Big 3900XT Summer AMD Ryzen 9 3900XT 12-Core testing with a MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS) and AMD Radeon RX 56/64 8GB on Ubuntu 20.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2107144-PTS-BIG3900X84&export=txt&grs&sro&rro .
Big 3900XT Summer Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Vulkan Compiler File-System Screen Resolution 1 2 3 4 AMD Ryzen 9 3900XT 12-Core @ 3.80GHz (12 Cores / 24 Threads) MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS) AMD Starship/Matisse 16GB 500GB Seagate FireCuda 520 SSD ZP500GM30002 AMD Radeon RX 56/64 8GB (1630/945MHz) AMD Vega 10 HDMI Audio ASUS MG28U Realtek Device 2600 + Realtek Device 3000 + Intel Wi-Fi 6 AX200 Ubuntu 20.10 5.11.0-rc1-phx (x86_64) 20201228 GNOME Shell 3.38.1 X Server 1.20.9 4.6 Mesa 20.2.1 (LLVM 11.0.0) 1.2.131 GCC 10.2.0 ext4 3840x2160 GCC 10.3.0 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - 1: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - 2: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - 3: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-poYruo/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-poYruo/gcc-10-10.3.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - 4: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-poYruo/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-poYruo/gcc-10-10.3.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021 Graphics Details - GLAMOR Java Details - OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2.20.10) Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Big 3900XT Summer build-gdb: Time To Compile build-ffmpeg: Time To Compile unvanquished: 800 x 600 - High srsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAM srsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAM unvanquished: 1280 x 1024 - Medium srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM unvanquished: 1024 x 768 - High srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAM unvanquished: 1920 x 1080 - Medium srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAM unvanquished: 1920 x 1200 - Medium vpxenc: Speed 0 - Bosphorus 1080p rocksdb: Rand Read unvanquished: 1280 x 1024 - High unvanquished: 1280 x 1024 - Ultra unvanquished: 2560 x 1440 - Ultra unvanquished: 1600 x 1200 - High gravitymark: 3840 x 2160 - Vulkan mnn: squeezenetv1.1 srsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAM srsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAM srsran: OFDM_Test unvanquished: 1024 x 768 - Medium ncnn: CPU - resnet18 unvanquished: 3840 x 2160 - Medium renaissance: Scala Dotty unvanquished: 1600 x 1200 - Ultra srsran: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM rocksdb: Read While Writing unvanquished: 800 x 600 - Medium srsran: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM renaissance: Genetic Algorithm Using Jenetics + Futures unvanquished: 800 x 600 - Ultra mnn: resnet-v2-50 renaissance: Rand Forest srsran: 4G PHY_DL_Test 100 PRB SISO 64-QAM ncnn: CPU - yolov4-tiny srsran: 4G PHY_DL_Test 100 PRB SISO 64-QAM blosc: blosclz unvanquished: 3840 x 2160 - Ultra renaissance: Savina Reactors.IO unvanquished: 1920 x 1200 - Ultra unvanquished: 1920 x 1200 - High mnn: MobileNetV2_224 rocksdb: Seq Fill unvanquished: 3840 x 2160 - High tnn: CPU - MobileNet v2 ncnn: CPU - vgg16 astcenc: Medium unvanquished: 2560 x 1440 - High rocksdb: Rand Fill unvanquished: 1024 x 768 - Ultra ncnn: CPU - blazeface gravitymark: 1600 x 1200 - Vulkan tnn: CPU - SqueezeNet v1.1 mnn: inception-v3 unvanquished: 1920 x 1080 - Ultra renaissance: In-Memory Database Shootout mnn: mobilenet-v1-1.0 ncnn: CPU - googlenet gravitymark: 2560 x 1440 - Vulkan ncnn: CPU - alexnet renaissance: Akka Unbalanced Cobwebbed Tree unvanquished: 1600 x 1200 - Medium ncnn: CPU - regnety_400m tnn: CPU - DenseNet gravitymark: 1920 x 1200 - Vulkan tnn: CPU - SqueezeNet v2 unvanquished: 1920 x 1080 - High ncnn: CPU - squeezenet_ssd unvanquished: 2560 x 1440 - Medium renaissance: Finagle HTTP Requests vpxenc: Speed 5 - Bosphorus 4K ncnn: CPU-v3-v3 - mobilenet-v3 renaissance: Apache Spark ALS mnn: SqueezeNetV1.0 mnn: mobilenetV3 brl-cad: VGR Performance Metric gravitymark: 1920 x 1080 - Vulkan renaissance: ALS Movie Lens renaissance: Apache Spark PageRank ncnn: CPU - shufflenet-v2 ncnn: CPU - efficientnet-b0 renaissance: Apache Spark Bayes ncnn: CPU - resnet50 ncnn: CPU - mnasnet rocksdb: Update Rand ncnn: CPU-v2-v2 - mobilenet-v2 ncnn: CPU - mobilenet vpxenc: Speed 0 - Bosphorus 4K rocksdb: Read Rand Write Rand astcenc: Thorough rocksdb: Rand Fill Sync astcenc: Exhaustive natron: Spaceship vpxenc: Speed 5 - Bosphorus 1080p 1 2 3 4 57.606 40.086 184.7 170.4 423.0 197.3 62.2 112.0 184.9 280.6 187.7 402.4 188.2 8.60 72732224 188.8 188.7 185.5 188.8 62.7 5.002 383.4 156.5 128633333 186.2 16.76 187.4 743.3 186.1 78.7 2629503 190.6 121.2 1943.1 188.6 36.486 880.2 237.0 27.67 376.9 15804.3 183.5 8319.2 181.5 186.8 4.660 1075179 183.1 256.792 62.45 4.6266 185.4 934075 185.3 2.13 108.2 226.565 33.524 183.2 3630.0 5.629 17.23 87.0 13.03 12785.8 189.0 12.10 3026.283 100.7 60.338 184.2 18.94 185.9 2101.5 10.05 4.66 1987.1 7.083 2.862 189712 101.0 8462.9 3610.9 5.11 7.04 2116.3 27.52 4.78 537472 5.37 16.51 5.73 1968077 6.6419 13100 61.6745 1.8 22.57 51.465 36.849 188.0 162.2 404.1 189.5 59.6 107.4 188.8 281.5 188.0 403.2 193.5 8.40 75316788 184.8 182.4 180.9 185.0 60.8 4.851 372.9 152.0 132400000 191.6 17.24 188.2 730.1 181.5 77.1 2636681 186.3 118.7 1961.2 185.7 36.604 898.6 237.5 27.13 377.4 16010.4 184.1 8304.7 183.4 183.4 4.578 1094396 184.3 252.648 62.60 4.5565 184.1 947946 183.6 2.13 108.3 223.516 33.086 182.9 3673.6 5.581 17.16 86.1 13.18 12713.3 189.5 12.03 3018.515 100.1 60.477 185.0 19.12 187.0 2094.2 10.06 4.62 2003.5 7.058 2.841 190007 101.5 8451.6 3612.3 5.09 7.02 2109.0 27.61 4.76 539663 5.35 16.57 5.75 1973866 6.6252 13080 61.5895 1.8 20.81 51.349 36.851 194.4 168.2 416.4 189.0 62.1 111.8 188.9 292.1 188.5 418.5 186.1 8.72 74599807 191.3 186.0 180.7 188.3 61.5 4.970 384.5 156.7 131366667 188.8 17.21 190.0 750.6 185.3 79.0 2693833 186.7 121.4 1917.9 184.6 35.830 880.1 241.9 27.56 384.4 15698.4 181.7 8468.0 184.9 186.3 4.607 1094225 186.2 256.897 61.64 4.5766 184.2 934477 186.2 2.16 107.8 223.780 33.257 183.7 3662.6 5.564 17.36 87.1 13.16 12856.4 191.1 12.16 2995.856 99.7 59.886 185.1 19.07 185.3 2082.8 10.14 4.65 2001.3 7.113 2.855 191092 101.3 8408.0 3589.4 5.12 7.06 2104.6 27.48 4.78 537820 5.35 16.51 5.73 1968072 6.6301 13107 61.6254 1.8 20.37 183.1 189.3 192.8 181.2 190.1 189.4 183.3 179.4 190.8 61.7 187.1 192.7 184.8 188.8 184.6 185.3 184.4 186.7 183.7 186.9 185.8 109.3 181.5 87.1 189.8 100.5 183.3 187 101.7 OpenBenchmarking.org
Timed GDB GNU Debugger Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GDB GNU Debugger Compilation 10.2 Time To Compile 3 2 1 13 26 39 52 65 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 51.35 51.47 57.61
Timed FFmpeg Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed FFmpeg Compilation 4.4 Time To Compile 3 2 1 9 18 27 36 45 SE +/- 0.08, N = 3 SE +/- 0.15, N = 3 SE +/- 0.04, N = 3 36.85 36.85 40.09
Unvanquished Resolution: 800 x 600 - Effects Quality: High OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 800 x 600 - Effects Quality: High 4 3 2 1 40 80 120 160 200 SE +/- 0.88, N = 3 SE +/- 2.24, N = 4 SE +/- 1.93, N = 3 183.1 194.4 188.0 184.7
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM 3 2 1 40 80 120 160 200 SE +/- 0.20, N = 3 SE +/- 0.03, N = 3 SE +/- 1.30, N = 8 168.2 162.2 170.4 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM 3 2 1 90 180 270 360 450 SE +/- 0.63, N = 3 SE +/- 1.15, N = 3 SE +/- 3.63, N = 8 416.4 404.1 423.0 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
Unvanquished Resolution: 1280 x 1024 - Effects Quality: Medium OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1280 x 1024 - Effects Quality: Medium 4 3 2 1 40 80 120 160 200 SE +/- 1.03, N = 3 SE +/- 1.26, N = 15 SE +/- 1.74, N = 3 189.3 189.0 189.5 197.3
srsRAN Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM 3 2 1 14 28 42 56 70 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.00, N = 3 62.1 59.6 62.2 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
srsRAN Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM 3 2 1 30 60 90 120 150 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 111.8 107.4 112.0 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
Unvanquished Resolution: 1024 x 768 - Effects Quality: High OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1024 x 768 - Effects Quality: High 4 3 2 1 40 80 120 160 200 SE +/- 1.08, N = 3 SE +/- 2.36, N = 3 SE +/- 1.18, N = 3 192.8 188.9 188.8 184.9
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM 3 2 1 60 120 180 240 300 SE +/- 0.64, N = 3 SE +/- 0.07, N = 3 SE +/- 0.40, N = 3 292.1 281.5 280.6 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
Unvanquished Resolution: 1920 x 1080 - Effects Quality: Medium OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1920 x 1080 - Effects Quality: Medium 4 3 2 1 40 80 120 160 200 SE +/- 1.85, N = 3 SE +/- 1.28, N = 3 SE +/- 1.27, N = 15 181.2 188.5 188.0 187.7
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM 3 2 1 90 180 270 360 450 SE +/- 1.36, N = 3 SE +/- 0.38, N = 3 SE +/- 0.15, N = 3 418.5 403.2 402.4 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
Unvanquished Resolution: 1920 x 1200 - Effects Quality: Medium OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1920 x 1200 - Effects Quality: Medium 4 3 2 1 40 80 120 160 200 SE +/- 1.91, N = 4 SE +/- 1.58, N = 3 SE +/- 2.01, N = 3 190.1 186.1 193.5 188.2
VP9 libvpx Encoding Speed: Speed 0 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.10.0 Speed: Speed 0 - Input: Bosphorus 1080p 3 2 1 2 4 6 8 10 SE +/- 0.08, N = 15 SE +/- 0.08, N = 3 SE +/- 0.09, N = 6 8.72 8.40 8.60 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
Facebook RocksDB Test: Random Read OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Random Read 3 2 1 16M 32M 48M 64M 80M SE +/- 298817.66, N = 3 SE +/- 1023472.59, N = 3 SE +/- 440149.97, N = 3 74599807 75316788 72732224 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Unvanquished Resolution: 1280 x 1024 - Effects Quality: High OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1280 x 1024 - Effects Quality: High 4 3 2 1 40 80 120 160 200 SE +/- 1.44, N = 3 SE +/- 1.59, N = 3 SE +/- 1.76, N = 6 189.4 191.3 184.8 188.8
Unvanquished Resolution: 1280 x 1024 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1280 x 1024 - Effects Quality: Ultra 4 3 2 1 40 80 120 160 200 SE +/- 1.50, N = 9 SE +/- 1.22, N = 3 SE +/- 2.01, N = 5 183.3 186.0 182.4 188.7
Unvanquished Resolution: 2560 x 1440 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 2560 x 1440 - Effects Quality: Ultra 4 3 2 1 40 80 120 160 200 SE +/- 0.62, N = 3 SE +/- 2.23, N = 3 SE +/- 1.05, N = 3 179.4 180.7 180.9 185.5
Unvanquished Resolution: 1600 x 1200 - Effects Quality: High OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1600 x 1200 - Effects Quality: High 4 3 2 1 40 80 120 160 200 SE +/- 1.69, N = 3 SE +/- 0.95, N = 3 SE +/- 1.82, N = 3 190.8 188.3 185.0 188.8
GravityMark Resolution: 3840 x 2160 - Renderer: Vulkan OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.2 Resolution: 3840 x 2160 - Renderer: Vulkan 4 3 2 1 14 28 42 56 70 SE +/- 0.13, N = 3 SE +/- 0.15, N = 3 SE +/- 0.13, N = 3 61.7 61.5 60.8 62.7
Mobile Neural Network Model: squeezenetv1.1 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: squeezenetv1.1 3 2 1 1.1255 2.251 3.3765 4.502 5.6275 SE +/- 0.017, N = 3 SE +/- 0.100, N = 3 SE +/- 0.001, N = 3 4.970 4.851 5.002 MIN: 4.78 / MAX: 14.12 MIN: 4.48 / MAX: 11.55 MIN: 4.81 / MAX: 14.82 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM 3 2 1 80 160 240 320 400 SE +/- 0.75, N = 3 SE +/- 1.00, N = 3 SE +/- 3.41, N = 7 384.5 372.9 383.4 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM 3 2 1 30 60 90 120 150 SE +/- 0.35, N = 3 SE +/- 0.12, N = 3 SE +/- 1.37, N = 7 156.7 152.0 156.5 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
srsRAN Test: OFDM_Test OpenBenchmarking.org Samples / Second, More Is Better srsRAN 21.04 Test: OFDM_Test 3 2 1 30M 60M 90M 120M 150M SE +/- 833333.33, N = 3 SE +/- 1541644.14, N = 4 SE +/- 851143.02, N = 3 131366667 132400000 128633333 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
Unvanquished Resolution: 1024 x 768 - Effects Quality: Medium OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1024 x 768 - Effects Quality: Medium 4 3 2 1 40 80 120 160 200 SE +/- 1.41, N = 15 SE +/- 0.49, N = 3 SE +/- 1.10, N = 3 187.1 188.8 191.6 186.2
NCNN Target: CPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: resnet18 3 2 1 4 8 12 16 20 SE +/- 0.25, N = 3 SE +/- 0.25, N = 3 SE +/- 0.08, N = 3 17.21 17.24 16.76 MIN: 16.13 / MAX: 26.21 MIN: 16.16 / MAX: 25.86 MIN: 16.13 / MAX: 26.23 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Unvanquished Resolution: 3840 x 2160 - Effects Quality: Medium OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 3840 x 2160 - Effects Quality: Medium 4 3 2 1 40 80 120 160 200 SE +/- 1.65, N = 3 SE +/- 1.82, N = 6 SE +/- 2.14, N = 3 192.7 190.0 188.2 187.4
Renaissance Test: Scala Dotty OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Scala Dotty 3 2 1 160 320 480 640 800 SE +/- 11.19, N = 15 SE +/- 8.25, N = 14 SE +/- 3.08, N = 3 750.6 730.1 743.3 MIN: 596.59 / MAX: 1359.39 MIN: 590.82 / MAX: 1443.97 MIN: 605.59 / MAX: 1440.44
Unvanquished Resolution: 1600 x 1200 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1600 x 1200 - Effects Quality: Ultra 4 3 2 1 40 80 120 160 200 SE +/- 1.03, N = 3 SE +/- 1.12, N = 3 SE +/- 0.32, N = 3 184.8 185.3 181.5 186.1
srsRAN Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM 3 2 1 20 40 60 80 100 SE +/- 0.09, N = 3 SE +/- 1.08, N = 3 SE +/- 0.06, N = 3 79.0 77.1 78.7 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
Facebook RocksDB Test: Read While Writing OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Read While Writing 3 2 1 600K 1200K 1800K 2400K 3000K SE +/- 1295.16, N = 3 SE +/- 2625.90, N = 3 SE +/- 19810.08, N = 3 2693833 2636681 2629503 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Unvanquished Resolution: 800 x 600 - Effects Quality: Medium OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 800 x 600 - Effects Quality: Medium 4 3 2 1 40 80 120 160 200 SE +/- 1.12, N = 3 SE +/- 0.95, N = 3 SE +/- 1.87, N = 15 188.8 186.7 186.3 190.6
srsRAN Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM 3 2 1 30 60 90 120 150 SE +/- 0.15, N = 3 SE +/- 1.62, N = 3 SE +/- 0.06, N = 3 121.4 118.7 121.2 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
Renaissance Test: Genetic Algorithm Using Jenetics + Futures OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Genetic Algorithm Using Jenetics + Futures 3 2 1 400 800 1200 1600 2000 SE +/- 17.01, N = 3 SE +/- 26.25, N = 3 SE +/- 19.12, N = 3 1917.9 1961.2 1943.1 MIN: 1854.98 / MAX: 1966.64 MIN: 1867.51 / MAX: 2023.77 MIN: 1822.38 / MAX: 2000.6
Unvanquished Resolution: 800 x 600 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 800 x 600 - Effects Quality: Ultra 4 3 2 1 40 80 120 160 200 SE +/- 0.72, N = 3 SE +/- 1.20, N = 3 SE +/- 0.54, N = 3 184.6 184.6 185.7 188.6
Mobile Neural Network Model: resnet-v2-50 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: resnet-v2-50 3 2 1 8 16 24 32 40 SE +/- 0.26, N = 3 SE +/- 0.29, N = 3 SE +/- 0.52, N = 3 35.83 36.60 36.49 MIN: 34.09 / MAX: 45.45 MIN: 34.86 / MAX: 47.44 MIN: 34.29 / MAX: 78.51 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Renaissance Test: Random Forest OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Random Forest 3 2 1 200 400 600 800 1000 SE +/- 9.92, N = 3 SE +/- 4.91, N = 3 SE +/- 9.62, N = 3 880.1 898.6 880.2 MIN: 720.89 / MAX: 1048.17 MIN: 785.72 / MAX: 1168.2 MIN: 780.5 / MAX: 1084.36
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM 3 2 1 50 100 150 200 250 SE +/- 1.47, N = 3 SE +/- 3.37, N = 3 SE +/- 2.64, N = 3 241.9 237.5 237.0 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: yolov4-tiny 3 2 1 7 14 21 28 35 SE +/- 0.44, N = 3 SE +/- 0.45, N = 3 SE +/- 0.51, N = 3 27.56 27.13 27.67 MIN: 25.9 / MAX: 37.93 MIN: 25.84 / MAX: 30.55 MIN: 25.66 / MAX: 55.86 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM 3 2 1 80 160 240 320 400 SE +/- 1.88, N = 3 SE +/- 5.38, N = 3 SE +/- 4.53, N = 3 384.4 377.4 376.9 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lbladeRF -lm -lfftw3f -lmbedcrypto
C-Blosc Compressor: blosclz OpenBenchmarking.org MB/s, More Is Better C-Blosc 2.0 Compressor: blosclz 3 2 1 3K 6K 9K 12K 15K SE +/- 144.74, N = 3 SE +/- 228.02, N = 3 SE +/- 67.33, N = 3 15698.4 16010.4 15804.3 1. (CC) gcc options: -std=gnu99 -O3 -pthread -lrt -lm
Unvanquished Resolution: 3840 x 2160 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 3840 x 2160 - Effects Quality: Ultra 4 3 2 1 40 80 120 160 200 SE +/- 0.90, N = 3 SE +/- 0.99, N = 3 SE +/- 1.38, N = 3 185.3 181.7 184.1 183.5
Renaissance Test: Savina Reactors.IO OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Savina Reactors.IO 3 2 1 2K 4K 6K 8K 10K SE +/- 90.43, N = 4 SE +/- 114.97, N = 3 SE +/- 21.91, N = 3 8468.0 8304.7 8319.2 MIN: 8258.92 / MAX: 12642.13 MIN: 8085.57 / MAX: 12128.74 MIN: 8276.65 / MAX: 11910.31
Unvanquished Resolution: 1920 x 1200 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1920 x 1200 - Effects Quality: Ultra 4 3 2 1 40 80 120 160 200 SE +/- 0.84, N = 3 SE +/- 0.61, N = 3 SE +/- 1.32, N = 3 184.4 184.9 183.4 181.5
Unvanquished Resolution: 1920 x 1200 - Effects Quality: High OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1920 x 1200 - Effects Quality: High 4 3 2 1 40 80 120 160 200 SE +/- 1.70, N = 3 SE +/- 1.21, N = 3 SE +/- 0.50, N = 3 186.7 186.3 183.4 186.8
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: MobileNetV2_224 3 2 1 1.0485 2.097 3.1455 4.194 5.2425 SE +/- 0.011, N = 3 SE +/- 0.029, N = 3 SE +/- 0.038, N = 3 4.607 4.578 4.660 MIN: 4.46 / MAX: 15.04 MIN: 4.39 / MAX: 14.16 MIN: 4.42 / MAX: 14.67 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Facebook RocksDB Test: Sequential Fill OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Sequential Fill 3 2 1 200K 400K 600K 800K 1000K SE +/- 4707.18, N = 3 SE +/- 7693.63, N = 3 SE +/- 10720.19, N = 3 1094225 1094396 1075179 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Unvanquished Resolution: 3840 x 2160 - Effects Quality: High OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 3840 x 2160 - Effects Quality: High 4 3 2 1 40 80 120 160 200 SE +/- 1.55, N = 8 SE +/- 1.59, N = 3 SE +/- 1.37, N = 3 183.7 186.2 184.3 183.1
TNN Target: CPU - Model: MobileNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: MobileNet v2 3 2 1 60 120 180 240 300 SE +/- 2.35, N = 3 SE +/- 1.84, N = 3 SE +/- 0.32, N = 3 256.90 252.65 256.79 MIN: 242.08 / MAX: 272.17 MIN: 242.31 / MAX: 288.61 MIN: 249.16 / MAX: 279.34 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
NCNN Target: CPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: vgg16 3 2 1 14 28 42 56 70 SE +/- 0.24, N = 3 SE +/- 0.15, N = 3 SE +/- 0.23, N = 3 61.64 62.60 62.45 MIN: 59.82 / MAX: 71 MIN: 60.87 / MAX: 72.1 MIN: 60.81 / MAX: 76.03 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
ASTC Encoder Preset: Medium OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 3.0 Preset: Medium 3 2 1 1.041 2.082 3.123 4.164 5.205 SE +/- 0.0251, N = 3 SE +/- 0.0483, N = 3 SE +/- 0.0229, N = 3 4.5766 4.5565 4.6266 1. (CXX) g++ options: -O3 -flto -pthread
Unvanquished Resolution: 2560 x 1440 - Effects Quality: High OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 2560 x 1440 - Effects Quality: High 4 3 2 1 40 80 120 160 200 SE +/- 1.39, N = 3 SE +/- 2.28, N = 3 SE +/- 1.97, N = 4 186.9 184.2 184.1 185.4
Facebook RocksDB Test: Random Fill OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Random Fill 3 2 1 200K 400K 600K 800K 1000K SE +/- 6546.09, N = 3 SE +/- 6724.59, N = 3 SE +/- 434.63, N = 3 934477 947946 934075 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Unvanquished Resolution: 1024 x 768 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1024 x 768 - Effects Quality: Ultra 4 3 2 1 40 80 120 160 200 SE +/- 2.06, N = 3 SE +/- 1.45, N = 3 SE +/- 1.16, N = 3 185.8 186.2 183.6 185.3
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: blazeface 3 2 1 0.486 0.972 1.458 1.944 2.43 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 2.16 2.13 2.13 MIN: 2.06 / MAX: 2.78 MIN: 2.05 / MAX: 2.72 MIN: 2.06 / MAX: 2.73 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
GravityMark Resolution: 1600 x 1200 - Renderer: Vulkan OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.2 Resolution: 1600 x 1200 - Renderer: Vulkan 4 3 2 1 20 40 60 80 100 SE +/- 0.35, N = 3 SE +/- 0.33, N = 3 SE +/- 0.25, N = 3 109.3 107.8 108.3 108.2
TNN Target: CPU - Model: SqueezeNet v1.1 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: SqueezeNet v1.1 3 2 1 50 100 150 200 250 SE +/- 0.23, N = 3 SE +/- 0.36, N = 3 SE +/- 2.35, N = 3 223.78 223.52 226.57 MIN: 223.02 / MAX: 224.54 MIN: 222.56 / MAX: 235.97 MIN: 223.49 / MAX: 246 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: inception-v3 3 2 1 8 16 24 32 40 SE +/- 0.16, N = 3 SE +/- 0.31, N = 3 SE +/- 0.16, N = 3 33.26 33.09 33.52 MIN: 31.73 / MAX: 43.47 MIN: 31.39 / MAX: 43.21 MIN: 32.04 / MAX: 61.01 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Unvanquished Resolution: 1920 x 1080 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1920 x 1080 - Effects Quality: Ultra 4 3 2 1 40 80 120 160 200 SE +/- 2.27, N = 4 SE +/- 1.40, N = 3 SE +/- 1.31, N = 12 181.5 183.7 182.9 183.2
Renaissance Test: In-Memory Database Shootout OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: In-Memory Database Shootout 3 2 1 800 1600 2400 3200 4000 SE +/- 26.71, N = 3 SE +/- 30.31, N = 15 SE +/- 17.50, N = 3 3662.6 3673.6 3630.0 MIN: 3358.36 / MAX: 4325.05 MIN: 3257.97 / MAX: 4270.45 MIN: 3237.23 / MAX: 4102.1
Mobile Neural Network Model: mobilenet-v1-1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: mobilenet-v1-1.0 3 2 1 1.2665 2.533 3.7995 5.066 6.3325 SE +/- 0.009, N = 3 SE +/- 0.012, N = 3 SE +/- 0.037, N = 3 5.564 5.581 5.629 MIN: 5.35 / MAX: 14.28 MIN: 5.37 / MAX: 15.44 MIN: 5.37 / MAX: 15.07 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
NCNN Target: CPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: googlenet 3 2 1 4 8 12 16 20 SE +/- 0.31, N = 3 SE +/- 0.27, N = 3 SE +/- 0.31, N = 2 17.36 17.16 17.23 MIN: 15.95 / MAX: 26.3 MIN: 15.78 / MAX: 43.01 MIN: 15.79 / MAX: 18.6 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
GravityMark Resolution: 2560 x 1440 - Renderer: Vulkan OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.2 Resolution: 2560 x 1440 - Renderer: Vulkan 4 3 2 1 20 40 60 80 100 SE +/- 0.43, N = 3 SE +/- 0.34, N = 3 SE +/- 0.23, N = 3 87.1 87.1 86.1 87.0
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: alexnet 3 2 1 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.16, N = 3 13.16 13.18 13.03 MIN: 12.44 / MAX: 59.78 MIN: 12.67 / MAX: 14.06 MIN: 12.38 / MAX: 20.49 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Renaissance Test: Akka Unbalanced Cobwebbed Tree OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Akka Unbalanced Cobwebbed Tree 3 2 1 3K 6K 9K 12K 15K SE +/- 64.09, N = 3 SE +/- 23.84, N = 3 SE +/- 61.04, N = 3 12856.4 12713.3 12785.8 MIN: 10060.42 / MAX: 12964.84 MIN: 9948.44 / MAX: 12758 MIN: 9998.74 / MAX: 12907.84
Unvanquished Resolution: 1600 x 1200 - Effects Quality: Medium OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1600 x 1200 - Effects Quality: Medium 4 3 2 1 40 80 120 160 200 SE +/- 1.03, N = 3 SE +/- 0.84, N = 3 SE +/- 0.27, N = 3 189.8 191.1 189.5 189.0
NCNN Target: CPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: regnety_400m 3 2 1 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 12.16 12.03 12.10 MIN: 11.67 / MAX: 31.98 MIN: 11.62 / MAX: 12.93 MIN: 11.67 / MAX: 21.19 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
TNN Target: CPU - Model: DenseNet OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: DenseNet 3 2 1 600 1200 1800 2400 3000 SE +/- 0.70, N = 3 SE +/- 4.93, N = 3 SE +/- 6.29, N = 3 2995.86 3018.52 3026.28 MIN: 2868.29 / MAX: 3109.37 MIN: 2913.92 / MAX: 3097.69 MIN: 2898.49 / MAX: 3152.3 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
GravityMark Resolution: 1920 x 1200 - Renderer: Vulkan OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.2 Resolution: 1920 x 1200 - Renderer: Vulkan 4 3 2 1 20 40 60 80 100 SE +/- 0.37, N = 3 SE +/- 0.58, N = 3 SE +/- 0.30, N = 3 100.5 99.7 100.1 100.7
TNN Target: CPU - Model: SqueezeNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: SqueezeNet v2 3 2 1 14 28 42 56 70 SE +/- 0.18, N = 3 SE +/- 0.73, N = 3 SE +/- 0.19, N = 3 59.89 60.48 60.34 MIN: 59.2 / MAX: 61.76 MIN: 57.39 / MAX: 69 MIN: 59.86 / MAX: 62.96 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
Unvanquished Resolution: 1920 x 1080 - Effects Quality: High OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 1920 x 1080 - Effects Quality: High 4 3 2 1 40 80 120 160 200 SE +/- 1.96, N = 4 SE +/- 1.19, N = 15 SE +/- 0.32, N = 3 183.3 185.1 185.0 184.2
NCNN Target: CPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: squeezenet_ssd 3 2 1 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.20, N = 3 19.07 19.12 18.94 MIN: 18.14 / MAX: 27.2 MIN: 18.16 / MAX: 28.23 MIN: 17.79 / MAX: 26.15 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Unvanquished Resolution: 2560 x 1440 - Effects Quality: Medium OpenBenchmarking.org Frames Per Second, More Is Better Unvanquished 0.52.1 Resolution: 2560 x 1440 - Effects Quality: Medium 4 3 2 1 40 80 120 160 200 SE +/- 1.55, N = 3 SE +/- 0.92, N = 3 SE +/- 0.39, N = 3 187.0 185.3 187.0 185.9
Renaissance Test: Finagle HTTP Requests OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Finagle HTTP Requests 3 2 1 500 1000 1500 2000 2500 SE +/- 13.98, N = 3 SE +/- 7.73, N = 3 SE +/- 20.86, N = 3 2082.8 2094.2 2101.5 MIN: 1907.01 / MAX: 2116.62 MIN: 1933.27 / MAX: 2132.22 MIN: 1932.3 / MAX: 2144.63
VP9 libvpx Encoding Speed: Speed 5 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.10.0 Speed: Speed 5 - Input: Bosphorus 4K 3 2 1 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.11, N = 5 SE +/- 0.12, N = 3 10.14 10.06 10.05 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
NCNN Target: CPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU-v3-v3 - Model: mobilenet-v3 3 2 1 1.0485 2.097 3.1455 4.194 5.2425 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 4.65 4.62 4.66 MIN: 4.46 / MAX: 13.85 MIN: 4.46 / MAX: 5.47 MIN: 4.47 / MAX: 5.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Renaissance Test: Apache Spark ALS OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Apache Spark ALS 3 2 1 400 800 1200 1600 2000 SE +/- 22.37, N = 4 SE +/- 10.04, N = 3 SE +/- 23.45, N = 3 2001.3 2003.5 1987.1 MIN: 1813.56 / MAX: 2210.24 MIN: 1817.74 / MAX: 2340.39 MIN: 1751.76 / MAX: 2358.98
Mobile Neural Network Model: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: SqueezeNetV1.0 3 2 1 2 4 6 8 10 SE +/- 0.020, N = 3 SE +/- 0.118, N = 3 SE +/- 0.016, N = 3 7.113 7.058 7.083 MIN: 6.81 / MAX: 16.04 MIN: 6.61 / MAX: 16.37 MIN: 6.85 / MAX: 16.79 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Mobile Neural Network Model: mobilenetV3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: mobilenetV3 3 2 1 0.644 1.288 1.932 2.576 3.22 SE +/- 0.019, N = 3 SE +/- 0.031, N = 3 SE +/- 0.041, N = 3 2.855 2.841 2.862 MIN: 2.75 / MAX: 10.08 MIN: 2.71 / MAX: 3.51 MIN: 2.7 / MAX: 13.18 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
BRL-CAD VGR Performance Metric OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.32.2 VGR Performance Metric 3 2 1 40K 80K 120K 160K 200K 191092 190007 189712 1. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -pthread -ldl -lm
GravityMark Resolution: 1920 x 1080 - Renderer: Vulkan OpenBenchmarking.org Frames Per Second, More Is Better GravityMark 1.2 Resolution: 1920 x 1080 - Renderer: Vulkan 4 3 2 1 20 40 60 80 100 SE +/- 0.21, N = 3 SE +/- 0.23, N = 3 SE +/- 0.06, N = 3 101.7 101.3 101.5 101.0
Renaissance Test: ALS Movie Lens OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: ALS Movie Lens 3 2 1 2K 4K 6K 8K 10K SE +/- 16.75, N = 3 SE +/- 9.70, N = 3 SE +/- 27.74, N = 3 8408.0 8451.6 8462.9 MIN: 8374.97 / MAX: 9304.53 MIN: 8322.99 / MAX: 9307.11 MIN: 8411.33 / MAX: 9538.23
Renaissance Test: Apache Spark PageRank OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Apache Spark PageRank 3 2 1 800 1600 2400 3200 4000 SE +/- 26.61, N = 3 SE +/- 36.74, N = 6 SE +/- 23.77, N = 3 3589.4 3612.3 3610.9 MIN: 3158.92 / MAX: 3728.47 MIN: 3174.88 / MAX: 3901.76 MIN: 3152.25 / MAX: 3810.45
NCNN Target: CPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: shufflenet-v2 3 2 1 1.152 2.304 3.456 4.608 5.76 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 5.12 5.09 5.11 MIN: 4.94 / MAX: 13.04 MIN: 4.9 / MAX: 13.69 MIN: 4.94 / MAX: 6.07 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: efficientnet-b0 3 2 1 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 7.06 7.02 7.04 MIN: 6.77 / MAX: 8.01 MIN: 6.7 / MAX: 15.34 MIN: 6.62 / MAX: 15.6 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Renaissance Test: Apache Spark Bayes OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Apache Spark Bayes 3 2 1 500 1000 1500 2000 2500 SE +/- 5.65, N = 3 SE +/- 6.39, N = 3 SE +/- 20.25, N = 6 2104.6 2109.0 2116.3 MIN: 1615.06 / MAX: 2114.96 MIN: 1596.2 / MAX: 2115.92 MIN: 1586.15 / MAX: 2474.25
NCNN Target: CPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: resnet50 3 2 1 6 12 18 24 30 SE +/- 0.37, N = 3 SE +/- 0.35, N = 3 SE +/- 0.39, N = 3 27.48 27.61 27.52 MIN: 25.87 / MAX: 36.81 MIN: 25.94 / MAX: 37.88 MIN: 25.91 / MAX: 36.45 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: mnasnet 3 2 1 1.0755 2.151 3.2265 4.302 5.3775 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 4.78 4.76 4.78 MIN: 4.59 / MAX: 5.58 MIN: 4.58 / MAX: 13.06 MIN: 4.59 / MAX: 5.51 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Facebook RocksDB Test: Update Random OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Update Random 3 2 1 120K 240K 360K 480K 600K SE +/- 825.41, N = 3 SE +/- 1157.03, N = 3 SE +/- 824.52, N = 3 537820 539663 537472 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
NCNN Target: CPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU-v2-v2 - Model: mobilenet-v2 3 2 1 1.2083 2.4166 3.6249 4.8332 6.0415 SE +/- 0.06, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 5.35 5.35 5.37 MIN: 5.04 / MAX: 24.85 MIN: 5.1 / MAX: 6.31 MIN: 5.11 / MAX: 14.59 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: CPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: CPU - Model: mobilenet 3 2 1 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.19, N = 3 16.51 16.57 16.51 MIN: 15.66 / MAX: 25.29 MIN: 15.64 / MAX: 27.2 MIN: 15.28 / MAX: 24.91 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
VP9 libvpx Encoding Speed: Speed 0 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.10.0 Speed: Speed 0 - Input: Bosphorus 4K 3 2 1 1.2938 2.5876 3.8814 5.1752 6.469 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 5.73 5.75 5.73 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
Facebook RocksDB Test: Read Random Write Random OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Read Random Write Random 3 2 1 400K 800K 1200K 1600K 2000K SE +/- 9674.59, N = 3 SE +/- 15430.39, N = 3 SE +/- 1027.04, N = 3 1968072 1973866 1968077 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
ASTC Encoder Preset: Thorough OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 3.0 Preset: Thorough 3 2 1 2 4 6 8 10 SE +/- 0.0026, N = 3 SE +/- 0.0048, N = 3 SE +/- 0.0023, N = 3 6.6301 6.6252 6.6419 1. (CXX) g++ options: -O3 -flto -pthread
Facebook RocksDB Test: Random Fill Sync OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Random Fill Sync 3 2 1 3K 6K 9K 12K 15K SE +/- 6.00, N = 3 SE +/- 21.42, N = 3 SE +/- 3.51, N = 3 13107 13080 13100 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
ASTC Encoder Preset: Exhaustive OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 3.0 Preset: Exhaustive 3 2 1 14 28 42 56 70 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 61.63 61.59 61.67 1. (CXX) g++ options: -O3 -flto -pthread
Natron Input: Spaceship OpenBenchmarking.org FPS, More Is Better Natron 2.4 Input: Spaceship 3 2 1 0.405 0.81 1.215 1.62 2.025 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 6 1.8 1.8 1.8
VP9 libvpx Encoding Speed: Speed 5 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.10.0 Speed: Speed 5 - Input: Bosphorus 1080p 3 2 1 5 10 15 20 25 SE +/- 0.67, N = 15 SE +/- 0.89, N = 12 SE +/- 0.49, N = 15 20.37 20.81 22.57 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
Phoronix Test Suite v10.8.5