / Max: 13247.05Min: 19151.86 / Avg: 19985.45 / Max: 20998.57

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarNVIDIA RTX 3060Radeon VII14002800420056007000SE +/- 0.29, N = 3SE +/- 58.06, N = 34480.116562.74
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarNVIDIA RTX 3060Radeon VII11002200330044005500Min: 4479.74 / Avg: 4480.11 / Max: 4480.68Min: 6477.62 / Avg: 6562.74 / Max: 6673.7

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4NVIDIA RTX 3060Radeon VII3K6K9K12K15KSE +/- 0.30, N = 3SE +/- 255.84, N = 39079.5012171.11
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4NVIDIA RTX 3060Radeon VII2K4K6K8K10KMin: 9079.1 / Avg: 9079.5 / Max: 9080.09Min: 11772.61 / Avg: 12171.11 / Max: 12648.33

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: YesNVIDIA RTX 3060Radeon VII1530456075SE +/- 0.06, N = 3SE +/- 0.33, N = 367.9051.03
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesNVIDIA RTX 3060Radeon VII1326395265Min: 67.79 / Avg: 67.9 / Max: 67.97Min: 50.37 / Avg: 51.03 / Max: 51.39

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoNVIDIA RTX 3060Radeon VII3691215SE +/- 0.003, N = 3SE +/- 0.054, N = 310.5058.658
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoNVIDIA RTX 3060Radeon VII3691215Min: 10.5 / Avg: 10.51 / Max: 10.51Min: 8.56 / Avg: 8.66 / Max: 8.74

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: YesNVIDIA RTX 3060Radeon VII1.28952.5793.86855.1586.4475SE +/- 0.004, N = 3SE +/- 0.018, N = 34.9745.731
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesNVIDIA RTX 3060Radeon VII246810Min: 4.97 / Avg: 4.97 / Max: 4.98Min: 5.7 / Avg: 5.73 / Max: 5.76

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 20210424fp16-scalarNVIDIA RTX 3060Radeon VII15003000450060007500SE +/- 17.95, N = 3SE +/- 43.13, N = 36849.646619.51
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarNVIDIA RTX 3060Radeon VII12002400360048006000Min: 6813.76 / Avg: 6849.64 / Max: 6868.53Min: 6539.34 / Avg: 6619.51 / Max: 6687.19

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: SingleNVIDIA RTX 3060612182430SE +/- 0.02, N = 323.151. (CXX) g++ options: -O3 -pthread

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 1.1.1NVIDIA RTX 30606K12K18K24K30KSE +/- 308.54, N = 3273371. (CXX) g++ options: -O3 -pthread