compulab-airtop-3-rtx-4000-compute

Intel Xeon E-2288G testing with a Compulab SBC-ATCFL v1.2 (ATOP3.PRD.0.29.2 BIOS) and NVIDIA Quadro RTX 4000 8GB on Ubuntu 20.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2010311-FI-COMPULABA24
Jump To Table - Results

View

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

Limit displaying results to tests within:

CPU Massive 2 Tests
Creator Workloads 2 Tests
HPC - High Performance Computing 3 Tests
Machine Learning 2 Tests
Multi-Core 3 Tests
NVIDIA GPU Compute 16 Tests
OpenCL 4 Tests
Vulkan Compute 4 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
1
October 30 2020
  9 Minutes
1a
October 30 2020
  9 Minutes
2
October 30 2020
 
1b
October 30 2020
  9 Minutes
1c
October 30 2020
  6 Minutes
1d
October 30 2020
  1 Hour, 19 Minutes
1e
October 30 2020
  3 Hours, 54 Minutes
NVIDIA Quadro RTX 4000
October 30 2020
  3 Hours, 48 Minutes
RTX 4000
October 30 2020
  3 Hours, 47 Minutes
NVIDIA RTX 4000
October 31 2020
  28 Minutes
Invert Hiding All Results Option
  1 Hour, 23 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):


compulab-airtop-3-rtx-4000-compute ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen Resolution11a21b1c1d1eNVIDIA Quadro RTX 4000RTX 4000NVIDIA RTX 4000Intel Xeon E-2288G @ 5.00GHz (8 Cores / 16 Threads)Compulab SBC-ATCFL v1.2 (ATOP3.PRD.0.29.2 BIOS)Intel Cannon Lake PCH64GBSamsung SSD 970 EVO Plus 250GBNVIDIA Quadro RTX 4000 8GB (1005/6500MHz)Intel Cannon Lake PCH cAVSVE228Intel I219-LM + Intel I210Ubuntu 20.105.8.0-26-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.9NVIDIA 455.284.6.0OpenCL 1.2 CUDA 11.1.961.2.142GCC 10.2.0ext41920x1080NVIDIA Quadro RTX 4000 8GB (300/405MHz)NVIDIA Quadro RTX 4000 8GB (1005/6500MHz)NVIDIA Quadro RTX 4000 8GB (300/405MHz)NVIDIA Quadro RTX 4000 8GB (1005/6500MHz)OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd6 - Thermald 2.3OpenCL Details- GPU Compute Cores: 2304Python Details- 1, 1a, 1b, 1d, 1e, NVIDIA Quadro RTX 4000, RTX 4000, NVIDIA RTX 4000: Python 3.8.6Security Details- itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Mitigation of TSX disabled + tsx_async_abort: Mitigation of TSX disabled

compulab-airtop-3-rtx-4000-compute realsr-ncnn: 4x - Yesclpeak: Single-Precision Floathashcat: 7-Ziphashcat: MD5realsr-ncnn: 4x - Nowaifu2x-ncnn: 2x - 3 - Yeshashcat: SHA1hashcat: TrueCrypt RIPEMD160 + XTShashcat: SHA-512clpeak: Integer Compute INTncnn: Vulkan GPU - resnet50vkfft: ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - vgg16redshift: blender: Classroom - CUDAncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - shufflenet-v2blender: Classroom - NVIDIA OptiXcl-mem: Writeblender: Barbershop - CUDAncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - blazefacecl-mem: Copyclpeak: Global Memory Bandwidthncnn: Vulkan GPU - mnasnetblender: Fishy Cat - NVIDIA OptiXplaidml: No - Inference - DenseNet 201 - OpenCLblender: Pabellon Barcelona - NVIDIA OptiXblender: Barbershop - NVIDIA OptiXblender: Fishy Cat - CUDAncnn: Vulkan GPU - efficientnet-b0blender: BMW27 - CUDAplaidml: No - Inference - Mobilenet - OpenCLplaidml: Yes - Inference - Mobilenet - OpenCLncnn: Vulkan GPU - squeezenetviennacl: OpenCL LU Factorizationplaidml: No - Inference - IMDB LSTM - OpenCLfahbench: mandelgpu: GPUarrayfire: Conjugate Gradient OpenCLblender: Pabellon Barcelona - CUDAcl-mem: Readclpeak: Double-Precision Doublefinancebench: Black-Scholes OpenCLncnn: Vulkan GPU-v2-v2 - mobilenet-v2neatbench: GPUblender: BMW27 - NVIDIA OptiXncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - mobilenetluxcorerender-cl: Rainbow Colors and Prismluxcorerender-cl: LuxCore Benchmarkluxcorerender-cl: Foodluxcorerender-cl: DLSC11a21b1c1d1eNVIDIA Quadro RTX 4000RTX 4000NVIDIA RTX 400080.4454469002512196666712.5525.5028704500000327233110226666725694325.5283.068.2883379.714.03481.1294443002494470000012.6045.5518642000000326300109526666725585320.4282.668.4737379.214.03281.5444426672483896666712.6655.5728615933333324033109103333325486321.1282.168.2546379.214.03782.46412.7015.67184.1744333002425903333312.9785.6948426866667317867107006666725027382321.1281.168.3795191.6264379.314.03610.423.401.523.9981.1746033.104430002487690000012.5805.571863350000032426710922000005712.253.89254572.188.79381218.663.321.33115.78322.5756.641.710.63282.1346.091.5258.38140.54160.821307.24112.742.7357.801490.381843.843.7768.4059423.50191.8417248122412.92.247459.35379.3259.6614.0341.4831.032.148.621.84.6610.783.501.574.0985.9266004.484246332384096666713.0785.785825403333331223310497333335741.923.93246842.219.14391224.853.401.35118.39318.4771.191.740.63278.7340.911.5459.12138.91160.951321.89114.002.7558.101475.981834.203.7868.0188421.03190.7594248177151.42.255460.64379.3259.5014.0341.4830.929.238.331.954.6710.793.471.554.0187.7936536.454177672350686666713.3505.861818156666730940010415000006013.594.08245382.279.02393223.813.371.36117.20319.4764.301.740.64278.8342.371.5458.69139.15159.091307.37113.882.7658.411480.201829.243.8068.0204420.80190.8199246857018.02.257459.08379.3259.3314.0341.4830.229.148.221.774.8610.733.471.554.0280.5494464002507546666712.4945.5408700533333327567110043333325593379323.0282.168.2894379.314.032OpenBenchmarking.org

RealSR-NCNN

RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1c1b1a120406080100SE +/- 0.27, N = 3SE +/- 0.37, N = 3SE +/- 0.45, N = 3SE +/- 0.36, N = 3SE +/- 0.36, N = 3SE +/- 0.40, N = 3SE +/- 0.40, N = 3SE +/- 0.36, N = 3SE +/- 0.35, N = 387.7980.5585.9381.1784.1782.4681.5481.1380.45
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1c1b1a120406080100Min: 87.29 / Avg: 87.79 / Max: 88.22Min: 79.88 / Avg: 80.55 / Max: 81.15Min: 85.1 / Avg: 85.93 / Max: 86.64Min: 80.53 / Avg: 81.17 / Max: 81.79Min: 83.5 / Avg: 84.17 / Max: 84.74Min: 81.75 / Avg: 82.46 / Max: 83.12Min: 80.83 / Avg: 81.54 / Max: 82.22Min: 80.49 / Avg: 81.13 / Max: 81.73Min: 79.84 / Avg: 80.45 / Max: 81.04

clpeak

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

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatRTX 4000NVIDIA Quadro RTX 40001e14002800420056007000SE +/- 97.61, N = 3SE +/- 55.06, N = 3SE +/- 35.64, N = 36536.456004.486033.101. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatRTX 4000NVIDIA Quadro RTX 40001e11002200330044005500Min: 6378.8 / Avg: 6536.45 / Max: 6715Min: 5897.91 / Avg: 6004.48 / Max: 6081.8Min: 5996.87 / Avg: 6033.1 / Max: 6104.381. (CXX) g++ options: -O3 -rdynamic -lOpenCL

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: 7-ZipRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a1100K200K300K400K500KSE +/- 233.33, N = 3SE +/- 200.00, N = 3SE +/- 463.08, N = 3SE +/- 1365.04, N = 3SE +/- 321.46, N = 3SE +/- 202.76, N = 3SE +/- 556.78, N = 3SE +/- 208.17, N = 3417767446400424633443000433300442667444300446900
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: 7-ZipRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a180K160K240K320K400KMin: 417300 / Avg: 417766.67 / Max: 418000Min: 446000 / Avg: 446400 / Max: 446600Min: 423800 / Avg: 424633.33 / Max: 425400Min: 440300 / Avg: 443000 / Max: 444700Min: 432800 / Avg: 433300 / Max: 433900Min: 442300 / Avg: 442666.67 / Max: 443000Min: 443200 / Avg: 444300 / Max: 445000Min: 446500 / Avg: 446900 / Max: 447200

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: MD5RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a15000M10000M15000M20000M25000MSE +/- 2355372.11, N = 3SE +/- 24626025.08, N = 3SE +/- 13574649.58, N = 3SE +/- 12698162.60, N = 3SE +/- 2643440.52, N = 3SE +/- 12651789.51, N = 3SE +/- 2051828.45, N = 3SE +/- 25031801.99, N = 32350686666725075466667238409666672487690000024259033333248389666672494470000025121966667
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: MD5RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a14000M8000M12000M16000M20000MMin: 23503100000 / Avg: 23506866666.67 / Max: 23511200000Min: 25032500000 / Avg: 25075466666.67 / Max: 25117800000Min: 23823500000 / Avg: 23840966666.67 / Max: 23867700000Min: 24853600000 / Avg: 24876900000 / Max: 24897300000Min: 24253800000 / Avg: 24259033333.33 / Max: 24262300000Min: 24814400000 / Avg: 24838966666.67 / Max: 24856500000Min: 24942500000 / Avg: 24944700000 / Max: 24948800000Min: 25086500000 / Avg: 25121966666.67 / Max: 25170300000

RealSR-NCNN

RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1c1b1a13691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 313.3512.4913.0812.5812.9812.7012.6712.6012.55
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1c1b1a148121620Min: 13.32 / Avg: 13.35 / Max: 13.37Min: 12.47 / Avg: 12.49 / Max: 12.52Min: 13.05 / Avg: 13.08 / Max: 13.12Min: 12.57 / Avg: 12.58 / Max: 12.6Min: 12.97 / Avg: 12.98 / Max: 12.99Min: 12.66 / Avg: 12.7 / Max: 12.77Min: 12.59 / Avg: 12.66 / Max: 12.71Min: 12.57 / Avg: 12.6 / Max: 12.63Min: 12.53 / Avg: 12.55 / Max: 12.58

Waifu2x-NCNN Vulkan

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1c1b1a11.31872.63743.95615.27486.5935SE +/- 0.014, N = 3SE +/- 0.041, N = 3SE +/- 0.018, N = 3SE +/- 0.017, N = 3SE +/- 0.008, N = 3SE +/- 0.047, N = 3SE +/- 0.022, N = 3SE +/- 0.023, N = 3SE +/- 0.008, N = 35.8615.5405.7855.5715.6945.6715.5725.5515.502
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1c1b1a1246810Min: 5.84 / Avg: 5.86 / Max: 5.89Min: 5.49 / Avg: 5.54 / Max: 5.62Min: 5.76 / Avg: 5.79 / Max: 5.82Min: 5.55 / Avg: 5.57 / Max: 5.6Min: 5.68 / Avg: 5.69 / Max: 5.71Min: 5.61 / Avg: 5.67 / Max: 5.76Min: 5.53 / Avg: 5.57 / Max: 5.6Min: 5.51 / Avg: 5.55 / Max: 5.59Min: 5.49 / Avg: 5.5 / Max: 5.51

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA1RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a12000M4000M6000M8000M10000MSE +/- 6590228.46, N = 3SE +/- 2630800.47, N = 3SE +/- 5691026.07, N = 3SE +/- 6005275.46, N = 3SE +/- 7846938.54, N = 3SE +/- 3773739.67, N = 3SE +/- 6847870.72, N = 3SE +/- 9832090.32, N = 381815666678700533333825403333386335000008426866667861593333386420000008704500000
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA1RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a11500M3000M4500M6000M7500MMin: 8170500000 / Avg: 8181566666.67 / Max: 8193300000Min: 8695400000 / Avg: 8700533333.33 / Max: 8704100000Min: 8247000000 / Avg: 8254033333.33 / Max: 8265300000Min: 8623000000 / Avg: 8633500000 / Max: 8643800000Min: 8415700000 / Avg: 8426866666.67 / Max: 8442000000Min: 8608400000 / Avg: 8615933333.33 / Max: 8620100000Min: 8633800000 / Avg: 8642000000 / Max: 8655600000Min: 8686600000 / Avg: 8704500000 / Max: 8720500000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: TrueCrypt RIPEMD160 + XTSRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a170K140K210K280K350KSE +/- 683.94, N = 3SE +/- 317.98, N = 3SE +/- 233.33, N = 3SE +/- 88.19, N = 3SE +/- 185.59, N = 3SE +/- 533.33, N = 3309400327567312233324267317867324033326300327233
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: TrueCrypt RIPEMD160 + XTSRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a160K120K180K240K300KMin: 326200 / Avg: 327566.67 / Max: 328300Min: 311600 / Avg: 312233.33 / Max: 312600Min: 323900 / Avg: 324266.67 / Max: 324700Min: 317700 / Avg: 317866.67 / Max: 318000Min: 323800 / Avg: 324033.33 / Max: 324400Min: 326700 / Avg: 327233.33 / Max: 328300

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA-512RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a1200M400M600M800M1000MSE +/- 953939.20, N = 3SE +/- 240370.09, N = 3SE +/- 1260070.54, N = 3SE +/- 1021436.90, N = 3SE +/- 1017076.42, N = 3SE +/- 643773.60, N = 3SE +/- 491030.66, N = 3SE +/- 819213.72, N = 310415000001100433333104973333310922000001070066667109103333310952666671102266667
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA-512RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a1200M400M600M800M1000MMin: 1039900000 / Avg: 1041500000 / Max: 1043200000Min: 1100100000 / Avg: 1100433333.33 / Max: 1100900000Min: 1047800000 / Avg: 1049733333.33 / Max: 1052100000Min: 1090300000 / Avg: 1092200000 / Max: 1093800000Min: 1068200000 / Avg: 1070066666.67 / Max: 1071700000Min: 1090200000 / Avg: 1091033333.33 / Max: 1092300000Min: 1094300000 / Avg: 1095266666.67 / Max: 1095900000Min: 1101000000 / Avg: 1102266666.67 / Max: 1103800000

clpeak

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

OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTRTX 4000NVIDIA Quadro RTX 40001e13002600390052006500SE +/- 102.19, N = 3SE +/- 46.13, N = 3SE +/- 68.26, N = 126013.595741.925712.251. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTRTX 4000NVIDIA Quadro RTX 40001e10002000300040005000Min: 5900.36 / Avg: 6013.59 / Max: 6217.55Min: 5678.83 / Avg: 5741.92 / Max: 5831.77Min: 5039.83 / Avg: 5712.25 / Max: 6051.861. (CXX) g++ options: -O3 -rdynamic -lOpenCL

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet50RTX 4000NVIDIA Quadro RTX 40001e0.9181.8362.7543.6724.59SE +/- 0.14, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 34.083.933.89MIN: 3.92 / MAX: 40.55MIN: 3.91 / MAX: 4.04MIN: 3.86 / MAX: 3.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet50RTX 4000NVIDIA Quadro RTX 40001e246810Min: 3.94 / Avg: 4.08 / Max: 4.36Min: 3.92 / Avg: 3.93 / Max: 3.94Min: 3.87 / Avg: 3.89 / Max: 3.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

VkFFT

VkFFT is a Fast Fourier Transform (FFT) Library that is GPU accelerated by means of the Vulkan API. The VkFFT benchmark runs FFT performance differences of many different sizes before returning an overall benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 2020-09-29RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a16K12K18K24K30KSE +/- 4.04, N = 3SE +/- 20.11, N = 3SE +/- 32.54, N = 3SE +/- 17.21, N = 3SE +/- 27.82, N = 3SE +/- 16.51, N = 3SE +/- 28.39, N = 32453825593246842545725027254862558525694
OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 2020-09-29RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a14K8K12K16K20KMin: 24533 / Avg: 24538 / Max: 24546Min: 25554 / Avg: 25593 / Max: 25621Min: 25412 / Avg: 25456.67 / Max: 25520Min: 25000 / Avg: 25027 / Max: 25059Min: 25450 / Avg: 25486.33 / Max: 25541Min: 25560 / Avg: 25584.67 / Max: 25616Min: 25639 / Avg: 25694.33 / Max: 25733

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: alexnetRTX 4000NVIDIA Quadro RTX 40001e0.51081.02161.53242.04322.554SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 32.272.212.18MIN: 2.15 / MAX: 23.82MIN: 1.91 / MAX: 6.96MIN: 1.91 / MAX: 11.431. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: alexnetRTX 4000NVIDIA Quadro RTX 40001e246810Min: 2.21 / Avg: 2.27 / Max: 2.33Min: 2.2 / Avg: 2.21 / Max: 2.22Min: 2.16 / Avg: 2.18 / Max: 2.211. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: vgg16RTX 4000NVIDIA Quadro RTX 40001e3691215SE +/- 0.04, N = 3SE +/- 0.13, N = 3SE +/- 0.04, N = 39.029.148.79MIN: 8.35 / MAX: 20.34MIN: 8.49 / MAX: 36.48MIN: 8.1 / MAX: 20.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: vgg16RTX 4000NVIDIA Quadro RTX 40001e3691215Min: 8.98 / Avg: 9.02 / Max: 9.1Min: 8.98 / Avg: 9.14 / Max: 9.39Min: 8.72 / Avg: 8.79 / Max: 8.841. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

RedShift Demo

This is a test of MAXON's RedShift demo build that currently requires NVIDIA GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRedShift Demo 3.0RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d90180270360450SE +/- 4.91, N = 3SE +/- 2.33, N = 3SE +/- 4.63, N = 3SE +/- 2.31, N = 3SE +/- 2.60, N = 3393379391381382
OpenBenchmarking.orgSeconds, Fewer Is BetterRedShift Demo 3.0RTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d70140210280350Min: 384 / Avg: 392.67 / Max: 401Min: 375 / Avg: 379.33 / Max: 383Min: 383 / Avg: 390.67 / Max: 399Min: 377 / Avg: 381 / Max: 385Min: 377 / Avg: 381.67 / Max: 386

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Classroom - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e50100150200250SE +/- 3.26, N = 3SE +/- 3.62, N = 3SE +/- 1.49, N = 3223.81224.85218.66
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Classroom - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e4080120160200Min: 218.62 / Avg: 223.81 / Max: 229.82Min: 219.03 / Avg: 224.85 / Max: 231.5Min: 216.02 / Avg: 218.66 / Max: 221.16

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: googlenetRTX 4000NVIDIA Quadro RTX 40001e0.7651.532.2953.063.825SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 33.373.403.32MIN: 3.35 / MAX: 3.44MIN: 3.33 / MAX: 20.26MIN: 3.29 / MAX: 3.431. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: googlenetRTX 4000NVIDIA Quadro RTX 40001e246810Min: 3.37 / Avg: 3.37 / Max: 3.37Min: 3.36 / Avg: 3.4 / Max: 3.44Min: 3.31 / Avg: 3.32 / Max: 3.321. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: shufflenet-v2RTX 4000NVIDIA Quadro RTX 40001e0.3060.6120.9181.2241.53SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.361.351.33MIN: 1.34 / MAX: 1.41MIN: 1.33 / MAX: 1.4MIN: 1.32 / MAX: 1.41. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: shufflenet-v2RTX 4000NVIDIA Quadro RTX 40001e246810Min: 1.35 / Avg: 1.36 / Max: 1.36Min: 1.35 / Avg: 1.35 / Max: 1.36Min: 1.33 / Avg: 1.33 / Max: 1.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Classroom - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e306090120150SE +/- 0.57, N = 3SE +/- 0.55, N = 3SE +/- 0.87, N = 3117.20118.39115.78
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Classroom - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e20406080100Min: 116.06 / Avg: 117.2 / Max: 117.88Min: 117.29 / Avg: 118.39 / Max: 118.95Min: 114.1 / Avg: 115.78 / Max: 117.01

cl-mem

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

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: WriteRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a170140210280350SE +/- 1.44, N = 3SE +/- 1.47, N = 3SE +/- 2.17, N = 3SE +/- 0.58, N = 3SE +/- 0.96, N = 3SE +/- 0.78, N = 3SE +/- 1.48, N = 3SE +/- 1.79, N = 3319.4323.0318.4322.5321.1321.1320.4325.51. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: WriteRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a160120180240300Min: 316.9 / Avg: 319.43 / Max: 321.9Min: 320.5 / Avg: 322.97 / Max: 325.6Min: 314.1 / Avg: 318.43 / Max: 320.8Min: 321.4 / Avg: 322.47 / Max: 323.4Min: 319.9 / Avg: 321.1 / Max: 323Min: 319.8 / Avg: 321.13 / Max: 322.5Min: 318.9 / Avg: 320.43 / Max: 323.4Min: 322.3 / Avg: 325.5 / Max: 328.51. (CC) gcc options: -O2 -flto -lOpenCL

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e170340510680850SE +/- 0.87, N = 3SE +/- 1.15, N = 3SE +/- 2.80, N = 3764.30771.19756.64
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e140280420560700Min: 762.69 / Avg: 764.3 / Max: 765.7Min: 768.95 / Avg: 771.19 / Max: 772.74Min: 751.24 / Avg: 756.64 / Max: 760.61

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3RTX 4000NVIDIA Quadro RTX 40001e0.39150.7831.17451.5661.9575SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.741.741.71MIN: 1.73 / MAX: 1.8MIN: 1.73 / MAX: 1.81MIN: 1.7 / MAX: 1.751. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3RTX 4000NVIDIA Quadro RTX 40001e246810Min: 1.74 / Avg: 1.74 / Max: 1.75Min: 1.74 / Avg: 1.74 / Max: 1.75Min: 1.71 / Avg: 1.71 / Max: 1.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: blazefaceRTX 4000NVIDIA Quadro RTX 40001e0.1440.2880.4320.5760.72SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 30.640.630.63MIN: 0.62 / MAX: 0.84MIN: 0.62 / MAX: 0.65MIN: 0.62 / MAX: 0.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: blazefaceRTX 4000NVIDIA Quadro RTX 40001e246810Min: 0.64 / Avg: 0.64 / Max: 0.64Min: 0.62 / Avg: 0.63 / Max: 0.64Min: 0.62 / Avg: 0.63 / Max: 0.631. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

cl-mem

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

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: CopyRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a160120180240300SE +/- 0.12, N = 3SE +/- 0.28, N = 3SE +/- 0.20, N = 3SE +/- 0.12, N = 3SE +/- 0.20, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.26, N = 3278.8282.1278.7282.1281.1282.1282.6283.01. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: CopyRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a150100150200250Min: 278.6 / Avg: 278.8 / Max: 279Min: 281.5 / Avg: 282.07 / Max: 282.4Min: 278.3 / Avg: 278.67 / Max: 279Min: 281.9 / Avg: 282.07 / Max: 282.3Min: 280.7 / Avg: 281.07 / Max: 281.4Min: 281.9 / Avg: 282.07 / Max: 282.3Min: 282.5 / Avg: 282.6 / Max: 282.7Min: 282.6 / Avg: 283 / Max: 283.51. (CC) gcc options: -O2 -flto -lOpenCL

clpeak

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

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory BandwidthRTX 4000NVIDIA Quadro RTX 40001e80160240320400SE +/- 5.13, N = 3SE +/- 4.44, N = 3SE +/- 4.72, N = 3342.37340.91346.091. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory BandwidthRTX 4000NVIDIA Quadro RTX 40001e60120180240300Min: 337.23 / Avg: 342.37 / Max: 352.63Min: 336.37 / Avg: 340.91 / Max: 349.79Min: 337.18 / Avg: 346.09 / Max: 353.271. (CXX) g++ options: -O3 -rdynamic -lOpenCL

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: mnasnetRTX 4000NVIDIA Quadro RTX 40001e0.34650.6931.03951.3861.7325SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.541.541.52MIN: 1.53 / MAX: 1.63MIN: 1.53 / MAX: 1.63MIN: 1.5 / MAX: 1.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: mnasnetRTX 4000NVIDIA Quadro RTX 40001e246810Min: 1.54 / Avg: 1.54 / Max: 1.54Min: 1.54 / Avg: 1.54 / Max: 1.54Min: 1.51 / Avg: 1.52 / Max: 1.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Fishy Cat - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e1326395265SE +/- 0.22, N = 3SE +/- 0.24, N = 3SE +/- 0.21, N = 358.6959.1258.38
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Fishy Cat - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e1224364860Min: 58.26 / Avg: 58.69 / Max: 58.95Min: 58.65 / Avg: 59.12 / Max: 59.38Min: 57.98 / Avg: 58.38 / Max: 58.69

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCLRTX 4000NVIDIA Quadro RTX 40001e306090120150SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.18, N = 3139.15138.91140.54
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCLRTX 4000NVIDIA Quadro RTX 40001e306090120150Min: 138.95 / Avg: 139.15 / Max: 139.35Min: 138.8 / Avg: 138.91 / Max: 139.12Min: 140.3 / Avg: 140.54 / Max: 140.89

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e4080120160200SE +/- 0.48, N = 3SE +/- 0.11, N = 3SE +/- 0.15, N = 3159.09160.95160.82
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e306090120150Min: 158.27 / Avg: 159.09 / Max: 159.93Min: 160.73 / Avg: 160.95 / Max: 161.06Min: 160.54 / Avg: 160.82 / Max: 161.07

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e30060090012001500SE +/- 0.53, N = 3SE +/- 1.05, N = 3SE +/- 4.68, N = 31307.371321.891307.24
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e2004006008001000Min: 1306.52 / Avg: 1307.37 / Max: 1308.35Min: 1319.86 / Avg: 1321.89 / Max: 1323.37Min: 1298.59 / Avg: 1307.24 / Max: 1314.65

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Fishy Cat - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e306090120150SE +/- 0.21, N = 3SE +/- 0.16, N = 3SE +/- 0.28, N = 3113.88114.00112.74
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Fishy Cat - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e20406080100Min: 113.48 / Avg: 113.88 / Max: 114.17Min: 113.69 / Avg: 114 / Max: 114.22Min: 112.24 / Avg: 112.74 / Max: 113.21

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: efficientnet-b0RTX 4000NVIDIA Quadro RTX 40001e0.6211.2421.8632.4843.105SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 32.762.752.73MIN: 2.75 / MAX: 3.23MIN: 2.74 / MAX: 3.38MIN: 2.7 / MAX: 8.241. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: efficientnet-b0RTX 4000NVIDIA Quadro RTX 40001e246810Min: 2.76 / Avg: 2.76 / Max: 2.76Min: 2.75 / Avg: 2.75 / Max: 2.76Min: 2.72 / Avg: 2.73 / Max: 2.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: BMW27 - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e1326395265SE +/- 0.10, N = 3SE +/- 0.14, N = 3SE +/- 0.09, N = 358.4158.1057.80
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: BMW27 - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e1224364860Min: 58.2 / Avg: 58.41 / Max: 58.54Min: 57.85 / Avg: 58.1 / Max: 58.32Min: 57.66 / Avg: 57.8 / Max: 57.97

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCLRTX 4000NVIDIA Quadro RTX 40001e30060090012001500SE +/- 5.32, N = 3SE +/- 5.69, N = 3SE +/- 5.16, N = 31480.201475.981490.38
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCLRTX 4000NVIDIA Quadro RTX 40001e30060090012001500Min: 1471.62 / Avg: 1480.2 / Max: 1489.95Min: 1467.95 / Avg: 1475.98 / Max: 1486.99Min: 1484.37 / Avg: 1490.38 / Max: 1500.65

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCLRTX 4000NVIDIA Quadro RTX 40001e400800120016002000SE +/- 8.76, N = 3SE +/- 2.10, N = 3SE +/- 9.92, N = 31829.241834.201843.84
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCLRTX 4000NVIDIA Quadro RTX 40001e30060090012001500Min: 1812.38 / Avg: 1829.24 / Max: 1841.83Min: 1830.25 / Avg: 1834.2 / Max: 1837.4Min: 1832.86 / Avg: 1843.84 / Max: 1863.64

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: squeezenetRTX 4000NVIDIA Quadro RTX 40001e0.8551.712.5653.424.275SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 33.803.783.77MIN: 3.74 / MAX: 10.31MIN: 3.72 / MAX: 3.84MIN: 3.71 / MAX: 3.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: squeezenetRTX 4000NVIDIA Quadro RTX 40001e246810Min: 3.79 / Avg: 3.8 / Max: 3.81Min: 3.75 / Avg: 3.78 / Max: 3.79Min: 3.75 / Avg: 3.77 / Max: 3.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile uses ViennaCL OpenCL support and runs the included computational benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterViennaCL 1.4.2OpenCL LU FactorizationRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a11530456075SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.28, N = 3SE +/- 0.12, N = 3SE +/- 0.12, N = 368.0268.2968.0268.4168.3868.2568.4768.291. (CXX) g++ options: -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterViennaCL 1.4.2OpenCL LU FactorizationRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a11326395265Min: 67.93 / Avg: 68.02 / Max: 68.14Min: 68.27 / Avg: 68.29 / Max: 68.33Min: 67.99 / Avg: 68.02 / Max: 68.04Min: 68.37 / Avg: 68.41 / Max: 68.43Min: 68.28 / Avg: 68.38 / Max: 68.5Min: 67.7 / Avg: 68.25 / Max: 68.6Min: 68.3 / Avg: 68.47 / Max: 68.7Min: 68.06 / Avg: 68.29 / Max: 68.431. (CXX) g++ options: -rdynamic -lOpenCL

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCLRTX 4000NVIDIA Quadro RTX 40001e90180270360450SE +/- 0.32, N = 3SE +/- 1.26, N = 3SE +/- 0.45, N = 3420.80421.03423.50
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCLRTX 4000NVIDIA Quadro RTX 40001e80160240320400Min: 420.38 / Avg: 420.8 / Max: 421.43Min: 419.12 / Avg: 421.03 / Max: 423.41Min: 422.94 / Avg: 423.5 / Max: 424.38

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 4000NVIDIA Quadro RTX 40001e1d4080120160200SE +/- 0.37, N = 3SE +/- 0.27, N = 3SE +/- 0.32, N = 3SE +/- 0.40, N = 3190.82190.76191.84191.63
OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2RTX 4000NVIDIA Quadro RTX 40001e1d4080120160200Min: 190.42 / Avg: 190.82 / Max: 191.57Min: 190.26 / Avg: 190.76 / Max: 191.18Min: 191.26 / Avg: 191.84 / Max: 192.37Min: 191.09 / Avg: 191.63 / Max: 192.42

MandelGPU

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

OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPURTX 4000NVIDIA Quadro RTX 40001e50M100M150M200M250MSE +/- 540319.59, N = 3SE +/- 308768.05, N = 3SE +/- 711502.39, N = 3246857018.0248177151.4248122412.91. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPURTX 4000NVIDIA Quadro RTX 40001e40M80M120M160M200MMin: 246299220.2 / Avg: 246857017.97 / Max: 247937466.6Min: 247673467.9 / Avg: 248177151.4 / Max: 248738412.5Min: 247045758.1 / Avg: 248122412.93 / Max: 249466542.91. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

ArrayFire

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

OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient OpenCLRTX 4000NVIDIA Quadro RTX 40001e0.50781.01561.52342.03122.539SE +/- 0.007, N = 3SE +/- 0.012, N = 3SE +/- 0.008, N = 32.2572.2552.2471. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient OpenCLRTX 4000NVIDIA Quadro RTX 40001e246810Min: 2.25 / Avg: 2.26 / Max: 2.27Min: 2.24 / Avg: 2.26 / Max: 2.28Min: 2.23 / Avg: 2.25 / Max: 2.261. (CXX) g++ options: -rdynamic

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Pabellon Barcelona - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e100200300400500SE +/- 0.47, N = 3SE +/- 1.47, N = 3SE +/- 1.17, N = 3459.08460.64459.35
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Pabellon Barcelona - Compute: CUDARTX 4000NVIDIA Quadro RTX 40001e80160240320400Min: 458.58 / Avg: 459.08 / Max: 460.03Min: 457.89 / Avg: 460.64 / Max: 462.93Min: 457.44 / Avg: 459.35 / Max: 461.49

cl-mem

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

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: ReadRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a180160240320400SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3379.3379.3379.3379.3379.3379.2379.2379.71. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: ReadRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a170140210280350Min: 379.3 / Avg: 379.3 / Max: 379.3Min: 379.2 / Avg: 379.27 / Max: 379.3Min: 379.3 / Avg: 379.3 / Max: 379.3Min: 379.2 / Avg: 379.27 / Max: 379.3Min: 379.3 / Avg: 379.3 / Max: 379.3Min: 379.2 / Avg: 379.23 / Max: 379.3Min: 379.1 / Avg: 379.2 / Max: 379.3Min: 379.7 / Avg: 379.73 / Max: 379.81. (CC) gcc options: -O2 -flto -lOpenCL

clpeak

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

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleRTX 4000NVIDIA Quadro RTX 40001e60120180240300SE +/- 0.02, N = 3SE +/- 0.18, N = 3SE +/- 0.31, N = 3259.33259.50259.661. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleRTX 4000NVIDIA Quadro RTX 40001e50100150200250Min: 259.31 / Avg: 259.33 / Max: 259.37Min: 259.3 / Avg: 259.5 / Max: 259.86Min: 259.32 / Avg: 259.66 / Max: 260.281. (CXX) g++ options: -O3 -rdynamic -lOpenCL

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-06-06Benchmark: Black-Scholes OpenCLRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a148121620SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 314.0314.0314.0314.0314.0414.0414.0314.031. (CXX) g++ options: -O3 -lOpenCL
OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-06-06Benchmark: Black-Scholes OpenCLRTX 4000NVIDIA RTX 4000NVIDIA Quadro RTX 40001e1d1b1a148121620Min: 14.03 / Avg: 14.03 / Max: 14.04Min: 14.03 / Avg: 14.03 / Max: 14.04Min: 14.03 / Avg: 14.03 / Max: 14.04Min: 14.03 / Avg: 14.03 / Max: 14.04Min: 14.03 / Avg: 14.04 / Max: 14.04Min: 14.03 / Avg: 14.04 / Max: 14.04Min: 14.03 / Avg: 14.03 / Max: 14.03Min: 14.03 / Avg: 14.03 / Max: 14.041. (CXX) g++ options: -O3 -lOpenCL

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2RTX 4000NVIDIA Quadro RTX 40001e0.3330.6660.9991.3321.665SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 31.481.481.48MIN: 1.47 / MAX: 1.54MIN: 1.46 / MAX: 1.5MIN: 1.44 / MAX: 20.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2RTX 4000NVIDIA Quadro RTX 40001e246810Min: 1.48 / Avg: 1.48 / Max: 1.48Min: 1.47 / Avg: 1.48 / Max: 1.48Min: 1.45 / Avg: 1.48 / Max: 1.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NeatBench

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

OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: GPURTX 4000NVIDIA Quadro RTX 40001e714212835SE +/- 0.63, N = 15SE +/- 0.69, N = 15SE +/- 0.66, N = 1530.230.931.0
OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: GPURTX 4000NVIDIA Quadro RTX 40001e714212835Min: 28.5 / Avg: 30.16 / Max: 34.7Min: 28.5 / Avg: 30.9 / Max: 35.3Min: 28.5 / Avg: 30.97 / Max: 35.4

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: BMW27 - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e714212835SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 3.24, N = 1529.1429.2332.14
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: BMW27 - Compute: NVIDIA OptiXRTX 4000NVIDIA Quadro RTX 40001e714212835Min: 29.03 / Avg: 29.14 / Max: 29.2Min: 29.05 / Avg: 29.23 / Max: 29.33Min: 28.81 / Avg: 32.14 / Max: 77.56

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: yolov4-tinyRTX 4000NVIDIA Quadro RTX 40001e246810SE +/- 0.00, N = 3SE +/- 0.09, N = 3SE +/- 0.37, N = 38.228.338.62MIN: 8.15 / MAX: 8.57MIN: 8.13 / MAX: 55.28MIN: 8.1 / MAX: 74.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: yolov4-tinyRTX 4000NVIDIA Quadro RTX 40001e3691215Min: 8.21 / Avg: 8.22 / Max: 8.22Min: 8.21 / Avg: 8.33 / Max: 8.51Min: 8.23 / Avg: 8.62 / Max: 9.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet18RTX 4000NVIDIA Quadro RTX 40001e0.43880.87761.31641.75522.194SE +/- 0.04, N = 3SE +/- 0.13, N = 3SE +/- 0.05, N = 21.771.951.80MIN: 1.71 / MAX: 24.18MIN: 1.7 / MAX: 20.49MIN: 1.69 / MAX: 21.821. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet18RTX 4000NVIDIA Quadro RTX 40001e246810Min: 1.73 / Avg: 1.77 / Max: 1.85Min: 1.8 / Avg: 1.95 / Max: 2.21Min: 1.7 / Avg: 1.75 / Max: 1.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: mobilenetRTX 4000NVIDIA Quadro RTX 40001e1.09352.1873.28054.3745.4675SE +/- 0.17, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 34.864.674.66MIN: 4.64 / MAX: 71.23MIN: 4.64 / MAX: 4.75MIN: 4.6 / MAX: 4.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: mobilenetRTX 4000NVIDIA Quadro RTX 40001e246810Min: 4.68 / Avg: 4.86 / Max: 5.2Min: 4.66 / Avg: 4.67 / Max: 4.68Min: 4.63 / Avg: 4.66 / Max: 4.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

LuxCoreRender OpenCL

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

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: Rainbow Colors and PrismRTX 4000NVIDIA Quadro RTX 40001e1d3691215SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.34, N = 1210.7310.7910.7810.42MIN: 9.75 / MAX: 11.24MIN: 10.45 / MAX: 11.21MIN: 10.09 / MAX: 11.23MIN: 3.45 / MAX: 11.19
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: Rainbow Colors and PrismRTX 4000NVIDIA Quadro RTX 40001e1d3691215Min: 10.71 / Avg: 10.73 / Max: 10.77Min: 10.68 / Avg: 10.79 / Max: 10.86Min: 10.75 / Avg: 10.78 / Max: 10.82Min: 6.73 / Avg: 10.42 / Max: 10.86

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: LuxCore BenchmarkRTX 4000NVIDIA Quadro RTX 40001e1d0.78751.5752.36253.153.9375SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 123.473.473.503.40MIN: 0.33 / MAX: 3.96MIN: 0.27 / MAX: 3.96MIN: 0.27 / MAX: 4MIN: 0.17 / MAX: 3.97
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: LuxCore BenchmarkRTX 4000NVIDIA Quadro RTX 40001e1d246810Min: 3.43 / Avg: 3.47 / Max: 3.5Min: 3.46 / Avg: 3.47 / Max: 3.5Min: 3.46 / Avg: 3.5 / Max: 3.52Min: 2.69 / Avg: 3.4 / Max: 3.51

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: FoodRTX 4000NVIDIA Quadro RTX 40001e1d0.35330.70661.05991.41321.7665SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 121.551.551.571.52MIN: 0.26 / MAX: 1.86MIN: 0.25 / MAX: 1.85MIN: 0.26 / MAX: 1.89MIN: 0.14 / MAX: 1.88
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: FoodRTX 4000NVIDIA Quadro RTX 40001e1d246810Min: 1.53 / Avg: 1.55 / Max: 1.56Min: 1.53 / Avg: 1.55 / Max: 1.56Min: 1.55 / Avg: 1.57 / Max: 1.58Min: 1.07 / Avg: 1.52 / Max: 1.58

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: DLSCRTX 4000NVIDIA Quadro RTX 40001e1d0.92031.84062.76093.68124.6015SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 124.024.014.093.99MIN: 3.82 / MAX: 4.2MIN: 3.83 / MAX: 4.21MIN: 3.82 / MAX: 4.25MIN: 1.12 / MAX: 4.22
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: DLSCRTX 4000NVIDIA Quadro RTX 40001e1d246810Min: 4.02 / Avg: 4.02 / Max: 4.03Min: 3.99 / Avg: 4.01 / Max: 4.05Min: 4.08 / Avg: 4.09 / Max: 4.11Min: 3.09 / Avg: 3.99 / Max: 4.09