Generally, any software, not in any special instructions, could have a consistent performance on different processor architectures.
LAIPE 2 is one of them, without any special instructions. A consistent performance could be expected.
This post uses timing results to show it. Matrix [A] is of order 15,by, , and matrix [B] is of order 11,by, , and matrix [C] is of order 15,by, The compilation option included -O3, and the executable was linked against neuloop4.
Test EQUATION SILENTIS HYGRO
Because L can run at a higher frequency when using 2 or less cores, processor's turbo boost was disabled for the purpose to measure speedup. Further, logical core e. Navigation Tree.
The computer has 48 cores. Speedup and parallel efficiency are listed in the table below. Matrix multiplication is a good example of parallel computing.
The detailed timing results are appended in this post for reference. Number of Cores.