In article <3C7BE5C6.22003.968563@localhost>, Tim Wegner <twegner@swbell.net> writes:
I wonder how well OpenGL handles the sort of pixel graphics we need.
The key to all modern graphics APIs is to avoid the "VESA memory-mapped model" of diddling pixels one at a time. You need to batch up pixels into blocks for updating. This is true of Direct3D and OpenGL, particularly if you're trying to manipulate pixels directly in video memory -- which is horribly slow. Its often faster to manipulate a block of pixels in the system memory and request the card do the transfer to video memory by DMA or something. -- Ask me about my upcoming book on Direct3D from Addison-Wesley! Direct3D Book <http://www.xmission.com/~legalize/book/> Don't Support Spammers! Boycott Fractal Painter 7! <http://www.xmission.com/~legalize/spammers.html>