Tests for a future article.
-O3 -march=native Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-O1 Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-O1 CFLAGS=-O1Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-O0 Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-O0 CFLAGS=-O0Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-Og Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-Og CFLAGS=-OgCompiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-O2 Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-O2 CFLAGS=-O2Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-O2 -march=native Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O2 -march=native" CFLAGS="-O2 -march=native"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-O2 -flto Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O2 -flto" CFLAGS="-O2 -flto"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-O3 Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-O3 CFLAGS=-O3Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-O3 -flto Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -flto" CFLAGS="-O3 -flto"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-O3 -march=native -flto Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native -flto" CFLAGS="-O3 -march=native -flto"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-Ofast -march=native Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-Ofast -march=native" CFLAGS="-Ofast -march=native"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
-Ofast Processor: Intel Core i9-11900K @ 5.10GHz (8 Cores / 16 Threads), Motherboard: ASUS ROG MAXIMUS XIII HERO (0707 BIOS), Chipset: Intel Tiger Lake-H, Memory: 32GB, Disk: 2000GB Corsair Force MP600 + 257GB Flash Drive, Graphics: AMD Radeon VII 16GB (1801/1000MHz), Audio: Intel Tiger Lake-H HD Audio, Monitor: ASUS MG28U, Network: 2 x Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411
OS: Fedora 34, Kernel: 5.12.9-300.fc34.x86_64 (x86_64), Desktop: GNOME Shell 40.1, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 21.1.1 (LLVM 12.0.0), Compiler: GCC 11.1.1 20210531, File-System: btrfs, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-Ofast CFLAGS=-OfastCompiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
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 2021.01.31 Threads: 1 - Buffer Length: 256 - Filter Length: 57 -O3 -march=native -O3 -march=native -flto -O2 -march=native -Ofast -march=native -O3 -O3 -flto -Ofast -O2 -O2 -flto -O1 -Og -O0 20M 40M 60M 80M 100M SE +/- 14836.14, N = 3 SE +/- 3711.84, N = 3 SE +/- 3282.95, N = 3 SE +/- 26576.93, N = 3 SE +/- 42880.97, N = 3 SE +/- 1089169.77, N = 3 SE +/- 15213.30, N = 3 SE +/- 996670.85, N = 3 SE +/- 1080865.75, N = 3 SE +/- 6806.86, N = 3 SE +/- 16556.30, N = 3 SE +/- 10115.99, N = 3 99844333 99749333 97389667 96017000 95463333 95045333 94333667 92384333 92353667 88411000 66759667 11024000 -march=native -march=native -flto -O2 -march=native -Ofast -march=native -flto -Ofast -O2 -O2 -flto -O1 -Og 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 2 - Buffer Length: 256 - Filter Length: 57 -O3 -march=native -flto -O3 -march=native -O2 -march=native -Ofast -march=native -Ofast -O3 -flto -O3 -O2 -flto -O2 -O1 -Og -O0 40M 80M 120M 160M 200M SE +/- 612943.72, N = 3 SE +/- 66416.20, N = 3 SE +/- 691182.56, N = 3 SE +/- 30550.50, N = 3 SE +/- 417026.51, N = 3 SE +/- 29627.31, N = 3 SE +/- 2010530.61, N = 3 SE +/- 1464631.16, N = 3 SE +/- 23333.33, N = 3 SE +/- 601728.99, N = 3 SE +/- 10000.00, N = 3 SE +/- 107734.76, N = 3 189120000 188003333 184670000 182180000 178193333 178046667 175400000 173536667 173196667 162046667 126890000 21299667 -march=native -flto -march=native -O2 -march=native -Ofast -march=native -Ofast -flto -O2 -flto -O2 -O1 -Og 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 4 - Buffer Length: 256 - Filter Length: 57 -O3 -march=native -O3 -march=native -flto -O2 -march=native -Ofast -march=native -O3 -flto -O3 -Ofast -O2 -O2 -flto -O1 -Og -O0 80M 160M 240M 320M 400M SE +/- 1410968.93, N = 3 SE +/- 495120.64, N = 3 SE +/- 772492.36, N = 3 SE +/- 763566.19, N = 3 SE +/- 1498358.36, N = 3 SE +/- 1167318.87, N = 3 SE +/- 946379.06, N = 3 SE +/- 1022518.68, N = 3 SE +/- 372394.65, N = 3 SE +/- 132035.35, N = 3 SE +/- 636273.88, N = 3 SE +/- 97119.06, N = 3 363760000 363303333 355393333 349460000 345733333 341970000 341200000 335913333 335793333 316710000 242253333 41181333 -march=native -march=native -flto -O2 -march=native -Ofast -march=native -flto -Ofast -O2 -O2 -flto -O1 -Og 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 8 - Buffer Length: 256 - Filter Length: 57 -O3 -march=native -O3 -march=native -flto -O2 -march=native -Ofast -march=native -O3 -flto -Ofast -O3 -O2 -flto -O2 -O1 -Og -O0 150M 300M 450M 600M 750M SE +/- 689597.31, N = 3 SE +/- 898319.42, N = 3 SE +/- 1217880.67, N = 3 SE +/- 1059124.38, N = 3 SE +/- 597559.85, N = 3 SE +/- 261809.94, N = 3 SE +/- 632200.74, N = 3 SE +/- 355074.33, N = 3 SE +/- 463692.91, N = 3 SE +/- 736168.76, N = 3 SE +/- 764293.43, N = 3 SE +/- 320728.51, N = 3 687846667 687796667 674180000 671553333 650283333 648636667 646743333 637113333 637103333 595816667 462596667 78172667 -march=native -march=native -flto -O2 -march=native -Ofast -march=native -flto -Ofast -O2 -flto -O2 -O1 -Og 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 16 - Buffer Length: 256 - Filter Length: 57 -O3 -march=native -O3 -march=native -flto -O2 -march=native -O3 -flto -O2 -O3 -O2 -flto -Ofast -march=native -O1 -Ofast -Og -O0 150M 300M 450M 600M 750M SE +/- 134824.99, N = 3 SE +/- 265015.72, N = 3 SE +/- 403993.95, N = 3 SE +/- 330201.96, N = 3 SE +/- 63857.48, N = 3 SE +/- 49103.07, N = 3 SE +/- 171302.20, N = 3 SE +/- 331679.23, N = 3 SE +/- 328295.26, N = 3 SE +/- 58594.65, N = 3 SE +/- 130766.97, N = 3 SE +/- 172619.36, N = 3 722756667 722200000 721436667 713980000 711676667 711086667 707806667 684946667 672296667 671610000 517440000 90397667 -march=native -march=native -flto -O2 -march=native -flto -O2 -O2 -flto -Ofast -march=native -O1 -Ofast -Og 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
FLAC Audio Encoding This test times how long it takes to encode a sample WAV file to FLAC format five times. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC -Ofast -march=native -O3 -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -O3 -flto -Ofast -O3 -O2 -O1 -Og -O0 10 20 30 40 50 SE +/- 0.004, N = 5 SE +/- 0.002, N = 5 SE +/- 0.005, N = 5 SE +/- 0.005, N = 5 SE +/- 0.003, N = 5 SE +/- 0.004, N = 5 SE +/- 0.005, N = 5 SE +/- 0.012, N = 5 SE +/- 0.010, N = 5 SE +/- 0.004, N = 5 SE +/- 0.004, N = 5 SE +/- 0.013, N = 5 5.923 5.937 5.938 5.941 6.063 6.073 6.082 6.106 6.108 6.590 6.740 41.971 -Ofast -march=native -O3 -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -O3 -flto -Ofast -O3 -O2 -O1 -Og -O0 1. (CXX) g++ options: -fvisibility=hidden -logg -lm
Crypto++ Crypto++ is a C++ class library of cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Keyed Algorithms -O3 -march=native -Ofast -march=native -O3 -Ofast -O2 -march=native -O3 -march=native -flto -O3 -flto -O2 -O2 -flto -O1 -Og -O0 200 400 600 800 1000 SE +/- 0.64, N = 3 SE +/- 0.32, N = 3 SE +/- 0.39, N = 3 SE +/- 1.09, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 1.27, N = 3 SE +/- 0.99, N = 3 SE +/- 0.80, N = 3 SE +/- 0.51, N = 3 SE +/- 0.17, N = 3 SE +/- 0.19, N = 3 924.21 922.64 910.86 910.43 844.68 842.10 828.89 825.82 794.97 751.48 676.03 135.96 -O3 -march=native -Ofast -march=native -O3 -Ofast -O2 -march=native -O3 -march=native -flto -O3 -flto -O2 -O2 -flto -O1 -Og -O0 1. (CXX) g++ options: -fPIC -pthread -pipe
Timed HMMer Search This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.2 Pfam Database Search -Ofast -march=native -Ofast -O3 -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -O3 -flto -O2 -O3 -O1 -Og -O0 120 240 360 480 600 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.22, N = 3 SE +/- 0.15, N = 3 89.95 93.79 99.48 99.81 100.40 101.85 102.32 102.36 103.13 103.74 114.92 532.07 -Ofast -march=native -Ofast -O3 -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -O3 -flto -O2 -O3 -O1 -Og -O0 1. (CC) gcc options: -pthread -lhmmer -leasel -lm -lmpi
Timed MrBayes Analysis This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis -Ofast -Ofast -march=native -O3 -flto -O3 -O3 -march=native -flto -O3 -march=native -O2 -O2 -flto -O2 -march=native -O1 -Og -O0 60 120 180 240 300 SE +/- 0.20, N = 3 SE +/- 0.27, N = 3 SE +/- 0.43, N = 3 SE +/- 0.35, N = 3 SE +/- 0.79, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.19, N = 3 SE +/- 0.14, N = 3 SE +/- 0.17, N = 3 SE +/- 0.06, N = 3 SE +/- 0.38, N = 3 46.56 50.81 80.56 81.37 83.30 83.43 84.36 85.58 88.44 88.53 93.34 265.31 -Ofast -Ofast -march=native -flto -march=native -flto -march=native -O2 -O2 -flto -O2 -march=native -O1 -Og -O0 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msha -maes -mavx -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mavx512ifma -mavx512vbmi -mrdrnd -mbmi -mbmi2 -madx -mmpx -mabm -O3 -std=c99 -pedantic -lm
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd 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 1.5.0 Compression Level: 3 - Decompression Speed -O2 -Ofast -O3 -O2 -march=native -O2 -flto -O3 -march=native -Ofast -march=native -O3 -flto -O3 -march=native -flto -O1 -Og -O0 1100 2200 3300 4400 5500 SE +/- 3.43, N = 3 SE +/- 20.70, N = 3 SE +/- 7.36, N = 3 SE +/- 4.97, N = 3 SE +/- 2.77, N = 3 SE +/- 19.31, N = 3 SE +/- 9.83, N = 3 SE +/- 20.13, N = 3 SE +/- 18.92, N = 3 SE +/- 8.75, N = 3 SE +/- 14.02, N = 3 SE +/- 1.51, N = 3 5191.9 5174.7 5151.9 5093.5 5087.0 4997.8 4991.4 4972.9 4935.5 4847.5 3791.4 1030.1 -O2 -Ofast -O3 -O2 -march=native -O2 -flto -O3 -march=native -Ofast -march=native -O3 -flto -O3 -march=native -flto -O1 -Og -O0 1. (CC) gcc options: -pthread -lz
Coremark This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second -Ofast -O3 -O3 -march=native -flto -Ofast -march=native -O3 -march=native -O2 -flto -O2 -O3 -flto -O2 -march=native -O1 -Og -O0 100K 200K 300K 400K 500K SE +/- 751.74, N = 3 SE +/- 295.16, N = 3 SE +/- 163.45, N = 3 SE +/- 387.90, N = 3 SE +/- 533.17, N = 3 SE +/- 236.18, N = 3 SE +/- 734.61, N = 3 SE +/- 491.58, N = 3 SE +/- 686.94, N = 3 SE +/- 661.73, N = 3 SE +/- 256.15, N = 3 SE +/- 138.09, N = 3 450629.16 440872.96 436079.65 435243.25 434724.85 429492.03 428580.32 427503.80 423569.01 366951.48 261573.76 90409.83 -Ofast -O3 -O3 -march=native -flto -Ofast -march=native -O3 -march=native -flto -O3 -flto -march=native -O1 -Og -O0 1. (CC) gcc options: -O2 -lrt" -lrt
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd 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 1.5.0 Compression Level: 8 - Decompression Speed -O2 -Ofast -O3 -O2 -march=native -Ofast -march=native -O2 -flto -O3 -march=native -O3 -flto -O3 -march=native -flto -O1 -Og -O0 1200 2400 3600 4800 6000 SE +/- 6.15, N = 3 SE +/- 9.23, N = 3 SE +/- 3.81, N = 3 SE +/- 6.35, N = 3 SE +/- 12.33, N = 3 SE +/- 91.74, N = 3 SE +/- 15.26, N = 3 SE +/- 27.37, N = 3 SE +/- 19.39, N = 3 SE +/- 13.17, N = 3 SE +/- 9.27, N = 3 SE +/- 0.86, N = 3 5394.8 5383.8 5332.9 5324.1 5209.0 5207.4 5189.9 5121.7 5105.5 5075.8 3939.4 1087.0 -O2 -Ofast -O3 -O2 -march=native -Ofast -march=native -O2 -flto -O3 -march=native -O3 -flto -O3 -march=native -flto -O1 -Og -O0 1. (CC) gcc options: -pthread -lz
Botan Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: CAST-256 - Decrypt -Ofast -O3 -O3 -flto -O3 -march=native -Ofast -march=native -O3 -march=native -flto -Og -O2 -march=native -O2 -flto -O2 -O1 -O0 40 80 120 160 200 SE +/- 0.11, N = 3 SE +/- 0.15, N = 3 SE +/- 0.15, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.25, N = 3 SE +/- 0.67, N = 3 SE +/- 1.87, N = 4 SE +/- 2.02, N = 6 SE +/- 3.81, N = 3 SE +/- 1.14, N = 15 SE +/- 0.05, N = 3 178.28 178.20 172.97 168.85 168.83 168.41 161.42 158.89 158.25 150.17 149.81 36.34 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: CAST-256 -Ofast -O3 -O3 -flto -O3 -march=native -flto -Ofast -march=native -O3 -march=native -Og -O2 -flto -O2 -O2 -march=native -O1 -O0 40 80 120 160 200 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 SE +/- 0.23, N = 3 SE +/- 0.13, N = 3 SE +/- 0.00, N = 3 SE +/- 0.06, N = 3 SE +/- 0.95, N = 3 SE +/- 1.50, N = 6 SE +/- 1.17, N = 3 SE +/- 1.79, N = 4 SE +/- 1.37, N = 15 SE +/- 0.03, N = 3 178.33 178.21 170.58 169.13 168.88 168.76 160.88 155.16 154.29 153.44 149.44 36.89 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd 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 1.5.0 Compression Level: 8, Long Mode - Decompression Speed -O2 -Ofast -O3 -O2 -march=native -O2 -flto -O3 -march=native -Ofast -march=native -O3 -march=native -flto -O3 -flto -O1 -Og -O0 1200 2400 3600 4800 6000 SE +/- 5.12, N = 3 SE +/- 2.92, N = 3 SE +/- 10.95, N = 3 SE +/- 7.83, N = 3 SE +/- 11.17, N = 3 SE +/- 6.10, N = 15 SE +/- 10.09, N = 3 SE +/- 6.94, N = 12 SE +/- 13.35, N = 11 SE +/- 9.52, N = 3 SE +/- 6.26, N = 3 SE +/- 0.28, N = 3 5749.7 5734.4 5676.9 5649.1 5628.6 5542.9 5529.2 5478.3 5474.8 5385.7 4236.9 1206.4 -O2 -Ofast -O3 -O2 -march=native -O2 -flto -O3 -march=native -Ofast -march=native -O3 -march=native -flto -O3 -flto -O1 -Og -O0 1. (CC) gcc options: -pthread -lz
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. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.3.1 WAV To Opus Encode -O3 -march=native -flto -O3 -march=native -O2 -march=native -O3 -O3 -flto -O2 -O2 -flto -O1 -Og -O0 6 12 18 24 30 SE +/- 0.003, N = 5 SE +/- 0.010, N = 5 SE +/- 0.002, N = 5 SE +/- 0.006, N = 5 SE +/- 0.006, N = 5 SE +/- 0.004, N = 5 SE +/- 0.011, N = 5 SE +/- 0.004, N = 5 SE +/- 0.002, N = 5 SE +/- 0.015, N = 5 5.550 5.595 5.906 6.152 6.175 6.436 6.632 6.828 7.699 25.496 -O3 -march=native -flto -O3 -march=native -O2 -march=native -O3 -O3 -flto -O2 -O2 -flto -O1 -Og -O0 1. (CXX) g++ options: -fvisibility=hidden -logg -lm
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd 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 1.5.0 Compression Level: 19, Long Mode - Decompression Speed -O2 -Ofast -O3 -O2 -flto -O2 -march=native -O3 -flto -O3 -march=native -flto -Ofast -march=native -O3 -march=native -O1 -Og -O0 1000 2000 3000 4000 5000 SE +/- 13.81, N = 3 SE +/- 8.36, N = 3 SE +/- 2.68, N = 3 SE +/- 5.20, N = 3 SE +/- 6.86, N = 3 SE +/- 16.05, N = 3 SE +/- 22.61, N = 3 SE +/- 11.52, N = 3 SE +/- 15.31, N = 3 SE +/- 3.19, N = 3 SE +/- 8.30, N = 3 SE +/- 0.44, N = 5 4791.0 4725.4 4720.0 4693.2 4676.5 4602.9 4591.0 4589.4 4540.6 4506.0 3476.6 1066.8 -O2 -Ofast -O3 -O2 -flto -O2 -march=native -O3 -flto -O3 -march=native -flto -Ofast -march=native -O3 -march=native -O1 -Og -O0 1. (CC) gcc options: -pthread -lz
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 1.3.33 Operation: Sharpen -Ofast -march=native -O2 -march=native -O3 -march=native -flto -O3 -march=native -O3 -flto -O3 -O2 -O2 -flto -Ofast -O1 -Og -O0 50 100 150 200 250 SE +/- 0.67, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 217 197 195 195 164 164 164 163 162 162 87 50 -Ofast -march=native -O2 -march=native -O3 -march=native -flto -O3 -march=native -O3 -flto -O3 -O2 -O2 -flto -Ofast -O1 -Og -O0 1. (CC) gcc options: -fopenmp -pthread -ljpeg -lz -lm -lpthread
Crypto++ Crypto++ is a C++ class library of cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: All Algorithms -O3 -march=native -Ofast -march=native -Ofast -O3 -O3 -march=native -flto -O3 -flto -O2 -march=native -O2 -O2 -flto -O1 -Og -O0 500 1000 1500 2000 2500 SE +/- 1.51, N = 3 SE +/- 2.30, N = 3 SE +/- 0.43, N = 3 SE +/- 0.85, N = 3 SE +/- 0.90, N = 3 SE +/- 1.95, N = 3 SE +/- 0.95, N = 3 SE +/- 1.37, N = 3 SE +/- 1.13, N = 3 SE +/- 0.38, N = 3 SE +/- 1.15, N = 3 SE +/- 0.18, N = 3 2346.36 2330.25 2320.94 2316.41 2278.38 2269.04 2261.01 2245.55 2224.10 2114.62 1975.63 542.23 -O3 -march=native -Ofast -march=native -Ofast -O3 -O3 -march=native -flto -O3 -flto -O2 -march=native -O2 -O2 -flto -O1 -Og -O0 1. (CXX) g++ options: -fPIC -pthread -pipe
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 -Ofast -march=native -O3 -march=native -O3 -march=native -flto -Ofast -O3 -flto -O3 -O2 -march=native -O2 -O2 -flto -O1 -Og -O0 40 80 120 160 200 SE +/- 0.17, N = 3 SE +/- 0.15, N = 3 SE +/- 0.16, N = 3 SE +/- 0.16, N = 3 SE +/- 0.19, N = 3 SE +/- 0.19, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 46.14 47.34 47.46 63.75 66.35 66.47 84.73 106.49 106.60 128.91 130.02 191.79 -Ofast -march=native -march=native -march=native -flto -Ofast -flto -O2 -march=native -O2 -O2 -flto -O1 -Og -O0 1. (CC) gcc options: -lm -lpthread -O3
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 1.3.33 Operation: Enhanced -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O2 -march=native -Ofast -O2 -O2 -flto -O1 -O3 -flto -O3 -Og -O0 70 140 210 280 350 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 299 270 270 265 228 219 218 218 214 213 213 80 -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O2 -march=native -Ofast -O2 -O2 -flto -O1 -O3 -flto -O3 -Og -O0 1. (CC) gcc options: -fopenmp -pthread -ljpeg -lz -lm -lpthread
Smallpt Smallpt is a C++ global illumination renderer written in less than 100 lines of code. Global illumination is done via unbiased Monte Carlo path tracing and there is multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 128 Samples -O2 -march=native -Ofast -march=native -O3 -march=native -O3 -march=native -flto -Ofast -O2 -flto -O2 -O3 -O3 -flto -O1 -Og -O0 7 14 21 28 35 SE +/- 0.019, N = 3 SE +/- 0.006, N = 3 SE +/- 0.009, N = 3 SE +/- 0.008, N = 3 SE +/- 0.006, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.025, N = 3 SE +/- 0.015, N = 3 SE +/- 0.002, N = 3 SE +/- 0.020, N = 3 SE +/- 0.037, N = 3 7.886 8.290 8.401 8.457 8.590 8.703 8.770 9.029 9.088 9.133 10.690 29.094 -O2 -march=native -Ofast -march=native -march=native -march=native -flto -Ofast -O2 -flto -O2 -flto -O1 -Og -O0 1. (CXX) g++ options: -fopenmp -O3
Botan Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: KASUMI - Decrypt -O3 -flto -O3 -Ofast -O2 -O2 -flto -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O2 -march=native -O1 -Og -O0 30 60 90 120 150 SE +/- 0.09, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.59, N = 3 SE +/- 0.36, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.27, N = 3 SE +/- 0.06, N = 3 SE +/- 0.80, N = 3 SE +/- 0.07, N = 3 119.06 118.83 118.79 115.40 114.98 112.37 112.04 112.03 111.71 106.48 98.83 32.60 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: KASUMI -Ofast -O3 -flto -O3 -O2 -O2 -flto -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O2 -march=native -O1 -Og -O0 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.22, N = 3 SE +/- 0.03, N = 3 SE +/- 0.58, N = 3 SE +/- 0.02, N = 3 124.16 124.15 124.04 119.47 119.38 116.05 116.01 115.82 112.45 108.28 100.68 34.46 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
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 -O3 -march=native -flto -Ofast -march=native -O3 -march=native -O3 -flto -Ofast -O3 -O2 -march=native -O2 -flto -O2 -O1 -Og -O0 5 10 15 20 25 SE +/- 0.005, N = 3 SE +/- 0.006, N = 3 SE +/- 0.008, N = 3 SE +/- 0.008, N = 3 SE +/- 0.015, N = 3 SE +/- 0.019, N = 3 SE +/- 0.005, N = 3 SE +/- 0.008, N = 3 SE +/- 0.074, N = 3 SE +/- 0.092, N = 4 SE +/- 0.068, N = 3 SE +/- 0.039, N = 3 5.386 5.456 5.473 5.818 5.861 5.884 6.965 7.106 7.357 7.675 8.480 19.012 -march=native -flto -Ofast -march=native -march=native -flto -Ofast -O2 -march=native -O2 -flto -O2 -O1 -Og -O0 1. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lm
Crypto++ Crypto++ is a C++ class library of cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Integer + Elliptic Curve Public Key Algorithms -O3 -flto -O2 -flto -O3 -march=native -flto -O3 -march=native -O2 -march=native -O2 -Ofast -march=native -Ofast -O3 -O1 -Og -O0 1600 3200 4800 6400 8000 SE +/- 3.89, N = 3 SE +/- 4.94, N = 3 SE +/- 7.22, N = 3 SE +/- 1.75, N = 3 SE +/- 6.19, N = 3 SE +/- 5.85, N = 3 SE +/- 0.23, N = 3 SE +/- 3.58, N = 3 SE +/- 7.04, N = 3 SE +/- 4.50, N = 3 SE +/- 7.34, N = 3 SE +/- 0.91, N = 3 7270.92 7257.05 7252.64 7194.86 7173.76 7128.98 7097.81 7091.95 7052.85 6862.79 6555.53 2145.95 -O3 -flto -O2 -flto -O3 -march=native -flto -O3 -march=native -O2 -march=native -O2 -Ofast -march=native -Ofast -O3 -O1 -Og -O0 1. (CXX) g++ options: -fPIC -pthread -pipe
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Unkeyed Algorithms -O3 -Ofast -O2 -O3 -march=native -O3 -flto -Ofast -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -O1 -Og -O0 110 220 330 440 550 SE +/- 0.66, N = 3 SE +/- 0.54, N = 3 SE +/- 0.40, N = 3 SE +/- 0.05, N = 3 SE +/- 0.32, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.11, N = 3 SE +/- 0.67, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 496.42 495.40 492.24 491.45 490.79 490.61 490.39 484.45 475.00 472.95 450.93 150.67 -O3 -Ofast -O2 -O3 -march=native -O3 -flto -Ofast -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -O1 -Og -O0 1. (CXX) g++ options: -fPIC -pthread -pipe
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 1.3.33 Operation: Resizing -O3 -march=native -flto -O2 -march=native -Ofast -march=native -O3 -march=native -O2 -O2 -flto -O3 -flto -Ofast -O3 -O1 -Og -O0 300 600 900 1200 1500 SE +/- 1.86, N = 3 SE +/- 6.17, N = 3 SE +/- 1.00, N = 3 SE +/- 2.33, N = 3 SE +/- 1.86, N = 3 SE +/- 3.38, N = 3 SE +/- 1.45, N = 3 SE +/- 3.06, N = 3 SE +/- 1.76, N = 3 SE +/- 1.00, N = 3 SE +/- 1.86, N = 3 1231 1230 1228 1222 1092 1089 1086 1078 1078 1021 763 422 -O3 -march=native -flto -O2 -march=native -Ofast -march=native -O3 -march=native -O2 -O2 -flto -O3 -flto -Ofast -O3 -O1 -Og -O0 1. (CC) gcc options: -fopenmp -pthread -ljpeg -lz -lm -lpthread
eSpeak-NG Speech Engine This test times how long it takes the eSpeak speech synthesizer to read Project Gutenberg's The Outline of Science and output to a WAV file. This test profile is now tracking the eSpeak-NG version of eSpeak. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better eSpeak-NG Speech Engine 20200907 Text-To-Speech Synthesis -O2 -O3 -flto -O2 -flto -Ofast -march=native -O3 -march=native -O2 -march=native -O3 -march=native -flto -Ofast -O3 -O1 -Og -O0 13 26 39 52 65 SE +/- 0.03, N = 4 SE +/- 0.05, N = 4 SE +/- 0.06, N = 4 SE +/- 0.06, N = 4 SE +/- 0.06, N = 4 SE +/- 0.08, N = 4 SE +/- 0.07, N = 4 SE +/- 0.06, N = 4 SE +/- 0.07, N = 4 SE +/- 0.07, N = 4 SE +/- 0.08, N = 4 SE +/- 0.06, N = 4 21.41 21.60 21.66 21.75 21.77 22.23 22.66 22.77 22.93 24.00 26.89 59.01 -O2 -O3 -flto -O2 -flto -Ofast -march=native -O3 -march=native -O2 -march=native -O3 -march=native -flto -Ofast -O3 -O1 -Og -O0 1. (CC) gcc options: -std=c99 -lpthread -lm
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd 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 1.5.0 Compression Level: 3 - Compression Speed -O3 -march=native -O2 -march=native -O2 -flto -O3 -march=native -flto -O3 -flto -Ofast -march=native -O2 -O3 -Ofast -O1 -Og -O0 600 1200 1800 2400 3000 SE +/- 14.92, N = 3 SE +/- 5.47, N = 3 SE +/- 18.81, N = 3 SE +/- 3.46, N = 3 SE +/- 18.53, N = 3 SE +/- 6.82, N = 3 SE +/- 14.51, N = 3 SE +/- 8.13, N = 3 SE +/- 6.54, N = 3 SE +/- 8.18, N = 3 SE +/- 7.00, N = 3 SE +/- 2.56, N = 3 2731.5 2729.4 2723.2 2719.6 2712.7 2711.4 2707.6 2660.5 2658.3 2568.0 2376.3 1005.1 -O3 -march=native -O2 -march=native -O2 -flto -O3 -march=native -flto -O3 -flto -Ofast -march=native -O2 -O3 -Ofast -O1 -Og -O0 1. (CC) gcc options: -pthread -lz
Botan Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: ChaCha20Poly1305 - Decrypt -O2 -march=native -O3 -march=native -flto -O3 -march=native -Ofast -march=native -O1 -O3 -flto -Og -Ofast -O3 -O2 -flto -O2 -O0 200 400 600 800 1000 SE +/- 1.76, N = 6 SE +/- 0.91, N = 6 SE +/- 0.23, N = 3 SE +/- 1.51, N = 6 SE +/- 1.73, N = 3 SE +/- 3.37, N = 6 SE +/- 1.20, N = 6 SE +/- 2.42, N = 6 SE +/- 3.00, N = 6 SE +/- 3.44, N = 6 SE +/- 3.49, N = 6 SE +/- 0.13, N = 6 1080.31 1016.89 1010.79 1008.85 1004.65 969.70 969.30 962.28 960.08 907.74 901.57 76.49 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Blowfish -O2 -flto -O2 -Ofast -march=native -O2 -march=native -O3 -O3 -march=native -flto -Ofast -O3 -march=native -O3 -flto -O1 -Og -O0 120 240 360 480 600 SE +/- 0.11, N = 3 SE +/- 0.19, N = 3 SE +/- 0.24, N = 3 SE +/- 0.23, N = 3 SE +/- 0.23, N = 3 SE +/- 0.13, N = 3 SE +/- 0.35, N = 3 SE +/- 0.20, N = 3 SE +/- 2.26, N = 3 SE +/- 0.93, N = 3 SE +/- 0.55, N = 3 SE +/- 0.13, N = 3 563.75 562.85 554.57 554.07 553.53 553.17 552.87 552.46 551.74 533.96 443.74 47.15 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Blowfish - Decrypt -O3 -march=native -flto -Ofast -march=native -O3 -march=native -O2 -flto -O2 -O3 -O3 -flto -Ofast -O2 -march=native -O1 -Og -O0 120 240 360 480 600 SE +/- 0.47, N = 3 SE +/- 0.18, N = 3 SE +/- 0.26, N = 3 SE +/- 0.21, N = 3 SE +/- 0.27, N = 3 SE +/- 0.20, N = 3 SE +/- 2.20, N = 3 SE +/- 0.54, N = 3 SE +/- 2.45, N = 3 SE +/- 1.04, N = 3 SE +/- 2.60, N = 3 SE +/- 0.14, N = 3 555.06 553.97 553.52 550.13 549.17 540.31 538.67 538.43 533.32 532.56 441.08 46.59 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Twofish -O2 -march=native -O3 -march=native -Ofast -march=native -O3 -march=native -flto -Ofast -O3 -O3 -flto -O2 -O2 -flto -O1 -Og -O0 100 200 300 400 500 SE +/- 0.08, N = 3 SE +/- 0.31, N = 3 SE +/- 1.26, N = 3 SE +/- 1.36, N = 3 SE +/- 0.29, N = 3 SE +/- 0.19, N = 3 SE +/- 0.23, N = 3 SE +/- 0.75, N = 3 SE +/- 0.06, N = 3 SE +/- 0.19, N = 3 SE +/- 1.58, N = 3 SE +/- 0.26, N = 3 470.94 464.47 463.94 463.01 450.00 449.44 448.62 448.57 448.40 430.95 384.86 43.45 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Twofish - Decrypt -O2 -march=native -O3 -march=native -flto -Ofast -O3 -O3 -march=native -O3 -flto -Ofast -march=native -O2 -O2 -flto -O1 -Og -O0 100 200 300 400 500 SE +/- 0.96, N = 3 SE +/- 0.11, N = 3 SE +/- 0.65, N = 3 SE +/- 0.22, N = 3 SE +/- 0.62, N = 3 SE +/- 0.25, N = 3 SE +/- 1.74, N = 3 SE +/- 0.35, N = 3 SE +/- 0.16, N = 3 SE +/- 0.13, N = 3 SE +/- 0.43, N = 3 SE +/- 0.10, N = 3 470.03 454.15 453.84 451.78 451.66 451.24 449.76 449.21 445.67 427.26 387.64 43.79 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
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 1.3.33 Operation: Swirl -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O2 -march=native -Ofast -O3 -O2 -O2 -flto -O3 -flto -O1 -Og -O0 160 320 480 640 800 SE +/- 3.28, N = 3 SE +/- 1.73, N = 3 SE +/- 2.67, N = 3 SE +/- 2.00, N = 3 SE +/- 1.76, N = 3 SE +/- 2.08, N = 3 SE +/- 1.67, N = 3 SE +/- 1.15, N = 3 SE +/- 0.58, N = 3 SE +/- 1.00, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 735 693 689 682 659 625 622 617 612 592 581 378 -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O2 -march=native -Ofast -O3 -O2 -O2 -flto -O3 -flto -O1 -Og -O0 1. (CC) gcc options: -fopenmp -pthread -ljpeg -lz -lm -lpthread
AOBench AOBench is a lightweight ambient occlusion renderer, written in C. The test profile is using a size of 2048 x 2048. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O2 -march=native -Ofast -O3 -flto -O3 -O2 -flto -O2 -O1 -Og -O0 9 18 27 36 45 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.17, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.27, N = 4 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 SE +/- 0.39, N = 3 21.54 21.54 21.56 22.65 22.99 23.37 23.69 24.27 24.43 24.61 34.36 40.88 -Ofast -march=native -march=native -flto -march=native -O2 -march=native -Ofast -flto -O2 -flto -O2 -O1 -Og -O0 1. (CC) gcc options: -lm -O3
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 1.3.33 Operation: Noise-Gaussian -Ofast -march=native -Ofast -O3 -march=native -flto -O2 -march=native -O3 -march=native -O1 -O3 -flto -O2 -O2 -flto -O3 -Og -O0 80 160 240 320 400 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 352 348 314 312 310 306 305 304 303 301 281 188 -Ofast -march=native -Ofast -O3 -march=native -flto -O2 -march=native -O3 -march=native -O1 -O3 -flto -O2 -O2 -flto -O3 -Og -O0 1. (CC) gcc options: -fopenmp -pthread -ljpeg -lz -lm -lpthread
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: shufflenet-v2 -O2 -march=native -Ofast -march=native -O3 -march=native -O1 -Og -O0 -O3 -Ofast -O2 -O3 -march=native -flto -O2 -flto -O3 -flto 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 3.25 3.25 3.26 3.45 3.45 3.46 3.46 3.47 3.48 5.68 5.97 6.08 -march=native - MIN: 3.19 / MAX: 6.89 -Ofast -march=native - MIN: 3.17 / MAX: 6.87 -O3 -march=native - MIN: 3.18 / MAX: 6.94 -O1 - MIN: 3.39 / MAX: 7.07 -Og - MIN: 3.39 / MAX: 7.06 -O0 - MIN: 3.4 / MAX: 7.21 -O3 - MIN: 3.42 / MAX: 6.99 -Ofast - MIN: 3.39 / MAX: 7.12 MIN: 3.38 / MAX: 7.1 -O3 -march=native -flto - MIN: 5.56 / MAX: 9.36 -flto - MIN: 5.83 / MAX: 9.65 -O3 -flto - MIN: 5.91 / MAX: 9.83 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
SQLite Speedtest This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better SQLite Speedtest 3.30 Timed Time - Size 1,000 -O3 -march=native -O2 -march=native -O2 -flto -O2 -O3 -O3 -flto -Ofast -march=native -O3 -march=native -flto -Ofast -O1 -Og -O0 20 40 60 80 100 SE +/- 0.15, N = 3 SE +/- 0.03, N = 3 SE +/- 0.13, N = 3 SE +/- 0.24, N = 3 SE +/- 0.10, N = 3 SE +/- 0.14, N = 3 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.26, N = 3 SE +/- 0.11, N = 3 SE +/- 0.24, N = 3 46.09 46.09 46.23 46.31 46.35 46.54 46.59 46.92 47.00 49.01 52.67 80.89 -O3 -march=native -O2 -march=native -O2 -flto -O2 -O3 -O3 -flto -Ofast -march=native -O3 -march=native -flto -Ofast -O1 -Og -O0 1. (CC) gcc options: -ldl -lz -lpthread
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: blazeface -O2 -march=native -O3 -march=native -Ofast -march=native -Og -O0 -Ofast -O1 -O3 -O2 -O3 -flto -O3 -march=native -flto -O2 -flto 0.4073 0.8146 1.2219 1.6292 2.0365 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 1.10 1.15 1.16 1.18 1.19 1.22 1.24 1.24 1.28 1.80 1.80 1.81 -march=native - MIN: 1.08 / MAX: 4.83 -O3 -march=native - MIN: 1.08 / MAX: 2 -Ofast -march=native - MIN: 1.08 / MAX: 4.68 -Og - MIN: 1.14 / MAX: 1.94 -O0 - MIN: 1.14 / MAX: 4.76 -Ofast - MIN: 1.15 / MAX: 1.3 -O1 - MIN: 1.21 / MAX: 5.59 -O3 - MIN: 1.21 / MAX: 4.79 MIN: 1.22 / MAX: 4.92 -O3 -flto - MIN: 1.74 / MAX: 1.88 -O3 -march=native -flto - MIN: 1.72 / MAX: 2.56 -flto - MIN: 1.77 / MAX: 1.94 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd 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 1.5.0 Compression Level: 19, Long Mode - Compression Speed -O3 -Ofast -Ofast -march=native -O3 -march=native -flto -O2 -march=native -O2 -O3 -flto -O2 -flto -Og -O0 8 16 24 32 40 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.13, N = 3 SE +/- 0.06, N = 3 SE +/- 0.17, N = 3 SE +/- 0.15, N = 3 SE +/- 0.09, N = 3 SE +/- 0.15, N = 3 SE +/- 0.03, N = 3 SE +/- 0.21, N = 5 33.2 33.1 33.1 32.9 32.9 32.8 32.7 32.5 31.9 20.2 -O3 -Ofast -Ofast -march=native -O3 -march=native -flto -O2 -march=native -O2 -O3 -flto -O2 -flto -Og -O0 1. (CC) gcc options: -pthread -lz
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mnasnet -O3 -march=native -O2 -march=native -Ofast -march=native -O3 -march=native -flto -O0 -Og -Ofast -O1 -O3 -O2 -O2 -flto -O3 -flto 0.765 1.53 2.295 3.06 3.825 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 2.22 2.23 2.25 2.28 3.09 3.11 3.14 3.17 3.18 3.21 3.39 3.40 -O3 -march=native - MIN: 2.17 / MAX: 2.35 -march=native - MIN: 2.19 / MAX: 5.93 -Ofast -march=native - MIN: 2.17 / MAX: 2.77 -O3 -march=native -flto - MIN: 2.22 / MAX: 5.98 -O0 - MIN: 3.05 / MAX: 6.68 -Og - MIN: 3.06 / MAX: 6.78 -Ofast - MIN: 3.06 / MAX: 6.77 -O1 - MIN: 3.14 / MAX: 6.8 -O3 - MIN: 3.09 / MAX: 6.86 MIN: 3.08 / MAX: 6.93 -flto - MIN: 3.29 / MAX: 6.96 -O3 -flto - MIN: 3.24 / MAX: 7.1 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
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 1.3.33 Operation: HWB Color Space -O3 -march=native -flto -O2 -flto -O3 -flto -O3 -march=native -Ofast -march=native -Ofast -O2 -march=native -O3 -O2 -Og -O1 -O0 300 600 900 1200 1500 SE +/- 1.53, N = 3 SE +/- 0.67, N = 3 SE +/- 1.20, N = 3 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 SE +/- 1.53, N = 3 SE +/- 0.33, N = 3 SE +/- 1.33, N = 3 SE +/- 0.33, N = 3 1349 1345 1343 1285 1275 1267 1245 1239 1235 1207 1207 881 -O3 -march=native -flto -O2 -flto -O3 -flto -O3 -march=native -Ofast -march=native -Ofast -O2 -march=native -O3 -O2 -Og -O1 -O0 1. (CC) gcc options: -fopenmp -pthread -ljpeg -lz -lm -lpthread
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mobilenet -O3 -march=native -O2 -march=native -Ofast -march=native -O3 -march=native -flto -Og -O2 -O1 -O0 -Ofast -O3 -O3 -flto -O2 -flto 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.16, N = 3 SE +/- 0.10, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.06, N = 3 SE +/- 0.13, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 11.76 11.88 11.92 13.66 14.92 15.01 15.02 15.06 15.08 15.12 16.92 16.96 -O3 -march=native - MIN: 11.54 / MAX: 15.41 -march=native - MIN: 11.77 / MAX: 15.49 -Ofast -march=native - MIN: 11.81 / MAX: 15.57 -O3 -march=native -flto - MIN: 13.25 / MAX: 17.52 -Og - MIN: 14.6 / MAX: 18.67 MIN: 14.87 / MAX: 18.64 -O1 - MIN: 14.88 / MAX: 18.66 -O0 - MIN: 14.84 / MAX: 18.7 -Ofast - MIN: 14.77 / MAX: 18.87 -O3 - MIN: 14.93 / MAX: 18.84 -O3 -flto - MIN: 16.64 / MAX: 24.79 -flto - MIN: 16.74 / MAX: 20.55 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v2-v2 - Model: mobilenet-v2 -O3 -march=native -O2 -march=native -Ofast -march=native -O3 -march=native -flto -O0 -O1 -Og -Ofast -O3 -O2 -O2 -flto -O3 -flto 1.0035 2.007 3.0105 4.014 5.0175 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 3.21 3.22 3.24 3.26 4.18 4.19 4.20 4.20 4.25 4.36 4.46 4.46 -O3 -march=native - MIN: 3.08 / MAX: 4.11 -march=native - MIN: 3.1 / MAX: 6.96 -Ofast -march=native - MIN: 3.13 / MAX: 4.21 -O3 -march=native -flto - MIN: 3.14 / MAX: 6.82 -O0 - MIN: 4.03 / MAX: 7.87 -O1 - MIN: 4.06 / MAX: 7.81 -Og - MIN: 4.06 / MAX: 10.02 -Ofast - MIN: 4.07 / MAX: 7.83 -O3 - MIN: 4.06 / MAX: 7.77 MIN: 4.12 / MAX: 8.14 -flto - MIN: 4.28 / MAX: 8.19 -O3 -flto - MIN: 4.27 / MAX: 8.14 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v3-v3 - Model: mobilenet-v3 -O3 -march=native -O2 -march=native -Ofast -march=native -O3 -march=native -flto -O0 -Og -O1 -Ofast -O3 -O2 -O3 -flto -O2 -flto 0.774 1.548 2.322 3.096 3.87 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 2.49 2.50 2.51 2.54 3.17 3.17 3.19 3.21 3.24 3.25 3.41 3.44 -O3 -march=native - MIN: 2.44 / MAX: 6.14 -march=native - MIN: 2.46 / MAX: 6.05 -Ofast -march=native - MIN: 2.44 / MAX: 6.19 -O3 -march=native -flto - MIN: 2.47 / MAX: 6.15 -O0 - MIN: 3.11 / MAX: 6.83 -Og - MIN: 3.12 / MAX: 5.46 -O1 - MIN: 3.16 / MAX: 4.05 -Ofast - MIN: 3.14 / MAX: 4.03 -O3 - MIN: 3.18 / MAX: 6.79 MIN: 3.16 / MAX: 6.87 -O3 -flto - MIN: 3.33 / MAX: 4.14 -flto - MIN: 3.38 / MAX: 7.08 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
dav1d Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better dav1d 0.9.0 Video Input: Summer Nature 4K -O3 -march=native -O2 -O1 -O2 -march=native -Og -O3 -Ofast -Ofast -march=native -O0 40 80 120 160 200 SE +/- 0.19, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.13, N = 3 SE +/- 0.11, N = 3 SE +/- 0.18, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 195.94 187.10 185.95 185.36 184.01 181.31 180.05 179.95 146.06 -O3 -march=native - MIN: 181.35 / MAX: 208.71 -O2 - MIN: 172.85 / MAX: 197.02 -O1 - MIN: 169.98 / MAX: 195.75 -O2 -march=native - MIN: 170.59 / MAX: 194.09 -Og - MIN: 171.24 / MAX: 196.19 -O3 - MIN: 165.68 / MAX: 190.05 -Ofast - MIN: 163.86 / MAX: 187.94 -Ofast -march=native - MIN: 165.19 / MAX: 188.36 -O0 - MIN: 135.09 / MAX: 168.72 1. (CC) gcc options: -pthread -lm
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: efficientnet-b0 -O3 -march=native -O2 -march=native -O3 -march=native -flto -Ofast -march=native -O0 -Og -O1 -O3 -Ofast -O2 -O2 -flto -O3 -flto 1.2465 2.493 3.7395 4.986 6.2325 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 4.24 4.27 4.31 4.33 5.23 5.23 5.24 5.30 5.32 5.36 5.54 5.54 -O3 -march=native - MIN: 4.19 / MAX: 7.9 -march=native - MIN: 4.2 / MAX: 7.94 -O3 -march=native -flto - MIN: 4.25 / MAX: 7.95 -Ofast -march=native - MIN: 4.21 / MAX: 7.88 -O0 - MIN: 5.15 / MAX: 8.89 -Og - MIN: 5.17 / MAX: 8.95 -O1 - MIN: 5.17 / MAX: 8.84 -O3 - MIN: 5.18 / MAX: 9.09 -Ofast - MIN: 5.19 / MAX: 11.06 MIN: 5.18 / MAX: 9.93 -flto - MIN: 5.45 / MAX: 9.2 -O3 -flto - MIN: 5.41 / MAX: 9.33 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd 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 1.5.0 Compression Level: 8, Long Mode - Compression Speed -Og -O3 -O3 -flto -Ofast -O2 -flto -O2 -O3 -march=native -flto -O3 -march=native -O1 -O2 -march=native -Ofast -march=native -O0 70 140 210 280 350 SE +/- 1.92, N = 3 SE +/- 3.28, N = 3 SE +/- 2.23, N = 11 SE +/- 2.55, N = 3 SE +/- 2.74, N = 3 SE +/- 1.34, N = 3 SE +/- 2.00, N = 12 SE +/- 2.25, N = 15 SE +/- 2.78, N = 3 SE +/- 2.77, N = 3 SE +/- 3.28, N = 3 SE +/- 1.88, N = 3 316.6 308.4 306.7 300.7 299.1 298.9 287.7 285.9 281.5 277.0 276.5 244.1 -Og -O3 -O3 -flto -Ofast -O2 -flto -O2 -O3 -march=native -flto -O3 -march=native -O1 -O2 -march=native -Ofast -march=native -O0 1. (CC) gcc options: -pthread -lz
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet50 -O3 -march=native -Ofast -march=native -O2 -march=native -O3 -march=native -flto -Og -O1 -Ofast -O2 -O0 -O3 -O3 -flto -O2 -flto 6 12 18 24 30 SE +/- 0.15, N = 3 SE +/- 0.19, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.23, N = 3 SE +/- 0.03, N = 3 SE +/- 0.24, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 18.23 18.39 18.51 18.75 22.12 22.29 22.29 22.34 22.36 22.41 23.39 23.42 -O3 -march=native - MIN: 17.79 / MAX: 22.11 -Ofast -march=native - MIN: 17.9 / MAX: 22.23 -march=native - MIN: 18.3 / MAX: 24.03 -O3 -march=native -flto - MIN: 18.58 / MAX: 22.39 -Og - MIN: 21.48 / MAX: 26.11 -O1 - MIN: 22.02 / MAX: 27 -Ofast - MIN: 21.67 / MAX: 26.11 MIN: 22.12 / MAX: 26.1 -O0 - MIN: 22.1 / MAX: 27.7 -O3 - MIN: 22.18 / MAX: 26.09 -O3 -flto - MIN: 23.2 / MAX: 27.18 -flto - MIN: 23.16 / MAX: 28.06 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: yolov4-tiny -O3 -march=native -O2 -march=native -Ofast -march=native -Og -O2 -O1 -Ofast -O0 -O3 -O3 -march=native -flto -O3 -flto -O2 -flto 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.16, N = 3 SE +/- 0.28, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.36, N = 3 SE +/- 0.14, N = 3 SE +/- 0.13, N = 3 SE +/- 0.12, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 20.21 20.32 20.66 20.96 21.24 21.26 21.27 21.37 21.41 23.82 24.50 24.58 -O3 -march=native - MIN: 20.03 / MAX: 23.86 -march=native - MIN: 20.15 / MAX: 24.06 -Ofast -march=native - MIN: 20.3 / MAX: 24.56 -Og - MIN: 20.29 / MAX: 25.35 MIN: 21.01 / MAX: 26.9 -O1 - MIN: 20.97 / MAX: 27.08 -Ofast - MIN: 20.47 / MAX: 26.5 -O0 - MIN: 21.06 / MAX: 25.18 -O3 - MIN: 21.1 / MAX: 25.22 -O3 -march=native -flto - MIN: 23.43 / MAX: 28.5 -O3 -flto - MIN: 24.27 / MAX: 28.19 -flto - MIN: 24.33 / MAX: 28.27 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: regnety_400m -O3 -march=native -Ofast -march=native -O2 -march=native -O3 -march=native -flto -O0 -O2 -Og -O3 -Ofast -O1 -O2 -flto -O3 -flto 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 2 SE +/- 0.06, N = 3 SE +/- 0.14, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 8.57 8.60 8.75 8.92 9.59 9.64 9.65 9.71 9.72 9.73 10.27 10.34 -O3 -march=native - MIN: 8.47 / MAX: 12.35 -Ofast -march=native - MIN: 8.43 / MAX: 12.94 -march=native - MIN: 8.52 / MAX: 15 -O3 -march=native -flto - MIN: 8.75 / MAX: 12.62 -O0 - MIN: 9.46 / MAX: 13.26 MIN: 9.46 / MAX: 13.35 -Og - MIN: 9.51 / MAX: 15.51 -O3 - MIN: 9.49 / MAX: 13.42 -Ofast - MIN: 9.38 / MAX: 13.46 -O1 - MIN: 9.55 / MAX: 14.41 -flto - MIN: 10.13 / MAX: 13.94 -O3 -flto - MIN: 10.1 / MAX: 17.97 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
PJSIP PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: OPTIONS, Stateless -O3 -flto -O3 -march=native -flto -O2 -flto -O3 -march=native -O3 -O2 -Ofast -march=native -Ofast -O2 -march=native -Og -O1 -O0 60K 120K 180K 240K 300K SE +/- 392.40, N = 3 SE +/- 381.59, N = 3 SE +/- 608.82, N = 3 SE +/- 711.03, N = 3 SE +/- 231.12, N = 3 SE +/- 97.41, N = 3 SE +/- 730.96, N = 3 SE +/- 590.54, N = 3 SE +/- 574.52, N = 3 SE +/- 1082.91, N = 3 SE +/- 520.47, N = 3 SE +/- 540.11, N = 3 256859 255381 255114 254610 253887 253466 252869 252647 252495 247454 247106 215182 -O3 -flto -O3 -march=native -flto -O2 -flto -O3 -march=native -O3 -O2 -Ofast -march=native -Ofast -O2 -march=native -Og -O1 -O0 1. (CC) gcc options: -lstdc++ -lssl -lcrypto -lm -lrt -lpthread
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: googlenet -O3 -march=native -Ofast -march=native -O2 -march=native -O3 -march=native -flto -Og -O0 -O1 -Ofast -O2 -O3 -O3 -flto -O2 -flto 3 6 9 12 15 SE +/- 0.17, N = 3 SE +/- 0.19, N = 3 SE +/- 0.00, N = 3 SE +/- 0.06, N = 3 SE +/- 0.17, N = 3 SE +/- 0.18, N = 3 SE +/- 0.02, N = 3 SE +/- 0.23, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.25, N = 3 SE +/- 0.07, N = 3 10.09 10.15 10.39 10.64 11.04 11.07 11.40 11.43 11.51 11.52 11.83 12.01 -O3 -march=native - MIN: 9.67 / MAX: 13.94 -Ofast -march=native - MIN: 9.84 / MAX: 15.83 -march=native - MIN: 10.3 / MAX: 14.02 -O3 -march=native -flto - MIN: 10.48 / MAX: 15.45 -Og - MIN: 10.78 / MAX: 14.99 -O0 - MIN: 10.74 / MAX: 15.65 -O1 - MIN: 11.29 / MAX: 14.99 -Ofast - MIN: 10.92 / MAX: 17.47 MIN: 11.34 / MAX: 15.19 -O3 - MIN: 11.38 / MAX: 15.27 -O3 -flto - MIN: 11.25 / MAX: 15.72 -flto - MIN: 11.84 / MAX: 15.72 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
TNN TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: MobileNet v2 -O3 -march=native -O2 -march=native -Ofast -march=native -O0 -O1 -Ofast -O3 -Og -O2 -O3 -march=native -flto -O3 -flto -O2 -flto 50 100 150 200 250 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.21, N = 3 SE +/- 0.13, N = 3 SE +/- 0.20, N = 3 SE +/- 0.17, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.15, N = 3 SE +/- 0.28, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 230.11 230.22 230.49 243.10 243.16 243.29 243.31 243.33 243.41 248.41 251.49 251.61 -O3 -march=native - MIN: 229.52 / MAX: 232.81 -march=native - MIN: 229.36 / MAX: 233.08 -Ofast -march=native - MIN: 229.69 / MAX: 233.56 -O0 - MIN: 241.4 / MAX: 244.76 -O1 - MIN: 241.63 / MAX: 246.21 -Ofast - MIN: 241.45 / MAX: 246.15 -O3 - MIN: 241.88 / MAX: 245.9 -Og - MIN: 241.41 / MAX: 252.51 MIN: 241.93 / MAX: 245.01 -O3 -march=native -flto - MIN: 247.34 / MAX: 252.91 -O3 -flto - MIN: 250.87 / MAX: 254.61 -flto - MIN: 251.13 / MAX: 254.92 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O2 -rdynamic -ldl
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: squeezenet_ssd -O3 -march=native -O2 -march=native -Ofast -march=native -O3 -march=native -flto -Og -O1 -O0 -O2 -O3 -Ofast -O2 -flto -O3 -flto 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.13, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 15.29 15.46 15.65 15.89 16.16 16.18 16.18 16.25 16.33 16.34 16.63 16.66 -O3 -march=native - MIN: 15.14 / MAX: 19 -march=native - MIN: 15.31 / MAX: 19.14 -Ofast -march=native - MIN: 15.35 / MAX: 19.25 -O3 -march=native -flto - MIN: 15.72 / MAX: 21.7 -Og - MIN: 15.99 / MAX: 20.07 -O1 - MIN: 16.02 / MAX: 19.89 -O0 - MIN: 16.05 / MAX: 19.86 MIN: 16.07 / MAX: 24.06 -O3 - MIN: 16.15 / MAX: 21.16 -Ofast - MIN: 16.12 / MAX: 20.09 -flto - MIN: 16.51 / MAX: 20.21 -O3 -flto - MIN: 16.52 / MAX: 20.45 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
LAMMPS Molecular Dynamics Simulator LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: Rhodopsin Protein -O3 -march=native -flto -O3 -march=native -Ofast -march=native -O2 -march=native -O1 -O3 -Og -O3 -flto -O0 -Ofast -O2 -O2 -flto 2 4 6 8 10 SE +/- 0.030, N = 3 SE +/- 0.026, N = 3 SE +/- 0.063, N = 3 SE +/- 0.053, N = 3 SE +/- 0.028, N = 3 SE +/- 0.048, N = 3 SE +/- 0.034, N = 3 SE +/- 0.034, N = 3 SE +/- 0.112, N = 3 SE +/- 0.103, N = 12 SE +/- 0.124, N = 12 SE +/- 0.071, N = 3 8.619 8.513 8.510 8.499 8.184 8.135 8.121 8.109 8.093 8.032 8.012 8.001 -O3 -march=native -flto -O3 -march=native -Ofast -march=native -march=native -O1 -O3 -Og -O3 -flto -O0 -Ofast -flto 1. (CXX) g++ options: -O2 -pthread -lm
TNN TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: SqueezeNet v1.1 -Ofast -march=native -O3 -march=native -O2 -march=native -Og -O1 -Ofast -O2 -O3 -O0 -O3 -march=native -flto -O3 -flto -O2 -flto 50 100 150 200 250 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.14, N = 3 SE +/- 0.15, N = 3 SE +/- 0.22, N = 3 SE +/- 0.27, N = 3 SE +/- 0.13, N = 3 SE +/- 0.37, N = 3 SE +/- 0.12, N = 3 SE +/- 0.19, N = 3 SE +/- 0.16, N = 3 227.41 227.46 227.59 235.77 235.96 235.99 236.33 236.43 236.47 242.69 244.59 244.63 -Ofast -march=native - MIN: 226.48 / MAX: 228.67 -O3 -march=native - MIN: 226.88 / MAX: 228.23 -march=native - MIN: 226.72 / MAX: 229.36 -Og - MIN: 234.57 / MAX: 237.7 -O1 - MIN: 234.76 / MAX: 237.84 -Ofast - MIN: 234.48 / MAX: 237.48 MIN: 234.45 / MAX: 237.44 -O3 - MIN: 234.65 / MAX: 237.59 -O0 - MIN: 234.7 / MAX: 238.18 -O3 -march=native -flto - MIN: 242.07 / MAX: 243.46 -O3 -flto - MIN: 243.8 / MAX: 245.52 -flto - MIN: 243.9 / MAX: 245.59 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O2 -rdynamic -ldl
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd 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 1.5.0 Compression Level: 8 - Compression Speed -O3 -flto -O2 -flto -O3 -march=native -flto -O3 -march=native -O3 -O2 -Og -O2 -march=native -Ofast -march=native -O1 -Ofast -O0 40 80 120 160 200 SE +/- 0.38, N = 3 SE +/- 0.38, N = 3 SE +/- 0.30, N = 3 SE +/- 0.90, N = 3 SE +/- 0.84, N = 3 SE +/- 0.75, N = 3 SE +/- 0.31, N = 3 SE +/- 0.82, N = 3 SE +/- 0.89, N = 3 SE +/- 0.57, N = 3 SE +/- 0.78, N = 3 SE +/- 0.79, N = 3 194.3 194.0 192.7 192.6 192.5 191.8 190.8 190.5 189.8 189.2 188.6 180.9 -O3 -flto -O2 -flto -O3 -march=native -flto -O3 -march=native -O3 -O2 -Og -O2 -march=native -Ofast -march=native -O1 -Ofast -O0 1. (CC) gcc options: -pthread -lz
PJSIP PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: INVITE -O2 -flto -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O3 -flto -Ofast -O2 -march=native -O1 -O2 -O3 -Og -O0 1100 2200 3300 4400 5500 SE +/- 33.42, N = 3 SE +/- 27.45, N = 3 SE +/- 22.84, N = 3 SE +/- 15.24, N = 3 SE +/- 5.04, N = 3 SE +/- 52.72, N = 3 SE +/- 43.32, N = 3 SE +/- 45.51, N = 3 SE +/- 68.29, N = 3 SE +/- 43.88, N = 3 SE +/- 65.58, N = 3 SE +/- 51.86, N = 4 5114 5089 5061 5060 5049 5035 5009 4993 4962 4953 4884 4816 -O2 -flto -Ofast -march=native -O3 -march=native -flto -O3 -march=native -O3 -flto -Ofast -O2 -march=native -O1 -O2 -O3 -Og -O0 1. (CC) gcc options: -lstdc++ -lssl -lcrypto -lm -lrt -lpthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU -Og -O1 -O0 -O3 -march=native -O3 -flto -Ofast -march=native -O3 -march=native -flto -O2 -flto -O2 -Ofast -O3 -O2 -march=native 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 10.59 11.03 11.07 11.20 11.22 11.22 11.23 11.23 11.24 11.24 11.24 11.24 -Og - MIN: 10.49 -O1 - MIN: 10.93 -O0 - MIN: 10.97 MIN: 11.11 -flto - MIN: 11.13 -Ofast - MIN: 11.13 -flto - MIN: 11.13 -flto - MIN: 11.13 MIN: 11.13 -Ofast - MIN: 11.14 MIN: 11.14 MIN: 11.13 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl
SVT-HEVC This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 1 - Input: Bosphorus 1080p -O3 -march=native -flto -O3 -flto -O2 -flto -O3 -march=native -O2 -march=native -Ofast -march=native -O2 -Og -Ofast -O0 -O1 -O3 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 9.74 9.51 9.51 9.48 9.47 9.46 9.21 9.21 9.20 9.20 9.20 9.19 -march=native -flto -flto -flto -march=native -march=native -Ofast -march=native -Og -Ofast -O0 -O1 1. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt
libjpeg-turbo tjbench tjbench is a JPEG decompression/compression benchmark that is part of libjpeg-turbo, a JPEG image codec library optimized for SIMD instructions on modern CPU architectures. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 2.1.0 Test: Decompression Throughput -Ofast -march=native -O2 -march=native -O3 -march=native -flto -O3 -march=native -O2 -Ofast -Og -O2 -flto -O1 -O3 -flto -O0 -O3 60 120 180 240 300 SE +/- 0.14, N = 3 SE +/- 0.41, N = 3 SE +/- 1.04, N = 3 SE +/- 0.45, N = 3 SE +/- 1.18, N = 3 SE +/- 0.59, N = 3 SE +/- 0.43, N = 3 SE +/- 0.51, N = 3 SE +/- 0.31, N = 3 SE +/- 0.65, N = 3 SE +/- 0.67, N = 3 SE +/- 1.16, N = 3 273.25 272.81 271.75 271.68 261.22 261.00 260.63 260.30 260.26 260.03 259.49 257.85 -Ofast -march=native -O2 -march=native -march=native -flto -lm -march=native -O2 -Ofast -Og -O2 -flto -lm -O1 -flto -lm -O0 1. (CC) gcc options: -O3 -rdynamic
LAMMPS Molecular Dynamics Simulator LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: 20k Atoms -O3 -march=native -flto -O2 -march=native -Ofast -march=native -O3 -march=native -O0 -Og -Ofast -O1 -O3 -flto -O2 -O3 -O2 -flto 2 4 6 8 10 SE +/- 0.006, N = 3 SE +/- 0.014, N = 3 SE +/- 0.011, N = 3 SE +/- 0.020, N = 3 SE +/- 0.005, N = 3 SE +/- 0.022, N = 3 SE +/- 0.010, N = 3 SE +/- 0.035, N = 3 SE +/- 0.026, N = 3 SE +/- 0.010, N = 3 SE +/- 0.016, N = 3 SE +/- 0.077, N = 9 8.757 8.755 8.739 8.737 8.368 8.356 8.348 8.345 8.335 8.314 8.295 8.267 -O3 -march=native -flto -march=native -Ofast -march=native -O3 -march=native -O0 -Og -Ofast -O1 -O3 -flto -O3 -flto 1. (CXX) g++ options: -O2 -pthread -lm
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet18 -O3 -march=native -Ofast -march=native -O2 -march=native -O0 -Og -Ofast -O3 -march=native -flto -O1 -O2 -O3 -O3 -flto -O2 -flto 3 6 9 12 15 SE +/- 0.14, N = 3 SE +/- 0.16, N = 3 SE +/- 0.02, N = 3 SE +/- 0.19, N = 3 SE +/- 0.20, N = 3 SE +/- 0.18, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 2 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 11.08 11.16 11.29 11.30 11.33 11.37 11.45 11.47 11.48 11.51 11.65 11.67 -O3 -march=native - MIN: 10.69 / MAX: 16.91 -Ofast -march=native - MIN: 10.75 / MAX: 14.91 -march=native - MIN: 11.17 / MAX: 14.94 -O0 - MIN: 10.85 / MAX: 15.09 -Og - MIN: 10.84 / MAX: 17.14 -Ofast - MIN: 10.94 / MAX: 15.16 -O3 -march=native -flto - MIN: 11.35 / MAX: 15.04 -O1 - MIN: 11.34 / MAX: 15.37 MIN: 11.37 / MAX: 15.06 -O3 - MIN: 11.4 / MAX: 16.95 -O3 -flto - MIN: 11.53 / MAX: 15.24 -flto - MIN: 11.56 / MAX: 15.3 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 3.0 Preset: Thorough -O3 -march=native -O2 -march=native -Ofast -march=native -O3 -march=native -flto -O1 -O3 -flto -O3 -Og -O2 -O2 -flto -O0 -Ofast 3 6 9 12 15 SE +/- 0.0151, N = 3 SE +/- 0.0180, N = 3 SE +/- 0.0113, N = 3 SE +/- 0.0219, N = 3 SE +/- 0.0228, N = 3 SE +/- 0.0190, N = 3 SE +/- 0.0093, N = 3 SE +/- 0.0166, N = 3 SE +/- 0.0276, N = 3 SE +/- 0.0141, N = 3 SE +/- 0.0078, N = 3 SE +/- 0.0209, N = 3 9.3601 9.3662 9.3948 9.3991 9.7734 9.8010 9.8071 9.8124 9.8158 9.8211 9.8270 9.8377 -O3 -march=native -march=native -Ofast -march=native -O3 -march=native -O1 -O3 -O3 -Og -O0 -Ofast 1. (CXX) g++ options: -O2 -flto -pthread
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p -O3 -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -O0 -O3 -flto -O1 -O2 -Og -Ofast -march=native -O3 -Ofast 40 80 120 160 200 SE +/- 0.17, N = 3 SE +/- 0.10, N = 3 SE +/- 0.48, N = 3 SE +/- 0.74, N = 3 SE +/- 0.43, N = 3 SE +/- 0.21, N = 3 SE +/- 0.07, N = 3 SE +/- 0.19, N = 3 SE +/- 0.06, N = 3 SE +/- 0.26, N = 3 SE +/- 0.10, N = 3 SE +/- 0.36, N = 3 204.96 199.57 198.90 198.86 198.38 198.37 198.18 197.89 197.85 197.26 196.43 195.23 -march=native -march=native -flto -march=native -flto -O0 -flto -O1 -Og -Ofast -march=native -Ofast 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm
x265 This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 4K -O3 -march=native -O2 -march=native -O2 -O1 -Ofast -O0 -O3 -O2 -flto -Ofast -march=native -O3 -flto -Og -O3 -march=native -flto 4 8 12 16 20 SE +/- 0.12, N = 3 SE +/- 0.16, N = 5 SE +/- 0.18, N = 4 SE +/- 0.17, N = 4 SE +/- 0.21, N = 3 SE +/- 0.10, N = 3 SE +/- 0.17, N = 4 SE +/- 0.13, N = 8 SE +/- 0.15, N = 3 SE +/- 0.17, N = 5 SE +/- 0.15, N = 3 SE +/- 0.10, N = 15 16.02 15.89 15.74 15.72 15.67 15.65 15.55 15.48 15.41 15.38 15.38 15.27 -O3 -march=native -march=native -O1 -Ofast -O0 -O3 -flto -Ofast -march=native -O3 -flto -Og -O3 -march=native -flto 1. (CXX) g++ options: -O2 -rdynamic -lpthread -lrt -ldl
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 1.3.33 Operation: Rotate -Og -O3 -march=native -O2 -flto -O1 -Ofast -march=native -O3 -O3 -march=native -flto -Ofast -O2 -march=native -O3 -flto -O2 -O0 200 400 600 800 1000 SE +/- 2.08, N = 3 SE +/- 2.03, N = 3 SE +/- 1.76, N = 3 SE +/- 1.20, N = 3 SE +/- 1.00, N = 3 SE +/- 0.67, N = 3 SE +/- 1.86, N = 3 SE +/- 2.40, N = 3 SE +/- 2.40, N = 3 SE +/- 2.00, N = 3 SE +/- 1.15, N = 3 1115 1094 1079 1078 1077 1077 1076 1075 1075 1074 1072 1063 -Og -O3 -march=native -O2 -flto -O1 -Ofast -march=native -O3 -O3 -march=native -flto -Ofast -O2 -march=native -O3 -flto -O2 -O0 1. (CC) gcc options: -fopenmp -pthread -ljpeg -lz -lm -lpthread
SVT-HEVC This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 1080p -O3 -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -O3 -flto -O0 -O1 -Og -O2 -Ofast -march=native -O3 -Ofast 60 120 180 240 300 SE +/- 0.60, N = 3 SE +/- 0.37, N = 3 SE +/- 0.18, N = 3 SE +/- 0.29, N = 3 SE +/- 0.32, N = 3 SE +/- 0.45, N = 3 SE +/- 0.19, N = 3 SE +/- 0.29, N = 3 SE +/- 0.15, N = 3 SE +/- 0.39, N = 3 SE +/- 0.11, N = 3 SE +/- 0.41, N = 3 279.12 275.57 274.64 274.39 273.85 273.72 271.99 271.78 271.66 271.62 270.19 267.46 -march=native -march=native -flto -march=native -flto -flto -O0 -O1 -Og -Ofast -march=native -Ofast 1. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 1080p -O3 -march=native -O3 -march=native -flto -O2 -march=native -O0 -Ofast -march=native -O2 -flto -Og -O3 -flto -O2 -O1 -O3 -Ofast 40 80 120 160 200 SE +/- 1.49, N = 10 SE +/- 0.22, N = 3 SE +/- 0.61, N = 3 SE +/- 0.42, N = 3 SE +/- 0.20, N = 3 SE +/- 0.36, N = 3 SE +/- 0.10, N = 3 SE +/- 0.30, N = 3 SE +/- 0.33, N = 3 SE +/- 1.54, N = 9 SE +/- 0.14, N = 3 SE +/- 0.39, N = 3 198.73 195.12 195.03 193.96 193.79 193.70 193.57 193.53 192.87 191.41 191.35 190.49 -march=native -march=native -flto -march=native -O0 -Ofast -march=native -flto -Og -flto -O1 -Ofast 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p -O3 -march=native -O3 -march=native -flto -O2 -march=native -Ofast -march=native -O2 -flto -O3 -flto -Og -O1 -O0 -O2 -O3 -Ofast 40 80 120 160 200 SE +/- 0.27, N = 3 SE +/- 0.27, N = 3 SE +/- 0.09, N = 3 SE +/- 0.17, N = 3 SE +/- 0.22, N = 3 SE +/- 0.21, N = 3 SE +/- 0.11, N = 3 SE +/- 0.29, N = 3 SE +/- 0.16, N = 3 SE +/- 0.11, N = 3 SE +/- 0.36, N = 3 SE +/- 0.18, N = 3 166.43 165.01 164.42 163.33 162.81 162.78 161.02 160.73 160.72 160.47 159.86 159.74 -march=native -march=native -flto -march=native -Ofast -march=native -flto -flto -Og -O1 -O0 -Ofast 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm
Mobile Neural Network MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: SqueezeNetV1.0 -O3 -march=native -flto -Ofast -march=native -O3 -march=native -O2 -march=native -O2 -flto -O3 -flto -O2 -O0 -O3 -Ofast -O1 -Og 0.8685 1.737 2.6055 3.474 4.3425 SE +/- 0.025, N = 3 SE +/- 0.019, N = 3 SE +/- 0.024, N = 3 SE +/- 0.026, N = 3 SE +/- 0.029, N = 3 SE +/- 0.041, N = 3 SE +/- 0.035, N = 3 SE +/- 0.043, N = 3 SE +/- 0.008, N = 3 SE +/- 0.005, N = 3 SE +/- 0.019, N = 3 SE +/- 0.004, N = 3 3.706 3.741 3.748 3.753 3.767 3.815 3.824 3.825 3.838 3.845 3.848 3.860 -march=native -flto - MIN: 3.61 / MAX: 7.92 -Ofast -march=native - MIN: 3.65 / MAX: 8.19 -march=native - MIN: 3.64 / MAX: 10.5 -march=native - MIN: 3.66 / MAX: 8.01 -flto - MIN: 3.69 / MAX: 8.26 -flto - MIN: 3.69 / MAX: 8.62 MIN: 3.71 / MAX: 8.07 -O0 - MIN: 3.69 / MAX: 8.15 MIN: 3.78 / MAX: 8.07 -Ofast - MIN: 3.78 / MAX: 8.38 -O1 - MIN: 3.75 / MAX: 8.08 -Og - MIN: 3.81 / MAX: 8.2 1. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: MobileNetV2_224 -O3 -march=native -flto -O3 -march=native -O2 -march=native -Ofast -march=native -O2 -flto -O2 -O3 -flto -O0 -O3 -Og -Ofast -O1 0.446 0.892 1.338 1.784 2.23 SE +/- 0.011, N = 3 SE +/- 0.008, N = 3 SE +/- 0.007, N = 3 SE +/- 0.007, N = 3 SE +/- 0.003, N = 3 SE +/- 0.008, N = 3 SE +/- 0.015, N = 3 SE +/- 0.014, N = 3 SE +/- 0.007, N = 3 SE +/- 0.010, N = 3 SE +/- 0.006, N = 3 SE +/- 0.011, N = 3 1.907 1.916 1.921 1.922 1.944 1.958 1.962 1.966 1.969 1.973 1.976 1.982 -march=native -flto - MIN: 1.84 / MAX: 6.27 -march=native - MIN: 1.87 / MAX: 6.22 -march=native - MIN: 1.88 / MAX: 6.22 -Ofast -march=native - MIN: 1.88 / MAX: 6.23 -flto - MIN: 1.9 / MAX: 6.25 MIN: 1.91 / MAX: 6.28 -flto - MIN: 1.9 / MAX: 6.25 -O0 - MIN: 1.91 / MAX: 6.27 MIN: 1.93 / MAX: 6.3 -Og - MIN: 1.93 / MAX: 6.26 -Ofast - MIN: 1.93 / MAX: 6.27 -O1 - MIN: 1.93 / MAX: 7.73 1. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl
SVT-HEVC This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 1080p -O3 -march=native -flto -O3 -flto -O3 -march=native -O2 -flto -O2 -march=native -Ofast -march=native -O0 -Og -O1 -O2 -O3 -Ofast 30 60 90 120 150 SE +/- 0.14, N = 3 SE +/- 0.12, N = 3 SE +/- 0.11, N = 3 SE +/- 0.14, N = 3 SE +/- 0.07, N = 3 SE +/- 0.11, N = 3 SE +/- 0.24, N = 3 SE +/- 0.23, N = 3 SE +/- 0.28, N = 3 SE +/- 0.20, N = 3 SE +/- 0.15, N = 3 SE +/- 0.10, N = 3 142.04 140.48 140.40 140.32 139.61 138.97 137.47 137.30 137.23 137.13 136.89 136.73 -march=native -flto -flto -march=native -flto -march=native -Ofast -march=native -O0 -Og -O1 -Ofast 1. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 3.0 Preset: Medium -O3 -march=native -Ofast -march=native -O2 -march=native -O3 -march=native -flto -O2 -Og -O3 -O1 -O3 -flto -O2 -flto -O0 -Ofast 0.9852 1.9704 2.9556 3.9408 4.926 SE +/- 0.0026, N = 3 SE +/- 0.0057, N = 3 SE +/- 0.0054, N = 3 SE +/- 0.0030, N = 3 SE +/- 0.0027, N = 3 SE +/- 0.0062, N = 3 SE +/- 0.0036, N = 3 SE +/- 0.0112, N = 3 SE +/- 0.0088, N = 3 SE +/- 0.0076, N = 3 SE +/- 0.0088, N = 3 SE +/- 0.0107, N = 3 4.2153 4.2296 4.2307 4.2359 4.3470 4.3511 4.3603 4.3606 4.3686 4.3697 4.3715 4.3786 -O3 -march=native -Ofast -march=native -march=native -O3 -march=native -Og -O3 -O1 -O3 -O0 -Ofast 1. (CXX) g++ options: -O2 -flto -pthread
PostMark This is a test of NetApp's PostMark benchmark designed to simulate small-file testing similar to the tasks endured by web and mail servers. This test profile will set PostMark to perform 25,000 transactions with 500 files simultaneously with the file sizes ranging between 5 and 512 kilobytes. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostMark 1.51 Disk Transaction Performance -O3 -march=native -Ofast -march=native -O3 -march=native -flto -O2 -flto -O2 -march=native -O2 -Og -O1 -Ofast -O3 -flto -O3 -O0 2K 4K 6K 8K 10K SE +/- 118.67, N = 3 SE +/- 110.33, N = 3 SE +/- 110.33, N = 3 SE +/- 110.33, N = 3 SE +/- 110.33, N = 3 9496 9259 9259 9259 9259 9259 9259 9259 9149 9149 9149 9149 -march=native -Ofast -march=native -march=native -flto -O2 -flto -O2 -march=native -O2 -Og -O1 -Ofast -flto -O0 1. (CC) gcc options: -O3
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 3.0 Preset: Exhaustive -O3 -march=native -O2 -march=native -Ofast -march=native -O3 -march=native -flto -O1 -Og -O0 -O2 -flto -O2 -O3 -Ofast -O3 -flto 12 24 36 48 60 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 51.49 51.49 51.51 51.54 53.25 53.27 53.28 53.28 53.28 53.29 53.29 53.30 -O3 -march=native -march=native -Ofast -march=native -O3 -march=native -O1 -Og -O0 -O3 -Ofast -O3 1. (CXX) g++ options: -O2 -flto -pthread
Mobile Neural Network MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: mobilenet-v1-1.0 -O3 -march=native -flto -Ofast -march=native -O3 -march=native -O2 -march=native -O2 -flto -O3 -flto -O0 -O2 -O1 -Ofast -O3 -Og 0.4345 0.869 1.3035 1.738 2.1725 SE +/- 0.004, N = 3 SE +/- 0.004, N = 3 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 SE +/- 0.005, N = 3 SE +/- 0.005, N = 3 SE +/- 0.006, N = 3 SE +/- 0.005, N = 3 SE +/- 0.004, N = 3 SE +/- 0.003, N = 3 SE +/- 0.008, N = 3 SE +/- 0.000, N = 3 1.878 1.881 1.883 1.884 1.917 1.918 1.919 1.920 1.921 1.921 1.924 1.931 -march=native -flto - MIN: 1.85 / MAX: 6.19 -Ofast -march=native - MIN: 1.85 / MAX: 6.17 -march=native - MIN: 1.85 / MAX: 7.81 -march=native - MIN: 1.85 / MAX: 6.17 -flto - MIN: 1.89 / MAX: 7.02 -flto - MIN: 1.89 / MAX: 6.19 -O0 - MIN: 1.89 / MAX: 6.19 MIN: 1.89 / MAX: 6.2 -O1 - MIN: 1.89 / MAX: 9.19 -Ofast - MIN: 1.89 / MAX: 6.22 MIN: 1.89 / MAX: 8.6 -Og - MIN: 1.9 / MAX: 8.69 1. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: vgg16 -O3 -march=native -O2 -march=native -O1 -Ofast -march=native -O2 -O0 -Og -O2 -flto -O3 -Ofast -O3 -flto -O3 -march=native -flto 13 26 39 52 65 SE +/- 0.11, N = 3 SE +/- 0.23, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.14, N = 3 SE +/- 0.09, N = 3 SE +/- 0.10, N = 3 SE +/- 0.08, N = 3 SE +/- 0.16, N = 3 SE +/- 0.18, N = 3 SE +/- 0.12, N = 3 SE +/- 0.04, N = 3 54.36 54.59 54.91 55.15 55.17 55.31 55.34 55.51 55.77 55.77 55.79 55.82 -O3 -march=native - MIN: 53.85 / MAX: 59.24 -march=native - MIN: 53.82 / MAX: 59.28 -O1 - MIN: 54.36 / MAX: 58.94 -Ofast -march=native - MIN: 54.82 / MAX: 60.18 MIN: 54.55 / MAX: 59.77 -O0 - MIN: 54.78 / MAX: 60.73 -Og - MIN: 54.84 / MAX: 62.65 -flto - MIN: 55.03 / MAX: 59.08 -O3 - MIN: 55.15 / MAX: 66.36 -Ofast - MIN: 55.26 / MAX: 59.78 -O3 -flto - MIN: 55.29 / MAX: 59.45 -O3 -march=native -flto - MIN: 55.41 / MAX: 61.05 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
Redis Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: GET -O2 -O3 -O2 -march=native -O3 -march=native -O3 -march=native -flto -Ofast -march=native -Ofast -O2 -flto -Og -O3 -flto -O1 -O0 900K 1800K 2700K 3600K 4500K SE +/- 3379.85, N = 3 SE +/- 13278.68, N = 3 SE +/- 19952.38, N = 3 SE +/- 18099.88, N = 3 SE +/- 15517.78, N = 3 SE +/- 13892.93, N = 3 SE +/- 28152.86, N = 3 SE +/- 50798.23, N = 3 SE +/- 29703.89, N = 3 SE +/- 39966.63, N = 3 SE +/- 33158.80, N = 3 SE +/- 48889.70, N = 3 4081087.33 4056377.08 4054268.33 4049394.67 4042208.58 4038893.75 4026735.50 4021727.25 4019272.00 4018545.17 3982525.83 3975306.33 -O2 -O2 -march=native -march=native -march=native -flto -Ofast -march=native -Ofast -O2 -flto -Og -flto -O1 -O0 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU -Og -O1 -O0 -O2 -O3 -march=native -O3 -flto -O3 -march=native -flto -Ofast -march=native -Ofast -O3 -O2 -march=native -O2 -flto 400 800 1200 1600 2000 SE +/- 1.91, N = 3 SE +/- 4.14, N = 3 SE +/- 1.24, N = 3 SE +/- 1.27, N = 3 SE +/- 1.46, N = 3 SE +/- 1.95, N = 3 SE +/- 0.62, N = 3 SE +/- 1.60, N = 3 SE +/- 0.97, N = 3 SE +/- 1.53, N = 3 SE +/- 0.46, N = 3 SE +/- 0.58, N = 3 1836.67 1854.40 1860.28 1876.16 1876.42 1877.44 1878.45 1880.54 1880.71 1880.75 1880.88 1883.34 -Og - MIN: 1823.46 -O1 - MIN: 1837.76 -O0 - MIN: 1849.08 MIN: 1864.8 MIN: 1865.18 -flto - MIN: 1864.71 -flto - MIN: 1869.12 -Ofast - MIN: 1868.92 -Ofast - MIN: 1870.91 MIN: 1870.28 MIN: 1870.79 -flto - MIN: 1872.75 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl
Mobile Neural Network MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: inception-v3 -O3 -march=native -Ofast -march=native -O3 -march=native -flto -O2 -march=native -O2 -flto -Ofast -Og -O3 -flto -O1 -O2 -O3 -O0 6 12 18 24 30 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.20, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.18, N = 3 SE +/- 0.20, N = 3 SE +/- 0.16, N = 3 22.51 22.53 22.53 22.79 22.83 22.91 22.93 22.93 22.94 23.07 23.07 23.08 -march=native - MIN: 22.19 / MAX: 27.64 -Ofast -march=native - MIN: 22.28 / MAX: 29.02 -march=native -flto - MIN: 22.28 / MAX: 27.89 -march=native - MIN: 22.14 / MAX: 27.48 -flto - MIN: 22.56 / MAX: 32.63 -Ofast - MIN: 22.68 / MAX: 28.85 -Og - MIN: 22.69 / MAX: 27.2 -flto - MIN: 22.56 / MAX: 29.28 -O1 - MIN: 22.65 / MAX: 29.53 MIN: 22.73 / MAX: 29.81 MIN: 22.71 / MAX: 30.08 -O0 - MIN: 22.77 / MAX: 27.58 1. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl
PJSIP PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: OPTIONS, Stateful -O2 -flto -O3 -flto -O3 -march=native -flto -Ofast -Ofast -march=native -O2 -march=native -O3 -O3 -march=native -O2 -O1 -Og -O0 2K 4K 6K 8K 10K SE +/- 3.28, N = 3 SE +/- 1.20, N = 3 SE +/- 6.08, N = 3 SE +/- 5.55, N = 3 SE +/- 6.43, N = 3 SE +/- 3.79, N = 3 SE +/- 3.06, N = 3 SE +/- 7.69, N = 3 SE +/- 4.36, N = 3 SE +/- 4.41, N = 3 SE +/- 7.86, N = 3 SE +/- 2.73, N = 3 9406 9405 9396 9384 9381 9380 9375 9375 9366 9333 9311 9189 -O2 -flto -O3 -flto -O3 -march=native -flto -Ofast -Ofast -march=native -O2 -march=native -O3 -O3 -march=native -O2 -O1 -Og -O0 1. (CC) gcc options: -lstdc++ -lssl -lcrypto -lm -lrt -lpthread
Mobile Neural Network MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: resnet-v2-50 -O3 -march=native -O3 -march=native -flto -Ofast -march=native -O2 -flto -O3 -flto -O2 -march=native -O1 -Og -Ofast -O0 -O2 -O3 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.14, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.16, N = 3 SE +/- 0.19, N = 3 SE +/- 0.14, N = 3 19.22 19.29 19.32 19.33 19.38 19.45 19.51 19.54 19.54 19.57 19.62 19.67 -march=native - MIN: 19.06 / MAX: 24.92 -march=native -flto - MIN: 19.06 / MAX: 25.79 -Ofast -march=native - MIN: 19.15 / MAX: 28.31 -flto - MIN: 19.14 / MAX: 23.64 -flto - MIN: 19.15 / MAX: 25.49 -march=native - MIN: 19.15 / MAX: 26.45 -O1 - MIN: 19.33 / MAX: 23.75 -Og - MIN: 19.39 / MAX: 23.78 -Ofast - MIN: 19.37 / MAX: 25.55 -O0 - MIN: 19.17 / MAX: 25.8 MIN: 19.25 / MAX: 24.23 MIN: 19.37 / MAX: 24.17 1. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: alexnet -O0 -Og -O2 -O3 -Ofast -march=native -O2 -march=native -Ofast -O3 -march=native -flto -O3 -flto -O2 -flto 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 9.63 9.64 9.65 9.68 9.68 9.69 9.72 9.74 9.82 9.83 -O0 - MIN: 9.54 / MAX: 13.14 -Og - MIN: 9.54 / MAX: 13.27 MIN: 9.55 / MAX: 14.36 -O3 - MIN: 9.59 / MAX: 13.26 -Ofast -march=native - MIN: 9.6 / MAX: 13.26 -march=native - MIN: 9.58 / MAX: 14.24 -Ofast - MIN: 9.62 / MAX: 13.28 -O3 -march=native -flto - MIN: 9.64 / MAX: 13.24 -O3 -flto - MIN: 9.75 / MAX: 13.41 -flto - MIN: 9.75 / MAX: 13.44 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU -Og -O1 -O0 -O3 -march=native -O2 -O2 -march=native -Ofast -O3 -march=native -flto -O3 -flto -O3 -Ofast -march=native -O2 -flto 700 1400 2100 2800 3500 SE +/- 1.46, N = 3 SE +/- 2.80, N = 3 SE +/- 0.67, N = 3 SE +/- 1.32, N = 3 SE +/- 1.81, N = 3 SE +/- 0.97, N = 3 SE +/- 1.46, N = 3 SE +/- 1.52, N = 3 SE +/- 1.80, N = 3 SE +/- 0.66, N = 3 SE +/- 0.15, N = 3 SE +/- 2.58, N = 3 3127.13 3133.28 3143.03 3165.60 3168.45 3171.69 3177.34 3177.57 3177.88 3177.98 3180.01 3182.48 -Og - MIN: 3113.66 -O1 - MIN: 3120.48 -O0 - MIN: 3131.71 MIN: 3154.25 MIN: 3155.33 MIN: 3161.55 -Ofast - MIN: 3164.15 -flto - MIN: 3166.18 -flto - MIN: 3165.88 MIN: 3167.31 -Ofast - MIN: 3168.64 -flto - MIN: 3168.44 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl
Chia Blockchain VDF Chia is a blockchain and smart transaction platform based on proofs of space and time rather than proofs of work with other cryptocurrencies. This test profile is benchmarking the CPU performance for Chia VDF performance using the Chia VDF benchmark. The Chia VDF is for the Chia Verifiable Delay Function (Proof of Time). Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IPS, More Is Better Chia Blockchain VDF 1.0.1 Test: Square Assembly Optimized -O3 -march=native -O2 -march=native -O3 -march=native -flto -O0 -O3 -O2 -flto -Ofast -march=native -Og -O1 -O3 -flto -Ofast -O2 50K 100K 150K 200K 250K SE +/- 1105.04, N = 3 SE +/- 200.00, N = 3 SE +/- 961.48, N = 3 SE +/- 520.68, N = 3 SE +/- 674.12, N = 3 SE +/- 484.19, N = 3 SE +/- 1020.35, N = 3 SE +/- 1816.90, N = 3 SE +/- 1471.21, N = 3 SE +/- 1809.54, N = 3 SE +/- 1922.09, N = 3 SE +/- 472.58, N = 3 250633 250600 249567 249533 248167 248033 247967 247933 247933 247867 246967 246700 1. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
Caffe This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: GoogleNet - Acceleration: CPU - Iterations: 100 -O3 -march=native -O2 -march=native -Ofast -march=native -O3 -flto -O2 -flto -O3 -march=native -flto -O1 -Og -O0 -Ofast -O3 -O2 20K 40K 60K 80K 100K SE +/- 43.97, N = 3 SE +/- 35.23, N = 3 SE +/- 22.39, N = 3 SE +/- 47.42, N = 3 SE +/- 32.64, N = 3 SE +/- 61.19, N = 3 SE +/- 10.17, N = 3 SE +/- 73.11, N = 3 SE +/- 84.95, N = 3 SE +/- 30.02, N = 3 SE +/- 30.68, N = 3 SE +/- 133.67, N = 3 83625 83638 83693 83713 83722 84271 84729 84809 84831 84843 84845 84895 -O3 -march=native -march=native -Ofast -march=native -O3 -flto -flto -O3 -march=native -flto -O1 -Og -O0 -Ofast -O3 1. (CXX) g++ options: -fPIC -O2 -rdynamic -lboost_system -lboost_thread -lboost_filesystem -lboost_chrono -lboost_date_time -lboost_atomic -lglog -lgflags -lprotobuf -lpthread -lhdf5_cpp -lhdf5 -lhdf5_hl_cpp -lhdf5_hl -llmdb -lopenblas
-O3 -march=native Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 9 June 2021 17:11 by user phoronix.
-O1 Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-O1 CFLAGS=-O1Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 10 June 2021 11:06 by user phoronix.
-O0 Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-O0 CFLAGS=-O0Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 10 June 2021 20:07 by user phoronix.
-Og Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-Og CFLAGS=-OgCompiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 11 June 2021 04:13 by user phoronix.
-O2 Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-O2 CFLAGS=-O2Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 11 June 2021 09:34 by user phoronix.
-O2 -march=native Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O2 -march=native" CFLAGS="-O2 -march=native"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 11 June 2021 14:55 by user phoronix.
-O2 -flto Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O2 -flto" CFLAGS="-O2 -flto"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 11 June 2021 19:43 by user phoronix.
-O3 Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-O3 CFLAGS=-O3Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 June 2021 04:33 by user phoronix.
-O3 -flto Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -flto" CFLAGS="-O3 -flto"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 June 2021 10:11 by user phoronix.
-O3 -march=native -flto Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-O3 -march=native -flto" CFLAGS="-O3 -march=native -flto"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 June 2021 15:57 by user phoronix.
-Ofast -march=native Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS="-Ofast -march=native" CFLAGS="-Ofast -march=native"Compiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 13 June 2021 04:39 by user phoronix.
-Ofast Processor: Intel Core i9-11900K @ 5.10GHz (8 Cores / 16 Threads), Motherboard: ASUS ROG MAXIMUS XIII HERO (0707 BIOS), Chipset: Intel Tiger Lake-H, Memory: 32GB, Disk: 2000GB Corsair Force MP600 + 257GB Flash Drive, Graphics: AMD Radeon VII 16GB (1801/1000MHz), Audio: Intel Tiger Lake-H HD Audio, Monitor: ASUS MG28U, Network: 2 x Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411
OS: Fedora 34, Kernel: 5.12.9-300.fc34.x86_64 (x86_64), Desktop: GNOME Shell 40.1, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 21.1.1 (LLVM 12.0.0), Compiler: GCC 11.1.1 20210531, File-System: btrfs, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseEnvironment Notes: CXXFLAGS=-Ofast CFLAGS=-OfastCompiler Notes: --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-driverDisk Notes: NONE / compress=zstd:1,relatime,rw,seclabel,space_cache,ssd,subvol=/home,subvolid=256 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x3c - Thermald 2.4.4Python Notes: Python 3.9.5Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 13 June 2021 10:35 by user phoronix.