RE: Eigen 3.3 vs 3.2 Performance (was RE: [eigen] 3.3-beta2 released!)

[ Thread Index | Date Index | More Archives ]

On Wed, 1 Aug 2018, Daniel.Vollmer@xxxxxx wrote:

Attached you will find an extracted subset, which may point to one problematic area with partial vectorization as for this example Eigen 3.3 is ~13% slower (unless I'm measuring incorrectly).

Not what you are looking for, but the main difference between gcc and clang here seems to be that gcc fails to turn the divisions by rho into multiplications by 1/rho (computed once).

Marc Glisse

Mail converted by MHonArc 2.6.19+