NVIDIA GH200 Compilers Clang and GCC benchmarks by Michael Larabel for a future article. ARMv8 Neoverse-V2 testing with a Quanta Cloud QuantaGrid S74G-2U 1S7GZ9Z0000 S7G MB (CG1) (3A06 BIOS) and ASPEED on Ubuntu 23.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2402098-NE-NVIDIAGH291&grr&sro .
NVIDIA GH200 Compilers Processor Motherboard Memory Disk Graphics Network OS Kernel Compiler File-System Screen Resolution GCC 13 Clang 17 ARMv8 Neoverse-V2 @ 3.39GHz (72 Cores) Quanta Cloud QuantaGrid S74G-2U 1S7GZ9Z0000 S7G MB (CG1) (3A06 BIOS) 1 x 480GB DRAM-6400MT/s 960GB SAMSUNG MZ1L2960HCJR-00A07 + 1920GB SAMSUNG MZTL21T9 ASPEED 2 x Mellanox MT2910 + 2 x QLogic FastLinQ QL41000 10/25/40/50GbE Ubuntu 23.10 6.8.0-060800rc3daily20240208-generic-64k (aarch64) GCC 13.2.0 ext4 1920x1200 Clang 17.0.2 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Environment Details - CXXFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2" CFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2" Compiler Details - GCC 13: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --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-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v Processor Details - Scaling Governor: cppc_cpufreq performance (Boost: Disabled) Python Details - Python 3.11.6 Security Details - gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected
NVIDIA GH200 Compilers securemark: SecureMark-TLS helsing: 14 digit graphics-magick: Sharpen lammps: 20k Atoms stress-ng: CPU Cache avifenc: 0 minibude: OpenMP - BM2 minibude: OpenMP - BM2 stress-ng: Fused Multiply-Add compress-zstd: 19, Long Mode - Decompression Speed compress-zstd: 19, Long Mode - Compression Speed compress-zstd: 19 - Decompression Speed compress-zstd: 19 - Compression Speed avifenc: 2 graphics-magick: Noise-Gaussian graphics-magick: Rotate graphics-magick: Resizing graphics-magick: Enhanced graphics-magick: Swirl graphics-magick: HWB Color Space encode-opus: WAV To Opus Encode webp: Quality 100, Lossless, Highest Compression encode-flac: WAV To FLAC primesieve: 1e13 quantlib: Single-Threaded liquid-dsp: 72 - 256 - 512 liquid-dsp: 64 - 256 - 512 liquid-dsp: 32 - 256 - 512 stress-ng: Floating Point stress-ng: Vector Floating Point stress-ng: Matrix Math stress-ng: Vector Math liquid-dsp: 72 - 256 - 57 liquid-dsp: 64 - 256 - 57 liquid-dsp: 72 - 256 - 32 liquid-dsp: 64 - 256 - 32 liquid-dsp: 32 - 256 - 57 liquid-dsp: 32 - 256 - 32 liquid-dsp: 1 - 256 - 512 liquid-dsp: 1 - 256 - 57 liquid-dsp: 1 - 256 - 32 quantlib: Multi-Threaded webp: Quality 100, Lossless stress-ng: Vector Shuffle minibude: OpenMP - BM1 minibude: OpenMP - BM1 lulesh: c-ray: Total Time - 4K, 16 Rays Per Pixel encode-mp3: WAV To MP3 webp: Quality 100, Highest Compression avifenc: 6, Lossless primesieve: 1e12 avifenc: 10, Lossless webp: Quality 100 webp: Default lammps: Rhodopsin Protein tscp: AI Chess Performance GCC 13 Clang 17 265718 68.102 882 48.233 949580.78 109.674 48.041 1201.027 161511818.72 1283.6 8.57 1237.4 14.7 67.073 1920 1764 8044 2170 3672 4731 33.037 0.52 16.872 35.191 3456.0 215500000 191886667 95931000 19830.04 83730.08 515044.15 387369.41 1767800000 1587900000 2952333333 2671533333 795033333 1362833333 3194567 26470667 45523000 232068.2 1.28 71014.33 47.755 1193.878 48090.643 6.006 5.474 3.86 3.754 2.891 2.851 9.44 13.95 55.360 2078407 267498 84.332 1761 49.521 932492.75 122.871 60.979 1524.466 157339813.59 1094.4 8.70 1031.1 14.9 79.371 1440 1820 7919 1542 3748 4360 31.420 0.56 16.078 35.597 3740.4 231370000 206270000 102853333 19566.28 141522.24 550915.68 450187.94 2406400000 2182700000 4386000000 4032833333 1096366667 2066233333 3414867 36488333 68921000 249451.3 1.31 62.075 1551.880 47590.091 6.749 6.287 4.66 3.717 2.929 2.871 10.19 15.42 56.335 2248073 OpenBenchmarking.org
SecureMark Benchmark: SecureMark-TLS OpenBenchmarking.org marks, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS Clang 17 GCC 13 60K 120K 180K 240K 300K SE +/- 54.60, N = 3 SE +/- 525.41, N = 3 267498 265718 1. (CC) gcc options: -pedantic -O3
Helsing Digit Range: 14 digit OpenBenchmarking.org Seconds, Fewer Is Better Helsing 1.0-beta Digit Range: 14 digit Clang 17 GCC 13 20 40 60 80 100 SE +/- 0.72, N = 3 SE +/- 0.40, N = 15 84.33 68.10 1. (CC) gcc options: -O2 -pthread
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Sharpen Clang 17 GCC 13 400 800 1200 1600 2000 SE +/- 11.15, N = 15 SE +/- 1.86, N = 3 1761 882 1. (CC) gcc options: -fopenmp -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread
LAMMPS Molecular Dynamics Simulator Model: 20k Atoms OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms Clang 17 GCC 13 11 22 33 44 55 SE +/- 0.56, N = 3 SE +/- 0.15, N = 3 49.52 48.23 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm -ldl
Stress-NG Test: CPU Cache OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: CPU Cache Clang 17 GCC 13 200K 400K 600K 800K 1000K SE +/- 44765.07, N = 15 SE +/- 34515.94, N = 15 932492.75 949580.78
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 0 Clang 17 GCC 13 30 60 90 120 150 SE +/- 0.34, N = 3 SE +/- 0.43, N = 3 122.87 109.67 1. (CXX) g++ options: -O3 -fPIC -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 Clang 17 GCC 13 14 28 42 56 70 SE +/- 0.68, N = 3 SE +/- 0.09, N = 3 60.98 48.04 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 Clang 17 GCC 13 300 600 900 1200 1500 SE +/- 16.91, N = 3 SE +/- 2.12, N = 3 1524.47 1201.03 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm
Stress-NG Test: Fused Multiply-Add OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Fused Multiply-Add Clang 17 GCC 13 30M 60M 90M 120M 150M SE +/- 1087775.35, N = 15 SE +/- 1745267.88, N = 3 157339813.59 161511818.72
Zstd Compression Compression Level: 19, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed Clang 17 GCC 13 300 600 900 1200 1500 SE +/- 2.48, N = 3 SE +/- 3.15, N = 3 1094.4 1283.6 -Qunused-arguments 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lz -llzma
Zstd Compression Compression Level: 19, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Compression Speed Clang 17 GCC 13 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 8.70 8.57 -Qunused-arguments 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lz -llzma
Zstd Compression Compression Level: 19 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed Clang 17 GCC 13 300 600 900 1200 1500 SE +/- 4.91, N = 3 SE +/- 13.01, N = 3 1031.1 1237.4 -Qunused-arguments 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lz -llzma
Zstd Compression Compression Level: 19 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Compression Speed Clang 17 GCC 13 4 8 12 16 20 SE +/- 0.17, N = 3 SE +/- 0.12, N = 3 14.9 14.7 -Qunused-arguments 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lz -llzma
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 2 Clang 17 GCC 13 20 40 60 80 100 SE +/- 0.27, N = 3 SE +/- 0.03, N = 3 79.37 67.07 1. (CXX) g++ options: -O3 -fPIC -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
GraphicsMagick Operation: Noise-Gaussian OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Noise-Gaussian Clang 17 GCC 13 400 800 1200 1600 2000 SE +/- 4.67, N = 3 SE +/- 0.67, N = 3 1440 1920 1. (CC) gcc options: -fopenmp -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Rotate Clang 17 GCC 13 400 800 1200 1600 2000 SE +/- 25.64, N = 3 SE +/- 9.35, N = 3 1820 1764 1. (CC) gcc options: -fopenmp -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Resizing Clang 17 GCC 13 2K 4K 6K 8K 10K SE +/- 70.92, N = 3 SE +/- 39.89, N = 3 7919 8044 1. (CC) gcc options: -fopenmp -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread
GraphicsMagick Operation: Enhanced OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Enhanced Clang 17 GCC 13 500 1000 1500 2000 2500 SE +/- 14.05, N = 3 SE +/- 11.57, N = 3 1542 2170 1. (CC) gcc options: -fopenmp -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread
GraphicsMagick Operation: Swirl OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Swirl Clang 17 GCC 13 800 1600 2400 3200 4000 SE +/- 48.77, N = 3 SE +/- 5.24, N = 3 3748 3672 1. (CC) gcc options: -fopenmp -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: HWB Color Space Clang 17 GCC 13 1000 2000 3000 4000 5000 SE +/- 16.18, N = 3 SE +/- 41.68, N = 3 4360 4731 1. (CC) gcc options: -fopenmp -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread
Opus Codec Encoding WAV To Opus Encode OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.4 WAV To Opus Encode Clang 17 GCC 13 8 16 24 32 40 SE +/- 0.01, N = 5 SE +/- 0.02, N = 5 31.42 33.04 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -fvisibility=hidden -logg -lm
WebP Image Encode Encode Settings: Quality 100, Lossless, Highest Compression OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless, Highest Compression Clang 17 GCC 13 0.126 0.252 0.378 0.504 0.63 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.56 0.52 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.4 WAV To FLAC Clang 17 GCC 13 4 8 12 16 20 SE +/- 0.13, N = 9 SE +/- 0.19, N = 5 16.08 16.87 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -fvisibility=hidden -logg -lm
Primesieve Length: 1e13 OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 8.0 Length: 1e13 Clang 17 GCC 13 8 16 24 32 40 SE +/- 0.34, N = 3 SE +/- 0.42, N = 3 35.60 35.19 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2
QuantLib Configuration: Single-Threaded OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Single-Threaded Clang 17 GCC 13 800 1600 2400 3200 4000 SE +/- 37.89, N = 6 SE +/- 35.85, N = 3 3740.4 3456.0 1. (CXX) g++ options: -O3 -march=native -fPIE -pie
Liquid-DSP Threads: 72 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 72 - Buffer Length: 256 - Filter Length: 512 Clang 17 GCC 13 50M 100M 150M 200M 250M SE +/- 91651.51, N = 3 SE +/- 79372.54, N = 3 231370000 215500000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 64 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 512 Clang 17 GCC 13 40M 80M 120M 160M 200M SE +/- 37859.39, N = 3 SE +/- 67659.28, N = 3 206270000 191886667 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 32 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 512 Clang 17 GCC 13 20M 40M 60M 80M 100M SE +/- 196751.39, N = 3 SE +/- 129816.54, N = 3 102853333 95931000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Stress-NG Test: Floating Point OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Floating Point Clang 17 GCC 13 4K 8K 12K 16K 20K SE +/- 4.32, N = 3 SE +/- 0.79, N = 3 19566.28 19830.04
Stress-NG Test: Vector Floating Point OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Floating Point Clang 17 GCC 13 30K 60K 90K 120K 150K SE +/- 38.13, N = 3 SE +/- 275.28, N = 3 141522.24 83730.08
Stress-NG Test: Matrix Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Matrix Math Clang 17 GCC 13 120K 240K 360K 480K 600K SE +/- 356.30, N = 3 SE +/- 2880.52, N = 3 550915.68 515044.15
Stress-NG Test: Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Math Clang 17 GCC 13 100K 200K 300K 400K 500K SE +/- 100.06, N = 3 SE +/- 25.00, N = 3 450187.94 387369.41
Liquid-DSP Threads: 72 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 72 - Buffer Length: 256 - Filter Length: 57 Clang 17 GCC 13 500M 1000M 1500M 2000M 2500M SE +/- 8154140.05, N = 3 SE +/- 5356304.70, N = 3 2406400000 1767800000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 64 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 57 Clang 17 GCC 13 500M 1000M 1500M 2000M 2500M SE +/- 4628534.69, N = 3 SE +/- 57735.03, N = 3 2182700000 1587900000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 72 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 72 - Buffer Length: 256 - Filter Length: 32 Clang 17 GCC 13 900M 1800M 2700M 3600M 4500M SE +/- 26602506.15, N = 3 SE +/- 13903516.74, N = 3 4386000000 2952333333 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 64 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 32 Clang 17 GCC 13 900M 1800M 2700M 3600M 4500M SE +/- 37196788.99, N = 3 SE +/- 18636374.23, N = 3 4032833333 2671533333 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 32 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 57 Clang 17 GCC 13 200M 400M 600M 800M 1000M SE +/- 120185.04, N = 3 SE +/- 153441.99, N = 3 1096366667 795033333 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 32 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 32 Clang 17 GCC 13 400M 800M 1200M 1600M 2000M SE +/- 993870.10, N = 3 SE +/- 1337078.07, N = 3 2066233333 1362833333 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 1 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 512 Clang 17 GCC 13 700K 1400K 2100K 2800K 3500K SE +/- 2635.86, N = 3 SE +/- 1117.04, N = 3 3414867 3194567 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 1 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 57 Clang 17 GCC 13 8M 16M 24M 32M 40M SE +/- 1666.67, N = 3 SE +/- 14240.01, N = 3 36488333 26470667 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 1 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 32 Clang 17 GCC 13 15M 30M 45M 60M 75M SE +/- 6082.76, N = 3 SE +/- 30138.57, N = 3 68921000 45523000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
QuantLib Configuration: Multi-Threaded OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Multi-Threaded Clang 17 GCC 13 50K 100K 150K 200K 250K SE +/- 2974.17, N = 3 SE +/- 3083.44, N = 3 249451.3 232068.2 1. (CXX) g++ options: -O3 -march=native -fPIE -pie
WebP Image Encode Encode Settings: Quality 100, Lossless OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless Clang 17 GCC 13 0.2948 0.5896 0.8844 1.1792 1.474 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.31 1.28 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
Stress-NG Test: Vector Shuffle OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Shuffle GCC 13 15K 30K 45K 60K 75K SE +/- 173.05, N = 3 71014.33 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -std=gnu99 -U_FORTIFY_SOURCE -O2 -lc
miniBUDE Implementation: OpenMP - Input Deck: BM1 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 Clang 17 GCC 13 14 28 42 56 70 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 62.08 47.76 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM1 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 Clang 17 GCC 13 300 600 900 1200 1500 SE +/- 1.53, N = 3 SE +/- 2.51, N = 3 1551.88 1193.88 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm
LULESH OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 Clang 17 GCC 13 10K 20K 30K 40K 50K SE +/- 108.91, N = 3 SE +/- 108.54, N = 3 47590.09 48090.64 1. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi
C-Ray Total Time - 4K, 16 Rays Per Pixel OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Clang 17 GCC 13 2 4 6 8 10 SE +/- 0.009, N = 3 SE +/- 0.003, N = 3 6.749 6.006 1. (CC) gcc options: -lm -lpthread -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 Clang 17 GCC 13 2 4 6 8 10 SE +/- 0.001, N = 3 SE +/- 0.003, N = 3 6.287 5.474 -lncurses -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr 1. (CC) gcc options: -O3 -pipe -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
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 Clang 17 GCC 13 1.0485 2.097 3.1455 4.194 5.2425 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 4.66 3.86 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6, Lossless Clang 17 GCC 13 0.8447 1.6894 2.5341 3.3788 4.2235 SE +/- 0.008, N = 3 SE +/- 0.007, N = 3 3.717 3.754 1. (CXX) g++ options: -O3 -fPIC -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
Primesieve Length: 1e12 OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 8.0 Length: 1e12 Clang 17 GCC 13 0.659 1.318 1.977 2.636 3.295 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 2.929 2.891 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 10, Lossless Clang 17 GCC 13 0.646 1.292 1.938 2.584 3.23 SE +/- 0.003, N = 3 SE +/- 0.006, N = 3 2.871 2.851 1. (CXX) g++ options: -O3 -fPIC -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
WebP Image Encode Encode Settings: Quality 100 OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100 Clang 17 GCC 13 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 10.19 9.44 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
WebP Image Encode Encode Settings: Default OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Default Clang 17 GCC 13 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 15.42 13.95 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
LAMMPS Molecular Dynamics Simulator Model: Rhodopsin Protein OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: Rhodopsin Protein Clang 17 GCC 13 13 26 39 52 65 SE +/- 0.11, N = 3 SE +/- 0.12, N = 3 56.34 55.36 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm -ldl
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance Clang 17 GCC 13 500K 1000K 1500K 2000K 2500K SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 2248073 2078407 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -march=native
Phoronix Test Suite v10.8.5