Recent content by Hyp-X

Sadly the review didn't say how they performed the Crysis benchmark. For example particle effects usually have high BW requirements. If they benchmarked a flyby demo they wouldn't get that many particle effects, but it would have a lot of streaming which can cause CPU limitations. For...

Have you tried using glDrawElements(...) instead of immediate mode or display lists?

Well it would be an improvement if you didn't use point sampling. It looks like you used GL_NEAREST_MIPMAP_LINEAR for GL_TEXTURE_MIN_FILTER and GL_LINEAR for GL_TEXTURE_MAG_FILTER. For some reason I can see lighter stripes in the shadow of rocks - looks like the shadow of the grass - but it...

Not if your eye is in the focal point (virtual view point). The problem is that it works with one eye only - the distance between the two eyes are too big (for normal screens), so you will always see distorsion no matter where you look at, or where you place your head. Yes I agree that IMAX...

Yes the GS program operates on a single primitive and can output multiple primitives. It can however use information from adjacent vertices if the topology information is provided (that means more indices per primitive). OTOH, I don't see why couldn't multiple primitives be run paralelly...

By the time low-end D3D10 hardware will have sufficiently high user-base (that is 3-4 years from now), Vista requirement might not be an issue... Until that the only pure engines will be in techdemos and benchmarks. But I'm sure some developers will start to work on some of those engines as...

Exactly. Those engines will be general enough to run on D3D9 and D3D10 so you'll lose most of the efficiency improvements of the D3D10 API.

That MS realizes that they are stupid not making D3D10 backward compatible with D3D9 cards.

I think it might be a good thing to provide a standard (I can dream) API that allows the game to access the AA and AF level set in the driver control panel and apply it in the game manually. The driver wouldn't really force these values on detecting the application supports the API. It...

If the vertex Z has 1/2^24 error that the interpolated values will also have 1/2^24 error even if you calculate them with 128-bit arithmetic. This is the same thing Humus is saying: Interpolation can't help you.

Yes, using a FP32 W buffer makes sense, and the Vertex Shaders definately has enough precision is that. Also what I meant is that Z is computed with a (z-N)/z like equation which result an approximate 24bit precision in the whole [0, 1] range. This is because how the FP subtraction behaves.

I know that NV2x cards had 4 32-bit DDR buses and they had a smallest access of 64 bit (8 byte). Their Z compression method were optimized for 64 bit access size. R300 introduced 4 64 bit DDR buses so I assumed thay had a smallest access of 128 bit (16 byte). Has the burst length of 4 been...

The thing is that Z has a range of [0, 1]. Z is computed in the vertex shader. The vertex shader uses FP32. FP32 will give you a precision of 24 bits at best (given how Z is usually calculated). So to make actual use of 32 bit Z we'd need a higher precision vertex shader.

Ok, I understand the message. However I was under the impression that the shortest read of the GPU's are usualy 16 byte. That's 648KB for 16x16 tiles which is about 26% above the optimal - I wonder if <=X850 cards have this inefficency, or if it is solved in the X1000 series. 26% is still a...

d) none of the above? My guess would be: 8x8/256B reads 2 MB 8x8/64B reads 1.125 MB 16x16/256B reads 1.125 MB 16x16/64B reads 800 KB