RTX 4070 SUPER

Intel Core i9-13900K testing with a ASUS TUF GAMING Z790-PRO WIFI (1401 BIOS) and NVIDIA GeForce RTX 3090 24GB on EndeavourOS rolling via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2402116-SADD-240207012&grr&rdt.

RTX 4070 SUPERProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLCompilerFile-SystemScreen ResolutionNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090Intel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads)ASUS TUF GAMING Z790-PRO WIFI (1401 BIOS)Intel Device 7a2732GB4001GB Seagate ZP4000GP304001ASUS NVIDIA GeForce RTX 4070 SUPER 12GBRealtek ALC1220ARZOPAIntel I226-V + Intel Device 7a70EndeavourOS rolling6.7.1-arch1-1 (x86_64)KDE Plasma 5.27.10X Server 1.21.1.11NVIDIA 550.40.074.6.0OpenCL 3.0 CUDA 12.4.74GCC 13.2.1 20230801ext41920x1080MSI NVIDIA GeForce RTX 4070 12GBGCC 13.2.1 20230801 + CUDA 12.3NVIDIA GeForce RTX 4070 Ti 12GBNVIDIA GeForce RTX 3090 24GBPI-KVM Video6.7.4-arch1-1 (x86_64)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: alwaysCompiler Details- NVIDIA RTX 4070 SUPER: --disable-libssp --disable-libstdcxx-pch --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet=auto --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-gnu-indirect-function --enable-gnu-unique-object --enable-languages=ada,c,c++,d,fortran,go,lto,objc,obj-c++ --enable-libstdcxx-backtrace --enable-link-serialization=1 --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-build-config=bootstrap-lto --with-linker-hash-style=gnu - NVIDIA RTX 4070: --disable-libssp --disable-libstdcxx-pch --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet=auto --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-gnu-indirect-function --enable-gnu-unique-object --enable-languages=ada,c,c++,d,fortran,go,lto,objc,obj-c++ --enable-libstdcxx-backtrace --enable-link-serialization=1 --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-build-config=bootstrap-lto --with-linker-hash-style=gnu - NVIDIA RTX 4070 TI: --disable-libssp --disable-libstdcxx-pch --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet=auto --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-gnu-indirect-function --enable-gnu-unique-object --enable-languages=ada,c,c++,d,fortran,go,lto,objc,obj-c++ --enable-libstdcxx-backtrace --enable-link-serialization=1 --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-build-config=bootstrap-lto --with-linker-hash-style=gnu - NVIDIA RTX 3090: --disable-libssp --disable-libstdcxx-pch --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet=auto --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-gnu-indirect-function --enable-gnu-unique-object --enable-languages=ada,c,c++,d,fortran,go,lto,m2,objc,obj-c++ --enable-libstdcxx-backtrace --enable-link-serialization=1 --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-build-config=bootstrap-lto --with-linker-hash-style=gnu Processor Details- Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x11dGraphics Details- NVIDIA RTX 4070 SUPER: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.04.69.00.c1- NVIDIA RTX 4070: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.04.3e.40.2a- NVIDIA RTX 4070 TI: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.04.31.00.36- NVIDIA RTX 3090: BAR1 / Visible vRAM Size: 256 MiB - vBIOS Version: 94.02.26.08.baSecurity Details- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected Environment Details- NVIDIA RTX 4070, NVIDIA RTX 4070 TI, NVIDIA RTX 3090: NVCC_PREPEND_FLAGS="-ccbin /opt/cuda/bin"Python Details- NVIDIA RTX 4070, NVIDIA RTX 4070 TI, NVIDIA RTX 3090: Python 3.11.6

RTX 4070 SUPERtensorflow: GPU - 256 - VGG-16tensorflow: GPU - 64 - VGG-16tensorflow: GPU - 32 - VGG-16tensorflow: GPU - 512 - AlexNettensorflow: GPU - 64 - ResNet-50tensorflow: GPU - 16 - VGG-16ncnn: Vulkan GPU - FastestDetncnn: Vulkan GPU - vision_transformerncnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU - mobilenetncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - blazefacencnn: Vulkan GPU-v3-v3 - mobilenet-v3tensorflow: GPU - 256 - AlexNettensorflow: GPU - 32 - ResNet-50tensorflow: GPU - 64 - GoogLeNettensorflow: GPU - 16 - ResNet-50vkpeak: int16-vec4vkpeak: int16-scalarvkpeak: int32-vec4vkpeak: int32-scalarvkpeak: fp64-vec4vkpeak: fp64-scalarvkpeak: fp16-vec4vkpeak: fp16-scalarvkpeak: fp32-vec4vkpeak: fp32-scalartensorflow: GPU - 32 - GoogLeNettensorflow: GPU - 64 - AlexNetgpuowl: 77936867gpuowl: 332220523octanebench: Total Scorevkfft: FFT + iFFT C2C multidimensional in single precisiongpuowl: 57885161tensorflow: GPU - 16 - GoogLeNetluxcorerender: DLSC - GPUvkfft: FFT + iFFT C2C Bluestein benchmark in double precisiontensorflow: GPU - 32 - AlexNetfahbench: vkresample: 2x - Doubleindigobench: OpenCL GPU - Bedroomvkfft: FFT + iFFT C2C 1D batched in double precisionindigobench: OpenCL GPU - Supercarluxcorerender: LuxCore Benchmark - GPUvkfft: FFT + iFFT C2C Bluestein in single precisionluxcorerender: Orange Juice - GPUluxcorerender: Danish Mood - GPUtensorflow: GPU - 1 - VGG-16blender: Barbershop - NVIDIA OptiXvkfft: FFT + iFFT C2C 1D batched in single precisionvkfft: FFT + iFFT C2C 1D batched in single precision, no reshufflingvkfft: FFT + iFFT R2C / C2Rtensorflow: GPU - 16 - AlexNetlibplacebo: av1_grain_laplibplacebo: hdr_lutlibplacebo: hdr_peakdetectlibplacebo: polar_nocomputelibplacebo: deband_heavyrealsr-ncnn: 4x - Yesvkfft: FFT + iFFT C2C 1D batched in half precisionnamd-cuda: ATPase Simulation - 327,506 Atomsblender: Fishy Cat - NVIDIA OptiXpytorch: NVIDIA CUDA GPU - 32 - Efficientnet_v2_lpytorch: NVIDIA CUDA GPU - 64 - Efficientnet_v2_ltensorflow: GPU - 1 - ResNet-50viennacl: CPU BLAS - dGEMM-TTviennacl: CPU BLAS - dGEMM-TNviennacl: CPU BLAS - dGEMM-NTviennacl: CPU BLAS - dGEMM-NNviennacl: CPU BLAS - dGEMV-Tviennacl: CPU BLAS - dGEMV-Nviennacl: CPU BLAS - dDOTviennacl: CPU BLAS - dAXPYviennacl: CPU BLAS - dCOPYviennacl: CPU BLAS - sDOTviennacl: CPU BLAS - sAXPYviennacl: CPU BLAS - sCOPYpytorch: NVIDIA CUDA GPU - 512 - Efficientnet_v2_lviennacl: OpenCL BLAS - dGEMM-TTviennacl: OpenCL BLAS - dGEMM-TNviennacl: OpenCL BLAS - dGEMM-NTviennacl: OpenCL BLAS - dGEMM-NNviennacl: OpenCL BLAS - dGEMV-Tviennacl: OpenCL BLAS - dGEMV-Nviennacl: OpenCL BLAS - dDOTviennacl: OpenCL BLAS - dAXPYviennacl: OpenCL BLAS - dCOPYviennacl: OpenCL BLAS - sDOTviennacl: OpenCL BLAS - sAXPYviennacl: OpenCL BLAS - sCOPYpytorch: NVIDIA CUDA GPU - 256 - Efficientnet_v2_ltensorflow: GPU - 1 - AlexNetblender: BMW27 - NVIDIA OptiXblender: Pabellon Barcelona - NVIDIA OptiXblender: Classroom - NVIDIA OptiXpytorch: NVIDIA CUDA GPU - 16 - Efficientnet_v2_lopencl-benchmark: Memory Bandwidth Coalesced Writeopencl-benchmark: Memory Bandwidth Coalesced Readopencl-benchmark: INT8 Computeopencl-benchmark: INT16 Computeopencl-benchmark: INT32 Computeopencl-benchmark: INT64 Computeopencl-benchmark: FP32 Computeopencl-benchmark: FP64 Computerealsr-ncnn: 4x - Nopytorch: NVIDIA CUDA GPU - 512 - ResNet-152pytorch: NVIDIA CUDA GPU - 64 - ResNet-152pytorch: NVIDIA CUDA GPU - 256 - ResNet-152pytorch: NVIDIA CUDA GPU - 1 - Efficientnet_v2_lpytorch: NVIDIA CUDA GPU - 32 - ResNet-152tensorflow: GPU - 1 - GoogLeNetvkresample: 2x - Singleclpeak: Double-Precision Doublepytorch: NVIDIA CUDA GPU - 16 - ResNet-152luxcorerender: Rainbow Colors and Prism - GPUpytorch: NVIDIA CUDA GPU - 1 - ResNet-50pytorch: NVIDIA CUDA GPU - 64 - ResNet-50rodinia: OpenCL Particle Filterhashcat: MD5hashcat: SHA1hashcat: SHA-512pytorch: NVIDIA CUDA GPU - 256 - ResNet-50pytorch: NVIDIA CUDA GPU - 1 - ResNet-152pytorch: NVIDIA CUDA GPU - 16 - ResNet-50pytorch: NVIDIA CUDA GPU - 512 - ResNet-50pytorch: NVIDIA CUDA GPU - 32 - ResNet-50hashcat: TrueCrypt RIPEMD160 + XTScl-mem: Copycl-mem: Writecl-mem: Readhashcat: 7-Zipwaifu2x-ncnn: 2x - 3 - Yesclpeak: Global Memory Bandwidthmandelgpu: GPUfinancebench: Black-Scholes OpenCLclpeak: Integer Compute INTclpeak: Single-Precision Floatneatbench: GPUarrayfire: Conjugate Gradient OpenCLNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30901.5035.105.551.482.86844.6111.116.8663.8246.2616.178.97117.815.073.852.313.038.6211.040.842.2534.165.5115.525.4615.6133.97646.41137.44720.97378950299869.0715.6713.59445133.4366.0576339.59319.8012431752.81312.821516611.7210.561.3551.3073929750785479431.594171.003905.983292.372327.552186.7034.8851317050.067919.45102.60102.604.3512211511711910910296.887.270.8165156132103.57613599584577389210458437423370392334103.1713.925.5714.2912.60455.01464.8614.30717.17019.8894.21438.5940.6216.323195.30196.07194.58106.37195.3912.6218.489630.11195.4027.67557.73507.453.48067583033333221326000003232733333504.67201.94509.45504.27501.50802967331.8407.5446.211764672.855437.65587219538.25.91218170.5435492.6940701.501.535.215.551.502.34281.566.505.2725.118.249.338.5854.543.462.222.114.6910.146.060.848.715.5515.545.4915.6333.93530.32112.61647.99786747212714.8015.6611.74388633.32317.1952415.16018.2032239048.51710.921371410.408.891.3658.4477774790574709731.454103.403927.113310.021972.781847.9842.8521377620.0749811.03102.90101.554.3411812112212210910396.786.871.0166153131101.43502494477473387209456455423362389330101.2414.046.2116.5514.86103.68459.43465.1812.11614.28416.3773.44331.7680.5107.092187.51186.63187.27107.59187.6912.7818.016515.17187.2623.26546.76458.364.09856147866667182024666672673300000459.93198.18458.39459.27459.94660967330.3406.7446.39769673.168437.21516770131.26.90614555.1928479.3940701.51.51.535.445.531.492.84390.185.975.3616.4714.323.745.4732.053.492.302.032.438.435.870.812.0934.615.5015.505.4615.8134.06676.59145.84735.94059351528919.1315.6913.95464733.29382.1637322.06420.2562543153.58913.231512511.8910.991.3850.7373942751415544631.704140.873971.613544.602461.232306.6733.6261362100.067889.0296.50103.204.32124125118117102.710396.487.371.3168156132103.50648634612604391211457437424365393336103.2414.795.4313.9712.30103.45457.17465.0715.73118.28121.0474.42040.9140.6605.962194.87197.02195.86108.59198.8212.7918.456667.05194.2927.71535.39505.623.29173312233333235324000003462500000201.19502.92504.66505.55858600333.3412.2446.312626332.854437.63619106132.55.22619821.1038691.7340701.511.511.535.585.571.496.38663.248.066.6326.8527.773.6917.41145.7213.872.244.172.6512.077.490.863.1934.465.5715.635.4916331.1613259.9719996.9220280.33638.72638.7039771.9720080.4726563.7220263.1315.6733.93645.99137.32674.25091250856866.3115.6812.99419533.53343.0199333.63920.9593091252.01413.121420512.1410.201.3854.301418761443114841831.984126.893313.264997.082119.892017.7530.3132732210.1082210.6499.0599.844.3511312111911311010395.286.270.2132.115413299.2559359459559237418765972460537649836399.4314.456.3117.3015.2698.11887.31864.1113.72717.00120.0273.13539.3950.6375.556164.35164.14161.01105.55163.7412.8210.323642.23164.1433.29525.12419.033.84467177300000213237333333081866667416.89197.12419.76416.20420.29797833360.8753.8825.810560003.202816.55484098913.85.74117923.3334906.793090OpenBenchmarking.org

TensorFlow

Device: GPU - Batch Size: 256 - Model: VGG-16

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 256 - Model: VGG-16NVIDIA RTX 4070 TINVIDIA RTX 30900.33980.67961.01941.35921.699SE +/- 0.00, N = 31.501.51

TensorFlow

Device: GPU - Batch Size: 64 - Model: VGG-16

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 64 - Model: VGG-16NVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.33980.67961.01941.35921.699SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.501.501.51

TensorFlow

Device: GPU - Batch Size: 32 - Model: VGG-16

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 32 - Model: VGG-16NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.33750.6751.01251.351.6875SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 2SE +/- 0.00, N = 31.501.501.501.50

TensorFlow

Device: GPU - Batch Size: 512 - Model: AlexNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 512 - Model: AlexNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090816243240SE +/- 0.02, N = 2SE +/- 0.03, N = 3SE +/- 0.09, N = 2SE +/- 0.01, N = 335.1035.2135.4435.58

TensorFlow

Device: GPU - Batch Size: 64 - Model: ResNet-50

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 64 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30901.25332.50663.75995.01326.2665SE +/- 0.01, N = 2SE +/- 0.00, N = 3SE +/- 0.01, N = 2SE +/- 0.01, N = 35.555.555.535.57

TensorFlow

Device: GPU - Batch Size: 16 - Model: VGG-16

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 16 - Model: VGG-16NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.33750.6751.01251.351.6875SE +/- 0.00, N = 2SE +/- 0.01, N = 2SE +/- 0.00, N = 31.481.501.491.49

NCNN

Target: Vulkan GPU - Model: FastestDet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: FastestDetNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER246810SE +/- 0.13, N = 12SE +/- 0.32, N = 10SE +/- 0.08, N = 8SE +/- 0.29, N = 92.673.042.502.86MIN: 2.04 / MAX: 3.47MIN: 2.14 / MAX: 847.58MIN: 2.1 / MAX: 32.36MIN: 2.17 / MAX: 577.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: vision_transformer

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vision_transformerNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER2004006008001000SE +/- 58.37, N = 12SE +/- 58.07, N = 12SE +/- 52.80, N = 9SE +/- 87.53, N = 9382.82497.66327.82844.61MIN: 46.51 / MAX: 1840MIN: 46.41 / MAX: 1886.67MIN: 46.48 / MAX: 1816.93MIN: 46.34 / MAX: 1866.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: regnety_400mNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER3691215SE +/- 0.24, N = 12SE +/- 0.18, N = 12SE +/- 0.32, N = 8SE +/- 3.28, N = 96.215.896.4711.11MIN: 5.53 / MAX: 8.99MIN: 5.42 / MAX: 7.57MIN: 5.44 / MAX: 9.3MIN: 5.49 / MAX: 4942.191. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: squeezenet_ssdNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER246810SE +/- 0.12, N = 12SE +/- 0.93, N = 12SE +/- 0.20, N = 9SE +/- 1.76, N = 95.186.135.206.86MIN: 4.67 / MAX: 6.88MIN: 4.75 / MAX: 1633.33MIN: 4.36 / MAX: 6.17MIN: 4.34 / MAX: 1630.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: yolov4-tinyNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER1428425670SE +/- 5.37, N = 12SE +/- 3.10, N = 12SE +/- 2.52, N = 9SE +/- 10.56, N = 920.7416.3713.3163.82MIN: 10.3 / MAX: 854.36MIN: 10.57 / MAX: 855.36MIN: 10.13 / MAX: 840.81MIN: 10.28 / MAX: 858.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet50NVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER1020304050SE +/- 0.49, N = 12SE +/- 4.00, N = 12SE +/- 0.12, N = 9SE +/- 14.70, N = 98.7212.258.2046.26MIN: 7.94 / MAX: 935.44MIN: 8 / MAX: 1777.17MIN: 7.69 / MAX: 11.69MIN: 7.71 / MAX: 1829.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: alexnetNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER48121620SE +/- 1.70, N = 12SE +/- 2.33, N = 12SE +/- 1.74, N = 9SE +/- 5.86, N = 95.786.076.2016.17MIN: 3.6 / MAX: 397.75MIN: 3.63 / MAX: 432.24MIN: 3.56 / MAX: 340MIN: 3.52 / MAX: 436.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet18NVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER48121620SE +/- 0.73, N = 12SE +/- 1.74, N = 12SE +/- 3.28, N = 9SE +/- 3.49, N = 95.117.748.948.97MIN: 3.99 / MAX: 916.69MIN: 3.96 / MAX: 898.97MIN: 4.04 / MAX: 900MIN: 3.94 / MAX: 922.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vgg16NVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER306090120150SE +/- 13.24, N = 12SE +/- 8.53, N = 12SE +/- 5.69, N = 9SE +/- 29.60, N = 945.5234.4924.45117.81MIN: 17.49 / MAX: 643.35MIN: 17.35 / MAX: 644.35MIN: 18.11 / MAX: 642.2MIN: 17.16 / MAX: 647.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: efficientnet-b0NVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER48121620SE +/- 0.20, N = 12SE +/- 0.07, N = 12SE +/- 0.07, N = 9SE +/- 0.97, N = 93.593.463.545.07MIN: 3.02 / MAX: 509.97MIN: 3.13 / MAX: 7.03MIN: 3.19 / MAX: 3.95MIN: 3.22 / MAX: 1124.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mnasnetNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER0.93151.8632.79453.7264.6575SE +/- 0.05, N = 10SE +/- 1.94, N = 12SE +/- 0.06, N = 9SE +/- 1.31, N = 92.244.142.303.85MIN: 2 / MAX: 2.6MIN: 1.95 / MAX: 1020MIN: 2.01 / MAX: 2.64MIN: 1.89 / MAX: 1093.291. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: shufflenet-v2NVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER0.93831.87662.81493.75324.6915SE +/- 0.09, N = 11SE +/- 0.08, N = 12SE +/- 0.10, N = 9SE +/- 0.34, N = 82.082.012.092.31MIN: 1.82 / MAX: 2.59MIN: 1.73 / MAX: 3.86MIN: 1.83 / MAX: 2.96MIN: 1.76 / MAX: 421.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2NVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER1.05532.11063.16594.22125.2765SE +/- 0.07, N = 12SE +/- 0.06, N = 12SE +/- 0.05, N = 9SE +/- 0.44, N = 92.482.542.673.03MIN: 2.02 / MAX: 5.82MIN: 2.07 / MAX: 15.69MIN: 2.44 / MAX: 22.79MIN: 2.38 / MAX: 970.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mobilenetNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER3691215SE +/- 0.21, N = 12SE +/- 0.25, N = 12SE +/- 0.22, N = 9SE +/- 0.47, N = 97.207.456.928.62MIN: 6.2 / MAX: 11.13MIN: 6.87 / MAX: 734.65MIN: 6.06 / MAX: 8.65MIN: 6.42 / MAX: 1101.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: googlenetNVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPERNVIDIA RTX 40703691215SE +/- 1.04, N = 12SE +/- 0.18, N = 9SE +/- 1.21, N = 9SE +/- 0.14, N = 97.376.1111.046.06MIN: 5.33 / MAX: 1653.33MIN: 5.25 / MAX: 9.16MIN: 5.28 / MAX: 1769.19MIN: 5.33 / MAX: 8.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: blazefaceNVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPERNVIDIA RTX 40700.19580.39160.58740.78320.979SE +/- 0.03, N = 12SE +/- 0.03, N = 9SE +/- 0.04, N = 9SE +/- 0.03, N = 90.820.840.840.84MIN: 0.63 / MAX: 2.58MIN: 0.63 / MAX: 1.13MIN: 0.65 / MAX: 4.63MIN: 0.64 / MAX: 0.961. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3NVIDIA RTX 4070NVIDIA RTX 3090NVIDIA RTX 4070 SUPERNVIDIA RTX 4070 TI246810SE +/- 0.08, N = 12SE +/- 0.09, N = 9SE +/- 0.16, N = 9SE +/- 0.09, N = 92.152.202.252.09MIN: 1.81 / MAX: 2.58MIN: 1.91 / MAX: 2.71MIN: 1.75 / MAX: 343.7MIN: 1.78 / MAX: 2.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

TensorFlow

Device: GPU - Batch Size: 256 - Model: AlexNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 256 - Model: AlexNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070 TINVIDIA RTX 3090816243240SE +/- 0.01, N = 3SE +/- 0.07, N = 2SE +/- 0.07, N = 334.1634.6134.46

TensorFlow

Device: GPU - Batch Size: 32 - Model: ResNet-50

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 32 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30901.25332.50663.75995.01326.2665SE +/- 0.01, N = 2SE +/- 0.01, N = 2SE +/- 0.02, N = 3SE +/- 0.01, N = 35.515.555.505.57

TensorFlow

Device: GPU - Batch Size: 64 - Model: GoogLeNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 64 - Model: GoogLeNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.07, N = 3SE +/- 0.06, N = 2SE +/- 0.08, N = 315.5215.5415.5015.63

TensorFlow

Device: GPU - Batch Size: 16 - Model: ResNet-50

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 16 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30901.23532.47063.70594.94126.1765SE +/- 0.00, N = 2SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 35.465.495.465.49

vkpeak

int16-vec4

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-vec4NVIDIA RTX 30904K8K12K16K20KSE +/- 9.58, N = 316338.23

vkpeak

int16-scalar

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-scalarNVIDIA RTX 30903K6K9K12K15KSE +/- 10.34, N = 313273.53

vkpeak

int32-vec4

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-vec4NVIDIA RTX 30904K8K12K16K20KSE +/- 15.96, N = 320017.06

vkpeak

int32-scalar

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-scalarNVIDIA RTX 30904K8K12K16K20KSE +/- 16.62, N = 320290.30

vkpeak

fp64-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-vec4NVIDIA RTX 3090140280420560700638.77

vkpeak

fp64-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-scalarNVIDIA RTX 3090140280420560700SE +/- 0.02, N = 3638.75

vkpeak

fp16-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-vec4NVIDIA RTX 30909K18K27K36K45KSE +/- 61.18, N = 339835.21

vkpeak

fp16-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-scalarNVIDIA RTX 30904K8K12K16K20KSE +/- 33.96, N = 320113.01

vkpeak

fp32-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-vec4NVIDIA RTX 30906K12K18K24K30KSE +/- 70.66, N = 326630.59

vkpeak

fp32-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-scalarNVIDIA RTX 30904K8K12K16K20KSE +/- 85.09, N = 320319.80

TensorFlow

Device: GPU - Batch Size: 32 - Model: GoogLeNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 32 - Model: GoogLeNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.01, N = 2SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 315.6115.6315.8115.67

TensorFlow

Device: GPU - Batch Size: 64 - Model: AlexNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 64 - Model: AlexNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090816243240SE +/- 0.14, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 333.9733.9334.0633.93

GpuOwl

Exponent: 77936867

OpenBenchmarking.orgIterations / Second, More Is BetterGpuOwl 7.2.1Exponent: 77936867NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090150300450600750SE +/- 0.00, N = 3SE +/- 0.09, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3646.41530.32676.59645.99

GpuOwl

Exponent: 332220523

OpenBenchmarking.orgIterations / Second, More Is BetterGpuOwl 7.2.1Exponent: 332220523NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090306090120150SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3137.44112.61145.84137.32

OctaneBench

Total Score

OpenBenchmarking.orgScore, More Is BetterOctaneBench 2020.1Total ScoreNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090160320480640800720.97648.00735.94674.25

VkFFT

Test: FFT + iFFT C2C multidimensional in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C multidimensional in single precisionNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309011K22K33K44K55KSE +/- 407.19, N = 15SE +/- 476.57, N = 5SE +/- 417.77, N = 15SE +/- 407.28, N = 15502994721251528508561. (CXX) g++ options: -O3 -lrt

GpuOwl

Exponent: 57885161

OpenBenchmarking.orgIterations / Second, More Is BetterGpuOwl 7.2.1Exponent: 57885161NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30902004006008001000SE +/- 1.26, N = 3SE +/- 0.00, N = 3SE +/- 2.53, N = 3SE +/- 2.01, N = 3869.07714.80919.13866.31

TensorFlow

Device: GPU - Batch Size: 16 - Model: GoogLeNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 16 - Model: GoogLeNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 315.6715.6615.6915.68

LuxCoreRender

Scene: DLSC - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: GPUNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 1.13, N = 1213.5911.7413.9512.99MIN: 12.52 / MAX: 13.84MIN: 11.35 / MAX: 11.83MIN: 13.67 / MAX: 14.14MIN: 0.52 / MAX: 14.69

VkFFT

Test: FFT + iFFT C2C Bluestein benchmark in double precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C Bluestein benchmark in double precisionNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309010002000300040005000SE +/- 12.55, N = 3SE +/- 4.51, N = 3SE +/- 11.35, N = 3SE +/- 9.84, N = 344513886464741951. (CXX) g++ options: -O3 -lrt

TensorFlow

Device: GPU - Batch Size: 32 - Model: AlexNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 32 - Model: AlexNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090816243240SE +/- 0.15, N = 2SE +/- 0.18, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 333.4033.3233.2933.53

FAHBench

OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309080160240320400SE +/- 0.39, N = 3SE +/- 0.12, N = 3SE +/- 0.26, N = 3SE +/- 0.26, N = 3366.06317.20382.16343.02

VkResample

Upscale: 2x - Precision: Double

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: DoubleNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309090180270360450SE +/- 0.30, N = 3SE +/- 0.77, N = 3SE +/- 0.35, N = 3SE +/- 0.30, N = 3339.59415.16322.06333.641. (CXX) g++ options: -O3

IndigoBench

Acceleration: OpenCL GPU - Scene: Bedroom

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: BedroomNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090510152025SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 319.8018.2020.2620.96

VkFFT

Test: FFT + iFFT C2C 1D batched in double precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in double precisionNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30907K14K21K28K35KSE +/- 146.69, N = 3SE +/- 125.94, N = 3SE +/- 302.46, N = 3SE +/- 50.66, N = 3243172239025431309121. (CXX) g++ options: -O3 -lrt

IndigoBench

Acceleration: OpenCL GPU - Scene: Supercar

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: SupercarNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30901224364860SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 352.8148.5253.5952.01

LuxCoreRender

Scene: LuxCore Benchmark - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: GPUNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30903691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 212.8210.9213.2313.12MIN: 4.84 / MAX: 14.62MIN: 4.45 / MAX: 12.42MIN: 5.41 / MAX: 15.13MIN: 4.85 / MAX: 15.21

VkFFT

Test: FFT + iFFT C2C Bluestein in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C Bluestein in single precisionNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30903K6K9K12K15KSE +/- 102.52, N = 3SE +/- 52.09, N = 3SE +/- 118.41, N = 3SE +/- 115.62, N = 3151661371415125142051. (CXX) g++ options: -O3 -lrt

LuxCoreRender

Scene: Orange Juice - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: GPUNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30903691215SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 311.7210.4011.8912.14MIN: 9.6 / MAX: 15.44MIN: 8.31 / MAX: 13.9MIN: 9.85 / MAX: 15.88MIN: 10.24 / MAX: 16.71

LuxCoreRender

Scene: Danish Mood - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: GPUNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30903691215SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 310.568.8910.9910.20MIN: 3.7 / MAX: 12.17MIN: 3.32 / MAX: 10.26MIN: 4.17 / MAX: 12.71MIN: 4.07 / MAX: 11.93

TensorFlow

Device: GPU - Batch Size: 1 - Model: VGG-16

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 1 - Model: VGG-16NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.31050.6210.93151.2421.5525SE +/- 0.01, N = 2SE +/- 0.01, N = 3SE +/- 0.01, N = 31.351.361.381.38

Blender

Blend File: Barbershop - Compute: NVIDIA OptiX

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Barbershop - Compute: NVIDIA OptiXNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30901326395265SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 251.3058.4450.7354.30

VkFFT

Test: FFT + iFFT C2C 1D batched in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in single precisionNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309030K60K90K120K150KSE +/- 7.94, N = 3SE +/- 13.72, N = 3SE +/- 0.88, N = 3SE +/- 9.64, N = 37392977774739421418761. (CXX) g++ options: -O3 -lrt

VkFFT

Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in single precision, no reshufflingNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309030K60K90K120K150KSE +/- 37.77, N = 3SE +/- 5.84, N = 3SE +/- 28.54, N = 3SE +/- 37.44, N = 37507879057751411443111. (CXX) g++ options: -O3 -lrt

VkFFT

Test: FFT + iFFT R2C / C2R

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT R2C / C2RNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309012K24K36K48K60KSE +/- 702.53, N = 15SE +/- 745.02, N = 13SE +/- 520.37, N = 3SE +/- 320.62, N = 3547944709755446484181. (CXX) g++ options: -O3 -lrt

TensorFlow

Device: GPU - Batch Size: 16 - Model: AlexNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 16 - Model: AlexNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090714212835SE +/- 0.17, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 331.5931.4531.7031.98

Libplacebo

Test: av1_grain_lap

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: av1_grain_lapNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER9001800270036004500SE +/- 72.31, N = 3SE +/- 35.33, N = 3SE +/- 42.82, N = 3SE +/- 5.52, N = 34126.404143.964096.484171.001. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF

Libplacebo

Test: hdr_lut

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: hdr_lutNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER9001800270036004500SE +/- 0.24, N = 3SE +/- 33.96, N = 3SE +/- 22.23, N = 3SE +/- 12.09, N = 33940.403976.043376.853905.981. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF

Libplacebo

Test: hdr_peakdetect

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: hdr_peakdetectNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER11002200330044005500SE +/- 144.09, N = 3SE +/- 165.74, N = 3SE +/- 13.97, N = 3SE +/- 3.65, N = 33452.433475.065104.103292.371. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF

Libplacebo

Test: polar_nocompute

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: polar_nocomputeNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER5001000150020002500SE +/- 0.29, N = 3SE +/- 1.70, N = 3SE +/- 3.16, N = 3SE +/- 0.24, N = 31969.192459.032116.502327.551. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF

Libplacebo

Test: deband_heavy

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: deband_heavyNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090NVIDIA RTX 4070 SUPER5001000150020002500SE +/- 0.15, N = 3SE +/- 0.40, N = 3SE +/- 2.90, N = 3SE +/- 2.26, N = 31844.082306.562015.932186.701. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF

RealSR-NCNN

Scale: 4x - TAA: Yes

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30901020304050SE +/- 0.02, N = 3SE +/- 0.23, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 334.8942.8533.6330.31

VkFFT

Test: FFT + iFFT C2C 1D batched in half precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in half precisionNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309060K120K180K240K300KSE +/- 159.17, N = 3SE +/- 1301.92, N = 3SE +/- 1708.38, N = 3SE +/- 160.60, N = 31317051377621362102732211. (CXX) g++ options: -O3 -lrt

NAMD CUDA

ATPase Simulation - 327,506 Atoms

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 AtomsNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.02430.04860.07290.09720.1215SE +/- 0.00031, N = 3SE +/- 0.00021, N = 3SE +/- 0.00061, N = 3SE +/- 0.00042, N = 30.067910.074980.067880.10822

Blender

Blend File: Fishy Cat - Compute: NVIDIA OptiX

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Fishy Cat - Compute: NVIDIA OptiXNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30903691215SE +/- 0.06, N = 13SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 99.4511.039.0210.64

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_lNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 6.65, N = 5SE +/- 0.13, N = 3102.60102.9096.5099.05MIN: 94.84 / MAX: 104.25MIN: 95.98 / MAX: 104.54MIN: 64.35 / MAX: 104.79MIN: 91.8 / MAX: 100.69

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_lNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 1.49, N = 2SE +/- 0.45, N = 3SE +/- 0.39, N = 2SE +/- 0.14, N = 3102.60101.55103.2099.84MIN: 79.69 / MAX: 105.28MIN: 93.44 / MAX: 103.08MIN: 95.31 / MAX: 105.27MIN: 92.73 / MAX: 101.46

TensorFlow

Device: GPU - Batch Size: 1 - Model: ResNet-50

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 1 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.97881.95762.93643.91524.894SE +/- 0.01, N = 3SE +/- 0.02, N = 2SE +/- 0.03, N = 34.354.344.324.35

ViennaCL

Test: CPU BLAS - dGEMM-TT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090306090120150SE +/- 2.08, N = 3SE +/- 1.20, N = 3SE +/- 2.08, N = 3SE +/- 0.88, N = 31221181241131. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMM-TN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TNNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090306090120150SE +/- 1.00, N = 2SE +/- 2.31, N = 3SE +/- 2.08, N = 3SE +/- 2.08, N = 31151211251211. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMM-NT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090306090120150SE +/- 2.08, N = 3SE +/- 1.76, N = 3SE +/- 1.20, N = 3SE +/- 3.28, N = 31171221181191. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMM-NN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NNNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090306090120150SE +/- 4.04, N = 3SE +/- 1.86, N = 3SE +/- 1.15, N = 3SE +/- 1.86, N = 31191221171131. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMV-T

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-TNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 6.30, N = 3SE +/- 0.33, N = 3109.0109.0102.7110.01. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMV-N

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-NNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.88, N = 31021031031031. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 0.09, N = 3SE +/- 0.22, N = 3SE +/- 0.58, N = 3SE +/- 0.84, N = 396.896.796.495.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPYNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 0.12, N = 3SE +/- 0.44, N = 3SE +/- 0.57, N = 3SE +/- 0.94, N = 387.286.887.386.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPYNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30901632486480SE +/- 0.32, N = 3SE +/- 0.25, N = 3SE +/- 0.74, N = 3SE +/- 0.72, N = 370.871.071.370.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30904080120160200SE +/- 2.73, N = 3SE +/- 3.76, N = 3SE +/- 2.40, N = 3SE +/- 35.40, N = 3165.0166.0168.0132.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPYNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090306090120150SE +/- 2.19, N = 3SE +/- 4.81, N = 3SE +/- 2.00, N = 3SE +/- 0.33, N = 31561531561541. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPYNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090306090120150SE +/- 1.20, N = 3SE +/- 1.20, N = 3SE +/- 0.88, N = 3SE +/- 1.20, N = 31321311321321. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_lNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 0.39, N = 3SE +/- 0.36, N = 2SE +/- 0.19, N = 3103.57101.43103.5099.25MIN: 95.95 / MAX: 105.54MIN: 93.27 / MAX: 103.58MIN: 94.95 / MAX: 105.61MIN: 91.16 / MAX: 101.18

ViennaCL

Test: OpenCL BLAS - dGEMM-TT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090140280420560700SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 36135026485931. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dGEMM-TN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090140280420560700SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 2.03, N = 35994946345941. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dGEMM-NT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090130260390520650SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 2.33, N = 35844776125951. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dGEMM-NN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090130260390520650SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 2.31, N = 35774736045921. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dGEMV-T

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-TNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309080160240320400SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 33893873913741. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dGEMV-N

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-NNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309050100150200250SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 32102092111871. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090140280420560700SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.88, N = 34584564576591. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090160320480640800SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 34374554377241. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPYNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090130260390520650SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.58, N = 34234234246051. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - sDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309080160240320400SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.58, N = 33703623653761. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - sAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPYNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090110220330440550SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.58, N = 33923893934981. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - sCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPYNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309080160240320400SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 1.00, N = 33343303363631. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_lNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 0.05, N = 2SE +/- 0.57, N = 3103.17101.24103.2499.43MIN: 95.79 / MAX: 105.15MIN: 93.33 / MAX: 102.92MIN: 95.41 / MAX: 104.9MIN: 90.49 / MAX: 101.97

TensorFlow

Device: GPU - Batch Size: 1 - Model: AlexNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 1 - Model: AlexNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.22, N = 2SE +/- 0.16, N = 3SE +/- 0.06, N = 2SE +/- 0.20, N = 1513.9214.0414.7914.45

Blender

Blend File: BMW27 - Compute: NVIDIA OptiX

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: BMW27 - Compute: NVIDIA OptiXNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090246810SE +/- 0.06, N = 13SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 145.576.215.436.31

Blender

Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 314.2916.5513.9717.30

Blender

Blend File: Classroom - Compute: NVIDIA OptiX

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Classroom - Compute: NVIDIA OptiXNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 312.6014.8612.3015.26

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_lNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 0.52, N = 2SE +/- 0.53, N = 2103.68103.4598.11MIN: 96.86 / MAX: 105.56MIN: 95.22 / MAX: 105.88MIN: 89.88 / MAX: 100.25

ProjectPhysX OpenCL-Benchmark

Operation: Memory Bandwidth Coalesced Write

OpenBenchmarking.orgGB/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: Memory Bandwidth Coalesced WriteNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30902004006008001000SE +/- 0.14, N = 3SE +/- 0.16, N = 3SE +/- 0.11, N = 3SE +/- 0.06, N = 3455.01459.43457.17887.311. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

ProjectPhysX OpenCL-Benchmark

Operation: Memory Bandwidth Coalesced Read

OpenBenchmarking.orgGB/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: Memory Bandwidth Coalesced ReadNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30902004006008001000SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3464.86465.18465.07864.111. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

ProjectPhysX OpenCL-Benchmark

Operation: INT8 Compute

OpenBenchmarking.orgTIOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: INT8 ComputeNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 314.3112.1215.7313.731. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

ProjectPhysX OpenCL-Benchmark

Operation: INT16 Compute

OpenBenchmarking.orgTIOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: INT16 ComputeNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309048121620SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 317.1714.2818.2817.001. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

ProjectPhysX OpenCL-Benchmark

Operation: INT32 Compute

OpenBenchmarking.orgTIOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: INT32 ComputeNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090510152025SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.06, N = 319.8916.3821.0520.031. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

ProjectPhysX OpenCL-Benchmark

Operation: INT64 Compute

OpenBenchmarking.orgTIOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: INT64 ComputeNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.99451.9892.98353.9784.9725SE +/- 0.015, N = 3SE +/- 0.004, N = 3SE +/- 0.016, N = 3SE +/- 0.003, N = 34.2143.4434.4203.1351. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

ProjectPhysX OpenCL-Benchmark

Operation: FP32 Compute

OpenBenchmarking.orgTFLOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: FP32 ComputeNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090918273645SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.10, N = 338.5931.7740.9139.401. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

ProjectPhysX OpenCL-Benchmark

Operation: FP64 Compute

OpenBenchmarking.orgTFLOPs/s, More Is BetterProjectPhysX OpenCL-Benchmark 1.2Operation: FP64 ComputeNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.14850.2970.44550.5940.7425SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 30.6210.5100.6600.6371. (CXX) g++ options: -std=c++17 -pthread -lOpenCL

RealSR-NCNN

Scale: 4x - TAA: No

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090246810SE +/- 0.150, N = 15SE +/- 0.006, N = 3SE +/- 0.039, N = 3SE +/- 0.016, N = 36.3237.0925.9625.556

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30904080120160200SE +/- 1.38, N = 2SE +/- 0.05, N = 3SE +/- 0.33, N = 2195.30187.51194.87164.35MIN: 182 / MAX: 199.43MIN: 181.57 / MAX: 188.05MIN: 180.8 / MAX: 198MIN: 149.91 / MAX: 166.09

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30904080120160200SE +/- 0.51, N = 3SE +/- 0.34, N = 3SE +/- 0.78, N = 2196.07186.63197.02164.14MIN: 171.95 / MAX: 199.96MIN: 180.51 / MAX: 187.79MIN: 183.92 / MAX: 200.54MIN: 149 / MAX: 165

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30904080120160200SE +/- 1.14, N = 2SE +/- 0.17, N = 3SE +/- 0.19, N = 2194.58187.27195.86161.01MIN: 183.74 / MAX: 198.52MIN: 179.9 / MAX: 188.08MIN: 181.64 / MAX: 199.2MIN: 138.12 / MAX: 165.16

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_lNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309020406080100SE +/- 0.55, N = 3SE +/- 0.33, N = 3106.37107.59108.59105.55MIN: 97.91 / MAX: 108.16MIN: 98.77 / MAX: 109.43MIN: 99.04 / MAX: 110.68MIN: 91.76 / MAX: 107.42

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30904080120160200SE +/- 0.29, N = 3195.39187.69198.82163.74MIN: 183.94 / MAX: 198.7MIN: 182.03 / MAX: 188.31MIN: 188.33 / MAX: 201.47MIN: 144.93 / MAX: 165.03

TensorFlow

Device: GPU - Batch Size: 1 - Model: GoogLeNet

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: GPU - Batch Size: 1 - Model: GoogLeNetNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30903691215SE +/- 0.17, N = 2SE +/- 0.10, N = 3SE +/- 0.30, N = 2SE +/- 0.07, N = 312.6212.7812.7912.82

VkResample

Upscale: 2x - Precision: Single

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090510152025SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 318.4918.0218.4610.321. (CXX) g++ options: -O3

clpeak

OpenCL Test: Double-Precision Double

OpenBenchmarking.orgGFLOPS, More Is Betterclpeak 1.1.2OpenCL Test: Double-Precision DoubleNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090140280420560700SE +/- 0.98, N = 3SE +/- 0.21, N = 3SE +/- 1.33, N = 3SE +/- 1.63, N = 3630.11515.17667.05642.231. (CXX) g++ options: -O3

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30904080120160200SE +/- 0.29, N = 3195.40187.26194.29164.14MIN: 186.09 / MAX: 197.7MIN: 179.81 / MAX: 188.21MIN: 182.25 / MAX: 197.39MIN: 145.67 / MAX: 165.38

LuxCoreRender

Scene: Rainbow Colors and Prism - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Rainbow Colors and Prism - Acceleration: GPUNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090816243240SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.36, N = 527.6723.2627.7133.29MIN: 24.87 / MAX: 29.03MIN: 20.92 / MAX: 24.3MIN: 25.01 / MAX: 29.15MIN: 30.4 / MAX: 36.21

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090120240360480600SE +/- 3.09, N = 3SE +/- 11.16, N = 12557.73546.76535.39525.12MIN: 513.63 / MAX: 563.37MIN: 195.25 / MAX: 556.94MIN: 428.43 / MAX: 572.99MIN: 458.54 / MAX: 542.46

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090110220330440550SE +/- 0.92, N = 3SE +/- 0.27, N = 3SE +/- 1.92, N = 3SE +/- 0.24, N = 3507.45458.36505.62419.03MIN: 423.41 / MAX: 512.88MIN: 404.89 / MAX: 461.01MIN: 426.6 / MAX: 513.25MIN: 376 / MAX: 422

Rodinia

Test: OpenCL Particle Filter

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle FilterNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.92211.84422.76633.68844.6105SE +/- 0.039, N = 4SE +/- 0.008, N = 3SE +/- 0.002, N = 3SE +/- 0.030, N = 153.4804.0983.2913.8441. (CXX) g++ options: -O2 -lOpenCL

Hashcat

Benchmark: MD5

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: MD5NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309016000M32000M48000M64000M80000MSE +/- 22430807.19, N = 3SE +/- 33772046.30, N = 3SE +/- 11283665.68, N = 3SE +/- 53667246.37, N = 367583033333561478666677331223333367177300000

Hashcat

Benchmark: SHA1

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA1NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30905000M10000M15000M20000M25000MSE +/- 5140363.15, N = 3SE +/- 6318315.53, N = 3SE +/- 15926811.78, N = 3SE +/- 26244639.66, N = 322132600000182024666672353240000021323733333

Hashcat

Benchmark: SHA-512

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA-512NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090700M1400M2100M2800M3500MSE +/- 1530068.99, N = 3SE +/- 1059874.21, N = 3SE +/- 721110.26, N = 3SE +/- 3288532.26, N = 33232733333267330000034625000003081866667

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 3090110220330440550SE +/- 1.39, N = 3SE +/- 0.34, N = 3SE +/- 0.14, N = 3504.67459.93416.89MIN: 412.34 / MAX: 514.07MIN: 403.65 / MAX: 462.74MIN: 329.77 / MAX: 420.82

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30904080120160200SE +/- 0.36, N = 3SE +/- 0.73, N = 3SE +/- 0.09, N = 2201.94198.18201.19197.12MIN: 183.53 / MAX: 206.5MIN: 181.27 / MAX: 200.06MIN: 180.79 / MAX: 203.92MIN: 137.37 / MAX: 198.9

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090110220330440550SE +/- 0.26, N = 3SE +/- 2.23, N = 3SE +/- 0.89, N = 2509.45458.39502.92419.76MIN: 430.1 / MAX: 516.48MIN: 404.5 / MAX: 461.01MIN: 415.65 / MAX: 520.39MIN: 376.2 / MAX: 422.17

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090110220330440550SE +/- 4.43, N = 2SE +/- 0.43, N = 2SE +/- 0.83, N = 2SE +/- 0.40, N = 3504.27459.27504.66416.20MIN: 418.22 / MAX: 512.44MIN: 405.48 / MAX: 461.88MIN: 424.27 / MAX: 509.08MIN: 355.45 / MAX: 419.05

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50NVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090110220330440550SE +/- 2.17, N = 2SE +/- 0.13, N = 2SE +/- 1.69, N = 3501.50459.94505.55420.29MIN: 415.94 / MAX: 510.69MIN: 403.65 / MAX: 462.59MIN: 419.93 / MAX: 512.69MIN: 376.81 / MAX: 421.58

Hashcat

Benchmark: TrueCrypt RIPEMD160 + XTS

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090200K400K600K800K1000KSE +/- 633.33, N = 3SE +/- 176.38, N = 3SE +/- 888.82, N = 3SE +/- 1757.21, N = 3802967660967858600797833

cl-mem

Benchmark: Copy

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: CopyNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 309080160240320400SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.00, N = 3SE +/- 0.22, N = 3331.8330.3333.3360.81. (CC) gcc options: -O2 -flto -lOpenCL

cl-mem

Benchmark: Write

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: WriteNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090160320480640800SE +/- 1.11, N = 3SE +/- 0.55, N = 3SE +/- 0.12, N = 3SE +/- 0.83, N = 3407.5406.7412.2753.81. (CC) gcc options: -O2 -flto -lOpenCL

cl-mem

Benchmark: Read

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: ReadNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30902004006008001000SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.32, N = 3446.2446.3446.3825.81. (CC) gcc options: -O2 -flto -lOpenCL

Hashcat

Benchmark: 7-Zip

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: 7-ZipNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090300K600K900K1200K1500KSE +/- 1991.93, N = 3SE +/- 2062.63, N = 3SE +/- 2339.04, N = 3SE +/- 1587.45, N = 3117646797696712626331056000

Waifu2x-NCNN Vulkan

Scale: 2x - Denoise: 3 - TAA: Yes

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30900.72051.4412.16152.8823.6025SE +/- 0.014, N = 3SE +/- 0.028, N = 3SE +/- 0.009, N = 3SE +/- 0.011, N = 32.8553.1682.8543.202

clpeak

OpenCL Test: Global Memory Bandwidth

OpenBenchmarking.orgGBPS, More Is Betterclpeak 1.1.2OpenCL Test: Global Memory BandwidthNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30902004006008001000SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3437.65437.21437.63816.551. (CXX) g++ options: -O3

MandelGPU

OpenCL Device: GPU

OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPUNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090130M260M390M520M650MSE +/- 467034.80, N = 3SE +/- 1783157.89, N = 3SE +/- 1202791.77, N = 3SE +/- 794770.01, N = 3587219538.2516770131.2619106132.5484098913.81. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

FinanceBench

Benchmark: Black-Scholes OpenCL

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCLNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 3090246810SE +/- 0.114, N = 15SE +/- 0.003, N = 3SE +/- 0.003, N = 3SE +/- 0.006, N = 35.9126.9065.2265.7411. (CXX) g++ options: -O3 -march=native -fopenmp

clpeak

OpenCL Test: Integer Compute INT

OpenBenchmarking.orgGIOPS, More Is Betterclpeak 1.1.2OpenCL Test: Integer Compute INTNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30904K8K12K16K20KSE +/- 3.14, N = 3SE +/- 15.26, N = 3SE +/- 2.50, N = 3SE +/- 16.49, N = 318170.5414555.1919821.1017923.331. (CXX) g++ options: -O3

clpeak

OpenCL Test: Single-Precision Float

OpenBenchmarking.orgGFLOPS, More Is Betterclpeak 1.1.2OpenCL Test: Single-Precision FloatNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30908K16K24K32K40KSE +/- 0.99, N = 3SE +/- 5.46, N = 3SE +/- 11.67, N = 3SE +/- 113.39, N = 335492.6928479.3938691.7334906.791. (CXX) g++ options: -O3

NeatBench

Acceleration: GPU

OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: GPUNVIDIA RTX 4070 SUPERNVIDIA RTX 4070NVIDIA RTX 4070 TINVIDIA RTX 30909001800270036004500SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 34070407040703090


Phoronix Test Suite v10.8.5