NVIDIA RTX 6000 Ada Generation

Benchmarks for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2409138-PTS-NVIDIART87
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
HPC - High Performance Computing 3 Tests
Multi-Core 7 Tests
NVIDIA GPU Compute 10 Tests
OpenCL 3 Tests
OpenMPI Tests 2 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 6000 Ada Generation
September 11
  4 Hours, 51 Minutes
4
September 12
  1 Hour, 34 Minutes
5
September 12
  1 Hour, 34 Minutes
RTX 4000 Ada Generation
September 12
  5 Hours, 30 Minutes
NVIDIA RTX 4000 Ada Generation
September 12
  1 Hour, 48 Minutes
RTX 2000 Ada Generation
September 12
  6 Hours, 42 Minutes
2
September 12
  6 Hours, 12 Minutes
2a
September 13
  6 Hours, 40 Minutes
Invert Hiding All Results Option
  4 Hours, 21 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 6000 Ada GenerationProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLCompilerFile-SystemScreen ResolutionRTX 6000 Ada Generation45RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 2000 Ada Generation22aAMD 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 6000 Ada Generation 48GBNVIDIA AD102 HD AudioDELL 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 2000 Ada Generation 16GBNVIDIA Device 22beOpenBenchmarking.orgKernel Details- nouveau.modeset=0 - 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 6000 Ada Generation: BAR1 / Visible vRAM Size: 65536 MiB - vBIOS Version: 95.02.3a.00.01- 4: BAR1 / Visible vRAM Size: 65536 MiB - vBIOS Version: 95.02.3a.00.01- 5: BAR1 / Visible vRAM Size: 65536 MiB - vBIOS Version: 95.02.3a.00.01- RTX 4000 Ada Generation: BAR1 / Visible vRAM Size: 32768 MiB - vBIOS Version: 95.04.5c.00.0d- NVIDIA RTX 4000 Ada Generation: BAR1 / Visible vRAM Size: 32768 MiB - vBIOS Version: 95.04.5c.00.0d- RTX 2000 Ada Generation: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.07.47.00.05- 2: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.07.47.00.05- 2a: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.07.47.00.05OpenCL Details- RTX 6000 Ada Generation: GPU Compute Cores: 18176- 4: GPU Compute Cores: 18176- 5: GPU Compute Cores: 18176- RTX 4000 Ada Generation: GPU Compute Cores: 6144- NVIDIA RTX 4000 Ada Generation: GPU Compute Cores: 6144- RTX 2000 Ada Generation: GPU Compute Cores: 2816- 2: GPU Compute Cores: 2816- 2a: GPU Compute Cores: 2816Python 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 6000 Ada Generation45RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 2000 Ada Generation22aResult OverviewPhoronix Test Suite100%245%391%536%682%vkpeakGpuOwlProjectPhysX OpenCL-BenchmarkFluidX3DFinanceBenchBlenderParaViewclpeakSPECViewPerf 2020FAHBenchRodinia

NVIDIA RTX 6000 Ada Generationclpeak: Single-Precision Computeclpeak: Integer 24-bit Computeclpeak: Double-Precision Computevkpeak: fp64-scalarvkpeak: fp32-scalaropencl-benchmark: FP64 Computevkpeak: fp64-vec4vkpeak: int32-scalarvkpeak: int32-vec4vkpeak: fp16-vec4vkpeak: fp32-vec4clpeak: Integer Computeopencl-benchmark: FP32 Computegpuowl: 332220523vkpeak: int16-scalarvkpeak: fp16-scalargpuowl: 57885161gpuowl: 77936867opencl-benchmark: INT32 Computevkpeak: int16-vec4financebench: Black-Scholes OpenCLopencl-benchmark: INT16 Computeblender: Pabellon Barcelona - NVIDIA CUDAspecviewperf2020: 2560 x 1440 - ENERGY-03opencl-benchmark: INT8 Computeblender: Fishy Cat - NVIDIA CUDAblender: Classroom - NVIDIA CUDArodinia: OpenCL Particle Filterluxcorerender: Danish Mood - GPUparaview: Wavelet Contour - 3000 - 3840 x 2160blender: Fishy Cat - NVIDIA OptiXparaview: Wavelet Volume - 3000 - 3840 x 2160v-ray: NVIDIA CUDA GPUfluidx3d: FP32-FP16Sparaview: Wavelet Contour - 3000 - 2560 x 1440fluidx3d: FP32-FP16Cclpeak: Global Memory Bandwidthopencl-benchmark: Memory Bandwidth Coalesced Readblender: Barbershop - NVIDIA CUDAluxcorerender: DLSC - GPUblender: BMW27 - NVIDIA CUDAopencl-benchmark: Memory Bandwidth Coalesced Writev-ray: NVIDIA RTX GPUfluidx3d: FP32-FP32luxcorerender: LuxCore Benchmark - GPUparaview: Many Spheres - 3000 - 3840 x 2160blender: Classroom - NVIDIA OptiXspecviewperf2020: 2560 x 1440 - SNX-04indigobench: OpenCL GPU - Bedroomparaview: Many Spheres - 3000 - 2560 x 1440blender: Pabellon Barcelona - NVIDIA OptiXblender: Barbershop - NVIDIA OptiXblender: Junkshop - NVIDIA CUDAspecviewperf2020: 2560 x 1440 - MEDICAL-O3specviewperf2020: 2560 x 1440 - SOLIDWORKS-07luxcorerender: Orange Juice - GPUblender: Junkshop - NVIDIA OptiXindigobench: OpenCL GPU - Supercarparaview: Wavelet Volume - 3000 - 2560 x 1440luxcorerender: Rainbow Colors and Prism - GPUspecviewperf2020: 2560 x 1440 - MAYA-06blender: BMW27 - NVIDIA OptiXspecviewperf2020: 2560 x 1440 - CATIA-06financebench: Monte-Carlo OpenCLspecviewperf2020: 2560 x 1440 - CREO-03fahbench: opencl-benchmark: INT64 Computeclpeak: Transfer Bandwidth enqueueReadBufferclpeak: Transfer Bandwidth enqueueWriteBufferrodinia: OpenCL Myocyteclpeak: Kernel Latencyparaview: Wavelet Volume - 3000 - 3840 x 2160paraview: Wavelet Volume - 3000 - 2560 x 1440paraview: Wavelet Contour - 3000 - 3840 x 2160paraview: Wavelet Contour - 3000 - 2560 x 1440paraview: Many Spheres - 3000 - 3840 x 2160paraview: Many Spheres - 3000 - 2560 x 1440RTX 6000 Ada Generation45RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 2000 Ada Generation22a83188.7940310.581534.701538.9448174.331.511533.6448838.8348137.8593562.4762508.5440360.7590.805322.4231027.6646258.432043.601490.3144.16138953.712.71032.29721.30156.1721.79210.6710.442.09117.72625.765.48639.67598410208811.1910585815.91865.5745.6518.955.57850.048174525217.78158.477.921026.5629.282167.618.8632.9111.08211.98578.1015.867.3473.905853.6835.65813.183.61236.9699.239287256.46462.31313.77422.8722.1120.1643.7910234.76213658.9056521.1528453.56715887.66716804.13882190.3642249.871526.621539.9448751.251.5021533.5748805.5747938.794400.5263072.9738762.5990.179322.5830661.8946494.292044.991490.3143.50438068.152.70231.16821.13156.3620.84410.6110.382.11517.78622.925.4634594810175806.7410545816.12866.1945.77195.55860.258159526717.79157.737.851029.4429.302165.378.8832.8311.07212.24579.4915.557.3773.808846.8635.77812.163.62237.3599.321999256.36462.97853.69322.6621.8920.3063.7810144.05313549.7966491.5268407.18115813.25816578.78583007.7139686.791532.811539.9948735.811.5021535.4848814.9148068.4493801.4863065.0538304.8990.172322.4830694.2946558.282044.991490.3143.86637707.092.69230.94621.13155.9120.84610.6410.342.08317.54625.145.41638.96595910209805.1710584815.88866.4745.66195.55860.038159526817.81157.747.821029.3729.258165.548.8232.9411.12211.55579.4115.897.3473.782863.2435.73812.53.61236.7499.347255.97467.31083.81223.8622.9820.273.7910223.37613811.8916514.7058390.82615813.78516596.14125836.2013239.24439.07440.8014015.620.433439.9113996.8513937.8327661.6518522.5313179.7126.94095.659337.3314009.36610.87451.0613.87512377.797.66011.38859.1962.799.70626.1926.634.9856.90267.5813.02269.9319463805345.924064307.31326.93108.147.9613.26333.72292720468.0488.9317.91423.8712.86496.7719.4472.3922.9197.36308.767.6914.5436.886474.4717.59430.487.24141.84199.828003170.61282.15942.87522.6922.2021.6843.734318.9077591.5892788.5453604.9088915.5889701.1742572813222.86439.76440.8614015.80.433439.9313995.613936.6127782.7918566.913209.7526.92195.749344.5514070.21611.25451.6713.87212378.897.64111.3859.1362.99.70726.1826.524.9847.07266.7812.92268.9819603804344.934066307.47326.92108.087.9713.25333.67293720478.0588.5717.82423.8512.8996.6119.3872.122.9997.38308.687.7114.5936.944468.317.57430.27.22143.77200.210999170.79282.15422.82421.7521.921.5093.74303.657492.8442780.1173594.5398879.4669685.69911704.575987.65217.30218.856944.900.215218.846959.066926.5713744.129185.795978.9713.43947.864635.636967.84308.36226.556.8096166.9316.2280005.924108.9631.694.77647.2845.899.1724.09145.5823.02151.4114222434193.512529195.11207.71188.784.622.52211.82203013314.6343.1628.61285.108.15047.1231.18114.4338.3565.49184.605.2122.3524.390286.6111.96276.2510.5283.87265.878001113.93207.95291.75012.9213.120.7603.772422.5394585.8191517.1152016.5814326.6324723.58711608.165951.11218.65218.766969.220.215218.706959.026926.8113799.419223.625936.0913.43847.864643.866995.37308.26226.606.8096166.5816.2253345.920108.9831.744.77647.2445.869.190145.8422.76151.282434193.972530195.10207.71189.1522.49211.84133243.2028.59285.5747.1931.20114.6438.4665.60184.6222.19287.09276.5810.4683.90266.163005114.14208.01541.73212.9313.1020.8173.762420.4264593.5341519.7532021.3694331.2674731.25711616.095954.25218.01218.846941.300.215218.766959.086926.8713744.299184.815939.2413.43947.864625.436967.87308.26226.606.8236166.8816.2270005.922108.9731.794.77647.2445.879.1874.06145.9922.79151.0914152434194.282530195.05207.71189.044.6122.55211.82202313314.6343.2828.55286.318.15747.3131.14114.6738.2965.73184.995.2122.1124.402287.4911.95277.1410.4483.57265.861664114.14207.66301.72812.9313.120.6943.782417.4264599.8971521.4082024.6104338.5484743.235OpenBenchmarking.org

clpeak

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

OpenBenchmarking.orgGFLOPS, More Is Betterclpeak 1.1.2OpenCL Test: Single-Precision Compute22aRTX 2000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation20K40K60K80K100KSE +/- 15.71, N = 3SE +/- 50.14, N = 3SE +/- 122.39, N = 3SE +/- 33.21, N = 3SE +/- 359.84, N = 1311608.1611616.0911704.5725728.0025836.2082190.3683007.7183188.791. (CXX) g++ options: -O3

OpenBenchmarking.orgGIOPS, More Is Betterclpeak 1.1.2OpenCL Test: Integer 24-bit Compute22aRTX 2000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation5RTX 6000 Ada Generation49K18K27K36K45KSE +/- 0.00, N = 3SE +/- 3.51, N = 3SE +/- 36.62, N = 3SE +/- 7.41, N = 3SE +/- 289.03, N = 155951.115954.255987.6513222.8613239.2439686.7940310.5842249.871. (CXX) g++ options: -O3

OpenBenchmarking.orgGFLOPS, More Is Betterclpeak 1.1.2OpenCL Test: Double-Precision ComputeRTX 2000 Ada Generation2a2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation30060090012001500SE +/- 0.28, N = 3SE +/- 0.24, N = 3SE +/- 0.13, N = 3SE +/- 0.63, N = 3SE +/- 2.12, N = 6217.30218.01218.65439.07439.761526.621532.811534.701. (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 20230730fp64-scalar22aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation4530060090012001500SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 3218.76218.84218.85440.80440.861538.941539.941539.99

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-scalar2aRTX 2000 Ada Generation2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation5410K20K30K40K50KSE +/- 14.03, N = 3SE +/- 12.38, N = 3SE +/- 0.00, N = 3SE +/- 2.02, N = 3SE +/- 370.02, N = 36941.306944.906969.2214015.6214015.8048174.3348735.8148751.25

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 Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation0.33980.67961.01941.35921.699SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 60.2150.2150.2150.4330.4331.5021.5021.5101. (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 20230730fp64-vec422aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation4RTX 6000 Ada Generation530060090012001500SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3218.70218.76218.84439.91439.931533.571533.641535.48

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-scalar2RTX 2000 Ada Generation2aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation10K20K30K40K50KSE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 1.02, N = 3SE +/- 4.83, N = 36959.026959.066959.0813995.6013996.8548805.5748814.9148838.83

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-vec4RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation10K20K30K40K50KSE +/- 0.25, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.20, N = 3SE +/- 34.51, N = 36926.576926.816926.8713936.6113937.8347938.7048068.4448137.85

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-vec4RTX 2000 Ada Generation2a2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation5420K40K60K80K100KSE +/- 27.72, N = 3SE +/- 27.86, N = 3SE +/- 0.07, N = 3SE +/- 60.23, N = 3SE +/- 692.18, N = 313744.1213744.2913799.4127661.6527782.7993562.4793801.4894400.52

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-vec42aRTX 2000 Ada Generation2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation5414K28K42K56K70KSE +/- 19.42, N = 3SE +/- 18.82, N = 3SE +/- 0.25, N = 3SE +/- 38.83, N = 3SE +/- 485.03, N = 39184.819185.799223.6218522.5318566.9062508.5463065.0563072.97

clpeak

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

OpenBenchmarking.orgGIOPS, More Is Betterclpeak 1.1.2OpenCL Test: Integer Compute22aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation54RTX 6000 Ada Generation9K18K27K36K45KSE +/- 3.27, N = 3SE +/- 0.10, N = 3SE +/- 39.53, N = 3SE +/- 19.73, N = 3SE +/- 462.74, N = 155936.095939.245978.9713179.7113209.7538304.8938762.5940360.751. (CXX) g++ options: -O3

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: FP32 Compute2RTX 2000 Ada Generation2aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation54RTX 6000 Ada Generation20406080100SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.08, N = 613.4413.4413.4426.9226.9490.1790.1890.811. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

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 Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation5470140210280350SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 1.45, N = 347.8647.8647.8695.6595.74322.42322.48322.581. (CXX) g++ options: -O3 -lgmp -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.orgGIOPS, More Is Bettervkpeak 20230730int16-scalar2aRTX 2000 Ada Generation2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation7K14K21K28K35KSE +/- 9.35, N = 3SE +/- 8.47, N = 3SE +/- 0.29, N = 3SE +/- 0.96, N = 3SE +/- 5.73, N = 34625.434635.634643.869337.339344.5530661.8930694.2931027.66

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-scalarRTX 2000 Ada Generation2a2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation4510K20K30K40K50KSE +/- 14.04, N = 3SE +/- 14.03, N = 3SE +/- 0.54, N = 3SE +/- 28.84, N = 3SE +/- 153.56, N = 36967.846967.876995.3714009.3614070.2146258.4346494.2946558.28

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: 5788516122aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation45400800120016002000SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.22, N = 3SE +/- 1.39, N = 3308.26308.26308.36610.87611.252043.602044.992044.991. (CXX) g++ options: -O3 -lgmp -lOpenCL

OpenBenchmarking.orgIterations / Second, More Is BetterGpuOwl 7.5Exponent: 77936867RTX 2000 Ada Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation4530060090012001500SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3226.55226.60226.60451.06451.671490.311490.311490.311. (CXX) g++ options: -O3 -lgmp -lOpenCL

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 ComputeRTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation1020304050SE +/- 0.014, N = 3SE +/- 0.014, N = 3SE +/- 0.000, N = 3SE +/- 0.002, N = 3SE +/- 0.291, N = 66.8096.8096.82313.87213.87543.50443.86644.1611. (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.orgGIOPS, More Is Bettervkpeak 20230730int16-vec422aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation54RTX 6000 Ada Generation8K16K24K32K40KSE +/- 0.28, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.88, N = 3SE +/- 89.71, N = 36166.586166.886166.9312377.7912378.8937707.0938068.1538953.71

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 Generation2a2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation4548121620SE +/- 0.002646, N = 3SE +/- 0.001155, N = 3SE +/- 0.002186, N = 3SE +/- 0.008950, N = 3SE +/- 0.003806, N = 1516.22800016.22700016.2253347.6600007.6410002.7100002.7020002.6920001. (CXX) g++ options: -O3 -march=native -fopenmp

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 Compute22aRTX 2000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation54RTX 6000 Ada Generation816243240SE +/- 0.001, N = 3SE +/- 0.003, N = 3SE +/- 0.002, N = 3SE +/- 0.012, N = 3SE +/- 0.127, N = 65.9205.9225.92411.38011.38830.94631.16832.2971. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Pabellon Barcelona - Compute: NVIDIA CUDA22aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation5420406080100SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 3108.98108.97108.9659.1959.1321.3021.1321.13

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: ENERGY-03RTX 2000 Ada Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation5RTX 6000 Ada Generation4306090120150SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.57, N = 331.6931.7431.7962.7962.90155.91156.17156.36

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 Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation510152025SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.002, N = 3SE +/- 0.068, N = 64.7764.7764.7769.7069.70720.84420.84621.7921. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Fishy Cat - Compute: NVIDIA CUDARTX 2000 Ada Generation2a2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation541122334455SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 547.2847.2447.2426.1926.1810.6710.6410.61

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Classroom - Compute: NVIDIA CUDARTX 2000 Ada Generation2a2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation451020304050SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 545.8945.8745.8626.6326.5210.4410.3810.34

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 Filter22aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation4RTX 6000 Ada Generation53691215SE +/- 0.021, N = 3SE +/- 0.022, N = 3SE +/- 0.012, N = 3SE +/- 0.008, N = 3SE +/- 0.004, N = 119.1909.1879.1724.9854.9842.1152.0912.0831. (CXX) g++ options: -O2 -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: Danish Mood - Acceleration: GPU2aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation5RTX 6000 Ada Generation448121620SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.17, N = 34.064.096.907.0717.5417.7217.78MIN: 1.82 / MAX: 4.59MIN: 1.57 / MAX: 4.63MIN: 3.19 / MAX: 7.9MIN: 3.54 / MAX: 7.94MIN: 6.24 / MAX: 20.16MIN: 6.81 / MAX: 20.39MIN: 7.14 / MAX: 20.22

ParaView

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Wavelet Contour - Frames: 3000 - Resolution: 3840 x 2160RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation140280420560700SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.45, N = 3145.58145.84145.99266.78267.58622.92625.14625.76

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Fishy Cat - Compute: NVIDIA OptiXRTX 2000 Ada Generation2a2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation54612182430SE +/- 0.27, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.11, N = 7SE +/- 0.05, N = 1523.0222.7922.7613.0212.925.485.415.40

ParaView

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Wavelet Volume - Frames: 3000 - Resolution: 3840 x 21602a2RTX 2000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation140280420560700SE +/- 0.15, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.17, N = 3SE +/- 6.56, N = 3151.09151.28151.41268.98269.93634.00638.96639.67

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 GPU2aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation13002600390052006500SE +/- 6.06, N = 3SE +/- 3.67, N = 3SE +/- 3.67, N = 3SE +/- 24.01, N = 31415142219461960594859595984

FluidX3D

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.17Test: FP32-FP16SRTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation4RTX 6000 Ada Generation52K4K6K8K10KSE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.88, N = 3SE +/- 2.00, N = 324342434243438043805101751020810209

ParaView

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Wavelet Contour - Frames: 3000 - Resolution: 2560 x 1440RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation54RTX 6000 Ada Generation2004006008001000SE +/- 0.43, N = 3SE +/- 0.37, N = 3SE +/- 0.41, N = 3SE +/- 0.71, N = 3SE +/- 0.63, N = 3193.51193.97194.28344.93345.92805.17806.74811.19

FluidX3D

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.17Test: FP32-FP16CRTX 2000 Ada Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation2K4K6K8K10KSE +/- 1.45, N = 3SE +/- 1.20, N = 3SE +/- 1.00, N = 3SE +/- 0.33, N = 3SE +/- 3.50, N = 425292530253040644066105451058410585

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 Bandwidth2a2RTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation5RTX 6000 Ada Generation42004006008001000SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.19, N = 3SE +/- 0.03, N = 11195.05195.10195.11307.31307.47815.88815.91816.121. (CXX) g++ options: -O3

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 Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation452004006008001000SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 6207.71207.71207.71326.92326.93865.57866.19866.471. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Barbershop - Compute: NVIDIA CUDA22aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation4080120160200SE +/- 0.15, N = 3SE +/- 0.27, N = 3SE +/- 0.15, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3189.15189.04188.78108.14108.0845.7745.6645.65

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 Generation2aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation45510152025SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 34.604.617.967.9718.9519.0019.00MIN: 4.38 / MAX: 4.75MIN: 4.41 / MAX: 4.79MIN: 7.65 / MAX: 8.16MIN: 7.76 / MAX: 8.16MIN: 18.15 / MAX: 19.18MIN: 18.23 / MAX: 19.15MIN: 18.27 / MAX: 19.15

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: BMW27 - Compute: NVIDIA CUDA2aRTX 2000 Ada Generation2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation54510152025SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 722.5522.5222.4913.2613.255.575.555.55

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 Generation2a2NVIDIA RTX 4000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation542004006008001000SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.26, N = 6211.82211.82211.84333.67333.72850.04860.03860.251. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

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 RTX GPU2aRTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation2K4K6K8K10KSE +/- 6.06, N = 3SE +/- 3.67, N = 3SE +/- 10.33, N = 3SE +/- 8.95, N = 32023203029272937815981598174

FluidX3D

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.17Test: FP32-FP32RTX 2000 Ada Generation2a2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation4511002200330044005500SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 16.33, N = 313311331133220462047525252675268

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 Generation2aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation4548121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.06, N = 34.634.638.048.0517.7817.7917.81MIN: 1.91 / MAX: 5.25MIN: 1.91 / MAX: 5.26MIN: 3.16 / MAX: 9.09MIN: 3.52 / MAX: 9.09MIN: 7.98 / MAX: 20.82MIN: 8.01 / MAX: 20.68MIN: 8.01 / MAX: 20.74

ParaView

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Many Spheres - Frames: 3000 - Resolution: 3840 x 2160RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation4080120160200SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 343.1643.2043.2888.5788.93157.73157.74158.47

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Classroom - Compute: NVIDIA OptiXRTX 2000 Ada Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 6000 Ada Generation45714212835SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 628.6128.5928.5517.9117.827.927.857.82

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: SNX-04RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation542004006008001000SE +/- 0.21, N = 3SE +/- 0.16, N = 3SE +/- 0.24, N = 3SE +/- 0.12, N = 3SE +/- 5.90, N = 3285.10285.57286.31423.85423.871026.561029.371029.44

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 Generation2aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation5RTX 6000 Ada Generation4714212835SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.006, N = 3SE +/- 0.029, N = 38.1508.15712.86412.89029.25829.28229.302

ParaView

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Many Spheres - Frames: 3000 - Resolution: 2560 x 1440RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation4080120160200SE +/- 0.18, N = 3SE +/- 0.13, N = 3SE +/- 0.16, N = 3SE +/- 0.05, N = 3SE +/- 0.74, N = 347.1247.1947.3196.6196.77165.37165.54167.61

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX2RTX 2000 Ada Generation2aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation4RTX 6000 Ada Generation5714212835SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 531.2031.1831.1419.4419.388.888.868.82

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Barbershop - Compute: NVIDIA OptiX2a2RTX 2000 Ada GenerationRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation5RTX 6000 Ada Generation4306090120150SE +/- 0.12, N = 3SE +/- 0.17, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3114.67114.64114.4372.3972.1032.9432.9132.83

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Junkshop - Compute: NVIDIA CUDA2RTX 2000 Ada Generation2aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation5RTX 6000 Ada Generation4918273645SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 438.4638.3538.2922.9922.9111.1211.0811.07

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: MEDICAL-O3RTX 2000 Ada Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation5RTX 6000 Ada Generation450100150200250SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.21, N = 365.4965.6065.7397.3697.38211.55211.98212.24

OpenBenchmarking.orgComposite Score, More Is BetterSPECViewPerf 2020 3.0Resolution: 2560 x 1440 - Viewset: SOLIDWORKS-07RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation54130260390520650SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.20, N = 3184.60184.62184.99308.68308.76578.10579.41579.49

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 Generation2aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation4RTX 6000 Ada Generation548121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 35.215.217.697.7115.5515.8615.89MIN: 4.17 / MAX: 6.54MIN: 4.19 / MAX: 6.52MIN: 6 / MAX: 9.96MIN: 6.04 / MAX: 9.86MIN: 13.62 / MAX: 21.43MIN: 13.68 / MAX: 21.62MIN: 13.71 / MAX: 21.62

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Junkshop - Compute: NVIDIA OptiXRTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation510152025SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 622.3522.1922.1114.5914.547.377.347.34

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 Generation2aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation54RTX 6000 Ada Generation1632486480SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 324.3924.4036.8936.9473.7873.8173.91

ParaView

OpenBenchmarking.orgFrames / Sec, More Is BetterParaView 5.13Test: Wavelet Volume - Frames: 3000 - Resolution: 2560 x 1440RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation4RTX 6000 Ada Generation52004006008001000SE +/- 0.33, N = 3SE +/- 0.97, N = 3SE +/- 0.37, N = 3SE +/- 1.78, N = 3SE +/- 2.56, N = 3286.61287.09287.49468.30474.47846.86853.68863.24

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: GPU2aRTX 2000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation54816243240SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 711.9511.9617.5717.5935.6535.7335.77MIN: 11.18 / MAX: 12.57MIN: 11.2 / MAX: 12.57MIN: 16.25 / MAX: 18.2MIN: 16.25 / MAX: 18.22MIN: 31.82 / MAX: 39.14MIN: 31.93 / MAX: 37.41MIN: 31.96 / MAX: 37.45

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: MAYA-06RTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation45RTX 6000 Ada Generation2004006008001000SE +/- 0.08, N = 3SE +/- 0.16, N = 3SE +/- 0.16, N = 3SE +/- 0.06, N = 3SE +/- 0.66, N = 3276.25276.58277.14430.20430.48812.16812.50813.18

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: BMW27 - Compute: NVIDIA OptiXRTX 2000 Ada Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation45RTX 6000 Ada Generation3691215SE +/- 0.08, N = 10SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 14SE +/- 0.00, N = 810.5210.4610.447.247.223.623.613.61

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-062aRTX 2000 Ada Generation2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation5RTX 6000 Ada Generation450100150200250SE +/- 0.04, N = 3SE +/- 0.49, N = 3SE +/- 0.30, N = 3SE +/- 0.04, N = 3SE +/- 1.01, N = 383.5783.8783.90141.84143.77236.74236.96237.35

FinanceBench

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

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Monte-Carlo OpenCL2RTX 2000 Ada Generation2aNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation54RTX 6000 Ada Generation60120180240300SE +/- 0.29, N = 3SE +/- 0.13, N = 3SE +/- 0.19, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 7266.16265.88265.86200.21199.8399.3599.3299.241. (CXX) g++ options: -O3 -march=native -fopenmp

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 Generation22aRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation54RTX 6000 Ada Generation60120180240300SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.16, N = 3SE +/- 0.47, N = 3113.93114.14114.14170.61170.79255.97256.36256.46

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 2.3.22aRTX 2000 Ada Generation2NVIDIA RTX 4000 Ada GenerationRTX 4000 Ada GenerationRTX 6000 Ada Generation45100200300400500SE +/- 0.51, N = 3SE +/- 0.29, N = 3SE +/- 0.12, N = 3SE +/- 0.67, N = 3SE +/- 0.94, N = 3207.66207.95208.02282.15282.16462.31462.98467.31

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: INT64 Compute2a2RTX 2000 Ada GenerationNVIDIA RTX 4000 Ada GenerationRTX 4000 Ada Generation4RTX 6000 Ada Generation50.85771.71542.57313.43084.2885SE +/- 0.003, N = 3SE +/- 0.002, N = 3SE +/- 0.002, N = 3SE +/- 0.088, N = 3SE +/- 0.021, N = 61.7281.7321.7502.8242.8753.6933.7743.8121. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

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: Transfer Bandwidth enqueueReadBufferRTX 2000 Ada Generation22aNVIDIA RTX 4000 Ada Generation4RTX 4000 Ada GenerationRTX 6000 Ada Generation5612182430SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.26, N = 15SE +/- 0.01, N = 312.9212.9312.9321.7522.6622.6922.8723.861. (CXX) g++ options: -O3

OpenBenchmarking.orgGBPS, More Is Betterclpeak 1.1.2OpenCL Test: Transfer Bandwidth enqueueWriteBufferRTX 2000 Ada Generation22a4NVIDIA RTX 4000 Ada GenerationRTX 6000 Ada GenerationRTX 4000 Ada Generation5612182430SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 313.1013.1013.1021.8921.9022.1122.2022.981. (CXX) g++ options: -O3

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 MyocyteRTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation2RTX 2000 Ada Generation2a45RTX 6000 Ada Generation510152025SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.00, N = 321.6821.5120.8220.7620.6920.3120.2720.161. (CXX) g++ options: -O2 -lOpenCL

clpeak

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

OpenBenchmarking.orgus, Fewer Is Betterclpeak 1.1.2OpenCL Test: Kernel Latency5RTX 6000 Ada Generation2a4RTX 2000 Ada Generation2RTX 4000 Ada GenerationNVIDIA RTX 4000 Ada Generation0.85281.70562.55843.41124.264SE +/- 0.01, N = 14SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 33.793.793.783.783.773.763.733.701. (CXX) g++ options: -O3

Meta Performance Per Watts

OpenBenchmarking.orgPerformance Per Watts, More Is BetterMeta Performance Per WattsPerformance Per WattsRTX 6000 Ada Generation2004006008001000983.03

GPU Temperature Monitor

OpenBenchmarking.orgCelsiusGPU Temperature MonitorPhoronix Test Suite System MonitoringRTX 6000 Ada Generation1632486480Min: 41 / Avg: 72.97 / Max: 87

GPU Power Consumption Monitor

OpenBenchmarking.orgWattsGPU Power Consumption MonitorPhoronix Test Suite System MonitoringRTX 6000 Ada Generation50100150200250Min: 21.72 / Avg: 191 / Max: 303.57

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

Test: tConvolve OpenCL

RTX 6000 Ada Generation: The test run did not produce a result.

4: The test run did not produce a result.

5: The test run did not produce a result.

RTX 4000 Ada Generation: The test run did not produce a result.

NVIDIA RTX 4000 Ada Generation: The test run did not produce a result.

RTX 2000 Ada Generation: The test run did not produce a result.

2: The test run did not produce a result.

2a: The test run did not produce a result.

ArrayFire

Test: Neural Network OpenCL FP16

RTX 6000 Ada Generation: The test run did not produce a result. E: ./arrayfire: 7: ./neural_network_opencl: not found

4: The test run did not produce a result. E: ./arrayfire: 7: ./neural_network_opencl: not found

5: The test run did not produce a result. E: ./arrayfire: 7: ./neural_network_opencl: not found

RTX 4000 Ada Generation: The test run did not produce a result. E: ./arrayfire: 7: ./neural_network_opencl: not found

NVIDIA RTX 4000 Ada Generation: The test run did not produce a result. E: ./arrayfire: 7: ./neural_network_opencl: not found

RTX 2000 Ada Generation: The test run did not produce a result. E: ./arrayfire: 7: ./neural_network_opencl: not found

2: The test run did not produce a result. E: ./arrayfire: 7: ./neural_network_opencl: not found

2a: The test run did not produce a result. E: ./arrayfire: 7: ./neural_network_opencl: not found