GCC Compiler Intel Sandy Bridge AVX fpmath GCC testing for a future article on Phoronix.com. Testing with CFLAGS/CXXFLAGS set to -O3 and -march=native. Benchmarks of GCC 4.7 RC1 when comparing the effects of building the GNU Compiler Collection with the --with-fpmath=avx to see how GCC 4.7 is impacted by the AVX floating-point arithmetic. Stock: Processor: Intel Core i7-3960X @ 3.20GHz (12 Cores), Motherboard: Intel DX79SI, Chipset: Intel Sandy DMI2, Memory: 16384MB, Disk: 240GB OCZ VERTEX3, Graphics: AMD Radeon HD 5700 1024MB, Audio: Realtek ALC892, Monitor: DELL P2210H, Network: Intel 82579LM Gigabit Connection OS: Ubuntu 12.04, Kernel: 3.2.0-18-generic (x86_64), Desktop: Unity 5.4.0, Display Server: X Server 1.11.3, Display Driver: radeon 6.14.99, OpenGL: 2.1 Mesa 8.0.1 Gallium 0.4, Compiler: GCC 4.7.0, File-System: ext4, Screen Resolution: 1920x1080 AVX fpmath: Processor: Intel Core i7-3960X @ 3.20GHz (12 Cores), Motherboard: Intel DX79SI, Chipset: Intel Sandy DMI2, Memory: 16384MB, Disk: 240GB OCZ VERTEX3, Graphics: AMD Radeon HD 5700 1024MB, Audio: Realtek ALC892, Monitor: DELL P2210H, Network: Intel 82579LM Gigabit Connection OS: Ubuntu 12.04, Kernel: 3.2.0-18-generic (x86_64), Desktop: Unity 5.4.0, Display Server: X Server 1.11.3, Display Driver: radeon 6.14.99, OpenGL: 2.1 Mesa 8.0.1 Gallium 0.4, Compiler: GCC 4.7.0, File-System: ext4, Screen Resolution: 1920x1080 C-Ray 1.1 Total Time Seconds < Lower Is Better Stock ...... 25.70 |=========================================================== AVX fpmath . 25.74 |=========================================================== Smallpt 1.0 Global Illumination Renderer; 100 Samples Seconds < Lower Is Better Stock ...... 20 |============================================================== AVX fpmath . 20 |============================================================== BYTE Unix Benchmark 3.6 Computational Test: Dhrystone 2 LPS > Higher Is Better Stock ...... 21255660.73 |===================================================== AVX fpmath . 21254777.57 |===================================================== Fhourstones 3.1 Complex Connect-4 Solving Kpos / sec > Higher Is Better Stock ...... 10353.83 |======================================================== AVX fpmath . 10235.50 |======================================================= CLOMP 3.3 Static OMP Speedup Speedup > Higher Is Better Stock ...... 6.38 |============================================================ AVX fpmath . 6.34 |============================================================ LZMA Compression 256MB File Compression Seconds < Lower Is Better Stock ...... 155.14 |========================================================== AVX fpmath . 155.14 |========================================================== Ogg Encoding 1.3.0 WAV To Ogg Seconds < Lower Is Better Stock ...... 9.43 |============================================================ AVX fpmath . 9.44 |============================================================ FLAC Audio Encoding 1.2.1 WAV To FLAC Seconds < Lower Is Better Stock ...... 6.29 |=========================================================== AVX fpmath . 6.38 |============================================================ FFmpeg 0.10 AVI To NTSC VCD Seconds < Lower Is Better Stock ...... 12.34 |=========================================================== AVX fpmath . 12.31 |=========================================================== x264 2011-12-06 H.264 Video Encoding Frames Per Second > Higher Is Better Stock ...... 171.68 |========================================================== AVX fpmath . 172.89 |========================================================== VP8 libvpx Encoding 0.9.7-p1 vpxenc Frames Per Second > Higher Is Better Stock ...... 23.58 |=========================================================== AVX fpmath . 23.42 |=========================================================== OpenSSL 1.0.0e RSA 4096-bit Performance Signs Per Second > Higher Is Better Stock ...... 75.45 |=========================================================== AVX fpmath . 75.45 |=========================================================== GMPbench 0.2 Total Time GMPbench Score > Higher Is Better Stock ...... 2833.40 |========================================================= AVX fpmath . 2834.00 |========================================================= Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better Stock ...... 9.25 |============================================================ AVX fpmath . 9.22 |============================================================ Timed MAFFT Alignment 6.864 Multiple Sequence Alignment Seconds < Lower Is Better Stock ...... 4.89 |========================================================== AVX fpmath . 5.06 |============================================================ Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better Stock ...... 1377.66 |========================================================= AVX fpmath . 1378.74 |========================================================= Minion 0.12 Benchmark: Bibd Seconds < Lower Is Better Stock ...... 161.22 |========================================================= AVX fpmath . 163.42 |========================================================== Minion 0.12 Benchmark: Graceful Seconds < Lower Is Better Stock ...... 90.02 |=========================================================== AVX fpmath . 89.23 |========================================================== Minion 0.12 Benchmark: Quasigroup Seconds < Lower Is Better Stock ...... 188.09 |========================================================== AVX fpmath . 189.65 |========================================================== Minion 0.12 Benchmark: Solitaire Seconds < Lower Is Better Stock ...... 145.82 |========================================================= AVX fpmath . 148.53 |========================================================== Open FMM Nero2D 2.0.2 Total Time Seconds < Lower Is Better Stock ...... 534.09 |========================================================== AVX fpmath . 532.10 |========================================================== GraphicsMagick 1.3.12 Operation: HWB Color Space Iterations Per Minute > Higher Is Better Stock ...... 163 |============================================================= AVX fpmath . 163 |============================================================= GraphicsMagick 1.3.12 Operation: Blur Iterations Per Minute > Higher Is Better Stock ...... 123 |============================================================= AVX fpmath . 123 |============================================================= GraphicsMagick 1.3.12 Operation: Local Adaptive Thresholding Iterations Per Minute > Higher Is Better Stock ...... 85 |============================================================== AVX fpmath . 85 |============================================================== GraphicsMagick 1.3.12 Operation: Resizing Iterations Per Minute > Higher Is Better Stock ...... 151 |============================================================= AVX fpmath . 151 |============================================================= GraphicsMagick 1.3.12 Operation: Sharpen Iterations Per Minute > Higher Is Better Stock ...... 98 |============================================================== AVX fpmath . 98 |============================================================== Apache Benchmark 2.2.21 Static Web Page Serving Requests Per Second > Higher Is Better Stock ...... 16263.52 |======================================================= AVX fpmath . 16427.85 |======================================================== TTSIOD 3D Renderer 2.2w Phong Rendering With Soft-Shadow Mapping FPS > Higher Is Better Stock ...... 158.67 |========================================================== AVX fpmath . 158.95 |========================================================== NAS Parallel Benchmarks 3.3 Test / Class: BT.A Total Mop/s > Higher Is Better Stock ...... 17009.49 |=================================================== AVX fpmath . 18631.50 |======================================================== NAS Parallel Benchmarks 3.3 Test / Class: FT.B Total Mop/s > Higher Is Better Stock ...... 10069.85 |===================================================== AVX fpmath . 10613.43 |======================================================== NAS Parallel Benchmarks 3.3 Test / Class: LU.A Total Mop/s > Higher Is Better Stock ...... 16396.35 |====================================================== AVX fpmath . 16999.52 |======================================================== NAS Parallel Benchmarks 3.3 Test / Class: MG.B Total Mop/s > Higher Is Better Stock ...... 10409.75 |======================================================== AVX fpmath . 10408.69 |======================================================== NAS Parallel Benchmarks 3.3 Test / Class: SP.A Total Mop/s > Higher Is Better Stock ...... 9410.41 |======================================================== AVX fpmath . 9625.70 |========================================================= NAS Parallel Benchmarks 3.3 Test / Class: UA.A Total Mop/s > Higher Is Better Stock ...... 73.66 |========================================================== AVX fpmath . 74.75 |=========================================================== Timed Linux Kernel Compilation 3.1 Time To Compile Seconds < Lower Is Better Stock ...... 70.92 |=========================================================== AVX fpmath . 70.92 |=========================================================== Timed PHP Compilation 5.2.9 Time To Compile Seconds < Lower Is Better Stock ...... 24.08 |=========================================================== AVX fpmath . 24.17 |===========================================================