Marti, Am 03.12.03, 16:35 +0100 schrieb Marti Maria:
> I've managed to set correct permissions. Seems latest versions of Cygwin does > support atributtes too. So, many thanks for point this > out. Thanks for changing this. > Regarding floating point, I really appreciate any contribution. But I'm afraid this > would be a bit dangerous :-) > > Please note than changing mode from assembler to floating point does result in > nothing but a general slowdown. > There was a lot of work to assure fixed and floating point math were giving same > results, so, you will not get any extra precission > by compiling in floating point mode. All will work same, but slower. The floating > point mode is intended as a way to describe > algorithms. Then, the optimized version of these algorithms is provided for speed. > Donald E.Knuth said "premature optimization is > the source of all evil" so, I keep original algorithms under USE_FLOAT compilation > mode. Just for clarity sake. The accurancy of I understand. That is fine to make the code more transparent. So I have more easy reading. > lcms is 16 bit, a lot more than strictly needed, which would be 11-12 bit at most. > You can still use TYPE_RGB_DBL, etc. despite the > engine are using fixed point. Oh, I have to look much deeper in this. Regarding the configure changes. Would You like to take over the assembler switch only. As You described, the assembler optimization only works for pentium based systems. Switching this part of could help make Your library more portable? I am not aware of compiling lcms on other processors. regards Kai-Uwe ------------------------------------------------------- This SF.net email is sponsored by OSDN's Audience Survey. Help shape OSDN's sites and tell us what you think. Take this five minute survey and you could win a $250 Gift Certificate. http://www.wrgsurveys.com/2003/osdntech03.php?site=8 _______________________________________________ Lcms-user mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/lcms-user
