RAM speed performance comparison Testing RAM speed impact for the new CSMM workstation 3200MHz CL 16-18-18-18-38 1T: Processor: AMD Ryzen 9 3900X 12-Core @ 3.80GHz (12 Cores / 24 Threads), Motherboard: MSI MEG X570 UNIFY (MS-7C35) v2.0 (A.30 BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 2 x 1024GB ADATA SX8200PNP, Graphics: AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (2100/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: S2719DGF + DELL U2312HM, Network: Realtek RTL8125 2.5GbE OS: Arch rolling, Kernel: 5.10.15-arch1-1 (x86_64), Display Server: X Server 1.20.10, OpenGL: 4.6 Mesa 20.3.4 (LLVM 11.0.1), Compiler: GCC 10.2.0 + Clang 11.0.1, File-System: xfs, Screen Resolution: 4480x1440 2666MHz CL 18-18-18-18-43 1T: Processor: AMD Ryzen 9 3900X 12-Core @ 3.80GHz (12 Cores / 24 Threads), Motherboard: MSI MEG X570 UNIFY (MS-7C35) v2.0 (A.30 BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 2 x 1024GB ADATA SX8200PNP, Graphics: AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (2100/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: S2719DGF + DELL U2312HM, Network: Realtek RTL8125 2.5GbE OS: Arch rolling, Kernel: 5.10.15-arch1-1 (x86_64), Display Server: X Server 1.20.10, OpenGL: 4.6 Mesa 20.3.4 (LLVM 11.0.1), Compiler: GCC 10.2.0 + Clang 11.0.1, File-System: xfs, Screen Resolution: 4480x1440 SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better 3200MHz CL 16-18-18-18-38 1T . 659.60 |======================================== 2666MHz CL 18-18-18-18-43 1T . 650.43 |======================================= SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better 3200MHz CL 16-18-18-18-38 1T . 160.65 |======================================== 2666MHz CL 18-18-18-18-43 1T . 160.47 |======================================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better 3200MHz CL 16-18-18-18-38 1T . 360.49 |======================================== 2666MHz CL 18-18-18-18-43 1T . 339.13 |====================================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better 3200MHz CL 16-18-18-18-38 1T . 809.59 |======================================== 2666MHz CL 18-18-18-18-43 1T . 783.14 |======================================= SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better 3200MHz CL 16-18-18-18-38 1T . 642.65 |======================================== 2666MHz CL 18-18-18-18-43 1T . 645.23 |======================================== SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better 3200MHz CL 16-18-18-18-38 1T . 1305.64 |====================================== 2666MHz CL 18-18-18-18-43 1T . 1324.18 |======================================= Timed LLVM Compilation 10.0 Time To Compile Seconds < Lower Is Better 3200MHz CL 16-18-18-18-38 1T . 417.28 |======================================= 2666MHz CL 18-18-18-18-43 1T . 430.58 |======================================== Xsbench 2017-07-06 Lookups/s > Higher Is Better 3200MHz CL 16-18-18-18-38 1T . 3509496 |======================================= 2666MHz CL 18-18-18-18-43 1T . 3352635 |===================================== Kripke 1.2.4 Throughput FoM > Higher Is Better 3200MHz CL 16-18-18-18-38 1T . 92424237 |====================================== 2666MHz CL 18-18-18-18-43 1T . 81082983 |=================================