RTX 4070 SUPER Intel Core i9-13900K testing with a ASUS TUF GAMING Z790-PRO WIFI (1630 BIOS) and ASUS NVIDIA GeForce RTX 4070 Ti SUPER 16GB on EndeavourOS rolling via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2402174-SADD-240211636&grs&sor .
RTX 4070 SUPER Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL OpenCL Compiler File-System Screen Resolution NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER Intel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads) ASUS TUF GAMING Z790-PRO WIFI (1401 BIOS) Intel Device 7a27 32GB 4001GB Seagate ZP4000GP304001 ASUS NVIDIA GeForce RTX 4070 SUPER 12GB Realtek ALC1220 ARZOPA Intel I226-V + Intel Device 7a70 EndeavourOS rolling 6.7.1-arch1-1 (x86_64) KDE Plasma 5.27.10 X Server 1.21.1.11 NVIDIA 550.40.07 4.6.0 OpenCL 3.0 CUDA 12.4.74 GCC 13.2.1 20230801 ext4 1920x1080 MSI NVIDIA GeForce RTX 4070 12GB GCC 13.2.1 20230801 + CUDA 12.3 NVIDIA GeForce RTX 4070 Ti 12GB NVIDIA GeForce RTX 3090 24GB PI-KVM Video 6.7.4-arch1-1 (x86_64) ASUS TUF GAMING Z790-PRO WIFI (1630 BIOS) Intel Raptor Lake-S PCH 4001GB Seagate ZP4000GP304001 + 0GB CD-ROM Drive ASUS NVIDIA GeForce RTX 4070 Ti SUPER 16GB Intel I226-V + Intel Raptor Lake-S PCH CNVi WiFi OpenCL 2.1 AMD-APP (3602.0) + OpenCL 3.0 CUDA 12.4.74 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: always Compiler 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 - NVIDIA RTX 4070 TI 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,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 - NVIDIA RTX 4070 SUPER: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x11d - NVIDIA RTX 4070: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x11d - NVIDIA RTX 4070 TI: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x11d - NVIDIA RTX 3090: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x11d - NVIDIA RTX 4070 TI SUPER: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x11f Graphics 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.ba - NVIDIA RTX 4070 TI SUPER: BAR1 / Visible vRAM Size: 256 MiB - vBIOS Version: 95.03.45.00.c5 Security 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, NVIDIA RTX 4070 TI SUPER: NVCC_PREPEND_FLAGS="-ccbin /opt/cuda/bin" Python Details - NVIDIA RTX 4070: Python 3.11.6 - NVIDIA RTX 4070 TI: Python 3.11.6 - NVIDIA RTX 3090: Python 3.11.6 - NVIDIA RTX 4070 TI SUPER: Python 3.11.7
RTX 4070 SUPER opencl-benchmark: Memory Bandwidth Coalesced Write vkfft: FFT + iFFT C2C 1D batched in single precision, no reshuffling vkfft: FFT + iFFT C2C 1D batched in single precision clpeak: Global Memory Bandwidth opencl-benchmark: Memory Bandwidth Coalesced Read cl-mem: Write cl-mem: Read vkresample: 2x - Single viennacl: OpenCL BLAS - dAXPY namd-cuda: ATPase Simulation - 327,506 Atoms libplacebo: hdr_peakdetect clpeak: Integer Compute INT clpeak: Single-Precision Float hashcat: MD5 opencl-benchmark: FP64 Compute clpeak: Double-Precision Double viennacl: OpenCL BLAS - dGEMM-TT hashcat: TrueCrypt RIPEMD160 + XTS hashcat: 7-Zip hashcat: SHA-512 opencl-benchmark: INT8 Compute vkresample: 2x - Double gpuowl: 332220523 hashcat: SHA1 opencl-benchmark: FP32 Compute viennacl: OpenCL BLAS - dGEMM-TN viennacl: OpenCL BLAS - dDOT opencl-benchmark: INT32 Compute viennacl: OpenCL BLAS - dGEMM-NT viennacl: OpenCL BLAS - dGEMM-NN gpuowl: 77936867 opencl-benchmark: INT16 Compute gpuowl: 57885161 luxcorerender: Rainbow Colors and Prism - GPU viennacl: OpenCL BLAS - dCOPY realsr-ncnn: 4x - Yes opencl-benchmark: INT64 Compute luxcorerender: Danish Mood - GPU vkfft: FFT + iFFT C2C 1D batched in double precision rodinia: OpenCL Particle Filter blender: Pabellon Barcelona - NVIDIA OptiX viennacl: CPU BLAS - dDOT blender: Classroom - NVIDIA OptiX viennacl: CPU BLAS - dAXPY mandelgpu: GPU libplacebo: deband_heavy viennacl: CPU BLAS - dCOPY octanebench: Total Score indigobench: OpenCL GPU - Bedroom libplacebo: polar_nocompute luxcorerender: LuxCore Benchmark - GPU blender: Fishy Cat - NVIDIA OptiX blender: Barbershop - NVIDIA OptiX viennacl: CPU BLAS - dGEMV-N luxcorerender: Orange Juice - GPU viennacl: CPU BLAS - sAXPY vkfft: FFT + iFFT C2C Bluestein benchmark in double precision vkpeak: int16-vec4 viennacl: OpenCL BLAS - sAXPY pytorch: NVIDIA CUDA GPU - 512 - ResNet-50 pytorch: NVIDIA CUDA GPU - 256 - ResNet-50 pytorch: NVIDIA CUDA GPU - 32 - ResNet-50 pytorch: NVIDIA CUDA GPU - 16 - ResNet-50 vkfft: FFT + iFFT C2C multidimensional in single precision indigobench: OpenCL GPU - Supercar vkfft: FFT + iFFT R2C / C2R pytorch: NVIDIA CUDA GPU - 64 - ResNet-50 blender: BMW27 - NVIDIA OptiX fahbench: pytorch: NVIDIA CUDA GPU - 256 - ResNet-152 viennacl: CPU BLAS - sCOPY pytorch: NVIDIA CUDA GPU - 32 - ResNet-152 pytorch: NVIDIA CUDA GPU - 16 - ResNet-152 tensorflow: GPU - 1 - AlexNet pytorch: NVIDIA CUDA GPU - 512 - ResNet-152 waifu2x-ncnn: 2x - 3 - Yes vkpeak: int16-scalar pytorch: NVIDIA CUDA GPU - 64 - ResNet-152 libplacebo: hdr_lut vkpeak: fp16-vec4 vkpeak: fp32-vec4 vkpeak: fp16-scalar vkpeak: int32-vec4 vkpeak: fp32-scalar vkpeak: int32-scalar vkfft: FFT + iFFT C2C Bluestein in single precision vkpeak: fp64-vec4 vkpeak: fp64-scalar viennacl: OpenCL BLAS - dGEMV-N viennacl: OpenCL BLAS - dGEMV-T viennacl: OpenCL BLAS - sDOT viennacl: OpenCL BLAS - sCOPY cl-mem: Copy viennacl: CPU BLAS - dGEMM-TT viennacl: CPU BLAS - dGEMM-TN viennacl: CPU BLAS - dGEMM-NN pytorch: NVIDIA CUDA GPU - 16 - Efficientnet_v2_l tensorflow: GPU - 1 - ResNet-50 tensorflow: GPU - 1 - GoogLeNet tensorflow: GPU - 32 - GoogLeNet tensorflow: GPU - 1 - VGG-16 tensorflow: GPU - 64 - ResNet-50 pytorch: NVIDIA CUDA GPU - 512 - Efficientnet_v2_l viennacl: CPU BLAS - dGEMM-NT tensorflow: GPU - 64 - GoogLeNet tensorflow: GPU - 32 - ResNet-50 pytorch: NVIDIA CUDA GPU - 256 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 64 - Efficientnet_v2_l tensorflow: GPU - 16 - VGG-16 tensorflow: GPU - 64 - VGG-16 tensorflow: GPU - 16 - ResNet-50 libplacebo: av1_grain_lap pytorch: NVIDIA CUDA GPU - 1 - Efficientnet_v2_l tensorflow: GPU - 16 - AlexNet tensorflow: GPU - 32 - VGG-16 tensorflow: GPU - 256 - VGG-16 tensorflow: GPU - 16 - GoogLeNet pytorch: NVIDIA CUDA GPU - 1 - ResNet-152 tensorflow: GPU - 32 - AlexNet tensorflow: GPU - 256 - AlexNet tensorflow: GPU - 512 - AlexNet tensorflow: GPU - 64 - AlexNet ncnn: Vulkan GPU - FastestDet ncnn: Vulkan GPU - vision_transformer ncnn: Vulkan GPU - regnety_400m ncnn: Vulkan GPU - squeezenet_ssd ncnn: Vulkan GPU - yolov4-tiny ncnn: Vulkan GPU - resnet50 ncnn: Vulkan GPU - alexnet ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU - vgg16 ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - blazeface ncnn: Vulkan GPU - efficientnet-b0 ncnn: Vulkan GPU - mnasnet ncnn: Vulkan GPU - shufflenet-v2 ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 ncnn: Vulkan GPU - mobilenet neatbench: GPU viennacl: CPU BLAS - dGEMV-T viennacl: CPU BLAS - sDOT financebench: Black-Scholes OpenCL luxcorerender: DLSC - GPU vkfft: FFT + iFFT C2C 1D batched in half precision realsr-ncnn: 4x - No pytorch: NVIDIA CUDA GPU - 32 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 1 - ResNet-50 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER 455.01 75078 73929 437.65 464.86 407.5 446.2 18.489 437 0.06791 3292.37 18170.54 35492.69 67583033333 0.621 630.11 613 802967 1176467 3232733333 14.307 339.593 137.44 22132600000 38.594 599 458 19.889 584 577 646.41 17.170 869.07 27.67 423 34.885 4.214 10.56 24317 3.480 14.29 96.8 12.60 87.2 587219538.2 2186.70 70.8 720.973789 19.801 2327.55 12.82 9.45 51.30 102 11.72 156 4451 392 504.27 504.67 501.50 509.45 50299 52.813 54794 507.45 5.57 366.0576 194.58 132 195.39 195.40 13.92 195.30 2.855 196.07 3905.98 15166 210 389 370 334 331.8 122 115 119 4.35 12.62 15.61 1.35 5.55 103.57 117 15.52 5.51 103.17 102.60 1.48 5.46 4171.00 106.37 31.59 1.50 15.67 201.94 33.4 34.16 35.10 33.97 2.86 844.61 11.11 6.86 63.82 46.26 16.17 8.97 117.81 11.04 0.84 5.07 3.85 2.31 2.25 3.03 8.62 4070 109 165 5.912 13.59 131705 6.323 102.60 557.73 459.43 79057 77774 437.21 465.18 406.7 446.3 18.016 455 0.07498 3329.26 14555.19 28479.39 56147866667 0.510 515.17 502 660967 976967 2673300000 12.116 415.160 112.61 18202466667 31.768 494 456 16.377 477 473 530.32 14.284 714.80 23.26 423 42.852 3.443 8.89 22390 4.098 16.55 96.7 14.86 86.8 516770131.2 1843.26 71.0 647.997867 18.203 1968.37 10.92 11.03 58.44 103 10.40 153 3886 389 459.27 459.93 459.94 458.39 47212 48.517 47097 458.36 6.21 317.1952 187.27 131 187.69 187.26 14.04 187.51 3.168 186.63 3946.90 13714 209 387 362 330 330.3 118 121 122 103.68 4.34 12.78 15.63 1.36 5.55 101.43 122 15.54 5.55 101.24 101.55 1.50 1.50 5.49 4152.41 107.59 31.45 1.5 15.66 198.18 33.32 35.21 33.93 2.67 382.82 6.21 5.18 20.74 8.72 5.78 5.11 45.52 6.06 0.84 3.59 2.24 2.08 2.15 2.48 7.20 4070 109 166 6.906 11.74 137762 7.092 102.90 546.76 457.17 75141 73942 437.63 465.07 412.2 446.3 18.456 437 0.06788 3475.06 19821.10 38691.73 73312233333 0.660 667.05 648 858600 1262633 3462500000 15.731 322.064 145.84 23532400000 40.914 634 457 21.047 612 604 676.59 18.281 919.13 27.71 424 33.626 4.420 10.99 25431 3.291 13.97 96.4 12.30 87.3 619106132.5 2306.56 71.3 735.940593 20.256 2459.03 13.23 9.02 50.73 103 11.89 156 4647 393 504.66 505.55 502.92 51528 53.589 55446 505.62 5.43 382.1637 195.86 132 198.82 194.29 14.79 194.87 2.854 197.02 3976.04 15125 211 391 365 336 333.3 124 125 117 103.45 4.32 12.79 15.81 1.38 5.53 103.50 118 15.50 5.50 103.24 103.20 1.49 1.5 5.46 4143.96 108.59 31.70 1.5 1.5 15.69 201.19 33.29 34.61 35.44 34.06 3.04 497.66 5.89 6.13 16.37 12.25 6.07 7.74 34.49 7.37 0.82 3.46 4.14 2.01 2.09 2.54 7.45 4070 102.7 168 5.226 13.95 136210 5.962 96.50 535.39 887.31 144311 141876 816.55 864.11 753.8 825.8 10.323 724 0.10822 5055.88 17923.33 34906.79 67177300000 0.637 642.23 593 797833 1056000 3081866667 13.727 333.639 137.32 21323733333 39.395 594 659 20.027 595 592 645.99 17.001 866.31 33.29 605 30.313 3.135 10.20 30912 3.844 17.30 95.2 15.26 86.2 484098913.8 2020.16 70.2 674.250912 20.959 2116.79 13.12 10.64 54.30 103 12.14 154 4195 16329.72 498 416.20 416.89 420.29 419.76 50856 52.014 48418 419.03 6.31 343.0199 161.01 132 163.74 164.14 14.45 164.35 3.202 13264.91 164.14 3369.88 39860.80 26699.66 20151.44 20009.73 20353.95 20295.27 14205 638.74 638.84 187 374 376 363 360.8 113 121 113 98.11 4.35 12.82 15.67 1.38 5.57 99.25 119 15.63 5.57 99.43 99.84 1.49 1.51 5.49 4100.36 105.55 31.98 1.5 1.51 15.68 197.12 33.53 34.46 35.58 33.93 2.65 354.57 6.73 4.90 11.29 12.70 3.60 4.12 17.88 6.14 0.87 3.34 2.16 2.04 2.21 2.34 7.27 3090 110 132.1 5.741 12.99 273221 5.556 99.05 525.12 608.94 105549 104003 582.84 619.03 551.9 595.2 13.363 585 0.07715 3913.34 22171.25 43244.79 82004966667 0.743 750.36 731 961733 1420700 3887033333 17.615 285.988 163.41 26388600000 45.950 714 575 23.660 689 681 761.61 20.503 1025.99 31.86 512 30.724 4.414 12.42 27947 2.973 12.56 70.8 11.20 64.3 656484783.7 2495.92 52.7 876.436994 24.570 2653.03 14.61 8.32 44.49 78.5 13.64 120 5047 21156.99 469 529.49 529.14 532.77 531.96 59790 61.338 59378 527.82 5.04 394.7356 198.70 107 197.82 198.58 12.26 198.01 2.660 15901.32 196.50 3822.16 47340.52 31635.47 23894.70 23768.27 23920.67 23888.02 16141 750.68 750.49 218 424 410 373 370.7 117 120 122 103.66 4.14 12.24 15.11 1.32 5.33 103.53 119 15.00 5.35 102.83 103.49 1.45 1.46 5.32 4044.72 105.86 31.10 1.46 1.47 15.29 200.46 32.88 33.95 35.02 33.55 2.55 312.10 6.59 5.19 17.20 8.79 4.41 7.58 24.85 6.46 0.88 3.48 2.26 2.13 2.16 2.70 7.48 2084.1 82.6 129 0.501 16.23 143992 5.633 102.83 558.82 OpenBenchmarking.org
ProjectPhysX OpenCL-Benchmark Operation: Memory Bandwidth Coalesced Write OpenBenchmarking.org GB/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.2 Operation: Memory Bandwidth Coalesced Write NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 200 400 600 800 1000 SE +/- 0.06, N = 3 SE +/- 0.57, N = 3 SE +/- 0.16, N = 3 SE +/- 0.11, N = 3 SE +/- 0.14, N = 3 887.31 608.94 459.43 457.17 455.01 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
VkFFT Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.2.31 Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 30K 60K 90K 120K 150K SE +/- 37.44, N = 3 SE +/- 20.80, N = 3 SE +/- 5.84, N = 3 SE +/- 28.54, N = 3 SE +/- 37.77, N = 3 144311 105549 79057 75141 75078 1. (CXX) g++ options: -O3 -lrt
VkFFT Test: FFT + iFFT C2C 1D batched in single precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.2.31 Test: FFT + iFFT C2C 1D batched in single precision NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 30K 60K 90K 120K 150K SE +/- 9.64, N = 3 SE +/- 33.60, N = 3 SE +/- 13.72, N = 3 SE +/- 0.88, N = 3 SE +/- 7.94, N = 3 141876 104003 77774 73942 73929 1. (CXX) g++ options: -O3 -lrt
clpeak OpenCL Test: Global Memory Bandwidth OpenBenchmarking.org GBPS, More Is Better clpeak 1.1.2 OpenCL Test: Global Memory Bandwidth NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 200 400 600 800 1000 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 816.55 582.84 437.65 437.63 437.21 1. (CXX) g++ options: -O3
ProjectPhysX OpenCL-Benchmark Operation: Memory Bandwidth Coalesced Read OpenBenchmarking.org GB/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.2 Operation: Memory Bandwidth Coalesced Read NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 200 400 600 800 1000 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 864.11 619.03 465.18 465.07 464.86 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
cl-mem Benchmark: Write OpenBenchmarking.org GB/s, More Is Better cl-mem 2017-01-13 Benchmark: Write NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 160 320 480 640 800 SE +/- 0.83, N = 3 SE +/- 0.25, N = 3 SE +/- 0.12, N = 3 SE +/- 1.11, N = 3 SE +/- 0.55, N = 3 753.8 551.9 412.2 407.5 406.7 1. (CC) gcc options: -O2 -flto -lOpenCL
cl-mem Benchmark: Read OpenBenchmarking.org GB/s, More Is Better cl-mem 2017-01-13 Benchmark: Read NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER 200 400 600 800 1000 SE +/- 0.32, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.12, N = 3 825.8 595.2 446.3 446.3 446.2 1. (CC) gcc options: -O2 -flto -lOpenCL
VkResample Upscale: 2x - Precision: Single OpenBenchmarking.org ms, Fewer Is Better VkResample 1.0 Upscale: 2x - Precision: Single NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 5 10 15 20 25 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 10.32 13.36 18.02 18.46 18.49 1. (CXX) g++ options: -O3
ViennaCL Test: OpenCL BLAS - dAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dAXPY NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 160 320 480 640 800 SE +/- 0.58, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 724 585 455 437 437 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
NAMD CUDA ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD CUDA 2.14 ATPase Simulation - 327,506 Atoms NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 0.0243 0.0486 0.0729 0.0972 0.1215 SE +/- 0.00061, N = 3 SE +/- 0.00031, N = 3 SE +/- 0.00021, N = 3 SE +/- 0.00018, N = 3 SE +/- 0.00042, N = 3 0.06788 0.06791 0.07498 0.07715 0.10822
Libplacebo Test: hdr_peakdetect OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: hdr_peakdetect NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER 1100 2200 3300 4400 5500 SE +/- 13.97, N = 3 SE +/- 28.18, N = 3 SE +/- 99.97, N = 3 SE +/- 144.09, N = 3 SE +/- 3.65, N = 3 5104.10 3931.57 3544.60 3452.43 3292.37 1. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF
clpeak OpenCL Test: Integer Compute INT OpenBenchmarking.org GIOPS, More Is Better clpeak 1.1.2 OpenCL Test: Integer Compute INT NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 5K 10K 15K 20K 25K SE +/- 28.14, N = 3 SE +/- 2.50, N = 3 SE +/- 3.14, N = 3 SE +/- 16.49, N = 3 SE +/- 15.26, N = 3 22171.25 19821.10 18170.54 17923.33 14555.19 1. (CXX) g++ options: -O3
clpeak OpenCL Test: Single-Precision Float OpenBenchmarking.org GFLOPS, More Is Better clpeak 1.1.2 OpenCL Test: Single-Precision Float NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 9K 18K 27K 36K 45K SE +/- 50.25, N = 3 SE +/- 11.67, N = 3 SE +/- 0.99, N = 3 SE +/- 113.39, N = 3 SE +/- 5.46, N = 3 43244.79 38691.73 35492.69 34906.79 28479.39 1. (CXX) g++ options: -O3
Hashcat Benchmark: MD5 OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: MD5 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 20000M 40000M 60000M 80000M 100000M SE +/- 97655010.68, N = 3 SE +/- 11283665.68, N = 3 SE +/- 22430807.19, N = 3 SE +/- 53667246.37, N = 3 SE +/- 33772046.30, N = 3 82004966667 73312233333 67583033333 67177300000 56147866667
ProjectPhysX OpenCL-Benchmark Operation: FP64 Compute OpenBenchmarking.org TFLOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.2 Operation: FP64 Compute NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 0.1672 0.3344 0.5016 0.6688 0.836 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 0.743 0.660 0.637 0.621 0.510 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
clpeak OpenCL Test: Double-Precision Double OpenBenchmarking.org GFLOPS, More Is Better clpeak 1.1.2 OpenCL Test: Double-Precision Double NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 160 320 480 640 800 SE +/- 1.26, N = 3 SE +/- 1.33, N = 3 SE +/- 1.63, N = 3 SE +/- 0.98, N = 3 SE +/- 0.21, N = 3 750.36 667.05 642.23 630.11 515.17 1. (CXX) g++ options: -O3
ViennaCL Test: OpenCL BLAS - dGEMM-TT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-TT NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 160 320 480 640 800 SE +/- 1.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 731 648 613 593 502 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Hashcat Benchmark: TrueCrypt RIPEMD160 + XTS OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: TrueCrypt RIPEMD160 + XTS NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 200K 400K 600K 800K 1000K SE +/- 392.99, N = 3 SE +/- 888.82, N = 3 SE +/- 633.33, N = 3 SE +/- 1757.21, N = 3 SE +/- 176.38, N = 3 961733 858600 802967 797833 660967
Hashcat Benchmark: 7-Zip OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: 7-Zip NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 300K 600K 900K 1200K 1500K SE +/- 1628.91, N = 3 SE +/- 2339.04, N = 3 SE +/- 1991.93, N = 3 SE +/- 1587.45, N = 3 SE +/- 2062.63, N = 3 1420700 1262633 1176467 1056000 976967
Hashcat Benchmark: SHA-512 OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: SHA-512 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 800M 1600M 2400M 3200M 4000M SE +/- 1098989.43, N = 3 SE +/- 721110.26, N = 3 SE +/- 1530068.99, N = 3 SE +/- 3288532.26, N = 3 SE +/- 1059874.21, N = 3 3887033333 3462500000 3232733333 3081866667 2673300000
ProjectPhysX OpenCL-Benchmark Operation: INT8 Compute OpenBenchmarking.org TIOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.2 Operation: INT8 Compute NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 17.62 15.73 14.31 13.73 12.12 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
VkResample Upscale: 2x - Precision: Double OpenBenchmarking.org ms, Fewer Is Better VkResample 1.0 Upscale: 2x - Precision: Double NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 90 180 270 360 450 SE +/- 0.02, N = 3 SE +/- 0.35, N = 3 SE +/- 0.30, N = 3 SE +/- 0.30, N = 3 SE +/- 0.77, N = 3 285.99 322.06 333.64 339.59 415.16 1. (CXX) g++ options: -O3
GpuOwl Exponent: 332220523 OpenBenchmarking.org Iterations / Second, More Is Better GpuOwl 7.2.1 Exponent: 332220523 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 40 80 120 160 200 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 163.41 145.84 137.44 137.32 112.61
Hashcat Benchmark: SHA1 OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: SHA1 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 6000M 12000M 18000M 24000M 30000M SE +/- 29067564.97, N = 3 SE +/- 15926811.78, N = 3 SE +/- 5140363.15, N = 3 SE +/- 26244639.66, N = 3 SE +/- 6318315.53, N = 3 26388600000 23532400000 22132600000 21323733333 18202466667
ProjectPhysX OpenCL-Benchmark Operation: FP32 Compute OpenBenchmarking.org TFLOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.2 Operation: FP32 Compute NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 10 20 30 40 50 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 45.95 40.91 39.40 38.59 31.77 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMM-TN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-TN NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 150 300 450 600 750 SE +/- 1.00, N = 3 SE +/- 0.67, N = 3 SE +/- 0.00, N = 3 SE +/- 2.03, N = 3 SE +/- 0.33, N = 3 714 634 599 594 494 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dDOT NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 140 280 420 560 700 SE +/- 0.88, N = 3 SE +/- 1.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 659 575 458 457 456 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ProjectPhysX OpenCL-Benchmark Operation: INT32 Compute OpenBenchmarking.org TIOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.2 Operation: INT32 Compute NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 23.66 21.05 20.03 19.89 16.38 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMM-NT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-NT NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 150 300 450 600 750 SE +/- 1.00, N = 3 SE +/- 0.33, N = 3 SE +/- 2.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 689 612 595 584 477 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMM-NN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMM-NN NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 150 300 450 600 750 SE +/- 1.33, N = 3 SE +/- 0.33, N = 3 SE +/- 2.31, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 681 604 592 577 473 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
GpuOwl Exponent: 77936867 OpenBenchmarking.org Iterations / Second, More Is Better GpuOwl 7.2.1 Exponent: 77936867 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 160 320 480 640 800 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.09, N = 3 761.61 676.59 646.41 645.99 530.32
ProjectPhysX OpenCL-Benchmark Operation: INT16 Compute OpenBenchmarking.org TIOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.2 Operation: INT16 Compute NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 20.50 18.28 17.17 17.00 14.28 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
GpuOwl Exponent: 57885161 OpenBenchmarking.org Iterations / Second, More Is Better GpuOwl 7.2.1 Exponent: 57885161 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 200 400 600 800 1000 SE +/- 0.35, N = 3 SE +/- 2.53, N = 3 SE +/- 1.26, N = 3 SE +/- 2.01, N = 3 SE +/- 0.00, N = 3 1025.99 919.13 869.07 866.31 714.80
LuxCoreRender Scene: Rainbow Colors and Prism - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: GPU NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 8 16 24 32 40 SE +/- 0.36, N = 5 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 33.29 31.86 27.71 27.67 23.26 MIN: 30.4 / MAX: 36.21 MIN: 28.57 / MAX: 33.29 MIN: 25.01 / MAX: 29.15 MIN: 24.87 / MAX: 29.03 MIN: 20.92 / MAX: 24.3
ViennaCL Test: OpenCL BLAS - dCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dCOPY NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER 130 260 390 520 650 SE +/- 0.58, N = 3 SE +/- 4.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 605 512 424 423 423 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
RealSR-NCNN Scale: 4x - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 10 20 30 40 50 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.23, N = 3 30.31 30.72 33.63 34.89 42.85
ProjectPhysX OpenCL-Benchmark Operation: INT64 Compute OpenBenchmarking.org TIOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.2 Operation: INT64 Compute NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 0.9945 1.989 2.9835 3.978 4.9725 SE +/- 0.016, N = 3 SE +/- 0.009, N = 3 SE +/- 0.015, N = 3 SE +/- 0.004, N = 3 SE +/- 0.003, N = 3 4.420 4.414 4.214 3.443 3.135 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
LuxCoreRender Scene: Danish Mood - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: GPU NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 12.42 10.99 10.56 10.20 8.89 MIN: 4.35 / MAX: 14.32 MIN: 4.17 / MAX: 12.71 MIN: 3.7 / MAX: 12.17 MIN: 4.07 / MAX: 11.93 MIN: 3.32 / MAX: 10.26
VkFFT Test: FFT + iFFT C2C 1D batched in double precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.2.31 Test: FFT + iFFT C2C 1D batched in double precision NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 7K 14K 21K 28K 35K SE +/- 50.66, N = 3 SE +/- 325.03, N = 3 SE +/- 302.46, N = 3 SE +/- 146.69, N = 3 SE +/- 125.94, N = 3 30912 27947 25431 24317 22390 1. (CXX) g++ options: -O3 -lrt
Rodinia Test: OpenCL Particle Filter OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenCL Particle Filter NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 0.9221 1.8442 2.7663 3.6884 4.6105 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 SE +/- 0.039, N = 4 SE +/- 0.030, N = 15 SE +/- 0.008, N = 3 2.973 3.291 3.480 3.844 4.098 1. (CXX) g++ options: -O2 -lOpenCL
Blender Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.0 Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 12.56 13.97 14.29 16.55 17.30
ViennaCL Test: CPU BLAS - dDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dDOT NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER 20 40 60 80 100 SE +/- 0.09, N = 3 SE +/- 0.22, N = 3 SE +/- 0.58, N = 3 SE +/- 0.84, N = 3 SE +/- 0.19, N = 3 96.8 96.7 96.4 95.2 70.8 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Blender Blend File: Classroom - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.0 Blend File: Classroom - Compute: NVIDIA OptiX NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 11.20 12.30 12.60 14.86 15.26
ViennaCL Test: CPU BLAS - dAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dAXPY NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER 20 40 60 80 100 SE +/- 0.57, N = 3 SE +/- 0.12, N = 3 SE +/- 0.44, N = 3 SE +/- 0.94, N = 3 SE +/- 0.12, N = 3 87.3 87.2 86.8 86.2 64.3 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
MandelGPU OpenCL Device: GPU OpenBenchmarking.org Samples/sec, More Is Better MandelGPU 1.3pts1 OpenCL Device: GPU NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 140M 280M 420M 560M 700M SE +/- 1096202.13, N = 3 SE +/- 1202791.77, N = 3 SE +/- 467034.80, N = 3 SE +/- 1783157.89, N = 3 SE +/- 794770.01, N = 3 656484783.7 619106132.5 587219538.2 516770131.2 484098913.8 1. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
Libplacebo Test: deband_heavy OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: deband_heavy NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 500 1000 1500 2000 2500 SE +/- 0.75, N = 3 SE +/- 0.56, N = 3 SE +/- 2.26, N = 3 SE +/- 2.92, N = 3 SE +/- 0.08, N = 3 2495.92 2306.67 2186.70 2024.61 1847.98 1. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF
ViennaCL Test: CPU BLAS - dCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dCOPY NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER 16 32 48 64 80 SE +/- 0.74, N = 3 SE +/- 0.25, N = 3 SE +/- 0.32, N = 3 SE +/- 0.72, N = 3 SE +/- 0.18, N = 3 71.3 71.0 70.8 70.2 52.7 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OctaneBench Total Score OpenBenchmarking.org Score, More Is Better OctaneBench 2020.1 Total Score NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 200 400 600 800 1000 876.44 735.94 720.97 674.25 648.00
IndigoBench Acceleration: OpenCL GPU - Scene: Bedroom OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: OpenCL GPU - Scene: Bedroom NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 24.57 20.96 20.26 19.80 18.20
Libplacebo Test: polar_nocompute OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: polar_nocompute NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 600 1200 1800 2400 3000 SE +/- 1.94, N = 3 SE +/- 0.26, N = 3 SE +/- 0.24, N = 3 SE +/- 3.45, N = 3 SE +/- 0.16, N = 3 2653.03 2461.23 2327.55 2126.31 1972.78 1. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF
LuxCoreRender Scene: LuxCore Benchmark - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: LuxCore Benchmark - Acceleration: GPU NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 2 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 14.61 13.23 13.12 12.82 10.92 MIN: 5.91 / MAX: 16.88 MIN: 5.41 / MAX: 15.13 MIN: 4.85 / MAX: 15.21 MIN: 4.84 / MAX: 14.62 MIN: 4.45 / MAX: 12.42
Blender Blend File: Fishy Cat - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.0 Blend File: Fishy Cat - Compute: NVIDIA OptiX NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 3 6 9 12 15 SE +/- 0.06, N = 13 SE +/- 0.01, N = 3 SE +/- 0.06, N = 13 SE +/- 0.08, N = 9 SE +/- 0.03, N = 3 8.32 9.02 9.45 10.64 11.03
Blender Blend File: Barbershop - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.0 Blend File: Barbershop - Compute: NVIDIA OptiX NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 13 26 39 52 65 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.10, N = 3 SE +/- 0.02, N = 2 SE +/- 0.04, N = 3 44.49 50.73 51.30 54.30 58.44
ViennaCL Test: CPU BLAS - dGEMV-N OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-N NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 20 40 60 80 100 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.46, N = 3 103.0 103.0 103.0 102.0 78.5 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
LuxCoreRender Scene: Orange Juice - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: GPU NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 4 8 12 16 20 SE +/- 0.15, N = 4 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 13.64 12.14 11.89 11.72 10.40 MIN: 11.16 / MAX: 18.46 MIN: 10.24 / MAX: 16.71 MIN: 9.85 / MAX: 15.88 MIN: 9.6 / MAX: 15.44 MIN: 8.31 / MAX: 13.9
ViennaCL Test: CPU BLAS - sAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sAXPY NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER 30 60 90 120 150 SE +/- 2.00, N = 3 SE +/- 2.19, N = 3 SE +/- 0.33, N = 3 SE +/- 4.81, N = 3 SE +/- 0.33, N = 3 156 156 154 153 120 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
VkFFT Test: FFT + iFFT C2C Bluestein benchmark in double precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.2.31 Test: FFT + iFFT C2C Bluestein benchmark in double precision NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 1100 2200 3300 4400 5500 SE +/- 11.37, N = 3 SE +/- 11.35, N = 3 SE +/- 12.55, N = 3 SE +/- 9.84, N = 3 SE +/- 4.51, N = 3 5047 4647 4451 4195 3886 1. (CXX) g++ options: -O3 -lrt
vkpeak int16-vec4 OpenBenchmarking.org GIOPS, More Is Better vkpeak 20230730 int16-vec4 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 5K 10K 15K 20K 25K SE +/- 1.60, N = 3 SE +/- 9.58, N = 3 21156.99 16338.23
ViennaCL Test: OpenCL BLAS - sAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - sAXPY NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 110 220 330 440 550 SE +/- 0.58, N = 3 SE +/- 1.20, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 498 469 393 392 389 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 110 220 330 440 550 SE +/- 1.16, N = 3 SE +/- 0.83, N = 2 SE +/- 4.43, N = 2 SE +/- 0.43, N = 2 SE +/- 0.40, N = 3 529.49 504.66 504.27 459.27 416.20 MIN: 410.12 / MAX: 537.25 MIN: 424.27 / MAX: 509.08 MIN: 418.22 / MAX: 512.44 MIN: 405.48 / MAX: 461.88 MIN: 355.45 / MAX: 419.05
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 110 220 330 440 550 SE +/- 0.54, N = 3 SE +/- 1.39, N = 3 SE +/- 0.34, N = 3 SE +/- 0.14, N = 3 529.14 504.67 459.93 416.89 MIN: 414.54 / MAX: 534.65 MIN: 412.34 / MAX: 514.07 MIN: 403.65 / MAX: 462.74 MIN: 329.77 / MAX: 420.82
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 120 240 360 480 600 SE +/- 0.70, N = 3 SE +/- 1.69, N = 3 SE +/- 2.17, N = 2 SE +/- 0.13, N = 2 532.77 505.55 501.50 459.94 420.29 MIN: 420.31 / MAX: 538.98 MIN: 419.93 / MAX: 512.69 MIN: 415.94 / MAX: 510.69 MIN: 403.65 / MAX: 462.59 MIN: 376.81 / MAX: 421.58
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 3090 120 240 360 480 600 SE +/- 1.33, N = 3 SE +/- 2.23, N = 3 SE +/- 0.26, N = 3 SE +/- 0.89, N = 2 531.96 509.45 502.92 458.39 419.76 MIN: 422.98 / MAX: 539.81 MIN: 430.1 / MAX: 516.48 MIN: 415.65 / MAX: 520.39 MIN: 404.5 / MAX: 461.01 MIN: 376.2 / MAX: 422.17
VkFFT Test: FFT + iFFT C2C multidimensional in single precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.2.31 Test: FFT + iFFT C2C multidimensional in single precision NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 13K 26K 39K 52K 65K SE +/- 251.10, N = 3 SE +/- 417.77, N = 15 SE +/- 407.28, N = 15 SE +/- 407.19, N = 15 SE +/- 476.57, N = 5 59790 51528 50856 50299 47212 1. (CXX) g++ options: -O3 -lrt
IndigoBench Acceleration: OpenCL GPU - Scene: Supercar OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: OpenCL GPU - Scene: Supercar NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 14 28 42 56 70 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 61.34 53.59 52.81 52.01 48.52
VkFFT Test: FFT + iFFT R2C / C2R OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.2.31 Test: FFT + iFFT R2C / C2R NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 13K 26K 39K 52K 65K SE +/- 772.47, N = 15 SE +/- 520.37, N = 3 SE +/- 702.53, N = 15 SE +/- 320.62, N = 3 SE +/- 745.02, N = 13 59378 55446 54794 48418 47097 1. (CXX) g++ options: -O3 -lrt
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 3090 110 220 330 440 550 SE +/- 1.58, N = 3 SE +/- 0.92, N = 3 SE +/- 1.92, N = 3 SE +/- 0.27, N = 3 SE +/- 0.24, N = 3 527.82 507.45 505.62 458.36 419.03 MIN: 419.39 / MAX: 534.44 MIN: 423.41 / MAX: 512.88 MIN: 426.6 / MAX: 513.25 MIN: 404.89 / MAX: 461.01 MIN: 376 / MAX: 422
Blender Blend File: BMW27 - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.0 Blend File: BMW27 - Compute: NVIDIA OptiX NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 2 4 6 8 10 SE +/- 0.06, N = 14 SE +/- 0.02, N = 3 SE +/- 0.06, N = 13 SE +/- 0.01, N = 3 SE +/- 0.06, N = 14 5.04 5.43 5.57 6.21 6.31
FAHBench OpenBenchmarking.org Ns Per Day, More Is Better FAHBench 2.3.2 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 90 180 270 360 450 SE +/- 0.22, N = 3 SE +/- 0.26, N = 3 SE +/- 0.39, N = 3 SE +/- 0.26, N = 3 SE +/- 0.12, N = 3 394.74 382.16 366.06 343.02 317.20
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 40 80 120 160 200 SE +/- 0.95, N = 3 SE +/- 0.19, N = 2 SE +/- 1.14, N = 2 SE +/- 0.17, N = 3 198.70 195.86 194.58 187.27 161.01 MIN: 185.21 / MAX: 203.36 MIN: 181.64 / MAX: 199.2 MIN: 183.74 / MAX: 198.52 MIN: 179.9 / MAX: 188.08 MIN: 138.12 / MAX: 165.16
ViennaCL Test: CPU BLAS - sCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sCOPY NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER 30 60 90 120 150 SE +/- 1.20, N = 3 SE +/- 0.88, N = 3 SE +/- 1.20, N = 3 SE +/- 1.20, N = 3 SE +/- 0.67, N = 3 132 132 132 131 107 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152 NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 40 80 120 160 200 SE +/- 0.28, N = 3 SE +/- 0.29, N = 3 198.82 197.82 195.39 187.69 163.74 MIN: 188.33 / MAX: 201.47 MIN: 176.19 / MAX: 201.63 MIN: 183.94 / MAX: 198.7 MIN: 182.03 / MAX: 188.31 MIN: 144.93 / MAX: 165.03
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 3090 40 80 120 160 200 SE +/- 0.33, N = 3 SE +/- 0.29, N = 3 198.58 195.40 194.29 187.26 164.14 MIN: 183.91 / MAX: 201.98 MIN: 186.09 / MAX: 197.7 MIN: 182.25 / MAX: 197.39 MIN: 179.81 / MAX: 188.21 MIN: 145.67 / MAX: 165.38
TensorFlow Device: GPU - Batch Size: 1 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 1 - Model: AlexNet NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 4 8 12 16 20 SE +/- 0.06, N = 2 SE +/- 0.20, N = 15 SE +/- 0.16, N = 3 SE +/- 0.22, N = 2 SE +/- 0.13, N = 15 14.79 14.45 14.04 13.92 12.26
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 3090 40 80 120 160 200 SE +/- 0.81, N = 3 SE +/- 1.38, N = 2 SE +/- 0.05, N = 3 SE +/- 0.33, N = 2 198.01 195.30 194.87 187.51 164.35 MIN: 185.3 / MAX: 202.59 MIN: 182 / MAX: 199.43 MIN: 180.8 / MAX: 198 MIN: 181.57 / MAX: 188.05 MIN: 149.91 / MAX: 166.09
Waifu2x-NCNN Vulkan Scale: 2x - Denoise: 3 - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better Waifu2x-NCNN Vulkan 20200818 Scale: 2x - Denoise: 3 - TAA: Yes NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 0.7205 1.441 2.1615 2.882 3.6025 SE +/- 0.028, N = 3 SE +/- 0.009, N = 3 SE +/- 0.014, N = 3 SE +/- 0.028, N = 3 SE +/- 0.011, N = 3 2.660 2.854 2.855 3.168 3.202
vkpeak int16-scalar OpenBenchmarking.org GIOPS, More Is Better vkpeak 20230730 int16-scalar NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 3K 6K 9K 12K 15K SE +/- 0.75, N = 3 SE +/- 10.34, N = 3 15901.32 13273.53
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152 NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 40 80 120 160 200 SE +/- 0.78, N = 2 SE +/- 0.20, N = 3 SE +/- 0.51, N = 3 SE +/- 0.34, N = 3 197.02 196.50 196.07 186.63 164.14 MIN: 183.92 / MAX: 200.54 MIN: 179.34 / MAX: 200 MIN: 171.95 / MAX: 199.96 MIN: 180.51 / MAX: 187.79 MIN: 149 / MAX: 165
Libplacebo Test: hdr_lut OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: hdr_lut NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 900 1800 2700 3600 4500 SE +/- 33.96, N = 3 SE +/- 6.47, N = 3 SE +/- 12.09, N = 3 SE +/- 17.88, N = 3 SE +/- 22.23, N = 3 3976.04 3946.90 3905.98 3845.51 3376.85 1. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF
vkpeak fp16-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20230730 fp16-vec4 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 10K 20K 30K 40K 50K SE +/- 0.16, N = 3 SE +/- 12.74, N = 3 47340.52 39860.80
vkpeak fp32-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20230730 fp32-vec4 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 7K 14K 21K 28K 35K SE +/- 0.73, N = 3 SE +/- 206.35, N = 3 31635.47 26767.21
vkpeak fp16-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20230730 fp16-scalar NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 5K 10K 15K 20K 25K SE +/- 0.21, N = 3 SE +/- 34.29, N = 3 23894.70 20151.44
vkpeak int32-vec4 OpenBenchmarking.org GIOPS, More Is Better vkpeak 20230730 int32-vec4 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 5K 10K 15K 20K 25K SE +/- 0.99, N = 3 SE +/- 15.96, N = 3 23768.27 20017.06
vkpeak fp32-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20230730 fp32-scalar NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 5K 10K 15K 20K 25K SE +/- 5.58, N = 3 SE +/- 123.67, N = 3 23920.67 20353.95
vkpeak int32-scalar OpenBenchmarking.org GIOPS, More Is Better vkpeak 20230730 int32-scalar NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 5K 10K 15K 20K 25K SE +/- 0.08, N = 3 SE +/- 16.76, N = 3 23888.02 20315.10
VkFFT Test: FFT + iFFT C2C Bluestein in single precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.2.31 Test: FFT + iFFT C2C Bluestein in single precision NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 3K 6K 9K 12K 15K SE +/- 73.00, N = 3 SE +/- 102.52, N = 3 SE +/- 118.41, N = 3 SE +/- 115.62, N = 3 SE +/- 52.09, N = 3 16141 15166 15125 14205 13714 1. (CXX) g++ options: -O3 -lrt
vkpeak fp64-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20230730 fp64-vec4 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 160 320 480 640 800 SE +/- 0.45, N = 3 SE +/- 0.72, N = 3 750.68 639.52
vkpeak fp64-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20230730 fp64-scalar NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 160 320 480 640 800 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 750.49 638.84
ViennaCL Test: OpenCL BLAS - dGEMV-N OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMV-N NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 50 100 150 200 250 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 218 211 210 209 187 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - dGEMV-T OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - dGEMV-T NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 90 180 270 360 450 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 424 391 389 387 374 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - sDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - sDOT NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 90 180 270 360 450 SE +/- 1.00, N = 3 SE +/- 0.58, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 410 376 370 365 362 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: OpenCL BLAS - sCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: OpenCL BLAS - sCOPY NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 80 160 240 320 400 SE +/- 0.33, N = 3 SE +/- 1.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 373 363 336 334 330 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
cl-mem Benchmark: Copy OpenBenchmarking.org GB/s, More Is Better cl-mem 2017-01-13 Benchmark: Copy NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 80 160 240 320 400 SE +/- 0.00, N = 3 SE +/- 0.22, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 370.7 360.8 333.3 331.8 330.3 1. (CC) gcc options: -O2 -flto -lOpenCL
ViennaCL Test: CPU BLAS - dGEMM-TT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TT NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 30 60 90 120 150 SE +/- 2.08, N = 3 SE +/- 2.08, N = 3 SE +/- 1.20, N = 3 SE +/- 2.91, N = 3 SE +/- 0.88, N = 3 124 122 118 117 113 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dGEMM-TN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TN NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER 30 60 90 120 150 SE +/- 2.08, N = 3 SE +/- 2.08, N = 3 SE +/- 2.31, N = 3 SE +/- 3.00, N = 2 SE +/- 1.00, N = 2 125 121 121 120 115 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - dGEMM-NN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NN NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 30 60 90 120 150 SE +/- 1.50, N = 2 SE +/- 1.86, N = 3 SE +/- 4.04, N = 3 SE +/- 1.15, N = 3 SE +/- 1.86, N = 3 122 122 119 117 113 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 20 40 60 80 100 SE +/- 0.33, N = 3 SE +/- 0.52, N = 2 SE +/- 0.53, N = 2 103.68 103.66 103.45 98.11 MIN: 96.86 / MAX: 105.56 MIN: 93.46 / MAX: 105.95 MIN: 95.22 / MAX: 105.88 MIN: 89.88 / MAX: 100.25
TensorFlow Device: GPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 1 - Model: ResNet-50 NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER 0.9788 1.9576 2.9364 3.9152 4.894 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 2 SE +/- 0.02, N = 3 4.35 4.35 4.34 4.32 4.14
TensorFlow Device: GPU - Batch Size: 1 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 1 - Model: GoogLeNet NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 3 6 9 12 15 SE +/- 0.07, N = 3 SE +/- 0.30, N = 2 SE +/- 0.10, N = 3 SE +/- 0.17, N = 2 SE +/- 0.05, N = 3 12.82 12.79 12.78 12.62 12.24
TensorFlow Device: GPU - Batch Size: 32 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 32 - Model: GoogLeNet NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 2 SE +/- 0.06, N = 3 15.81 15.67 15.63 15.61 15.11
TensorFlow Device: GPU - Batch Size: 1 - Model: VGG-16 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 1 - Model: VGG-16 NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 0.3105 0.621 0.9315 1.242 1.5525 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 2 SE +/- 0.00, N = 3 1.38 1.38 1.36 1.35 1.32
TensorFlow Device: GPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 64 - Model: ResNet-50 NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER 1.2533 2.5066 3.7599 5.0132 6.2665 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 2 SE +/- 0.01, N = 2 SE +/- 0.02, N = 3 5.57 5.55 5.55 5.53 5.33
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 3090 20 40 60 80 100 SE +/- 0.36, N = 2 SE +/- 0.39, N = 3 SE +/- 0.19, N = 3 103.57 103.53 103.50 101.43 99.25 MIN: 95.95 / MAX: 105.54 MIN: 88.81 / MAX: 104.8 MIN: 94.95 / MAX: 105.61 MIN: 93.27 / MAX: 103.58 MIN: 91.16 / MAX: 101.18
ViennaCL Test: CPU BLAS - dGEMM-NT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NT NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 30 60 90 120 150 SE +/- 1.76, N = 3 SE +/- 3.50, N = 2 SE +/- 3.28, N = 3 SE +/- 1.20, N = 3 SE +/- 2.08, N = 3 122 119 119 118 117 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
TensorFlow Device: GPU - Batch Size: 64 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 64 - Model: GoogLeNet NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER 4 8 12 16 20 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 2 SE +/- 0.09, N = 3 15.63 15.54 15.52 15.50 15.00
TensorFlow Device: GPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 32 - Model: ResNet-50 NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER 1.2533 2.5066 3.7599 5.0132 6.2665 SE +/- 0.01, N = 3 SE +/- 0.01, N = 2 SE +/- 0.01, N = 2 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 5.57 5.55 5.51 5.50 5.35
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 20 40 60 80 100 SE +/- 0.18, N = 3 SE +/- 0.05, N = 2 SE +/- 0.57, N = 3 103.24 103.17 102.83 101.24 99.43 MIN: 95.41 / MAX: 104.9 MIN: 95.79 / MAX: 105.15 MIN: 93.16 / MAX: 105.07 MIN: 93.33 / MAX: 102.92 MIN: 90.49 / MAX: 101.97
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 20 40 60 80 100 SE +/- 0.13, N = 3 SE +/- 0.39, N = 2 SE +/- 1.49, N = 2 SE +/- 0.45, N = 3 SE +/- 0.14, N = 3 103.49 103.20 102.60 101.55 99.84 MIN: 93.23 / MAX: 105.43 MIN: 95.31 / MAX: 105.27 MIN: 79.69 / MAX: 105.28 MIN: 93.44 / MAX: 103.08 MIN: 92.73 / MAX: 101.46
TensorFlow Device: GPU - Batch Size: 16 - Model: VGG-16 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 16 - Model: VGG-16 NVIDIA RTX 4070 NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 0.3375 0.675 1.0125 1.35 1.6875 SE +/- 0.01, N = 2 SE +/- 0.00, N = 3 SE +/- 0.00, N = 2 SE +/- 0.00, N = 3 1.50 1.49 1.49 1.48 1.45
TensorFlow Device: GPU - Batch Size: 64 - Model: VGG-16 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 64 - Model: VGG-16 NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER 0.3398 0.6796 1.0194 1.3592 1.699 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.51 1.50 1.50 1.46
TensorFlow Device: GPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 16 - Model: ResNet-50 NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 1.2353 2.4706 3.7059 4.9412 6.1765 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 2 SE +/- 0.02, N = 3 5.49 5.49 5.46 5.46 5.32
Libplacebo Test: av1_grain_lap OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: av1_grain_lap NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER 900 1800 2700 3600 4500 SE +/- 5.52, N = 3 SE +/- 16.20, N = 3 SE +/- 35.33, N = 3 SE +/- 12.99, N = 3 SE +/- 39.01, N = 3 4171.00 4152.41 4143.96 4126.89 4057.41 1. (CXX) g++ options: -lm -pthread -ldl -fvisibility=hidden -std=c++20 -O2 -fno-math-errno -fPIC -MD -MQ -MF
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 20 40 60 80 100 SE +/- 0.55, N = 3 SE +/- 0.24, N = 2 SE +/- 0.33, N = 3 108.59 107.59 106.37 105.86 105.55 MIN: 99.04 / MAX: 110.68 MIN: 98.77 / MAX: 109.43 MIN: 97.91 / MAX: 108.16 MIN: 95.05 / MAX: 107.6 MIN: 91.76 / MAX: 107.42
TensorFlow Device: GPU - Batch Size: 16 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 16 - Model: AlexNet NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER 7 14 21 28 35 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.17, N = 3 SE +/- 0.07, N = 3 31.98 31.70 31.59 31.45 31.10
TensorFlow Device: GPU - Batch Size: 32 - Model: VGG-16 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 32 - Model: VGG-16 NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 0.3375 0.675 1.0125 1.35 1.6875 SE +/- 0.00, N = 3 SE +/- 0.00, N = 2 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.50 1.50 1.50 1.50 1.46
TensorFlow Device: GPU - Batch Size: 256 - Model: VGG-16 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 256 - Model: VGG-16 NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER 0.3398 0.6796 1.0194 1.3592 1.699 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.51 1.50 1.47
TensorFlow Device: GPU - Batch Size: 16 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 16 - Model: GoogLeNet NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER 4 8 12 16 20 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 15.69 15.68 15.67 15.66 15.29
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 40 80 120 160 200 SE +/- 0.73, N = 3 SE +/- 0.38, N = 3 SE +/- 0.36, N = 3 SE +/- 0.09, N = 2 201.94 201.19 200.46 198.18 197.12 MIN: 183.53 / MAX: 206.5 MIN: 180.79 / MAX: 203.92 MIN: 177.25 / MAX: 203.31 MIN: 181.27 / MAX: 200.06 MIN: 137.37 / MAX: 198.9
TensorFlow Device: GPU - Batch Size: 32 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 32 - Model: AlexNet NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER 8 16 24 32 40 SE +/- 0.05, N = 3 SE +/- 0.15, N = 2 SE +/- 0.18, N = 3 SE +/- 0.04, N = 3 SE +/- 0.19, N = 3 33.53 33.40 33.32 33.29 32.88
TensorFlow Device: GPU - Batch Size: 256 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 256 - Model: AlexNet NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 8 16 24 32 40 SE +/- 0.07, N = 2 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 34.61 34.46 34.16 33.95
TensorFlow Device: GPU - Batch Size: 512 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 512 - Model: AlexNet NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI SUPER 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.09, N = 2 SE +/- 0.03, N = 3 SE +/- 0.02, N = 2 SE +/- 0.01, N = 3 35.58 35.44 35.21 35.10 35.02
TensorFlow Device: GPU - Batch Size: 64 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: GPU - Batch Size: 64 - Model: AlexNet NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER 8 16 24 32 40 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.14, N = 3 SE +/- 0.06, N = 3 34.06 33.97 33.93 33.93 33.55
NCNN Target: Vulkan GPU - Model: FastestDet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: FastestDet NVIDIA RTX 4070 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 2 4 6 8 10 SE +/- 0.10, N = 9 SE +/- 0.08, N = 8 SE +/- 0.26, N = 3 SE +/- 0.12, N = 8 SE +/- 0.29, N = 9 2.34 2.50 2.54 2.84 2.86 MIN: 2 / MAX: 3.86 MIN: 2.1 / MAX: 32.36 MIN: 2.14 / MAX: 4.21 MIN: 2.4 / MAX: 5.07 MIN: 2.17 / MAX: 577.17 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: vision_transformer OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: vision_transformer NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 200 400 600 800 1000 SE +/- 61.31, N = 9 SE +/- 57.46, N = 12 SE +/- 52.80, N = 9 SE +/- 25.65, N = 9 SE +/- 87.53, N = 9 281.56 312.10 327.82 390.18 844.61 MIN: 46.48 / MAX: 1913.33 MIN: 47.85 / MAX: 1850.09 MIN: 46.48 / MAX: 1816.93 MIN: 46.49 / MAX: 1816.77 MIN: 46.34 / MAX: 1866.93 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: regnety_400m NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER 3 6 9 12 15 SE +/- 0.18, N = 12 SE +/- 0.24, N = 12 SE +/- 0.32, N = 8 SE +/- 0.26, N = 12 SE +/- 3.28, N = 9 5.89 6.21 6.47 6.59 11.11 MIN: 5.42 / MAX: 7.57 MIN: 5.53 / MAX: 8.99 MIN: 5.44 / MAX: 9.3 MIN: 5.45 / MAX: 9.09 MIN: 5.49 / MAX: 4942.19 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: squeezenet_ssd NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 2 4 6 8 10 SE +/- 0.19, N = 3 SE +/- 0.54, N = 3 SE +/- 0.12, N = 12 SE +/- 0.29, N = 9 SE +/- 1.76, N = 9 4.90 5.11 5.18 5.36 6.86 MIN: 4.47 / MAX: 5.27 MIN: 4.43 / MAX: 8.4 MIN: 4.67 / MAX: 6.88 MIN: 4.55 / MAX: 496.3 MIN: 4.34 / MAX: 1630.01 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: yolov4-tiny NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER 14 28 42 56 70 SE +/- 0.21, N = 3 SE +/- 3.14, N = 3 SE +/- 3.10, N = 12 SE +/- 5.37, N = 12 SE +/- 10.56, N = 9 11.29 14.26 16.37 20.74 63.82 MIN: 10.82 / MAX: 11.93 MIN: 10.89 / MAX: 673.37 MIN: 10.57 / MAX: 855.36 MIN: 10.3 / MAX: 854.36 MIN: 10.28 / MAX: 858.44 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: resnet50 NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 10 20 30 40 50 SE +/- 0.12, N = 9 SE +/- 0.10, N = 9 SE +/- 0.22, N = 3 SE +/- 4.00, N = 12 SE +/- 14.70, N = 9 8.20 8.24 8.58 12.25 46.26 MIN: 7.69 / MAX: 11.69 MIN: 7.87 / MAX: 9.87 MIN: 8.2 / MAX: 11.07 MIN: 8 / MAX: 1777.17 MIN: 7.71 / MAX: 1829.99 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: alexnet NVIDIA RTX 3090 NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER 4 8 12 16 20 SE +/- 0.09, N = 3 SE +/- 0.03, N = 9 SE +/- 0.03, N = 3 SE +/- 1.70, N = 12 SE +/- 5.86, N = 9 3.60 3.74 4.38 5.78 16.17 MIN: 3.44 / MAX: 3.79 MIN: 3.61 / MAX: 3.98 MIN: 4.29 / MAX: 6.18 MIN: 3.6 / MAX: 397.75 MIN: 3.52 / MAX: 436.52 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: resnet18 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 4 8 12 16 20 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.73, N = 12 SE +/- 1.33, N = 9 SE +/- 3.49, N = 9 4.12 4.64 5.11 5.47 8.97 MIN: 3.97 / MAX: 4.51 MIN: 4.46 / MAX: 7.98 MIN: 3.99 / MAX: 916.69 MIN: 3.95 / MAX: 726.67 MIN: 3.94 / MAX: 922.04 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: vgg16 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER 30 60 90 120 150 SE +/- 0.25, N = 3 SE +/- 0.19, N = 3 SE +/- 11.81, N = 9 SE +/- 13.24, N = 12 SE +/- 29.60, N = 9 17.88 21.76 32.05 45.52 117.81 MIN: 17.3 / MAX: 18.57 MIN: 21.34 / MAX: 23.45 MIN: 17.34 / MAX: 644.35 MIN: 17.49 / MAX: 643.35 MIN: 17.16 / MAX: 647.67 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: googlenet NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER 3 6 9 12 15 SE +/- 0.14, N = 9 SE +/- 0.14, N = 9 SE +/- 0.18, N = 9 SE +/- 0.24, N = 3 SE +/- 1.21, N = 9 5.87 6.06 6.11 6.25 11.04 MIN: 5.2 / MAX: 6.88 MIN: 5.33 / MAX: 8.36 MIN: 5.25 / MAX: 9.16 MIN: 5.84 / MAX: 6.86 MIN: 5.28 / MAX: 1769.19 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: blazeface NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER 0.198 0.396 0.594 0.792 0.99 SE +/- 0.03, N = 9 SE +/- 0.04, N = 9 SE +/- 0.03, N = 9 SE +/- 0.03, N = 9 SE +/- 0.05, N = 3 0.81 0.84 0.84 0.84 0.86 MIN: 0.61 / MAX: 1.19 MIN: 0.65 / MAX: 4.63 MIN: 0.64 / MAX: 0.96 MIN: 0.63 / MAX: 1.13 MIN: 0.75 / MAX: 2.68 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: efficientnet-b0 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 4 8 12 16 20 SE +/- 0.17, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 9 SE +/- 0.07, N = 12 SE +/- 0.97, N = 9 3.34 3.36 3.46 3.46 5.07 MIN: 3.14 / MAX: 4 MIN: 3.21 / MAX: 3.57 MIN: 2.91 / MAX: 3.79 MIN: 3.13 / MAX: 7.03 MIN: 3.22 / MAX: 1124.2 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: mnasnet NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 0.9315 1.863 2.7945 3.726 4.6575 SE +/- 0.14, N = 3 SE +/- 0.13, N = 3 SE +/- 0.08, N = 8 SE +/- 0.05, N = 9 SE +/- 1.31, N = 9 2.16 2.21 2.22 2.30 3.85 MIN: 2.01 / MAX: 2.55 MIN: 2.01 / MAX: 3.93 MIN: 1.83 / MAX: 2.54 MIN: 2.15 / MAX: 2.58 MIN: 1.89 / MAX: 1093.29 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: shufflenet-v2 NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER 0.9383 1.8766 2.8149 3.7532 4.6915 SE +/- 0.08, N = 12 SE +/- 0.21, N = 3 SE +/- 0.19, N = 3 SE +/- 0.09, N = 11 SE +/- 0.34, N = 8 2.01 2.04 2.05 2.08 2.31 MIN: 1.73 / MAX: 3.86 MIN: 1.8 / MAX: 5.8 MIN: 1.83 / MAX: 6.6 MIN: 1.82 / MAX: 2.59 MIN: 1.76 / MAX: 421.42 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.09, N = 9 SE +/- 0.08, N = 12 SE +/- 0.09, N = 9 SE +/- 0.16, N = 9 1.87 2.09 2.15 2.20 2.25 MIN: 1.81 / MAX: 5.21 MIN: 1.78 / MAX: 2.85 MIN: 1.81 / MAX: 2.58 MIN: 1.91 / MAX: 2.71 MIN: 1.75 / MAX: 343.7 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER 1.0553 2.1106 3.1659 4.2212 5.2765 SE +/- 0.15, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 9 SE +/- 0.07, N = 12 SE +/- 0.44, N = 9 2.34 2.42 2.43 2.48 3.03 MIN: 2.04 / MAX: 2.63 MIN: 2.24 / MAX: 9.23 MIN: 2.09 / MAX: 5.8 MIN: 2.02 / MAX: 5.82 MIN: 2.38 / MAX: 970.87 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: Vulkan GPU - Model: mobilenet NVIDIA RTX 4070 TI SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.22, N = 9 SE +/- 0.21, N = 12 SE +/- 0.25, N = 12 SE +/- 0.47, N = 9 6.28 6.92 7.20 7.45 8.62 MIN: 6.16 / MAX: 8.09 MIN: 6.06 / MAX: 8.65 MIN: 6.2 / MAX: 11.13 MIN: 6.87 / MAX: 734.65 MIN: 6.42 / MAX: 1101.3 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NeatBench Acceleration: GPU OpenBenchmarking.org FPS, More Is Better NeatBench 5 Acceleration: GPU NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER 900 1800 2700 3600 4500 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 512.75, N = 16 4070.0 4070.0 4070.0 3090.0 2084.1
ViennaCL Test: CPU BLAS - dGEMV-T OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-T NVIDIA RTX 3090 NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 TI SUPER 20 40 60 80 100 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 6.30, N = 3 SE +/- 0.47, N = 3 110.0 109.0 109.0 102.7 82.6 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
ViennaCL Test: CPU BLAS - sDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sDOT NVIDIA RTX 4070 TI NVIDIA RTX 4070 NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER 40 80 120 160 200 SE +/- 2.40, N = 3 SE +/- 3.76, N = 3 SE +/- 2.73, N = 3 SE +/- 35.40, N = 3 SE +/- 0.58, N = 3 168.0 166.0 165.0 132.1 129.0 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
FinanceBench Benchmark: Black-Scholes OpenCL OpenBenchmarking.org ms, Fewer Is Better FinanceBench 2016-07-25 Benchmark: Black-Scholes OpenCL NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 3090 NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 2 4 6 8 10 SE +/- 0.000, N = 3 SE +/- 0.003, N = 3 SE +/- 0.006, N = 3 SE +/- 0.114, N = 15 SE +/- 0.003, N = 3 0.501 5.226 5.741 5.912 6.906 1. (CXX) g++ options: -O3 -march=native -fopenmp
LuxCoreRender Scene: DLSC - Acceleration: GPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: GPU NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 1.13, N = 12 SE +/- 0.01, N = 3 16.23 13.95 13.59 12.99 11.74 MIN: 15.91 / MAX: 16.36 MIN: 13.67 / MAX: 14.14 MIN: 12.52 / MAX: 13.84 MIN: 0.52 / MAX: 14.69 MIN: 11.35 / MAX: 11.83
VkFFT Test: FFT + iFFT C2C 1D batched in half precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.2.31 Test: FFT + iFFT C2C 1D batched in half precision NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER 60K 120K 180K 240K 300K SE +/- 160.60, N = 3 SE +/- 3524.05, N = 12 SE +/- 1301.92, N = 3 SE +/- 1708.38, N = 3 SE +/- 159.17, N = 3 273221 143992 137762 136210 131705 1. (CXX) g++ options: -O3 -lrt
RealSR-NCNN Scale: 4x - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No NVIDIA RTX 3090 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 TI NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 2 4 6 8 10 SE +/- 0.016, N = 3 SE +/- 0.003, N = 3 SE +/- 0.039, N = 3 SE +/- 0.150, N = 15 SE +/- 0.006, N = 3 5.556 5.633 5.962 6.323 7.092
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l NVIDIA RTX 4070 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 3090 NVIDIA RTX 4070 TI 20 40 60 80 100 SE +/- 0.62, N = 3 SE +/- 0.13, N = 3 SE +/- 6.65, N = 5 102.90 102.83 102.60 99.05 96.50 MIN: 95.98 / MAX: 104.54 MIN: 92.44 / MAX: 105.47 MIN: 94.84 / MAX: 104.25 MIN: 91.8 / MAX: 100.69 MIN: 64.35 / MAX: 104.79
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50 NVIDIA RTX 4070 TI SUPER NVIDIA RTX 4070 SUPER NVIDIA RTX 4070 NVIDIA RTX 4070 TI NVIDIA RTX 3090 120 240 360 480 600 SE +/- 3.07, N = 3 SE +/- 3.09, N = 3 SE +/- 11.16, N = 12 558.82 557.73 546.76 535.39 525.12 MIN: 473.77 / MAX: 573.46 MIN: 513.63 / MAX: 563.37 MIN: 195.25 / MAX: 556.94 MIN: 428.43 / MAX: 572.99 MIN: 458.54 / MAX: 542.46
Phoronix Test Suite v10.8.5