Clang znver1 Scheduler Model Tests x86-64 Haswell march AMD Ryzen 7 1700 Eight-Core testing with the new Zen znver1 scheduler model. Tests by Michael Larabel for a future article. -O3 -march=x86-64: Processor: AMD Ryzen 7 1700 Eight-Core @ 3.00GHz (16 Cores), Motherboard: MSI B350 TOMAHAWK (MS-7A34) v1.0, Chipset: AMD Device 1450, Memory: 16384MB, Disk: 120GB Samsung SSD 840, Graphics: HIS AMD Radeon HD 7750/8740 / R7 250E 1024MB, Audio: AMD Cape Verde/Pitcairn, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 17.04, Kernel: 4.12.0-999-generic (x86_64) 20170622, Desktop: Unity 7.5.0, Display Driver: modesetting 1.19.3, Compiler: Clang 6.0.0 (SVN 308494), File-System: ext4, Screen Resolution: 1024x768 -O3 -march=znver1: Processor: AMD Ryzen 7 1700 Eight-Core @ 3.00GHz (16 Cores), Motherboard: MSI B350 TOMAHAWK (MS-7A34) v1.0, Chipset: AMD Device 1450, Memory: 16384MB, Disk: 120GB Samsung SSD 840, Graphics: HIS AMD Radeon HD 7750/8740 / R7 250E 1024MB, Audio: AMD Cape Verde/Pitcairn, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 17.04, Kernel: 4.12.0-999-generic (x86_64) 20170622, Desktop: Unity 7.5.0, Display Driver: modesetting 1.19.3, Compiler: Clang 6.0.0 (SVN 308494), File-System: ext4, Screen Resolution: 1024x768 -O3 -march=haswell: Processor: AMD Ryzen 7 1700 Eight-Core @ 3.00GHz (16 Cores), Motherboard: MSI B350 TOMAHAWK (MS-7A34) v1.0, Chipset: AMD Device 1450, Memory: 16384MB, Disk: 120GB Samsung SSD 840, Graphics: HIS AMD Radeon HD 7750/8740 / R7 250E 1024MB, Audio: AMD Cape Verde/Pitcairn, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 17.04, Kernel: 4.13.0-999-generic (x86_64) 20170622, Desktop: Unity 7.5.0, Display Driver: modesetting 1.19.3, Compiler: Clang 6.0.0 (SVN 308494), File-System: ext4, Screen Resolution: 1024x768 FFTW 3.3.4 Build: Float + SSE - Size: 2D FFT Size 4096 Mflops > Higher Is Better -O3 -march=x86-64 .. 13690 |=================================================== -O3 -march=znver1 .. 13398 |================================================== -O3 -march=haswell . 13133 |================================================= SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better -O3 -march=x86-64 .. 2129.48 |================================================ -O3 -march=znver1 .. 2152.36 |================================================= -O3 -march=haswell . 2129.77 |================================================ SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better -O3 -march=x86-64 .. 642.55 |================================================= -O3 -march=znver1 .. 657.96 |================================================== -O3 -march=haswell . 653.54 |================================================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better -O3 -march=x86-64 .. 133.36 |============================ -O3 -march=znver1 .. 134.24 |============================ -O3 -march=haswell . 240.21 |================================================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better -O3 -march=x86-64 .. 2623.75 |================================================= -O3 -march=znver1 .. 2597.54 |================================================ -O3 -march=haswell . 2627.46 |================================================= SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better -O3 -march=x86-64 .. 5572.72 |================================================ -O3 -march=znver1 .. 5692.83 |================================================= -O3 -march=haswell . 5454.00 |=============================================== SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better -O3 -march=x86-64 .. 1675.03 |================================================= -O3 -march=znver1 .. 1679.21 |================================================= -O3 -march=haswell . 1673.65 |================================================= TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better -O3 -march=x86-64 .. 1103331 |================================================= -O3 -march=znver1 .. 1096303 |================================================= -O3 -march=haswell . 1090223 |================================================ x264 2017-05-19 H.264 Video Encoding Frames Per Second > Higher Is Better -O3 -march=x86-64 .. 252.33 |================================================== -O3 -march=znver1 .. 252.52 |================================================== -O3 -march=haswell . 233.92 |============================================== GraphicsMagick 1.3.19 Operation: Blur Iterations Per Minute > Higher Is Better -O3 -march=x86-64 .. 136 |=================================================== -O3 -march=znver1 .. 140 |===================================================== -O3 -march=haswell . 132 |================================================== GraphicsMagick 1.3.19 Operation: Sharpen Iterations Per Minute > Higher Is Better -O3 -march=x86-64 .. 137 |==================================================== -O3 -march=znver1 .. 139 |===================================================== -O3 -march=haswell . 139 |===================================================== GraphicsMagick 1.3.19 Operation: HWB Color Space Iterations Per Minute > Higher Is Better -O3 -march=x86-64 .. 173 |==================================================== -O3 -march=znver1 .. 178 |===================================================== -O3 -march=haswell . 176 |==================================================== GraphicsMagick 1.3.19 Operation: Local Adaptive Thresholding Iterations Per Minute > Higher Is Better -O3 -march=x86-64 .. 133 |==================================================== -O3 -march=znver1 .. 136 |===================================================== -O3 -march=haswell . 126 |================================================= Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better -O3 -march=x86-64 .. 1110.92 |================================================ -O3 -march=znver1 .. 1022.65 |============================================= -O3 -march=haswell . 1123.43 |================================================= ebizzy 0.3 Records/s > Higher Is Better -O3 -march=x86-64 .. 408509 |================================================== -O3 -march=znver1 .. 382960 |=============================================== -O3 -march=haswell . 400765 |================================================= C-Ray 1.1 Total Time Seconds < Lower Is Better -O3 -march=x86-64 .. 13.68 |=================================================== -O3 -march=znver1 .. 13.62 |================================================== -O3 -march=haswell . 13.77 |=================================================== Stockfish 2014-11-26 Total Time ms < Lower Is Better -O3 -march=x86-64 .. 3710 |==================================================== -O3 -march=znver1 .. 3670 |=================================================== -O3 -march=haswell . 3745 |==================================================== Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better -O3 -march=x86-64 .. 2.86 |==================================================== -O3 -march=znver1 .. 2.78 |=================================================== -O3 -march=haswell . 2.72 |================================================= Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better -O3 -march=x86-64 .. 4.62 |==================================================== -O3 -march=znver1 .. 4.53 |=================================================== -O3 -march=haswell . 4.57 |=================================================== Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better -O3 -march=x86-64 .. 5.27 |==================================================== -O3 -march=znver1 .. 5.10 |================================================== -O3 -march=haswell . 5.20 |=================================================== Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better -O3 -march=x86-64 .. 4.84 |==================================================== -O3 -march=znver1 .. 4.64 |================================================== -O3 -march=haswell . 4.47 |================================================ Bullet Physics Engine 2.81 Test: 136 Ragdolls Seconds < Lower Is Better -O3 -march=x86-64 .. 2.96 |==================================================== -O3 -march=znver1 .. 2.93 |=================================================== -O3 -march=haswell . 2.93 |=================================================== Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better -O3 -march=x86-64 .. 1.00 |==================================================== -O3 -march=znver1 .. 0.99 |=================================================== -O3 -march=haswell . 0.98 |=================================================== Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better -O3 -march=x86-64 .. 1.20 |==================================================== -O3 -march=znver1 .. 1.16 |================================================== -O3 -march=haswell . 1.12 |================================================= FLAC Audio Encoding 1.3.1 WAV To FLAC Seconds < Lower Is Better -O3 -march=x86-64 .. 6.77 |==================================================== -O3 -march=znver1 .. 5.69 |============================================ -O3 -march=haswell . 5.67 |============================================ LAME MP3 Encoding 3.99.3 WAV To MP3 Seconds < Lower Is Better -O3 -march=x86-64 .. 9.75 |============================================= -O3 -march=znver1 .. 10.92 |=================================================== -O3 -march=haswell . 10.95 |=================================================== WavPack Audio Encoding 5.1 WAV To WavPack Seconds < Lower Is Better -O3 -march=x86-64 .. 6.52 |==================================================== -O3 -march=znver1 .. 6.45 |=================================================== -O3 -march=haswell . 6.50 |==================================================== Tachyon 0.98.9 Total Time Seconds < Lower Is Better -O3 -march=x86-64 .. 5.29 |==================================================== -O3 -march=znver1 .. 5.29 |==================================================== -O3 -march=haswell . 5.30 |==================================================== libjpeg-turbo tjbench 1.5.1 Test: Decompression Throughput Megapixels/sec > Higher Is Better -O3 -march=x86-64 .. 161.78 |================================================ -O3 -march=znver1 .. 165.30 |================================================= -O3 -march=haswell . 167.20 |================================================== PostgreSQL pgbench 9.6.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write TPS > Higher Is Better -O3 -march=x86-64 .. 1865.09 |================================================= -O3 -march=znver1 .. 1813.34 |================================================ -O3 -march=haswell . 1662.61 |============================================ PostgreSQL pgbench 9.6.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Write TPS > Higher Is Better -O3 -march=x86-64 .. 226.04 |================================================== -O3 -march=znver1 .. 224.89 |================================================== -O3 -march=haswell . 216.83 |================================================ PostgreSQL pgbench 9.6.3 Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write TPS > Higher Is Better -O3 -march=x86-64 .. 1938.93 |================================================= -O3 -march=znver1 .. 1875.27 |=============================================== -O3 -march=haswell . 1488.59 |====================================== Redis 3.0.1 Test: LPOP Requests Per Second > Higher Is Better -O3 -march=x86-64 .. 1968693.38 |============================================= -O3 -march=znver1 .. 1960971.50 |============================================= -O3 -march=haswell . 2004499.67 |============================================== Redis 3.0.1 Test: GET Requests Per Second > Higher Is Better -O3 -march=x86-64 .. 1953368.38 |============================================== -O3 -march=znver1 .. 1973143.29 |============================================== -O3 -march=haswell . 1916990.88 |============================================= Redis 3.0.1 Test: SET Requests Per Second > Higher Is Better -O3 -march=x86-64 .. 1385134.04 |============================================= -O3 -march=znver1 .. 1413713.29 |============================================== -O3 -march=haswell . 1411858.83 |============================================== Hierarchical INTegration 1.0 Test: FLOAT QUIPs > Higher Is Better -O3 -march=x86-64 .. 301795000.85 |============================================ -O3 -march=znver1 .. 300351844.05 |============================================ -O3 -march=haswell . 299157100.64 |============================================ Apache Benchmark 2.4.7 Static Web Page Serving Requests Per Second > Higher Is Better -O3 -march=x86-64 .. 26994.80 |=============================================== -O3 -march=znver1 .. 27700.93 |================================================ -O3 -march=haswell . 26277.48 |==============================================