new radeon drivers

[Doomtrooper]

All my games play fine what exactly are you looking for here ??

 

[Dolemite]

All my games play fine what exactly are you looking for here ??

Read my post above...or read it here, either way.

Hey OpenGL Guy,

I was hoping you could give us the inside perspective on the issue Carmack described in his .plan update back in February. Specifically, the unexpectedly low performance in a high poly count scene that may or may not be a hardware issue:

February 11, 2002

...

A high polygon count scene that was more representative of real game graphics under heavy load gave a surprising result. I was expecting ATI to clobber Nvidia here due to the much lower triangle count and MUCH lower state change functional overhead from the single pass interaction rendering, but they came out slower. ATI has identified an issue that is likely causing the unexpected performance, but it may not be something that can be worked around on current hardware.

Some people came to associate what he said with the low performance under GLExcess test #6 (I think), another high poly count test, which we all remember as the infamous "high poly count bug." That issue was fixed with a driver update, but I don't think we can safely make the assumption that this was the same problem, and was thus actually fixed.

I've asked Reverend to ask JC, but I'd be interested what you might know about it as well, especially on whether this is something that has been fixed, will be fixed, or is "unfixable."

I'm asking whether ATI was able to fix the problem that Carmack had with a "high polygon count scene that was more representative of real game [doom3] graphics," or whether they had come to the conclusion that it was "fixable" in the drivers or an actual hardware limitation.

So far I'm not getting an answer... , which leads to me to assume the worst.

 

[Doomtrooper]

Well a 8500 is running the same frames as a Ti500 in the leaked UT 2003 demo, which is a modern high poly game. This Demo doesn't even have specific optimaztions either i.e PS etc ..ATI doesn't just pop in and answer question when you want it , but it would be nice if they would

Believe me, I try

 

[Dolemite]

Hmmm, well from what I've seen UT2003 doesn't look like it has any more geometric complexity than Quake3, especially the leaked demo.

In any case, I think this issue may have more to do with Doom3 rendering specifics rather than simply polygon throughput in general.

Believe me, I don't expect the 8500 driver guys to just jump in here and explain it to me, but I was hoping OpenGL guy would refer me to somebody at ATI. I've also sent several emails and I'm waiting on a response.