NVIDIA RTX 2000 / 4000 / 6000 Ada Benchmarks

Benchmarks for a future article by Michael Larabel of NVIDIA RTX 6000 Ada Generation and other cards against Radeon PRO W7000 series on Linux.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2409140-PTS-NVIDIART32
Jump To Table - Results

View

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

Limit displaying results to tests within:

CPU Massive 3 Tests
Creator Workloads 4 Tests
Multi-Core 5 Tests
NVIDIA GPU Compute 9 Tests
OpenCL 3 Tests
Renderers 4 Tests
Server CPU Tests 2 Tests
Common Workstation Benchmarks 3 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
RTX 2000 Ada Generation
September 13
  6 Hours, 12 Minutes
RTX 4000 Ada Generation
September 13
  5 Hours, 1 Minute
RTX 6000 Ada Generation
September 11
  4 Hours, 22 Minutes
Radeon PRO W7500
September 14
  9 Hours, 55 Minutes
Radeon PRO W7600
September 14
  6 Hours, 2 Minutes
Radeon PRO W7700
September 13
  5 Hours, 2 Minutes
Radeon PRO W7900
September 14
  5 Hours, 37 Minutes
W7900 rep
September 14
  1 Hour, 46 Minutes
Invert Hiding All Results Option
  5 Hours, 30 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


NVIDIA RTX 2000 / 4000 / 6000 Ada BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLCompilerFile-SystemScreen ResolutionRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 repAMD Ryzen 9 9950X 16-Core @ 8.18GHz (16 Cores / 32 Threads)ASUS ROG STRIX X670E-E GAMING WIFI (2308 BIOS)AMD Device 14d82 x 32GB DDR5-6400MT/s Corsair CMK64GX5M2B6400C32Western Digital WD_BLACK SN850X 2000GBNVIDIA RTX 2000 Ada Generation 16GBNVIDIA Device 22beDELL U2723QEIntel I225-V + Intel Wi-Fi 6EUbuntu 24.046.8.0-41-generic (x86_64)GNOME Shell 46.0X Server 1.21.1.11NVIDIA 560.35.034.6.0OpenCL 3.0 CUDA 12.6.65GCC 13.2.0ext43840x2160NVIDIA RTX 4000 Ada Generation 20GBNVIDIA Device 22bcNVIDIA RTX 6000 Ada Generation 48GBNVIDIA AD102 HD AudioAMD Radeon PRO W7500 8GBAMD Navi 31 HDMI/DPX Server 1.21.1.11 + Wayland4.6 Mesa 24.2.0-devel (LLVM 18.1.7 DRM 3.58)OpenCL 2.1 AMD-APP (3625.0)AMD Radeon PRO W7600 8GBAMD Radeon PRO W7700 15GBAMD Radeon PRO W7900 45GBOpenBenchmarking.orgKernel Details- RTX 2000 Ada Generation: nouveau.modeset=0 - Transparent Huge Pages: madvise- RTX 4000 Ada Generation: nouveau.modeset=0 - Transparent Huge Pages: madvise- RTX 6000 Ada Generation: nouveau.modeset=0 - Transparent Huge Pages: madvise- Radeon PRO W7500: Transparent Huge Pages: madvise- Radeon PRO W7600: Transparent Huge Pages: madvise- Radeon PRO W7700: Transparent Huge Pages: madvise- Radeon PRO W7900: Transparent Huge Pages: madvise- W7900 rep: Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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-epp performance (EPP: performance) - CPU Microcode: 0xb40401cGraphics Details- RTX 2000 Ada Generation: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.07.47.00.05- RTX 4000 Ada Generation: BAR1 / Visible vRAM Size: 32768 MiB - vBIOS Version: 95.04.5c.00.0d- RTX 6000 Ada Generation: BAR1 / Visible vRAM Size: 65536 MiB - vBIOS Version: 95.02.3a.00.01- Radeon PRO W7500: BAR1 / Visible vRAM Size: 8176 MB- Radeon PRO W7600: BAR1 / Visible vRAM Size: 8176 MB- Radeon PRO W7700: BAR1 / Visible vRAM Size: 15344 MB- Radeon PRO W7900: BAR1 / Visible vRAM Size: 46064 MB- W7900 rep: BAR1 / Visible vRAM Size: 46064 MBOpenCL Details- RTX 2000 Ada Generation: GPU Compute Cores: 2816- RTX 4000 Ada Generation: GPU Compute Cores: 6144- RTX 6000 Ada Generation: GPU Compute Cores: 18176Python Details- Python 3.12.3Security Details- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected

RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 repResult OverviewPhoronix Test Suite100%299%499%698%897%LuxCoreRendervkpeakclpeakProjectPhysX OpenCL-BenchmarkGpuOwlFluidX3DSPECViewPerf 2020IndigoBenchFinanceBenchParaView

RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 repPer Watt Result OverviewPhoronix Test Suite100%150%200%249%299%ParaViewLuxCoreRenderProjectPhysX OpenCL-BenchmarkFluidX3DclpeakGpuOwlSPECViewPerf 2020vkpeakIndigoBenchP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.M

NVIDIA RTX 2000 / 4000 / 6000 Ada Benchmarksrodinia: OpenCL Particle Filterv-ray: NVIDIA CUDA GPUv-ray: NVIDIA RTX GPUblender: BMW27 - NVIDIA CUDAblender: BMW27 - NVIDIA OptiXblender: Classroom - NVIDIA CUDAblender: Classroom - NVIDIA OptiXblender: Fishy Cat - NVIDIA CUDAblender: Fishy Cat - NVIDIA OptiXblender: Pabellon Barcelona - NVIDIA OptiXblender: Barbershop - NVIDIA CUDAblender: Barbershop - NVIDIA OptiXblender: Junkshop - NVIDIA CUDAblender: Junkshop - NVIDIA OptiXblender: BMW27 - Radeon HIPblender: Fishy Cat - Radeon HIPblender: Pabellon Barcelona - Radeon HIPblender: Barbershop - Radeon HIPblender: Junkshop - Radeon HIPgpuowl: 77936867blender: Classroom - Radeon HIPindigobench: OpenCL GPU - Supercaropencl-benchmark: Memory Bandwidth Coalesced Readindigobench: OpenCL GPU - Bedroomluxcorerender: DLSC - GPUopencl-benchmark: INT8 Computeluxcorerender: Rainbow Colors and Prism - GPUopencl-benchmark: INT16 Computeopencl-benchmark: FP32 Computeopencl-benchmark: INT64 Computeluxcorerender: LuxCore Benchmark - GPUopencl-benchmark: Memory Bandwidth Coalesced Writeluxcorerender: Orange Juice - GPUluxcorerender: Danish Mood - GPUfinancebench: Black-Scholes OpenCLfinancebench: Monte-Carlo OpenCLclpeak: Global Memory Bandwidthclpeak: Single-Precision Computeclpeak: Double-Precision Computeclpeak: Integer 24-bit Computevkpeak: fp32-scalargpuowl: 57885161fahbench: vkpeak: fp32-vec4vkpeak: fp16-scalaropencl-benchmark: FP64 Computevkpeak: fp16-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4vkpeak: int16-scalarvkpeak: int16-vec4vkpeak: int32-scalarvkpeak: int32-vec4fluidx3d: FP32-FP32fluidx3d: FP32-FP16Sopencl-benchmark: FP16 Computefluidx3d: FP32-FP16Cparaview: Many Spheres - 3000 - 2560 x 1440paraview: Many Spheres - 3000 - 3840 x 2160specviewperf2020: 2560 x 1440 - CREO-03specviewperf2020: 2560 x 1440 - MAYA-06specviewperf2020: 2560 x 1440 - ENERGY-03specviewperf2020: 2560 x 1440 - SNX-04specviewperf2020: 2560 x 1440 - MEDICAL-O3specviewperf2020: 2560 x 1440 - SOLIDWORKS-07opencl-benchmark: INT32 Computeparaview: Wavelet Contour - 3000 - 2560 x 1440paraview: Wavelet Contour - 3000 - 3840 x 2160paraview: Wavelet Volume - 3000 - 2560 x 1440paraview: Wavelet Volume - 3000 - 3840 x 2160gpuowl: 332220523specviewperf2020: 2560 x 1440 - CATIA-06blender: Pabellon Barcelona - NVIDIA CUDAparaview: Many Spheres - 3000 - 2560 x 1440paraview: Many Spheres - 3000 - 3840 x 2160paraview: Wavelet Contour - 3000 - 2560 x 1440paraview: Wavelet Contour - 3000 - 3840 x 2160paraview: Wavelet Volume - 3000 - 2560 x 1440paraview: Wavelet Volume - 3000 - 3840 x 2160RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep9.2061419202322.5310.4545.8728.5547.2622.7431.07188.76114.4538.4022.30226.6424.399207.718.1614.614.78711.925.94713.4391.7354.63211.815.214.0916.228143266.035143195.0011596.82217.615965.806944.94308.39207.85009186.816968.170.21513745.30218.82218.824634.076167.1513312434253147.2743.29113.82276.4431.73285.2265.60184.73194.01145.70285.73150.8347.8583.69108.954739.4664339.6572021.8211518.3564571.6352413.3354.9911960292713.267.1926.5617.8626.1312.9019.40107.8572.3622.9614.62449.5136.915326.8512.8787.969.69217.6211.48326.9612.8248.03333.787.706.927.693199.926860307.5425785.68439.3213209.0613990.27612.37281.646418498.6413985.470.43327650.70440.87440.019328.0112381.1120473770403996.5889.04170.61430.1062.74423.6697.35308.72346.77267.98466.70269.0395.30142.9559.209682.3218926.8063613.7112792.6547467.0904304.5142.091598481745.573.6110.447.9210.675.488.8645.6532.9111.087.341490.3173.905865.5729.28218.9521.79235.6532.29790.8053.77417.78850.0415.8617.722.71099.239287815.9183188.791534.7040310.5848174.332043.60462.313162508.5446258.431.5193562.471538.941533.6431027.6638953.7152521020810585167.61158.47256.46813.18156.171026.56211.98578.10811.19625.76853.68639.67322.42236.9621.3016804.13815887.6678453.5676521.15213658.90510234.76245.1479.12181.37348.2663.38226.9184.3918.503146.926.3701.922.7276.924.8556.4801.0472.19149.231.051.9012.201327.081569139.396072.16238.565462.754887.92315.994345.635005.480.2378701.87235.12229.924897.028451.358561630168255.0149.08113.67229.6514.53209.9216.02162.17218.56160.57291.53151.9749.0867.845514.7724920.5482277.7051673.2594664.4032431.53634.4058.45135.28258.0647.97345.1163.0125.080237.728.7402.543.8209.276.7318.9191.4633.01213.461.292.559.724261.737998220.397710.95332.836985.957784.73467.656876.757839.680.32415112.53332.94329.527740.8815006.1211772194231775.0768.77151.46316.2523.03263.6925.27216.52274.72212.95398.03229.3366.1887.877526.3726894.4582862.8772219.2026368.4913669.29422.4539.9490.92164.7331.36545.2640.4536.452403.7212.0334.096.11514.9810.69314.3532.3644.90308.932.314.166.779152.220712391.2112304.24462.2511408.6612385.23735.1110997.0712392.950.46925066.85484.51484.5312377.1425011.69163229112865109.7296.28208.57501.3836.67360.2333.26336.76400.30320.57564.16340.94102.10137.5610999.5409652.0824171.5743340.7259026.5705455.08113.5221.4247.2191.4418.671088.5322.2854.969590.2419.3486.2812.56620.6922.45931.0314.2267.69459.222.936.623.382100.719712496.4225936.651096.5922802.7425619.411460.5822828.3725838.081.05749210.091118.441117.4725473.8948689.68279646505503168.48147.76276.10770.3567.46582.3053.71530.89734.05582.89653.03589.04206.30203.2216890.75614813.5257649.7396074.42610448.3919424.65113.3321.4147.4991.9718.761090.5122.0355.228595.4419.3246.4812.46320.8222.25430.3494.2877.88454.442.936.623.3697.539001493.7426088.31099.2122557.9325643.971470.5922365.7525688.061.07948905.971117.741117.3625460.7648657.966492.056554.462878477856.8165612167.57147.89786.1468.26581.8353.95532.836.013730.54580.31644.4549.68206.44203.9316799.29514827.0957613.0936047.55210310.3828794.923OpenBenchmarking.org

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 FilterRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation3691215SE +/- 0.009, N = 5SE +/- 0.005, N = 7SE +/- 0.004, N = 119.2064.9912.0911. (CXX) g++ options: -O2 -lOpenCL

Test: OpenCL Particle Filter

Radeon PRO W7500: The test run did not produce a result. E: ERROR: clEnqueueNDRangeKernel(kernel_likelihood)=>-54 failed

Radeon PRO W7600: The test run did not produce a result. E: ERROR: clEnqueueNDRangeKernel(kernel_likelihood)=>-54 failed

Radeon PRO W7700: The test run did not produce a result. E: ERROR: clEnqueueNDRangeKernel(kernel_likelihood)=>-54 failed

Radeon PRO W7900: The test run did not produce a result. E: ERROR: clEnqueueNDRangeKernel(kernel_likelihood)=>-54 failed

W7900 rep: The test run did not produce a result. E: ERROR: clEnqueueNDRangeKernel(kernel_likelihood)=>-54 failed

Chaos Group V-RAY

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

OpenBenchmarking.orgvpaths, More Is BetterChaos Group V-RAY 6.0Mode: NVIDIA CUDA GPURTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation13002600390052006500SE +/- 3.67, N = 3SE +/- 0.00, N = 3SE +/- 24.01, N = 3141919605984

OpenBenchmarking.orgvpaths, More Is BetterChaos Group V-RAY 6.0Mode: NVIDIA RTX GPURTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation2K4K6K8K10KSE +/- 0.00, N = 3SE +/- 13.86, N = 3SE +/- 8.95, N = 3202329278174

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: BMW27 - Compute: NVIDIA CUDARTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation510152025SE +/- 0.02, N = 3SE +/- 0.02, N = 4SE +/- 0.01, N = 722.5313.265.57

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: BMW27 - Compute: NVIDIA OptiXRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation3691215SE +/- 0.01, N = 5SE +/- 0.01, N = 6SE +/- 0.00, N = 810.457.193.61

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Classroom - Compute: NVIDIA CUDARTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation1020304050SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 545.8726.5610.44

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Classroom - Compute: NVIDIA OptiXRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation714212835SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 628.5517.867.92

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Fishy Cat - Compute: NVIDIA CUDARTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation1122334455SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 547.2626.1310.67

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Fishy Cat - Compute: NVIDIA OptiXRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation510152025SE +/- 0.01, N = 3SE +/- 0.01, N = 4SE +/- 0.05, N = 1522.7412.905.48

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation714212835SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 531.0719.408.86

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Barbershop - Compute: NVIDIA CUDARTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation4080120160200SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3188.76107.8545.65

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Barbershop - Compute: NVIDIA OptiXRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation306090120150SE +/- 0.06, N = 3SE +/- 0.15, N = 3SE +/- 0.05, N = 3114.4572.3632.91

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Junkshop - Compute: NVIDIA CUDARTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation918273645SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 438.4022.9611.08

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Junkshop - Compute: NVIDIA OptiXRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation510152025SE +/- 0.04, N = 3SE +/- 0.02, N = 4SE +/- 0.01, N = 622.3014.627.34

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: BMW27 - Compute: Radeon HIPRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep1020304050SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 445.1434.4022.4513.5213.33

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Fishy Cat - Compute: Radeon HIPRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep20406080100SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.04, N = 379.1258.4539.9421.4221.41

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Pabellon Barcelona - Compute: Radeon HIPRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep4080120160200SE +/- 0.25, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.10, N = 3181.37135.2890.9247.2147.49

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Barbershop - Compute: Radeon HIPRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep80160240320400SE +/- 0.08, N = 3SE +/- 0.71, N = 3SE +/- 0.16, N = 3SE +/- 0.06, N = 3348.26258.06164.7391.4491.97

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Junkshop - Compute: Radeon HIPRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep1428425670SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.01, N = 363.3847.9731.3618.6718.76

GpuOwl

GpuOwl is a Mersenne primality tester leveraging OpenCL for cross-vendor GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations / Second, More Is BetterGpuOwl 7.5Exponent: 77936867RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep30060090012001500SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.00, N = 3SE +/- 0.22, N = 3SE +/- 1.00, N = 3SE +/- 0.60, N = 3SE +/- 0.40, N = 3226.64449.511490.31226.91345.11545.261088.531090.511. (CXX) g++ options: -O3 -lgmp -lOpenCL

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Classroom - Compute: Radeon HIPRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep20406080100SE +/- 0.16, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 384.3963.0140.4522.2822.03

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: SupercarRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep1632486480SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 324.4036.9273.9118.5025.0836.4554.9755.23

ProjectPhysX OpenCL-Benchmark

ProjectPhysX OpenCL-Benchmark provides various OpenCL compute and memory bandwidth micro-benchmarks Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: Memory Bandwidth Coalesced ReadRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2004006008001000SE +/- 0.00, N = 3SE +/- 0.00, N = 4SE +/- 0.06, N = 6SE +/- 0.32, N = 3SE +/- 2.74, N = 3SE +/- 10.49, N = 3SE +/- 2.98, N = 5207.71326.85865.57146.92237.72403.72590.24595.441. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

IndigoBench

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

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: BedroomRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep714212835SE +/- 0.002, N = 3SE +/- 0.009, N = 3SE +/- 0.029, N = 3SE +/- 0.003, N = 3SE +/- 0.002, N = 3SE +/- 0.004, N = 3SE +/- 0.007, N = 38.16112.87829.2826.3708.74012.03319.34819.324

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: GPURTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep510152025SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 12SE +/- 0.07, N = 12SE +/- 0.11, N = 12SE +/- 0.17, N = 124.617.9618.951.922.544.096.286.48MIN: 4.41 / MAX: 4.76MIN: 7.71 / MAX: 8.17MIN: 18.15 / MAX: 19.18MIN: 0.52 / MAX: 2.75MIN: 0.84 / MAX: 4.42MIN: 1.27 / MAX: 6.75MIN: 6.33 / MAX: 6.79

ProjectPhysX OpenCL-Benchmark

ProjectPhysX OpenCL-Benchmark provides various OpenCL compute and memory bandwidth micro-benchmarks Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTIOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: INT8 ComputeRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep510152025SE +/- 0.011, N = 3SE +/- 0.017, N = 4SE +/- 0.068, N = 6SE +/- 0.035, N = 3SE +/- 0.070, N = 3SE +/- 0.012, N = 3SE +/- 0.050, N = 54.7879.69221.7922.7273.8206.11512.56612.4631. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

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: Rainbow Colors and Prism - Acceleration: GPURTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep816243240SE +/- 0.02, N = 4SE +/- 0.04, N = 5SE +/- 0.09, N = 7SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 5SE +/- 0.08, N = 511.9217.6235.656.929.2714.9820.6920.82MIN: 10.89 / MAX: 12.56MIN: 16.21 / MAX: 18.4MIN: 31.82 / MAX: 39.14MIN: 6.36 / MAX: 7.09MIN: 8.58 / MAX: 9.52MIN: 13.88 / MAX: 15.48MIN: 19.26 / MAX: 21.65MIN: 19.25 / MAX: 21.63

ProjectPhysX OpenCL-Benchmark

ProjectPhysX OpenCL-Benchmark provides various OpenCL compute and memory bandwidth micro-benchmarks Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTIOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: INT16 ComputeRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep816243240SE +/- 0.011, N = 3SE +/- 0.004, N = 4SE +/- 0.127, N = 6SE +/- 0.078, N = 3SE +/- 0.221, N = 3SE +/- 0.062, N = 3SE +/- 0.144, N = 55.94711.48332.2974.8556.73110.69322.45922.2541. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

OpenBenchmarking.orgTFLOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: FP32 ComputeRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep20406080100SE +/- 0.000, N = 3SE +/- 0.002, N = 4SE +/- 0.081, N = 6SE +/- 0.102, N = 3SE +/- 0.135, N = 3SE +/- 0.040, N = 3SE +/- 0.188, N = 513.43926.96190.8056.4808.91914.35331.03130.3491. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

OpenBenchmarking.orgTIOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: INT64 ComputeRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep0.96461.92922.89383.85844.823SE +/- 0.009, N = 3SE +/- 0.060, N = 4SE +/- 0.021, N = 6SE +/- 0.005, N = 3SE +/- 0.004, N = 3SE +/- 0.006, N = 3SE +/- 0.016, N = 51.7352.8243.7741.0471.4632.3644.2264.2871. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

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: LuxCore Benchmark - Acceleration: GPURTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 12SE +/- 0.07, N = 12SE +/- 0.12, N = 12SE +/- 0.18, N = 124.638.0317.782.193.014.907.697.88MIN: 1.89 / MAX: 5.26MIN: 3.12 / MAX: 9.08MIN: 7.98 / MAX: 20.82MIN: 0.35 / MAX: 2.52MIN: 0.48 / MAX: 3.47MIN: 0.86 / MAX: 5.64MIN: 1.31 / MAX: 8.88MIN: 3.53 / MAX: 8.87

ProjectPhysX OpenCL-Benchmark

ProjectPhysX OpenCL-Benchmark provides various OpenCL compute and memory bandwidth micro-benchmarks Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: Memory Bandwidth Coalesced WriteRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2004006008001000SE +/- 0.01, N = 3SE +/- 0.02, N = 4SE +/- 0.26, N = 6SE +/- 1.37, N = 3SE +/- 2.65, N = 3SE +/- 3.49, N = 3SE +/- 1.25, N = 5211.81333.78850.04149.23213.46308.93459.22454.441. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

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: Orange Juice - Acceleration: GPURTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep48121620SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 35.217.7015.861.051.292.312.932.93MIN: 4.19 / MAX: 6.51MIN: 6.01 / MAX: 9.92MIN: 13.68 / MAX: 21.62MIN: 2.09 / MAX: 3.15MIN: 2.69 / MAX: 3.71MIN: 2.7 / MAX: 3.7

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: GPURTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep48121620SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.17, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 34.096.9217.721.902.554.166.626.62MIN: 1.83 / MAX: 4.63MIN: 3.18 / MAX: 7.9MIN: 6.81 / MAX: 20.39MIN: 0.73 / MAX: 2.13MIN: 1.07 / MAX: 2.84MIN: 1.91 / MAX: 4.68MIN: 2.89 / MAX: 7.51MIN: 3.12 / MAX: 7.46

FinanceBench

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

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCLRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep48121620SE +/- 0.001644, N = 14SE +/- 0.041867, N = 15SE +/- 0.003806, N = 15SE +/- 0.061748, N = 14SE +/- 0.061700, N = 15SE +/- 0.015787, N = 15SE +/- 0.017790, N = 1516.2281437.6930002.71000012.2010009.7240006.7790003.3820003.3600001. (CXX) g++ options: -O3 -march=native -fopenmp

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Monte-Carlo OpenCLRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep70140210280350SE +/- 0.11, N = 7SE +/- 0.04, N = 7SE +/- 0.04, N = 7SE +/- 0.35, N = 7SE +/- 0.99, N = 7SE +/- 0.22, N = 7SE +/- 0.25, N = 7266.04199.9399.24327.08261.74152.22100.7297.541. (CXX) g++ options: -O3 -march=native -fopenmp

clpeak

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

OpenBenchmarking.orgGBPS, More Is Betterclpeak 1.1.2OpenCL Test: Global Memory BandwidthRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2004006008001000SE +/- 0.06, N = 7SE +/- 0.05, N = 8SE +/- 0.03, N = 11SE +/- 0.08, N = 5SE +/- 0.18, N = 7SE +/- 0.98, N = 8SE +/- 1.00, N = 9195.00307.54815.91139.39220.39391.21496.42493.741. (CXX) g++ options: -O3

OpenBenchmarking.orgGFLOPS, More Is Betterclpeak 1.1.2OpenCL Test: Single-Precision ComputeRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep20K40K60K80K100KSE +/- 14.33, N = 15SE +/- 26.99, N = 13SE +/- 359.84, N = 13SE +/- 9.29, N = 12SE +/- 14.09, N = 12SE +/- 27.07, N = 12SE +/- 73.46, N = 1211596.8225785.6883188.796072.167710.9512304.2425936.6526088.301. (CXX) g++ options: -O3

OpenBenchmarking.orgGFLOPS, More Is Betterclpeak 1.1.2OpenCL Test: Double-Precision ComputeRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep30060090012001500SE +/- 0.22, N = 5SE +/- 0.36, N = 5SE +/- 2.12, N = 6SE +/- 0.17, N = 7SE +/- 0.86, N = 7SE +/- 1.47, N = 7SE +/- 0.78, N = 8217.61439.321534.70238.56332.83462.251096.591099.211. (CXX) g++ options: -O3

OpenBenchmarking.orgGIOPS, More Is Betterclpeak 1.1.2OpenCL Test: Integer 24-bit ComputeRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep9K18K27K36K45KSE +/- 17.53, N = 13SE +/- 6.66, N = 13SE +/- 289.03, N = 15SE +/- 25.04, N = 12SE +/- 31.15, N = 13SE +/- 27.65, N = 13SE +/- 99.46, N = 135965.8013209.0640310.585462.756985.9511408.6622802.7422557.931. (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 20230730fp32-scalarRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep10K20K30K40K50KSE +/- 12.68, N = 3SE +/- 16.39, N = 3SE +/- 370.02, N = 3SE +/- 42.52, N = 8SE +/- 53.71, N = 3SE +/- 10.33, N = 3SE +/- 289.96, N = 46944.9413990.2748174.334887.927784.7312385.2325619.4125643.97

GpuOwl

GpuOwl is a Mersenne primality tester leveraging OpenCL for cross-vendor GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations / Second, More Is BetterGpuOwl 7.5Exponent: 57885161RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep400800120016002000SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 1.39, N = 3SE +/- 0.23, N = 3SE +/- 0.07, N = 3SE +/- 0.18, N = 3SE +/- 2.85, N = 3308.39612.372043.60315.99467.65735.111460.581470.591. (CXX) g++ options: -O3 -lgmp -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 2.3.2RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation100200300400500SE +/- 0.16, N = 3SE +/- 0.14, N = 3SE +/- 0.94, N = 3207.85281.65462.31

Radeon PRO W7500: The test run did not produce a result.

Radeon PRO W7600: The test run did not produce a result.

Radeon PRO W7700: The test run did not produce a result.

Radeon PRO W7900: The test run did not produce a result.

W7900 rep: The test run did not produce a result.

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 20230730fp32-vec4RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep13K26K39K52K65KSE +/- 18.50, N = 3SE +/- 31.57, N = 3SE +/- 485.03, N = 3SE +/- 32.25, N = 8SE +/- 16.67, N = 3SE +/- 6.54, N = 3SE +/- 342.60, N = 49186.8118498.6462508.544345.636876.7510997.0722828.3722365.75

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-scalarRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep10K20K30K40K50KSE +/- 14.03, N = 3SE +/- 9.85, N = 3SE +/- 153.56, N = 3SE +/- 32.58, N = 8SE +/- 11.58, N = 3SE +/- 3.75, N = 3SE +/- 158.36, N = 46968.1713985.4746258.435005.487839.6812392.9525838.0825688.06

ProjectPhysX OpenCL-Benchmark

ProjectPhysX OpenCL-Benchmark provides various OpenCL compute and memory bandwidth micro-benchmarks Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTFLOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: FP64 ComputeRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep0.33980.67961.01941.35921.699SE +/- 0.000, N = 3SE +/- 0.000, N = 4SE +/- 0.000, N = 6SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.002, N = 50.2150.4331.5100.2370.3240.4691.0571.0791. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

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 20230730fp16-vec4RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep20K40K60K80K100KSE +/- 27.78, N = 3SE +/- 16.63, N = 3SE +/- 692.18, N = 3SE +/- 53.19, N = 8SE +/- 2.46, N = 3SE +/- 8.54, N = 3SE +/- 135.83, N = 413745.3027650.7093562.478701.8715112.5325066.8549210.0948905.97

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-scalarRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep30060090012001500SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.14, N = 3SE +/- 0.81, N = 8SE +/- 0.01, N = 3SE +/- 0.12, N = 3SE +/- 0.27, N = 4218.82440.871538.94235.12332.94484.511118.441117.74

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-vec4RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep30060090012001500SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.73, N = 8SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.18, N = 4218.82440.011533.64229.92329.52484.531117.471117.36

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-scalarRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep7K14K21K28K35KSE +/- 8.93, N = 3SE +/- 11.44, N = 3SE +/- 5.73, N = 3SE +/- 15.50, N = 8SE +/- 0.62, N = 3SE +/- 0.51, N = 3SE +/- 13.76, N = 44634.079328.0131027.664897.027740.8812377.1425473.8925460.76

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-vec4RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep10K20K30K40K50KSE +/- 0.02, N = 3SE +/- 0.51, N = 3SE +/- 89.71, N = 3SE +/- 28.67, N = 8SE +/- 2.93, N = 3SE +/- 5.00, N = 3SE +/- 18.90, N = 46167.1512381.1138953.718451.3515006.1225011.6948689.6848657.96

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-scalarW7900 rep140028004200560070006492.05

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-vec4W7900 rep140028004200560070006554.46

FluidX3D

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.17Test: FP32-FP32RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep11002200330044005500SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 16.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 1.15, N = 3SE +/- 7.22, N = 31331204752528561177163227962878

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.17Test: FP32-FP16SRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2K4K6K8K10KSE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 2.00, N = 3SE +/- 2.31, N = 3SE +/- 0.88, N = 3SE +/- 1.45, N = 3SE +/- 12.73, N = 3243437701020816302194291146504778

ProjectPhysX OpenCL-Benchmark

ProjectPhysX OpenCL-Benchmark provides various OpenCL compute and memory bandwidth micro-benchmarks Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTFLOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: FP16 ComputeW7900 rep132639526556.821. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

FluidX3D

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.17Test: FP32-FP16CRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2K4K6K8K10KSE +/- 1.67, N = 3SE +/- 2.03, N = 3SE +/- 3.50, N = 4SE +/- 0.00, N = 3SE +/- 3.84, N = 3SE +/- 2.73, N = 3SE +/- 73.37, N = 3253140391058516822317286555035612

ParaView

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Many Spheres - Frames: 3000 - Resolution: 2560 x 1440RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep4080120160200SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.74, N = 3SE +/- 0.05, N = 3SE +/- 0.25, N = 3SE +/- 0.18, N = 3SE +/- 0.41, N = 347.2796.58167.6155.0175.07109.72168.48167.57

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Many Spheres - Frames: 3000 - Resolution: 3840 x 2160RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep4080120160200SE +/- 0.02, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.11, N = 343.2989.04158.4749.0868.7796.28147.76147.89

SPECViewPerf 2020

This test runs SPECViewPerf 2020 if available on your system. SPECViewPerf is made up of real-world OpenGL workstation tests such as CATIA and SolidWorks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: CREO-03RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W790060120180240300SE +/- 0.07, N = 3SE +/- 0.17, N = 3SE +/- 0.47, N = 3SE +/- 0.13, N = 3SE +/- 0.08, N = 3SE +/- 0.14, N = 3SE +/- 0.53, N = 3113.82170.61256.46113.67151.46208.57276.10

Resolution: 2560 x 1440 - Viewset: CREO-03

W7900 rep: The test run did not produce a result.

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: MAYA-06RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2004006008001000SE +/- 0.26, N = 3SE +/- 0.09, N = 3SE +/- 0.66, N = 3SE +/- 0.58, N = 3SE +/- 0.09, N = 3SE +/- 1.26, N = 3SE +/- 4.21, N = 3276.44430.10813.18229.65316.25501.38770.35786.14

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: ENERGY-03RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep306090120150SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.57, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 331.7362.74156.1714.5323.0336.6767.4668.26

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: SNX-04RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2004006008001000SE +/- 0.62, N = 3SE +/- 0.27, N = 3SE +/- 5.90, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.42, N = 3285.22423.661026.56209.92263.69360.23582.30581.83

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: MEDICAL-O3RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep50100150200250SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.21, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 365.6097.35211.9816.0225.2733.2653.7153.95

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: SOLIDWORKS-07RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep120240360480600SE +/- 0.10, N = 3SE +/- 0.07, N = 3SE +/- 0.20, N = 3SE +/- 0.25, N = 3SE +/- 0.09, N = 3SE +/- 0.28, N = 3SE +/- 0.42, N = 3184.73308.72578.10162.17216.52336.76530.89532.83

ProjectPhysX OpenCL-Benchmark

ProjectPhysX OpenCL-Benchmark provides various OpenCL compute and memory bandwidth micro-benchmarks Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTIOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: INT32 ComputeW7900 rep2468106.0131. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

ParaView

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Wavelet Contour - Frames: 3000 - Resolution: 2560 x 1440RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2004006008001000SE +/- 0.41, N = 3SE +/- 0.38, N = 3SE +/- 0.63, N = 3SE +/- 0.14, N = 3SE +/- 0.38, N = 3SE +/- 0.87, N = 3SE +/- 1.91, N = 3194.01346.77811.19218.56274.72400.30734.05730.54

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Wavelet Contour - Frames: 3000 - Resolution: 3840 x 2160RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep140280420560700SE +/- 0.11, N = 3SE +/- 0.16, N = 3SE +/- 0.45, N = 3SE +/- 0.07, N = 3SE +/- 0.10, N = 3SE +/- 0.57, N = 3SE +/- 0.90, N = 3145.70267.98625.76160.57212.95320.57582.89580.31

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Wavelet Volume - Frames: 3000 - Resolution: 2560 x 1440RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep2004006008001000SE +/- 0.37, N = 3SE +/- 0.99, N = 3SE +/- 2.56, N = 3SE +/- 0.21, N = 3SE +/- 1.21, N = 3SE +/- 1.43, N = 3SE +/- 2.07, N = 3285.73466.70853.68291.53398.03564.16653.03644.40

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Wavelet Volume - Frames: 3000 - Resolution: 3840 x 2160RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep140280420560700SE +/- 0.22, N = 3SE +/- 0.11, N = 3SE +/- 6.56, N = 3SE +/- 0.19, N = 3SE +/- 0.07, N = 3SE +/- 0.42, N = 3SE +/- 2.09, N = 3150.83269.03639.67151.97229.33340.94589.04549.68

GpuOwl

GpuOwl is a Mersenne primality tester leveraging OpenCL for cross-vendor GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations / Second, More Is BetterGpuOwl 7.5Exponent: 332220523RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep70140210280350SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 1.45, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 347.8595.30322.4249.0866.18102.10206.30206.441. (CXX) g++ options: -O3 -lgmp -lOpenCL

SPECViewPerf 2020

This test runs SPECViewPerf 2020 if available on your system. SPECViewPerf is made up of real-world OpenGL workstation tests such as CATIA and SolidWorks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: CATIA-06RTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep50100150200250SE +/- 0.26, N = 3SE +/- 0.05, N = 3SE +/- 1.01, N = 3SE +/- 0.57, N = 3SE +/- 0.37, N = 3SE +/- 0.33, N = 3SE +/- 1.26, N = 383.69142.95236.9667.8487.87137.56203.22203.93

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Pabellon Barcelona - Compute: NVIDIA CUDARTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation20406080100SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 3108.9559.2021.30

GPU Power Consumption Monitor

OpenBenchmarking.orgWattsGPU Power Consumption MonitorPhoronix Test Suite System MonitoringRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep50100150200250Min: 7.09 / Avg: 47.88 / Max: 70Min: 6.46 / Avg: 78.35 / Max: 130.8Min: 21.72 / Avg: 191 / Max: 303.57Min: 1 / Avg: 43.46 / Max: 54Min: 1 / Avg: 79.44 / Max: 101Min: 5 / Avg: 111.28 / Max: 154Min: 8 / Avg: 189.77 / Max: 249Min: 8 / Avg: 147.42 / Max: 248

GPU Temperature Monitor

OpenBenchmarking.orgCelsiusGPU Temperature MonitorPhoronix Test Suite System MonitoringRTX 2000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada GenerationRadeon PRO W7500Radeon PRO W7600Radeon PRO W7700Radeon PRO W7900W7900 rep1632486480Min: 36 / Avg: 66.43 / Max: 79Min: 29 / Avg: 63.59 / Max: 80Min: 41 / Avg: 72.97 / Max: 87Min: 30 / Avg: 74.13 / Max: 83Min: 27 / Avg: 76.2 / Max: 86Min: 31 / Avg: 68.93 / Max: 83Min: 27 / Avg: 70.28 / Max: 80Min: 29 / Avg: 64.53 / Max: 82

Meta Performance Per Watts

OpenBenchmarking.orgPerformance Per Watts, More Is BetterMeta Performance Per WattsPerformance Per WattsW7900 rep140280420560700634.64

76 Results Shown

Rodinia
Chaos Group V-RAY:
  NVIDIA CUDA GPU
  NVIDIA RTX GPU
Blender:
  BMW27 - NVIDIA CUDA
  BMW27 - NVIDIA OptiX
  Classroom - NVIDIA CUDA
  Classroom - NVIDIA OptiX
  Fishy Cat - NVIDIA CUDA
  Fishy Cat - NVIDIA OptiX
  Pabellon Barcelona - NVIDIA OptiX
  Barbershop - NVIDIA CUDA
  Barbershop - NVIDIA OptiX
  Junkshop - NVIDIA CUDA
  Junkshop - NVIDIA OptiX
  BMW27 - Radeon HIP
  Fishy Cat - Radeon HIP
  Pabellon Barcelona - Radeon HIP
  Barbershop - Radeon HIP
  Junkshop - Radeon HIP
GpuOwl
Blender
IndigoBench
ProjectPhysX OpenCL-Benchmark
IndigoBench
LuxCoreRender
ProjectPhysX OpenCL-Benchmark
LuxCoreRender
ProjectPhysX OpenCL-Benchmark:
  INT16 Compute
  FP32 Compute
  INT64 Compute
LuxCoreRender
ProjectPhysX OpenCL-Benchmark
LuxCoreRender:
  Orange Juice - GPU
  Danish Mood - GPU
FinanceBench:
  Black-Scholes OpenCL
  Monte-Carlo OpenCL
clpeak:
  Global Memory Bandwidth
  Single-Precision Compute
  Double-Precision Compute
  Integer 24-bit Compute
vkpeak
GpuOwl
FAHBench
vkpeak:
  fp32-vec4
  fp16-scalar
ProjectPhysX OpenCL-Benchmark
vkpeak:
  fp16-vec4
  fp64-scalar
  fp64-vec4
  int16-scalar
  int16-vec4
  int32-scalar
  int32-vec4
FluidX3D:
  FP32-FP32
  FP32-FP16S
ProjectPhysX OpenCL-Benchmark
FluidX3D
ParaView:
  Many Spheres - 3000 - 2560 x 1440
  Many Spheres - 3000 - 3840 x 2160
SPECViewPerf 2020:
  2560 x 1440 - CREO-03
  2560 x 1440 - MAYA-06
  2560 x 1440 - ENERGY-03
  2560 x 1440 - SNX-04
  2560 x 1440 - MEDICAL-O3
  2560 x 1440 - SOLIDWORKS-07
ProjectPhysX OpenCL-Benchmark
ParaView:
  Wavelet Contour - 3000 - 2560 x 1440
  Wavelet Contour - 3000 - 3840 x 2160
  Wavelet Volume - 3000 - 2560 x 1440
  Wavelet Volume - 3000 - 3840 x 2160
GpuOwl
SPECViewPerf 2020
Blender
GPU Power Consumption Monitor:
  Phoronix Test Suite System Monitoring:
    Watts
    Celsius
  Performance Per Watts:
    Performance Per Watts