NVIDIA GPU Compute

AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3402 BIOS) and NVIDIA TITAN RTX 24GB on Ubuntu 21.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2106042-IB-TITAN666028
Jump To Table - Results

Statistics

Remove Outliers Before Calculating Averages

Graph Settings

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

Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
TITAN RTX
June 04 2021
  4 Hours, 35 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


NVIDIA GPU ComputeOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads)ASUS ROG CROSSHAIR VIII HERO (3402 BIOS)AMD Starship/Matisse16GB1000GB Sabrent Rocket 4.0 Plus + 2000GBNVIDIA TITAN RTX 24GBNVIDIA TU102 HD AudioASUS VP28URealtek RTL8125 2.5GbE + Intel I211Ubuntu 21.045.11.0-17-generic (x86_64)GNOME Shell 3.38.4X Server 1.20.11NVIDIA 465.314.6.0OpenCL 3.0 CUDA 11.3.1161.2.168GCC 10.3.0 + CUDA 11.3ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionNVIDIA GPU Compute BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/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-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009 - GPU Compute Cores: 4608- Python 3.9.5- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

NVIDIA GPU Computeplaidml: No - Inference - ResNet 50 - OpenCLplaidml: No - Inference - VGG16 - OpenCLplaidml: No - Inference - VGG19 - OpenCLcl-mem: Readcl-mem: Writecl-mem: Copyviennacl: OpenCL BLAS - sCOPYviennacl: OpenCL BLAS - sAXPYviennacl: OpenCL BLAS - dCOPYviennacl: OpenCL BLAS - dAXPYviennacl: OpenCL BLAS - dDOTshoc: OpenCL - Texture Read Bandwidthviennacl: OpenCL BLAS - sDOTviennacl: OpenCL BLAS - dGEMV-Nviennacl: OpenCL BLAS - dGEMV-Tclpeak: Global Memory Bandwidthmixbench: NVIDIA CUDA - Double Precisionmixbench: NVIDIA CUDA - Half Precisionarrayfire: BLAS OpenCLmixbench: NVIDIA CUDA - Single Precisionshoc: OpenCL - FFT SPshoc: OpenCL - GEMM SGEMM_Nshoc: OpenCL - S3Dclpeak: Single-Precision Floatclpeak: Double-Precision Doublevkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp16-scalarvkpeak: fp16-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4viennacl: OpenCL BLAS - dGEMM-NNviennacl: OpenCL BLAS - dGEMM-NTviennacl: OpenCL BLAS - dGEMM-TNviennacl: OpenCL BLAS - dGEMM-TTshoc: OpenCL - MD5 Hashclpeak: Integer Compute INTvkpeak: int32-scalarvkpeak: int32-vec4vkpeak: int16-scalarmixbench: NVIDIA CUDA - Integervkpeak: int16-vec4hashcat: MD5hashcat: SHA1hashcat: SHA-512hashcat: 7-Ziphashcat: TrueCrypt RIPEMD160 + XTSindigobench: OpenCL GPU - Supercarindigobench: OpenCL GPU - Bedroomluxcorerender: DLSC - GPUluxcorerender: Rainbow Colors and Prism - GPUluxcorerender: LuxCore Benchmark - GPUluxcorerender: Orange Juice - GPUluxcorerender: Danish Mood - GPUlczero: OpenCLfahbench: octanebench: Total Scorev-ray: NVIDIA CUDA GPUv-ray: NVIDIA RTX GPUnamd-cuda: ATPase Simulation - 327,506 Atomsarrayfire: Conjugate Gradient OpenCLvkresample: 2x - Singleblender: BMW27 - CUDAblender: BMW27 - NVIDIA OptiXblender: Classroom - CUDAblender: Classroom - NVIDIA OptiXblender: Fishy Cat - CUDAblender: Fishy Cat - NVIDIA OptiXblender: Pabellon Barcelona - CUDAblender: Pabellon Barcelona - NVIDIA OptiXblender: Barbershop - CUDAblender: Barbershop - NVIDIA OptiXrealsr-ncnn: 4x - Yesrealsr-ncnn: 4x - Nowaifu2x-ncnn: 2x - 3 - Yesbetsy: ETC1 - HighestTITAN RTX653.17247.51196.56567.8495.2319.33104024955415301162.96316461371530.37425.7928449.565005.0515409.841579.485163.49288.56914361.16547.2317175.2917203.0317182.7233537.08535.94536.5749549148549336.657513827.2917081.0116983.4511259.3015616.3314742.215672534285718683816667258771666794245067782935.32811.5837.0516.826.887.235.3028858307.6535382.299286100513880.131441.63713.56836.7818.70152.7273.6369.4532.73281.1997.27551.44906.4141.8967.3473.8543.703OpenBenchmarking.org

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: ResNet 50 - Device: OpenCLTITAN RTX140280420560700SE +/- 1.07, N = 6653.17

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: OpenCLTITAN RTX50100150200250SE +/- 0.66, N = 3247.51

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: OpenCLTITAN RTX4080120160200SE +/- 0.58, N = 3196.56

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: ReadTITAN RTX120240360480600SE +/- 0.10, N = 9567.81. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: WriteTITAN RTX110220330440550SE +/- 0.76, N = 9495.21. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: CopyTITAN RTX70140210280350SE +/- 0.10, N = 9319.31. (CC) gcc options: -O2 -flto -lOpenCL

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPYTITAN RTX70140210280350SE +/- 1.20, N = 33101. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

SHOC Scalable HeterOgeneous Computing

OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthTITAN RTX2468107.428

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPYTITAN RTX90180270360450SE +/- 0.33, N = 34021. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPYTITAN RTX1102203304405504951. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYTITAN RTX1202403604806005411. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTTITAN RTX1102203304405505301. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthTITAN RTX30060090012001500SE +/- 2.01, N = 31162.961. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTTITAN RTX701402102803503161. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-NTITAN RTX100200300400500SE +/- 0.33, N = 34611. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-TTITAN RTX80160240320400SE +/- 0.33, N = 33711. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

clpeak

OpenBenchmarking.orgGBPS Per Watt, More Is BetterclpeakOpenCL Test: Global Memory BandwidthTITAN RTX2468107.11

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory BandwidthTITAN RTX110220330440550SE +/- 0.03, N = 13530.371. (CXX) g++ options: -O3 -rdynamic -lOpenCL

Mixbench

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

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: Double PrecisionTITAN RTX90180270360450SE +/- 12.04, N = 15425.791. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: Half PrecisionTITAN RTX6K12K18K24K30KSE +/- 392.63, N = 1528449.561. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

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.orgGFLOPS, More Is BetterArrayFire 3.7Test: BLAS OpenCLTITAN RTX11002200330044005500SE +/- 13.00, N = 45005.051. (CXX) g++ options: -rdynamic

Mixbench

OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: Single PrecisionTITAN RTX60120180240300288.73

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: Single PrecisionTITAN RTX3K6K9K12K15KSE +/- 550.35, N = 1515409.841. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: FFT SPTITAN RTX30060090012001500SE +/- 0.53, N = 121579.481. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_NTITAN RTX11002200330044005500SE +/- 44.05, N = 155163.491. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3DTITAN RTX60120180240300SE +/- 0.11, N = 13288.571. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

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 FloatTITAN RTX3K6K9K12K15KSE +/- 180.32, N = 1514361.161. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleTITAN RTX120240360480600SE +/- 0.04, N = 3547.231. (CXX) g++ options: -O3 -rdynamic -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 20210424fp32-scalarTITAN RTX4K8K12K16K20KSE +/- 130.00, N = 317175.29

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4TITAN RTX4K8K12K16K20KSE +/- 134.80, N = 317203.03

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarTITAN RTX4K8K12K16K20KSE +/- 93.03, N = 317182.72

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4TITAN RTX7K14K21K28K35KSE +/- 87.22, N = 333537.08

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarTITAN RTX120240360480600SE +/- 1.42, N = 3535.94

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4TITAN RTX120240360480600SE +/- 1.21, N = 3536.57

ViennaCL

OpenBenchmarking.orgGFLOPs/s Per Watt, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTTITAN RTX0.65271.30541.95812.61083.26352.901

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNTITAN RTX110220330440550SE +/- 1.20, N = 34951. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTTITAN RTX110220330440550SE +/- 1.33, N = 34911. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNTITAN RTX100200300400500SE +/- 1.33, N = 34851. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTTITAN RTX1102203304405504931. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 HashTITAN RTX816243240SE +/- 0.04, N = 1436.661. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

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 INTTITAN RTX3K6K9K12K15KSE +/- 132.02, N = 1513827.291. (CXX) g++ options: -O3 -rdynamic -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 20210424int32-scalarTITAN RTX4K8K12K16K20KSE +/- 45.39, N = 317081.01

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4TITAN RTX4K8K12K16K20KSE +/- 47.86, N = 316983.45

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarTITAN RTX2K4K6K8K10KSE +/- 29.00, N = 311259.30

Mixbench

OpenBenchmarking.orgGIOPS Per Watt, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: IntegerTITAN RTX70140210280350310.23

vkpeak

OpenBenchmarking.orgGIOPS Per Watt, More Is Bettervkpeak 20210424int16-vec4TITAN RTX153045607569.13

Mixbench

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

OpenBenchmarking.orgGIOPS, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: IntegerTITAN RTX3K6K9K12K15KSE +/- 315.99, N = 1515616.331. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

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 20210424int16-vec4TITAN RTX3K6K9K12K15KSE +/- 14.87, N = 314742.21

Hashcat

OpenBenchmarking.orgH/s Per Watt, More Is BetterHashcat 6.1.1Benchmark: SHA1TITAN RTX30M60M90M120M150M126675761.31

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: MD5TITAN RTX12000M24000M36000M48000M60000MSE +/- 90782909.18, N = 756725342857

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA1TITAN RTX4000M8000M12000M16000M20000MSE +/- 27886393.54, N = 618683816667

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA-512TITAN RTX600M1200M1800M2400M3000MSE +/- 2648574.04, N = 62587716667

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: 7-ZipTITAN RTX200K400K600K800K1000KSE +/- 600.42, N = 10942450

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: TrueCrypt RIPEMD160 + XTSTITAN RTX150K300K450K600K750KSE +/- 690.66, N = 7677829

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: SupercarTITAN RTX816243240SE +/- 0.05, N = 335.33

OpenBenchmarking.orgM samples/s Per Watt, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: BedroomTITAN RTX0.01170.02340.03510.04680.05850.052

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: BedroomTITAN RTX3691215SE +/- 0.00, N = 311.58

LuxCoreRender

OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.5Scene: Rainbow Colors and Prism - Acceleration: GPUTITAN RTX0.02340.04680.07020.09360.1170.104

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: DLSC - Acceleration: GPUTITAN RTX246810SE +/- 0.00, N = 37.05MIN: 6.55 / MAX: 7.28

OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.5Scene: Orange Juice - Acceleration: GPUTITAN RTX0.00810.01620.02430.03240.04050.036

OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.5Scene: Danish Mood - Acceleration: GPUTITAN RTX0.00630.01260.01890.02520.03150.028

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Rainbow Colors and Prism - Acceleration: GPUTITAN RTX48121620SE +/- 0.01, N = 516.82MIN: 15.77 / MAX: 18.21

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: LuxCore Benchmark - Acceleration: GPUTITAN RTX246810SE +/- 0.00, N = 36.88MIN: 2.2 / MAX: 8.15

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Orange Juice - Acceleration: GPUTITAN RTX246810SE +/- 0.02, N = 37.23MIN: 5.72 / MAX: 9.06

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Danish Mood - Acceleration: GPUTITAN RTX1.19252.3853.57754.775.9625SE +/- 0.05, N = 65.30MIN: 1.66 / MAX: 6.37

LeelaChessZero

OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.26Backend: OpenCLTITAN RTX20406080100107.91

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: OpenCLTITAN RTX6K12K18K24K30KSE +/- 77.21, N = 3288581. (CXX) g++ options: -flto -pthread

FAHBench

OpenBenchmarking.orgNs Per Day Per Watt, More Is BetterFAHBench 2.3.2TITAN RTX0.41420.82841.24261.65682.0711.841

OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2TITAN RTX70140210280350SE +/- 0.84, N = 3307.65

OctaneBench

OpenBenchmarking.orgScore Per Watt, More Is BetterOctaneBench 2020.1Total ScoreTITAN RTX0.36180.72361.08541.44721.8091.608

OpenBenchmarking.orgScore, More Is BetterOctaneBench 2020.1Total ScoreTITAN RTX80160240320400382.30

Chaos Group V-RAY

OpenBenchmarking.orgvpaths Per Watt, More Is BetterChaos Group V-RAY 5Mode: NVIDIA CUDA GPUTITAN RTX1.24972.49943.74914.99886.24855.554

OpenBenchmarking.orgvpaths, More Is BetterChaos Group V-RAY 5Mode: NVIDIA CUDA GPUTITAN RTX20040060080010001005

OpenBenchmarking.orgvrays, More Is BetterChaos Group V-RAY 5Mode: NVIDIA RTX GPUTITAN RTX30060090012001500SE +/- 0.67, N = 31388

NAMD CUDA

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. This version of the NAMD test profile uses CUDA GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 AtomsTITAN RTX0.02960.05920.08880.11840.148SE +/- 0.00043, N = 50.13144

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 OpenCLTITAN RTX0.36830.73661.10491.47321.8415SE +/- 0.004, N = 111.6371. (CXX) g++ options: -rdynamic

VkResample

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

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleTITAN RTX3691215SE +/- 0.02, N = 513.571. (CXX) g++ options: -O3 -pthread

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CUDATITAN RTX816243240SE +/- 0.03, N = 336.78

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: NVIDIA OptiXTITAN RTX510152025SE +/- 0.02, N = 318.70

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CUDATITAN RTX306090120150SE +/- 0.17, N = 3152.72

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: NVIDIA OptiXTITAN RTX1632486480SE +/- 0.09, N = 373.63

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CUDATITAN RTX1530456075SE +/- 0.02, N = 369.45

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: NVIDIA OptiXTITAN RTX816243240SE +/- 0.05, N = 332.73

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: CUDATITAN RTX60120180240300SE +/- 0.06, N = 3281.19

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXTITAN RTX20406080100SE +/- 0.04, N = 397.27

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CUDATITAN RTX120240360480600SE +/- 0.57, N = 3551.44

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: NVIDIA OptiXTITAN RTX2004006008001000SE +/- 1.64, N = 3906.41

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: YesTITAN RTX1020304050SE +/- 0.18, N = 341.90

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoTITAN RTX246810SE +/- 0.018, N = 67.347

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: YesTITAN RTX0.86721.73442.60163.46884.336SE +/- 0.005, N = 83.854

Betsy GPU Compressor

Betsy is an open-source GPU compressor of various GPU compression techniques. Betsy is written in GLSL for Vulkan/OpenGL (compute shader) support for GPU-based texture compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC1 - Quality: HighestTITAN RTX0.83321.66642.49963.33284.166SE +/- 0.009, N = 93.7031. (CXX) g++ options: -O3 -O2 -lpthread -ldl

91 Results Shown

PlaidML:
  No - Inference - ResNet 50 - OpenCL
  No - Inference - VGG16 - OpenCL
  No - Inference - VGG19 - OpenCL
cl-mem:
  Read
  Write
  Copy
ViennaCL
SHOC Scalable HeterOgeneous Computing
ViennaCL:
  OpenCL BLAS - sAXPY
  OpenCL BLAS - dCOPY
  OpenCL BLAS - dAXPY
  OpenCL BLAS - dDOT
SHOC Scalable HeterOgeneous Computing
ViennaCL:
  OpenCL BLAS - sDOT
  OpenCL BLAS - dGEMV-N
  OpenCL BLAS - dGEMV-T
clpeak
clpeak
Mixbench:
  NVIDIA CUDA - Double Precision
  NVIDIA CUDA - Half Precision
ArrayFire
Mixbench
Mixbench
SHOC Scalable HeterOgeneous Computing:
  OpenCL - FFT SP
  OpenCL - GEMM SGEMM_N
  OpenCL - S3D
clpeak:
  Single-Precision Float
  Double-Precision Double
vkpeak:
  fp32-scalar
  fp32-vec4
  fp16-scalar
  fp16-vec4
  fp64-scalar
  fp64-vec4
ViennaCL
ViennaCL:
  OpenCL BLAS - dGEMM-NN
  OpenCL BLAS - dGEMM-NT
  OpenCL BLAS - dGEMM-TN
  OpenCL BLAS - dGEMM-TT
SHOC Scalable HeterOgeneous Computing
clpeak
vkpeak:
  int32-scalar
  int32-vec4
  int16-scalar
Mixbench:
  NVIDIA CUDA - Integer
  int16-vec4
Mixbench
vkpeak
Hashcat
Hashcat:
  MD5
  SHA1
  SHA-512
  7-Zip
  TrueCrypt RIPEMD160 + XTS
IndigoBench
IndigoBench
IndigoBench
LuxCoreRender
LuxCoreRender
LuxCoreRender:
  Orange Juice - GPU
  Danish Mood - GPU
LuxCoreRender:
  Rainbow Colors and Prism - GPU
  LuxCore Benchmark - GPU
  Orange Juice - GPU
  Danish Mood - GPU
LeelaChessZero
LeelaChessZero
FAHBench
FAHBench
OctaneBench
OctaneBench
Chaos Group V-RAY
Chaos Group V-RAY:
  NVIDIA CUDA GPU
  NVIDIA RTX GPU
NAMD CUDA
ArrayFire
VkResample
Blender:
  BMW27 - CUDA
  BMW27 - NVIDIA OptiX
  Classroom - CUDA
  Classroom - NVIDIA OptiX
  Fishy Cat - CUDA
  Fishy Cat - NVIDIA OptiX
  Pabellon Barcelona - CUDA
  Pabellon Barcelona - NVIDIA OptiX
  Barbershop - CUDA
  Barbershop - NVIDIA OptiX
RealSR-NCNN:
  4x - Yes
  4x - No
Waifu2x-NCNN Vulkan
Betsy GPU Compressor