matthiasm wrote: > > On May 15, 2007, at 10:36 AM, MacArthur, Ian ((SELEX)) ((UK)) wrote: > >> This window corruption the Mariwan and Albrecht have seen... Is anybody >> else seeing it? I do not see it here - maybe it has been fixed already? >> Or is it some peculiarity of their set-ups? >> > > I have not looked at the source code yet, but this problem has been > reappearing many times. The problem is, that during high CPU loads (and > the color chooser creates a high load), multiple refresh requests are > sent (WM_PAINT), but only the first one is served correctly. > > I have fixed this at least a dozen times, but somehow either the code is > changed back, or other side effects keep messing up the fix. Or there > simply is no 100% fix, because I do occasionally see this problem on > Microsoft software as well. > > Anyway, I will dive into this yet again...
Maybe I can help with this additional (though very strange) info: My system is a dual processor "Pentium(R) D CPU 3.20 GHz" with 2 GB RAM, WinXP SP2, IE 7 installed, and I can easily reproduce the problem. On a one processor system "Pentium(R) 4 CPU 2.40 GHz" with 512 MB RAM, WinXP SP1, I had to try really hard, but I can sometimes reproduce it. As I wrote before, this seems only to apply to single buffered windows. Another strange observation: test/doublebuffer.exe seems not to have the problem in either window (single and double buffered), but it flickers very much in the single buffered window. Maybe it does always draw the whole window ? Albrecht _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
