Things are working, but I'm not entirely sure why things are so slow. Surely, a bunch of glBegin/glEnd's are going to be slower but I'm getting like ~23 FPS with this image. Each unit cube has their own begin and end. Perhaps it's because they're 128x128 textures each... but that shouldn't be too bad for a GeForce GTX 560 SE... I can run Fallout: New Vegas.
I'm wondering how much VBOs/vertex arrays will help, and if a move toward the modern pipeline will speed things up... but still... a 50x50 set of cubes is pretty pathetic... I used to run far higher when I was doing my Minecraft-in-space game years ago.
I am running in a VM, which slows it "somewhat" (and does remove features like AA), but I'm pretty sure my Minecraft game was also in a VM... maybe not... I don't remember it's been so long.
Each bitmap has its own projection. So if you create the backbuffer, and then create another sub-bitmap of the backbuffer. They will both have their own, separate projections but will draw to the same surface.
That's a very confusion statement. How can a bitmap have a completely different projection from the global OpenGL stack?
I've also seen the same type of problem with the Nouveau driver, if you're on linux.
I've had some really crappy luck with them before as well. Lots of screen corruption issues.