More pages: 1 2
3 4 5 6 7
New dynamic brancing demo
Thursday, July 1, 2004 | Permalink
Here's a demo that does dynamic branching, without the need for pixel shader 3.0, and still receives the huge performance boost that pixel shader 3.0 dynamic branching supposedly gives when utilized in a similar fashion.
Enjoy!
2004-07-03:
Since it seems people have lost the ability to understand smilies I removed the first line of this post. Couldn't have imagined that people would feel so seriously offended by it, and I certainly didn't foresee the amount of crap I would recieved for it.
2004-07-04: Did a minor update to the demo to work around the performance drop issue on nVidia cards. The demo will now let you choose between doing a full stencil clear or simply zero it for surviving fragment. The former method seems to be required for this technique to see any performance gains at all on nVidia hardware, while both methods run fast on ATI card, the latter at a higher speed though. In order to get maximum performance it will choose zeroing as default for ATI and full stencil clear on everyone else. I don't know if that assumption holds true for other vendors though.
Da3dalus
Friday, July 2, 2004
Very nice demo, Humus
g__day
Friday, July 2, 2004
Once again Humus - very impressive!
Anonymous
Friday, July 2, 2004
nVidia owns ATI this time round...
Wester547
Friday, July 2, 2004
LOL, pretty persistent to get around nVidia's future-proof shader technology, aren't you humus? That's one damn impressive demo, but keep in mind the jump that Shader Model 2.0/2.x took to Shader Model 3.0 isn't as great as 1.x to 2.0/2.x, just like that jump itself isnt as great as the fixed Hardware T & L pipeline was to the Programmable Shaders introduction, so especially with Shader Model 2.x(the enhanced iteration with 2.0 used with the nVidia GeForce FX and ATi Radeon X800 series), this demo is still very possible.
Richteralan
Friday, July 2, 2004
Obviously my problem been ignored @ Beyond3D.
If being mature, I'd rather not say the first sentence in your demo description.
Kombatant
Friday, July 2, 2004
Excellent as always pal; always eager to try out your new demos
Humus
Saturday, July 3, 2004
Sunray & sqrt[-1],
I've been experimenting some with scissors and clipping planes at work today. Of course the benefits goes down when you use other means to reduce the workloads too, and that's equally true for ps 3.0. But there's still a very significant performance increase with this technique, even when you enclose a bounding box of six clipping planes around the light and enclose the light in screen space with a scissor rectangle. We may no longer be talking about 3x performance boost, but still easily 30-50%.
Regarding texture lookups, well, we won't be able to do everything vs3.0 can, but we can do some similar stuff the day we get the super_buffers extension (if ever, seems to be progressing very slowly ...).
Humus
Saturday, July 3, 2004
Anonymous,
yes, it has been around for a while. It has been used for implementing high level shading languages on lesser hardware, but I don't know if it ever has been used as a means to boost performance. But that's not a very interesting point anyway, what matters is that it works and is fast.
More pages: 1 2
3 4 5 6 7
