PDA

View Full Version : NV3X subpixel precision in 3DMark03


AnteP
05-14-03, 09:26 AM
I just installed the 44.03 aka Detonator FX drivers on my testing rig and fired up the Abit FX 5800 (which has a really silent fan compared to other FX solutions I've heard btw).

The old problems with the dull sky etc. are gone. Performance is super (4700 point, this is a non Ultra btw).
But, therre seems to be a problem with the subpixel precision. Crawling pixels here and there. Especially noticable in GT1 on the planes and mostly in the scene in Mother Nature where there's a stone table.

Just miscolored "dancing" pixels, often close to the edges of the polygons.
Sort of looks like when you've overclocked to much and start to get bad pixels, but of course it's not anywhere near as bad as that.

I tried capturing some screenshots of it but it didn't show what I saw in motion. I'll try some more, perhaps until someone who's using an FX with these drivers can look and see if they are experiencing the same thing?

Thanks.

StealthHawk
05-14-03, 12:32 PM
Can you try the WHQL 43.51 drivers and see if you have the same problem? Also it would be interesting to see how those drivers affect your score.

AnteP
05-14-03, 12:44 PM
Originally posted by StealthHawk
Can you try the WHQL 43.51 drivers and see if you have the same problem? Also it would be interesting to see how those drivers affect your score.

Sure thing, bbl

SurfMonkey
05-14-03, 05:12 PM
If they don't show in screen captures then they are probably post processing glitches.

John Reynolds
05-14-03, 05:17 PM
No, those dancing pixels are being caused by something else, something that will probably be revealed fairly soon.

SurfMonkey
05-14-03, 05:22 PM
It could also be alternate frame rendering going wrong. Would nvidia could corners as drastically as that?

AnteP
05-14-03, 05:33 PM
Originally posted by John Reynolds
No, those dancing pixels are being caused by something else, something that will probably be revealed fairly soon.

Yeah but when? :)
I've postponed the review I'm working on until I find out more about this...

digitalwanderer
05-14-03, 05:35 PM
Originally posted by John Reynolds
No, those dancing pixels are being caused by something else, something that will probably be revealed fairly soon.

*COUGH-COUGH*motionbasedantialiasing*COUGH-COUGH* ;)

SurfMonkey
05-14-03, 05:50 PM
Originally posted by digitalwanderer
*COUGH-COUGH*motionbasedantialiasing*COUGH-COUGH* ;)

motion based antialiasing can actually be quite cool, I don't think that would be a particulary bad thing. It would be pretty intensive though, you have to do both spatial and temporal antialiasing and then post-filter it.

What would be bad is if you were using lower precision (fixed point?) in the vertex shader in order to achieve the speed increase

digitalwanderer
05-14-03, 06:04 PM
Originally posted by SurfMonkey
motion based antialiasing can actually be quite cool, I don't think that would be a particulary bad thing. It would be pretty intensive though, you have to do both spatial and temporal antialiasing and then post-filter it.

What would be bad is if you were using lower precision (fixed point?) in the vertex shader in order to achieve the speed increase

There ain't no easy way to test what precision they're using in the vertex shaders to find out either, I'll wager. :(

StealthHawk
05-14-03, 11:26 PM
Could this be the answer? (http://www.nvnews.net/vbulletin/showthread.php?s=&threadid=11826)

MrNasty
05-15-03, 12:20 AM
Originally posted by StealthHawk
Could this be the answer? (http://www.nvnews.net/vbulletin/showthread.php?s=&threadid=11826)

Just reading the title of this thread it was a nobrainer for me.