if i have an image that is square and a power of 2 ie 512 x 512 should I not use mipmapping? It seems to be rather slow and uses more memory. if I have an image that is 1024 * 768 is mipmapping the most efficient way?
if i have an image that is square and a power of 2 ie 512 x 512 should I not use mipmapping? It seems to be rather slow and uses more memory.
I can’t say whether you should use mipmaps or not, since that depends on what you use the texture for. Mipmaps consumes more memory, but they can also improve image quality and performance. If you don’t think it’s worth the extra memory usage, don’t use it.
if I have an image that is 1024 * 768 is mipmapping the most efficient way?
All textures must be a power of two, and 768 is not a power of two. So you must rescale that image, or upload the data as a subimage in a larger texture.
[This message has been edited by Bob (edited 10-04-2003).]
I dont think you understand my 2nd question. I realize 768 is not power of two but what I am asking is should i use gluBuild2DMipmaps to scale the image or should i scale it before in an editor before I load it. mipmapping will take care of that for me but it seems to be slow
When pre-scaling the image you have more control over the scaling process, but at the cost of more disk space usage. Loading time is faster, since you don’t have to rescale the image. If you think it’s worth it, consider building all your mipmaps, not just the base level, and just pass them to OpenGL. Costs more disk space, but can load even faster.