rtx3090

AMD Ryzen 9 7950X 16-Core testing with a ASUS ROG CROSSHAIR X670E HERO (0805 BIOS) and MSI NVIDIA GeForce RTX 4090 24GB on Linuxmint 21.1 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 2301238-NE-2301232NE06
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 3 Tests
C++ Boost Tests 2 Tests
CPU Massive 4 Tests
Creator Workloads 5 Tests
Game Development 2 Tests
HPC - High Performance Computing 8 Tests
Machine Learning 5 Tests
Multi-Core 7 Tests
NVIDIA GPU Compute 30 Tests
OpenCL 6 Tests
OpenMPI Tests 2 Tests
Python Tests 3 Tests
Renderers 3 Tests
Scientific Computing 2 Tests
Server CPU Tests 2 Tests
Vulkan Compute 8 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
rtx3090-gpu
January 18 2023
  9 Hours, 41 Minutes
rtx4090
January 23 2023
  11 Hours, 18 Minutes
Invert Hiding All Results Option
  10 Hours, 30 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


rtx3090ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen Resolutionrtx3090-gpurtx4090AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR X670E HERO (0805 BIOS)AMD Device 14d862GB2000GB Sabrent Rocket 4.0 Plus + 2000GB CT2000P5PSSD8 + 2000GB Samsung SSD 980 PRO 2TBeVGA NVIDIA GeForce RTX 3090 24GBNVIDIA GA102 HD AudioPHL 328E1 + LG ULTRAWIDEIntel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411Linuxmint 21.15.15.0-58-generic (x86_64)Cinnamon 5.6.7X Server 1.21.1.3NVIDIA 525.60.114.6.0OpenCL 3.0 CUDA 12.0.891.3.224GCC 11.3.0ext48960x2160MSI NVIDIA GeForce RTX 4090 24GBNVIDIA Device 22baNVIDIA 525.78.01OpenCL 3.0 CUDA 12.0.133OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --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-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/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 Processor Details- Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa601203Graphics Details- rtx3090-gpu: BAR1 / Visible vRAM Size: 256 MiB - vBIOS Version: 94.02.42.c0.15- rtx4090: BAR1 / Visible vRAM Size: 256 MiB - vBIOS Version: 95.02.18.80.74OpenCL Details- rtx3090-gpu: GPU Compute Cores: 10496- rtx4090: GPU Compute Cores: 16384Python Details- Python 3.10.8Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

rtx3090-gpu vs. rtx4090 ComparisonPhoronix Test SuiteBaseline+6396.3%+6396.3%+12792.6%+12792.6%+19188.9%+19188.9%353%320.2%231.1%25585.2%163.9%154%144.5%141.7%139.4%138.3%137.4%134.6%134.2%131.3%130.6%124.6%122.9%122.8%122.4%122.4%122.2%122.2%122.2%122.1%122.1%121.8%121.3%121.1%119.3%103.7%100.8%100.6%99.8%99.4%99.2%99.1%98.9%98%94.1%92.2%89.6%84.7%80.8%80.2%73.9%71.1%71%57.7%57.1%52.6%48.5%42.6%41.5%40.9%35.1%34.7%32.4%31.4%31%30.9%30.9%29.9%21.8%21.4%20.7%20%19.8%19%16%15%14.7%11.9%9.9%9.6%9.1%8.3%7.3%7.2%7.1%6.8%6.6%4.9%2.1%Vulkan GPU - googlenetVulkan GPU - squeezenet_ssdOpenCL - GEMM SGEMM_NGPUOpenCL - Reduction7-Zipint16-vec4T.R.XS.P.FSHA-512I.C.IOpenCL - Max SP FlopsSHA1OpenCL BLAS - dGEMM-TTMD5BMW27 - NVIDIA OptiX126%int16-scalarfp64-vec4fp64-scalarint32-scalarfp32-vec4int32-vec4OpenCL BLAS - dGEMM-TNfp16-scalarOpenCL - MD5 Hashfp16-vec4D.P.Dfp32-scalar2x - DoubleOpenCL BLAS - dGEMM-NTVulkan GPU - mobilenet116.8%B.S.OVulkan GPU - mnasnet103.3%OpenCL - Bus Speed DownloadOpenCL - TriadOpenCL - Bus Speed ReadbackPabellon Barcelona - NVIDIA OptiXOpenCL BLAS - dGEMM-NNClassroom - NVIDIA OptiXTotal ScoreDanish Mood - GPUDLSC - GPUFishy Cat - NVIDIA OptiXVulkan GPU-v3-v3 - mobilenet-v391.5%C.G.OVulkan GPU - shufflenet-v288.9%OpenCLVulkan GPU - alexnet82.9%OpenCL GPU - BedroomO.P.FBarbershop - NVIDIA OptiXOrange Juice - GPULuxCore Benchmark - GPUVulkan GPU - vision_transformer70.5%Vulkan GPU - FastestDet67.6%OpenCL GPU - Supercar4x - YesOpenCL - S3DVulkan GPU-v2-v2 - mobilenet-v248.6%Vulkan GPU - resnet5046.5%Vulkan GPU - regnety_400mVulkan GPU - efficientnet-b0OpenCL - T.R.BVulkan GPU - yolov4-tinyGPUOpenCL - FFT SPOpenCL BLAS - sCOPY2x - 3 - Yes4x - NoR.C.a.P - GPUOpenCL BLAS - sAXPYOpenCL BLAS - sDOT2x - SingleOpenCL BLAS - dGEMV-TATPase Simulation - 327,506 AtomsOpenCL BLAS - dGEMV-NVulkan GPU - resnet18CopyCPU BLAS - sDOTOpenCL BLAS - dDOTCPU BLAS - sAXPYOpenCL BLAS - dCOPYCPU BLAS - dGEMV-TVulkan GPU - vgg168.7%WriteReadOpenCL BLAS - dAXPYVulkan GPU - blazefaceG.M.BCPU BLAS - sCOPYCPU BLAS - dGEMV-NCPU BLAS - dDOTNCNNNCNNSHOC Scalable HeterOgeneous ComputingMandelGPUSHOC Scalable HeterOgeneous ComputingHashcatvkpeakHashcatclpeakHashcatclpeakSHOC Scalable HeterOgeneous ComputingHashcatViennaCLHashcatBlendervkpeakvkpeakvkpeakvkpeakvkpeakvkpeakViennaCLvkpeakSHOC Scalable HeterOgeneous ComputingvkpeakclpeakvkpeakVkResampleViennaCLNCNNFinanceBenchNCNNSHOC Scalable HeterOgeneous ComputingSHOC Scalable HeterOgeneous ComputingSHOC Scalable HeterOgeneous ComputingBlenderViennaCLBlenderOctaneBenchLuxCoreRenderLuxCoreRenderBlenderNCNNArrayFireNCNNLeelaChessZeroNCNNIndigoBenchRodiniaBlenderLuxCoreRenderLuxCoreRenderNCNNNCNNIndigoBenchRealSR-NCNNSHOC Scalable HeterOgeneous ComputingNCNNVkFFTNCNNNCNNNCNNSHOC Scalable HeterOgeneous ComputingNCNNFAHBenchNeatBenchSHOC Scalable HeterOgeneous ComputingViennaCLWaifu2x-NCNN VulkanRealSR-NCNNLuxCoreRenderViennaCLViennaCLVkResampleViennaCLNAMD CUDAViennaCLNCNNcl-memViennaCLViennaCLViennaCLViennaCLViennaCLNCNNcl-memcl-memViennaCLNCNNclpeakViennaCLViennaCLViennaCLrtx3090-gpurtx4090

rtx3090lczero: OpenCLshoc: 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 Bandwidthncnn: 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_400mncnn: Vulkan GPU - vision_transformerncnn: Vulkan GPU - FastestDetrodinia: OpenCL Particle Filterarrayfire: Conjugate Gradient OpenCLblender: BMW27 - NVIDIA OptiXblender: Classroom - NVIDIA OptiXblender: Fishy Cat - NVIDIA OptiXblender: Barbershop - NVIDIA OptiXblender: Pabellon Barcelona - NVIDIA OptiXneatbench: GPUindigobench: OpenCL GPU - Bedroomindigobench: OpenCL GPU - Supercarluxcorerender: DLSC - GPUluxcorerender: Danish Mood - GPUluxcorerender: Orange Juice - GPUluxcorerender: LuxCore Benchmark - GPUluxcorerender: Rainbow Colors and Prism - GPUfahbench: hashcat: MD5hashcat: SHA1hashcat: 7-Ziphashcat: SHA-512hashcat: TrueCrypt RIPEMD160 + XTSnamd-cuda: ATPase Simulation - 327,506 Atomsoctanebench: Total Scorefinancebench: Black-Scholes OpenCLcl-mem: Copycl-mem: Readcl-mem: Writeclpeak: Integer Compute INTclpeak: Single-Precision Floatclpeak: Double-Precision Doubleclpeak: Global Memory Bandwidthmandelgpu: GPUviennacl: 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-TTrealsr-ncnn: 4x - Norealsr-ncnn: 4x - Yesvkfft: vkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp16-scalarvkpeak: fp16-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4vkpeak: int32-scalarvkpeak: int32-vec4vkpeak: int16-scalarvkpeak: int16-vec4vkresample: 2x - Doublevkresample: 2x - Singlewaifu2x-ncnn: 2x - 3 - Yesrtx3090-gpurtx409040156427.91113.10252141.0243.1315395.8548428.6839391.913.359413.19132199.143.221.791.531.531.235.350.9110.513.101.881.172.5612.7834.292.11158.722.223.7671.5985.7814.0010.1551.6115.43309018.64349.23811.289.1710.1811.1431.03328.48206818063333321815700000109273327359333337908000.09694686.7661355.853362.3827.1751.518146.3335529.52653.69814.223132109.615222319047.671.571.577.790.61111081171103685013796067246541893756016026036015.61329.6714274720868.1427457.9620781.0741153.13652.97652.8820754.8820661.0713692.2416740.59118.9899.3503.26374185653.05826.28682812.7695.79891044.7227907.792405.026.830226.35043098.676.982.662.932.892.503.780.852.323.371.622.143.759.468.161.48270.653.722.0910.842713.067.035.2829.687.74409033.70177.65721.9018.1617.4219.0540.30442.4502157233333333510939000002775433652040000019115000.080951366.295182.874416.7887.5813.743083.6085069.351449.77869.72804490079.616224521848.572.473.081.598.811110811611048261046066477673222545011971320134013904.28918.8866346546187.7761074.8346166.6491395.921455.071454.9746167.6645916.9930756.4040926.0553.8157.7462.492OpenBenchmarking.org

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./betsy: 3: ./betsy: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./betsy: 3: ./betsy: not found

Codec: ETC2 RGB - Quality: Highest

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./betsy: 3: ./betsy: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./betsy: 3: ./betsy: not found

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ImportError: cannot import name 'Iterable' from 'collections' (/home/james/miniconda3/lib/python3.10/collections/__init__.py)

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: OpenCLrtx4090rtx3090-gpu16K32K48K64K80KSE +/- 538.71, N = 3SE +/- 467.99, N = 374185401561. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: OpenCLrtx4090rtx3090-gpu13K26K39K52K65KMin: 73125 / Avg: 74185.33 / Max: 74881Min: 39311 / Avg: 40156.33 / Max: 409271. (CXX) g++ options: -flto -pthread

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./caffe: 3: ./tools/caffe: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7fc4409d178f google::LogMessageFatal::~LogMessageFatal()

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./caffe: 3: ./tools/caffe: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f20f505478f google::LogMessageFatal::~LogMessageFatal()

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./caffe: 3: ./tools/caffe: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f383c70a78f google::LogMessageFatal::~LogMessageFatal()

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./caffe: 3: ./tools/caffe: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f6231b4778f google::LogMessageFatal::~LogMessageFatal()

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./caffe: 3: ./tools/caffe: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f44a267a78f google::LogMessageFatal::~LogMessageFatal()

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./caffe: 3: ./tools/caffe: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f1f0cb1078f google::LogMessageFatal::~LogMessageFatal()

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: S3Drtx4090rtx3090-gpu140280420560700SE +/- 0.46, N = 3SE +/- 0.53, N = 3653.06427.911. (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: S3Drtx4090rtx3090-gpu120240360480600Min: 652.17 / Avg: 653.06 / Max: 653.7Min: 427.08 / Avg: 427.91 / Max: 428.911. (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: Triadrtx4090rtx3090-gpu612182430SE +/- 0.01, N = 3SE +/- 0.01, N = 326.2913.101. (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: Triadrtx4090rtx3090-gpu612182430Min: 26.27 / Avg: 26.29 / Max: 26.31Min: 13.1 / Avg: 13.1 / Max: 13.121. (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 SPrtx4090rtx3090-gpu6001200180024003000SE +/- 1.85, N = 3SE +/- 0.44, N = 32812.762141.021. (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 SPrtx4090rtx3090-gpu5001000150020002500Min: 2810.36 / Avg: 2812.76 / Max: 2816.4Min: 2140.15 / Avg: 2141.02 / Max: 2141.61. (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 Hashrtx4090rtx3090-gpu20406080100SE +/- 1.04, N = 4SE +/- 0.06, N = 395.8043.131. (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 Hashrtx4090rtx3090-gpu20406080100Min: 94.72 / Avg: 95.8 / Max: 98.92Min: 43.08 / Avg: 43.13 / Max: 43.241. (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: Reductionrtx4090rtx3090-gpu2004006008001000SE +/- 0.05, N = 3SE +/- 0.54, N = 31044.72395.851. (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: Reductionrtx4090rtx3090-gpu2004006008001000Min: 1044.63 / Avg: 1044.72 / Max: 1044.79Min: 395.04 / Avg: 395.85 / Max: 396.871. (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_Nrtx4090rtx3090-gpu6K12K18K24K30KSE +/- 311.60, N = 4SE +/- 58.26, N = 327907.708428.681. (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_Nrtx4090rtx3090-gpu5K10K15K20K25KMin: 27570.2 / Avg: 27907.65 / Max: 28841.2Min: 8316.33 / Avg: 8428.68 / Max: 8511.641. (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 Flopsrtx4090rtx3090-gpu20K40K60K80K100KSE +/- 131.92, N = 3SE +/- 204.68, N = 392405.039391.91. (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 Flopsrtx4090rtx3090-gpu16K32K48K64K80KMin: 92269.6 / Avg: 92405 / Max: 92668.8Min: 38992.4 / Avg: 39391.87 / Max: 39669.11. (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 Downloadrtx4090rtx3090-gpu612182430SE +/- 0.00, N = 3SE +/- 0.00, N = 326.8313.361. (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 Downloadrtx4090rtx3090-gpu612182430Min: 26.83 / Avg: 26.83 / Max: 26.83Min: 13.36 / Avg: 13.36 / Max: 13.361. (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 Readbackrtx4090rtx3090-gpu612182430SE +/- 0.00, N = 3SE +/- 0.00, N = 326.3513.191. (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: Bus Speed Readbackrtx4090rtx3090-gpu612182430Min: 26.35 / Avg: 26.35 / Max: 26.35Min: 13.19 / Avg: 13.19 / Max: 13.191. (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: Texture Read Bandwidthrtx4090rtx3090-gpu7001400210028003500SE +/- 1.63, N = 3SE +/- 2.10, N = 33098.672199.141. (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 Bandwidthrtx4090rtx3090-gpu5001000150020002500Min: 3095.43 / Avg: 3098.67 / Max: 3100.58Min: 2197.04 / Avg: 2199.14 / Max: 2203.341. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

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 20220729Target: Vulkan GPU - Model: mobilenetrtx4090rtx3090-gpu246810SE +/- 4.06, N = 9SE +/- 0.03, N = 66.983.22MIN: 2.61 / MAX: 47.34MIN: 2.81 / MAX: 31.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: mobilenetrtx4090rtx3090-gpu3691215Min: 2.73 / Avg: 6.98 / Max: 39.47Min: 3.15 / Avg: 3.22 / Max: 3.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2rtx4090rtx3090-gpu0.59851.1971.79552.3942.9925SE +/- 1.31, N = 9SE +/- 0.03, N = 62.661.79MIN: 1.16 / MAX: 20.75MIN: 1.11 / MAX: 14.961. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2rtx4090rtx3090-gpu246810Min: 1.22 / Avg: 2.66 / Max: 13.17Min: 1.67 / Avg: 1.79 / Max: 1.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3rtx4090rtx3090-gpu0.65931.31861.97792.63723.2965SE +/- 1.19, N = 9SE +/- 0.03, N = 62.931.53MIN: 1.25 / MAX: 41.37MIN: 1.39 / MAX: 15.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3rtx4090rtx3090-gpu246810Min: 1.49 / Avg: 2.93 / Max: 12.41Min: 1.45 / Avg: 1.53 / Max: 1.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: shufflenet-v2rtx4090rtx3090-gpu0.65031.30061.95092.60123.2515SE +/- 1.51, N = 9SE +/- 0.05, N = 62.891.53MIN: 1.17 / MAX: 21.26MIN: 1.34 / MAX: 14.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: shufflenet-v2rtx4090rtx3090-gpu246810Min: 1.31 / Avg: 2.89 / Max: 14.95Min: 1.41 / Avg: 1.53 / Max: 1.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: mnasnetrtx4090rtx3090-gpu0.56251.1251.68752.252.8125SE +/- 1.41, N = 9SE +/- 0.01, N = 62.501.23MIN: 0.98 / MAX: 22.9MIN: 1.13 / MAX: 7.821. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: mnasnetrtx4090rtx3090-gpu246810Min: 1.04 / Avg: 2.5 / Max: 13.77Min: 1.18 / Avg: 1.23 / Max: 1.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: efficientnet-b0rtx4090rtx3090-gpu1.20382.40763.61144.81526.019SE +/- 1.43, N = 9SE +/- 0.96, N = 63.785.35MIN: 1.76 / MAX: 23.24MIN: 2.22 / MAX: 29.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: efficientnet-b0rtx4090rtx3090-gpu246810Min: 2.08 / Avg: 3.78 / Max: 15.2Min: 2.75 / Avg: 5.35 / Max: 9.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: blazefacertx4090rtx3090-gpu0.20480.40960.61440.81921.024SE +/- 0.00, N = 9SE +/- 0.01, N = 60.850.91MIN: 0.78 / MAX: 6.72MIN: 0.83 / MAX: 8.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: blazefacertx4090rtx3090-gpu246810Min: 0.84 / Avg: 0.85 / Max: 0.88Min: 0.87 / Avg: 0.91 / Max: 0.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: googlenetrtx4090rtx3090-gpu3691215SE +/- 0.04, N = 9SE +/- 0.10, N = 62.3210.51MIN: 1.45 / MAX: 14.91MIN: 1.94 / MAX: 27.241. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: googlenetrtx4090rtx3090-gpu3691215Min: 2.09 / Avg: 2.32 / Max: 2.47Min: 10.16 / Avg: 10.51 / Max: 10.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: vgg16rtx4090rtx3090-gpu0.75831.51662.27493.03323.7915SE +/- 1.30, N = 9SE +/- 0.20, N = 63.373.10MIN: 1.47 / MAX: 26.06MIN: 1.59 / MAX: 25.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: vgg16rtx4090rtx3090-gpu246810Min: 1.85 / Avg: 3.37 / Max: 13.8Min: 2.49 / Avg: 3.1 / Max: 3.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: resnet18rtx4090rtx3090-gpu0.4230.8461.2691.6922.115SE +/- 0.04, N = 7SE +/- 0.11, N = 61.621.88MIN: 1.17 / MAX: 14.28MIN: 1.14 / MAX: 8.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: resnet18rtx4090rtx3090-gpu246810Min: 1.5 / Avg: 1.62 / Max: 1.78Min: 1.61 / Avg: 1.88 / Max: 2.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: alexnetrtx4090rtx3090-gpu0.48150.9631.44451.9262.4075SE +/- 0.17, N = 9SE +/- 0.01, N = 62.141.17MIN: 1.02 / MAX: 20MIN: 1.08 / MAX: 7.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: alexnetrtx4090rtx3090-gpu246810Min: 1.76 / Avg: 2.14 / Max: 3.33Min: 1.15 / Avg: 1.17 / Max: 1.211. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: resnet50rtx4090rtx3090-gpu0.84381.68762.53143.37524.219SE +/- 1.60, N = 9SE +/- 0.09, N = 63.752.56MIN: 1.46 / MAX: 25.78MIN: 1.73 / MAX: 14.191. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: resnet50rtx4090rtx3090-gpu246810Min: 1.89 / Avg: 3.75 / Max: 16.56Min: 2.32 / Avg: 2.56 / Max: 2.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: yolov4-tinyrtx4090rtx3090-gpu3691215SE +/- 3.93, N = 9SE +/- 3.53, N = 69.4612.78MIN: 4.95 / MAX: 48.71MIN: 5.02 / MAX: 48.891. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: yolov4-tinyrtx4090rtx3090-gpu48121620Min: 5.3 / Avg: 9.46 / Max: 40.92Min: 6.81 / Avg: 12.78 / Max: 29.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: squeezenet_ssdrtx4090rtx3090-gpu816243240SE +/- 4.27, N = 9SE +/- 1.16, N = 68.1634.29MIN: 2.99 / MAX: 53.35MIN: 7.52 / MAX: 75.031. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: squeezenet_ssdrtx4090rtx3090-gpu714212835Min: 3.66 / Avg: 8.16 / Max: 42.33Min: 29.48 / Avg: 34.29 / Max: 37.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: regnety_400mrtx4090rtx3090-gpu0.47480.94961.42441.89922.374SE +/- 0.02, N = 9SE +/- 0.17, N = 61.482.11MIN: 1.35 / MAX: 14.94MIN: 1.7 / MAX: 19.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: regnety_400mrtx4090rtx3090-gpu246810Min: 1.42 / Avg: 1.48 / Max: 1.6Min: 1.79 / Avg: 2.11 / Max: 2.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: vision_transformerrtx4090rtx3090-gpu60120180240300SE +/- 127.68, N = 9SE +/- 15.25, N = 6270.65158.72MIN: 99.36 / MAX: 1493.37MIN: 115.97 / MAX: 1243.691. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: vision_transformerrtx4090rtx3090-gpu50100150200250Min: 141.12 / Avg: 270.65 / Max: 1292.07Min: 142.17 / Avg: 158.72 / Max: 234.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: FastestDetrtx4090rtx3090-gpu0.8371.6742.5113.3484.185SE +/- 1.45, N = 9SE +/- 0.06, N = 63.722.22MIN: 1.65 / MAX: 21.61MIN: 1.54 / MAX: 13.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: FastestDetrtx4090rtx3090-gpu246810Min: 2.08 / Avg: 3.72 / Max: 15.3Min: 1.99 / Avg: 2.22 / Max: 2.451. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./gromacs: 5: /cuda-build/run-gromacs: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./gromacs: 5: /cuda-build/run-gromacs: not found

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 Filterrtx4090rtx3090-gpu0.84761.69522.54283.39044.238SE +/- 0.037, N = 62.0913.7671. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle Filterrtx4090rtx3090-gpu246810Min: 3.68 / Avg: 3.77 / Max: 3.941. (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 OpenCLrtx4090rtx3090-gpu0.35960.71921.07881.43841.798SE +/- 0.0009, N = 3SE +/- 0.0053, N = 30.84271.59801. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient OpenCLrtx4090rtx3090-gpu246810Min: 0.84 / Avg: 0.84 / Max: 0.84Min: 1.59 / Avg: 1.6 / Max: 1.611. (CXX) g++ options: -rdynamic

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: BMW27 - Compute: NVIDIA OptiXrtx4090rtx3090-gpu3691215SE +/- 9.70, N = 12SE +/- 0.01, N = 313.065.78
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: BMW27 - Compute: NVIDIA OptiXrtx4090rtx3090-gpu48121620Min: 3.34 / Avg: 13.06 / Max: 119.75Min: 5.76 / Avg: 5.78 / Max: 5.79

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Classroom - Compute: NVIDIA OptiXrtx4090rtx3090-gpu48121620SE +/- 0.02, N = 3SE +/- 0.02, N = 37.0314.00
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Classroom - Compute: NVIDIA OptiXrtx4090rtx3090-gpu48121620Min: 6.99 / Avg: 7.03 / Max: 7.05Min: 13.97 / Avg: 14 / Max: 14.05

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Fishy Cat - Compute: NVIDIA OptiXrtx4090rtx3090-gpu3691215SE +/- 0.06, N = 13SE +/- 0.01, N = 35.2810.15
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Fishy Cat - Compute: NVIDIA OptiXrtx4090rtx3090-gpu3691215Min: 5.2 / Avg: 5.28 / Max: 5.94Min: 10.13 / Avg: 10.15 / Max: 10.17

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Barbershop - Compute: NVIDIA OptiXrtx4090rtx3090-gpu1224364860SE +/- 0.06, N = 3SE +/- 0.13, N = 329.6851.61
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Barbershop - Compute: NVIDIA OptiXrtx4090rtx3090-gpu1020304050Min: 29.59 / Avg: 29.68 / Max: 29.79Min: 51.47 / Avg: 51.61 / Max: 51.87

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXrtx4090rtx3090-gpu48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 37.7415.43
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXrtx4090rtx3090-gpu48121620Min: 7.73 / Avg: 7.74 / Max: 7.75Min: 15.41 / Avg: 15.43 / Max: 15.44

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.

OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: GPUrtx4090rtx3090-gpu9001800270036004500SE +/- 0.00, N = 3SE +/- 0.00, N = 340903090
OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: GPUrtx4090rtx3090-gpu7001400210028003500Min: 4090 / Avg: 4090 / Max: 4090Min: 3090 / Avg: 3090 / Max: 3090

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: Bedroomrtx4090rtx3090-gpu816243240SE +/- 0.06, N = 3SE +/- 0.04, N = 333.7018.64
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: Bedroomrtx4090rtx3090-gpu714212835Min: 33.58 / Avg: 33.7 / Max: 33.77Min: 18.56 / Avg: 18.64 / Max: 18.72

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: Supercarrtx4090rtx3090-gpu20406080100SE +/- 0.02, N = 3SE +/- 0.15, N = 377.6649.24
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: Supercarrtx4090rtx3090-gpu1530456075Min: 77.62 / Avg: 77.66 / Max: 77.69Min: 48.94 / Avg: 49.24 / Max: 49.41

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 2.6Scene: DLSC - Acceleration: GPUrtx4090rtx3090-gpu510152025SE +/- 0.02, N = 3SE +/- 0.01, N = 321.9011.28MIN: 19.4 / MAX: 22.16MIN: 10.52 / MAX: 11.61
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: GPUrtx4090rtx3090-gpu510152025Min: 21.86 / Avg: 21.9 / Max: 21.93Min: 11.27 / Avg: 11.28 / Max: 11.29

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: GPUrtx4090rtx3090-gpu48121620SE +/- 0.03, N = 3SE +/- 0.02, N = 318.169.17MIN: 7.14 / MAX: 21.2MIN: 3.4 / MAX: 10.68
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: GPUrtx4090rtx3090-gpu510152025Min: 18.11 / Avg: 18.16 / Max: 18.2Min: 9.14 / Avg: 9.17 / Max: 9.2

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: GPUrtx4090rtx3090-gpu48121620SE +/- 0.13, N = 3SE +/- 0.02, N = 317.4210.18MIN: 13.8 / MAX: 24.29MIN: 8.51 / MAX: 13.53
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: GPUrtx4090rtx3090-gpu48121620Min: 17.16 / Avg: 17.42 / Max: 17.56Min: 10.14 / Avg: 10.18 / Max: 10.22

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: GPUrtx4090rtx3090-gpu510152025SE +/- 0.03, N = 3SE +/- 0.03, N = 319.0511.14MIN: 6.93 / MAX: 22.91MIN: 3.22 / MAX: 12.94
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: GPUrtx4090rtx3090-gpu510152025Min: 19 / Avg: 19.05 / Max: 19.1Min: 11.09 / Avg: 11.14 / Max: 11.18

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Rainbow Colors and Prism - Acceleration: GPUrtx4090rtx3090-gpu918273645SE +/- 0.26, N = 3SE +/- 0.14, N = 340.3031.03MIN: 36.9 / MAX: 46MIN: 29.8 / MAX: 34.64
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Rainbow Colors and Prism - Acceleration: GPUrtx4090rtx3090-gpu816243240Min: 40 / Avg: 40.3 / Max: 40.83Min: 30.84 / Avg: 31.03 / Max: 31.29

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 2.3.2rtx4090rtx3090-gpu100200300400500SE +/- 0.52, N = 3SE +/- 0.29, N = 3442.45328.48
OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2rtx4090rtx3090-gpu80160240320400Min: 441.85 / Avg: 442.45 / Max: 443.49Min: 327.96 / Avg: 328.48 / Max: 328.96

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: MD5rtx4090rtx3090-gpu30000M60000M90000M120000M150000MSE +/- 120185042.52, N = 3SE +/- 25348328.90, N = 315723333333368180633333
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: MD5rtx4090rtx3090-gpu30000M60000M90000M120000M150000MMin: 157000000000 / Avg: 157233333333.33 / Max: 157400000000Min: 68137100000 / Avg: 68180633333.33 / Max: 68224900000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA1rtx4090rtx3090-gpu11000M22000M33000M44000M55000MSE +/- 75227011.99, N = 3SE +/- 45720163.31, N = 35109390000021815700000
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA1rtx4090rtx3090-gpu9000M18000M27000M36000M45000MMin: 50986900000 / Avg: 51093900000 / Max: 51239000000Min: 21725700000 / Avg: 21815700000 / Max: 21874700000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: 7-Ziprtx4090rtx3090-gpu600K1200K1800K2400K3000KSE +/- 8294.64, N = 3SE +/- 14591.13, N = 327754331092733
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: 7-Ziprtx4090rtx3090-gpu500K1000K1500K2000K2500KMin: 2766400 / Avg: 2775433.33 / Max: 2792000Min: 1063700 / Avg: 1092733.33 / Max: 1109800

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA-512rtx4090rtx3090-gpu1400M2800M4200M5600M7000MSE +/- 2150193.79, N = 3SE +/- 3640665.38, N = 365204000002735933333
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA-512rtx4090rtx3090-gpu1100M2200M3300M4400M5500MMin: 6518200000 / Avg: 6520400000 / Max: 6524700000Min: 2731900000 / Avg: 2735933333.33 / Max: 2743200000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSrtx4090rtx3090-gpu400K800K1200K1600K2000KSE +/- 18569.78, N = 6SE +/- 8982.20, N = 31911500790800
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSrtx4090rtx3090-gpu300K600K900K1200K1500KMin: 1818700 / Avg: 1911500 / Max: 1931900Min: 779800 / Avg: 790800 / Max: 808600

Mixbench

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

Backend: OpenCL - Benchmark: Integer

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: NVIDIA CUDA - Benchmark: Integer

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: OpenCL - Benchmark: Double Precision

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: OpenCL - Benchmark: Single Precision

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: NVIDIA CUDA - Benchmark: Half Precision

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: NVIDIA CUDA - Benchmark: Double Precision

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: NVIDIA CUDA - Benchmark: Single Precision

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

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 Atomsrtx4090rtx3090-gpu0.02180.04360.06540.08720.109SE +/- 0.00092, N = 15SE +/- 0.00139, N = 150.080950.09694
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 Atomsrtx4090rtx3090-gpu12345Min: 0.08 / Avg: 0.08 / Max: 0.09Min: 0.09 / Avg: 0.1 / Max: 0.11

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 Scorertx4090rtx3090-gpu300600900120015001366.30686.77

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.

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./redshift: 3: /usr/redshift/bin/redshiftBenchmark: not found

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./redshift: 3: /usr/redshift/bin/redshiftBenchmark: not found

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 OpenCLrtx4090rtx3090-gpu1.31692.63383.95075.26766.5845SE +/- 0.029, N = 6SE +/- 0.004, N = 32.8745.8531. (CXX) g++ options: -O3 -march=native -fopenmp
OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCLrtx4090rtx3090-gpu246810Min: 2.84 / Avg: 2.87 / Max: 3.02Min: 5.85 / Avg: 5.85 / Max: 5.861. (CXX) g++ options: -O3 -march=native -fopenmp

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: Copyrtx4090rtx3090-gpu90180270360450SE +/- 0.47, N = 3SE +/- 0.35, N = 3416.7362.31. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Copyrtx4090rtx3090-gpu70140210280350Min: 415.8 / Avg: 416.7 / Max: 417.4Min: 361.6 / Avg: 362.3 / Max: 362.71. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Readrtx4090rtx3090-gpu2004006008001000SE +/- 0.15, N = 3SE +/- 0.59, N = 3887.5827.11. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Readrtx4090rtx3090-gpu160320480640800Min: 887.3 / Avg: 887.5 / Max: 887.8Min: 826 / Avg: 827.13 / Max: 8281. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Writertx4090rtx3090-gpu2004006008001000SE +/- 1.59, N = 3SE +/- 0.58, N = 3813.7751.51. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Writertx4090rtx3090-gpu140280420560700Min: 811 / Avg: 813.7 / Max: 816.5Min: 750.3 / Avg: 751.47 / Max: 752.11. (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 INTrtx4090rtx3090-gpu9K18K27K36K45KSE +/- 401.20, N = 3SE +/- 152.55, N = 343083.6018146.331. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTrtx4090rtx3090-gpu7K14K21K28K35KMin: 42401.98 / Avg: 43083.6 / Max: 43791.05Min: 17958.83 / Avg: 18146.33 / Max: 18448.511. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision Floatrtx4090rtx3090-gpu20K40K60K80K100KSE +/- 31.18, N = 3SE +/- 198.17, N = 385069.3535529.521. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision Floatrtx4090rtx3090-gpu15K30K45K60K75KMin: 85019.92 / Avg: 85069.35 / Max: 85126.99Min: 35159.62 / Avg: 35529.52 / Max: 35837.741. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision Doublertx4090rtx3090-gpu30060090012001500SE +/- 0.83, N = 3SE +/- 0.79, N = 31449.77653.691. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision Doublertx4090rtx3090-gpu30060090012001500Min: 1448.68 / Avg: 1449.77 / Max: 1451.39Min: 652.15 / Avg: 653.69 / Max: 654.741. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory Bandwidthrtx4090rtx3090-gpu2004006008001000SE +/- 0.40, N = 3SE +/- 0.03, N = 3869.72814.221. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory Bandwidthrtx4090rtx3090-gpu150300450600750Min: 868.92 / Avg: 869.72 / Max: 870.22Min: 814.16 / Avg: 814.22 / Max: 814.281. (CXX) g++ options: -O3 -rdynamic -lOpenCL

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: GPUrtx4090rtx3090-gpu200M400M600M800M1000MSE +/- 5072564.87, N = 3SE +/- 14.10, N = 3804490079.63132109.61. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPUrtx4090rtx3090-gpu140M280M420M560M700MMin: 796195969.3 / Avg: 804490079.6 / Max: 813696635.1Min: 3132082 / Avg: 3132109.6 / Max: 3132128.41. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

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 - sCOPYrtx4090rtx3090-gpu4080120160200SE +/- 1.86, N = 3SE +/- 1.01, N = 151621521. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPYrtx4090rtx3090-gpu306090120150Min: 160 / Avg: 162.33 / Max: 166Min: 143 / Avg: 151.67 / Max: 1561. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPYrtx4090rtx3090-gpu50100150200250SE +/- 5.13, N = 3SE +/- 2.26, N = 152452231. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPYrtx4090rtx3090-gpu4080120160200Min: 235 / Avg: 245 / Max: 252Min: 209 / Avg: 222.53 / Max: 2391. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOTrtx4090rtx3090-gpu50100150200250SE +/- 6.66, N = 3SE +/- 2.19, N = 152181901. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOTrtx4090rtx3090-gpu4080120160200Min: 207 / Avg: 218 / Max: 230Min: 173 / Avg: 190.33 / Max: 2041. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPYrtx4090rtx3090-gpu1122334455SE +/- 0.23, N = 3SE +/- 0.37, N = 1548.547.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPYrtx4090rtx3090-gpu1020304050Min: 48.1 / Avg: 48.47 / Max: 48.9Min: 44.8 / Avg: 47.55 / Max: 49.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPYrtx4090rtx3090-gpu1632486480SE +/- 1.57, N = 3SE +/- 0.55, N = 1572.471.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPYrtx4090rtx3090-gpu1428425670Min: 69.3 / Avg: 72.43 / Max: 74Min: 67.5 / Avg: 71.53 / Max: 741. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOTrtx4090rtx3090-gpu1632486480SE +/- 0.74, N = 3SE +/- 0.81, N = 1573.071.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOTrtx4090rtx3090-gpu1428425670Min: 71.5 / Avg: 72.97 / Max: 73.8Min: 64.6 / Avg: 71.48 / Max: 75.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-Nrtx4090rtx3090-gpu20406080100SE +/- 7.55, N = 3SE +/- 4.91, N = 1581.577.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-Nrtx4090rtx3090-gpu1632486480Min: 66.4 / Avg: 81.5 / Max: 89.1Min: 16 / Avg: 77.67 / Max: 87.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-Trtx4090rtx3090-gpu20406080100SE +/- 0.96, N = 3SE +/- 2.55, N = 1598.890.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-Trtx4090rtx3090-gpu20406080100Min: 96.9 / Avg: 98.8 / Max: 100Min: 58.9 / Avg: 90.61 / Max: 98.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NNrtx4090rtx3090-gpu20406080100SE +/- 0.88, N = 3SE +/- 0.48, N = 151111111. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NNrtx4090rtx3090-gpu20406080100Min: 110 / Avg: 111.33 / Max: 113Min: 108 / Avg: 110.87 / Max: 1141. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NTrtx4090rtx3090-gpu20406080100SE +/- 0.33, N = 3SE +/- 0.34, N = 151081081. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NTrtx4090rtx3090-gpu20406080100Min: 107 / Avg: 107.67 / Max: 108Min: 106 / Avg: 107.73 / Max: 1101. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TNrtx4090rtx3090-gpu306090120150SE +/- 0.58, N = 3SE +/- 0.59, N = 151161171. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TNrtx4090rtx3090-gpu20406080100Min: 115 / Avg: 116 / Max: 117Min: 114 / Avg: 116.87 / Max: 1211. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TTrtx4090rtx3090-gpu20406080100SE +/- 0.58, N = 3SE +/- 0.61, N = 101101101. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TTrtx4090rtx3090-gpu20406080100Min: 109 / Avg: 110 / Max: 111Min: 106 / Avg: 110 / Max: 1121. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPYrtx4090rtx3090-gpu100200300400500SE +/- 0.67, N = 3SE +/- 0.67, N = 34823681. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPYrtx4090rtx3090-gpu90180270360450Min: 481 / Avg: 481.67 / Max: 483Min: 367 / Avg: 367.67 / Max: 3691. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPYrtx4090rtx3090-gpu130260390520650SE +/- 0.58, N = 3SE +/- 1.86, N = 36105011. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPYrtx4090rtx3090-gpu110220330440550Min: 609 / Avg: 610 / Max: 611Min: 497 / Avg: 500.67 / Max: 5031. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTrtx4090rtx3090-gpu100200300400500SE +/- 0.33, N = 3SE +/- 0.58, N = 34603791. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTrtx4090rtx3090-gpu80160240320400Min: 459 / Avg: 459.67 / Max: 460Min: 378 / Avg: 379 / Max: 3801. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPYrtx4090rtx3090-gpu140280420560700SE +/- 0.00, N = 3SE +/- 0.58, N = 36646061. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPYrtx4090rtx3090-gpu120240360480600Min: 664 / Avg: 664 / Max: 664Min: 605 / Avg: 606 / Max: 6071. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYrtx4090rtx3090-gpu2004006008001000SE +/- 0.33, N = 3SE +/- 0.33, N = 37767241. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYrtx4090rtx3090-gpu140280420560700Min: 775 / Avg: 775.67 / Max: 776Min: 723 / Avg: 723.67 / Max: 7241. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTrtx4090rtx3090-gpu160320480640800SE +/- 0.58, N = 3SE +/- 0.58, N = 37326541. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTrtx4090rtx3090-gpu130260390520650Min: 731 / Avg: 732 / Max: 733Min: 653 / Avg: 654 / Max: 6551. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-Nrtx4090rtx3090-gpu50100150200250SE +/- 0.33, N = 3SE +/- 0.33, N = 32251891. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-Nrtx4090rtx3090-gpu4080120160200Min: 224 / Avg: 224.67 / Max: 225Min: 188 / Avg: 188.67 / Max: 1891. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-Trtx4090rtx3090-gpu100200300400500SE +/- 0.33, N = 3SE +/- 0.00, N = 24503751. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-Trtx4090rtx3090-gpu80160240320400Min: 449 / Avg: 449.67 / Max: 450Min: 375 / Avg: 375 / Max: 3751. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNrtx4090rtx3090-gpu30060090012001500SE +/- 3.33, N = 3SE +/- 1.86, N = 311976011. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNrtx4090rtx3090-gpu2004006008001000Min: 1190 / Avg: 1196.67 / Max: 1200Min: 599 / Avg: 601.33 / Max: 6051. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTrtx4090rtx3090-gpu30060090012001500SE +/- 0.00, N = 3SE +/- 4.41, N = 313206021. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTrtx4090rtx3090-gpu2004006008001000Min: 1320 / Avg: 1320 / Max: 1320Min: 594 / Avg: 602.33 / Max: 6091. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNrtx4090rtx3090-gpu30060090012001500SE +/- 0.00, N = 3SE +/- 1.53, N = 313406031. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNrtx4090rtx3090-gpu2004006008001000Min: 1340 / Avg: 1340 / Max: 1340Min: 601 / Avg: 603 / Max: 6061. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTrtx4090rtx3090-gpu30060090012001500SE +/- 0.00, N = 313906011. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTrtx4090rtx3090-gpu2004006008001000Min: 1390 / Avg: 1390 / Max: 13901. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

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.

rtx3090-gpu: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: fatal: Failed initializing vulkan device

rtx4090: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: fatal: Failed initializing vulkan device

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: Nortx4090rtx3090-gpu1.26292.52583.78875.05166.3145SE +/- 0.028, N = 3SE +/- 0.011, N = 34.2895.613
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: Nortx4090rtx3090-gpu246810Min: 4.24 / Avg: 4.29 / Max: 4.34Min: 5.59 / Avg: 5.61 / Max: 5.63

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: Yesrtx4090rtx3090-gpu714212835SE +/- 0.01, N = 3SE +/- 0.05, N = 318.8929.67
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: Yesrtx4090rtx3090-gpu714212835Min: 18.87 / Avg: 18.89 / Max: 18.9Min: 29.57 / Avg: 29.67 / Max: 29.73

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 1.1.1rtx4090rtx3090-gpu14K28K42K56K70KSE +/- 1362.49, N = 6SE +/- 394.43, N = 363465427471. (CXX) g++ options: -O3
OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1rtx4090rtx3090-gpu11K22K33K44K55KMin: 58633 / Avg: 63465 / Max: 67375Min: 42007 / Avg: 42746.67 / Max: 433541. (CXX) g++ options: -O3

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-scalarrtx4090rtx3090-gpu10K20K30K40K50KSE +/- 73.04, N = 3SE +/- 62.27, N = 346187.7720868.14
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarrtx4090rtx3090-gpu8K16K24K32K40KMin: 46044.22 / Avg: 46187.77 / Max: 46282.99Min: 20804.36 / Avg: 20868.14 / Max: 20992.66

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4rtx4090rtx3090-gpu13K26K39K52K65KSE +/- 47.86, N = 3SE +/- 65.98, N = 361074.8327457.96
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4rtx4090rtx3090-gpu11K22K33K44K55KMin: 60982.97 / Avg: 61074.83 / Max: 61144.05Min: 27387 / Avg: 27457.96 / Max: 27589.79

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarrtx4090rtx3090-gpu10K20K30K40K50KSE +/- 46.75, N = 3SE +/- 10.69, N = 346166.6420781.07
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarrtx4090rtx3090-gpu8K16K24K32K40KMin: 46073.23 / Avg: 46166.64 / Max: 46216.84Min: 20759.7 / Avg: 20781.07 / Max: 20791.8

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4rtx4090rtx3090-gpu20K40K60K80K100KSE +/- 128.78, N = 3SE +/- 12.43, N = 391395.9241153.13
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4rtx4090rtx3090-gpu16K32K48K64K80KMin: 91204.15 / Avg: 91395.92 / Max: 91640.71Min: 41128.27 / Avg: 41153.13 / Max: 41166.14

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarrtx4090rtx3090-gpu30060090012001500SE +/- 0.06, N = 3SE +/- 0.37, N = 31455.07652.97
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarrtx4090rtx3090-gpu30060090012001500Min: 1454.96 / Avg: 1455.07 / Max: 1455.14Min: 652.23 / Avg: 652.97 / Max: 653.35

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4rtx4090rtx3090-gpu30060090012001500SE +/- 0.10, N = 3SE +/- 0.37, N = 31454.97652.88
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4rtx4090rtx3090-gpu30060090012001500Min: 1454.82 / Avg: 1454.97 / Max: 1455.16Min: 652.15 / Avg: 652.88 / Max: 653.3

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarrtx4090rtx3090-gpu10K20K30K40K50KSE +/- 9.81, N = 3SE +/- 9.80, N = 346167.6620754.88
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarrtx4090rtx3090-gpu8K16K24K32K40KMin: 46150.4 / Avg: 46167.66 / Max: 46184.37Min: 20735.27 / Avg: 20754.88 / Max: 20764.86

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4rtx4090rtx3090-gpu10K20K30K40K50KSE +/- 42.39, N = 3SE +/- 6.94, N = 345916.9920661.07
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4rtx4090rtx3090-gpu8K16K24K32K40KMin: 45832.67 / Avg: 45916.99 / Max: 45966.8Min: 20647.19 / Avg: 20661.07 / Max: 20668.23

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarrtx4090rtx3090-gpu7K14K21K28K35KSE +/- 26.05, N = 3SE +/- 7.48, N = 330756.4013692.24
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarrtx4090rtx3090-gpu5K10K15K20K25KMin: 30704.44 / Avg: 30756.4 / Max: 30785.65Min: 13677.33 / Avg: 13692.24 / Max: 13700.82

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4rtx4090rtx3090-gpu9K18K27K36K45KSE +/- 22.76, N = 3SE +/- 2.73, N = 340926.0516740.59
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4rtx4090rtx3090-gpu7K14K21K28K35KMin: 40902.97 / Avg: 40926.05 / Max: 40971.57Min: 16736.96 / Avg: 16740.59 / Max: 16745.94

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: Doublertx4090rtx3090-gpu306090120150SE +/- 0.02, N = 3SE +/- 0.16, N = 353.82118.991. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Doublertx4090rtx3090-gpu20406080100Min: 53.77 / Avg: 53.82 / Max: 53.85Min: 118.73 / Avg: 118.99 / Max: 119.281. (CXX) g++ options: -O3

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Singlertx4090rtx3090-gpu3691215SE +/- 0.019, N = 3SE +/- 0.063, N = 37.7469.3501. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Singlertx4090rtx3090-gpu3691215Min: 7.72 / Avg: 7.75 / Max: 7.78Min: 9.28 / Avg: 9.35 / Max: 9.481. (CXX) g++ options: -O3

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

rtx3090-gpu: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

rtx4090: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: Yesrtx4090rtx3090-gpu0.73421.46842.20262.93683.671SE +/- 0.009, N = 3SE +/- 0.003, N = 32.4923.263
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: Yesrtx4090rtx3090-gpu246810Min: 2.48 / Avg: 2.49 / Max: 2.51Min: 3.26 / Avg: 3.26 / Max: 3.27

100 Results Shown

LeelaChessZero
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
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
  Vulkan GPU - vision_transformer
  Vulkan GPU - FastestDet
Rodinia
ArrayFire
Blender:
  BMW27 - NVIDIA OptiX
  Classroom - NVIDIA OptiX
  Fishy Cat - NVIDIA OptiX
  Barbershop - NVIDIA OptiX
  Pabellon Barcelona - NVIDIA OptiX
NeatBench
IndigoBench:
  OpenCL GPU - Bedroom
  OpenCL GPU - Supercar
LuxCoreRender:
  DLSC - GPU
  Danish Mood - GPU
  Orange Juice - GPU
  LuxCore Benchmark - GPU
  Rainbow Colors and Prism - GPU
FAHBench
Hashcat:
  MD5
  SHA1
  7-Zip
  SHA-512
  TrueCrypt RIPEMD160 + XTS
NAMD CUDA
OctaneBench
FinanceBench
cl-mem:
  Copy
  Read
  Write
clpeak:
  Integer Compute INT
  Single-Precision Float
  Double-Precision Double
  Global Memory Bandwidth
MandelGPU
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
RealSR-NCNN:
  4x - No
  4x - Yes
VkFFT
vkpeak:
  fp32-scalar
  fp32-vec4
  fp16-scalar
  fp16-vec4
  fp64-scalar
  fp64-vec4
  int32-scalar
  int32-vec4
  int16-scalar
  int16-vec4
VkResample:
  2x - Double
  2x - Single
Waifu2x-NCNN Vulkan