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.
GCC 13 Processor: ARMv8 Neoverse-V2 @ 3.39GHz (72 Cores), Motherboard: Quanta Cloud QuantaGrid S74G-2U 1S7GZ9Z0000 S7G MB (CG1) (3A06 BIOS), Memory: 1 x 480GB DRAM-6400MT/s, Disk: 960GB SAMSUNG MZ1L2960HCJR-00A07 + 1920GB SAMSUNG MZTL21T9, Graphics: ASPEED, Network: 2 x Mellanox MT2910 + 2 x QLogic FastLinQ QL41000 10/25/40/50GbE
OS: Ubuntu 23.10, Kernel: 6.8.0-060800rc3daily20240208-generic-64k (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2" CFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2"Compiler Notes: --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 -vProcessor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)Python Notes: Python 3.11.6Security Notes: 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
Clang 17 OS: Ubuntu 23.10, Kernel: 6.8.0-060800rc3daily20240208-generic-64k (aarch64), Compiler: Clang 17.0.2, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2" CFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2"Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)Python Notes: Python 3.11.6Security Notes: 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 OpenBenchmarking.org Phoronix Test Suite 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 Clang 17.0.2 ext4 1920x1200 Processor Motherboard Memory Disk Graphics Network OS Kernel Compilers File-System Screen Resolution NVIDIA GH200 Compilers Performance System Logs - Transparent Huge Pages: madvise - CXXFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2" CFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2" - 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 - Scaling Governor: cppc_cpufreq performance (Boost: Disabled) - Python 3.11.6 - 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
GCC 13 vs. Clang 17 Comparison Phoronix Test Suite Baseline +24.9% +24.9% +49.8% +49.8% +74.7% +74.7% +99.6% +99.6% 99.7% 69% 51.6% 51.4% 51% 48.6% 37.9% 37.8% 37.5% 36.1% 30% 30% 26.9% 26.9% 20.7% 16.2% 10.5% 8.2% 8.2% 7.9% 7.7% 7.5% 7.5% 7.4% 7.2% 7% 6.9% 5.1% 4.9% 3.2% 2.7% 2.3% 2.1% Sharpen V.F.P 32 - 256 - 32 1 - 256 - 32 64 - 256 - 32 72 - 256 - 32 Enhanced 40.7% 32 - 256 - 57 1 - 256 - 57 64 - 256 - 57 72 - 256 - 57 Noise-Gaussian 33.3% OpenMP - BM1 OpenMP - BM1 OpenMP - BM2 OpenMP - BM2 14 digit 23.8% Q.1.H.C 19 - D.S 20% 2 18.3% 19, Long Mode - D.S 17.3% Vector Math WAV To MP3 14.9% Total Time - 4.1.R.P.P 12.4% 12% Default HWB Color Space 8.5% Single-Threaded A.C.P Quality 100 Q.1.L.H.C 64 - 256 - 512 Multi-Threaded 72 - 256 - 512 32 - 256 - 512 Matrix Math 1 - 256 - 512 WAV To Opus Encode WAV To FLAC Rotate 20k Atoms Fused Multiply-Add 2.7% Q.1.L Swirl GraphicsMagick Stress-NG Liquid-DSP Liquid-DSP Liquid-DSP Liquid-DSP GraphicsMagick Liquid-DSP Liquid-DSP Liquid-DSP Liquid-DSP GraphicsMagick miniBUDE miniBUDE miniBUDE miniBUDE Helsing WebP Image Encode Zstd Compression libavif avifenc Zstd Compression Stress-NG LAME MP3 Encoding C-Ray libavif avifenc WebP Image Encode GraphicsMagick QuantLib TSCP WebP Image Encode WebP Image Encode Liquid-DSP QuantLib Liquid-DSP Liquid-DSP Stress-NG Liquid-DSP Opus Codec Encoding FLAC Audio Encoding GraphicsMagick LAMMPS Molecular Dynamics Simulator Stress-NG WebP Image Encode GraphicsMagick GCC 13 Clang 17
NVIDIA GH200 Compilers quantlib: Multi-Threaded quantlib: Single-Threaded minibude: OpenMP - BM1 minibude: OpenMP - BM1 minibude: OpenMP - BM2 minibude: OpenMP - BM2 lammps: 20k Atoms lammps: Rhodopsin Protein lulesh: compress-zstd: 19 - Compression Speed compress-zstd: 19 - Decompression Speed compress-zstd: 19, Long Mode - Compression Speed compress-zstd: 19, Long Mode - Decompression Speed webp: Default webp: Quality 100 webp: Quality 100, Lossless webp: Quality 100, Highest Compression webp: Quality 100, Lossless, Highest Compression tscp: AI Chess Performance graphics-magick: Swirl graphics-magick: Rotate graphics-magick: Sharpen graphics-magick: Enhanced graphics-magick: Resizing graphics-magick: Noise-Gaussian graphics-magick: HWB Color Space avifenc: 0 avifenc: 2 avifenc: 6, Lossless avifenc: 10, Lossless c-ray: Total Time - 4K, 16 Rays Per Pixel primesieve: 1e12 primesieve: 1e13 encode-flac: WAV To FLAC encode-mp3: WAV To MP3 encode-opus: WAV To Opus Encode helsing: 14 digit securemark: SecureMark-TLS liquid-dsp: 1 - 256 - 32 liquid-dsp: 1 - 256 - 57 liquid-dsp: 1 - 256 - 512 liquid-dsp: 32 - 256 - 32 liquid-dsp: 32 - 256 - 57 liquid-dsp: 64 - 256 - 32 liquid-dsp: 64 - 256 - 57 liquid-dsp: 72 - 256 - 32 liquid-dsp: 72 - 256 - 57 liquid-dsp: 32 - 256 - 512 liquid-dsp: 64 - 256 - 512 liquid-dsp: 72 - 256 - 512 stress-ng: CPU Cache stress-ng: Matrix Math stress-ng: Vector Math stress-ng: Floating Point stress-ng: Vector Shuffle stress-ng: Fused Multiply-Add stress-ng: Vector Floating Point GCC 13 Clang 17 232068.2 3456.0 1193.878 47.755 1201.027 48.041 48.233 55.360 48090.643 14.7 1237.4 8.57 1283.6 13.95 9.44 1.28 3.86 0.52 2078407 3672 1764 882 2170 8044 1920 4731 109.674 67.073 3.754 2.851 6.006 2.891 35.191 16.872 5.474 33.037 68.102 265718 45523000 26470667 3194567 1362833333 795033333 2671533333 1587900000 2952333333 1767800000 95931000 191886667 215500000 949580.78 515044.15 387369.41 19830.04 71014.33 161511818.72 83730.08 249451.3 3740.4 1551.880 62.075 1524.466 60.979 49.521 56.335 47590.091 14.9 1031.1 8.70 1094.4 15.42 10.19 1.31 4.66 0.56 2248073 3748 1820 1761 1542 7919 1440 4360 122.871 79.371 3.717 2.871 6.749 2.929 35.597 16.078 6.287 31.420 84.332 267498 68921000 36488333 3414867 2066233333 1096366667 4032833333 2182700000 4386000000 2406400000 102853333 206270000 231370000 932492.75 550915.68 450187.94 19566.28 157339813.59 141522.24 OpenBenchmarking.org
QuantLib QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Multi-Threaded GCC 13 Clang 17 50K 100K 150K 200K 250K SE +/- 3083.44, N = 3 SE +/- 2974.17, N = 3 232068.2 249451.3 1. (CXX) g++ options: -O3 -march=native -fPIE -pie
OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Single-Threaded GCC 13 Clang 17 800 1600 2400 3200 4000 SE +/- 35.85, N = 3 SE +/- 37.89, N = 6 3456.0 3740.4 1. (CXX) g++ options: -O3 -march=native -fPIE -pie
miniBUDE MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 GCC 13 Clang 17 300 600 900 1200 1500 SE +/- 2.51, N = 3 SE +/- 1.53, N = 3 1193.88 1551.88 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm
OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 GCC 13 Clang 17 14 28 42 56 70 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 47.76 62.08 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm
OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 GCC 13 Clang 17 300 600 900 1200 1500 SE +/- 2.12, N = 3 SE +/- 16.91, N = 3 1201.03 1524.47 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm
OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 GCC 13 Clang 17 14 28 42 56 70 SE +/- 0.09, N = 3 SE +/- 0.68, N = 3 48.04 60.98 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm
Zstd Compression This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Compression Speed GCC 13 Clang 17 4 8 12 16 20 SE +/- 0.12, N = 3 SE +/- 0.17, N = 3 14.7 14.9 -Qunused-arguments 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lz -llzma
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed GCC 13 Clang 17 300 600 900 1200 1500 SE +/- 13.01, N = 3 SE +/- 4.91, N = 3 1237.4 1031.1 -Qunused-arguments 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lz -llzma
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Compression Speed GCC 13 Clang 17 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 8.57 8.70 -Qunused-arguments 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lz -llzma
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed GCC 13 Clang 17 300 600 900 1200 1500 SE +/- 3.15, N = 3 SE +/- 2.48, N = 3 1283.6 1094.4 -Qunused-arguments 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lz -llzma
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100 GCC 13 Clang 17 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 9.44 10.19 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless GCC 13 Clang 17 0.2948 0.5896 0.8844 1.1792 1.474 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.28 1.31 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Highest Compression GCC 13 Clang 17 1.0485 2.097 3.1455 4.194 5.2425 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3.86 4.66 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless, Highest Compression GCC 13 Clang 17 0.126 0.252 0.378 0.504 0.63 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.52 0.56 1. (CC) gcc options: -fvisibility=hidden -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
TSCP This is a performance test of TSCP, Tom Kerrigan's Simple Chess Program, which has a built-in performance benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance GCC 13 Clang 17 500K 1000K 1500K 2000K 2500K SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 2078407 2248073 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -march=native
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Swirl GCC 13 Clang 17 800 1600 2400 3200 4000 SE +/- 5.24, N = 3 SE +/- 48.77, N = 3 3672 3748 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
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Rotate GCC 13 Clang 17 400 800 1200 1600 2000 SE +/- 9.35, N = 3 SE +/- 25.64, N = 3 1764 1820 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
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Sharpen GCC 13 Clang 17 400 800 1200 1600 2000 SE +/- 1.86, N = 3 SE +/- 11.15, N = 15 882 1761 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
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Enhanced GCC 13 Clang 17 500 1000 1500 2000 2500 SE +/- 11.57, N = 3 SE +/- 14.05, N = 3 2170 1542 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
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Resizing GCC 13 Clang 17 2K 4K 6K 8K 10K SE +/- 39.89, N = 3 SE +/- 70.92, N = 3 8044 7919 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
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Noise-Gaussian GCC 13 Clang 17 400 800 1200 1600 2000 SE +/- 0.67, N = 3 SE +/- 4.67, N = 3 1920 1440 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
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: HWB Color Space GCC 13 Clang 17 1000 2000 3000 4000 5000 SE +/- 41.68, N = 3 SE +/- 16.18, N = 3 4731 4360 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
OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 2 GCC 13 Clang 17 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.27, N = 3 67.07 79.37 1. (CXX) g++ options: -O3 -fPIC -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6, Lossless GCC 13 Clang 17 0.8447 1.6894 2.5341 3.3788 4.2235 SE +/- 0.007, N = 3 SE +/- 0.008, N = 3 3.754 3.717 1. (CXX) g++ options: -O3 -fPIC -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 10, Lossless GCC 13 Clang 17 0.646 1.292 1.938 2.584 3.23 SE +/- 0.006, N = 3 SE +/- 0.003, N = 3 2.851 2.871 1. (CXX) g++ options: -O3 -fPIC -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
C-Ray This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel GCC 13 Clang 17 2 4 6 8 10 SE +/- 0.003, N = 3 SE +/- 0.009, N = 3 6.006 6.749 1. (CC) gcc options: -lm -lpthread -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2
Primesieve Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 8.0 Length: 1e12 GCC 13 Clang 17 0.659 1.318 1.977 2.636 3.295 SE +/- 0.002, N = 3 SE +/- 0.001, N = 3 2.891 2.929 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2
OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 8.0 Length: 1e13 GCC 13 Clang 17 8 16 24 32 40 SE +/- 0.42, N = 3 SE +/- 0.34, N = 3 35.19 35.60 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2
LAME MP3 Encoding LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 GCC 13 Clang 17 2 4 6 8 10 SE +/- 0.003, N = 3 SE +/- 0.001, N = 3 5.474 6.287 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -lncurses 1. (CC) gcc options: -O3 -pipe -mtune=neoverse-v2 -mcpu=neoverse-v2 -lm
Opus Codec Encoding Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus five times. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.4 WAV To Opus Encode GCC 13 Clang 17 8 16 24 32 40 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 33.04 31.42 1. (CXX) g++ options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -fvisibility=hidden -logg -lm
SecureMark SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions developed by EEMBC. SecureMark-TLS is benchmarking Transport Layer Security performance with a focus on IoT/edge computing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org marks, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS GCC 13 Clang 17 60K 120K 180K 240K 300K SE +/- 525.41, N = 3 SE +/- 54.60, N = 3 265718 267498 1. (CC) gcc options: -pedantic -O3
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 32 GCC 13 Clang 17 15M 30M 45M 60M 75M SE +/- 30138.57, N = 3 SE +/- 6082.76, N = 3 45523000 68921000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 57 GCC 13 Clang 17 8M 16M 24M 32M 40M SE +/- 14240.01, N = 3 SE +/- 1666.67, N = 3 26470667 36488333 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 512 GCC 13 Clang 17 700K 1400K 2100K 2800K 3500K SE +/- 1117.04, N = 3 SE +/- 2635.86, N = 3 3194567 3414867 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 32 GCC 13 Clang 17 400M 800M 1200M 1600M 2000M SE +/- 1337078.07, N = 3 SE +/- 993870.10, N = 3 1362833333 2066233333 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 57 GCC 13 Clang 17 200M 400M 600M 800M 1000M SE +/- 153441.99, N = 3 SE +/- 120185.04, N = 3 795033333 1096366667 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 32 GCC 13 Clang 17 900M 1800M 2700M 3600M 4500M SE +/- 18636374.23, N = 3 SE +/- 37196788.99, N = 3 2671533333 4032833333 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 57 GCC 13 Clang 17 500M 1000M 1500M 2000M 2500M SE +/- 57735.03, N = 3 SE +/- 4628534.69, N = 3 1587900000 2182700000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 72 - Buffer Length: 256 - Filter Length: 32 GCC 13 Clang 17 900M 1800M 2700M 3600M 4500M SE +/- 13903516.74, N = 3 SE +/- 26602506.15, N = 3 2952333333 4386000000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 72 - Buffer Length: 256 - Filter Length: 57 GCC 13 Clang 17 500M 1000M 1500M 2000M 2500M SE +/- 5356304.70, N = 3 SE +/- 8154140.05, N = 3 1767800000 2406400000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 512 GCC 13 Clang 17 20M 40M 60M 80M 100M SE +/- 129816.54, N = 3 SE +/- 196751.39, N = 3 95931000 102853333 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 512 GCC 13 Clang 17 40M 80M 120M 160M 200M SE +/- 67659.28, N = 3 SE +/- 37859.39, N = 3 191886667 206270000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 72 - Buffer Length: 256 - Filter Length: 512 GCC 13 Clang 17 50M 100M 150M 200M 250M SE +/- 79372.54, N = 3 SE +/- 91651.51, N = 3 215500000 231370000 1. (CC) gcc options: -O3 -mtune=neoverse-v2 -mcpu=neoverse-v2 -pthread -lm -lc -lliquid
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 Test: Vector Shuffle
Clang 17: The test run did not produce a result.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Fused Multiply-Add GCC 13 Clang 17 30M 60M 90M 120M 150M SE +/- 1745267.88, N = 3 SE +/- 1087775.35, N = 15 161511818.72 157339813.59
GCC 13 Processor: ARMv8 Neoverse-V2 @ 3.39GHz (72 Cores), Motherboard: Quanta Cloud QuantaGrid S74G-2U 1S7GZ9Z0000 S7G MB (CG1) (3A06 BIOS), Memory: 1 x 480GB DRAM-6400MT/s, Disk: 960GB SAMSUNG MZ1L2960HCJR-00A07 + 1920GB SAMSUNG MZTL21T9, Graphics: ASPEED, Network: 2 x Mellanox MT2910 + 2 x QLogic FastLinQ QL41000 10/25/40/50GbE
OS: Ubuntu 23.10, Kernel: 6.8.0-060800rc3daily20240208-generic-64k (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2" CFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2"Compiler Notes: --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 -vProcessor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)Python Notes: Python 3.11.6Security Notes: 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
Testing initiated at 9 February 2024 15:23 by user x.
Clang 17 Processor: ARMv8 Neoverse-V2 @ 3.39GHz (72 Cores), Motherboard: Quanta Cloud QuantaGrid S74G-2U 1S7GZ9Z0000 S7G MB (CG1) (3A06 BIOS), Memory: 1 x 480GB DRAM-6400MT/s, Disk: 960GB SAMSUNG MZ1L2960HCJR-00A07 + 1920GB SAMSUNG MZTL21T9, Graphics: ASPEED, Network: 2 x Mellanox MT2910 + 2 x QLogic FastLinQ QL41000 10/25/40/50GbE
OS: Ubuntu 23.10, Kernel: 6.8.0-060800rc3daily20240208-generic-64k (aarch64), Compiler: Clang 17.0.2, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2" CFLAGS="-O3 -mtune=neoverse-v2 -mcpu=neoverse-v2"Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)Python Notes: Python 3.11.6Security Notes: 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
Testing initiated at 9 February 2024 19:05 by user x.