Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 1701204-TA-1701179TA92 Cortex A53 vs Cortex A5 - Phoronix Test Suite Cortex A53 vs Cortex A5 ARMv7 codegen options comparison
HTML result view exported from: https://openbenchmarking.org/result/1701204-TA-1701179TA92&rdt&grw .
Cortex A53 vs Cortex A5 Processor Motherboard Memory Disk OS Kernel Compiler File-System Screen Resolution Display Server thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 AArch64 rev 4 @ 1.50GHz (4 Cores) Amlogic 2048MB 32GB 00000 + 16GB NCard Ubuntu 16.04 3.14.29 (aarch64) GCC 7.0.0 20170110 + Clang 3.8.0-2ubuntu4 + LLVM 3.8.0 ext4 1920x3240 AArch64 rev 4 @ 1.55GHz (4 Cores) GCC 7.0.0 20170113 + Clang 3.8.0-2ubuntu4 + LLVM 3.8.0 ARMv7 rev 1 @ 1.73GHz (4 Cores) ODROIDC 916MB 64GB 00000 Ubuntu 14.04 3.10.96-149 (armv7l) X Server 1.15.1 GCC 7.0.0 20170114 + Clang 3.6.0-2ubuntu1~trusty1 + LLVM 3.6.0 1280x1024 GCC 4.9.4 + Clang 3.6.0-2ubuntu1~trusty1 + LLVM 3.6.0 OpenBenchmarking.org Compiler Details - thunderx/vectorize, pre-patch: --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new - A53 vectorize, pre-patch: --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new - A53 vectorize/LTO, pre patch: --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new - Cortex-A53: --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new - Cortex-A53 vectorize: --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new - Cortex-A5 VFPV4: --build=arm-linux-gnueabihf --disable-browser-plugin --disable-libitm --disable-libquadmath --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-multilib --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=arm-linux-gnueabihf --target=arm-linux-gnueabihf --with-arch-directory=arm --with-arch=armv7-a --with-default-libstdcxx-abi=gcc4-compatible --with-float=hard --with-fpu=vfpv3 --with-mode=arm -v - Cortex-A5 Neon: --build=arm-linux-gnueabihf --disable-browser-plugin --disable-libitm --disable-libquadmath --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-multilib --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=arm-linux-gnueabihf --target=arm-linux-gnueabihf --with-arch-directory=arm --with-arch=armv7-a --with-default-libstdcxx-abi=gcc4-compatible --with-float=hard --with-fpu=vfpv3 --with-mode=arm -v - Cortex-A5 NEON-vfpv4: --build=arm-linux-gnueabihf --disable-browser-plugin --disable-libitm --disable-libquadmath --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-multilib --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=arm-linux-gnueabihf --target=arm-linux-gnueabihf --with-arch-directory=arm --with-arch=armv7-a --with-default-libstdcxx-abi=gcc4-compatible --with-float=hard --with-fpu=vfpv3 --with-mode=arm -v - GCC 4.9 NEON-vfpv4: --build=arm-linux-gnueabihf --disable-browser-plugin --disable-libitm --disable-libquadmath --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,c++,java,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=arm-linux-gnueabihf --target=arm-linux-gnueabihf --with-arch-directory=arm --with-arch=armv7-a --with-float=hard --with-fpu=vfpv3-d16 --with-mode=thumb -v - GCC 4.9 VFPv4: --build=arm-linux-gnueabihf --disable-browser-plugin --disable-libitm --disable-libquadmath --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,c++,java,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=arm-linux-gnueabihf --target=arm-linux-gnueabihf --with-arch-directory=arm --with-arch=armv7-a --with-float=hard --with-fpu=vfpv3-d16 --with-mode=thumb -v Disk Details - thunderx/vectorize, pre-patch: CFQ / commit=30,errors=remount-ro,noatime,nodiratime,rw - A53 vectorize, pre-patch: CFQ / commit=30,errors=remount-ro,noatime,nodiratime,rw - A53 vectorize/LTO, pre patch: CFQ / commit=30,errors=remount-ro,noatime,nodiratime,rw - Cortex-A53: CFQ / commit=30,errors=remount-ro,noatime,nodiratime,rw - Cortex-A53 vectorize: CFQ / commit=30,errors=remount-ro,noatime,nodiratime,rw - Cortex-A5 VFPV4: CFQ / commit=25,errors=remount-ro,noatime,nodiratime,rw - Cortex-A5 Neon: CFQ / commit=25,errors=remount-ro,noatime,nodiratime,rw - Cortex-A5 NEON-vfpv4: CFQ / commit=25,errors=remount-ro,noatime,nodiratime,rw - GCC 4.9 NEON-vfpv4: CFQ / commit=25,errors=remount-ro,noatime,nodiratime,rw - GCC 4.9 VFPv4: CFQ / commit=25,errors=remount-ro,noatime,nodiratime,rw Processor Details - thunderx/vectorize, pre-patch: Scaling Governor: meson_cpufreq performance - A53 vectorize, pre-patch: Scaling Governor: meson_cpufreq performance - A53 vectorize/LTO, pre patch: Scaling Governor: meson_cpufreq performance - Cortex-A53: Scaling Governor: meson_cpufreq performance - Cortex-A53 vectorize: Scaling Governor: meson_cpufreq interactive - Cortex-A5 VFPV4: Scaling Governor: meson_cpufreq performance - Cortex-A5 Neon: Scaling Governor: meson_cpufreq performance - Cortex-A5 NEON-vfpv4: Scaling Governor: meson_cpufreq performance - GCC 4.9 NEON-vfpv4: Scaling Governor: meson_cpufreq performance - GCC 4.9 VFPv4: Scaling Governor: meson_cpufreq performance
Cortex A53 vs Cortex A5 ramspeed: Copy - Integer ramspeed: Copy - Floating Point fftw: Stock - 2D FFT Size 2048 mafft: Multiple Sequence Alignment postmark: Disk Transaction Performance primesieve: 1e12 Prime Number Generation tachyon: Total Time c-ray: Total Time ttsiod-renderer: Phong Rendering With Soft-Shadow Mapping smallpt: Global Illumination Renderer; 100 Samples openssl: RSA 4096-bit Performance redis: GET fhourstones: Complex Connect-4 Solving gmpbench: Total Time sudokut: Total Time thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 2821.43 2817.45 190.63 34.46 1351 566.21 71.41 149.82 23.01 167 21.50 318926.02 3210.20 554.83 102.75 4581.32 4580.39 196.90 35.42 1363 543.16 69.27 187.97 23.16 167 21.50 310344.73 3212.10 552.84 101.95 4829.91 4825.13 180.53 33.16 1378 540.95 67.64 184.81 23.77 168 21.50 311785.02 3213.77 554.37 101.75 4965.06 4964.66 186.21 33.06 1381 553.13 69.40 186.69 23.47 168 21.50 309030.64 3209.67 552.56 101.88 4955.97 4965.60 184.81 32.17 1378 573.13 69.34 186.61 23.49 167 21.50 313438.91 3205.40 555.10 102.17 1476.43 1365.27 135.07 43.74 725 836.59 107.33 273.18 16.91 306 20.23 174917.94 1211.33 221.56 167.97 1262.11 1049.66 128.75 45.21 717 1021.13 247.96 235.23 15.75 289 20.30 177754.08 1246.77 221.63 165.14 1482.21 1369.05 132.50 46.12 744 1029.57 108.33 244.62 15.77 291 20.30 179240.55 1228.77 221.48 167.41 1466.11 1302.02 127.22 41.68 692 894.65 109.85 265.13 15.78 287 20.33 171955.12 1186.23 221.04 169.11 1567.32 1114.65 134.44 31.59 720 948.93 109.81 290.51 16.47 306 20.30 165811.52 1046.73 221.70 170.29 OpenBenchmarking.org
RAMspeed SMP Type: Copy - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Copy - Benchmark: Integer thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 1100 2200 3300 4400 5500 2821.43 4581.32 4829.91 4965.06 4955.97 1476.43 1262.11 1482.21 1466.11 1567.32
RAMspeed SMP Type: Copy - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Copy - Benchmark: Floating Point thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 1100 2200 3300 4400 5500 2817.45 4580.39 4825.13 4964.66 4965.60 1365.27 1049.66 1369.05 1302.02 1114.65
FFTW Build: Stock - Size: 2D FFT Size 2048 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.4 Build: Stock - Size: 2D FFT Size 2048 thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 40 80 120 160 200 SE +/- 1.10, N = 5 SE +/- 0.99, N = 5 SE +/- 0.49, N = 5 SE +/- 0.08, N = 5 SE +/- 0.21, N = 5 SE +/- 1.16, N = 5 SE +/- 0.03, N = 5 SE +/- 0.04, N = 5 SE +/- 0.07, N = 5 SE +/- 0.08, N = 5 190.63 196.90 180.53 186.21 184.81 135.07 128.75 132.50 127.22 134.44 -Ofast -mcpu=thunderx -march=armv8-a+crc -ftree-vectorize -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -O3 -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -flto -ffat-lto-objects -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -Ofast -mtune=cortex-a53 -march=armv8-a+crc -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -std=gnu99 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -std=gnu99 -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 1. (CC) gcc options: -fomit-frame-pointer -fipa-pta -lm
Timed MAFFT Alignment Multiple Sequence Alignment OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 6.864 Multiple Sequence Alignment thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 10 20 30 40 50 SE +/- 0.73, N = 6 SE +/- 0.80, N = 6 SE +/- 0.70, N = 6 SE +/- 0.71, N = 6 SE +/- 0.01, N = 3 SE +/- 1.04, N = 6 SE +/- 1.09, N = 6 SE +/- 0.16, N = 3 SE +/- 3.47, N = 6 SE +/- 3.83, N = 6 34.46 35.42 33.16 33.06 32.17 43.74 45.21 46.12 41.68 31.59 1. (CC) gcc options: -O3 -lm -lpthread
PostMark Disk Transaction Performance OpenBenchmarking.org TPS, More Is Better PostMark 1.51 Disk Transaction Performance thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 300 600 900 1200 1500 SE +/- 0.00, N = 3 SE +/- 2.67, N = 3 SE +/- 2.67, N = 3 SE +/- 4.33, N = 3 SE +/- 5.00, N = 3 SE +/- 11.49, N = 6 SE +/- 10.35, N = 3 SE +/- 1.67, N = 3 SE +/- 10.49, N = 3 SE +/- 9.40, N = 6 1351 1363 1378 1381 1378 725 717 744 692 720 1. (CC) gcc options: -O3
Primesieve 1e12 Prime Number Generation OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 5.4.2 1e12 Prime Number Generation thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 200 400 600 800 1000 SE +/- 2.99, N = 3 SE +/- 3.01, N = 3 SE +/- 8.42, N = 3 SE +/- 9.14, N = 3 SE +/- 6.92, N = 3 SE +/- 6.29, N = 3 SE +/- 11.32, N = 3 SE +/- 31.67, N = 6 SE +/- 14.88, N = 3 SE +/- 9.13, N = 3 566.21 543.16 540.95 553.13 573.13 836.59 1021.13 1029.57 894.65 948.93 -Ofast -mcpu=thunderx -march=armv8-a+crc -ftree-vectorize -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -O3 -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -flto -ffat-lto-objects -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -Ofast -mtune=cortex-a53 -march=armv8-a+crc -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 1. (CXX) g++ options: -fomit-frame-pointer -fipa-pta -fopenmp
Tachyon Total Time OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.98.9 Total Time thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 50 100 150 200 250 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 SE +/- 0.12, N = 3 SE +/- 0.10, N = 3 SE +/- 0.62, N = 3 SE +/- 0.35, N = 3 SE +/- 0.10, N = 3 SE +/- 0.18, N = 3 SE +/- 0.21, N = 3 71.41 69.27 67.64 69.40 69.34 107.33 247.96 108.33 109.85 109.81
C-Ray Total Time OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 60 120 180 240 300 SE +/- 1.37, N = 3 SE +/- 0.69, N = 3 SE +/- 0.17, N = 3 SE +/- 0.14, N = 3 SE +/- 0.12, N = 3 SE +/- 0.14, N = 3 SE +/- 0.29, N = 3 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 SE +/- 0.13, N = 3 149.82 187.97 184.81 186.69 186.61 273.18 235.23 244.62 265.13 290.51 -Ofast -mcpu=thunderx -march=armv8-a+crc -ftree-vectorize -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -flto -ffat-lto-objects -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -Ofast -mtune=cortex-a53 -march=armv8-a+crc -ftree-vectorize -mcpu=cortex-a5 -marm -mfpu=vfpv4 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -marm -mfpu=vfpv4 1. (CC) gcc options: -lm -lpthread -O3 -fomit-frame-pointer -fipa-pta
TTSIOD 3D Renderer Phong Rendering With Soft-Shadow Mapping OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3a Phong Rendering With Soft-Shadow Mapping thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 6 12 18 24 30 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 2 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 23.01 23.16 23.77 23.47 23.49 16.91 15.75 15.77 15.78 16.47 -Ofast -mcpu=thunderx -march=armv8-a+crc -ftree-vectorize -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -O3 -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -ffat-lto-objects -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -Ofast -mtune=cortex-a53 -march=armv8-a+crc -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 1. (CXX) g++ options: -fomit-frame-pointer -fipa-pta -ffast-math -mtune=native -flto -lSDL -lstdc++
Smallpt Global Illumination Renderer; 100 Samples OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 100 Samples thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 70 140 210 280 350 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 +/- 0.33, N = 3 SE +/- 1.86, N = 3 SE +/- 1.53, N = 3 SE +/- 1.00, N = 3 167 167 168 168 167 306 289 291 287 306 -Ofast -mcpu=thunderx -march=armv8-a+crc -ftree-vectorize -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -O3 -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -flto -ffat-lto-objects -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -Ofast -mtune=cortex-a53 -march=armv8-a+crc -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 1. (CXX) g++ options: -fopenmp -fomit-frame-pointer -fipa-pta
OpenSSL RSA 4096-bit Performance OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.0.1g RSA 4096-bit Performance thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 5 10 15 20 25 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 +/- 0.07, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 21.50 21.50 21.50 21.50 21.50 20.23 20.30 20.30 20.33 20.30 -fomit-frame-pointer -fomit-frame-pointer -fomit-frame-pointer -fomit-frame-pointer -fomit-frame-pointer -march=armv7-a -march=armv7-a -march=armv7-a -march=armv7-a -march=armv7-a 1. (CC) gcc options: -O3 -lssl -lcrypto -ldl
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 3.0.1 Test: GET thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 70K 140K 210K 280K 350K SE +/- 2784.59, N = 3 SE +/- 4662.92, N = 6 SE +/- 2239.53, N = 3 SE +/- 1052.91, N = 3 SE +/- 1967.34, N = 3 SE +/- 3419.06, N = 3 SE +/- 2106.27, N = 3 SE +/- 1282.01, N = 3 SE +/- 2381.85, N = 6 SE +/- 2848.52, N = 6 318926.02 310344.73 311785.02 309030.64 313438.91 174917.94 177754.08 179240.55 171955.12 165811.52 -O2 -Ofast -mcpu=thunderx -march=armv8-a+crc -ftree-vectorize -O2 -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -O2 -O3 -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -flto -ffat-lto-objects -O2 -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -O2 -Ofast -mtune=cortex-a53 -march=armv8-a+crc -ftree-vectorize -std=gnu99 -pipe -g3 -O3 -funroll-loops -mcpu=cortex-a5 -marm -mfpu=vfpv4 -O2 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon -ftree-vectorize -O2 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -O2 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -O2 -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl -fomit-frame-pointer -fipa-pta
Fhourstones Complex Connect-4 Solving OpenBenchmarking.org Kpos / sec, More Is Better Fhourstones 3.1 Complex Connect-4 Solving thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 700 1400 2100 2800 3500 SE +/- 0.76, N = 3 SE +/- 0.35, N = 3 SE +/- 0.22, N = 3 SE +/- 1.47, N = 3 SE +/- 1.81, N = 3 SE +/- 17.08, N = 3 SE +/- 0.22, N = 3 SE +/- 18.53, N = 3 SE +/- 1.73, N = 3 SE +/- 0.49, N = 3 3210.20 3212.10 3213.77 3209.67 3205.40 1211.33 1246.77 1228.77 1186.23 1046.73 1. (CC) gcc options: -O3
GMPbench Total Time OpenBenchmarking.org GMPbench Score, More Is Better GMPbench 0.2 Total Time thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 120 240 360 480 600 554.83 552.84 554.37 552.56 555.10 221.56 221.63 221.48 221.04 221.70 -Ofast -mcpu=thunderx -march=armv8-a+crc -ftree-vectorize -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -O3 -mcpu=cortex-a53 -march=armv8-a+crc -ftree-vectorize -flto -ffat-lto-objects -Ofast -mcpu=cortex-a53 -march=armv8-a+crc -Ofast -mtune=cortex-a53 -march=armv8-a+crc -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 -mcpu=cortex-a5 -Ofast -marm -mfpu=neon -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -Ofast -marm -mfpu=neon-vfpv4 -ftree-vectorize -mcpu=cortex-a5 -O3 -marm -mfpu=vfpv4 1. (CC) gcc options: -fomit-frame-pointer -fipa-pta -lm
Sudokut Total Time OpenBenchmarking.org Seconds, Fewer Is Better Sudokut 0.4 Total Time thunderx/vectorize, pre-patch A53 vectorize, pre-patch A53 vectorize/LTO, pre patch Cortex-A53 Cortex-A53 vectorize Cortex-A5 VFPV4 Cortex-A5 Neon Cortex-A5 NEON-vfpv4 GCC 4.9 NEON-vfpv4 GCC 4.9 VFPv4 40 80 120 160 200 SE +/- 0.76, N = 3 SE +/- 0.20, N = 3 SE +/- 0.21, N = 3 SE +/- 0.10, N = 3 SE +/- 0.09, N = 3 SE +/- 2.21, N = 6 SE +/- 0.34, N = 3 SE +/- 2.25, N = 6 SE +/- 2.79, N = 6 SE +/- 2.80, N = 6 102.75 101.95 101.75 101.88 102.17 167.97 165.14 167.41 169.11 170.29
Phoronix Test Suite v10.8.4