joevenzon Wrote:This isn't related to the problem and solution in this post, which has to do with the depth buffer setting, which isn't a video mode issue (even though it does throw up a video mode set failed error).
sorry should have started a new thread, i knew this isn't related
joevenzon Wrote:However, to answer your question, the way things currently work is a list of candidate modes is stored in that file, and then the video card is queried to see if it can support each mode.
yes, this is also how i understood the codewhat's weird that it goes trough the candidate list twice....
joevenzon Wrote:Modes that are unsupported aren't shown for you to pick. It's been a while since I wrote that code, but I believe the reason things are done this way rather than by using the SDL_ListModes function directly was that you get really unreliable behavior if you're running in a window rather than full screen -- like it'll just show one resolution for you to pick from.
so why don't simply just always query SDL_ListModes() with fullscreen as parameter?you should be (technically) able to run in any window size (that's smaller than the screen and a multiple of 16), but it's a common convention to let people choose the resolutions for windowed mode that their monitor can handle fullscreen...the current situation with the candidates is really less than ideal because all macs and many other machines currently ship with widescreen monitors....i really have to edit the config file by hand to get an acceptable fullscreen resolutioni also don't believe extending the videomodes.txt file with more resolutions is the proper solution...you will always miss some, and SDL_ListModes() already provides you exactly with the list you want, that is the resolutions provided by the machine you are running on.should i just try if i can get something working reasonably well and submit a patch...?
