Frank van Maarseveen schreef: > > > Not that it should matter much in this particular case but on modern > hardware FP can be just as fast or even faster than int, depending on > what you need to calculate. Concur to that. I've done in my own dlraw project a resize with filtering. Tried all lots of combinations on my PC and it turns out that the float was the fastest ! > Nowadays there's a lot of silicon dedicated > to FP on x86. On the other hand I would expect that integer arithmetic > has more opportunities for future SIMD. > > What does matter anyway is careful coding (an occasional look at the > generated asm is helpful), (data) cache footprint and openmp, I have > some dcraw_api.cc work pending in that area which I'll post soon. > >
------------------------------------------------------------------------------ Come build with us! The BlackBerry(R) Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9 - 12, 2009. Register now! http://p.sf.net/sfu/devconference _______________________________________________ ufraw-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ufraw-devel
