powly k

Things faster than A* include, for example, summing two numbers, calculating a dot product or doing nothing at all. Most problems have many good solutions, and the best one must be picked by you depending on the specifics. And even better than a good solution is avoiding the problem altogether, which is also sometimes possible.

Winding is probably not it - if you don't specifically enable backface culling, it's not on. What I'd try is first negative z values for the vertices and then ditching the matrix multiplication to see if the matrix is really okay.

What you do in your shaders tends to matter a lot more than how many triangles you rasterize. The only actual way to know these is, like you said, to test it yourself.

That very much seems to be the issue, if averaging 4 images that should be different doesn't produce a blurred result. Can you actually somehow verify that the 4 samplers have different textures attached?

Okay, since you can't actually read from an FBO but only use an FBO to render to a texture and then read that texture, here's the question: do you set up the active textures, bindings and uniforms correctly? How this should go is like this:

set active texture to N

bind your texture[N] to GL_TEXTURE_2D

glUniform1i(uniform N location, N)

N loops through 1...4, obviously. What this sounds like is you forgetting to call the gluniform1i - all the uniforms samplers are texture unit 0 unless you specify otherwise.

And please, do "uniform sampler2D Blur1, Blur2, Blur3, Blur4;" instead of what you do now. And to be extra sure, always set the alpha to something.