3950X April AMD Ryzen 9 3950X 16-Core testing with a ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS) and Sapphire AMD Radeon RX 470/480/570/570X/580/580X/590 4GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2204120-NE-3950XAPRI19&sor&grs .
3950X April Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Vulkan Compiler File-System Screen Resolution A B 3 AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads) ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS) AMD Starship/Matisse 16GB Samsung SSD 970 EVO 250GB Sapphire AMD Radeon RX 470/480/570/570X/580/580X/590 4GB (1260/1750MHz) AMD Ellesmere HDMI Audio DELL S2409W Intel I211 + Realtek RTL8822BE 802.11a/b/g/n/ac Ubuntu 20.04 5.11.0-43-generic (x86_64) GNOME Shell 3.36.4 X Server 1.20.13 1.2.128 GCC 9.3.0 ext4 1920x1080 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8701021 Java Details - OpenJDK Runtime Environment (build 11.0.13+8-Ubuntu-0ubuntu1.20.04) 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
3950X April perf-bench: Epoll Wait ospray: gravity_spheres_volume/dim_512/ao/real_time ospray: gravity_spheres_volume/dim_512/scivis/real_time onednn: IP Shapes 3D - f32 - CPU perf-bench: Syscall Basic aom-av1: Speed 0 Two-Pass - Bosphorus 4K perf-bench: Memset 1MB perf-bench: Futex Lock-Pi aom-av1: Speed 9 Realtime - Bosphorus 4K onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU aom-av1: Speed 8 Realtime - Bosphorus 4K onednn: Deconvolution Batch shapes_1d - f32 - CPU perf-bench: Memcpy 1MB onednn: Recurrent Neural Network Inference - u8s8f32 - CPU aom-av1: Speed 0 Two-Pass - Bosphorus 1080p aom-av1: Speed 8 Realtime - Bosphorus 1080p aom-av1: Speed 6 Realtime - Bosphorus 4K avifenc: 6, Lossless aom-av1: Speed 9 Realtime - Bosphorus 1080p aom-av1: Speed 6 Realtime - Bosphorus 1080p aom-av1: Speed 10 Realtime - Bosphorus 1080p avifenc: 6 compress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compression avifenc: 10, Lossless onednn: Recurrent Neural Network Training - f32 - CPU onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU rocksdb: Read While Writing onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU ospray: particle_volume/ao/real_time ospray: particle_volume/scivis/real_time rocksdb: Rand Read avifenc: 2 onednn: Deconvolution Batch shapes_3d - f32 - CPU rocksdb: Read Rand Write Rand onednn: Recurrent Neural Network Training - u8s8f32 - CPU aom-av1: Speed 6 Two-Pass - Bosphorus 1080p build-mplayer: Time To Compile perf-bench: Futex Hash ospray-studio: 2 - 1080p - 32 - Path Tracer aom-av1: Speed 4 Two-Pass - Bosphorus 4K onednn: Convolution Batch Shapes Auto - f32 - CPU libgav1: Chimera 1080p ospray-studio: 2 - 1080p - 16 - Path Tracer ospray: gravity_spheres_volume/dim_512/pathtracer/real_time ospray-studio: 3 - 1080p - 1 - Path Tracer onednn: IP Shapes 3D - u8s8f32 - CPU ospray: particle_volume/pathtracer/real_time ospray-studio: 3 - 1080p - 16 - Path Tracer onednn: IP Shapes 1D - f32 - CPU avifenc: 0 libgav1: Chimera 1080p 10-bit ospray-studio: 1 - 1080p - 16 - Path Tracer libgav1: Summer Nature 1080p rocksdb: Update Rand ospray-studio: 1 - 1080p - 1 - Path Tracer aom-av1: Speed 4 Two-Pass - Bosphorus 1080p aom-av1: Speed 6 Two-Pass - Bosphorus 4K onednn: IP Shapes 1D - u8s8f32 - CPU perf-bench: Sched Pipe ospray-studio: 2 - 1080p - 1 - Path Tracer libgav1: Summer Nature 4K ospray-studio: 1 - 1080p - 32 - Path Tracer ospray-studio: 3 - 1080p - 32 - Path Tracer java-jmh: Throughput onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU aom-av1: Speed 10 Realtime - Bosphorus 4K A B 3 29976 2.55935 2.46619 11.0447 20487873 0.19 73.520223 482 49.39 2653.01 2677.16 0.679415 41.47 4.17083 14.470071 2600.06 0.45 95.77 14.47 9.998 115.9 8.52 124.99 6.459 4.342 5.622 5129.54 23.2413 5135.11 3222053 2.5431 0.789443 23.4656 22.4134 81933471 56.074 4.37595 2368822 5117.44 28.26 21.951 4799340 66843 6.03 20.6164 146.78 30384 4.06373 2216 0.790537 241.187 35312 4.66061 112.112 51.81 29566 198.41 654775 1854 10.28 10.92 1.29867 403883 1904 63.77 65388 76950 32629111123.553 1.84739 49.58 32098 2.61399 2.50856 10.6479 21592179 0.18 72.613683 461 51.51 2596.92 2638.95 0.680813 40.21 4.29655 14.439690 2610.48 0.45 94.49 14.35 9.814 118.04 8.37 123.89 6.555 4.400 5.666 5136.73 23.0397 5149.07 3254000 2.55214 0.792154 23.5908 22.5209 81329052 56.381 4.40032 2383189 5119.53 28.26 22.065 4819775 67035 6.00 20.5759 147.37 30283 4.06095 2219 0.788060 241.551 35299 4.66399 112.407 51.69 29498 198.03 656248 1852 10.29 10.94 1.30043 403760 1906 63.67 65291 76891 32638201689.918 1.84743 51.47 33429 2.35183 2.26083 10.2818 20407433 0.18 69.681475 461 50.85 2556.63 2581.96 0.701676 40.53 4.21857 14.871727 2540.71 0.44 96.49 14.64 9.899 117.74 8.45 122.97 6.565 4.348 5.596 5075.49 22.9890 5093.46 3256160 2.56316 0.795531 23.6432 22.5830 81500459 56.019 4.37283 2380613 5091.02 28.11 21.965 4823523 66701 6.00 20.5250 146.99 30296 4.05074 2223 0.789330 241.931 35404 4.65093 112.243 51.75 29541 198.48 654806 1850 10.27 10.94 1.30085 403227 1907 63.71 65296 76892 32625899282.111 1.84706 52.26 OpenBenchmarking.org
perf-bench Benchmark: Epoll Wait OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Epoll Wait 3 B A 7K 14K 21K 28K 35K SE +/- 354.56, N = 3 SE +/- 344.15, N = 3 33429 32098 29976 1. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma
OSPray Benchmark: gravity_spheres_volume/dim_512/ao/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPray 2.9 Benchmark: gravity_spheres_volume/dim_512/ao/real_time B A 3 0.5881 1.1762 1.7643 2.3524 2.9405 SE +/- 0.00062, N = 3 SE +/- 0.00623, N = 3 2.61399 2.55935 2.35183
OSPray Benchmark: gravity_spheres_volume/dim_512/scivis/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPray 2.9 Benchmark: gravity_spheres_volume/dim_512/scivis/real_time B A 3 0.5644 1.1288 1.6932 2.2576 2.822 SE +/- 0.00142, N = 3 SE +/- 0.00092, N = 3 2.50856 2.46619 2.26083
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU 3 B A 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 10.28 10.65 11.04 MIN: 9.94 MIN: 10.46 MIN: 10.91 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
perf-bench Benchmark: Syscall Basic OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Syscall Basic B A 3 5M 10M 15M 20M 25M SE +/- 156358.13, N = 3 SE +/- 147824.81, N = 3 21592179 20487873 20407433 1. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K A 3 B 0.0428 0.0856 0.1284 0.1712 0.214 SE +/- 0.00, N = 4 SE +/- 0.00, N = 3 0.19 0.18 0.18 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
perf-bench Benchmark: Memset 1MB OpenBenchmarking.org GB/sec, More Is Better perf-bench Benchmark: Memset 1MB A B 3 16 32 48 64 80 SE +/- 0.10, N = 3 SE +/- 0.77, N = 7 73.52 72.61 69.68 1. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma
perf-bench Benchmark: Futex Lock-Pi OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Futex Lock-Pi A 3 B 100 200 300 400 500 SE +/- 5.17, N = 3 SE +/- 5.21, N = 6 482 461 461 1. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K B 3 A 12 24 36 48 60 SE +/- 0.52, N = 15 SE +/- 0.75, N = 3 51.51 50.85 49.39 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU 3 B A 600 1200 1800 2400 3000 SE +/- 15.15, N = 3 SE +/- 20.31, N = 3 2556.63 2596.92 2653.01 MIN: 2510.01 MIN: 2539.82 MIN: 2610.08 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU 3 B A 600 1200 1800 2400 3000 SE +/- 28.67, N = 3 SE +/- 26.03, N = 3 2581.96 2638.95 2677.16 MIN: 2502.81 MIN: 2574.31 MIN: 2661.56 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU A B 3 0.1579 0.3158 0.4737 0.6316 0.7895 SE +/- 0.001462, N = 3 SE +/- 0.005860, N = 3 0.679415 0.680813 0.701676 MIN: 0.62 MIN: 0.61 MIN: 0.63 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K A 3 B 9 18 27 36 45 SE +/- 0.66, N = 3 SE +/- 0.41, N = 3 41.47 40.53 40.21 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU A 3 B 0.9667 1.9334 2.9001 3.8668 4.8335 SE +/- 0.02789, N = 3 SE +/- 0.02203, N = 3 4.17083 4.21857 4.29655 MIN: 3.93 MIN: 3.9 MIN: 3.93 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
perf-bench Benchmark: Memcpy 1MB OpenBenchmarking.org GB/sec, More Is Better perf-bench Benchmark: Memcpy 1MB 3 A B 4 8 12 16 20 SE +/- 0.09, N = 3 SE +/- 0.20, N = 3 14.87 14.47 14.44 1. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU 3 A B 600 1200 1800 2400 3000 SE +/- 7.94, N = 3 SE +/- 41.03, N = 3 2540.71 2600.06 2610.48 MIN: 2495.84 MIN: 2582.7 MIN: 2529.82 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p B A 3 0.1013 0.2026 0.3039 0.4052 0.5065 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.45 0.45 0.44 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p 3 A B 20 40 60 80 100 SE +/- 0.30, N = 3 SE +/- 0.34, N = 3 96.49 95.77 94.49 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K 3 A B 4 8 12 16 20 SE +/- 0.18, N = 3 SE +/- 0.11, N = 15 14.64 14.47 14.35 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 6, Lossless B 3 A 3 6 9 12 15 SE +/- 0.019, N = 3 SE +/- 0.107, N = 3 9.814 9.899 9.998 1. (CXX) g++ options: -O3 -fPIC -lm
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p B 3 A 30 60 90 120 150 SE +/- 0.58, N = 3 SE +/- 0.43, N = 3 118.04 117.74 115.90 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p A 3 B 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 8.52 8.45 8.37 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p A B 3 30 60 90 120 150 SE +/- 0.98, N = 14 SE +/- 1.64, N = 4 124.99 123.89 122.97 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 6 A B 3 2 4 6 8 10 SE +/- 0.010, N = 3 SE +/- 0.013, N = 3 6.459 6.555 6.565 1. (CXX) g++ options: -O3 -fPIC -lm
Parallel BZIP2 Compression FreeBSD-13.0-RELEASE-amd64-memstick.img Compression OpenBenchmarking.org Seconds, Fewer Is Better Parallel BZIP2 Compression 1.1.13 FreeBSD-13.0-RELEASE-amd64-memstick.img Compression A 3 B 0.99 1.98 2.97 3.96 4.95 SE +/- 0.060, N = 4 SE +/- 0.013, N = 3 4.342 4.348 4.400 1. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 10, Lossless 3 A B 1.2749 2.5498 3.8247 5.0996 6.3745 SE +/- 0.019, N = 3 SE +/- 0.046, N = 3 5.596 5.622 5.666 1. (CXX) g++ options: -O3 -fPIC -lm
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU 3 A B 1100 2200 3300 4400 5500 SE +/- 15.57, N = 3 SE +/- 21.94, N = 3 5075.49 5129.54 5136.73 MIN: 5015.99 MIN: 5095.07 MIN: 5068.06 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU 3 B A 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 22.99 23.04 23.24 MIN: 22.71 MIN: 22.84 MIN: 23.08 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU 3 A B 1100 2200 3300 4400 5500 SE +/- 56.24, N = 3 SE +/- 12.81, N = 3 5093.46 5135.11 5149.07 MIN: 4939.21 MIN: 5111.68 MIN: 5102.61 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
Facebook RocksDB Test: Read While Writing OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 7.0.1 Test: Read While Writing 3 B A 700K 1400K 2100K 2800K 3500K SE +/- 10292.52, N = 3 SE +/- 9707.64, N = 3 3256160 3254000 3222053 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU A B 3 0.5767 1.1534 1.7301 2.3068 2.8835 SE +/- 0.00703, N = 3 SE +/- 0.01307, N = 3 2.54310 2.55214 2.56316 MIN: 2.48 MIN: 2.44 MIN: 2.46 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU A B 3 0.179 0.358 0.537 0.716 0.895 SE +/- 0.004119, N = 3 SE +/- 0.000562, N = 3 0.789443 0.792154 0.795531 MIN: 0.74 MIN: 0.74 MIN: 0.75 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
OSPray Benchmark: particle_volume/ao/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPray 2.9 Benchmark: particle_volume/ao/real_time 3 B A 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 23.64 23.59 23.47
OSPray Benchmark: particle_volume/scivis/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPray 2.9 Benchmark: particle_volume/scivis/real_time 3 B A 5 10 15 20 25 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 22.58 22.52 22.41
Facebook RocksDB Test: Random Read OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 7.0.1 Test: Random Read A 3 B 20M 40M 60M 80M 100M SE +/- 310806.70, N = 3 SE +/- 886019.83, N = 3 81933471 81500459 81329052 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 2 3 A B 13 26 39 52 65 SE +/- 0.08, N = 3 SE +/- 0.26, N = 3 56.02 56.07 56.38 1. (CXX) g++ options: -O3 -fPIC -lm
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU 3 A B 0.9901 1.9802 2.9703 3.9604 4.9505 SE +/- 0.01112, N = 3 SE +/- 0.02200, N = 3 4.37283 4.37595 4.40032 MIN: 4.26 MIN: 4.28 MIN: 4.28 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
Facebook RocksDB Test: Read Random Write Random OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 7.0.1 Test: Read Random Write Random B 3 A 500K 1000K 1500K 2000K 2500K SE +/- 18963.04, N = 3 SE +/- 5611.13, N = 3 2383189 2380613 2368822 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU 3 A B 1100 2200 3300 4400 5500 SE +/- 19.61, N = 3 SE +/- 34.54, N = 3 5091.02 5117.44 5119.53 MIN: 5027.14 MIN: 5086.78 MIN: 5034.39 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p B A 3 7 14 21 28 35 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 28.26 28.26 28.11 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Timed MPlayer Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed MPlayer Compilation 1.5 Time To Compile A 3 B 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 21.95 21.97 22.07
perf-bench Benchmark: Futex Hash OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Futex Hash 3 B A 1000K 2000K 3000K 4000K 5000K SE +/- 6656.67, N = 3 SE +/- 3374.73, N = 3 4823523 4819775 4799340 1. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma
OSPray Studio Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer 3 A B 14K 28K 42K 56K 70K SE +/- 128.74, N = 3 SE +/- 63.26, N = 3 66701 66843 67035 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K A 3 B 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 6.03 6.00 6.00 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU 3 B A 5 10 15 20 25 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 20.53 20.58 20.62 MIN: 20.37 MIN: 20.41 MIN: 20.48 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
libgav1 Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better libgav1 0.17 Video Input: Chimera 1080p B 3 A 30 60 90 120 150 SE +/- 0.37, N = 3 SE +/- 0.08, N = 3 147.37 146.99 146.78 1. (CXX) g++ options: -O3 -lpthread -lrt
OSPray Studio Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer B 3 A 7K 14K 21K 28K 35K SE +/- 17.17, N = 3 SE +/- 11.15, N = 3 30283 30296 30384 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
OSPray Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPray 2.9 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time A B 3 0.9143 1.8286 2.7429 3.6572 4.5715 SE +/- 0.00423, N = 3 SE +/- 0.00418, N = 3 4.06373 4.06095 4.05074
OSPray Studio Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer A B 3 500 1000 1500 2000 2500 SE +/- 1.76, N = 3 2216 2219 2223 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU B 3 A 0.1779 0.3558 0.5337 0.7116 0.8895 SE +/- 0.004439, N = 3 SE +/- 0.006963, N = 3 0.788060 0.789330 0.790537 MIN: 0.73 MIN: 0.73 MIN: 0.75 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
OSPray Benchmark: particle_volume/pathtracer/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPray 2.9 Benchmark: particle_volume/pathtracer/real_time 3 B A 50 100 150 200 250 SE +/- 0.34, N = 3 SE +/- 0.20, N = 3 241.93 241.55 241.19
OSPray Studio Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer B A 3 8K 16K 24K 32K 40K SE +/- 21.28, N = 3 SE +/- 45.40, N = 3 35299 35312 35404 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU 3 A B 1.0494 2.0988 3.1482 4.1976 5.247 SE +/- 0.00958, N = 3 SE +/- 0.00368, N = 3 4.65093 4.66061 4.66399 MIN: 4.43 MIN: 4.46 MIN: 4.46 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 0 A 3 B 30 60 90 120 150 SE +/- 0.34, N = 3 SE +/- 0.41, N = 3 112.11 112.24 112.41 1. (CXX) g++ options: -O3 -fPIC -lm
libgav1 Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better libgav1 0.17 Video Input: Chimera 1080p 10-bit A 3 B 12 24 36 48 60 SE +/- 0.01, N = 3 SE +/- 0.10, N = 3 51.81 51.75 51.69 1. (CXX) g++ options: -O3 -lpthread -lrt
OSPray Studio Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer B 3 A 6K 12K 18K 24K 30K SE +/- 13.72, N = 3 29498 29541 29566 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
libgav1 Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better libgav1 0.17 Video Input: Summer Nature 1080p 3 A B 40 80 120 160 200 SE +/- 0.16, N = 3 SE +/- 0.14, N = 3 198.48 198.41 198.03 1. (CXX) g++ options: -O3 -lpthread -lrt
Facebook RocksDB Test: Update Random OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 7.0.1 Test: Update Random B 3 A 140K 280K 420K 560K 700K SE +/- 1055.36, N = 3 SE +/- 1748.11, N = 3 656248 654806 654775 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OSPray Studio Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer 3 B A 400 800 1200 1600 2000 SE +/- 0.67, N = 3 SE +/- 0.67, N = 3 1850 1852 1854 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p B A 3 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 10.29 10.28 10.27 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K 3 B A 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 10.94 10.94 10.92 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
oneDNN Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU A B 3 0.2927 0.5854 0.8781 1.1708 1.4635 SE +/- 0.00156, N = 3 SE +/- 0.00084, N = 3 1.29867 1.30043 1.30085 MIN: 1.28 MIN: 1.28 MIN: 1.28 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
perf-bench Benchmark: Sched Pipe OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Sched Pipe A B 3 90K 180K 270K 360K 450K SE +/- 3062.78, N = 3 SE +/- 2490.82, N = 3 403883 403760 403227 1. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma
OSPray Studio Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer A B 3 400 800 1200 1600 2000 SE +/- 1.86, N = 3 SE +/- 2.65, N = 3 1904 1906 1907 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
libgav1 Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better libgav1 0.17 Video Input: Summer Nature 4K A 3 B 14 28 42 56 70 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 63.77 63.71 63.67 1. (CXX) g++ options: -O3 -lpthread -lrt
OSPray Studio Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer B 3 A 14K 28K 42K 56K 70K SE +/- 76.21, N = 3 SE +/- 54.01, N = 3 65291 65296 65388 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
OSPray Studio Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer OpenBenchmarking.org ms, Fewer Is Better OSPray Studio 0.10 Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer B 3 A 16K 32K 48K 64K 80K SE +/- 55.43, N = 3 SE +/- 50.93, N = 3 76891 76892 76950 1. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread
Java JMH Throughput OpenBenchmarking.org Ops/s, More Is Better Java JMH Throughput B A 3 7000M 14000M 21000M 28000M 35000M 32638201689.92 32629111123.55 32625899282.11
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU 3 A B 0.4157 0.8314 1.2471 1.6628 2.0785 SE +/- 0.00506, N = 3 SE +/- 0.00549, N = 3 1.84706 1.84739 1.84743 MIN: 1.79 MIN: 1.78 MIN: 1.78 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.3 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K 3 B A 12 24 36 48 60 SE +/- 0.82, N = 15 SE +/- 0.66, N = 15 52.26 51.47 49.58 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Phoronix Test Suite v10.8.5