nvidia-linux-gpu-performance-20212

AMD Ryzen 9 5950X 16-Core testing with a ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3202 BIOS) and NVIDIA GeForce RTX 2080 Ti 11GB on Ubuntu 20.10 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 2102075-PTS-NVIDIALI44
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
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
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
RTX 3080
February 06 2021
  5 Hours, 28 Minutes
RTX 3060 Ti
February 06 2021
  2 Hours, 24 Minutes
NVIDIA GeForce RTX 3060 Ti
February 06 2021
  15 Minutes
RTX 2070 SUPER
February 06 2021
  2 Hours, 45 Minutes
RTX 2060 SUPER
February 06 2021
  5 Hours, 28 Minutes
RTX 2080 SUPER
February 07 2021
  2 Hours, 48 Minutes
RTX 2080 Ti
February 07 2021
  4 Hours, 48 Minutes
Invert Behavior (Only Show Selected Data)
  3 Hours, 25 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):


nvidia-linux-gpu-performance-20212 OpenBenchmarking.orgPhoronix Test SuiteAMD 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 + 2000GBNVIDIA GeForce RTX 3080 10GB (1710/9501MHz)NVIDIA GeForce RTX 3060 Ti 8GB (345/405MHz)NVIDIA GeForce RTX 3060 Ti 8GB (360/405MHz)NVIDIA GeForce RTX 2070 SUPER 8GB (1605/7000MHz)NVIDIA GeForce RTX 2060 SUPER 8GB (390/405MHz)NVIDIA GeForce RTX 2080 SUPER 8GB (1650/7750MHz)NVIDIA GeForce RTX 2080 Ti 11GB (1350/7000MHz)NVIDIA Device 1aefNVIDIA Device 228bNVIDIA TU104 HD AudioNVIDIA TU106 HD AudioNVIDIA TU102 HD AudioASUS MG28URealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.105.8.0-41-generic (x86_64)GNOME Shell 3.38.2X Server 1.20.9NVIDIA 460.394.6.0OpenCL 1.2 CUDA 11.2.1361.2.155GCC 10.2.0 + Clang 11.0.1-1~oibaf~gext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudiosMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionNvidia-linux-gpu-performance-20212 BenchmarksSystem Logs- Transparent Huge Pages: madvise- --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 - RTX 3080: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 - RTX 3060 Ti: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa201009 - NVIDIA GeForce RTX 3060 Ti: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa201009 - RTX 2070 SUPER: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 - RTX 2060 SUPER: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 - RTX 2080 SUPER: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 - RTX 2080 Ti: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 - RTX 3080: GPU Compute Cores: 8704- RTX 3060 Ti: GPU Compute Cores: 4864- RTX 2070 SUPER: GPU Compute Cores: 2560- RTX 2060 SUPER: GPU Compute Cores: 2176- RTX 2080 SUPER: GPU Compute Cores: 3072- RTX 2080 Ti: GPU Compute Cores: 4352- RTX 3080, RTX 3060 Ti, RTX 2070 SUPER, RTX 2060 SUPER, RTX 2080 SUPER, RTX 2080 Ti: Python 3.8.6 - 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

nvidia-linux-gpu-performance-20212 lczero: OpenCLredshift: vkfft: ncnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU - mobilenetoctanebench: Total Scorefahbench: warsow: 2560 x 1440warsow: 1920 x 1200warsow: 1920 x 1080warsow: 3840 x 2160v-ray: NVIDIA CUDA GPUv-ray: NVIDIA RTX GPUindigobench: OpenCL GPU - Bedroomindigobench: OpenCL GPU - Supercarrealsr-ncnn: 4x - Yesvkresample: 2x - Doublertiv: 3840 x 2160 - Lucy In One Weekendrtiv: 3840 x 2160 - Cornell Box + Lucyrtiv: 2560 x 1440 - Lucy In One Weekendrtiv: 1920 x 1200 - Lucy In One Weekendrtiv: 1920 x 1080 - Lucy In One Weekendrtiv: 2560 x 1440 - Cornell Box + Lucyrtiv: 1920 x 1200 - Cornell Box + Lucyrtiv: 1920 x 1080 - Cornell Box + Lucyrtiv: 3840 x 2160 - Planets In One Weekendrtiv: 3840 x 2160 - Ray Tracing In One Weekendrtiv: 2560 x 1440 - Planets In One Weekendrtiv: 3840 x 2160 - Cornell Boxrtiv: 1920 x 1080 - Planets In One Weekendrtiv: 1920 x 1200 - Planets In One Weekendrtiv: 2560 x 1440 - Ray Tracing In One Weekendrtiv: 1920 x 1080 - Ray Tracing In One Weekendrtiv: 1920 x 1200 - Ray Tracing In One Weekendrtiv: 2560 x 1440 - Cornell Boxrtiv: 1920 x 1200 - Cornell Boxrtiv: 1920 x 1080 - Cornell Boxparaview: Many Spheres - 3840 x 2160paraview: Many Spheres - 3840 x 2160paraview: Many Spheres - 1920 x 1080paraview: Many Spheres - 1920 x 1080paraview: Wavelet Contour - 3840 x 2160paraview: Wavelet Contour - 3840 x 2160realsr-ncnn: 4x - Novkresample: 2x - Singlenamd-cuda: ATPase Simulation - 327,506 Atomsparaview: Wavelet Contour - 1920 x 1080paraview: Wavelet Contour - 1920 x 1080paraview: Wavelet Volume - 3840 x 2160paraview: Wavelet Volume - 3840 x 2160hashcat: TrueCrypt RIPEMD160 + XTSfinancebench: Monte-Carlo OpenCLparaview: Wavelet Volume - 1920 x 1080paraview: Wavelet Volume - 1920 x 1080betsy: ETC2 RGB - Highesthashcat: SHA-512hashcat: SHA1hashcat: MD5betsy: ETC1 - Highestwaifu2x-ncnn: 2x - 3 - Yesdarktable: Boat - OpenCLdarktable: Masskrug - OpenCLhashcat: 7-Zipdarktable: Server Room - OpenCLfinancebench: Black-Scholes OpenCLdarktable: Server Rack - OpenCLRTX 3080RTX 3060 TiNVIDIA GeForce RTX 3060 TiRTX 2070 SUPERRTX 2060 SUPERRTX 2080 SUPERRTX 2080 Ti318971645658317.8514.4821.3824.9711.1214.4358.8312.821.855.413.994.444.214.4812.20562.818641321.9986984.8985.7963.6830.21666218317.65046.40534.114148.00516.431216.441333.054548.693957.822333.942248.429059.092017.551217.135936.536231.966863.796957.148235.618562.731255.541767.475997.1997117.9197680.53976.617841.29978.213478.835333.826.25011.2320.132915373.698515.655430.654339.42721867386.04699710606.699662.924.416242040000019098866667560396666673.3283.4391.2822.0909905670.7067.3070.0952413522818.0014.7423.3826.9811.7214.7858.9013.591.885.423.984.444.144.5013.27382.736651237.0623983.3974.5980.4499.654.096266.2109.607819.3931419.528729.099933.719619.545127.894634.06839.996529.7603820.975718.494336.736532.599220.481035.882331.803338.598555.766168.32054451.59344.404475.50144.642251.061216.018.76117.6580.136143169.404304.133342.870208.93426700461.6763417578.143473.636.794141463333311068233333329403666675.0094.3721.7132.1215859670.77713.1380.1011185155011.45233.9923483206317.9614.2722.3925.6312.0914.7058.4913.441.865.504.074.464.214.5812.87261.261302232.7775953.3986.0979.2437.67859777.77624.70663.020262.2497.931536.7771715.655022.790226.275314.103020.063524.596410.826210.637122.745316.791239.982535.465422.261239.143334.658735.275050.731261.70594407.22643.964499.12744.882299.463220.659.90119.4690.148233691.347354.213288.286205.52426200816.0099907715.261482.207.330142420000011121466667353981666675.4564.6891.7092.1565963670.77612.8040.0973793100017.9814.4721.9425.3911.1514.4458.9912.851.875.454.014.464.214.5012.53241.445325208.7075696.3985.2977.5353.34927337.39223.31175.755309.1856.872065.9760213.756820.172523.307012.437317.845021.83699.178868.9963319.334314.184833.888630.180918.952633.208329.486129.887042.937252.15343748.65037.393824.55738.151859.637178.4511.50122.2340.158572862.737274.702847.497177.97351200752.4139816827.820426.748.39411879333339341766667292802666676.0325.2621.7162.1184955670.72316.4863330.0973283471918.6514.6723.0726.6312.2615.1561.1113.571.875.574.094.574.254.6113.31262.843677257.9896983.2985.7979.7494.27429057.84525.32654.895217.3118.826167.1478417.149824.589028.447114.870821.119725.752913.146712.871927.512320.336148.345243.046326.953847.523042.064942.880061.680975.24465323.26553.105446.21254.322616.200251.059.01017.7790.148944225.680405.493853.392240.83520673843.4526578551.751534.486.831170843333313420400000423930333335.2964.3471.5912.1696925000.75610.6100.0972464215218.0014.4122.0824.9811.2014.5159.0213.291.845.454.034.474.194.5012.97358.549904307.2028985.7985.9979.3553.5926124611.01233.23643.816155.74312.467310.536924.371935.017840.417821.800531.024237.786217.637017.307336.757127.575064.497057.429036.079463.304656.239557.953584.0012102.0607296.59272.787469.29274.503059.954293.637.50614.7800.140844690.748450.114935.743308.48657067649.8206589606.532600.415.008223926666717718433333552981000003.8223.7761.4852.0848948670.7109.2260.096OpenBenchmarking.org

LeelaChessZero

LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: OpenCLRTX 30807K14K21K28K35KSE +/- 256.06, N = 3318971. (CXX) g++ options: -flto -pthread

RedShift Demo

This is a test of MAXON's RedShift demo build that currently requires NVIDIA GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRedShift Demo 3.0RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER80160240320400SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3164241246328348379

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 1.1.1RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER12K24K36K48K60KSE +/- 23.59, N = 3SE +/- 17.02, N = 3SE +/- 109.70, N = 3SE +/- 9.33, N = 3SE +/- 134.01, N = 3SE +/- 4.98, N = 35658335228421523471932063310001. (CXX) g++ options: -O3 -pthread

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 20201218Target: Vulkan GPU - Model: regnety_400mRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER510152025SE +/- 0.02, N = 15SE +/- 0.07, N = 3SE +/- 0.07, N = 13SE +/- 0.15, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 1517.8518.0018.0018.6517.9617.98MIN: 17.56 / MAX: 27.39MIN: 17.41 / MAX: 27.83MIN: 17.4 / MAX: 35.68MIN: 17.68 / MAX: 41.61MIN: 17.61 / MAX: 19.15MIN: 17.54 / MAX: 27.131. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssdRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER48121620SE +/- 0.07, N = 14SE +/- 0.21, N = 3SE +/- 0.05, N = 13SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 1514.4814.7414.4114.6714.2714.47MIN: 13.79 / MAX: 24.15MIN: 14.07 / MAX: 16.1MIN: 13.6 / MAX: 27.43MIN: 14.1 / MAX: 19.06MIN: 13.95 / MAX: 14.61MIN: 13.92 / MAX: 16.431. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tinyRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER612182430SE +/- 0.22, N = 15SE +/- 0.09, N = 3SE +/- 0.31, N = 13SE +/- 0.78, N = 3SE +/- 0.59, N = 3SE +/- 0.27, N = 1521.3823.3822.0823.0722.3921.94MIN: 20.7 / MAX: 38.06MIN: 22.85 / MAX: 25.03MIN: 20.68 / MAX: 34.43MIN: 20.78 / MAX: 30.08MIN: 20.98 / MAX: 23.83MIN: 20.72 / MAX: 25.981. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet50RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER612182430SE +/- 0.22, N = 15SE +/- 0.14, N = 3SE +/- 0.25, N = 13SE +/- 0.60, N = 3SE +/- 0.36, N = 3SE +/- 0.27, N = 1524.9726.9824.9826.6325.6325.39MIN: 24.2 / MAX: 30.79MIN: 24.63 / MAX: 46.48MIN: 23.79 / MAX: 34.58MIN: 24.4 / MAX: 46.26MIN: 24.31 / MAX: 35.26MIN: 24.28 / MAX: 43.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnetRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER3691215SE +/- 0.10, N = 15SE +/- 0.57, N = 3SE +/- 0.19, N = 13SE +/- 0.55, N = 3SE +/- 0.61, N = 3SE +/- 0.14, N = 1511.1211.7211.2012.2612.0911.15MIN: 10.81 / MAX: 24.77MIN: 10.8 / MAX: 44.26MIN: 10.71 / MAX: 13.13MIN: 10.76 / MAX: 16.88MIN: 10.8 / MAX: 13.2MIN: 10.78 / MAX: 13.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet18RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER48121620SE +/- 0.02, N = 15SE +/- 0.45, N = 3SE +/- 0.15, N = 13SE +/- 0.15, N = 3SE +/- 0.36, N = 3SE +/- 0.06, N = 1514.4314.7814.5115.1514.7014.44MIN: 14.25 / MAX: 27.77MIN: 14.12 / MAX: 17.41MIN: 14.09 / MAX: 16.91MIN: 14.24 / MAX: 28.92MIN: 14.18 / MAX: 23.66MIN: 14.17 / MAX: 32.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg16RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1428425670SE +/- 0.06, N = 15SE +/- 0.18, N = 3SE +/- 0.05, N = 13SE +/- 0.50, N = 3SE +/- 0.40, N = 3SE +/- 0.04, N = 1558.8358.9059.0261.1158.4958.99MIN: 57.03 / MAX: 117.59MIN: 57.26 / MAX: 90.67MIN: 57.58 / MAX: 72.34MIN: 57.69 / MAX: 129.41MIN: 56.35 / MAX: 60.41MIN: 57.66 / MAX: 83.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenetRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER3691215SE +/- 0.02, N = 15SE +/- 0.77, N = 3SE +/- 0.25, N = 13SE +/- 0.21, N = 3SE +/- 0.68, N = 3SE +/- 0.04, N = 1512.8213.5913.2913.5713.4412.85MIN: 12.43 / MAX: 22.33MIN: 12.49 / MAX: 16.77MIN: 12.5 / MAX: 27.44MIN: 12.61 / MAX: 15.4MIN: 12.44 / MAX: 24.56MIN: 12.46 / MAX: 21.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazefaceRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER0.4230.8461.2691.6922.115SE +/- 0.01, N = 15SE +/- 0.03, N = 3SE +/- 0.01, N = 13SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 151.851.881.841.871.861.87MIN: 1.8 / MAX: 2.32MIN: 1.8 / MAX: 7.07MIN: 1.8 / MAX: 2.4MIN: 1.8 / MAX: 2.36MIN: 1.8 / MAX: 2.1MIN: 1.79 / MAX: 2.291. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b0RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1.25332.50663.75995.01326.2665SE +/- 0.01, N = 15SE +/- 0.02, N = 3SE +/- 0.04, N = 13SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 155.415.425.455.575.505.45MIN: 5.3 / MAX: 14.74MIN: 5.3 / MAX: 9.74MIN: 5.28 / MAX: 6.74MIN: 5.34 / MAX: 7.4MIN: 5.31 / MAX: 9.38MIN: 5.31 / MAX: 14.051. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnetRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER0.92031.84062.76093.68124.6015SE +/- 0.00, N = 15SE +/- 0.01, N = 3SE +/- 0.04, N = 13SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.01, N = 153.993.984.034.094.074.01MIN: 3.85 / MAX: 4.5MIN: 3.84 / MAX: 5.19MIN: 3.82 / MAX: 5.26MIN: 3.88 / MAX: 5.64MIN: 3.84 / MAX: 5.04MIN: 3.85 / MAX: 6.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v2RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1.02832.05663.08494.11325.1415SE +/- 0.01, N = 15SE +/- 0.02, N = 3SE +/- 0.02, N = 12SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 154.444.444.474.574.464.46MIN: 4.35 / MAX: 6.25MIN: 4.34 / MAX: 5.43MIN: 4.32 / MAX: 5.58MIN: 4.41 / MAX: 6MIN: 4.36 / MAX: 6.1MIN: 4.36 / MAX: 19.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER0.95631.91262.86893.82524.7815SE +/- 0.02, N = 15SE +/- 0.01, N = 3SE +/- 0.01, N = 13SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 154.214.144.194.254.214.21MIN: 4.1 / MAX: 6.19MIN: 4.06 / MAX: 5.35MIN: 4.08 / MAX: 5.41MIN: 4.1 / MAX: 8.02MIN: 4.12 / MAX: 5.49MIN: 4.09 / MAX: 5.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1.03732.07463.11194.14925.1865SE +/- 0.00, N = 15SE +/- 0.01, N = 3SE +/- 0.00, N = 13SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.00, N = 154.484.504.504.614.584.50MIN: 4.3 / MAX: 5.1MIN: 4.28 / MAX: 5.73MIN: 4.29 / MAX: 6.08MIN: 4.31 / MAX: 6.2MIN: 4.3 / MAX: 28.29MIN: 4.3 / MAX: 5.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mobilenetRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER3691215SE +/- 0.09, N = 15SE +/- 0.19, N = 3SE +/- 0.09, N = 13SE +/- 0.09, N = 3SE +/- 0.14, N = 3SE +/- 0.10, N = 1512.2013.2712.9713.3112.8712.53MIN: 11.81 / MAX: 13.49MIN: 12.64 / MAX: 14.87MIN: 12.36 / MAX: 22.3MIN: 12.41 / MAX: 30.9MIN: 12.38 / MAX: 13.65MIN: 11.86 / MAX: 261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OctaneBench

OctaneBench is a test of the OctaneRender on the GPU and requires the use of NVIDIA CUDA. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterOctaneBench 2020.1Total ScoreRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER120240360480600562.82382.74358.55262.84261.26241.45

FAHBench

FAHBench is a Folding@Home benchmark on the GPU. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER70140210280350SE +/- 0.54, N = 3SE +/- 0.24, N = 3SE +/- 0.96, N = 3SE +/- 0.56, N = 3SE +/- 0.78, N = 3SE +/- 0.46, N = 3322.00237.06307.20257.99232.78208.71

Warsow

This is a benchmark of Warsow, a popular open-source first-person shooter. This game uses the QFusion engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 2560 x 1440RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER2004006008001000SE +/- 0.15, N = 3SE +/- 0.68, N = 3SE +/- 0.33, N = 3SE +/- 0.29, N = 3SE +/- 0.97, N = 3SE +/- 2.47, N = 3984.8983.3985.7983.2953.3696.3

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 1920 x 1200RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER2004006008001000SE +/- 0.37, N = 3SE +/- 5.47, N = 3SE +/- 0.20, N = 3SE +/- 0.55, N = 3SE +/- 0.46, N = 3SE +/- 0.52, N = 3985.7974.5985.9985.7986.0985.2

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 1920 x 1080RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER2004006008001000SE +/- 13.18, N = 3SE +/- 5.01, N = 3SE +/- 6.45, N = 3SE +/- 6.93, N = 3SE +/- 6.42, N = 3SE +/- 6.77, N = 3963.6980.4979.3979.7979.2977.5

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 3840 x 2160RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER2004006008001000SE +/- 2.63, N = 3SE +/- 1.22, N = 3SE +/- 0.61, N = 3SE +/- 0.52, N = 3SE +/- 0.30, N = 3SE +/- 0.73, N = 3830.2499.6553.5494.2437.6353.3

Chaos Group V-RAY

This is a test of Chaos Group's V-RAY benchmark. V-RAY is a commercial renderer that can integrate with various creator software products like SketchUp and 3ds Max. The V-RAY benchmark is standalone and supports CPU and NVIDIA CUDA/RTX based rendering. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgvpaths, More Is BetterChaos Group V-RAY 5Mode: NVIDIA CUDA GPURTX 3080RTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPERNVIDIA GeForce RTX 3060 Ti400800120016002000SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 316669267427854921185

OpenBenchmarking.orgvrays, More Is BetterChaos Group V-RAY 5Mode: NVIDIA RTX GPURTX 3080RTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPERNVIDIA GeForce RTX 3060 Ti5001000150020002500SE +/- 5.69, N = 3SE +/- 1.00, N = 3SE +/- 1.45, N = 3218312469059777331550

IndigoBench

This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: BedroomRTX 3080RTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPERNVIDIA GeForce RTX 3060 Ti48121620SE +/- 0.045, N = 3SE +/- 0.010, N = 3SE +/- 0.002, N = 3SE +/- 0.004, N = 3SE +/- 0.011, N = 3SE +/- 0.011, N = 317.65011.0127.8457.7767.39211.452

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: SupercarRTX 3080RTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPERNVIDIA GeForce RTX 3060 Ti1122334455SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 346.4133.2425.3324.7123.3133.99

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: YesRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER20406080100SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.23, N = 3SE +/- 0.16, N = 3SE +/- 0.27, N = 3SE +/- 0.30, N = 334.1154.1043.8254.9063.0275.76

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: DoubleRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER70140210280350SE +/- 0.11, N = 3SE +/- 0.15, N = 3SE +/- 0.08, N = 3SE +/- 0.53, N = 3SE +/- 0.63, N = 3SE +/- 0.24, N = 3148.01266.21155.74217.31262.25309.191. (CXX) g++ options: -O3 -pthread

Ray Tracing In Vulkan

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 3840 x 2160 - Scene: Lucy In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER48121620SE +/- 0.01030, N = 3SE +/- 0.00111, N = 3SE +/- 0.00880, N = 3SE +/- 0.00442, N = 3SE +/- 0.00227, N = 3SE +/- 0.00320, N = 316.431209.6078112.467308.826167.931536.872061. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 3840 x 2160 - Scene: Cornell Box + LucyRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER48121620SE +/- 0.00971, N = 3SE +/- 0.00090, N = 3SE +/- 0.00654, N = 3SE +/- 0.00445, N = 3SE +/- 0.01081, N = 3SE +/- 0.00374, N = 316.441309.3931410.536907.147846.777175.976021. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 2560 x 1440 - Scene: Lucy In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER816243240SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 333.0519.5324.3717.1515.6613.761. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1200 - Scene: Lucy In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1122334455SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 348.6929.1035.0224.5922.7920.171. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1080 - Scene: Lucy In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1326395265SE +/- 0.17, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 357.8233.7240.4228.4526.2823.311. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 2560 x 1440 - Scene: Cornell Box + LucyRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER816243240SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 333.9419.5521.8014.8714.1012.441. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1200 - Scene: Cornell Box + LucyRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1122334455SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 348.4327.8931.0221.1220.0617.851. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1080 - Scene: Cornell Box + LucyRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1326395265SE +/- 0.11, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 359.0934.0737.7925.7524.6021.841. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 3840 x 2160 - Scene: Planets In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER48121620SE +/- 0.00529, N = 3SE +/- 0.00065, N = 3SE +/- 0.00677, N = 3SE +/- 0.00055, N = 3SE +/- 0.00074, N = 3SE +/- 0.00967, N = 317.551209.9965217.6370013.1467010.826209.178861. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 3840 x 2160 - Scene: Ray Tracing In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER48121620SE +/- 0.00227, N = 3SE +/- 0.00160, N = 3SE +/- 0.00581, N = 3SE +/- 0.00003, N = 3SE +/- 0.02386, N = 3SE +/- 0.00026, N = 317.135909.7603817.3073012.8719010.637108.996331. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 2560 x 1440 - Scene: Planets In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER816243240SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 336.5420.9836.7627.5122.7519.331. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 3840 x 2160 - Scene: Cornell BoxRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER714212835SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 331.9718.4927.5820.3416.7914.181. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1080 - Scene: Planets In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1428425670SE +/- 0.10, N = 3SE +/- 0.01, N = 3SE +/- 0.22, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 363.8036.7464.5048.3539.9833.891. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1200 - Scene: Planets In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1326395265SE +/- 0.15, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 357.1532.6057.4343.0535.4730.181. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 2560 x 1440 - Scene: Ray Tracing In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER816243240SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 335.6220.4836.0826.9522.2618.951. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1080 - Scene: Ray Tracing In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1428425670SE +/- 0.23, N = 3SE +/- 0.00, N = 3SE +/- 0.15, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 362.7335.8863.3047.5239.1433.211. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1200 - Scene: Ray Tracing In One WeekendRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1326395265SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 355.5431.8056.2442.0634.6629.491. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 2560 x 1440 - Scene: Cornell BoxRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1530456075SE +/- 0.18, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 367.4838.6057.9542.8835.2829.891. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1200 - Scene: Cornell BoxRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER20406080100SE +/- 0.04, N = 3SE +/- 0.12, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 397.2055.7784.0061.6850.7342.941. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

OpenBenchmarking.orgFPS, More Is BetterRay Tracing In Vulkan r6Resolution: 1920 x 1080 - Scene: Cornell BoxRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER306090120150SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.11, N = 3SE +/- 0.01, N = 3117.9268.32102.0675.2461.7152.151. (CXX) g++ options: -O3 -lbacktrace -lstdc++fs -lm -ldl -lpthread

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.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 3840 x 2160RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER16003200480064008000SE +/- 11.77, N = 3SE +/- 5.00, N = 3SE +/- 7.19, N = 3SE +/- 1.58, N = 3SE +/- 1.45, N = 3SE +/- 2.20, N = 37680.544451.597296.595323.274407.233748.65

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 3840 x 2160RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER20406080100SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 376.6144.4072.7853.1043.9637.39

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 1920 x 1080RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER2K4K6K8K10KSE +/- 20.20, N = 3SE +/- 3.55, N = 3SE +/- 17.40, N = 3SE +/- 16.45, N = 3SE +/- 10.38, N = 3SE +/- 9.03, N = 37841.304475.507469.295446.214499.133824.56

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 1920 x 1080RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER20406080100SE +/- 0.20, N = 3SE +/- 0.04, N = 3SE +/- 0.17, N = 3SE +/- 0.16, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 378.2144.6474.5054.3244.8838.15

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 3840 x 2160RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER7001400210028003500SE +/- 28.95, N = 15SE +/- 25.69, N = 3SE +/- 12.62, N = 3SE +/- 20.11, N = 10SE +/- 11.62, N = 3SE +/- 12.72, N = 33478.842251.063059.952616.202299.461859.64

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 3840 x 2160RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER70140210280350SE +/- 2.78, N = 15SE +/- 2.47, N = 3SE +/- 1.21, N = 3SE +/- 1.93, N = 10SE +/- 1.12, N = 3SE +/- 1.22, N = 3333.82216.01293.63251.05220.65178.45

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: NoRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER3691215SE +/- 0.044, N = 3SE +/- 0.053, N = 3SE +/- 0.058, N = 3SE +/- 0.240, N = 14SE +/- 0.038, N = 3SE +/- 0.118, N = 56.2508.7617.5069.0109.90111.501

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: SingleRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER510152025SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 311.2317.6614.7817.7819.4722.231. (CXX) g++ options: -O3 -pthread

NAMD CUDA

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. This version of the NAMD test profile uses CUDA GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 AtomsRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER0.03570.07140.10710.14280.1785SE +/- 0.00128, N = 3SE +/- 0.00015, N = 3SE +/- 0.00010, N = 3SE +/- 0.00049, N = 3SE +/- 0.00031, N = 3SE +/- 0.00029, N = 30.132910.136140.140840.148940.148230.15857

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.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 1920 x 1080RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER12002400360048006000SE +/- 37.60, N = 3SE +/- 16.12, N = 3SE +/- 7.41, N = 3SE +/- 36.34, N = 3SE +/- 10.22, N = 3SE +/- 14.35, N = 35373.703169.404690.754225.683691.352862.74

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 1920 x 1080RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER110220330440550SE +/- 3.61, N = 3SE +/- 1.55, N = 3SE +/- 0.71, N = 3SE +/- 3.49, N = 3SE +/- 0.98, N = 3SE +/- 1.38, N = 3515.65304.13450.11405.49354.21274.70

OpenBenchmarking.orgMiVoxels / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 3840 x 2160RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER12002400360048006000SE +/- 21.74, N = 3SE +/- 20.25, N = 3SE +/- 27.20, N = 3SE +/- 14.19, N = 3SE +/- 7.19, N = 3SE +/- 3.77, N = 35430.653342.874935.743853.393288.292847.50

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 3840 x 2160RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER70140210280350SE +/- 1.36, N = 3SE +/- 1.26, N = 3SE +/- 1.70, N = 3SE +/- 0.89, N = 3SE +/- 0.45, N = 3SE +/- 0.24, N = 3339.42208.93308.48240.83205.52177.97

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: TrueCrypt RIPEMD160 + XTSRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER150K300K450K600K750KSE +/- 560.75, N = 3SE +/- 1006.64, N = 3SE +/- 592.55, N = 3SE +/- 4356.82, N = 15SE +/- 360.56, N = 3SE +/- 173.21, N = 3721867426700657067520673426200351200

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Monte-Carlo OpenCLRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER2004006008001000SE +/- 1.21, N = 3SE +/- 1.66, N = 3SE +/- 0.61, N = 3SE +/- 0.99, N = 3SE +/- 0.43, N = 3SE +/- 0.57, N = 3386.05461.68649.82843.45816.01752.411. (CXX) g++ options: -O3 -march=native -fopenmp

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.orgMiVoxels / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 1920 x 1080RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER2K4K6K8K10KSE +/- 34.46, N = 3SE +/- 81.03, N = 5SE +/- 71.88, N = 3SE +/- 14.53, N = 3SE +/- 22.07, N = 3SE +/- 26.87, N = 310606.707578.149606.538551.757715.266827.82

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 1920 x 1080RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER140280420560700SE +/- 2.15, N = 3SE +/- 5.06, N = 5SE +/- 4.49, N = 3SE +/- 0.91, N = 3SE +/- 1.38, N = 3SE +/- 1.68, N = 3662.92473.63600.41534.48482.20426.74

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: ETC2 RGB - Quality: HighestRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER246810SE +/- 0.011, N = 3SE +/- 0.009, N = 3SE +/- 0.006, N = 3SE +/- 0.006, N = 3SE +/- 0.011, N = 3SE +/- 0.011, N = 34.4166.7945.0086.8317.3308.3941. (CXX) g++ options: -O3 -O2 -lpthread -ldl

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA-512RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER500M1000M1500M2000M2500MSE +/- 2514623.90, N = 3SE +/- 536449.23, N = 3SE +/- 982061.32, N = 3SE +/- 1980179.57, N = 3SE +/- 1276714.53, N = 3SE +/- 633333.33, N = 3242040000014146333332239266667170843333314242000001187933333

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA1RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER4000M8000M12000M16000M20000MSE +/- 26464462.54, N = 3SE +/- 7377744.31, N = 3SE +/- 29255901.59, N = 3SE +/- 1527525.23, N = 3SE +/- 12109546.28, N = 3SE +/- 5454763.46, N = 319098866667110682333331771843333313420400000111214666679341766667

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: MD5RTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER12000M24000M36000M48000M60000MSE +/- 64449006.54, N = 3SE +/- 46443813.13, N = 3SE +/- 36241182.84, N = 3SE +/- 44071015.92, N = 3SE +/- 25610045.77, N = 3SE +/- 14634472.24, N = 3560396666673294036666755298100000423930333333539816666729280266667

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: HighestRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER246810SE +/- 0.024, N = 3SE +/- 0.041, N = 3SE +/- 0.036, N = 7SE +/- 0.028, N = 3SE +/- 0.021, N = 3SE +/- 0.016, N = 33.3285.0093.8225.2965.4566.0321. (CXX) g++ options: -O3 -O2 -lpthread -ldl

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: YesRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER1.1842.3683.5524.7365.92SE +/- 0.006, N = 3SE +/- 0.004, N = 3SE +/- 0.012, N = 3SE +/- 0.020, N = 3SE +/- 0.017, N = 3SE +/- 0.066, N = 33.4394.3723.7764.3474.6895.262

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Boat - Acceleration: OpenCLRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER0.38610.77221.15831.54441.9305SE +/- 0.006, N = 3SE +/- 0.004, N = 3SE +/- 0.010, N = 3SE +/- 0.007, N = 3SE +/- 0.008, N = 3SE +/- 0.005, N = 31.2821.7131.4851.5911.7091.716

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Masskrug - Acceleration: OpenCLRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER0.4880.9761.4641.9522.44SE +/- 0.006, N = 3SE +/- 0.004, N = 3SE +/- 0.006, N = 3SE +/- 0.010, N = 3SE +/- 0.007, N = 3SE +/- 0.003, N = 32.0902.1212.0842.1692.1562.118

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: 7-ZipRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER200K400K600K800K1000KSE +/- 1449.52, N = 3SE +/- 1047.75, N = 3SE +/- 2234.08, N = 3SE +/- 6986.65, N = 3SE +/- 1146.49, N = 3SE +/- 1083.72, N = 3990567585967894867692500596367495567

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Server Room - Acceleration: OpenCLRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER0.17480.34960.52440.69920.874SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.007, N = 3SE +/- 0.004, N = 3SE +/- 0.004, N = 3SE +/- 0.002, N = 30.7060.7770.7100.7560.7760.723

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCLRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER48121620SE +/- 0.014468, N = 3SE +/- 0.179234, N = 3SE +/- 0.016231, N = 3SE +/- 0.084311, N = 9SE +/- 0.083319, N = 3SE +/- 0.196835, N = 37.30700013.1380009.22600010.61000012.80400016.4863331. (CXX) g++ options: -O3 -march=native -fopenmp

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Server Rack - Acceleration: OpenCLRTX 3080RTX 3060 TiRTX 2080 TiRTX 2080 SUPERRTX 2070 SUPERRTX 2060 SUPER0.02270.04540.06810.09080.1135SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 30.0950.1010.0960.0970.0970.097

79 Results Shown

LeelaChessZero
RedShift Demo
VkFFT
NCNN:
  Vulkan GPU - regnety_400m
  Vulkan GPU - squeezenet_ssd
  Vulkan GPU - yolov4-tiny
  Vulkan GPU - resnet50
  Vulkan GPU - alexnet
  Vulkan GPU - resnet18
  Vulkan GPU - vgg16
  Vulkan GPU - googlenet
  Vulkan GPU - blazeface
  Vulkan GPU - efficientnet-b0
  Vulkan GPU - mnasnet
  Vulkan GPU - shufflenet-v2
  Vulkan GPU-v3-v3 - mobilenet-v3
  Vulkan GPU-v2-v2 - mobilenet-v2
  Vulkan GPU - mobilenet
OctaneBench
FAHBench
Warsow:
  2560 x 1440
  1920 x 1200
  1920 x 1080
  3840 x 2160
Chaos Group V-RAY:
  NVIDIA CUDA GPU
  NVIDIA RTX GPU
IndigoBench:
  OpenCL GPU - Bedroom
  OpenCL GPU - Supercar
RealSR-NCNN
VkResample
Ray Tracing In Vulkan:
  3840 x 2160 - Lucy In One Weekend
  3840 x 2160 - Cornell Box + Lucy
  2560 x 1440 - Lucy In One Weekend
  1920 x 1200 - Lucy In One Weekend
  1920 x 1080 - Lucy In One Weekend
  2560 x 1440 - Cornell Box + Lucy
  1920 x 1200 - Cornell Box + Lucy
  1920 x 1080 - Cornell Box + Lucy
  3840 x 2160 - Planets In One Weekend
  3840 x 2160 - Ray Tracing In One Weekend
  2560 x 1440 - Planets In One Weekend
  3840 x 2160 - Cornell Box
  1920 x 1080 - Planets In One Weekend
  1920 x 1200 - Planets In One Weekend
  2560 x 1440 - Ray Tracing In One Weekend
  1920 x 1080 - Ray Tracing In One Weekend
  1920 x 1200 - Ray Tracing In One Weekend
  2560 x 1440 - Cornell Box
  1920 x 1200 - Cornell Box
  1920 x 1080 - Cornell Box
ParaView:
  Many Spheres - 3840 x 2160:
    MiPolys / Sec
    Frames / Sec
  Many Spheres - 1920 x 1080:
    MiPolys / Sec
    Frames / Sec
  Wavelet Contour - 3840 x 2160:
    MiPolys / Sec
    Frames / Sec
RealSR-NCNN
VkResample
NAMD CUDA
ParaView:
  Wavelet Contour - 1920 x 1080:
    MiPolys / Sec
    Frames / Sec
  Wavelet Volume - 3840 x 2160:
    MiVoxels / Sec
    Frames / Sec
Hashcat
FinanceBench
ParaView:
  Wavelet Volume - 1920 x 1080:
    MiVoxels / Sec
    Frames / Sec
Betsy GPU Compressor
Hashcat:
  SHA-512
  SHA1
  MD5
Betsy GPU Compressor
Waifu2x-NCNN Vulkan
Darktable:
  Boat - OpenCL
  Masskrug - OpenCL
Hashcat
Darktable
FinanceBench
Darktable