• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

2900XT / 8800GTS Oblivion bench = On Par

fornowagain said:
Lol, thanks Tom. I know how it works and the two implementations are not directly comparable, other than graphically sort of. If you want to go frame to frame stick to 4xMSAA.

I know what your getting at the two brands use different methods to achieve 16xAA samples. I can do a bench with 4xMSAA tommorow but maybe will can test 16xAA nowto see which card can perform the best 16xAA sampling.
 
willhub said:
As for as most people are concerned it is since if someone wants to use 16xAA on either card they aint gonna use 4xMSAA.

With my 8800 2xMSAA gives about the same performance hit as 16xAA
The two may have a similar (debatable) graphical appearance at various 'named' resolves. But higher resolves are not implemented in the same way by the two cards. Both cards use the exact same sampling at 4xMSAA. Although different, looking at it the NV 8QxMSAA and Ati 8xMSAA Box filter seem near enough for testing. 16x is a waste of time for direct comparison, if you want to see the separation of the cards you need to increase resolutions with the same standard MSAA.



Has it occurred to you why a 16x coverage sample has a similar hit to a lower multisample?

They're both based on 4 or 8 multisamples and processed, i.e no further samples taken. For 16x the sample processing for the two cards is different. NV increases the color/z/stencil sample count and ATi uses data from neighboring multisamples to increase the count. Also iirc some of the ATi CF filters enable alpha transparencies as well.


NV.
16xCSAA Uses a 4x multisample + 16 Coverage samples
16QxCSAA Uses a 8x multisample + 16 Coverage samples
16SxSSAA Uses a 2X2 (rendered twice the res in xy) + 4x multisample.

ATi.
16xCFAA Wide tent uses a 8x multisample + 8 Custom Filter samples.

 
Last edited:
Well anyway on my pc, increasing the CPU to 3.4 got these:

Frames, Time (ms), Min, Max, Avg
4883, 92863, 32, 97, 52.583

So CPU aint much good been increased.

The 2900XT will proberbly be better at higher res, but I'm sorry, I wont be going out and buying a better monitor, I cant afford it.
 
Im also pretty sure that ATi's AA Resolve is better than nvidias IQ wise. ATi's AF is not as good though?.

From that review:

This CFAA mode with 8 samples produces extremely clean edges and does an excellent job of resolving very fine geometry, like the tips of the spires on the cathedral. Even 16X CSAA can't match it. Also, have a look at the tree leaves in these shots. They use alpha transparency, and I don't have transparency AA enabled, so you see some jagged edges on the GeForce 8800. The wide tent filter's subtle blending takes care of these edges, even without transparency AA.


I prefer the 4X MSAA + wide tent filter to anything Nvidia offers, in spite of myself. I've found that it looks great on the 30" wide-screen LCD attached to my GPU test rig

Oblivion has never looked better on the PC than it does on the Radeon HD 2900 XT.

Sum's it up TBFH. :)
 
Last edited:
It dont look much different on either card unless you take screenshots and zoom in to see which aint got as much jaggys :D
 
willhub said:
It dont look much different on either card unless you take screenshots and zoom in to see which aint got as much jaggys :D

I prefer ATi's as you can see the jaggies are almost none existant and blended as the sampling goes up with the CFAA. NV's are still slightly visible.
 
Tom|Nbk said:
I prefer ATi's as you can see the jaggies are almost none existant and blended as the sampling goes up with the CFAA. NV's are still slightly visible.


Maybe my eye's are not as good as yours, but I see no jaggys with 16xAA.
 
Back
Top Bottom