Ryzen 9 5950X Clang 12 vs. GCC 11 Benchmarks

GCC 11.1 versus LLVM Clang 12 on AMD Ryzen 9 5950X. Benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2105198-IB-11900KCOM08&sgm=1&imw=1&grs.

ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionGCC 11.1Clang 12 -O2 -O3 -march=native -O3 -march=native -flto -O2 -O3 -march=native -O3 -march=native -fltoAMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3302 BIOS)AMD Starship/Matisse32GB500GB Western Digital WDS500G3X0C-00SJG0AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (2100/875MHz)AMD Navi 10 HDMI AudioASUS MG28URealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Fedora 345.11.20-300.fc34.x86_64 (x86_64)GNOME Shell 40.1X Server + Wayland4.6 Mesa 21.0.3 (LLVM 12.0.0)GCC 11.1.1 20210428 + Clang 12.0.0btrfs3840x2160Clang 12.0.0OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseEnvironment Details- GCC 11.1: -O2: CXXFLAGS=-O2 CFLAGS=-O2- GCC 11.1: -O3 -march=native: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"- GCC 11.1: -O3 -march=native -flto: CXXFLAGS="-O3 -march=native -flto" CFLAGS="-O3 -march=native -flto"- Clang 12: -O2: CXXFLAGS=-O2 CFLAGS=-O2- Clang 12: -O3 -march=native: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"- Clang 12: -O3 -march=native -flto: CXXFLAGS="-O3 -march=native -flto" CFLAGS="-O3 -march=native -flto"Compiler Details- GCC 11.1: -O2, GCC 11.1: -O3 -march=native, GCC 11.1: -O3 -march=native -flto: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver Processor Details- Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009Security Details- SELinux + 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 IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

c-ray: Total Time - 4K, 16 Rays Per Pixelgraphics-magick: Sharpenncnn: CPU - blazefacetnn: CPU - MobileNet v2ncnn: CPU - shufflenet-v2ncnn: CPU - regnety_400mastcenc: Mediumncnn: CPU-v3-v3 - mobilenet-v3encode-flac: WAV To FLACncnn: CPU-v2-v2 - mobilenet-v2aobench: 2048 x 2048 - Total Timencnn: CPU - mnasnetencode-mp3: WAV To MP3ncnn: CPU - efficientnet-b0astcenc: Thoroughgraphics-magick: Resizingncnn: CPU - mobilenetencode-opus: WAV To Opus Encodecoremark: CoreMark Size 666 - Iterations Per Secondncnn: CPU - squeezenet_ssdncnn: CPU - resnet50ncnn: CPU - googlenetncnn: CPU - yolov4-tinywebp: Quality 100, Highest Compressiongraphics-magick: Rotateastcenc: Exhaustivegraphics-magick: Enhancedx265: Bosphorus 4Ksvt-hevc: 7 - Bosphorus 1080pliquid-dsp: 8 - 256 - 57webp: Quality 100, Losslesshmmer: Pfam Database Searchncnn: CPU - resnet18mrbayes: Primate Phylogeny Analysistnn: CPU - SqueezeNet v1.1webp: Quality 100, Lossless, Highest Compressionpjsip: INVITEliquid-dsp: 16 - 256 - 57lammps: Rhodopsin Proteinsqlite-speedtest: Timed Time - Size 1,000pjsip: OPTIONS, Statelesstjbench: Decompression Throughputncnn: CPU - vgg16svt-hevc: 10 - Bosphorus 1080psvt-vp9: Visual Quality Optimized - Bosphorus 1080pncnn: CPU - alexnetsvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080psvt-vp9: VMAF Optimized - Bosphorus 1080ppjsip: OPTIONS, Statefulhimeno: Poisson Pressure SolverGCC 11.1Clang 12 -O2 -O3 -march=native -O3 -march=native -flto -O2 -O3 -march=native -O3 -march=native -flto60.9312261.85219.1304.3317.244.62964.155.8744.5431.7323.976.8025.417.6841180312.956.515830811.53211614.6126.1813.1521.065.35598557.144542226.24217.3259485000013.18199.72214.5096.785208.78027.4244815104120000012.85547.404230599277.49680256.82365.79222.6311.16235.27230.5378605120.94026225.4053701.80227.7684.3917.184.62773.846.2374.4325.4793.955.5035.377.5718212012.705.387808580.86265013.9325.3312.8521.005.200103656.641144925.91221.7161156000013.55097.74414.3392.973214.27728.3274671108500000012.74946.917221759268.11345657.61369.54223.1110.99235.59231.1679825314.73390425.4823822.55215.3416.0118.534.69953.866.1954.4626.3123.925.4035.367.5299208513.745.475849671.95944214.5025.8413.2623.945.18496456.190044926.17227.0461939333313.25497.71414.8093.337202.57227.9494616109326666712.81346.479222581270.90915057.84368.55222.3311.13234.70230.4379165445.63522149.2752061.54271.0283.8112.933.33673.367.6873.5633.7783.266.5654.529.3950169311.585.874712758.35782612.6122.9911.8821.974.77892152.204841127.77225.1557666000013.64796.09214.0593.529208.55529.1194575105640000013.08747.792221107267.55306857.26368.55225.5711.09235.64231.3579424877.30641045.0102351.52355.9253.7412.593.31793.075.6663.5730.2343.266.0984.519.3549176711.465.508722694.00627512.3523.2511.7721.734.70398151.823045227.75229.2757605000014.10595.21413.9090.785215.78128.6404708106190000013.29947.630222572272.51355956.93375.95227.7311.16239.88234.7179585192.19004244.7422371.46261.8363.6912.143.37282.935.7563.4229.7303.085.8104.309.4432176211.055.558714933.85898312.4322.9011.6221.584.72296151.897445128.47236.0558780000013.63693.31113.8692.405207.92527.400107346666713.34248.584266.22982956.11375.08227.6611.25238.13232.665289.917571OpenBenchmarking.org

C-Ray

Total Time - 4K, 16 Rays Per Pixel

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per PixelGCC 11.1Clang 121428425670SE +/- 0.18, N = 3SE +/- 0.16, N = 3SE +/- 0.08, N = 3SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.17, N = 360.9349.2825.4145.0125.4844.741. (CC) gcc options: -lm -lpthread -O3 -march=native -flto

GraphicsMagick

Operation: Sharpen

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SharpenGCC 11.1Clang 1280160240320400SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 1.20, N = 3SE +/- 0.33, N = 3SE +/- 0.88, N = 3SE +/- 0.67, N = 32262063702353822371. (CC) gcc options: -fopenmp -O3 -march=native -flto -pthread -ljpeg -lz -lm -lpthread

NCNN

Target: CPU - Model: blazeface

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazefaceGCC 11.1Clang 120.57381.14761.72142.29522.869SE +/- 0.01, N = 15SE +/- 0.04, N = 3SE +/- 0.01, N = 15SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 31.851.541.801.522.551.461. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

TNN

Target: CPU - Model: MobileNet v2

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v2GCC 11.1Clang 1280160240320400SE +/- 1.57, N = 3SE +/- 0.41, N = 3SE +/- 0.86, N = 3SE +/- 0.48, N = 3SE +/- 1.27, N = 3SE +/- 0.55, N = 3219.13271.03227.77355.93215.34261.841. (CXX) g++ options: -O3 -march=native -flto -pthread -fvisibility=hidden -O2 -rdynamic -ldl

NCNN

Target: CPU - Model: shufflenet-v2

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v2GCC 11.1Clang 12246810SE +/- 0.01, N = 14SE +/- 0.05, N = 3SE +/- 0.15, N = 15SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 34.333.814.393.746.013.691. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

NCNN

Target: CPU - Model: regnety_400m

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400mGCC 11.1Clang 12510152025SE +/- 0.06, N = 15SE +/- 0.20, N = 3SE +/- 0.05, N = 15SE +/- 0.16, N = 3SE +/- 0.09, N = 3SE +/- 0.14, N = 317.2412.9317.1812.5918.5312.141. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

ASTC Encoder

Preset: Medium

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: MediumGCC 11.1Clang 121.05742.11483.17224.22965.287SE +/- 0.0337, N = 3SE +/- 0.0156, N = 3SE +/- 0.0292, N = 3SE +/- 0.0243, N = 3SE +/- 0.0045, N = 3SE +/- 0.0302, N = 34.62963.33674.62773.31794.69953.37281. (CXX) g++ options: -O3 -march=native -flto -O2 -pthread

NCNN

Target: CPU-v3-v3 - Model: mobilenet-v3

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3GCC 11.1Clang 120.93381.86762.80143.73524.669SE +/- 0.03, N = 15SE +/- 0.07, N = 3SE +/- 0.02, N = 15SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 34.153.363.843.073.862.931. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

FLAC Audio Encoding

WAV To FLAC

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.2WAV To FLACGCC 11.1Clang 12246810SE +/- 0.045, N = 5SE +/- 0.024, N = 5SE +/- 0.020, N = 5SE +/- 0.005, N = 5SE +/- 0.036, N = 5SE +/- 0.030, N = 55.8747.6876.2375.6666.1955.7561. (CXX) g++ options: -O3 -march=native -flto -logg -lm

NCNN

Target: CPU-v2-v2 - Model: mobilenet-v2

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v2GCC 11.1Clang 121.02152.0433.06454.0865.1075SE +/- 0.04, N = 15SE +/- 0.11, N = 3SE +/- 0.01, N = 15SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 34.543.564.433.574.463.421. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

AOBench

Size: 2048 x 2048 - Total Time

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total TimeGCC 11.1Clang 12816243240SE +/- 0.14, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.43, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 331.7333.7825.4830.2326.3129.731. (CC) gcc options: -lm -O3 -march=native -flto

NCNN

Target: CPU - Model: mnasnet

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnetGCC 11.1Clang 120.89331.78662.67993.57324.4665SE +/- 0.04, N = 15SE +/- 0.08, N = 3SE +/- 0.04, N = 15SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 33.973.263.953.263.923.081. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

LAME MP3 Encoding

WAV To MP3

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3GCC 11.1Clang 12246810SE +/- 0.084, N = 3SE +/- 0.068, N = 4SE +/- 0.071, N = 3SE +/- 0.045, N = 3SE +/- 0.010, N = 3SE +/- 0.056, N = 36.8026.5655.5036.0985.4035.8101. (CC) gcc options: -O3 -pipe -march=native -flto -lm

NCNN

Target: CPU - Model: efficientnet-b0

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b0GCC 11.1Clang 121.21732.43463.65194.86926.0865SE +/- 0.05, N = 15SE +/- 0.06, N = 3SE +/- 0.03, N = 15SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 35.414.525.374.515.364.301. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

ASTC Encoder

Preset: Thorough

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: ThoroughGCC 11.1Clang 123691215SE +/- 0.0109, N = 3SE +/- 0.0279, N = 3SE +/- 0.0351, N = 3SE +/- 0.0407, N = 3SE +/- 0.0109, N = 3SE +/- 0.0114, N = 37.68419.39507.57189.35497.52999.44321. (CXX) g++ options: -O3 -march=native -flto -O2 -pthread

GraphicsMagick

Operation: Resizing

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: ResizingGCC 11.1Clang 125001000150020002500SE +/- 0.88, N = 3SE +/- 4.48, N = 3SE +/- 1.53, N = 3SE +/- 3.61, N = 3SE +/- 3.71, N = 3SE +/- 2.85, N = 31803169321201767208517621. (CC) gcc options: -fopenmp -O3 -march=native -flto -pthread -ljpeg -lz -lm -lpthread

NCNN

Target: CPU - Model: mobilenet

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenetGCC 11.1Clang 1248121620SE +/- 0.11, N = 15SE +/- 0.03, N = 3SE +/- 0.13, N = 15SE +/- 0.11, N = 3SE +/- 0.17, N = 3SE +/- 0.12, N = 312.9511.5812.7011.4613.7411.051. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

Opus Codec Encoding

WAV To Opus Encode

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodeGCC 11.1Clang 12246810SE +/- 0.035, N = 5SE +/- 0.052, N = 5SE +/- 0.048, N = 5SE +/- 0.045, N = 5SE +/- 0.023, N = 5SE +/- 0.016, N = 56.5155.8745.3875.5085.4755.5581. (CXX) g++ options: -O3 -march=native -flto -logg -lm

Coremark

CoreMark Size 666 - Iterations Per Second

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondGCC 11.1Clang 12200K400K600K800K1000KSE +/- 989.68, N = 3SE +/- 789.22, N = 3SE +/- 2230.96, N = 3SE +/- 845.83, N = 3SE +/- 836.68, N = 3SE +/- 581.76, N = 3830811.53712758.36808580.86722694.01849671.96714933.861. (CC) gcc options: -O2 -O3 -march=native -flto -lrt" -lrt

NCNN

Target: CPU - Model: squeezenet_ssd

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssdGCC 11.1Clang 1248121620SE +/- 0.10, N = 15SE +/- 0.31, N = 3SE +/- 0.09, N = 15SE +/- 0.20, N = 3SE +/- 0.06, N = 3SE +/- 0.20, N = 314.6112.6113.9312.3514.5012.431. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

NCNN

Target: CPU - Model: resnet50

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50GCC 11.1Clang 12612182430SE +/- 0.18, N = 15SE +/- 0.11, N = 3SE +/- 0.10, N = 15SE +/- 0.06, N = 3SE +/- 0.21, N = 3SE +/- 0.09, N = 326.1822.9925.3323.2525.8422.901. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

NCNN

Target: CPU - Model: googlenet

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenetGCC 11.1Clang 123691215SE +/- 0.22, N = 15SE +/- 0.13, N = 3SE +/- 0.12, N = 15SE +/- 0.16, N = 3SE +/- 0.44, N = 3SE +/- 0.09, N = 313.1511.8812.8511.7713.2611.621. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

NCNN

Target: CPU - Model: yolov4-tiny

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tinyGCC 11.1Clang 12612182430SE +/- 0.11, N = 15SE +/- 0.18, N = 3SE +/- 0.14, N = 15SE +/- 0.12, N = 3SE +/- 0.41, N = 3SE +/- 0.10, N = 321.0621.9721.0021.7323.9421.581. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

WebP Image Encode

Encode Settings: Quality 100, Highest Compression

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionGCC 11.1Clang 121.20492.40983.61474.81966.0245SE +/- 0.021, N = 3SE +/- 0.059, N = 4SE +/- 0.023, N = 3SE +/- 0.026, N = 3SE +/- 0.036, N = 3SE +/- 0.027, N = 35.3554.7785.2004.7035.1844.7221. (CC) gcc options: -fvisibility=hidden -O3 -march=native -flto -pthread -lm -ljpeg

GraphicsMagick

Operation: Rotate

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: RotateGCC 11.1Clang 122004006008001000SE +/- 11.70, N = 3SE +/- 5.55, N = 3SE +/- 5.57, N = 3SE +/- 0.58, N = 3SE +/- 3.48, N = 3SE +/- 4.37, N = 398592110369819649611. (CC) gcc options: -fopenmp -O3 -march=native -flto -pthread -ljpeg -lz -lm -lpthread

ASTC Encoder

Preset: Exhaustive

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: ExhaustiveGCC 11.1Clang 121326395265SE +/- 0.09, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.11, N = 3SE +/- 0.10, N = 3SE +/- 0.07, N = 357.1452.2056.6451.8256.1951.901. (CXX) g++ options: -O3 -march=native -flto -O2 -pthread

GraphicsMagick

Operation: Enhanced

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: EnhancedGCC 11.1Clang 12100200300400500SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 34224114494524494511. (CC) gcc options: -fopenmp -O3 -march=native -flto -pthread -ljpeg -lz -lm -lpthread

x265

Video Input: Bosphorus 4K

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KGCC 11.1Clang 12714212835SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.10, N = 326.2427.7725.9127.7526.1728.471. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread -lrt -ldl

SVT-HEVC

Tuning: 7 - Input: Bosphorus 1080p

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pGCC 11.1Clang 1250100150200250SE +/- 0.49, N = 3SE +/- 0.95, N = 3SE +/- 0.83, N = 3SE +/- 0.32, N = 3SE +/- 0.03, N = 3SE +/- 1.15, N = 3217.32225.15221.71229.27227.04236.051. (CC) gcc options: -O3 -march=native -flto -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

Liquid-DSP

Threads: 8 - Buffer Length: 256 - Filter Length: 57

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 8 - Buffer Length: 256 - Filter Length: 57GCC 11.1Clang 12130M260M390M520M650MSE +/- 185202.59, N = 3SE +/- 1824399.44, N = 3SE +/- 4455270.29, N = 3SE +/- 4072558.57, N = 3SE +/- 1343755.10, N = 3SE +/- 1855020.22, N = 35948500005766600006115600005760500006193933335878000001. (CC) gcc options: -O3 -march=native -flto -pthread -lm -lc -lliquid

WebP Image Encode

Encode Settings: Quality 100, Lossless

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessGCC 11.1Clang 1248121620SE +/- 0.04, N = 3SE +/- 0.19, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.17, N = 3SE +/- 0.15, N = 413.1813.6513.5514.1113.2513.641. (CC) gcc options: -fvisibility=hidden -O3 -march=native -flto -pthread -lm -ljpeg

Timed HMMer Search

Pfam Database Search

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.2Pfam Database SearchGCC 11.1Clang 1220406080100SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 399.7296.0997.7495.2197.7193.311. (CC) gcc options: -O3 -march=native -flto -pthread -lhmmer -leasel -lm -lmpi

NCNN

Target: CPU - Model: resnet18

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18GCC 11.1Clang 1248121620SE +/- 0.11, N = 15SE +/- 0.23, N = 3SE +/- 0.09, N = 15SE +/- 0.13, N = 3SE +/- 0.22, N = 3SE +/- 0.13, N = 314.5014.0514.3313.9014.8013.861. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

Timed MrBayes Analysis

Primate Phylogeny Analysis

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisGCC 11.1Clang 1220406080100SE +/- 0.13, N = 3SE +/- 0.21, N = 3SE +/- 0.29, N = 3SE +/- 0.18, N = 3SE +/- 0.13, N = 3SE +/- 0.19, N = 396.7993.5392.9790.7993.3492.411. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -O3 -std=c99 -pedantic -march=native -flto -lm

TNN

Target: CPU - Model: SqueezeNet v1.1

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1GCC 11.1Clang 1250100150200250SE +/- 0.08, N = 3SE +/- 0.23, N = 3SE +/- 0.50, N = 3SE +/- 0.69, N = 3SE +/- 0.11, N = 3SE +/- 0.41, N = 3208.78208.56214.28215.78202.57207.931. (CXX) g++ options: -O3 -march=native -flto -pthread -fvisibility=hidden -O2 -rdynamic -ldl

WebP Image Encode

Encode Settings: Quality 100, Lossless, Highest Compression

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionGCC 11.1Clang 12714212835SE +/- 0.03, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 327.4229.1228.3328.6427.9527.401. (CC) gcc options: -fvisibility=hidden -O3 -march=native -flto -pthread -lm -ljpeg

PJSIP

Method: INVITE

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITEGCC 11.1Clang 1210002000300040005000SE +/- 16.33, N = 3SE +/- 42.78, N = 3SE +/- 32.67, N = 15SE +/- 66.36, N = 3SE +/- 65.77, N = 3481545754671470846161. (CC) gcc options: -lstdc++ -lssl -lcrypto -lm -lrt -lpthread -O3 -march=native

Liquid-DSP

Threads: 16 - Buffer Length: 256 - Filter Length: 57

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57GCC 11.1Clang 12200M400M600M800M1000MSE +/- 4650806.38, N = 3SE +/- 2690724.81, N = 3SE +/- 1252996.41, N = 3SE +/- 493288.29, N = 3SE +/- 783865.07, N = 3SE +/- 3219903.38, N = 31041200000105640000010850000001061900000109326666710734666671. (CC) gcc options: -O3 -march=native -flto -pthread -lm -lc -lliquid

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinGCC 11.1Clang 123691215SE +/- 0.11, N = 8SE +/- 0.08, N = 3SE +/- 0.14, N = 4SE +/- 0.02, N = 3SE +/- 0.09, N = 15SE +/- 0.01, N = 312.8613.0912.7513.3012.8113.341. (CXX) g++ options: -O3 -march=native -flto -O2 -pthread -lm

SQLite Speedtest

Timed Time - Size 1,000

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000GCC 11.1Clang 121122334455SE +/- 0.02, N = 3SE +/- 0.13, N = 3SE +/- 0.16, N = 3SE +/- 0.50, N = 3SE +/- 0.54, N = 3SE +/- 0.18, N = 347.4047.7946.9247.6346.4848.581. (CC) gcc options: -O3 -march=native -flto -ldl -lz -lpthread

PJSIP

Method: OPTIONS, Stateless

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatelessGCC 11.1Clang 1250K100K150K200K250KSE +/- 586.86, N = 3SE +/- 1827.48, N = 3SE +/- 1749.48, N = 3SE +/- 962.04, N = 3SE +/- 705.50, N = 32305992211072217592225722225811. (CC) gcc options: -lstdc++ -lssl -lcrypto -lm -lrt -lpthread -O3 -march=native

libjpeg-turbo tjbench

Test: Decompression Throughput

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.1.0Test: Decompression ThroughputGCC 11.1Clang 1260120180240300SE +/- 1.92, N = 3SE +/- 1.65, N = 3SE +/- 1.27, N = 3SE +/- 0.59, N = 3SE +/- 0.45, N = 3SE +/- 0.54, N = 3277.50267.55268.11272.51270.91266.231. (CC) gcc options: -O3 -march=native -flto -rdynamic -lm

NCNN

Target: CPU - Model: vgg16

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg16GCC 11.1Clang 121326395265SE +/- 0.08, N = 15SE +/- 0.08, N = 3SE +/- 0.08, N = 15SE +/- 0.53, N = 3SE +/- 0.23, N = 3SE +/- 0.45, N = 356.8257.2657.6156.9357.8456.111. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pGCC 11.1Clang 1280160240320400SE +/- 1.24, N = 3SE +/- 0.23, N = 3SE +/- 1.07, N = 3SE +/- 1.66, N = 3SE +/- 0.57, N = 3SE +/- 0.90, N = 3365.79368.55369.54375.95368.55375.081. (CC) gcc options: -O3 -march=native -flto -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080pGCC 11.1Clang 1250100150200250SE +/- 0.82, N = 3SE +/- 1.28, N = 3SE +/- 0.87, N = 3SE +/- 0.75, N = 3SE +/- 0.97, N = 3SE +/- 0.89, N = 3222.63225.57223.11227.73222.33227.661. (CC) gcc options: -O3 -fcommon -march=native -flto -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm

NCNN

Target: CPU - Model: alexnet

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnetGCC 11.1Clang 123691215SE +/- 0.06, N = 15SE +/- 0.03, N = 3SE +/- 0.04, N = 15SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 311.1611.0910.9911.1611.1311.251. (CXX) g++ options: -O3 -march=native -flto -O2 -rdynamic -lpthread

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pGCC 11.1Clang 1250100150200250SE +/- 0.37, N = 3SE +/- 0.18, N = 3SE +/- 0.23, N = 3SE +/- 0.40, N = 3SE +/- 0.33, N = 3SE +/- 0.48, N = 3235.27235.64235.59239.88234.70238.131. (CC) gcc options: -O3 -fcommon -march=native -flto -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 1080p

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pGCC 11.1Clang 1250100150200250SE +/- 0.09, N = 3SE +/- 0.33, N = 3SE +/- 0.46, N = 3SE +/- 0.32, N = 3SE +/- 0.18, N = 3SE +/- 0.70, N = 3230.53231.35231.16234.71230.43232.661. (CC) gcc options: -O3 -fcommon -march=native -flto -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm

PJSIP

Method: OPTIONS, Stateful

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatefulGCC 11.1Clang 122K4K6K8K10KSE +/- 72.34, N = 3SE +/- 27.78, N = 3SE +/- 21.79, N = 3SE +/- 47.54, N = 3SE +/- 56.75, N = 3786079427982795879161. (CC) gcc options: -lstdc++ -lssl -lcrypto -lm -lrt -lpthread -O3 -march=native

Himeno Benchmark

Poisson Pressure Solver

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverGCC 11.1Clang 1212002400360048006000SE +/- 83.12, N = 15SE +/- 62.44, N = 15SE +/- 90.91, N = 15SE +/- 66.92, N = 15SE +/- 68.97, N = 15SE +/- 96.57, N = 135120.944877.315314.735192.195445.645289.921. (CC) gcc options: -O3 -march=native -flto -mavx2

Geometric Mean Of All Test Results

Result Composite - Ryzen 9 5950X Clang 12 vs. GCC 11 Benchmarks

-O2-O3 -march=native-O3 -march=native -fltoOpenBenchmarking.orgGeometric Mean, More Is BetterGeometric Mean Of All Test ResultsResult Composite - Ryzen 9 5950X Clang 12 vs. GCC 11 BenchmarksGCC 11.1Clang 12122436486050.7152.5153.5153.7852.5154.73

Phoronix Test Suite v10.8.4