Tests for a future article by Michael Larabel.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2212135-NE-AOCC40AMD35 AOCC 4.0 AMD EPYC 9374F 2P Compiler Benchmarks - Phoronix Test Suite AOCC 4.0 AMD EPYC 9374F 2P Compiler Benchmarks Tests for a future article by Michael Larabel.
HTML result view exported from: https://openbenchmarking.org/result/2212135-NE-AOCC40AMD35&export=pdf&sor&grr .
AOCC 4.0 AMD EPYC 9374F 2P Compiler Benchmarks Processor Motherboard Chipset Memory Disk Graphics Monitor Network OS Kernel Desktop Display Server Vulkan Compiler File-System Screen Resolution AOCC 4.0 GCC 12.2 2 x AMD EPYC 9374F 32-Core @ 4.31GHz (64 Cores / 128 Threads) AMD Titanite_4G (RTI1002E BIOS) AMD Device 14a4 1520GB 800GB INTEL SSDPF21Q800GB ASPEED VGA HDMI Broadcom NetXtreme BCM5720 PCIe Ubuntu 22.10 5.19.0-26-generic (x86_64) GNOME Shell 43.0 X Server 1.21.1.4 1.3.224 Clang 14.0.6 ext4 1920x1080 GCC 12.2.0 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Environment Details - CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native" Compiler Details - AOCC 4.0: Optimized build with assertions; Default target: x86_64-unknown-linux-gnu; Host CPU: znver4 - GCC 12.2: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-gcn/usr --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: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d Python Details - Python 3.10.7 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
AOCC 4.0 AMD EPYC 9374F 2P Compiler Benchmarks lczero: Eigen cryptopp: Keyed Algorithms lczero: BLAS securemark: SecureMark-TLS aom-av1: Speed 6 Two-Pass - Bosphorus 4K lammps: 20k Atoms vpxenc: Speed 0 - Bosphorus 4K onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU openssl: SHA256 compress-zstd: 19 - Decompression Speed compress-zstd: 19 - Compression Speed vpxenc: Speed 5 - Bosphorus 4K onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU tnn: CPU - MobileNet v2 cryptopp: Unkeyed Algorithms graphics-magick: Sharpen graphics-magick: Rotate graphics-magick: Enhanced openssl: RSA4096 openssl: RSA4096 simdjson: DistinctUserID simdjson: TopTweet simdjson: PartialTweets stargate: 192000 - 512 avifenc: 0 stargate: 192000 - 1024 aom-av1: Speed 9 Realtime - Bosphorus 4K jpegxl-decode: 1 onednn: IP Shapes 1D - bf16bf16bf16 - CPU compress-zstd: 8 - Decompression Speed compress-zstd: 8 - Compression Speed stargate: 96000 - 512 gpaw: Carbon Nanotube kripke: stargate: 96000 - 1024 avifenc: 2 minibude: OpenMP - BM2 minibude: OpenMP - BM2 srsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAM svt-vp9: PSNR/SSIM Optimized - Bosphorus 4K srsran: OFDM_Test jpegxl-decode: All onednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPU svt-vp9: Visual Quality Optimized - Bosphorus 4K liquid-dsp: 128 - 256 - 57 liquid-dsp: 64 - 256 - 57 liquid-dsp: 32 - 256 - 57 svt-vp9: VMAF Optimized - Bosphorus 4K svt-hevc: 10 - Bosphorus 4K kvazaar: Bosphorus 4K - Medium tnn: CPU - SqueezeNet v1.1 onednn: Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPU svt-av1: Preset 13 - Bosphorus 4K onednn: IP Shapes 3D - bf16bf16bf16 - CPU svt-av1: Preset 8 - Bosphorus 4K kvazaar: Bosphorus 4K - Very Fast kvazaar: Bosphorus 4K - Ultra Fast astcenc: Thorough onednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPU webp: Quality 100, Highest Compression svt-hevc: 7 - Bosphorus 4K tnn: CPU - SqueezeNet v2 avifenc: 6, Lossless astcenc: Medium avifenc: 10, Lossless onednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPU avifenc: 6 webp: Default AOCC 4.0 GCC 12.2 16982 796.930644 11900 370262 20.98 43.544 7.99 292.766 122861312743 4123.1 104.6 19.51 524.101 343.974 534.930083 1150 760 1393 1270745.6 19492.8 9.55 8.54 9.17 3.057509 53.154 3.411187 37.39 60.57 0.45944 4722.5 6155.2 4.962698 32.671 340106367 5.425232 29.833 185.534 4638.340 476.6 189.03 192833333 297.49 1.72218 160.53 5868866667 5559600000 2884000000 193.13 157.12 34.23 287.194 0.112382 220.119 0.278467 96.552 71.17 82.71 64.0015 0.320302 4.86 147.93 54.618 4.342 457.1693 3.256 0.587671 2.421 22.78 11707 779.239837 11110 339503 17.66 42.170 7.89 664.985 119198709000 4203.8 101.4 18.99 869.007 229.793 448.186823 864 732 1407 1266612.9 19467.8 8.03 7.59 7.84 2.813028 57.292 3.221105 36.32 54.96 4.28328 4913.7 6060.7 4.477606 35.003 308884107 4.960055 31.974 181.582 4539.557 481.0 169.47 189500000 280.47 2.03703 120.03 5451600000 5280200000 2675133333 173.56 125.58 32.42 226.006 0.250318 187.192 1.56802 92.543 66.46 82.07 62.9653 0.399774 3.59 121.95 52.410 4.737 403.5316 3.422 0.637696 2.573 21.89 OpenBenchmarking.org
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.28 Backend: Eigen AOCC 4.0 GCC 12.2 4K 8K 12K 16K 20K SE +/- 173.43, N = 9 SE +/- 95.82, N = 9 16982 11707 1. (CXX) g++ options: -flto -O3 -march=native -pthread
Crypto++ Test: Keyed Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Keyed Algorithms AOCC 4.0 GCC 12.2 200 400 600 800 1000 SE +/- 1.21, N = 3 SE +/- 0.35, N = 3 796.93 779.24 1. (CXX) g++ options: -O3 -march=native -fPIC -pthread -pipe
LeelaChessZero Backend: BLAS OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.28 Backend: BLAS AOCC 4.0 GCC 12.2 3K 6K 9K 12K 15K SE +/- 62.91, N = 3 SE +/- 56.50, N = 3 11900 11110 1. (CXX) g++ options: -flto -O3 -march=native -pthread
SecureMark Benchmark: SecureMark-TLS OpenBenchmarking.org marks, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS AOCC 4.0 GCC 12.2 80K 160K 240K 320K 400K SE +/- 398.31, N = 3 SE +/- 1573.48, N = 3 370262 339503 1. (CC) gcc options: -pedantic -O3
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 5 10 15 20 25 SE +/- 0.37, N = 12 SE +/- 0.18, N = 15 20.98 17.66 1. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm
LAMMPS Molecular Dynamics Simulator Model: 20k Atoms OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms AOCC 4.0 GCC 12.2 10 20 30 40 50 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 43.54 42.17 1. (CXX) g++ options: -O3 -march=native -lm -ldl
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 AOCC 4.0 GCC 12.2 2 4 6 8 10 SE +/- 0.06, N = 12 SE +/- 0.06, N = 3 7.99 7.89 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=native -fPIC -U_FORTIFY_SOURCE -std=gnu++11
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU AOCC 4.0 GCC 12.2 140 280 420 560 700 SE +/- 0.18, N = 3 SE +/- 4.69, N = 12 292.77 664.99 -fopenmp=libomp - MIN: 283.8 -fopenmp - MIN: 635.26 1. (CXX) g++ options: -O3 -march=native -msse4.1 -fPIC -pie -ldl -lpthread
OpenSSL Algorithm: SHA256 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.0 Algorithm: SHA256 AOCC 4.0 GCC 12.2 30000M 60000M 90000M 120000M 150000M SE +/- 7229089.97, N = 3 SE +/- 3427994.13, N = 3 122861312743 119198709000 -Qunused-arguments 1. (CC) gcc options: -pthread -m64 -O3 -march=native -lssl -lcrypto -ldl
Zstd Compression Compression Level: 19 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Decompression Speed GCC 12.2 AOCC 4.0 900 1800 2700 3600 4500 SE +/- 53.58, N = 3 SE +/- 17.78, N = 12 4203.8 4123.1 1. (CC) gcc options: -O3 -march=native -pthread -lz -llzma
Zstd Compression Compression Level: 19 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Compression Speed AOCC 4.0 GCC 12.2 20 40 60 80 100 SE +/- 1.53, N = 12 SE +/- 0.61, N = 3 104.6 101.4 1. (CC) gcc options: -O3 -march=native -pthread -lz -llzma
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 AOCC 4.0 GCC 12.2 5 10 15 20 25 SE +/- 0.17, N = 3 SE +/- 0.23, N = 15 19.51 18.99 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=native -fPIC -U_FORTIFY_SOURCE -std=gnu++11
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU AOCC 4.0 GCC 12.2 200 400 600 800 1000 SE +/- 0.35, N = 3 SE +/- 6.48, N = 3 524.10 869.01 -fopenmp=libomp - MIN: 510.54 -fopenmp - MIN: 839.81 1. (CXX) g++ options: -O3 -march=native -msse4.1 -fPIC -pie -ldl -lpthread
TNN Target: CPU - Model: MobileNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: MobileNet v2 GCC 12.2 AOCC 4.0 70 140 210 280 350 SE +/- 0.12, N = 3 SE +/- 4.23, N = 15 229.79 343.97 -fopenmp - MIN: 227.65 / MAX: 234.53 -fopenmp=libomp - MIN: 274.57 / MAX: 495.33 1. (CXX) g++ options: -O3 -march=native -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl
Crypto++ Test: Unkeyed Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Unkeyed Algorithms AOCC 4.0 GCC 12.2 120 240 360 480 600 SE +/- 0.08, N = 3 SE +/- 0.18, N = 3 534.93 448.19 1. (CXX) g++ options: -O3 -march=native -fPIC -pthread -pipe
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Sharpen AOCC 4.0 GCC 12.2 200 400 600 800 1000 SE +/- 2.65, N = 3 SE +/- 3.61, N = 3 1150 864 1. (CC) gcc options: -fopenmp -O3 -march=native -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -lpthread
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Rotate AOCC 4.0 GCC 12.2 160 320 480 640 800 SE +/- 0.67, N = 3 SE +/- 1.53, N = 3 760 732 1. (CC) gcc options: -fopenmp -O3 -march=native -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -lpthread
GraphicsMagick Operation: Enhanced OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Enhanced GCC 12.2 AOCC 4.0 300 600 900 1200 1500 SE +/- 1.86, N = 3 SE +/- 6.33, N = 3 1407 1393 1. (CC) gcc options: -fopenmp -O3 -march=native -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -lpthread
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.0 Algorithm: RSA4096 AOCC 4.0 GCC 12.2 300K 600K 900K 1200K 1500K SE +/- 66.97, N = 3 SE +/- 2377.84, N = 3 1270745.6 1266612.9 -Qunused-arguments 1. (CC) gcc options: -pthread -m64 -O3 -march=native -lssl -lcrypto -ldl
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.0 Algorithm: RSA4096 AOCC 4.0 GCC 12.2 4K 8K 12K 16K 20K SE +/- 0.40, N = 3 SE +/- 18.34, N = 3 19492.8 19467.8 -Qunused-arguments 1. (CC) gcc options: -pthread -m64 -O3 -march=native -lssl -lcrypto -ldl
simdjson Throughput Test: DistinctUserID OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: DistinctUserID AOCC 4.0 GCC 12.2 3 6 9 12 15 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 9.55 8.03 1. (CXX) g++ options: -O3 -march=native
simdjson Throughput Test: TopTweet OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: TopTweet AOCC 4.0 GCC 12.2 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 8.54 7.59 1. (CXX) g++ options: -O3 -march=native
simdjson Throughput Test: PartialTweets OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: PartialTweets AOCC 4.0 GCC 12.2 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 9.17 7.84 1. (CXX) g++ options: -O3 -march=native
Stargate Digital Audio Workstation Sample Rate: 192000 - Buffer Size: 512 OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 512 AOCC 4.0 GCC 12.2 0.6879 1.3758 2.0637 2.7516 3.4395 SE +/- 0.001667, N = 3 SE +/- 0.012940, N = 3 3.057509 2.813028 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 0 AOCC 4.0 GCC 12.2 13 26 39 52 65 SE +/- 0.15, N = 3 SE +/- 0.49, N = 3 53.15 57.29 1. (CXX) g++ options: -O3 -fPIC -march=native -lm
Stargate Digital Audio Workstation Sample Rate: 192000 - Buffer Size: 1024 OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 1024 AOCC 4.0 GCC 12.2 0.7675 1.535 2.3025 3.07 3.8375 SE +/- 0.006797, N = 3 SE +/- 0.009147, N = 3 3.411187 3.221105 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 9 18 27 36 45 SE +/- 0.45, N = 3 SE +/- 0.49, N = 15 37.39 36.32 1. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm
JPEG XL Decoding libjxl CPU Threads: 1 OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.7 CPU Threads: 1 AOCC 4.0 GCC 12.2 14 28 42 56 70 SE +/- 0.22, N = 3 SE +/- 0.05, N = 3 60.57 54.96
oneDNN Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU AOCC 4.0 GCC 12.2 0.9637 1.9274 2.8911 3.8548 4.8185 SE +/- 0.00415, N = 3 SE +/- 0.05940, N = 15 0.45944 4.28328 -fopenmp - MIN: 2.97 1. (CXX) g++ options: -O3 -march=native -msse4.1 -fPIC -pie -ldl -lpthread
Zstd Compression Compression Level: 8 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 8 - Decompression Speed GCC 12.2 AOCC 4.0 1100 2200 3300 4400 5500 SE +/- 29.00, N = 4 SE +/- 10.29, N = 3 4913.7 4722.5 1. (CC) gcc options: -O3 -march=native -pthread -lz -llzma
Zstd Compression Compression Level: 8 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 8 - Compression Speed AOCC 4.0 GCC 12.2 1300 2600 3900 5200 6500 SE +/- 60.88, N = 3 SE +/- 65.79, N = 4 6155.2 6060.7 1. (CC) gcc options: -O3 -march=native -pthread -lz -llzma
Stargate Digital Audio Workstation Sample Rate: 96000 - Buffer Size: 512 OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 512 AOCC 4.0 GCC 12.2 1.1166 2.2332 3.3498 4.4664 5.583 SE +/- 0.008965, N = 3 SE +/- 0.016850, N = 3 4.962698 4.477606 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
GPAW Input: Carbon Nanotube OpenBenchmarking.org Seconds, Fewer Is Better GPAW 22.1 Input: Carbon Nanotube AOCC 4.0 GCC 12.2 8 16 24 32 40 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 32.67 35.00 1. (CC) gcc options: -shared -fwrapv -O2 -O3 -march=native -lxc -lblas -lmpi
Kripke OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.4 AOCC 4.0 GCC 12.2 70M 140M 210M 280M 350M SE +/- 1235045.49, N = 3 SE +/- 4347432.93, N = 15 340106367 308884107 -fopenmp=libomp -fopenmp 1. (CXX) g++ options: -O3 -march=native
Stargate Digital Audio Workstation Sample Rate: 96000 - Buffer Size: 1024 OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 1024 AOCC 4.0 GCC 12.2 1.2207 2.4414 3.6621 4.8828 6.1035 SE +/- 0.005282, N = 3 SE +/- 0.012443, N = 3 5.425232 4.960055 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 2 AOCC 4.0 GCC 12.2 7 14 21 28 35 SE +/- 0.28, N = 3 SE +/- 0.21, N = 3 29.83 31.97 1. (CXX) g++ options: -O3 -fPIC -march=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 AOCC 4.0 GCC 12.2 40 80 120 160 200 SE +/- 0.16, N = 3 SE +/- 0.39, N = 3 185.53 181.58 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 AOCC 4.0 GCC 12.2 1000 2000 3000 4000 5000 SE +/- 4.03, N = 3 SE +/- 9.63, N = 3 4638.34 4539.56 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM GCC 12.2 AOCC 4.0 100 200 300 400 500 SE +/- 1.94, N = 3 SE +/- 0.50, N = 3 481.0 476.6 -latomic 1. (CXX) g++ options: -O3 -march=native -std=c++14 -fno-strict-aliasing -mfpmath=sse -mavx2 -fvisibility=hidden -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
SVT-VP9 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 40 80 120 160 200 SE +/- 1.57, N = 9 SE +/- 1.94, N = 15 189.03 169.47 1. (CC) gcc options: -O3 -fcommon -march=native -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
srsRAN Test: OFDM_Test OpenBenchmarking.org Samples / Second, More Is Better srsRAN 22.04.1 Test: OFDM_Test AOCC 4.0 GCC 12.2 40M 80M 120M 160M 200M SE +/- 1299145.02, N = 3 SE +/- 781024.97, N = 3 192833333 189500000 -latomic 1. (CXX) g++ options: -O3 -march=native -std=c++14 -fno-strict-aliasing -mfpmath=sse -mavx2 -fvisibility=hidden -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
JPEG XL Decoding libjxl CPU Threads: All OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.7 CPU Threads: All AOCC 4.0 GCC 12.2 60 120 180 240 300 SE +/- 2.41, N = 3 SE +/- 0.55, N = 3 297.49 280.47
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU AOCC 4.0 GCC 12.2 0.4583 0.9166 1.3749 1.8332 2.2915 SE +/- 0.00581, N = 3 SE +/- 0.00916, N = 3 1.72218 2.03703 -fopenmp=libomp - MIN: 1.53 -fopenmp - MIN: 1.86 1. (CXX) g++ options: -O3 -march=native -msse4.1 -fPIC -pie -ldl -lpthread
SVT-VP9 Tuning: Visual Quality Optimized - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 40 80 120 160 200 SE +/- 1.60, N = 15 SE +/- 0.72, N = 3 160.53 120.03 1. (CC) gcc options: -O3 -fcommon -march=native -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Liquid-DSP Threads: 128 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 128 - Buffer Length: 256 - Filter Length: 57 AOCC 4.0 GCC 12.2 1300M 2600M 3900M 5200M 6500M SE +/- 2562117.18, N = 3 SE +/- 5550075.07, N = 3 5868866667 5451600000 1. (CC) gcc options: -O3 -march=native -pthread -lm -lc -lliquid
Liquid-DSP Threads: 64 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 64 - Buffer Length: 256 - Filter Length: 57 AOCC 4.0 GCC 12.2 1200M 2400M 3600M 4800M 6000M SE +/- 7275300.68, N = 3 SE +/- 18956353.38, N = 3 5559600000 5280200000 1. (CC) gcc options: -O3 -march=native -pthread -lm -lc -lliquid
Liquid-DSP Threads: 32 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 32 - Buffer Length: 256 - Filter Length: 57 AOCC 4.0 GCC 12.2 600M 1200M 1800M 2400M 3000M SE +/- 4590206.97, N = 3 SE +/- 12676794.20, N = 3 2884000000 2675133333 1. (CC) gcc options: -O3 -march=native -pthread -lm -lc -lliquid
SVT-VP9 Tuning: VMAF Optimized - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 40 80 120 160 200 SE +/- 1.04, N = 3 SE +/- 1.38, N = 15 193.13 173.56 1. (CC) gcc options: -O3 -fcommon -march=native -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
SVT-HEVC Tuning: 10 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 30 60 90 120 150 SE +/- 1.46, N = 3 SE +/- 0.86, N = 15 157.12 125.58 1. (CC) gcc options: -O3 -march=native -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt
Kvazaar Video Input: Bosphorus 4K - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Medium AOCC 4.0 GCC 12.2 8 16 24 32 40 SE +/- 0.00, N = 3 SE +/- 0.10, N = 3 34.23 32.42 -lpthread 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -march=native -lm -lrt
TNN Target: CPU - Model: SqueezeNet v1.1 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: SqueezeNet v1.1 GCC 12.2 AOCC 4.0 60 120 180 240 300 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 226.01 287.19 -fopenmp - MIN: 225.65 / MAX: 227.03 -fopenmp=libomp - MIN: 286.77 / MAX: 288 1. (CXX) g++ options: -O3 -march=native -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU AOCC 4.0 GCC 12.2 0.0563 0.1126 0.1689 0.2252 0.2815 SE +/- 0.001257, N = 3 SE +/- 0.002764, N = 4 0.112382 0.250318 -fopenmp=libomp - MIN: 0.1 -fopenmp - MIN: 0.22 1. (CXX) g++ options: -O3 -march=native -msse4.1 -fPIC -pie -ldl -lpthread
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 13 - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 50 100 150 200 250 SE +/- 2.79, N = 3 SE +/- 2.73, N = 15 220.12 187.19 1. (CXX) g++ options: -O3 -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
oneDNN Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU AOCC 4.0 GCC 12.2 0.3528 0.7056 1.0584 1.4112 1.764 SE +/- 0.000429, N = 3 SE +/- 0.018545, N = 3 0.278467 1.568020 -fopenmp=libomp - MIN: 0.24 -fopenmp - MIN: 1.2 1. (CXX) g++ options: -O3 -march=native -msse4.1 -fPIC -pie -ldl -lpthread
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 8 - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 20 40 60 80 100 SE +/- 0.24, N = 3 SE +/- 1.00, N = 3 96.55 92.54 1. (CXX) g++ options: -O3 -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Kvazaar Video Input: Bosphorus 4K - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Very Fast AOCC 4.0 GCC 12.2 16 32 48 64 80 SE +/- 0.20, N = 3 SE +/- 0.07, N = 3 71.17 66.46 -lpthread 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -march=native -lm -lrt
Kvazaar Video Input: Bosphorus 4K - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast AOCC 4.0 GCC 12.2 20 40 60 80 100 SE +/- 0.41, N = 3 SE +/- 0.55, N = 3 82.71 82.07 -lpthread 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -march=native -lm -lrt
ASTC Encoder Preset: Thorough OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Thorough AOCC 4.0 GCC 12.2 14 28 42 56 70 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 64.00 62.97 1. (CXX) g++ options: -O3 -march=native -flto -pthread
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU AOCC 4.0 GCC 12.2 0.0899 0.1798 0.2697 0.3596 0.4495 SE +/- 0.000615, N = 3 SE +/- 0.002555, N = 3 0.320302 0.399774 -fopenmp=libomp - MIN: 0.31 -fopenmp - MIN: 0.36 1. (CXX) g++ options: -O3 -march=native -msse4.1 -fPIC -pie -ldl -lpthread
WebP Image Encode Encode Settings: Quality 100, Highest Compression OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Highest Compression AOCC 4.0 GCC 12.2 1.0935 2.187 3.2805 4.374 5.4675 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 4.86 3.59 1. (CC) gcc options: -fvisibility=hidden -O3 -march=native -lm
SVT-HEVC Tuning: 7 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 4K AOCC 4.0 GCC 12.2 30 60 90 120 150 SE +/- 1.90, N = 3 SE +/- 1.03, N = 3 147.93 121.95 1. (CC) gcc options: -O3 -march=native -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt
TNN Target: CPU - Model: SqueezeNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: SqueezeNet v2 GCC 12.2 AOCC 4.0 12 24 36 48 60 SE +/- 0.31, N = 3 SE +/- 0.59, N = 5 52.41 54.62 -fopenmp - MIN: 51.53 / MAX: 54.74 -fopenmp=libomp - MIN: 52.11 / MAX: 57.47 1. (CXX) g++ options: -O3 -march=native -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 6, Lossless AOCC 4.0 GCC 12.2 1.0658 2.1316 3.1974 4.2632 5.329 SE +/- 0.008, N = 3 SE +/- 0.028, N = 3 4.342 4.737 1. (CXX) g++ options: -O3 -fPIC -march=native -lm
ASTC Encoder Preset: Medium OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Medium AOCC 4.0 GCC 12.2 100 200 300 400 500 SE +/- 0.25, N = 3 SE +/- 0.24, N = 3 457.17 403.53 1. (CXX) g++ options: -O3 -march=native -flto -pthread
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 10, Lossless AOCC 4.0 GCC 12.2 0.77 1.54 2.31 3.08 3.85 SE +/- 0.004, N = 3 SE +/- 0.011, N = 3 3.256 3.422 1. (CXX) g++ options: -O3 -fPIC -march=native -lm
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU AOCC 4.0 GCC 12.2 0.1435 0.287 0.4305 0.574 0.7175 SE +/- 0.001000, N = 3 SE +/- 0.003124, N = 3 0.587671 0.637696 -fopenmp=libomp - MIN: 0.54 -fopenmp - MIN: 0.61 1. (CXX) g++ options: -O3 -march=native -msse4.1 -fPIC -pie -ldl -lpthread
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 6 AOCC 4.0 GCC 12.2 0.5789 1.1578 1.7367 2.3156 2.8945 SE +/- 0.019, N = 3 SE +/- 0.011, N = 3 2.421 2.573 1. (CXX) g++ options: -O3 -fPIC -march=native -lm
WebP Image Encode Encode Settings: Default OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Default AOCC 4.0 GCC 12.2 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 22.78 21.89 1. (CC) gcc options: -fvisibility=hidden -O3 -march=native -lm
Phoronix Test Suite v10.8.4