Re: [AD] About Elias' bug

["Eric Botcazou" <ebotcazou@xxxxxxxxxx>]

> The patched-normal has almost no speed loss (might be inside the accuracy
> of my timing function),

11% for you though...

> and the patched-keep is a bit slower than with the blit2.diff.

Yes, I had to work around the weird optimization problem with GCC.

> Also, I noticed that my test program crashes with the blit3.diff, whenever
> I try to convert from 8bit to any other color-depth. (It may be a bug in
> my program, but I couldn't find anything yet)

No, my fault, thanks for pointing it out.

> I was thinking about the function get_replacement_color() - it's probably
> easier to just define the replacement color as makecol(255,1,255) for true
> and makecol(255,8,255) for highcolor.

The get_replacement_color() function from blit3.diff is an useless mess and
doesn't work well for conversion to 8-bit (as it was meant to). I never
wrote it ;-)
Anyway I'm a perfectionist and just for the sake of the 16-bit mode I'd
rather use your original function than makecol(255, 8, 255).

> If makecol is changed to never return 0 in palette modes, then there a
> standard makecol will get a replacement as well.

Quite right. But I just changed my mind about makecol8() :-) See the next
thread.

Corrected patch attached.

--
Eric Botcazou
ebotcazou@xxxxxxxxxx

Attachment: blit3b.zip
Description: Zip compressed data