SDL presents a very simple interface to the display framebuffer. The +framebuffer is represented as an offscreen surface to which you can write +directly. If you want the screen to show what you have written, call the update function which will +guarantee that the desired portion of the screen is updated.
Before you call any of the SDL video functions, you must first call +SDL_Init(SDL_INIT_VIDEO), which initializes the video +and events in the SDL library. Check the return code, which should be +0, to see if there were any errors in starting up.
If you use both sound and video in your application, you need to call +SDL_Init(SDL_INIT_AUDIO | SDL_INIT_VIDEO) before opening the +sound device, otherwise under Win32 DirectX, you won't be able to set +full-screen display modes.
After you have initialized the library, you can start up the video display in a +number of ways. The easiest way is to pick a common screen resolution and +depth and just initialize the video, checking for errors. You will probably +get what you want, but SDL may be emulating your requested mode and converting +the display on update. The best way is to +query, for the best +video mode closest to the desired one, and then +convert +your images to that pixel format.
SDL currently supports any bit depth >= 8 bits per pixel. 8 bpp formats are +considered 8-bit palettized modes, while 12, 15, 16, 24, and 32 bits per pixel +are considered "packed pixel" modes, meaning each pixel contains the RGB color +components packed in the bits of the pixel.
After you have initialized your video mode, you can take the surface that was +returned, and write to it like any other framebuffer, calling the update +routine as you go.
When you have finished your video access and are ready to quit your +application, you should call "SDL_Quit()" to shutdown the +video and events.