ngc rtx 3090

AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3402 BIOS) and NVIDIA GeForce RTX 3090 24GB on Ubuntu 20.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2104059-IB-NGCRTX30939
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

BLAS (Basic Linear Algebra Sub-Routine) Tests 2 Tests
CPU Massive 4 Tests
Creator Workloads 4 Tests
Game Development 2 Tests
HPC - High Performance Computing 5 Tests
Machine Learning 3 Tests
Multi-Core 5 Tests
NVIDIA GPU Compute 24 Tests
OpenCL 6 Tests
Renderers 3 Tests
Server CPU Tests 2 Tests
Vulkan Compute 6 Tests
Common Workstation Benchmarks 2 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
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
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
1
April 05 2021
  4 Hours, 56 Minutes
2
April 05 2021
  4 Hours, 47 Minutes
3
April 05 2021
  18 Minutes
Invert Hiding All Results Option
  3 Hours, 20 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):


ngc rtx 3090ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen Resolution123AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads)ASUS ROG CROSSHAIR VIII HERO (3402 BIOS)AMD Starship/Matisse16GB1000GB Sabrent Rocket 4.0 Plus + 2000GBNVIDIA GeForce RTX 3090 24GBNVIDIA Device 1aefASUS VP28URealtek RTL8125 2.5GbE + Intel I211Ubuntu 20.045.8.0-48-generic (x86_64)GNOME Shell 3.36.7X Server 1.20.9NVIDIA 460.674.6.0OpenCL 1.2 CUDA 11.2.1621.2.155GCC 9.3.0 + CUDA 11.2ext43840x2160OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 OpenCL Details- GPU Compute Cores: 10496Python Details- 1: Python 3.8.5Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

123Result OverviewPhoronix Test Suite100%101%102%103%104%MixbenchVkFFTHashcatRealSR-NCNNWaifu2x-NCNN VulkanSHOC Scalable HeterOgeneous Computing

ngc rtx 3090arrayfire: Conjugate Gradient OpenCLbetsy: ETC1 - Highestbetsy: ETC2 RGB - Highestblender: 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 OptiXv-ray: NVIDIA RTX GPUv-ray: NVIDIA CUDA GPUcl-mem: Copycl-mem: Readcl-mem: Writeclpeak: Integer Compute INTclpeak: Single-Precision Floatclpeak: Double-Precision Doubleclpeak: Global Memory Bandwidthfahbench: financebench: Black-Scholes OpenCLgromacs-gpu: Water Benchmarkhashcat: MD5hashcat: SHA1hashcat: 7-Ziphashcat: SHA-512hashcat: TrueCrypt RIPEMD160 + XTSindigobench: OpenCL GPU - Bedroomindigobench: OpenCL GPU - Supercarlczero: OpenCLluxcorerender-cl: DLSCluxcorerender-cl: Foodluxcorerender-cl: LuxCore Benchmarkluxcorerender-cl: Rainbow Colors and Prismmandelgpu: GPUmixbench: OpenCL - Integermixbench: OpenCL - Double Precisionmixbench: OpenCL - Single Precisionnamd-cuda: ATPase Simulation - 327,506 Atomsncnn: 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_400moctanebench: Total Scorerealsr-ncnn: 4x - Norealsr-ncnn: 4x - Yesredshift: rodinia: OpenCL Particle Filtershoc: 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 Bandwidthviennacl: 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-TTvkfft: vkresample: 2x - Doublevkresample: 2x - Singlewaifu2x-ncnn: 2x - 3 - Yes1231.4773.0404.05418.3551.4234.60373.509.7630.5220.43335.89120.4846.6826011963363.5825.3742.017922.3535173.57654.89810.12343.61826.2349.64366413233333225832333331149333285303333384696720.86552.9293914811.204.709.0826.12466253751.520685.77496.2934053.470.1288412.784.424.114.784.125.691.8813.1655.2413.9811.0624.7521.9214.9316.75680.7230235.86629.6231413.743428.91625.47642347.2144.1934390.9578202.7539924.926.310126.35852224.0162.691.613622.833.443.877.883.253.552.956.755.336650437560772265023737860360560143686122.7959.2823.4391.4913.0154.04918.3651.4234.61374.219.7730.3320.42335.87120.4946.7226101963363.0825.9740.817951.7335204.07657.03813.37344.72066.2659.67066432866667225857000001144667285126666784586720.87052.7203899611.174.689.1326.29468356717.819683.19467.5334079.350.1259912.924.454.184.774.045.571.8313.1156.1013.6511.0824.6021.9915.1916.74680.1860865.86729.5981413.729430.51225.46342342.0843.9588389.8998272.7739799.626.310626.38732226.4762.591.513822.533.343.278.183.153.252.956.555.136450337460772265123737660260460060543194122.8739.2843.44466617033333226658666671147133285736666784916720677.83459.2635867.335.84529.574429.95225.46982344.2644.0181391.2758229.33447503.440OpenBenchmarking.org

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 OpenCL120.33550.6711.00651.3421.6775SE +/- 0.003, N = 3SE +/- 0.003, N = 31.4771.4911. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient OpenCL12246810Min: 1.47 / Avg: 1.48 / Max: 1.48Min: 1.49 / Avg: 1.49 / Max: 1.491. (CXX) g++ options: -rdynamic

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: Highest210.6841.3682.0522.7363.42SE +/- 0.016, N = 3SE +/- 0.014, N = 33.0153.0401. (CXX) g++ options: -O3 -O2 -lpthread -ldl
OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC1 - Quality: Highest21246810Min: 3 / Avg: 3.02 / Max: 3.05Min: 3.02 / Avg: 3.04 / Max: 3.071. (CXX) g++ options: -O3 -O2 -lpthread -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC2 RGB - Quality: Highest210.91221.82442.73663.64884.561SE +/- 0.017, N = 3SE +/- 0.029, N = 34.0494.0541. (CXX) g++ options: -O3 -O2 -lpthread -ldl
OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC2 RGB - Quality: Highest21246810Min: 4.03 / Avg: 4.05 / Max: 4.08Min: 4 / Avg: 4.05 / Max: 4.11. (CXX) g++ options: -O3 -O2 -lpthread -ldl

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 OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CUDA12510152025SE +/- 0.03, N = 3SE +/- 0.02, N = 318.3518.36
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CUDA12510152025Min: 18.32 / Avg: 18.35 / Max: 18.41Min: 18.33 / Avg: 18.36 / Max: 18.39

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CUDA121224364860SE +/- 0.04, N = 3SE +/- 0.05, N = 351.4251.42
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CUDA121020304050Min: 51.33 / Avg: 51.42 / Max: 51.47Min: 51.33 / Avg: 51.42 / Max: 51.5

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CUDA12816243240SE +/- 0.01, N = 3SE +/- 0.01, N = 334.6034.61
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CUDA12714212835Min: 34.59 / Avg: 34.6 / Max: 34.62Min: 34.6 / Avg: 34.61 / Max: 34.62

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CUDA1280160240320400SE +/- 0.19, N = 3SE +/- 0.37, N = 3373.50374.21
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CUDA1270140210280350Min: 373.11 / Avg: 373.5 / Max: 373.69Min: 373.56 / Avg: 374.21 / Max: 374.85

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: NVIDIA OptiX123691215SE +/- 0.03, N = 3SE +/- 0.04, N = 39.769.77
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: NVIDIA OptiX123691215Min: 9.72 / Avg: 9.76 / Max: 9.83Min: 9.7 / Avg: 9.77 / Max: 9.85

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: NVIDIA OptiX21714212835SE +/- 0.15, N = 3SE +/- 0.08, N = 330.3330.52
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: NVIDIA OptiX21714212835Min: 30.07 / Avg: 30.33 / Max: 30.59Min: 30.36 / Avg: 30.52 / Max: 30.63

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: NVIDIA OptiX21510152025SE +/- 0.03, N = 3SE +/- 0.01, N = 320.4220.43
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: NVIDIA OptiX21510152025Min: 20.37 / Avg: 20.42 / Max: 20.46Min: 20.4 / Avg: 20.43 / Max: 20.44

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: NVIDIA OptiX2170140210280350SE +/- 0.22, N = 3SE +/- 0.72, N = 3335.87335.89
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: NVIDIA OptiX2160120180240300Min: 335.47 / Avg: 335.87 / Max: 336.22Min: 334.7 / Avg: 335.89 / Max: 337.18

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: CUDA12306090120150SE +/- 0.05, N = 3SE +/- 0.07, N = 3120.48120.49
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: CUDA1220406080100Min: 120.39 / Avg: 120.48 / Max: 120.54Min: 120.35 / Avg: 120.49 / Max: 120.56

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX121122334455SE +/- 0.12, N = 3SE +/- 0.08, N = 346.6846.72
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX121020304050Min: 46.43 / Avg: 46.68 / Max: 46.8Min: 46.57 / Avg: 46.72 / Max: 46.83

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 GPU216001200180024003000SE +/- 8.62, N = 3SE +/- 6.17, N = 326102601
OpenBenchmarking.orgvrays, More Is BetterChaos Group V-RAY 5Mode: NVIDIA RTX GPU215001000150020002500Min: 2599 / Avg: 2610 / Max: 2627Min: 2589 / Avg: 2600.67 / Max: 2610

OpenBenchmarking.orgvpaths, More Is BetterChaos Group V-RAY 5Mode: NVIDIA CUDA GPU2140080012001600200019631963

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: Copy1280160240320400SE +/- 0.20, N = 3SE +/- 0.17, N = 3363.5363.01. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Copy1260120180240300Min: 363.3 / Avg: 363.5 / Max: 363.9Min: 362.7 / Avg: 363 / Max: 363.31. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Read212004006008001000SE +/- 0.67, N = 3SE +/- 0.52, N = 3825.9825.31. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Read21150300450600750Min: 824.8 / Avg: 825.9 / Max: 827.1Min: 824.4 / Avg: 825.33 / Max: 826.21. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Write12160320480640800SE +/- 0.38, N = 3SE +/- 0.75, N = 3742.0740.81. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Write12130260390520650Min: 741.3 / Avg: 741.97 / Max: 742.6Min: 739.3 / Avg: 740.8 / Max: 741.61. (CC) gcc options: -O2 -flto -lOpenCL

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 INT214K8K12K16K20KSE +/- 168.52, N = 7SE +/- 138.19, N = 317951.7317922.351. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INT213K6K9K12K15KMin: 17520.5 / Avg: 17951.73 / Max: 18669.37Min: 17646.85 / Avg: 17922.35 / Max: 18079.321. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision Float218K16K24K32K40KSE +/- 21.78, N = 3SE +/- 51.98, N = 335204.0735173.571. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision Float216K12K18K24K30KMin: 35160.52 / Avg: 35204.07 / Max: 35226.3Min: 35069.61 / Avg: 35173.57 / Max: 35225.71. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision Double21140280420560700SE +/- 0.12, N = 3SE +/- 0.01, N = 3657.03654.891. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision Double21120240360480600Min: 656.86 / Avg: 657.03 / Max: 657.26Min: 654.87 / Avg: 654.89 / Max: 654.921. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory Bandwidth212004006008001000SE +/- 0.07, N = 3SE +/- 3.33, N = 3813.37810.121. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory Bandwidth21140280420560700Min: 813.22 / Avg: 813.37 / Max: 813.44Min: 803.46 / Avg: 810.12 / Max: 813.491. (CXX) g++ options: -O3 -rdynamic -lOpenCL

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 2.3.22170140210280350SE +/- 0.21, N = 3SE +/- 0.11, N = 3344.72343.62
OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.22160120180240300Min: 344.43 / Avg: 344.72 / Max: 345.13Min: 343.4 / Avg: 343.62 / Max: 343.76

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 OpenCL12246810SE +/- 0.020, N = 3SE +/- 0.003, N = 36.2346.2651. (CXX) g++ options: -O3 -march=native -fopenmp
OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCL123691215Min: 6.2 / Avg: 6.23 / Max: 6.26Min: 6.26 / Avg: 6.27 / Max: 6.271. (CXX) g++ options: -O3 -march=native -fopenmp

GROMACS

The CUDA version of the Gromacs molecular dynamics package. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water Benchmark213691215SE +/- 0.017, N = 3SE +/- 0.061, N = 39.6709.6431. (CXX) g++ options: -O3 -lpthread -ldl -lrt -lm
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water Benchmark213691215Min: 9.64 / Avg: 9.67 / Max: 9.7Min: 9.52 / Avg: 9.64 / Max: 9.721. (CXX) g++ options: -O3 -lpthread -ldl -lrt -lm

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: MD532114000M28000M42000M56000M70000MSE +/- 55041388.47, N = 3SE +/- 75053099.72, N = 3SE +/- 73641979.27, N = 3666170333336643286666766413233333
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: MD532112000M24000M36000M48000M60000MMin: 66524100000 / Avg: 66617033333.33 / Max: 66714600000Min: 66353800000 / Avg: 66432866666.67 / Max: 66582900000Min: 66319100000 / Avg: 66413233333.33 / Max: 66558400000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA13215000M10000M15000M20000M25000MSE +/- 19772062.22, N = 3SE +/- 18693403.47, N = 3SE +/- 20398229.77, N = 3226658666672258570000022583233333
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA13214000M8000M12000M16000M20000MMin: 22629500000 / Avg: 22665866666.67 / Max: 22697500000Min: 22559800000 / Avg: 22585700000 / Max: 22622000000Min: 22555800000 / Avg: 22583233333.33 / Max: 22623100000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: 7-Zip132200K400K600K800K1000KSE +/- 1071.34, N = 3SE +/- 3069.38, N = 3SE +/- 3483.45, N = 3114933311471331144667
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: 7-Zip132200K400K600K800K1000KMin: 1147400 / Avg: 1149333.33 / Max: 1151100Min: 1141400 / Avg: 1147133.33 / Max: 1151900Min: 1138300 / Avg: 1144666.67 / Max: 1150300

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA-512312600M1200M1800M2400M3000MSE +/- 2834509.56, N = 3SE +/- 3284475.27, N = 3SE +/- 2406472.75, N = 3285736666728530333332851266667
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA-512312500M1000M1500M2000M2500MMin: 2854000000 / Avg: 2857366666.67 / Max: 2863000000Min: 2849600000 / Avg: 2853033333.33 / Max: 2859600000Min: 2847600000 / Avg: 2851266666.67 / Max: 2855800000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: TrueCrypt RIPEMD160 + XTS312200K400K600K800K1000KSE +/- 883.80, N = 3SE +/- 1168.09, N = 3SE +/- 88.19, N = 3849167846967845867
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: TrueCrypt RIPEMD160 + XTS312150K300K450K600K750KMin: 848000 / Avg: 849166.67 / Max: 850900Min: 845700 / Avg: 846966.67 / Max: 849300Min: 845700 / Avg: 845866.67 / Max: 846000

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: Bedroom21510152025SE +/- 0.04, N = 3SE +/- 0.05, N = 320.8720.87
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: Bedroom21510152025Min: 20.8 / Avg: 20.87 / Max: 20.93Min: 20.81 / Avg: 20.87 / Max: 20.96

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: Supercar121224364860SE +/- 0.01, N = 3SE +/- 0.04, N = 352.9352.72
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: Supercar121122334455Min: 52.9 / Avg: 52.93 / Max: 52.94Min: 52.64 / Avg: 52.72 / Max: 52.78

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: OpenCL128K16K24K32K40KSE +/- 165.32, N = 3SE +/- 192.20, N = 339148389961. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: OpenCL127K14K21K28K35KMin: 38818 / Avg: 39148.33 / Max: 39326Min: 38612 / Avg: 38996 / Max: 392031. (CXX) g++ options: -flto -pthread

LuxCoreRender OpenCL

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on OpenCL accelerators/GPUs. The alternative luxcorerender test profile is for CPU execution due to a difference in tests, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: DLSC123691215SE +/- 0.02, N = 3SE +/- 0.01, N = 311.2011.17MIN: 10.85 / MAX: 11.44MIN: 10.84 / MAX: 11.33
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: DLSC123691215Min: 11.17 / Avg: 11.2 / Max: 11.23Min: 11.15 / Avg: 11.17 / Max: 11.19

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: Food121.05752.1153.17254.235.2875SE +/- 0.05, N = 3SE +/- 0.04, N = 34.704.68MIN: 0.33 / MAX: 5.9MIN: 0.3 / MAX: 5.88
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: Food12246810Min: 4.6 / Avg: 4.7 / Max: 4.78Min: 4.61 / Avg: 4.68 / Max: 4.75

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: LuxCore Benchmark213691215SE +/- 0.06, N = 3SE +/- 0.07, N = 39.139.08MIN: 0.23 / MAX: 10.53MIN: 0.23 / MAX: 10.56
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: LuxCore Benchmark213691215Min: 9.01 / Avg: 9.13 / Max: 9.2Min: 8.99 / Avg: 9.08 / Max: 9.21

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: Rainbow Colors and Prism21612182430SE +/- 0.03, N = 3SE +/- 0.08, N = 326.2926.12MIN: 23.57 / MAX: 27.71MIN: 23.66 / MAX: 27.71
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: Rainbow Colors and Prism21612182430Min: 26.25 / Avg: 26.29 / Max: 26.34Min: 25.97 / Avg: 26.12 / Max: 26.25

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: GPU21100M200M300M400M500MSE +/- 4569465.07, N = 5SE +/- 4478729.17, N = 3468356717.8466253751.51. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPU2180M160M240M320M400MMin: 458726994.9 / Avg: 468356717.76 / Max: 485499437.3Min: 459849231.4 / Avg: 466253751.53 / Max: 474879445.61. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

Mixbench

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

OpenBenchmarking.orgGIOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: Integer1324K8K12K16K20KSE +/- 8.12, N = 3SE +/- 13.56, N = 3SE +/- 404.51, N = 1220685.7720677.8319683.191. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2
OpenBenchmarking.orgGIOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: Integer1324K8K12K16K20KMin: 20676.49 / Avg: 20685.77 / Max: 20701.95Min: 20655.05 / Avg: 20677.83 / Max: 20701.95Min: 17455.85 / Avg: 19683.19 / Max: 20676.491. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: Double Precision123110220330440550SE +/- 9.28, N = 15SE +/- 0.00, N = 3SE +/- 5.63, N = 3496.29467.53459.261. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2
OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: Double Precision12390180270360450Min: 467.53 / Avg: 496.29 / Max: 548.25Min: 467.53 / Avg: 467.53 / Max: 467.54Min: 448.5 / Avg: 459.26 / Max: 467.531. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: Single Precision3218K16K24K32K40KSE +/- 623.98, N = 15SE +/- 17.98, N = 3SE +/- 22.01, N = 335867.3334079.3534053.471. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2
OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: Single Precision3216K12K18K24K30KMin: 34028.31 / Avg: 35867.33 / Max: 39184.11Min: 34043.38 / Avg: 34079.35 / Max: 34097.33Min: 34028.31 / Avg: 34053.47 / Max: 34097.331. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

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 Atoms210.0290.0580.0870.1160.145SE +/- 0.00057, N = 3SE +/- 0.00063, N = 30.125990.12884
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 Atoms2112345Min: 0.12 / Avg: 0.13 / Max: 0.13Min: 0.13 / Avg: 0.13 / Max: 0.13

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: mobilenet123691215SE +/- 0.14, N = 3SE +/- 0.09, N = 312.7812.92MIN: 11.91 / MAX: 28.97MIN: 12.1 / MAX: 29.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mobilenet1248121620Min: 12.52 / Avg: 12.78 / Max: 12.99Min: 12.74 / Avg: 12.92 / Max: 13.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2121.00132.00263.00394.00525.0065SE +/- 0.04, N = 3SE +/- 0.07, N = 34.424.45MIN: 4.08 / MAX: 17.92MIN: 4.12 / MAX: 6.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v212246810Min: 4.37 / Avg: 4.42 / Max: 4.5Min: 4.37 / Avg: 4.45 / Max: 4.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3120.94051.8812.82153.7624.7025SE +/- 0.05, N = 3SE +/- 0.04, N = 34.114.18MIN: 3.79 / MAX: 6MIN: 3.89 / MAX: 28.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v312246810Min: 4.04 / Avg: 4.11 / Max: 4.2Min: 4.11 / Avg: 4.18 / Max: 4.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v2211.07552.1513.22654.3025.3775SE +/- 0.04, N = 3SE +/- 0.08, N = 34.774.78MIN: 4.53 / MAX: 6.88MIN: 4.44 / MAX: 5.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v221246810Min: 4.73 / Avg: 4.77 / Max: 4.86Min: 4.65 / Avg: 4.78 / Max: 4.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnet210.9271.8542.7813.7084.635SE +/- 0.13, N = 3SE +/- 0.00, N = 34.044.12MIN: 3.67 / MAX: 25.57MIN: 3.76 / MAX: 27.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnet21246810Min: 3.87 / Avg: 4.04 / Max: 4.3Min: 4.12 / Avg: 4.12 / Max: 4.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b0211.28032.56063.84095.12126.4015SE +/- 0.05, N = 3SE +/- 0.06, N = 35.575.69MIN: 5.16 / MAX: 23.76MIN: 5.25 / MAX: 7.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b021246810Min: 5.47 / Avg: 5.57 / Max: 5.66Min: 5.58 / Avg: 5.69 / Max: 5.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazeface210.4230.8461.2691.6922.115SE +/- 0.03, N = 3SE +/- 0.07, N = 31.831.88MIN: 1.72 / MAX: 2.92MIN: 1.74 / MAX: 3.191. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazeface21246810Min: 1.79 / Avg: 1.83 / Max: 1.88Min: 1.81 / Avg: 1.88 / Max: 2.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenet213691215SE +/- 0.08, N = 3SE +/- 0.18, N = 313.1113.16MIN: 11.82 / MAX: 48.41MIN: 12.01 / MAX: 35.571. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenet2148121620Min: 12.99 / Avg: 13.11 / Max: 13.26Min: 12.86 / Avg: 13.16 / Max: 13.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg16121326395265SE +/- 0.23, N = 3SE +/- 0.20, N = 355.2456.10MIN: 51.44 / MAX: 87.52MIN: 51.69 / MAX: 136.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg16121122334455Min: 55.01 / Avg: 55.24 / Max: 55.69Min: 55.72 / Avg: 56.1 / Max: 56.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet182148121620SE +/- 0.16, N = 3SE +/- 0.05, N = 313.6513.98MIN: 12.7 / MAX: 30.04MIN: 12.86 / MAX: 46.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet182148121620Min: 13.34 / Avg: 13.65 / Max: 13.84Min: 13.92 / Avg: 13.98 / Max: 14.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnet123691215SE +/- 0.04, N = 3SE +/- 0.04, N = 311.0611.08MIN: 10.13 / MAX: 35.54MIN: 10.15 / MAX: 28.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnet123691215Min: 10.97 / Avg: 11.06 / Max: 11.11Min: 11.03 / Avg: 11.08 / Max: 11.161. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet5021612182430SE +/- 0.05, N = 3SE +/- 0.24, N = 324.6024.75MIN: 23.3 / MAX: 51.24MIN: 22.52 / MAX: 45.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet5021612182430Min: 24.5 / Avg: 24.6 / Max: 24.68Min: 24.28 / Avg: 24.75 / Max: 25.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tiny12510152025SE +/- 0.28, N = 3SE +/- 0.08, N = 321.9221.99MIN: 19.58 / MAX: 55.4MIN: 19.81 / MAX: 50.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tiny12510152025Min: 21.37 / Avg: 21.92 / Max: 22.3Min: 21.84 / Avg: 21.99 / Max: 22.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssd1248121620SE +/- 0.18, N = 3SE +/- 0.13, N = 314.9315.19MIN: 13.47 / MAX: 53.36MIN: 13.88 / MAX: 66.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssd1248121620Min: 14.57 / Avg: 14.93 / Max: 15.15Min: 14.96 / Avg: 15.19 / Max: 15.41. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: regnety_400m2148121620SE +/- 0.16, N = 3SE +/- 0.29, N = 316.7416.75MIN: 15.52 / MAX: 52.59MIN: 15.48 / MAX: 39.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: regnety_400m2148121620Min: 16.46 / Avg: 16.74 / Max: 17.02Min: 16.33 / Avg: 16.75 / Max: 17.321. (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 Score12150300450600750680.72680.19

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: No3121.32012.64023.96035.28046.6005SE +/- 0.079, N = 3SE +/- 0.079, N = 3SE +/- 0.071, N = 45.8455.8665.867
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: No312246810Min: 5.76 / Avg: 5.85 / Max: 6Min: 5.79 / Avg: 5.87 / Max: 6.02Min: 5.78 / Avg: 5.87 / Max: 6.08

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: Yes321714212835SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 329.5729.6029.62
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: Yes321714212835Min: 29.53 / Avg: 29.57 / Max: 29.65Min: 29.52 / Avg: 29.6 / Max: 29.7Min: 29.55 / Avg: 29.62 / Max: 29.69

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 3.012306090120150SE +/- 0.67, N = 3SE +/- 0.67, N = 3141141
OpenBenchmarking.orgSeconds, Fewer Is BetterRedShift Demo 3.012306090120150Min: 140 / Avg: 140.67 / Max: 142Min: 140 / Avg: 140.67 / Max: 142

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 Filter210.84221.68442.52663.36884.211SE +/- 0.012, N = 3SE +/- 0.027, N = 33.7293.7431. (CXX) g++ options: -m64 -lm -lcuda -lcudart -lcudadevrt -lcudart_static -lrt -lpthread -ldl
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle Filter21246810Min: 3.71 / Avg: 3.73 / Max: 3.75Min: 3.7 / Avg: 3.74 / Max: 3.81. (CXX) g++ options: -m64 -lm -lcuda -lcudart -lcudadevrt -lcudart_static -lrt -lpthread -ldl

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: S3D23190180270360450SE +/- 0.47, N = 3SE +/- 1.17, N = 3SE +/- 0.34, N = 3430.51429.95428.921. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3D23180160240320400Min: 429.95 / Avg: 430.51 / Max: 431.45Min: 427.72 / Avg: 429.95 / Max: 431.66Min: 428.23 / Avg: 428.92 / Max: 429.261. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Triad132612182430SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 325.4825.4725.461. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Triad132612182430Min: 25.47 / Avg: 25.48 / Max: 25.49Min: 25.46 / Avg: 25.47 / Max: 25.49Min: 25.45 / Avg: 25.46 / Max: 25.481. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: FFT SP1325001000150020002500SE +/- 1.05, N = 3SE +/- 2.35, N = 3SE +/- 1.31, N = 32347.212344.262342.081. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: FFT SP132400800120016002000Min: 2345.11 / Avg: 2347.21 / Max: 2348.26Min: 2341.04 / Avg: 2344.26 / Max: 2348.84Min: 2340.34 / Avg: 2342.08 / Max: 2344.641. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 Hash1321020304050SE +/- 0.12, N = 3SE +/- 0.18, N = 3SE +/- 0.00, N = 344.1944.0243.961. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 Hash132918273645Min: 43.96 / Avg: 44.19 / Max: 44.31Min: 43.74 / Avg: 44.02 / Max: 44.35Min: 43.96 / Avg: 43.96 / Max: 43.961. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Reduction31280160240320400SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.10, N = 3391.28390.96389.901. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Reduction31270140210280350Min: 391.16 / Avg: 391.28 / Max: 391.43Min: 390.84 / Avg: 390.96 / Max: 391.12Min: 389.71 / Avg: 389.9 / Max: 390.061. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_N2312K4K6K8K10KSE +/- 86.21, N = 4SE +/- 65.52, N = 3SE +/- 88.75, N = 48272.778229.338202.751. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_N23114002800420056007000Min: 8119.35 / Avg: 8272.77 / Max: 8519.7Min: 8101.42 / Avg: 8229.33 / Max: 8317.95Min: 8011.53 / Avg: 8202.75 / Max: 8421.261. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Max SP Flops129K18K27K36K45KSE +/- 222.52, N = 3SE +/- 353.13, N = 339924.939799.61. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Max SP Flops127K14K21K28K35KMin: 39646.4 / Avg: 39924.93 / Max: 40364.8Min: 39267.6 / Avg: 39799.63 / Max: 40467.91. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed Download21612182430SE +/- 0.03, N = 3SE +/- 0.03, N = 326.3126.311. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed Download21612182430Min: 26.25 / Avg: 26.31 / Max: 26.34Min: 26.26 / Avg: 26.31 / Max: 26.341. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed Readback21612182430SE +/- 0.01, N = 3SE +/- 0.02, N = 326.3926.361. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed Readback21612182430Min: 26.37 / Avg: 26.39 / Max: 26.4Min: 26.33 / Avg: 26.36 / Max: 26.381. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read Bandwidth215001000150020002500SE +/- 2.17, N = 3SE +/- 2.03, N = 32226.472224.011. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read Bandwidth21400800120016002000Min: 2223.86 / Avg: 2226.47 / Max: 2230.78Min: 2220.18 / Avg: 2224.01 / Max: 2227.081. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

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 - sCOPY121428425670SE +/- 0.49, N = 3SE +/- 0.61, N = 362.662.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPY121224364860Min: 61.7 / Avg: 62.6 / Max: 63.4Min: 61.4 / Avg: 62.53 / Max: 63.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPY1220406080100SE +/- 2.08, N = 3SE +/- 2.72, N = 391.691.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPY1220406080100Min: 88 / Avg: 91.6 / Max: 95.2Min: 86.1 / Avg: 91.53 / Max: 94.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOT21306090120150SE +/- 3.18, N = 3SE +/- 1.67, N = 31381361. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOT21306090120150Min: 132 / Avg: 138.33 / Max: 142Min: 133 / Avg: 136.33 / Max: 1381. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPY12510152025SE +/- 0.17, N = 3SE +/- 0.63, N = 322.822.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPY12510152025Min: 22.5 / Avg: 22.8 / Max: 23.1Min: 21.3 / Avg: 22.53 / Max: 23.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPY12816243240SE +/- 1.09, N = 3SE +/- 0.90, N = 333.433.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPY12714212835Min: 31.2 / Avg: 33.37 / Max: 34.7Min: 31.5 / Avg: 33.3 / Max: 34.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOT121020304050SE +/- 0.84, N = 3SE +/- 1.32, N = 343.843.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOT12918273645Min: 42.1 / Avg: 43.77 / Max: 44.8Min: 40.6 / Avg: 43.23 / Max: 44.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-N2120406080100SE +/- 0.26, N = 3SE +/- 0.26, N = 378.177.81. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-N211530456075Min: 77.6 / Avg: 78.07 / Max: 78.5Min: 77.4 / Avg: 77.8 / Max: 78.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-T1220406080100SE +/- 0.12, N = 3SE +/- 0.47, N = 383.283.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-T121632486480Min: 83 / Avg: 83.17 / Max: 83.4Min: 82.2 / Avg: 83.13 / Max: 83.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NN121224364860SE +/- 0.64, N = 3SE +/- 0.75, N = 353.553.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NN121122334455Min: 52.4 / Avg: 53.53 / Max: 54.6Min: 51.9 / Avg: 53.23 / Max: 54.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NT211224364860SE +/- 0.18, N = 3SE +/- 0.34, N = 352.952.91. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NT211122334455Min: 52.6 / Avg: 52.87 / Max: 53.2Min: 52.2 / Avg: 52.87 / Max: 53.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TN121326395265SE +/- 0.06, N = 3SE +/- 0.15, N = 356.756.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TN121122334455Min: 56.6 / Avg: 56.7 / Max: 56.8Min: 56.3 / Avg: 56.53 / Max: 56.81. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TT121224364860SE +/- 0.06, N = 3SE +/- 0.15, N = 355.355.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TT121122334455Min: 55.2 / Avg: 55.3 / Max: 55.4Min: 54.9 / Avg: 55.13 / Max: 55.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPY1280160240320400SE +/- 0.67, N = 33663641. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPY1270140210280350Min: 363 / Avg: 363.67 / Max: 3651. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPY121102203304405505045031. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOT1280160240320400SE +/- 0.88, N = 33753741. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOT1270140210280350Min: 374 / Avg: 375.33 / Max: 3771. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPY21130260390520650SE +/- 0.67, N = 3SE +/- 0.67, N = 36076071. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPY21110220330440550Min: 606 / Avg: 607.33 / Max: 608Min: 606 / Avg: 607.33 / Max: 6081. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPY21160320480640800SE +/- 0.33, N = 3SE +/- 0.58, N = 37227221. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPY21130260390520650Min: 721 / Avg: 721.67 / Max: 722Min: 721 / Avg: 722 / Max: 7231. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOT21140280420560700SE +/- 0.33, N = 36516501. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOT21110220330440550Min: 650 / Avg: 650.67 / Max: 6511. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-N2150100150200250SE +/- 0.67, N = 3SE +/- 0.67, N = 32372371. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-N214080120160200Min: 236 / Avg: 237.33 / Max: 238Min: 236 / Avg: 237.33 / Max: 2381. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-T1280160240320400SE +/- 0.67, N = 33783761. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-T1270140210280350Min: 375 / Avg: 376.33 / Max: 3771. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NN12130260390520650SE +/- 1.67, N = 3SE +/- 1.20, N = 36036021. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NN12110220330440550Min: 600 / Avg: 603.33 / Max: 605Min: 600 / Avg: 602.33 / Max: 6041. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NT12130260390520650SE +/- 1.53, N = 3SE +/- 1.67, N = 36056041. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NT12110220330440550Min: 602 / Avg: 605 / Max: 607Min: 601 / Avg: 604.33 / Max: 6061. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TN12130260390520650SE +/- 1.33, N = 3SE +/- 1.53, N = 36016001. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TN12110220330440550Min: 598 / Avg: 600.67 / Max: 602Min: 597 / Avg: 600 / Max: 6021. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TT2130260390520650SE +/- 0.50, N = 26051. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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 1.1.131210K20K30K40K50KSE +/- 562.76, N = 3SE +/- 463.57, N = 5SE +/- 334.75, N = 34475043686431941. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.13128K16K24K32K40KMin: 44109 / Avg: 44750.33 / Max: 45872Min: 41891 / Avg: 43686 / Max: 44579Min: 42574 / Avg: 43193.67 / Max: 437231. (CXX) g++ options: -O3 -pthread

VkResample

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

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Double12306090120150SE +/- 0.03, N = 3SE +/- 0.04, N = 3122.80122.871. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Double1220406080100Min: 122.75 / Avg: 122.8 / Max: 122.85Min: 122.8 / Avg: 122.87 / Max: 122.921. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Single123691215SE +/- 0.006, N = 3SE +/- 0.005, N = 39.2829.2841. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Single123691215Min: 9.27 / Avg: 9.28 / Max: 9.29Min: 9.27 / Avg: 9.28 / Max: 9.291. (CXX) g++ options: -O3 -pthread

Waifu2x-NCNN Vulkan

Waifu2x-NCNN is an NCNN neural network implementation of the Waifu2x converter project and accelerated using the Vulkan API. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: Yes1320.77491.54982.32473.09963.8745SE +/- 0.007, N = 3SE +/- 0.008, N = 3SE +/- 0.013, N = 33.4393.4403.444
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: Yes132246810Min: 3.43 / Avg: 3.44 / Max: 3.45Min: 3.42 / Avg: 3.44 / Max: 3.45Min: 3.42 / Avg: 3.44 / Max: 3.46

100 Results Shown

ArrayFire
Betsy GPU Compressor:
  ETC1 - Highest
  ETC2 RGB - Highest
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
Chaos Group V-RAY:
  NVIDIA RTX GPU
  NVIDIA CUDA GPU
cl-mem:
  Copy
  Read
  Write
clpeak:
  Integer Compute INT
  Single-Precision Float
  Double-Precision Double
  Global Memory Bandwidth
FAHBench
FinanceBench
GROMACS
Hashcat:
  MD5
  SHA1
  7-Zip
  SHA-512
  TrueCrypt RIPEMD160 + XTS
IndigoBench:
  OpenCL GPU - Bedroom
  OpenCL GPU - Supercar
LeelaChessZero
LuxCoreRender OpenCL:
  DLSC
  Food
  LuxCore Benchmark
  Rainbow Colors and Prism
MandelGPU
Mixbench:
  OpenCL - Integer
  OpenCL - Double Precision
  OpenCL - Single Precision
NAMD CUDA
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
OctaneBench
RealSR-NCNN:
  4x - No
  4x - Yes
RedShift Demo
Rodinia
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
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
VkFFT
VkResample:
  2x - Double
  2x - Single
Waifu2x-NCNN Vulkan