GCC 9.1 Compiler Tuning Threadripper AMD znver1 AMD Ryzen Threadripper 2990WX compiler benchmarks on GCC 9.1 with Ubuntu Linux by Michael Larabel. -O3 -march=native: Processor: AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH EXTREME (1701 BIOS), Chipset: AMD 17h, Memory: 32768MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: AMD Radeon RX 64 8GB (1590/800MHz), Audio: Realtek ALC1220, Monitor: ASUS VP28U, Network: Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad OS: Ubuntu 18.04, Kernel: 4.18.0-18-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: amdgpu 18.1.0, OpenGL: 4.5 Mesa 18.2.8 (LLVM 7.0.0), Compiler: GCC 9.1.0, File-System: ext4, Screen Resolution: 3840x2160 -O3 -march=athlon64-sse3: Processor: AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH EXTREME (1701 BIOS), Chipset: AMD 17h, Memory: 32768MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: AMD Radeon RX 64 8GB (1590/800MHz), Audio: Realtek ALC1220, Monitor: ASUS VP28U, Network: Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad OS: Ubuntu 18.04, Kernel: 4.18.0-18-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: amdgpu 18.1.0, OpenGL: 4.5 Mesa 18.2.8 (LLVM 7.0.0), Compiler: GCC 9.1.0, File-System: ext4, Screen Resolution: 3840x2160 -O3 -march=athlon64: Processor: AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH EXTREME (1701 BIOS), Chipset: AMD 17h, Memory: 32768MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: AMD Radeon RX 64 8GB (1590/800MHz), Audio: Realtek ALC1220, Monitor: ASUS VP28U, Network: Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad OS: Ubuntu 18.04, Kernel: 4.18.0-18-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: amdgpu 18.1.0, OpenGL: 4.5 Mesa 18.2.8 (LLVM 7.0.0), Compiler: GCC 9.1.0, File-System: ext4, Screen Resolution: 3840x2160 -O3 -march=native -flto: Processor: AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH EXTREME (1701 BIOS), Chipset: AMD 17h, Memory: 32768MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: AMD Radeon RX 64 8GB (1590/800MHz), Audio: Realtek ALC1220, Monitor: ASUS VP28U, Network: Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad OS: Ubuntu 18.04, Kernel: 4.18.0-18-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: amdgpu 18.1.0, OpenGL: 4.5 Mesa 18.2.8 (LLVM 7.0.0), Compiler: GCC 9.1.0, File-System: ext4, Screen Resolution: 3840x2160 -O2 -march=native: Processor: AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH EXTREME (1701 BIOS), Chipset: AMD 17h, Memory: 32768MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: AMD Radeon RX 64 8GB (1590/800MHz), Audio: Realtek ALC1220, Monitor: ASUS VP28U, Network: Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad OS: Ubuntu 18.04, Kernel: 4.18.0-18-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: amdgpu 18.1.0, OpenGL: 4.5 Mesa 18.2.8 (LLVM 7.0.0), Compiler: GCC 9.1.0, File-System: ext4, Screen Resolution: 3840x2160 -O2 -march=athlon64: Processor: AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH EXTREME (1701 BIOS), Chipset: AMD 17h, Memory: 32768MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: AMD Radeon RX 64 8GB (1590/800MHz), Audio: Realtek ALC1220, Monitor: ASUS VP28U, Network: Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad OS: Ubuntu 18.04, Kernel: 4.18.0-18-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: amdgpu 18.1.0, OpenGL: 4.5 Mesa 18.2.8 (LLVM 7.0.0), Compiler: GCC 9.1.0, File-System: ext4, Screen Resolution: 3840x2160 MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB MiB/s > Higher Is Better -O3 -march=native ........ 12851 |============================================ -O3 -march=athlon64-sse3 . 12699 |=========================================== -O3 -march=athlon64 ...... 12930 |============================================ -O3 -march=native -flto .. 13176 |============================================= -O2 -march=native ........ 12795 |============================================ -O2 -march=athlon64 ...... 12632 |=========================================== MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB MiB/s > Higher Is Better -O3 -march=native ........ 6685 |============================================= -O3 -march=athlon64-sse3 . 6681 |============================================= -O3 -march=athlon64 ...... 6755 |============================================== -O3 -march=native -flto .. 6721 |============================================== -O2 -march=native ........ 6778 |============================================== -O2 -march=athlon64 ...... 6526 |============================================ t-test1 2017-01-13 Threads: 1 Seconds < Lower Is Better -O3 -march=native ........ 26.23 |======================================== -O3 -march=athlon64-sse3 . 27.12 |========================================= -O3 -march=athlon64 ...... 29.66 |============================================= -O3 -march=native -flto .. 26.31 |======================================== -O2 -march=native ........ 26.07 |======================================== -O2 -march=athlon64 ...... 25.60 |======================================= t-test1 2017-01-13 Threads: 2 Seconds < Lower Is Better -O3 -march=native ........ 9.01 |=========================================== -O3 -march=athlon64-sse3 . 8.82 |========================================== -O3 -march=athlon64 ...... 9.59 |============================================== -O3 -march=native -flto .. 8.89 |=========================================== -O2 -march=native ........ 9.40 |============================================= -O2 -march=athlon64 ...... 8.68 |========================================== High Performance Conjugate Gradient 3.0 GFLOP/s > Higher Is Better -O3 -march=native ........ 0.83 |======================================= -O3 -march=athlon64-sse3 . 0.85 |======================================== -O3 -march=athlon64 ...... 0.92 |=========================================== -O3 -march=native -flto .. 0.98 |============================================== -O2 -march=native ........ 0.81 |====================================== -O2 -march=athlon64 ...... 0.89 |========================================== FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 Mflops > Higher Is Better -O3 -march=native ........ 6600 |=========================================== -O3 -march=athlon64-sse3 . 4463 |============================= -O3 -march=athlon64 ...... 4361 |============================= -O3 -march=native -flto .. 7019 |============================================== -O2 -march=native ........ 6335 |========================================== -O2 -march=athlon64 ...... 4342 |============================ FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 Mflops > Higher Is Better -O3 -march=native ....... 14927 |========================================== -O3 -march=native -flto . 14864 |========================================== -O2 -march=native ....... 16263 |============================================== Timed MAFFT Alignment 7.392 Multiple Sequence Alignment Seconds < Lower Is Better -O3 -march=native ........ 2.69 |============================================== -O3 -march=athlon64-sse3 . 2.55 |============================================ -O3 -march=athlon64 ...... 2.58 |============================================ -O3 -march=native -flto .. 2.65 |============================================= -O2 -march=native ........ 2.58 |============================================ -O2 -march=athlon64 ...... 2.60 |============================================ LuaJIT 2.1-git Test: Composite Mflops > Higher Is Better -O3 -march=native ........ 1495 |============================================== -O3 -march=athlon64-sse3 . 1496 |============================================== -O3 -march=athlon64 ...... 1489 |============================================== -O3 -march=native -flto .. 1497 |============================================== -O2 -march=native ........ 1467 |============================================= -O2 -march=athlon64 ...... 1491 |============================================== LuaJIT 2.1-git Test: Monte Carlo Mflops > Higher Is Better -O3 -march=native ........ 499 |=============================================== -O3 -march=athlon64-sse3 . 499 |=============================================== -O3 -march=athlon64 ...... 498 |=============================================== -O3 -march=native -flto .. 500 |=============================================== -O2 -march=native ........ 490 |============================================== -O2 -march=athlon64 ...... 498 |=============================================== LuaJIT 2.1-git Test: Fast Fourier Transform Mflops > Higher Is Better -O3 -march=native ........ 286 |=============================================== -O3 -march=athlon64-sse3 . 287 |=============================================== -O3 -march=athlon64 ...... 287 |=============================================== -O3 -march=native -flto .. 287 |=============================================== -O2 -march=native ........ 281 |============================================== -O2 -march=athlon64 ...... 286 |=============================================== LuaJIT 2.1-git Test: Sparse Matrix Multiply Mflops > Higher Is Better -O3 -march=native ........ 1208 |============================================== -O3 -march=athlon64-sse3 . 1207 |============================================== -O3 -march=athlon64 ...... 1200 |============================================== -O3 -march=native -flto .. 1204 |============================================== -O2 -march=native ........ 1182 |============================================= -O2 -march=athlon64 ...... 1183 |============================================= LuaJIT 2.1-git Test: Dense LU Matrix Factorization Mflops > Higher Is Better -O3 -march=native ........ 3611 |============================================== -O3 -march=athlon64-sse3 . 3619 |============================================== -O3 -march=athlon64 ...... 3590 |============================================== -O3 -march=native -flto .. 3624 |============================================== -O2 -march=native ........ 3550 |============================================= -O2 -march=athlon64 ...... 3623 |============================================== LuaJIT 2.1-git Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better -O3 -march=native ........ 1868 |============================================== -O3 -march=athlon64-sse3 . 1868 |============================================== -O3 -march=athlon64 ...... 1867 |============================================== -O3 -march=native -flto .. 1868 |============================================== -O2 -march=native ........ 1830 |============================================= -O2 -march=athlon64 ...... 1865 |============================================== SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better -O3 -march=native ........ 2514 |============================================= -O3 -march=athlon64-sse3 . 2039 |===================================== -O3 -march=athlon64 ...... 2021 |===================================== -O3 -march=native -flto .. 2543 |============================================== -O2 -march=native ........ 1981 |==================================== -O2 -march=athlon64 ...... 1782 |================================ SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better -O3 -march=native ........ 732 |================== -O3 -march=athlon64-sse3 . 737 |================== -O3 -march=athlon64 ...... 736 |================== -O3 -march=native -flto .. 1904 |============================================== -O2 -march=native ........ 721 |================= -O2 -march=athlon64 ...... 723 |================= SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better -O3 -march=native ........ 270 |=========================================== -O3 -march=athlon64-sse3 . 294 |=============================================== -O3 -march=athlon64 ...... 294 |=============================================== -O3 -march=native -flto .. 270 |=========================================== -O2 -march=native ........ 265 |========================================== -O2 -march=athlon64 ...... 291 |=============================================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better -O3 -march=native ........ 3174 |============================================== -O3 -march=athlon64-sse3 . 3082 |============================================= -O3 -march=athlon64 ...... 2874 |========================================== -O3 -march=native -flto .. 2951 |=========================================== -O2 -march=native ........ 3105 |============================================= -O2 -march=athlon64 ...... 3119 |============================================= SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better -O3 -march=native ........ 5989 |============================================== -O3 -march=athlon64-sse3 . 4239 |================================= -O3 -march=athlon64 ...... 4274 |================================= -O3 -march=native -flto .. 5388 |========================================= -O2 -march=native ........ 4507 |=================================== -O2 -march=athlon64 ...... 3593 |============================ SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better -O3 -march=native ........ 2218 |============================================== -O3 -march=athlon64-sse3 . 1842 |====================================== -O3 -march=athlon64 ...... 1842 |====================================== -O3 -march=native -flto .. 2202 |============================================== -O2 -march=native ........ 1306 |=========================== -O2 -march=athlon64 ...... 1186 |========================= TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better -O3 -march=native ........ 1116747 |========================================== -O3 -march=athlon64-sse3 . 1115841 |========================================== -O3 -march=athlon64 ...... 1115390 |========================================== -O3 -march=native -flto .. 1135626 |=========================================== -O2 -march=native ........ 1102013 |========================================== -O2 -march=athlon64 ...... 1094211 |========================================= MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: f32 ms < Lower Is Better -O3 -march=native ........ 71.26 |============================================ -O3 -march=athlon64-sse3 . 71.02 |============================================ -O3 -march=athlon64 ...... 70.94 |============================================ -O2 -march=native ........ 72.08 |============================================= -O2 -march=athlon64 ...... 71.98 |============================================= AOM AV1 2019-02-11 AV1 Video Encoding Frames Per Second > Higher Is Better -O3 -march=native ........ 0.22 |============================================== -O3 -march=athlon64-sse3 . 0.21 |============================================ -O3 -march=athlon64 ...... 0.20 |========================================== -O3 -march=native -flto .. 0.22 |============================================== -O2 -march=native ........ 0.21 |============================================ -O2 -march=athlon64 ...... 0.20 |========================================== SVT-AV1 2019-03-07 1080p 8-bit YUV To AV1 Video Encode Frames Per Second > Higher Is Better -O3 -march=native ........ 20.27 |============================================= -O3 -march=athlon64-sse3 . 18.66 |========================================= -O3 -march=athlon64 ...... 18.84 |========================================== -O3 -march=native -flto .. 20.41 |============================================= -O2 -march=native ........ 18.63 |========================================= -O2 -march=athlon64 ...... 18.51 |========================================= SVT-HEVC 2019-02-03 1080p 8-bit YUV To HEVC Video Encode Frames Per Second > Higher Is Better -O3 -march=native ........ 165 |============================================= -O3 -march=athlon64-sse3 . 166 |============================================= -O3 -march=athlon64 ...... 168 |============================================== -O3 -march=native -flto .. 165 |============================================= -O2 -march=native ........ 163 |============================================= -O2 -march=athlon64 ...... 172 |=============================================== SVT-VP9 2019-02-17 1080p 8-bit YUV To VP9 Video Encode Frames Per Second > Higher Is Better -O3 -march=native ........ 102.91 |=========================================== -O3 -march=athlon64-sse3 . 97.81 |========================================= -O3 -march=athlon64 ...... 103.45 |============================================ -O3 -march=native -flto .. 104.42 |============================================ -O2 -march=native ........ 100.82 |========================================== -O2 -march=athlon64 ...... 101.13 |=========================================== VP9 libvpx Encoding 1.8.0 vpxenc VP9 1080p Video Encode Frames Per Second > Higher Is Better -O3 -march=native ........ 26.37 |============================================= -O3 -march=athlon64-sse3 . 25.58 |============================================ -O3 -march=athlon64 ...... 26.12 |============================================= -O2 -march=native ........ 26.02 |============================================ -O2 -march=athlon64 ...... 25.44 |=========================================== x264 2018-09-25 H.264 Video Encoding Frames Per Second > Higher Is Better -O3 -march=native ........ 147 |=============================================== -O3 -march=athlon64-sse3 . 146 |=============================================== -O3 -march=athlon64 ...... 146 |=============================================== -O2 -march=native ........ 143 |============================================== -O2 -march=athlon64 ...... 146 |=============================================== x265 3.0 H.265 1080p Video Encoding Frames Per Second > Higher Is Better -O3 -march=native ........ 33.76 |============================================= -O3 -march=athlon64-sse3 . 33.88 |============================================= -O3 -march=athlon64 ...... 33.89 |============================================= -O3 -march=native -flto .. 33.68 |============================================= -O2 -march=native ........ 33.44 |============================================ -O2 -march=athlon64 ...... 33.53 |============================================= GraphicsMagick 1.3.30 Operation: Swirl Iterations Per Minute > Higher Is Better -O3 -march=native ........ 247 |============================================== -O3 -march=athlon64-sse3 . 237 |============================================= -O3 -march=athlon64 ...... 237 |============================================= -O3 -march=native -flto .. 250 |=============================================== -O2 -march=native ........ 245 |============================================== -O2 -march=athlon64 ...... 244 |============================================== GraphicsMagick 1.3.30 Operation: Rotate Iterations Per Minute > Higher Is Better -O3 -march=native ........ 249 |=============================================== -O3 -march=athlon64-sse3 . 240 |============================================= -O3 -march=athlon64 ...... 238 |============================================= -O3 -march=native -flto .. 248 |=============================================== -O2 -march=native ........ 249 |=============================================== -O2 -march=athlon64 ...... 243 |============================================== GraphicsMagick 1.3.30 Operation: Sharpen Iterations Per Minute > Higher Is Better -O3 -march=native ........ 219 |=============================================== -O3 -march=athlon64-sse3 . 191 |========================================= -O3 -march=athlon64 ...... 190 |======================================== -O3 -march=native -flto .. 221 |=============================================== -O2 -march=native ........ 217 |============================================== -O2 -march=athlon64 ...... 192 |========================================= GraphicsMagick 1.3.30 Operation: Enhanced Iterations Per Minute > Higher Is Better -O3 -march=native ........ 232 |=============================================== -O3 -march=athlon64-sse3 . 194 |======================================= -O3 -march=athlon64 ...... 195 |======================================= -O3 -march=native -flto .. 233 |=============================================== -O2 -march=native ........ 231 |=============================================== -O2 -march=athlon64 ...... 198 |======================================== GraphicsMagick 1.3.30 Operation: Resizing Iterations Per Minute > Higher Is Better -O3 -march=native ........ 243 |=============================================== -O3 -march=athlon64-sse3 . 231 |============================================= -O3 -march=athlon64 ...... 233 |============================================= -O3 -march=native -flto .. 240 |============================================== -O2 -march=native ........ 238 |============================================== -O2 -march=athlon64 ...... 235 |============================================= GraphicsMagick 1.3.30 Operation: Noise-Gaussian Iterations Per Minute > Higher Is Better -O3 -march=native ........ 204 |=============================================== -O3 -march=athlon64-sse3 . 198 |============================================== -O3 -march=athlon64 ...... 199 |============================================== -O3 -march=native -flto .. 203 |=============================================== -O2 -march=native ........ 200 |============================================== -O2 -march=athlon64 ...... 202 |=============================================== GraphicsMagick 1.3.30 Operation: HWB Color Space Iterations Per Minute > Higher Is Better -O3 -march=native ........ 272 |=============================================== -O3 -march=athlon64-sse3 . 263 |============================================= -O3 -march=athlon64 ...... 261 |============================================= -O3 -march=native -flto .. 274 |=============================================== -O2 -march=native ........ 271 |============================================== -O2 -march=athlon64 ...... 272 |=============================================== Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better -O3 -march=native ........ 1313 |============================================= -O3 -march=athlon64-sse3 . 1316 |============================================== -O3 -march=athlon64 ...... 1316 |============================================== -O3 -march=native -flto .. 1304 |============================================= -O2 -march=native ........ 1319 |============================================== -O2 -march=athlon64 ...... 1328 |============================================== Stockfish 9 Total Time Nodes Per Second > Higher Is Better -O3 -march=native ........ 68200164 |========================================== -O3 -march=athlon64-sse3 . 67571150 |========================================== -O3 -march=athlon64 ...... 67513602 |========================================== -O3 -march=native -flto .. 67450689 |========================================== -O2 -march=native ........ 66697487 |========================================= -O2 -march=athlon64 ...... 66890687 |========================================= Timed ImageMagick Compilation 6.9.0 Time To Compile Seconds < Lower Is Better -O3 -march=native ........ 19.15 |========== -O3 -march=athlon64-sse3 . 19.73 |========== -O3 -march=athlon64 ...... 19.77 |========== -O3 -march=native -flto .. 88.89 |============================================= -O2 -march=native ........ 17.58 |========= -O2 -march=athlon64 ...... 17.41 |========= Timed LLVM Compilation 6.0.1 Time To Compile Seconds < Lower Is Better -O3 -march=native ........ 221 |=========== -O3 -march=athlon64-sse3 . 224 |=========== -O3 -march=athlon64 ...... 225 |=========== -O3 -march=native -flto .. 951 |=============================================== -O2 -march=native ........ 221 |=========== -O2 -march=athlon64 ...... 229 |=========== Timed PHP Compilation 7.1.9 Time To Compile Seconds < Lower Is Better -O3 -march=native ........ 63.25 |============================================= -O3 -march=athlon64-sse3 . 62.93 |============================================= -O3 -march=athlon64 ...... 63.19 |============================================= -O2 -march=native ........ 44.44 |================================ -O2 -march=athlon64 ...... 43.86 |=============================== C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Seconds < Lower Is Better -O3 -march=native ........ 18.00 |================== -O3 -march=athlon64-sse3 . 29.53 |============================= -O3 -march=athlon64 ...... 29.39 |============================= -O3 -march=native -flto .. 17.85 |================== -O2 -march=native ........ 34.22 |================================== -O2 -march=athlon64 ...... 45.43 |============================================= Smallpt 1.0 Global Illumination Renderer; 128 Samples Seconds < Lower Is Better -O3 -march=native ........ 3.87 |======================================= -O3 -march=athlon64-sse3 . 4.54 |============================================== -O3 -march=athlon64 ...... 4.51 |============================================= -O3 -march=native -flto .. 3.85 |======================================= -O2 -march=native ........ 3.89 |======================================= -O2 -march=athlon64 ...... 4.57 |============================================== AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better -O3 -march=native ........ 39.11 |===================================== -O3 -march=athlon64-sse3 . 44.98 |=========================================== -O3 -march=athlon64 ...... 45.20 |=========================================== -O3 -march=native -flto .. 39.92 |====================================== -O2 -march=native ........ 42.34 |======================================== -O2 -march=athlon64 ...... 47.28 |============================================= Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better -O3 -march=native ........ 2.35 |=========================================== -O3 -march=athlon64-sse3 . 2.46 |============================================= -O3 -march=athlon64 ...... 2.46 |============================================= -O3 -march=native -flto .. 2.30 |========================================== -O2 -march=native ........ 2.38 |============================================ -O2 -march=athlon64 ...... 2.49 |============================================== Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better -O3 -march=native ........ 3.84 |============================================ -O3 -march=athlon64-sse3 . 3.91 |============================================= -O3 -march=athlon64 ...... 3.92 |============================================= -O3 -march=native -flto .. 3.97 |============================================== -O2 -march=native ........ 3.95 |============================================== -O2 -march=athlon64 ...... 3.96 |============================================== Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better -O3 -march=native ........ 4.42 |========================================== -O3 -march=athlon64-sse3 . 4.62 |============================================ -O3 -march=athlon64 ...... 4.65 |============================================ -O3 -march=native -flto .. 4.82 |============================================== -O2 -march=native ........ 4.57 |============================================ -O2 -march=athlon64 ...... 4.68 |============================================= Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better -O3 -march=native ........ 3.97 |========================================= -O3 -march=athlon64-sse3 . 4.45 |============================================= -O3 -march=athlon64 ...... 4.45 |============================================= -O3 -march=native -flto .. 3.90 |======================================== -O2 -march=native ........ 4.06 |========================================== -O2 -march=athlon64 ...... 4.50 |============================================== Bullet Physics Engine 2.81 Test: 136 Ragdolls Seconds < Lower Is Better -O3 -march=native ........ 2.30 |=========================================== -O3 -march=athlon64-sse3 . 2.35 |============================================ -O3 -march=athlon64 ...... 2.36 |============================================ -O3 -march=native -flto .. 2.44 |============================================== -O2 -march=native ........ 2.39 |============================================= -O2 -march=athlon64 ...... 2.42 |============================================== Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better -O3 -march=native ........ 0.84 |============================================ -O3 -march=athlon64-sse3 . 0.86 |============================================= -O3 -march=athlon64 ...... 0.86 |============================================= -O3 -march=native -flto .. 0.82 |=========================================== -O2 -march=native ........ 0.84 |============================================ -O2 -march=athlon64 ...... 0.87 |============================================== Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better -O3 -march=native ........ 1.00 |=========================================== -O3 -march=athlon64-sse3 . 1.06 |============================================== -O3 -march=athlon64 ...... 1.06 |============================================== -O3 -march=native -flto .. 0.97 |========================================== -O2 -march=native ........ 1.02 |============================================ -O2 -march=athlon64 ...... 1.07 |============================================== XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 Seconds < Lower Is Better -O3 -march=native ........ 25.70 |=========================================== -O3 -march=athlon64-sse3 . 25.93 |============================================ -O3 -march=athlon64 ...... 26.50 |============================================= -O3 -march=native -flto .. 25.62 |=========================================== -O2 -march=native ........ 26.78 |============================================= -O2 -march=athlon64 ...... 26.75 |============================================= Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Seconds < Lower Is Better -O3 -march=native ........ 19.09 |============================================ -O3 -march=athlon64-sse3 . 18.38 |=========================================== -O3 -march=athlon64 ...... 17.74 |========================================= -O3 -march=native -flto .. 17.94 |========================================== -O2 -march=native ........ 19.31 |============================================= -O2 -march=athlon64 ...... 18.02 |========================================== FLAC Audio Encoding 1.3.2 WAV To FLAC Seconds < Lower Is Better -O3 -march=native ........ 9.53 |============================ -O3 -march=athlon64-sse3 . 15.43 |============================================= -O3 -march=athlon64 ...... 15.48 |============================================= -O3 -march=native -flto .. 9.48 |=========================== -O2 -march=native ........ 9.78 |============================ -O2 -march=athlon64 ...... 15.58 |============================================= LAME MP3 Encoding 3.100 WAV To MP3 Seconds < Lower Is Better -O3 -march=native ........ 8.00 |================================ -O3 -march=athlon64-sse3 . 9.36 |===================================== -O3 -march=athlon64 ...... 9.43 |===================================== -O3 -march=native -flto .. 7.98 |=============================== -O2 -march=native ........ 10.80 |=========================================== -O2 -march=athlon64 ...... 11.41 |============================================= OpenSSL 1.1.1 RSA 4096-bit Performance Signs Per Second > Higher Is Better -O3 -march=native ........ 5825 |============================================== -O3 -march=athlon64-sse3 . 5838 |============================================== -O3 -march=athlon64 ...... 5837 |============================================== -O3 -march=native -flto .. 5833 |============================================== -O2 -march=native ........ 5831 |============================================== -O2 -march=athlon64 ...... 5832 |============================================== PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only TPS > Higher Is Better -O3 -march=native ........ 466723 |=========================================== -O3 -march=athlon64-sse3 . 459596 |=========================================== -O3 -march=athlon64 ...... 463840 |=========================================== -O3 -march=native -flto .. 473620 |============================================ -O2 -march=native ........ 458551 |=========================================== -O2 -march=athlon64 ...... 453605 |========================================== PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write TPS > Higher Is Better -O3 -march=native ........ 16339 |============================================= -O3 -march=athlon64-sse3 . 16252 |============================================= -O3 -march=athlon64 ...... 15616 |=========================================== -O3 -march=native -flto .. 14989 |========================================= -O2 -march=native ........ 13558 |===================================== -O2 -march=athlon64 ...... 5098 |============== CppPerformanceBenchmarks 9 Test: Atol Seconds < Lower Is Better -O3 -march=native ........ 69.03 |============================================= -O3 -march=athlon64-sse3 . 68.42 |============================================ -O3 -march=athlon64 ...... 68.37 |============================================ -O3 -march=native -flto .. 69.25 |============================================= -O2 -march=native ........ 69.80 |============================================= -O2 -march=athlon64 ...... 69.08 |============================================= CppPerformanceBenchmarks 9 Test: Ctype Seconds < Lower Is Better -O3 -march=native ........ 33.88 |============================================ -O3 -march=athlon64-sse3 . 33.31 |=========================================== -O3 -march=athlon64 ...... 33.28 |=========================================== -O3 -march=native -flto .. 32.32 |========================================== -O2 -march=native ........ 34.50 |============================================= -O2 -march=athlon64 ...... 33.37 |============================================ CppPerformanceBenchmarks 9 Test: Math Library Seconds < Lower Is Better -O3 -march=native ........ 351 |========================================= -O3 -march=athlon64-sse3 . 399 |=============================================== -O3 -march=athlon64 ...... 398 |=============================================== -O3 -march=native -flto .. 352 |========================================= -O2 -march=native ........ 356 |========================================== -O2 -march=athlon64 ...... 399 |=============================================== CppPerformanceBenchmarks 9 Test: Random Numbers Seconds < Lower Is Better -O3 -march=native ........ 1023 |============================================ -O3 -march=athlon64-sse3 . 1055 |============================================= -O3 -march=athlon64 ...... 1070 |============================================== -O3 -march=native -flto .. 1011 |=========================================== -O2 -march=native ........ 1041 |============================================= -O2 -march=athlon64 ...... 1072 |============================================== CppPerformanceBenchmarks 9 Test: Stepanov Vector Seconds < Lower Is Better -O3 -march=native ........ 74.93 |============================================ -O3 -march=athlon64-sse3 . 75.56 |============================================= -O3 -march=athlon64 ...... 75.53 |============================================= -O3 -march=native -flto .. 74.92 |============================================ -O2 -march=native ........ 75.99 |============================================= -O2 -march=athlon64 ...... 75.79 |============================================= CppPerformanceBenchmarks 9 Test: Function Objects Seconds < Lower Is Better -O3 -march=native ........ 15.40 |=========================================== -O3 -march=athlon64-sse3 . 16.14 |============================================= -O3 -march=athlon64 ...... 16.17 |============================================= -O3 -march=native -flto .. 15.50 |=========================================== -O2 -march=native ........ 15.61 |=========================================== -O2 -march=athlon64 ...... 16.20 |============================================= CppPerformanceBenchmarks 9 Test: Stepanov Abstraction Seconds < Lower Is Better -O3 -march=native ........ 28.22 |============================================ -O3 -march=athlon64-sse3 . 28.28 |============================================ -O3 -march=athlon64 ...... 28.26 |============================================ -O3 -march=native -flto .. 28.25 |============================================ -O2 -march=native ........ 28.69 |============================================= -O2 -march=athlon64 ...... 28.33 |============================================ Redis 4.0.8 Test: LPOP Requests Per Second > Higher Is Better -O3 -march=native ........ 2573823 |========================================== -O3 -march=athlon64-sse3 . 2506276 |========================================= -O3 -march=athlon64 ...... 2544762 |========================================= -O3 -march=native -flto .. 2616703 |========================================== -O2 -march=native ........ 2340595 |====================================== -O2 -march=athlon64 ...... 2654025 |=========================================== Redis 4.0.8 Test: SADD Requests Per Second > Higher Is Better -O3 -march=native ........ 2046557 |========================================== -O3 -march=athlon64-sse3 . 2032921 |========================================== -O3 -march=athlon64 ...... 2051129 |========================================== -O3 -march=native -flto .. 2084089 |=========================================== -O2 -march=native ........ 2000709 |========================================= -O2 -march=athlon64 ...... 1975322 |========================================= Redis 4.0.8 Test: LPUSH Requests Per Second > Higher Is Better -O3 -march=native ........ 1540257 |=========================================== -O3 -march=athlon64-sse3 . 1522952 |========================================== -O3 -march=athlon64 ...... 1520276 |========================================== -O3 -march=native -flto .. 1549647 |=========================================== -O2 -march=native ........ 1463388 |========================================= -O2 -march=athlon64 ...... 1548556 |=========================================== Redis 4.0.8 Test: GET Requests Per Second > Higher Is Better -O3 -march=native ........ 2502182 |========================================== -O3 -march=athlon64-sse3 . 2461725 |========================================== -O3 -march=athlon64 ...... 2540398 |=========================================== -O3 -march=native -flto .. 2509433 |========================================== -O2 -march=native ........ 2275019 |======================================= -O2 -march=athlon64 ...... 2384143 |======================================== Redis 4.0.8 Test: SET Requests Per Second > Higher Is Better -O3 -march=native ........ 1807270 |========================================== -O3 -march=athlon64-sse3 . 1755590 |========================================= -O3 -march=athlon64 ...... 1840806 |=========================================== -O3 -march=native -flto .. 1755472 |========================================= -O2 -march=native ........ 1744979 |========================================= -O2 -march=athlon64 ...... 1730702 |======================================== ctx_clock Context Switch Time Clocks < Lower Is Better -O3 -march=native ........ 150 |=============================================== -O3 -march=athlon64-sse3 . 150 |=============================================== -O3 -march=athlon64 ...... 150 |=============================================== -O3 -march=native -flto .. 150 |=============================================== -O2 -march=native ........ 150 |=============================================== -O2 -march=athlon64 ...... 150 |=============================================== Memcached mcperf 1.5.10 Method: Add Operations Per Second > Higher Is Better -O3 -march=native ........ 44903 |===================================== -O3 -march=athlon64-sse3 . 34822 |============================= -O3 -march=athlon64 ...... 53106 |============================================ -O3 -march=native -flto .. 50055 |========================================= -O2 -march=native ........ 34138 |============================ -O2 -march=athlon64 ...... 54600 |============================================= Memcached mcperf 1.5.10 Method: Get Operations Per Second > Higher Is Better -O3 -march=native ........ 57652 |====================================== -O3 -march=athlon64-sse3 . 69004 |============================================= -O3 -march=athlon64 ...... 56837 |===================================== -O3 -march=native -flto .. 68644 |============================================= -O2 -march=native ........ 55791 |==================================== -O2 -march=athlon64 ...... 55647 |==================================== Memcached mcperf 1.5.10 Method: Set Operations Per Second > Higher Is Better -O3 -march=native ........ 38646 |============================= -O3 -march=athlon64-sse3 . 48548 |==================================== -O3 -march=athlon64 ...... 34880 |========================== -O3 -march=native -flto .. 43570 |================================= -O2 -march=native ........ 34446 |========================== -O2 -march=athlon64 ...... 59924 |============================================= Memcached mcperf 1.5.10 Method: Append Operations Per Second > Higher Is Better -O3 -march=native ........ 42455 |=============================== -O3 -march=athlon64-sse3 . 45716 |================================== -O3 -march=athlon64 ...... 56172 |========================================= -O3 -march=native -flto .. 45339 |================================= -O2 -march=native ........ 35058 |========================== -O2 -march=athlon64 ...... 61219 |============================================= Memcached mcperf 1.5.10 Method: Delete Operations Per Second > Higher Is Better -O3 -march=native ........ 58969 |====================================== -O3 -march=athlon64-sse3 . 69509 |============================================= -O3 -march=athlon64 ...... 56822 |===================================== -O3 -march=native -flto .. 68696 |============================================ -O2 -march=native ........ 56141 |==================================== -O2 -march=athlon64 ...... 55747 |==================================== Memcached mcperf 1.5.10 Method: Prepend Operations Per Second > Higher Is Better -O3 -march=native ........ 43824 |=========================================== -O3 -march=athlon64-sse3 . 45691 |============================================= -O3 -march=athlon64 ...... 35968 |=================================== -O3 -march=native -flto .. 45587 |============================================= -O2 -march=native ........ 35552 |=================================== -O2 -march=athlon64 ...... 35467 |=================================== Memcached mcperf 1.5.10 Method: Replace Operations Per Second > Higher Is Better -O3 -march=native ........ 53865 |============================================= -O3 -march=athlon64-sse3 . 45707 |====================================== -O3 -march=athlon64 ...... 36231 |============================== -O3 -march=native -flto .. 45591 |====================================== -O2 -march=native ........ 35486 |============================== -O2 -march=athlon64 ...... 35829 |============================== NGINX Benchmark 1.9.9 Static Web Page Serving Requests Per Second > Higher Is Better -O3 -march=native ........ 29274 |============================================ -O3 -march=athlon64-sse3 . 29281 |============================================ -O3 -march=athlon64 ...... 29726 |============================================= -O3 -march=native -flto .. 27352 |========================================= -O2 -march=native ........ 27834 |========================================== -O2 -march=athlon64 ...... 29704 |=============================================