The program vectopt could be applied to check efficiency of several implementations of the vector operation:
<Vector 1> += <Vector 2>.It produces results like these:
Results: (for dimension 1000000, cycles 100)
--------
implementation CPU time
-----------------------------------------
LASPack 2.518e-01 s = 100.0 %
1 3.548e-01 s = 140.9 %
2 3.196e-01 s = 126.9 %
3 2.481e-01 s = 98.5 %
4 2.392e-01 s = 95.0 %
5 2.395e-01 s = 95.1 %
-----------------------------------------
For details to the implementations look at the file laspack/examples/vectopt/testproc.c. The current LASPack version corresponds to the implementation 2 and 3 depending on the computer architecture, respectively. They are, in comparison with LASPack , simplified and can be therefore a little faster.