GCC 9.1 Compiler Tuning Threadripper AMD znver1 AMD Ryzen Threadripper 2990WX compiler benchmarks on GCC 9.1 with Ubuntu Linux by Michael Larabel. -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 -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=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 -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 -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=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 PGO: 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 AMD Ryzen Threadripper 2990WX 32-Core: 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 -O2 -march=athlon64 ................... 12632 |=============================== -O3 -march=athlon64 ................... 12930 |=============================== -O3 -march=athlon64-sse3 .............. 12699 |=============================== -O2 -march=native ..................... 12795 |=============================== -O3 -march=native ..................... 12851 |=============================== -O3 -march=native -flto ............... 13176 |================================ PGO ................................... 12406 |============================== AMD Ryzen Threadripper 2990WX 32-Core . 12618 |=============================== MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB MiB/s > Higher Is Better -O2 -march=athlon64 ................... 6526 |================================ -O3 -march=athlon64 ................... 6755 |================================= -O3 -march=athlon64-sse3 .............. 6681 |================================= -O2 -march=native ..................... 6778 |================================= -O3 -march=native ..................... 6685 |================================= -O3 -march=native -flto ............... 6721 |================================= PGO ................................... 6753 |================================= AMD Ryzen Threadripper 2990WX 32-Core . 6596 |================================ t-test1 2017-01-13 Threads: 1 Seconds < Lower Is Better -O2 -march=athlon64 ................... 25.60 |============================ -O3 -march=athlon64 ................... 29.66 |================================ -O3 -march=athlon64-sse3 .............. 27.12 |============================= -O2 -march=native ..................... 26.07 |============================ -O3 -march=native ..................... 26.23 |============================ -O3 -march=native -flto ............... 26.31 |============================ PGO ................................... 28.74 |=============================== AMD Ryzen Threadripper 2990WX 32-Core . 27.83 |============================== t-test1 2017-01-13 Threads: 2 Seconds < Lower Is Better -O2 -march=athlon64 ................... 8.68 |==================== -O3 -march=athlon64 ................... 9.59 |====================== -O3 -march=athlon64-sse3 .............. 8.82 |==================== -O2 -march=native ..................... 9.40 |===================== -O3 -march=native ..................... 9.01 |===================== -O3 -march=native -flto ............... 8.89 |==================== PGO ................................... 9.41 |===================== AMD Ryzen Threadripper 2990WX 32-Core . 14.05 |================================ High Performance Conjugate Gradient 3.0 GFLOP/s > Higher Is Better -O2 -march=athlon64 ................... 0.89 |============================== -O3 -march=athlon64 ................... 0.92 |=============================== -O3 -march=athlon64-sse3 .............. 0.85 |============================= -O2 -march=native ..................... 0.81 |=========================== -O3 -march=native ..................... 0.83 |============================ -O3 -march=native -flto ............... 0.98 |================================= PGO ................................... 0.91 |=============================== AMD Ryzen Threadripper 2990WX 32-Core . 0.90 |============================== FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 Mflops > Higher Is Better -O2 -march=athlon64 ................... 4342 |==================== -O3 -march=athlon64 ................... 4361 |===================== -O3 -march=athlon64-sse3 .............. 4463 |===================== -O2 -march=native ..................... 6335 |============================== -O3 -march=native ..................... 6600 |=============================== -O3 -march=native -flto ............... 7019 |================================= PGO ................................... 6717 |================================ AMD Ryzen Threadripper 2990WX 32-Core . 6411 |============================== FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 Mflops > Higher Is Better -O2 -march=native ..................... 16263 |================================ -O3 -march=native ..................... 14927 |============================= -O3 -march=native -flto ............... 14864 |============================= PGO ................................... 15287 |============================== AMD Ryzen Threadripper 2990WX 32-Core . 15098 |============================== Timed MAFFT Alignment 7.392 Multiple Sequence Alignment Seconds < Lower Is Better -O2 -march=athlon64 ................... 2.60 |=============================== -O3 -march=athlon64 ................... 2.58 |=============================== -O3 -march=athlon64-sse3 .............. 2.55 |=============================== -O2 -march=native ..................... 2.58 |=============================== -O3 -march=native ..................... 2.69 |================================ -O3 -march=native -flto ............... 2.65 |================================ PGO ................................... 2.63 |================================ AMD Ryzen Threadripper 2990WX 32-Core . 2.74 |================================= LuaJIT 2.1-git Test: Composite Mflops > Higher Is Better -O2 -march=athlon64 ...... 1491 |============================================== -O3 -march=athlon64 ...... 1489 |============================================== -O3 -march=athlon64-sse3 . 1496 |============================================== -O2 -march=native ........ 1467 |============================================= -O3 -march=native ........ 1495 |============================================== -O3 -march=native -flto .. 1497 |============================================== PGO ...................... 1499 |============================================== LuaJIT 2.1-git Test: Monte Carlo Mflops > Higher Is Better -O2 -march=athlon64 ...... 498 |=============================================== -O3 -march=athlon64 ...... 498 |=============================================== -O3 -march=athlon64-sse3 . 499 |=============================================== -O2 -march=native ........ 490 |============================================== -O3 -march=native ........ 499 |=============================================== -O3 -march=native -flto .. 500 |=============================================== PGO ...................... 499 |=============================================== LuaJIT 2.1-git Test: Fast Fourier Transform Mflops > Higher Is Better -O2 -march=athlon64 ...... 286 |=============================================== -O3 -march=athlon64 ...... 287 |=============================================== -O3 -march=athlon64-sse3 . 287 |=============================================== -O2 -march=native ........ 281 |============================================== -O3 -march=native ........ 286 |=============================================== -O3 -march=native -flto .. 287 |=============================================== PGO ...................... 286 |=============================================== LuaJIT 2.1-git Test: Sparse Matrix Multiply Mflops > Higher Is Better -O2 -march=athlon64 ...... 1183 |============================================= -O3 -march=athlon64 ...... 1200 |============================================== -O3 -march=athlon64-sse3 . 1207 |============================================== -O2 -march=native ........ 1182 |============================================= -O3 -march=native ........ 1208 |============================================== -O3 -march=native -flto .. 1204 |============================================== PGO ...................... 1203 |============================================== LuaJIT 2.1-git Test: Dense LU Matrix Factorization Mflops > Higher Is Better -O2 -march=athlon64 ...... 3623 |============================================== -O3 -march=athlon64 ...... 3590 |============================================= -O3 -march=athlon64-sse3 . 3619 |============================================== -O2 -march=native ........ 3550 |============================================= -O3 -march=native ........ 3611 |============================================== -O3 -march=native -flto .. 3624 |============================================== PGO ...................... 3636 |============================================== LuaJIT 2.1-git Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better -O2 -march=athlon64 ...... 1865 |============================================== -O3 -march=athlon64 ...... 1867 |============================================== -O3 -march=athlon64-sse3 . 1868 |============================================== -O2 -march=native ........ 1830 |============================================= -O3 -march=native ........ 1868 |============================================== -O3 -march=native -flto .. 1868 |============================================== PGO ...................... 1870 |============================================== SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better -O2 -march=athlon64 ................... 1782 |======================= -O3 -march=athlon64 ................... 2021 |========================== -O3 -march=athlon64-sse3 .............. 2039 |========================== -O2 -march=native ..................... 1981 |========================== -O3 -march=native ..................... 2514 |================================ -O3 -march=native -flto ............... 2543 |================================= PGO ................................... 2555 |================================= AMD Ryzen Threadripper 2990WX 32-Core . 2257 |============================= SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better -O2 -march=athlon64 ................... 723 |============= -O3 -march=athlon64 ................... 736 |============= -O3 -march=athlon64-sse3 .............. 737 |============= -O2 -march=native ..................... 721 |============ -O3 -march=native ..................... 732 |============= -O3 -march=native -flto ............... 1904 |================================= PGO ................................... 728 |============= AMD Ryzen Threadripper 2990WX 32-Core . 255 |==== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better -O2 -march=athlon64 ................... 291 |================================== -O3 -march=athlon64 ................... 294 |================================== -O3 -march=athlon64-sse3 .............. 294 |================================== -O2 -march=native ..................... 265 |=============================== -O3 -march=native ..................... 270 |=============================== -O3 -march=native -flto ............... 270 |=============================== PGO ................................... 261 |============================== AMD Ryzen Threadripper 2990WX 32-Core . 260 |============================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better -O2 -march=athlon64 ................... 3119 |================================ -O3 -march=athlon64 ................... 2874 |============================= -O3 -march=athlon64-sse3 .............. 3082 |================================ -O2 -march=native ..................... 3105 |================================ -O3 -march=native ..................... 3174 |================================= -O3 -march=native -flto ............... 2951 |============================== PGO ................................... 3220 |================================= AMD Ryzen Threadripper 2990WX 32-Core . 3153 |================================ SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better -O2 -march=athlon64 ................... 3593 |=================== -O3 -march=athlon64 ................... 4274 |====================== -O3 -march=athlon64-sse3 .............. 4239 |====================== -O2 -march=native ..................... 4507 |======================= -O3 -march=native ..................... 5989 |=============================== -O3 -march=native -flto ............... 5388 |============================ PGO ................................... 6356 |================================= AMD Ryzen Threadripper 2990WX 32-Core . 5429 |============================ SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better -O2 -march=athlon64 ................... 1186 |================== -O3 -march=athlon64 ................... 1842 |=========================== -O3 -march=athlon64-sse3 .............. 1842 |=========================== -O2 -march=native ..................... 1306 |=================== -O3 -march=native ..................... 2218 |================================= -O3 -march=native -flto ............... 2202 |================================= PGO ................................... 2208 |================================= AMD Ryzen Threadripper 2990WX 32-Core . 2190 |================================= TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better -O2 -march=athlon64 ................... 1094211 |============================= -O3 -march=athlon64 ................... 1115390 |============================= -O3 -march=athlon64-sse3 .............. 1115841 |============================= -O2 -march=native ..................... 1102013 |============================= -O3 -march=native ..................... 1116747 |============================== -O3 -march=native -flto ............... 1135626 |============================== PGO ................................... 1109114 |============================= AMD Ryzen Threadripper 2990WX 32-Core . 981778 |========================== MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: f32 ms < Lower Is Better -O2 -march=athlon64 ................... 71.98 |================================ -O3 -march=athlon64 ................... 70.94 |=============================== -O3 -march=athlon64-sse3 .............. 71.02 |================================ -O2 -march=native ..................... 72.08 |================================ -O3 -march=native ..................... 71.26 |================================ PGO ................................... 70.94 |=============================== AMD Ryzen Threadripper 2990WX 32-Core . 71.00 |================================ AOM AV1 2019-02-11 AV1 Video Encoding Frames Per Second > Higher Is Better -O2 -march=athlon64 ................... 0.20 |============================== -O3 -march=athlon64 ................... 0.20 |============================== -O3 -march=athlon64-sse3 .............. 0.21 |================================ -O2 -march=native ..................... 0.21 |================================ -O3 -march=native ..................... 0.22 |================================= -O3 -march=native -flto ............... 0.22 |================================= PGO ................................... 0.22 |================================= AMD Ryzen Threadripper 2990WX 32-Core . 0.08 |============ SVT-AV1 2019-03-07 1080p 8-bit YUV To AV1 Video Encode Frames Per Second > Higher Is Better -O2 -march=athlon64 ................... 18.51 |============================= -O3 -march=athlon64 ................... 18.84 |============================== -O3 -march=athlon64-sse3 .............. 18.66 |============================= -O2 -march=native ..................... 18.63 |============================= -O3 -march=native ..................... 20.27 |================================ -O3 -march=native -flto ............... 20.41 |================================ PGO ................................... 19.57 |=============================== AMD Ryzen Threadripper 2990WX 32-Core . 18.77 |============================= SVT-HEVC 2019-02-03 1080p 8-bit YUV To HEVC Video Encode Frames Per Second > Higher Is Better -O2 -march=athlon64 ................... 172.00 |============================= -O3 -march=athlon64 ................... 168.00 |============================ -O3 -march=athlon64-sse3 .............. 166.00 |============================ -O2 -march=native ..................... 163.00 |=========================== -O3 -march=native ..................... 165.00 |============================ -O3 -march=native -flto ............... 165.00 |============================ PGO ................................... 185.00 |=============================== AMD Ryzen Threadripper 2990WX 32-Core . 23.01 |==== SVT-VP9 2019-02-17 1080p 8-bit YUV To VP9 Video Encode Frames Per Second > Higher Is Better -O2 -march=athlon64 ................... 101.13 |============================== -O3 -march=athlon64 ................... 103.45 |=============================== -O3 -march=athlon64-sse3 .............. 97.81 |============================= -O2 -march=native ..................... 100.82 |============================== -O3 -march=native ..................... 102.91 |=============================== -O3 -march=native -flto ............... 104.42 |=============================== PGO ................................... 103.97 |=============================== AMD Ryzen Threadripper 2990WX 32-Core . 5.07 |== VP9 libvpx Encoding 1.8.0 vpxenc VP9 1080p Video Encode Frames Per Second > Higher Is Better -O2 -march=athlon64 ...... 25.44 |=========================================== -O3 -march=athlon64 ...... 26.12 |============================================= -O3 -march=athlon64-sse3 . 25.58 |============================================ -O2 -march=native ........ 26.02 |============================================ -O3 -march=native ........ 26.37 |============================================= PGO ...................... 26.36 |============================================= x264 2018-09-25 H.264 Video Encoding Frames Per Second > Higher Is Better -O2 -march=athlon64 ...... 146 |=============================================== -O3 -march=athlon64 ...... 146 |=============================================== -O3 -march=athlon64-sse3 . 146 |=============================================== -O2 -march=native ........ 143 |============================================== -O3 -march=native ........ 147 |=============================================== PGO ...................... 145 |============================================== x265 3.0 H.265 1080p Video Encoding Frames Per Second > Higher Is Better -O2 -march=athlon64 ................... 33.53 |================================ -O3 -march=athlon64 ................... 33.89 |================================ -O3 -march=athlon64-sse3 .............. 33.88 |================================ -O2 -march=native ..................... 33.44 |================================ -O3 -march=native ..................... 33.76 |================================ -O3 -march=native -flto ............... 33.68 |================================ PGO ................................... 33.79 |================================ AMD Ryzen Threadripper 2990WX 32-Core . 9.95 |========= GraphicsMagick 1.3.30 Operation: Swirl Iterations Per Minute > Higher Is Better -O2 -march=athlon64 ................... 244 |================================= -O3 -march=athlon64 ................... 237 |================================ -O3 -march=athlon64-sse3 .............. 237 |================================ -O2 -march=native ..................... 245 |================================= -O3 -march=native ..................... 247 |================================== -O3 -march=native -flto ............... 250 |================================== PGO ................................... 250 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 26 |==== GraphicsMagick 1.3.30 Operation: Rotate Iterations Per Minute > Higher Is Better -O2 -march=athlon64 ................... 243 |================================= -O3 -march=athlon64 ................... 238 |================================ -O3 -march=athlon64-sse3 .............. 240 |================================= -O2 -march=native ..................... 249 |================================== -O3 -march=native ..................... 249 |================================== -O3 -march=native -flto ............... 248 |================================== PGO ................................... 251 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 175 |======================== GraphicsMagick 1.3.30 Operation: Sharpen Iterations Per Minute > Higher Is Better -O2 -march=athlon64 ................... 192 |============================== -O3 -march=athlon64 ................... 190 |============================= -O3 -march=athlon64-sse3 .............. 191 |============================= -O2 -march=native ..................... 217 |================================= -O3 -march=native ..................... 219 |================================== -O3 -march=native -flto ............... 221 |================================== PGO ................................... 220 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 4 |= GraphicsMagick 1.3.30 Operation: Enhanced Iterations Per Minute > Higher Is Better -O2 -march=athlon64 ................... 198 |============================= -O3 -march=athlon64 ................... 195 |============================ -O3 -march=athlon64-sse3 .............. 194 |============================ -O2 -march=native ..................... 231 |================================== -O3 -march=native ..................... 232 |================================== -O3 -march=native -flto ............... 233 |================================== PGO ................................... 234 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 8 |= GraphicsMagick 1.3.30 Operation: Resizing Iterations Per Minute > Higher Is Better -O2 -march=athlon64 ................... 235 |================================= -O3 -march=athlon64 ................... 233 |================================= -O3 -march=athlon64-sse3 .............. 231 |================================ -O2 -march=native ..................... 238 |================================= -O3 -march=native ..................... 243 |================================== -O3 -march=native -flto ............... 240 |================================== PGO ................................... 243 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 31 |==== GraphicsMagick 1.3.30 Operation: Noise-Gaussian Iterations Per Minute > Higher Is Better -O2 -march=athlon64 ................... 202 |================================== -O3 -march=athlon64 ................... 199 |================================= -O3 -march=athlon64-sse3 .............. 198 |================================= -O2 -march=native ..................... 200 |================================= -O3 -march=native ..................... 204 |================================== -O3 -march=native -flto ............... 203 |================================== PGO ................................... 203 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 40 |======= GraphicsMagick 1.3.30 Operation: HWB Color Space Iterations Per Minute > Higher Is Better -O2 -march=athlon64 ................... 272 |================================== -O3 -march=athlon64 ................... 261 |================================ -O3 -march=athlon64-sse3 .............. 263 |================================= -O2 -march=native ..................... 271 |================================== -O3 -march=native ..................... 272 |================================== -O3 -march=native -flto ............... 274 |================================== PGO ................................... 274 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 59 |======= Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better -O2 -march=athlon64 ................... 1328 |================================= -O3 -march=athlon64 ................... 1316 |================================= -O3 -march=athlon64-sse3 .............. 1316 |================================= -O2 -march=native ..................... 1319 |================================= -O3 -march=native ..................... 1313 |================================= -O3 -march=native -flto ............... 1304 |================================ PGO ................................... 1321 |================================= AMD Ryzen Threadripper 2990WX 32-Core . 1322 |================================= Stockfish 9 Total Time Nodes Per Second > Higher Is Better -O2 -march=athlon64 ...... 66890687 |========================================= -O3 -march=athlon64 ...... 67513602 |========================================== -O3 -march=athlon64-sse3 . 67571150 |========================================== -O2 -march=native ........ 66697487 |========================================= -O3 -march=native ........ 68200164 |========================================== -O3 -march=native -flto .. 67450689 |========================================== PGO ...................... 67841877 |========================================== Timed ImageMagick Compilation 6.9.0 Time To Compile Seconds < Lower Is Better -O2 -march=athlon64 ................... 17.41 |====== -O3 -march=athlon64 ................... 19.77 |======= -O3 -march=athlon64-sse3 .............. 19.73 |======= -O2 -march=native ..................... 17.58 |====== -O3 -march=native ..................... 19.15 |======= -O3 -march=native -flto ............... 88.89 |================================ PGO ................................... 19.09 |======= AMD Ryzen Threadripper 2990WX 32-Core . 19.69 |======= Timed LLVM Compilation 6.0.1 Time To Compile Seconds < Lower Is Better -O2 -march=athlon64 ................... 229 |======== -O3 -march=athlon64 ................... 225 |======== -O3 -march=athlon64-sse3 .............. 224 |======== -O2 -march=native ..................... 221 |======== -O3 -march=native ..................... 221 |======== -O3 -march=native -flto ............... 951 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 290 |========== Timed PHP Compilation 7.1.9 Time To Compile Seconds < Lower Is Better -O2 -march=athlon64 ................... 43.86 |================ -O3 -march=athlon64 ................... 63.19 |======================== -O3 -march=athlon64-sse3 .............. 62.93 |======================== -O2 -march=native ..................... 44.44 |================= -O3 -march=native ..................... 63.25 |======================== PGO ................................... 63.01 |======================== AMD Ryzen Threadripper 2990WX 32-Core . 85.40 |================================ C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Seconds < Lower Is Better -O2 -march=athlon64 ................... 45.43 | -O3 -march=athlon64 ................... 29.39 | -O3 -march=athlon64-sse3 .............. 29.53 | -O2 -march=native ..................... 34.22 | -O3 -march=native ..................... 18.00 | -O3 -march=native -flto ............... 17.85 | PGO ................................... 17.96 | AMD Ryzen Threadripper 2990WX 32-Core . 4024.62 |============================== Smallpt 1.0 Global Illumination Renderer; 128 Samples Seconds < Lower Is Better -O2 -march=athlon64 ................... 4.57 | -O3 -march=athlon64 ................... 4.51 | -O3 -march=athlon64-sse3 .............. 4.54 | -O2 -march=native ..................... 3.89 | -O3 -march=native ..................... 3.87 | -O3 -march=native -flto ............... 3.85 | PGO ................................... 3.83 | AMD Ryzen Threadripper 2990WX 32-Core . 564.07 |=============================== AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better -O2 -march=athlon64 ................... 47.28 |================================ -O3 -march=athlon64 ................... 45.20 |=============================== -O3 -march=athlon64-sse3 .............. 44.98 |============================== -O2 -march=native ..................... 42.34 |============================= -O3 -march=native ..................... 39.11 |========================== -O3 -march=native -flto ............... 39.92 |=========================== PGO ................................... 39.10 |========================== AMD Ryzen Threadripper 2990WX 32-Core . 43.36 |============================= Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better -O2 -march=athlon64 ...... 2.49 |============================================== -O3 -march=athlon64 ...... 2.46 |============================================= -O3 -march=athlon64-sse3 . 2.46 |============================================= -O2 -march=native ........ 2.38 |============================================ -O3 -march=native ........ 2.35 |=========================================== -O3 -march=native -flto .. 2.30 |========================================== PGO ...................... 2.34 |=========================================== Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better -O2 -march=athlon64 ...... 3.96 |============================================== -O3 -march=athlon64 ...... 3.92 |============================================= -O3 -march=athlon64-sse3 . 3.91 |============================================= -O2 -march=native ........ 3.95 |============================================== -O3 -march=native ........ 3.84 |============================================ -O3 -march=native -flto .. 3.97 |============================================== PGO ...................... 3.83 |============================================ Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better -O2 -march=athlon64 ...... 4.68 |============================================= -O3 -march=athlon64 ...... 4.65 |============================================ -O3 -march=athlon64-sse3 . 4.62 |============================================ -O2 -march=native ........ 4.57 |============================================ -O3 -march=native ........ 4.42 |========================================== -O3 -march=native -flto .. 4.82 |============================================== PGO ...................... 4.41 |========================================== Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better -O2 -march=athlon64 ...... 4.50 |============================================== -O3 -march=athlon64 ...... 4.45 |============================================= -O3 -march=athlon64-sse3 . 4.45 |============================================= -O2 -march=native ........ 4.06 |========================================== -O3 -march=native ........ 3.97 |========================================= -O3 -march=native -flto .. 3.90 |======================================== PGO ...................... 3.97 |========================================= Bullet Physics Engine 2.81 Test: 136 Ragdolls Seconds < Lower Is Better -O2 -march=athlon64 ...... 2.42 |============================================== -O3 -march=athlon64 ...... 2.36 |============================================ -O3 -march=athlon64-sse3 . 2.35 |============================================ -O2 -march=native ........ 2.39 |============================================= -O3 -march=native ........ 2.30 |=========================================== -O3 -march=native -flto .. 2.44 |============================================== PGO ...................... 2.32 |============================================ Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better -O2 -march=athlon64 ...... 0.87 |============================================== -O3 -march=athlon64 ...... 0.86 |============================================= -O3 -march=athlon64-sse3 . 0.86 |============================================= -O2 -march=native ........ 0.84 |============================================ -O3 -march=native ........ 0.84 |============================================ -O3 -march=native -flto .. 0.82 |=========================================== PGO ...................... 0.83 |============================================ Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better -O2 -march=athlon64 ...... 1.07 |============================================== -O3 -march=athlon64 ...... 1.06 |============================================== -O3 -march=athlon64-sse3 . 1.06 |============================================== -O2 -march=native ........ 1.02 |============================================ -O3 -march=native ........ 1.00 |=========================================== -O3 -march=native -flto .. 0.97 |========================================== PGO ...................... 1.00 |=========================================== XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 Seconds < Lower Is Better -O2 -march=athlon64 ................... 26.75 |= -O3 -march=athlon64 ................... 26.50 |= -O3 -march=athlon64-sse3 .............. 25.93 |= -O2 -march=native ..................... 26.78 |= -O3 -march=native ..................... 25.70 |= -O3 -march=native -flto ............... 25.62 |= PGO ................................... 26.09 |= AMD Ryzen Threadripper 2990WX 32-Core . 708.15 |=============================== Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Seconds < Lower Is Better -O2 -march=athlon64 ................... 18.02 |== -O3 -march=athlon64 ................... 17.74 |== -O3 -march=athlon64-sse3 .............. 18.38 |== -O2 -march=native ..................... 19.31 |== -O3 -march=native ..................... 19.09 |== -O3 -march=native -flto ............... 17.94 |== PGO ................................... 17.36 |== AMD Ryzen Threadripper 2990WX 32-Core . 265.79 |=============================== FLAC Audio Encoding 1.3.2 WAV To FLAC Seconds < Lower Is Better -O2 -march=athlon64 ................... 15.58 |================================ -O3 -march=athlon64 ................... 15.48 |================================ -O3 -march=athlon64-sse3 .............. 15.43 |================================ -O2 -march=native ..................... 9.78 |==================== -O3 -march=native ..................... 9.53 |==================== -O3 -march=native -flto ............... 9.48 |=================== PGO ................................... 9.44 |=================== AMD Ryzen Threadripper 2990WX 32-Core . 9.81 |==================== LAME MP3 Encoding 3.100 WAV To MP3 Seconds < Lower Is Better -O2 -march=athlon64 ................... 11.41 |================================ -O3 -march=athlon64 ................... 9.43 |========================== -O3 -march=athlon64-sse3 .............. 9.36 |========================== -O2 -march=native ..................... 10.80 |============================== -O3 -march=native ..................... 8.00 |====================== -O3 -march=native -flto ............... 7.98 |====================== PGO ................................... 7.98 |====================== AMD Ryzen Threadripper 2990WX 32-Core . 11.09 |=============================== OpenSSL 1.1.1 RSA 4096-bit Performance Signs Per Second > Higher Is Better -O2 -march=athlon64 ................... 5832 |================================= -O3 -march=athlon64 ................... 5837 |================================= -O3 -march=athlon64-sse3 .............. 5838 |================================= -O2 -march=native ..................... 5831 |================================= -O3 -march=native ..................... 5825 |================================= -O3 -march=native -flto ............... 5833 |================================= PGO ................................... 5830 |================================= AMD Ryzen Threadripper 2990WX 32-Core . 5791 |================================= PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only TPS > Higher Is Better -O2 -march=athlon64 ................... 453605 |============================== -O3 -march=athlon64 ................... 463840 |============================== -O3 -march=athlon64-sse3 .............. 459596 |============================== -O2 -march=native ..................... 458551 |============================== -O3 -march=native ..................... 466723 |=============================== -O3 -march=native -flto ............... 473620 |=============================== PGO ................................... 460414 |============================== AMD Ryzen Threadripper 2990WX 32-Core . 259058 |================= PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write TPS > Higher Is Better -O2 -march=athlon64 ................... 5098 |========== -O3 -march=athlon64 ................... 15616 |=============================== -O3 -march=athlon64-sse3 .............. 16252 |================================ -O2 -march=native ..................... 13558 |=========================== -O3 -march=native ..................... 16339 |================================ -O3 -march=native -flto ............... 14989 |============================= PGO ................................... 6281 |============ AMD Ryzen Threadripper 2990WX 32-Core . 6472 |============= CppPerformanceBenchmarks 9 Test: Atol Seconds < Lower Is Better -O2 -march=athlon64 ................... 69.08 |================================ -O3 -march=athlon64 ................... 68.37 |=============================== -O3 -march=athlon64-sse3 .............. 68.42 |=============================== -O2 -march=native ..................... 69.80 |================================ -O3 -march=native ..................... 69.03 |================================ -O3 -march=native -flto ............... 69.25 |================================ PGO ................................... 69.31 |================================ AMD Ryzen Threadripper 2990WX 32-Core . 69.25 |================================ CppPerformanceBenchmarks 9 Test: Ctype Seconds < Lower Is Better -O2 -march=athlon64 ................... 33.37 |============================ -O3 -march=athlon64 ................... 33.28 |============================ -O3 -march=athlon64-sse3 .............. 33.31 |============================ -O2 -march=native ..................... 34.50 |============================= -O3 -march=native ..................... 33.88 |============================= -O3 -march=native -flto ............... 32.32 |=========================== PGO ................................... 34.02 |============================= AMD Ryzen Threadripper 2990WX 32-Core . 37.67 |================================ CppPerformanceBenchmarks 9 Test: Math Library Seconds < Lower Is Better -O2 -march=athlon64 ................... 399 |================================== -O3 -march=athlon64 ................... 398 |================================== -O3 -march=athlon64-sse3 .............. 399 |================================== -O2 -march=native ..................... 356 |============================== -O3 -march=native ..................... 351 |============================== -O3 -march=native -flto ............... 352 |============================== PGO ................................... 353 |============================== AMD Ryzen Threadripper 2990WX 32-Core . 355 |============================== CppPerformanceBenchmarks 9 Test: Random Numbers Seconds < Lower Is Better -O2 -march=athlon64 ................... 1072 |=============================== -O3 -march=athlon64 ................... 1070 |=============================== -O3 -march=athlon64-sse3 .............. 1055 |============================== -O2 -march=native ..................... 1041 |============================== -O3 -march=native ..................... 1023 |============================= -O3 -march=native -flto ............... 1011 |============================= PGO ................................... 1027 |============================= AMD Ryzen Threadripper 2990WX 32-Core . 1154 |================================= CppPerformanceBenchmarks 9 Test: Stepanov Vector Seconds < Lower Is Better -O2 -march=athlon64 ................... 75.79 |=============================== -O3 -march=athlon64 ................... 75.53 |=============================== -O3 -march=athlon64-sse3 .............. 75.56 |=============================== -O2 -march=native ..................... 75.99 |=============================== -O3 -march=native ..................... 74.93 |=============================== -O3 -march=native -flto ............... 74.92 |=============================== PGO ................................... 75.27 |=============================== AMD Ryzen Threadripper 2990WX 32-Core . 78.14 |================================ CppPerformanceBenchmarks 9 Test: Function Objects Seconds < Lower Is Better -O2 -march=athlon64 ................... 16.20 |======================= -O3 -march=athlon64 ................... 16.17 |======================= -O3 -march=athlon64-sse3 .............. 16.14 |======================= -O2 -march=native ..................... 15.61 |======================= -O3 -march=native ..................... 15.40 |====================== -O3 -march=native -flto ............... 15.50 |====================== PGO ................................... 15.46 |====================== AMD Ryzen Threadripper 2990WX 32-Core . 22.07 |================================ CppPerformanceBenchmarks 9 Test: Stepanov Abstraction Seconds < Lower Is Better -O2 -march=athlon64 ................... 28.33 |=============================== -O3 -march=athlon64 ................... 28.26 |=============================== -O3 -march=athlon64-sse3 .............. 28.28 |=============================== -O2 -march=native ..................... 28.69 |================================ -O3 -march=native ..................... 28.22 |=============================== -O3 -march=native -flto ............... 28.25 |=============================== PGO ................................... 28.36 |=============================== AMD Ryzen Threadripper 2990WX 32-Core . 29.08 |================================ Redis 4.0.8 Test: LPOP Requests Per Second > Higher Is Better -O2 -march=athlon64 ...... 2654025 |=========================================== -O3 -march=athlon64 ...... 2544762 |========================================= -O3 -march=athlon64-sse3 . 2506276 |========================================= -O2 -march=native ........ 2340595 |====================================== -O3 -march=native ........ 2573823 |========================================== -O3 -march=native -flto .. 2616703 |========================================== PGO ...................... 2648345 |=========================================== Redis 4.0.8 Test: SADD Requests Per Second > Higher Is Better -O2 -march=athlon64 ...... 1975322 |========================================= -O3 -march=athlon64 ...... 2051129 |========================================== -O3 -march=athlon64-sse3 . 2032921 |========================================== -O2 -march=native ........ 2000709 |========================================= -O3 -march=native ........ 2046557 |========================================== -O3 -march=native -flto .. 2084089 |=========================================== PGO ...................... 2083430 |=========================================== Redis 4.0.8 Test: LPUSH Requests Per Second > Higher Is Better -O2 -march=athlon64 ...... 1548556 |=========================================== -O3 -march=athlon64 ...... 1520276 |========================================== -O3 -march=athlon64-sse3 . 1522952 |========================================== -O2 -march=native ........ 1463388 |========================================= -O3 -march=native ........ 1540257 |=========================================== -O3 -march=native -flto .. 1549647 |=========================================== PGO ...................... 1532038 |=========================================== Redis 4.0.8 Test: GET Requests Per Second > Higher Is Better -O2 -march=athlon64 ...... 2384143 |======================================== -O3 -march=athlon64 ...... 2540398 |=========================================== -O3 -march=athlon64-sse3 . 2461725 |========================================== -O2 -march=native ........ 2275019 |======================================= -O3 -march=native ........ 2502182 |========================================== -O3 -march=native -flto .. 2509433 |========================================== PGO ...................... 2533027 |=========================================== Redis 4.0.8 Test: SET Requests Per Second > Higher Is Better -O2 -march=athlon64 ...... 1730702 |======================================== -O3 -march=athlon64 ...... 1840806 |=========================================== -O3 -march=athlon64-sse3 . 1755590 |========================================= -O2 -march=native ........ 1744979 |========================================= -O3 -march=native ........ 1807270 |========================================== -O3 -march=native -flto .. 1755472 |========================================= PGO ...................... 1798800 |========================================== ctx_clock Context Switch Time Clocks < Lower Is Better -O2 -march=athlon64 ................... 150 |================================== -O3 -march=athlon64 ................... 150 |================================== -O3 -march=athlon64-sse3 .............. 150 |================================== -O2 -march=native ..................... 150 |================================== -O3 -march=native ..................... 150 |================================== -O3 -march=native -flto ............... 150 |================================== PGO ................................... 150 |================================== AMD Ryzen Threadripper 2990WX 32-Core . 150 |================================== Memcached mcperf 1.5.10 Method: Add Operations Per Second > Higher Is Better -O2 -march=athlon64 ................... 54600 |================================ -O3 -march=athlon64 ................... 53106 |=============================== -O3 -march=athlon64-sse3 .............. 34822 |==================== -O2 -march=native ..................... 34138 |==================== -O3 -march=native ..................... 44903 |========================== -O3 -march=native -flto ............... 50055 |============================= PGO ................................... 47774 |============================ AMD Ryzen Threadripper 2990WX 32-Core . 35174 |===================== Memcached mcperf 1.5.10 Method: Get Operations Per Second > Higher Is Better -O2 -march=athlon64 ................... 55647 |========================== -O3 -march=athlon64 ................... 56837 |========================== -O3 -march=athlon64-sse3 .............. 69004 |================================ -O2 -march=native ..................... 55791 |========================== -O3 -march=native ..................... 57652 |=========================== -O3 -march=native -flto ............... 68644 |================================ PGO ................................... 68426 |================================ AMD Ryzen Threadripper 2990WX 32-Core . 56324 |========================== Memcached mcperf 1.5.10 Method: Set Operations Per Second > Higher Is Better -O2 -march=athlon64 ................... 59924 |================================ -O3 -march=athlon64 ................... 34880 |=================== -O3 -march=athlon64-sse3 .............. 48548 |========================== -O2 -march=native ..................... 34446 |================== -O3 -march=native ..................... 38646 |===================== -O3 -march=native -flto ............... 43570 |======================= PGO ................................... 46396 |========================= AMD Ryzen Threadripper 2990WX 32-Core . 35239 |=================== Memcached mcperf 1.5.10 Method: Append Operations Per Second > Higher Is Better -O2 -march=athlon64 ................... 61219 |================================ -O3 -march=athlon64 ................... 56172 |============================= -O3 -march=athlon64-sse3 .............. 45716 |======================== -O2 -march=native ..................... 35058 |================== -O3 -march=native ..................... 42455 |====================== -O3 -march=native -flto ............... 45339 |======================== PGO ................................... 46106 |======================== AMD Ryzen Threadripper 2990WX 32-Core . 35920 |=================== Memcached mcperf 1.5.10 Method: Delete Operations Per Second > Higher Is Better -O2 -march=athlon64 ................... 55747 |========================== -O3 -march=athlon64 ................... 56822 |========================== -O3 -march=athlon64-sse3 .............. 69509 |================================ -O2 -march=native ..................... 56141 |========================== -O3 -march=native ..................... 58969 |=========================== -O3 -march=native -flto ............... 68696 |================================ PGO ................................... 68980 |================================ AMD Ryzen Threadripper 2990WX 32-Core . 56797 |========================== Memcached mcperf 1.5.10 Method: Prepend Operations Per Second > Higher Is Better -O2 -march=athlon64 ................... 35467 |======================== -O3 -march=athlon64 ................... 35968 |======================== -O3 -march=athlon64-sse3 .............. 45691 |=============================== -O2 -march=native ..................... 35552 |======================== -O3 -march=native ..................... 43824 |============================== -O3 -march=native -flto ............... 45587 |=============================== PGO ................................... 47085 |================================ AMD Ryzen Threadripper 2990WX 32-Core . 35579 |======================== Memcached mcperf 1.5.10 Method: Replace Operations Per Second > Higher Is Better -O2 -march=athlon64 ................... 35829 |===================== -O3 -march=athlon64 ................... 36231 |====================== -O3 -march=athlon64-sse3 .............. 45707 |=========================== -O2 -march=native ..................... 35486 |===================== -O3 -march=native ..................... 53865 |================================ -O3 -march=native -flto ............... 45591 |=========================== PGO ................................... 45956 |=========================== AMD Ryzen Threadripper 2990WX 32-Core . 35978 |===================== NGINX Benchmark 1.9.9 Static Web Page Serving Requests Per Second > Higher Is Better -O2 -march=athlon64 ...... 29704 |============================================= -O3 -march=athlon64 ...... 29726 |============================================= -O3 -march=athlon64-sse3 . 29281 |============================================ -O2 -march=native ........ 27834 |========================================== -O3 -march=native ........ 29274 |============================================ -O3 -march=native -flto .. 27352 |=========================================