NVIDIA GeForce RTX 3090

AMD Ryzen 9 5950X 16-Core testing with a ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3801 BIOS) and NVIDIA GeForce RTX 3090 24GB on Ubuntu 21.10 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 2112069-PTS-NVIDIAGE39
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

Limit displaying results to tests within:

BLAS (Basic Linear Algebra Sub-Routine) Tests 3 Tests
C++ Boost Tests 2 Tests
CPU Massive 5 Tests
Creator Workloads 6 Tests
Desktop Graphics 2 Tests
Game Development 2 Tests
HPC - High Performance Computing 8 Tests
Machine Learning 5 Tests
Multi-Core 8 Tests
NVIDIA GPU Compute 31 Tests
OpenCL 6 Tests
OpenMPI Tests 2 Tests
Python Tests 3 Tests
Renderers 4 Tests
Scientific Computing 2 Tests
Server CPU Tests 2 Tests
Vulkan Compute 8 Tests
Common Workstation Benchmarks 3 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
Run
  Test
  Duration
RTX 3090
December 05 2021
  2 Hours, 29 Minutes
NVIDIA RTX 3090
December 05 2021
  7 Hours, 31 Minutes
NVIDIA 3090
December 06 2021
  2 Hours, 25 Minutes
Invert Hiding All Results Option
  4 Hours, 8 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 GeForce RTX 3090OpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3801 BIOS)AMD Starship/Matisse32GB1000GB Sabrent Rocket 4.0 PlusNVIDIA GeForce RTX 3090 24GBNVIDIA GA102 HD AudioASUS MG28URealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 21.105.13.0-22-generic (x86_64)GNOME Shell 40.5X Server 1.20.13NVIDIA 495.444.6.0OpenCL 3.0 CUDA 11.5.1001.2.186GCC 11.2.0 + Clang 13.0.0-2ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionNVIDIA GeForce RTX 3090 BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/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-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 - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201016 - BAR1 / Visible vRAM Size: 32768 MiB- GPU Compute Cores: 10496- Python 3.9.7- 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

RTX 3090NVIDIA RTX 3090NVIDIA 3090Result OverviewPhoronix Test Suite100%103%105%108%LuxCoreRenderRodiniaVkFFTNCNNMandelGPUWaifu2x-NCNN VulkanclpeakLeelaChessZeroNAMD CUDARealSR-NCNNArrayFirevkpeakXonoticViennaCLOctaneBenchVkResampleParaViewChaos Group V-RAYET: LegacyWarsowHashcatFAHBenchUnvanquishedSHOC Scalable HeterOgeneous ComputingBlendercl-memIndigoBenchFinanceBench

NVIDIA GeForce RTX 3090vkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp16-scalarvkpeak: fp16-vec4vkpeak: fp64-scalarvkpeak: int32-scalarvkpeak: int32-vec4vkpeak: int16-scalarvkpeak: int16-vec4realsr-ncnn: 4x - Norealsr-ncnn: 4x - Yeswaifu2x-ncnn: 2x - 3 - Yesvkfft: hashcat: MD5hashcat: SHA1hashcat: 7-Ziphashcat: SHA-512hashcat: TrueCrypt RIPEMD160 + XTSshoc: OpenCL - S3Dshoc: OpenCL - Triadshoc: OpenCL - FFT SPshoc: OpenCL - MD5 Hashshoc: OpenCL - Reductionshoc: OpenCL - GEMM SGEMM_Nshoc: OpenCL - Max SP Flopsshoc: OpenCL - Bus Speed Downloadshoc: OpenCL - Bus Speed Readbackshoc: OpenCL - Texture Read Bandwidthcl-mem: Copycl-mem: Readcl-mem: Writenamd-cuda: ATPase Simulation - 327,506 Atomsvkresample: 2x - Doublevkresample: 2x - Singleetlegacy: 1920 x 1080etlegacy: 1920 x 1200etlegacy: 2560 x 1440etlegacy: 3840 x 2160unvanquished: 1920 x 1080 - Highunvanquished: 1920 x 1200 - Highunvanquished: 2560 x 1440 - Highunvanquished: 3840 x 2160 - Highunvanquished: 1920 x 1080 - Ultraunvanquished: 1920 x 1200 - Ultraunvanquished: 2560 x 1440 - Ultraunvanquished: 3840 x 2160 - Ultraunvanquished: 1920 x 1080 - Mediumunvanquished: 1920 x 1200 - Mediumunvanquished: 2560 x 1440 - Mediumunvanquished: 3840 x 2160 - Mediumwarsow: 1920 x 1080warsow: 1920 x 1200warsow: 2560 x 1440warsow: 3840 x 2160xonotic: 3840 x 2160 - Lowxonotic: 3840 x 2160 - Highxonotic: 3840 x 2160 - Ultraxonotic: 3840 x 2160 - Ultimateoctanebench: Total Scoreparaview: Many Spheres - 1920 x 1080paraview: Many Spheres - 1920 x 1080paraview: Many Spheres - 1920 x 1200paraview: Many Spheres - 1920 x 1200paraview: Many Spheres - 2560 x 1440paraview: Many Spheres - 2560 x 1440paraview: Many Spheres - 3840 x 2160paraview: Many Spheres - 3840 x 2160paraview: Wavelet Volume - 1920 x 1080paraview: Wavelet Volume - 1920 x 1080paraview: Wavelet Volume - 1920 x 1200paraview: Wavelet Volume - 1920 x 1200paraview: Wavelet Volume - 2560 x 1440paraview: Wavelet Volume - 2560 x 1440paraview: Wavelet Volume - 3840 x 2160paraview: Wavelet Volume - 3840 x 2160paraview: Wavelet Contour - 1920 x 1080paraview: Wavelet Contour - 1920 x 1080paraview: Wavelet Contour - 1920 x 1200paraview: Wavelet Contour - 1920 x 1200paraview: Wavelet Contour - 2560 x 1440paraview: Wavelet Contour - 2560 x 1440paraview: Wavelet Contour - 3840 x 2160paraview: Wavelet Contour - 3840 x 2160fahbench: lczero: OpenCLrodinia: OpenCL Particle Filterarrayfire: Conjugate Gradient OpenCLluxcorerender: DLSC - GPUluxcorerender: Danish Mood - GPUluxcorerender: Orange Juice - GPUluxcorerender: LuxCore Benchmark - GPUluxcorerender: Rainbow Colors and Prism - GPUfinancebench: Black-Scholes OpenCLviennacl: CPU BLAS - sCOPYviennacl: CPU BLAS - sAXPYviennacl: CPU BLAS - sDOTviennacl: CPU BLAS - dCOPYviennacl: CPU BLAS - dAXPYviennacl: CPU BLAS - dDOTviennacl: CPU BLAS - dGEMV-Nviennacl: CPU BLAS - dGEMV-Tviennacl: CPU BLAS - dGEMM-NNviennacl: CPU BLAS - dGEMM-NTviennacl: CPU BLAS - dGEMM-TNviennacl: CPU BLAS - dGEMM-TTviennacl: OpenCL BLAS - sCOPYviennacl: OpenCL BLAS - sAXPYviennacl: OpenCL BLAS - sDOTviennacl: OpenCL BLAS - dCOPYviennacl: OpenCL BLAS - dAXPYviennacl: OpenCL BLAS - dDOTviennacl: OpenCL BLAS - dGEMV-Nviennacl: OpenCL BLAS - dGEMV-Tviennacl: OpenCL BLAS - dGEMM-NNviennacl: OpenCL BLAS - dGEMM-NTviennacl: OpenCL BLAS - dGEMM-TNviennacl: OpenCL BLAS - dGEMM-TTncnn: 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_400mblender: BMW27 - CUDAblender: Classroom - CUDAblender: Fishy Cat - CUDAblender: Barbershop - CUDAblender: BMW27 - NVIDIA OptiXblender: Classroom - NVIDIA OptiXblender: Fishy Cat - NVIDIA OptiXblender: Barbershop - NVIDIA OptiXblender: Pabellon Barcelona - CUDAblender: Pabellon Barcelona - NVIDIA OptiXindigobench: OpenCL GPU - Bedroomindigobench: OpenCL GPU - Supercarmandelgpu: GPUclpeak: Integer Compute INTclpeak: Single-Precision Floatclpeak: Double-Precision Doubleclpeak: Global Memory Bandwidthv-ray: NVIDIA RTX GPUv-ray: NVIDIA CUDA GPUvkpeak: fp64-vec4RTX 3090NVIDIA RTX 3090NVIDIA 309020927.9927806.7420949.6441184.89653.6120769.2720672.1913708.5816885.795.62128.9493.31345831714469000002259930000011385002887000000846100429.08225.47812100.5144.1932391.2098102.3840753.926.334227.12522246.17364.1794.6743.10.12792117.8179.272652.3645.3640.2648.4482464.9469463.6470.8463461.7456.8496.1473.6492.5490.1984.3984.8984.7960.8655.142588556.9297001499.2632615374.4286255686.23125893.959419.39993.919414.68693.29343.81791.469169.636679.6710874.708634.4710151.478559.058944.746384.846157.361602.096274.519561.325849.675514.935366.18396.744134.488353.8915227393.8431.479.1810.4532.336.25966.199.614023.334.942.468.176.181.588.493.891.93685033756067226602393776016045996024.281.983.171.942.083.261.065.854.171.671.913.557.0215.962.5411.0422.7622.8691.136.9417.411.2256.1948.1117.9620.96153.438481322759.617725.1435227.21658.16813.422856211920826.6627457.9020717.3140926.17649.5320689.7220594.7813657.3616880.015.71529.0643.36244252714113666672267850000011401002884666667825633430.34725.48402101.0844.5453391.5738336.7140238.826.328227.12522240.09364.2794.3744.70.12912118.5179.292654.5659.1647.6638.7479.0478.3467.6471.0469.5469.8452.5461.2487.3488.5488.5485.6975.5980.3985.7980.7659.9059394567.7214300492.4660277369.9242194685.10965694.059429.53694.289451.90393.339356.71091.949217.228689.8111036.994643.7510300.044544.988719.750376.776028.366595.526206.035563.055867.721515.785374.983390.644070.881353.3272227113.6941.47011.579.2010.4111.2032.606.25865.598.213723.234.741.667.175.581.984.492.590.73645013716057226572403776006056016014.221.972.241.82.083.251.036.074.181.671.923.556.7217.312.5311.0222.6922.8791.296.9117.4311.3256.3548.1818.0020.95153.500475794831.718044.2535136.60657.71813.4528562112651.1720960.2527806.5520953.7241496.68658.520924.9820672.2113709.4117012.235.67328.8913.31844147714362000002281310000011415002892900000816300430.27925.45052101.8244.5358392.038098.4640566.226.320227.09042245.23362.8795.2742.80.12779119.0739.291659.8657.8643.7635.5440474.6475.7472.9474.5471.1466.6469.7491.9484481.2483984.8984.9985.5978.6645.8515908566.1228693490.2778714367.1052632681.12610893.799402.6392.139236.62993.579380.93292.039226.854696.4711143.486656.1210497.894565.69049.543381.696107.065599.526247.735560.035836.216521.095430.386391.634081.298354.4623230293.6271.48511.619.0110.4311.2232.256.25665.398.313723.234.342.267.575.482.686.892.890.83645003716087236372373776046066026054.231.972.241.82.083.421.055.754.171.71.93.567.4319.732.5311.0922.6422.7890.366.9217.9811.2256.3448.0917.9620.94353.6472928214.818742.735225.4658.16813.4728292115658.51OpenBenchmarking.org

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 20210424fp32-scalarNVIDIA 3090RTX 3090NVIDIA RTX 30904K8K12K16K20KSE +/- 65.72, N = 320960.2520927.9920826.66

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4RTX 3090NVIDIA 3090NVIDIA RTX 30906K12K18K24K30KSE +/- 69.31, N = 327806.7427806.5527457.90

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarNVIDIA 3090RTX 3090NVIDIA RTX 30904K8K12K16K20KSE +/- 45.34, N = 320953.7220949.6420717.31

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4NVIDIA 3090RTX 3090NVIDIA RTX 30909K18K27K36K45KSE +/- 52.39, N = 341496.6841184.8940926.17

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarNVIDIA 3090RTX 3090NVIDIA RTX 3090140280420560700SE +/- 0.83, N = 3658.50653.61649.53

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarNVIDIA 3090RTX 3090NVIDIA RTX 30904K8K12K16K20KSE +/- 44.97, N = 320924.9820769.2720689.72

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4NVIDIA 3090RTX 3090NVIDIA RTX 30904K8K12K16K20KSE +/- 44.80, N = 320672.2120672.1920594.78

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarNVIDIA 3090RTX 3090NVIDIA RTX 30903K6K9K12K15KSE +/- 29.87, N = 313709.4113708.5813657.36

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4NVIDIA 3090RTX 3090NVIDIA RTX 30904K8K12K16K20KSE +/- 2.98, N = 317012.2316885.7916880.01

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 3090NVIDIA 3090NVIDIA RTX 30901.28592.57183.85775.14366.4295SE +/- 0.022, N = 35.6215.6735.715

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesNVIDIA 3090RTX 3090NVIDIA RTX 3090714212835SE +/- 0.06, N = 328.8928.9529.06

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.

Scale: 2x - Denoise: 3 - TAA: No

RTX 3090: Test failed to run.

NVIDIA RTX 3090: Test failed to run.

NVIDIA 3090: Test failed to run.

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesRTX 3090NVIDIA 3090NVIDIA RTX 30900.75651.5132.26953.0263.7825SE +/- 0.004, N = 33.3133.3183.362

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 3090NVIDIA RTX 3090NVIDIA 309010K20K30K40K50KSE +/- 384.18, N = 94583144252441471. (CXX) g++ options: -O3

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 6.2.4Benchmark: MD5RTX 3090NVIDIA 3090NVIDIA RTX 309015000M30000M45000M60000M75000MSE +/- 78065236.25, N = 3714469000007143620000071411366667

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA1NVIDIA 3090NVIDIA RTX 3090RTX 30905000M10000M15000M20000M25000MSE +/- 32122473.96, N = 3228131000002267850000022599300000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: 7-ZipNVIDIA 3090NVIDIA RTX 3090RTX 3090200K400K600K800K1000KSE +/- 4014.97, N = 3114150011401001138500

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA-512NVIDIA 3090RTX 3090NVIDIA RTX 3090600M1200M1800M2400M3000MSE +/- 2179704.36, N = 3289290000028870000002884666667

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSRTX 3090NVIDIA RTX 3090NVIDIA 3090200K400K600K800K1000KSE +/- 4520.08, N = 3846100825633816300

Mixbench

A benchmark suite for GPUs on mixed operational intensity kernels. Learn more via the OpenBenchmarking.org test page.

Backend: OpenCL - Benchmark: Integer

RTX 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

NVIDIA RTX 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

NVIDIA 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: NVIDIA CUDA - Benchmark: Integer

RTX 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

NVIDIA RTX 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

NVIDIA 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: OpenCL - Benchmark: Double Precision

RTX 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

NVIDIA RTX 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

NVIDIA 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: OpenCL - Benchmark: Single Precision

RTX 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

NVIDIA RTX 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

NVIDIA 3090: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: NVIDIA CUDA - Benchmark: Half Precision

RTX 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

NVIDIA RTX 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

NVIDIA 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: NVIDIA CUDA - Benchmark: Double Precision

RTX 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

NVIDIA RTX 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

NVIDIA 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: NVIDIA CUDA - Benchmark: Single Precision

RTX 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

NVIDIA RTX 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

NVIDIA 3090: ./mixbench: 3: ./mixbench-cuda-ro: not found

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3DNVIDIA RTX 3090NVIDIA 3090RTX 309090180270360450SE +/- 0.21, N = 3430.35430.28429.081. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: TriadNVIDIA RTX 3090RTX 3090NVIDIA 3090612182430SE +/- 0.00, N = 325.4825.4825.451. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: FFT SPNVIDIA 3090NVIDIA RTX 3090RTX 30905001000150020002500SE +/- 0.42, N = 32101.822101.082100.511. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 HashNVIDIA RTX 3090NVIDIA 3090RTX 30901020304050SE +/- 0.00, N = 344.5544.5444.191. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: ReductionNVIDIA 3090NVIDIA RTX 3090RTX 309090180270360450SE +/- 0.19, N = 3392.03391.57391.211. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_NNVIDIA RTX 3090RTX 3090NVIDIA 30902K4K6K8K10KSE +/- 67.62, N = 38336.718102.388098.461. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Max SP FlopsRTX 3090NVIDIA 3090NVIDIA RTX 30909K18K27K36K45KSE +/- 297.72, N = 340753.940566.240238.81. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed DownloadRTX 3090NVIDIA RTX 3090NVIDIA 3090612182430SE +/- 0.01, N = 326.3326.3326.321. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed ReadbackNVIDIA RTX 3090RTX 3090NVIDIA 3090612182430SE +/- 0.00, N = 327.1327.1327.091. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthRTX 3090NVIDIA 3090NVIDIA RTX 30905001000150020002500SE +/- 3.07, N = 32246.172245.232240.091. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

Libplacebo

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

RTX 3090: The test quit with a non-zero exit status.

NVIDIA RTX 3090: The test quit with a non-zero exit status.

NVIDIA 3090: The test quit with a non-zero exit status.

cl-mem

A basic OpenCL memory benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: CopyNVIDIA RTX 3090RTX 3090NVIDIA 309080160240320400SE +/- 0.19, N = 3364.2364.1362.81. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: ReadNVIDIA 3090RTX 3090NVIDIA RTX 30902004006008001000SE +/- 0.84, N = 3795.2794.6794.31. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: WriteNVIDIA RTX 3090RTX 3090NVIDIA 3090160320480640800SE +/- 0.33, N = 3744.7743.1742.81. (CC) gcc options: -O2 -flto -lOpenCL

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 AtomsNVIDIA 3090RTX 3090NVIDIA RTX 30900.02910.05820.08730.11640.1455SE +/- 0.00047, N = 30.127790.127920.12912

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.

Codec: ETC1 - Quality: Highest

RTX 3090: ./betsy: 3: ./betsy: not found

NVIDIA RTX 3090: ./betsy: 3: ./betsy: not found

NVIDIA 3090: ./betsy: 3: ./betsy: not found

Codec: ETC2 RGB - Quality: Highest

RTX 3090: ./betsy: 3: ./betsy: not found

NVIDIA RTX 3090: ./betsy: 3: ./betsy: not found

NVIDIA 3090: ./betsy: 3: ./betsy: not found

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 3090NVIDIA RTX 3090NVIDIA 3090306090120150SE +/- 0.08, N = 3117.82118.52119.071. (CXX) g++ options: -O3

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleRTX 3090NVIDIA 3090NVIDIA RTX 30903691215SE +/- 0.003, N = 39.2729.2919.2921. (CXX) g++ options: -O3

ET: Legacy

ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.78Resolution: 1920 x 1080NVIDIA 3090NVIDIA RTX 3090RTX 3090140280420560700SE +/- 3.73, N = 3659.8654.5652.3

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.78Resolution: 1920 x 1200NVIDIA RTX 3090NVIDIA 3090RTX 3090140280420560700SE +/- 2.46, N = 3659.1657.8645.3

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.78Resolution: 2560 x 1440NVIDIA RTX 3090NVIDIA 3090RTX 3090140280420560700SE +/- 7.88, N = 4647.6643.7640.2

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.78Resolution: 3840 x 2160RTX 3090NVIDIA RTX 3090NVIDIA 3090140280420560700SE +/- 6.04, N = 3648.4638.7635.5

Unvanquished

Unvanquished is a modern fork of the Tremulous first person shooter. Unvanquished is powered by the Daemon engine, a combination of the ioquake3 engine with the graphically-beautiful XreaL engine. Unvanquished supports a modern OpenGL 3 renderer and other advanced graphics features for this open-source game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 1920 x 1080 - Effects Quality: HighRTX 3090NVIDIA RTX 3090NVIDIA 3090100200300400500SE +/- 2.19, N = 3482.0479.0440.0

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 1920 x 1200 - Effects Quality: HighNVIDIA RTX 3090NVIDIA 3090RTX 3090100200300400500SE +/- 4.62, N = 3478.3474.6464.9

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 2560 x 1440 - Effects Quality: HighNVIDIA 3090RTX 3090NVIDIA RTX 3090100200300400500SE +/- 2.97, N = 3475.7469.0467.6

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 3840 x 2160 - Effects Quality: HighNVIDIA 3090NVIDIA RTX 3090RTX 3090100200300400500SE +/- 3.53, N = 3472.9471.0463.6

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 1920 x 1080 - Effects Quality: UltraNVIDIA 3090RTX 3090NVIDIA RTX 3090100200300400500SE +/- 1.10, N = 3474.5470.8469.5

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 1920 x 1200 - Effects Quality: UltraNVIDIA 3090NVIDIA RTX 3090RTX 3090100200300400500SE +/- 5.55, N = 3471.1469.8463.0

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 2560 x 1440 - Effects Quality: UltraNVIDIA 3090RTX 3090NVIDIA RTX 3090100200300400500SE +/- 14.16, N = 12466.6461.7452.5

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 3840 x 2160 - Effects Quality: UltraNVIDIA 3090NVIDIA RTX 3090RTX 3090100200300400500SE +/- 4.41, N = 3469.7461.2456.8

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 1920 x 1080 - Effects Quality: MediumRTX 3090NVIDIA 3090NVIDIA RTX 3090110220330440550SE +/- 3.91, N = 3496.1491.9487.3

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 1920 x 1200 - Effects Quality: MediumNVIDIA RTX 3090NVIDIA 3090RTX 3090110220330440550SE +/- 1.98, N = 3488.5484.0473.6

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 2560 x 1440 - Effects Quality: MediumRTX 3090NVIDIA RTX 3090NVIDIA 3090110220330440550SE +/- 3.75, N = 3492.5488.5481.2

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.52.1Resolution: 3840 x 2160 - Effects Quality: MediumRTX 3090NVIDIA RTX 3090NVIDIA 3090110220330440550SE +/- 2.17, N = 3490.1485.6483.0

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: 1920 x 1080NVIDIA 3090RTX 3090NVIDIA RTX 30902004006008001000SE +/- 5.54, N = 3984.8984.3975.5

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 1920 x 1200NVIDIA 3090RTX 3090NVIDIA RTX 30902004006008001000SE +/- 5.60, N = 3984.9984.8980.3

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 2560 x 1440NVIDIA RTX 3090NVIDIA 3090RTX 30902004006008001000SE +/- 0.35, N = 3985.7985.5984.7

OpenBenchmarking.orgFrames Per Second, More Is BetterWarsow 2.5 BetaResolution: 3840 x 2160NVIDIA RTX 3090NVIDIA 3090RTX 30902004006008001000SE +/- 0.78, N = 3980.7978.6960.8

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: LowNVIDIA RTX 3090RTX 3090NVIDIA 3090140280420560700SE +/- 0.55, N = 3659.91655.14645.85MIN: 109 / MAX: 1303MIN: 103 / MAX: 1300MIN: 117 / MAX: 1283

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: HighNVIDIA RTX 3090NVIDIA 3090RTX 3090120240360480600SE +/- 1.97, N = 3567.72566.12556.93MIN: 91 / MAX: 1163MIN: 113 / MAX: 1151MIN: 114 / MAX: 1122

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltraRTX 3090NVIDIA RTX 3090NVIDIA 3090110220330440550SE +/- 1.15, N = 3499.26492.47490.28MIN: 117 / MAX: 913MIN: 93 / MAX: 902MIN: 122 / MAX: 891

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltimateRTX 3090NVIDIA RTX 3090NVIDIA 309080160240320400SE +/- 1.53, N = 3374.43369.92367.11MIN: 65 / MAX: 751MIN: 58 / MAX: 764MIN: 65 / MAX: 749

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 3090NVIDIA RTX 3090NVIDIA 3090150300450600750686.23685.11681.13

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.

RTX 3090: The test quit with a non-zero exit status.

NVIDIA RTX 3090: The test quit with a non-zero exit status.

NVIDIA 3090: The test quit with a non-zero exit status.

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 5.9Test: Many Spheres - Resolution: 1920 x 1080NVIDIA RTX 3090RTX 3090NVIDIA 309020406080100SE +/- 0.05, N = 394.0593.9593.79

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 1920 x 1080NVIDIA RTX 3090RTX 3090NVIDIA 30902K4K6K8K10KSE +/- 4.72, N = 39429.549419.409402.63

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 1920 x 1200NVIDIA RTX 3090RTX 3090NVIDIA 309020406080100SE +/- 0.06, N = 394.2893.9192.13

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 1920 x 1200NVIDIA RTX 3090RTX 3090NVIDIA 30902K4K6K8K10KSE +/- 5.73, N = 39451.909414.699236.63

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 2560 x 1440NVIDIA 3090NVIDIA RTX 3090RTX 309020406080100SE +/- 0.08, N = 393.5793.3393.20

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 2560 x 1440NVIDIA 3090NVIDIA RTX 3090RTX 30902K4K6K8K10KSE +/- 7.77, N = 39380.939356.719343.82

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 3840 x 2160NVIDIA 3090NVIDIA RTX 3090RTX 309020406080100SE +/- 0.06, N = 392.0391.9491.46

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Many Spheres - Resolution: 3840 x 2160NVIDIA 3090NVIDIA RTX 3090RTX 30902K4K6K8K10KSE +/- 6.01, N = 39226.859217.239169.64

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 1920 x 1080NVIDIA 3090NVIDIA RTX 3090RTX 3090150300450600750SE +/- 5.50, N = 3696.47689.81679.67

OpenBenchmarking.orgMiVoxels / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 1920 x 1080NVIDIA 3090NVIDIA RTX 3090RTX 30902K4K6K8K10KSE +/- 88.00, N = 311143.4911036.9910874.71

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 1920 x 1200NVIDIA 3090NVIDIA RTX 3090RTX 3090140280420560700SE +/- 4.68, N = 3656.12643.75634.47

OpenBenchmarking.orgMiVoxels / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 1920 x 1200NVIDIA 3090NVIDIA RTX 3090RTX 30902K4K6K8K10KSE +/- 74.83, N = 310497.8910300.0410151.48

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 2560 x 1440NVIDIA 3090RTX 3090NVIDIA RTX 3090120240360480600SE +/- 3.11, N = 3565.60559.05544.98

OpenBenchmarking.orgMiVoxels / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 2560 x 1440NVIDIA 3090RTX 3090NVIDIA RTX 30902K4K6K8K10KSE +/- 49.80, N = 39049.548944.758719.75

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 3840 x 2160RTX 3090NVIDIA 3090NVIDIA RTX 309080160240320400SE +/- 4.06, N = 5384.84381.69376.77

OpenBenchmarking.orgMiVoxels / Sec, More Is BetterParaView 5.9Test: Wavelet Volume - Resolution: 3840 x 2160RTX 3090NVIDIA 3090NVIDIA RTX 309013002600390052006500SE +/- 65.03, N = 56157.366107.076028.37

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 1920 x 1080RTX 3090NVIDIA 3090NVIDIA RTX 3090130260390520650SE +/- 5.27, N = 3602.09599.52595.52

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 1920 x 1080RTX 3090NVIDIA 3090NVIDIA RTX 309013002600390052006500SE +/- 54.90, N = 36274.526247.746206.04

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 1920 x 1200NVIDIA RTX 3090RTX 3090NVIDIA 3090120240360480600SE +/- 5.24, N = 3563.05561.32560.03

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 1920 x 1200NVIDIA RTX 3090RTX 3090NVIDIA 309013002600390052006500SE +/- 54.66, N = 35867.725849.685836.22

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 2560 x 1440NVIDIA 3090NVIDIA RTX 3090RTX 3090110220330440550SE +/- 1.60, N = 3521.09515.78514.93

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 2560 x 1440NVIDIA 3090NVIDIA RTX 3090RTX 309012002400360048006000SE +/- 16.63, N = 35430.395374.985366.18

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 3840 x 2160RTX 3090NVIDIA 3090NVIDIA RTX 309090180270360450SE +/- 2.99, N = 3396.74391.63390.64

OpenBenchmarking.orgMiPolys / Sec, More Is BetterParaView 5.9Test: Wavelet Contour - Resolution: 3840 x 2160RTX 3090NVIDIA 3090NVIDIA RTX 30909001800270036004500SE +/- 31.15, N = 34134.494081.304070.88

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 2.3.2NVIDIA 3090RTX 3090NVIDIA RTX 309080160240320400SE +/- 0.32, N = 3354.46353.89353.33

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 0.28Backend: OpenCLNVIDIA 3090RTX 3090NVIDIA RTX 30905K10K15K20K25KSE +/- 98.83, N = 32302922739227111. (CXX) g++ options: -flto -pthread

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle FilterNVIDIA 3090NVIDIA RTX 3090RTX 30900.86471.72942.59413.45884.3235SE +/- 0.013, N = 33.6273.6943.8431. (CXX) g++ options: -O2 -lOpenCL

ArrayFire

ArrayFire is an GPU and CPU numeric processing library, this test uses the built-in CPU and OpenCL ArrayFire benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient OpenCLRTX 3090NVIDIA RTX 3090NVIDIA 30900.33410.66821.00231.33641.6705SE +/- 0.006, N = 31.4701.4701.4851. (CXX) g++ options: -rdynamic

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: DLSC - Acceleration: GPUNVIDIA 3090NVIDIA RTX 30903691215SE +/- 0.03, N = 311.6111.57MIN: 11.41 / MAX: 11.72MIN: 11.39 / MAX: 11.73

Scene: DLSC - Acceleration: GPU

RTX 3090: Test failed to run.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Danish Mood - Acceleration: GPUNVIDIA RTX 3090RTX 3090NVIDIA 30903691215SE +/- 0.08, N = 39.209.189.01MIN: 3.03 / MAX: 10.91MIN: 3.53 / MAX: 10.86MIN: 3.3 / MAX: 10.77

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Orange Juice - Acceleration: GPURTX 3090NVIDIA 3090NVIDIA RTX 30903691215SE +/- 0.10, N = 310.4510.4310.41MIN: 8.53 / MAX: 13.8MIN: 8.54 / MAX: 13.75MIN: 8.47 / MAX: 13.76

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: LuxCore Benchmark - Acceleration: GPUNVIDIA 3090NVIDIA RTX 30903691215SE +/- 0.01, N = 311.2211.20MIN: 3.57 / MAX: 13.17MIN: 3.54 / MAX: 13.16

Scene: LuxCore Benchmark - Acceleration: GPU

RTX 3090: [LuxCore][5.612] PhotonGI estimated current indirect photon error: 23.06%

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Rainbow Colors and Prism - Acceleration: GPUNVIDIA RTX 3090RTX 3090NVIDIA 3090816243240SE +/- 0.21, N = 332.6032.3332.25MIN: 30.12 / MAX: 34.6MIN: 30.03 / MAX: 34.38MIN: 29.03 / MAX: 34.55

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 OpenCLNVIDIA 3090NVIDIA RTX 3090RTX 3090246810SE +/- 0.004, N = 36.2566.2586.2591. (CXX) g++ options: -O3 -march=native -fopenmp

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPYRTX 3090NVIDIA RTX 3090NVIDIA 30901530456075SE +/- 0.09, N = 366.165.565.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPYRTX 3090NVIDIA 3090NVIDIA RTX 309020406080100SE +/- 0.68, N = 399.698.398.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOTRTX 3090NVIDIA 3090NVIDIA RTX 3090306090120150SE +/- 0.67, N = 31401371371. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPYRTX 3090NVIDIA 3090NVIDIA RTX 3090612182430SE +/- 0.00, N = 323.323.223.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPYRTX 3090NVIDIA RTX 3090NVIDIA 3090816243240SE +/- 0.00, N = 334.934.734.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOTRTX 3090NVIDIA 3090NVIDIA RTX 30901020304050SE +/- 0.42, N = 342.442.241.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-NRTX 3090NVIDIA 3090NVIDIA RTX 30901530456075SE +/- 0.17, N = 368.167.567.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-TRTX 3090NVIDIA RTX 3090NVIDIA 309020406080100SE +/- 0.36, N = 376.175.575.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NNNVIDIA 3090NVIDIA RTX 3090RTX 309020406080100SE +/- 0.18, N = 382.681.981.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NTRTX 3090NVIDIA 3090NVIDIA RTX 309020406080100SE +/- 0.12, N = 388.486.884.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TNRTX 3090NVIDIA 3090NVIDIA RTX 309020406080100SE +/- 0.32, N = 393.892.892.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TTRTX 3090NVIDIA 3090NVIDIA RTX 309020406080100SE +/- 0.26, N = 391.990.890.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPYRTX 3090NVIDIA 3090NVIDIA RTX 309080160240320400SE +/- 0.88, N = 33683643641. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPYRTX 3090NVIDIA RTX 3090NVIDIA 3090110220330440550SE +/- 0.33, N = 35035015001. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTRTX 3090NVIDIA 3090NVIDIA RTX 309080160240320400SE +/- 0.58, N = 33753713711. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPYNVIDIA 3090RTX 3090NVIDIA RTX 3090130260390520650SE +/- 0.33, N = 36086066051. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYNVIDIA 3090NVIDIA RTX 3090RTX 3090160320480640800SE +/- 0.00, N = 37237227221. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTRTX 3090NVIDIA RTX 3090NVIDIA 3090140280420560700SE +/- 0.58, N = 36606576371. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-NNVIDIA RTX 3090RTX 3090NVIDIA 309050100150200250SE +/- 0.33, N = 32402392371. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-TNVIDIA 3090NVIDIA RTX 3090RTX 309080160240320400SE +/- 0.33, N = 33773773771. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNNVIDIA 3090RTX 3090NVIDIA RTX 30901302603905206506046016001. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTNVIDIA 3090NVIDIA RTX 3090RTX 3090130260390520650SE +/- 1.67, N = 36066056041. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNNVIDIA 3090NVIDIA RTX 3090RTX 3090130260390520650SE +/- 1.33, N = 36026015991. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTNVIDIA 3090RTX 3090NVIDIA RTX 30901302603905206506056026011. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

Implementation: NVIDIA CUDA GPU - Input: water_GMX50_bare

RTX 3090: ./gromacs: 5: /cuda-build/run-gromacs: not found

NVIDIA RTX 3090: ./gromacs: 5: /cuda-build/run-gromacs: not found

NVIDIA 3090: ./gromacs: 5: /cuda-build/run-gromacs: not found

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 100

RTX 3090: @ 0x7f673b18935f google::LogMessageFatal::~LogMessageFatal()

NVIDIA RTX 3090: @ 0x7ff1662db35f google::LogMessageFatal::~LogMessageFatal()

NVIDIA 3090: @ 0x7fa229dc935f google::LogMessageFatal::~LogMessageFatal()

Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 200

RTX 3090: @ 0x7ff7c4d3335f google::LogMessageFatal::~LogMessageFatal()

NVIDIA RTX 3090: @ 0x7efc6619835f google::LogMessageFatal::~LogMessageFatal()

NVIDIA 3090: @ 0x7f70627d835f google::LogMessageFatal::~LogMessageFatal()

Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 1000

RTX 3090: @ 0x7f2baa38435f google::LogMessageFatal::~LogMessageFatal()

NVIDIA RTX 3090: @ 0x7fa1401b935f google::LogMessageFatal::~LogMessageFatal()

NVIDIA 3090: @ 0x7f06ca10535f google::LogMessageFatal::~LogMessageFatal()

Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 100

RTX 3090: @ 0x7fcf7a66c35f google::LogMessageFatal::~LogMessageFatal()

NVIDIA RTX 3090: @ 0x7fc8cad7235f google::LogMessageFatal::~LogMessageFatal()

NVIDIA 3090: @ 0x7fc82391235f google::LogMessageFatal::~LogMessageFatal()

Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 200

RTX 3090: @ 0x7f07fc1da35f google::LogMessageFatal::~LogMessageFatal()

NVIDIA RTX 3090: @ 0x7fe536ef535f google::LogMessageFatal::~LogMessageFatal()

NVIDIA 3090: @ 0x7fcccd8ca35f google::LogMessageFatal::~LogMessageFatal()

Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 1000

RTX 3090: @ 0x7f9e6bb6a35f google::LogMessageFatal::~LogMessageFatal()

NVIDIA RTX 3090: @ 0x7f9599f2935f google::LogMessageFatal::~LogMessageFatal()

NVIDIA 3090: @ 0x7f1a673f235f google::LogMessageFatal::~LogMessageFatal()

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: mobilenetNVIDIA RTX 3090NVIDIA 3090RTX 30900.9631.9262.8893.8524.815SE +/- 0.01, N = 34.224.234.28MIN: 4.13 / MAX: 4.48MIN: 4.16 / MAX: 4.41MIN: 4.15 / MAX: 15.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2NVIDIA RTX 3090NVIDIA 3090RTX 30900.44550.8911.33651.7822.2275SE +/- 0.00, N = 31.971.971.98MIN: 1.94 / MAX: 3.15MIN: 1.94 / MAX: 2.26MIN: 1.94 / MAX: 6.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3NVIDIA RTX 3090NVIDIA 3090RTX 30900.71331.42662.13992.85323.5665SE +/- 0.00, N = 32.242.243.17MIN: 2.21 / MAX: 3.37MIN: 2.21 / MAX: 4.61MIN: 2.21 / MAX: 20.431. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: shufflenet-v2NVIDIA RTX 3090NVIDIA 3090RTX 30900.43650.8731.30951.7462.1825SE +/- 0.00, N = 31.801.801.94MIN: 1.76 / MAX: 3MIN: 1.77 / MAX: 2.69MIN: 1.78 / MAX: 2.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mnasnetRTX 3090NVIDIA RTX 3090NVIDIA 30900.4680.9361.4041.8722.34SE +/- 0.01, N = 32.082.082.08MIN: 2.06 / MAX: 2.29MIN: 2.05 / MAX: 3.23MIN: 2.06 / MAX: 2.511. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: efficientnet-b0NVIDIA RTX 3090RTX 3090NVIDIA 30900.76951.5392.30853.0783.8475SE +/- 0.00, N = 33.253.263.42MIN: 3.22 / MAX: 4.42MIN: 3.23 / MAX: 3.42MIN: 3.24 / MAX: 4.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: blazefaceNVIDIA RTX 3090NVIDIA 3090RTX 30900.23850.4770.71550.9541.1925SE +/- 0.00, N = 31.031.051.06MIN: 1 / MAX: 2.07MIN: 1.01 / MAX: 2.18MIN: 1.02 / MAX: 2.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: googlenetNVIDIA 3090RTX 3090NVIDIA RTX 3090246810SE +/- 0.26, N = 35.755.856.07MIN: 3.72 / MAX: 31MIN: 3.71 / MAX: 29.49MIN: 3.71 / MAX: 31.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: vgg16RTX 3090NVIDIA 3090NVIDIA RTX 30900.94051.8812.82153.7624.7025SE +/- 0.02, N = 34.174.174.18MIN: 4.12 / MAX: 4.86MIN: 4.13 / MAX: 4.34MIN: 4.12 / MAX: 11.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet18RTX 3090NVIDIA RTX 3090NVIDIA 30900.38250.7651.14751.531.9125SE +/- 0.00, N = 31.671.671.70MIN: 1.64 / MAX: 1.9MIN: 1.64 / MAX: 4.75MIN: 1.65 / MAX: 9.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: alexnetNVIDIA 3090RTX 3090NVIDIA RTX 30900.4320.8641.2961.7282.16SE +/- 0.01, N = 31.901.911.92MIN: 1.87 / MAX: 2.6MIN: 1.86 / MAX: 9.04MIN: 1.87 / MAX: 7.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet50RTX 3090NVIDIA RTX 3090NVIDIA 30900.8011.6022.4033.2044.005SE +/- 0.00, N = 33.553.553.56MIN: 3.52 / MAX: 3.69MIN: 3.52 / MAX: 3.75MIN: 3.53 / MAX: 3.691. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: yolov4-tinyNVIDIA RTX 3090RTX 3090NVIDIA 3090246810SE +/- 0.04, N = 36.727.027.43MIN: 6.3 / MAX: 19.53MIN: 6.34 / MAX: 30.81MIN: 6.36 / MAX: 36.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: squeezenet_ssdRTX 3090NVIDIA RTX 3090NVIDIA 3090510152025SE +/- 1.24, N = 315.9617.3119.73MIN: 5.77 / MAX: 36.07MIN: 5.63 / MAX: 36.32MIN: 7.36 / MAX: 35.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: regnety_400mNVIDIA RTX 3090NVIDIA 3090RTX 30900.57151.1431.71452.2862.8575SE +/- 0.01, N = 32.532.532.54MIN: 2.49 / MAX: 3.64MIN: 2.51 / MAX: 2.76MIN: 2.5 / MAX: 3.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

FP16: No - Mode: Training - Network: Mobilenet - Device: OpenCL

RTX 3090: Test failed to run.

NVIDIA RTX 3090: Test failed to run.

NVIDIA 3090: Test failed to run.

FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL

RTX 3090: Test failed to run.

NVIDIA RTX 3090: Test failed to run.

NVIDIA 3090: Test failed to run.

FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL

RTX 3090: Test failed to run.

NVIDIA RTX 3090: Test failed to run.

NVIDIA 3090: Test failed to run.

FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL

RTX 3090: Test failed to run.

NVIDIA RTX 3090: Test failed to run.

NVIDIA 3090: Test failed to run.

FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL

RTX 3090: Test failed to run.

NVIDIA RTX 3090: Test failed to run.

NVIDIA 3090: Test failed to run.

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: BMW27 - Compute: CUDANVIDIA RTX 3090RTX 3090NVIDIA 30903691215SE +/- 0.01, N = 311.0211.0411.09

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: Classroom - Compute: CUDANVIDIA 3090NVIDIA RTX 3090RTX 3090510152025SE +/- 0.03, N = 322.6422.6922.76

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: Fishy Cat - Compute: CUDANVIDIA 3090RTX 3090NVIDIA RTX 3090510152025SE +/- 0.04, N = 322.7822.8622.87

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: Barbershop - Compute: CUDANVIDIA 3090RTX 3090NVIDIA RTX 309020406080100SE +/- 0.17, N = 390.3691.1391.29

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: BMW27 - Compute: NVIDIA OptiXNVIDIA RTX 3090NVIDIA 3090RTX 3090246810SE +/- 0.01, N = 36.916.926.94

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: Classroom - Compute: NVIDIA OptiXRTX 3090NVIDIA RTX 3090NVIDIA 309048121620SE +/- 0.02, N = 317.4017.4317.98

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: Fishy Cat - Compute: NVIDIA OptiXRTX 3090NVIDIA 3090NVIDIA RTX 30903691215SE +/- 0.09, N = 311.2211.2211.32

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: Barbershop - Compute: NVIDIA OptiXRTX 3090NVIDIA 3090NVIDIA RTX 30901326395265SE +/- 0.06, N = 356.1956.3456.35

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: Pabellon Barcelona - Compute: CUDANVIDIA 3090RTX 3090NVIDIA RTX 30901122334455SE +/- 0.01, N = 348.0948.1148.18

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.0Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXRTX 3090NVIDIA 3090NVIDIA RTX 309048121620SE +/- 0.02, N = 317.9617.9618.00

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 3090NVIDIA RTX 3090NVIDIA 3090510152025SE +/- 0.01, N = 320.9620.9520.94

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: SupercarNVIDIA 3090NVIDIA RTX 3090RTX 30901224364860SE +/- 0.03, N = 353.6053.5053.44

MandelGPU

MandelGPU is an OpenCL benchmark and this test runs with the OpenCL rendering float4 kernel with a maximum of 4096 iterations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPURTX 3090NVIDIA RTX 3090NVIDIA 3090100M200M300M400M500MSE +/- 1595019.30, N = 3481322759.6475794831.7472928214.81. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

clpeak

Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTNVIDIA 3090NVIDIA RTX 3090RTX 30904K8K12K16K20KSE +/- 115.05, N = 318742.7018044.2517725.141. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatRTX 3090NVIDIA 3090NVIDIA RTX 30908K16K24K32K40KSE +/- 88.80, N = 335227.2135225.4035136.601. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleNVIDIA 3090RTX 3090NVIDIA RTX 3090140280420560700SE +/- 0.03, N = 3658.16658.16657.711. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory BandwidthNVIDIA 3090NVIDIA RTX 3090RTX 30902004006008001000SE +/- 0.02, N = 3813.47813.45813.421. (CXX) g++ options: -O3 -rdynamic -lOpenCL

NeatBench

NeatBench is a benchmark of the cross-platform Neat Video software on the CPU and optional GPU (OpenCL / CUDA) support. Learn more via the OpenBenchmarking.org test page.

Acceleration: GPU

RTX 3090: Test failed to run.

NVIDIA RTX 3090: Test failed to run.

NVIDIA 3090: Test failed to run.

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.orgvrays, More Is BetterChaos Group V-RAY 5Mode: NVIDIA RTX GPUNVIDIA RTX 3090RTX 3090NVIDIA 30906001200180024003000SE +/- 11.72, N = 3285628562829

OpenBenchmarking.orgvpaths, More Is BetterChaos Group V-RAY 5Mode: NVIDIA CUDA GPURTX 3090NVIDIA 3090NVIDIA RTX 30905001000150020002500SE +/- 0.33, N = 3211921152112

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 3090NVIDIA RTX 3090140280420560700SE +/- 1.42, N = 3658.51651.17

152 Results Shown

vkpeak:
  fp32-scalar
  fp32-vec4
  fp16-scalar
  fp16-vec4
  fp64-scalar
  int32-scalar
  int32-vec4
  int16-scalar
  int16-vec4
RealSR-NCNN:
  4x - No
  4x - Yes
Waifu2x-NCNN Vulkan
VkFFT
Hashcat:
  MD5
  SHA1
  7-Zip
  SHA-512
  TrueCrypt RIPEMD160 + XTS
SHOC Scalable HeterOgeneous Computing:
  OpenCL - S3D
  OpenCL - Triad
  OpenCL - FFT SP
  OpenCL - MD5 Hash
  OpenCL - Reduction
  OpenCL - GEMM SGEMM_N
  OpenCL - Max SP Flops
  OpenCL - Bus Speed Download
  OpenCL - Bus Speed Readback
  OpenCL - Texture Read Bandwidth
cl-mem:
  Copy
  Read
  Write
NAMD CUDA
VkResample:
  2x - Double
  2x - Single
ET: Legacy:
  1920 x 1080
  1920 x 1200
  2560 x 1440
  3840 x 2160
Unvanquished:
  1920 x 1080 - High
  1920 x 1200 - High
  2560 x 1440 - High
  3840 x 2160 - High
  1920 x 1080 - Ultra
  1920 x 1200 - Ultra
  2560 x 1440 - Ultra
  3840 x 2160 - Ultra
  1920 x 1080 - Medium
  1920 x 1200 - Medium
  2560 x 1440 - Medium
  3840 x 2160 - Medium
Warsow:
  1920 x 1080
  1920 x 1200
  2560 x 1440
  3840 x 2160
Xonotic:
  3840 x 2160 - Low
  3840 x 2160 - High
  3840 x 2160 - Ultra
  3840 x 2160 - Ultimate
OctaneBench
ParaView:
  Many Spheres - 1920 x 1080:
    Frames / Sec
    MiPolys / Sec
  Many Spheres - 1920 x 1200:
    Frames / Sec
    MiPolys / Sec
  Many Spheres - 2560 x 1440:
    Frames / Sec
    MiPolys / Sec
  Many Spheres - 3840 x 2160:
    Frames / Sec
    MiPolys / Sec
  Wavelet Volume - 1920 x 1080:
    Frames / Sec
    MiVoxels / Sec
  Wavelet Volume - 1920 x 1200:
    Frames / Sec
    MiVoxels / Sec
  Wavelet Volume - 2560 x 1440:
    Frames / Sec
    MiVoxels / Sec
  Wavelet Volume - 3840 x 2160:
    Frames / Sec
    MiVoxels / Sec
  Wavelet Contour - 1920 x 1080:
    Frames / Sec
    MiPolys / Sec
  Wavelet Contour - 1920 x 1200:
    Frames / Sec
    MiPolys / Sec
  Wavelet Contour - 2560 x 1440:
    Frames / Sec
    MiPolys / Sec
  Wavelet Contour - 3840 x 2160:
    Frames / Sec
    MiPolys / Sec
FAHBench
LeelaChessZero
Rodinia
ArrayFire
LuxCoreRender:
  DLSC - GPU
  Danish Mood - GPU
  Orange Juice - GPU
  LuxCore Benchmark - GPU
  Rainbow Colors and Prism - GPU
FinanceBench
ViennaCL:
  CPU BLAS - sCOPY
  CPU BLAS - sAXPY
  CPU BLAS - sDOT
  CPU BLAS - dCOPY
  CPU BLAS - dAXPY
  CPU BLAS - dDOT
  CPU BLAS - dGEMV-N
  CPU BLAS - dGEMV-T
  CPU BLAS - dGEMM-NN
  CPU BLAS - dGEMM-NT
  CPU BLAS - dGEMM-TN
  CPU BLAS - dGEMM-TT
  OpenCL BLAS - sCOPY
  OpenCL BLAS - sAXPY
  OpenCL BLAS - sDOT
  OpenCL BLAS - dCOPY
  OpenCL BLAS - dAXPY
  OpenCL BLAS - dDOT
  OpenCL BLAS - dGEMV-N
  OpenCL BLAS - dGEMV-T
  OpenCL BLAS - dGEMM-NN
  OpenCL BLAS - dGEMM-NT
  OpenCL BLAS - dGEMM-TN
  OpenCL BLAS - dGEMM-TT
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
Blender:
  BMW27 - CUDA
  Classroom - CUDA
  Fishy Cat - CUDA
  Barbershop - CUDA
  BMW27 - NVIDIA OptiX
  Classroom - NVIDIA OptiX
  Fishy Cat - NVIDIA OptiX
  Barbershop - NVIDIA OptiX
  Pabellon Barcelona - CUDA
  Pabellon Barcelona - NVIDIA OptiX
IndigoBench:
  OpenCL GPU - Bedroom
  OpenCL GPU - Supercar
MandelGPU
clpeak:
  Integer Compute INT
  Single-Precision Float
  Double-Precision Double
  Global Memory Bandwidth
Chaos Group V-RAY:
  NVIDIA RTX GPU
  NVIDIA CUDA GPU
vkpeak