I'm also having trouble with this. If I compile the shaders at startup I don't catch all the states that's probably in the "shader key" and I get bad performance.
Even if I let the program run a...

I tried 320.18 again and this time it worked perfectly...:confused:.
I haven't touched my program/game since last time.
Maybe it has something to do with some internal shader-caching that needs to...

Well it seems that you know what your're talking about.
Just one question... how do you test that normal is zero?

What texture format do you have for your position-buffer?

Well it could be anything but when I have similar problems I usually render the values from those buffers to get a visual view of the content.
result = max( vec3( 0.0 ), normal.xyz );
...and so on....

Yes its good.
Because when sampling on the edge of the mountain its important to not read the pixels that belonging to the sky, and with linear sampling you do just that.

Are you using nearest/point sampling when reading from those textures?

Thanks for the help.
The problem is that it works perfectly with other versions so before digging deep I have to ask others, because it could be a driver thing.
I switched back to 320.00(beta) and...

I forgot to say that I'm running 4.2 core version.
The problem also exists in debug-context, and I get no errors from that.

Well I get the feeling that I'm alone with this problem ;).
The scenes are also rather "heavy" with lots of shaders, textures and so on.
I don't think I have this problem in simple scenes, but it...

Anyone else experienced performance slowdown in this case?
Personally I have a 4x(fps) slowdown. Every other version have worked perfectly until 320.18.

Win7(64),gtx580

Do you mean eigen-vectors that defines the orientation of the box that minimizes the volume wich enclosing the point cloud?

There is also this:
http://www.matthiasmueller.info/publications/FTLHairFur.pdf

Try GL_RGB32F as internalformat for glTexStorage2D.

Maybe you are missing the "-1" in:
glReadPixels(x, viewport[3] - y, 1, 1, GL_RGB, GL_UNSIGNED_BYTE, pixel);
Try:
glReadPixels(x, viewport[3] - y - 1, 1, 1, GL_RGB, GL_UNSIGNED_BYTE, pixel);

You are sending in "(GLvoid *)0" into glDrawElements, do you use buffer objects?
If not then you should pass in "indices" instead.

Completely agree

This link describes a scenario that looks like yours.
http://forum-archive.developer.nvidia.com/index.php?showtopic=6720&mode=threaded&pid=21223

Regarding PBO and orphaning, its just that you...

Perhaps using PBO with orphaning might help.

But maybe a 4/3 ratio...

Have you tested with multiple FBO:s, one for each surface?

You could look at an implementation directly:
http://www.cs.berkeley.edu/~jfc/mirtich/massProps.html

http://www.multires.caltech.edu/teaching...rtich96fast.pdf

Declare the sampler as "uniform usampler2D myTexture"

What's the performance when you disable blending?