In my iPhone app, I have about 100 full screen images for image sequence core animation. The reason that I use such a large image sequence animation is that video playback is not flexible and powerful enough in SDK 3.0.
My images are not fully covered. For OpenGL, we can pack the valid content of several images into one large and power-of-2 texture (use different texture coordinates to access different parts of the image), so that the images are compressed and performance is highly improved. (named Texture Atlas in Cocos2d.)
For Core Animation, we don't have the texture coordinate mechanism, how can we do the same packing optimization?