New AA options in Detonator

[trinibwoy]

Horizontal lines (NV40 8x) are much better than vertical. EER is 8x4, so ATi (6x with EER 6x6) gives better results at vertical lines, horizontal lines are worse (compared to NV40 8x).

Yep that's what I found too. So this EER thing is what I usually call "sample resolution" huh? What's EER stand for? I'm guessing the R is resolution!

 

[Xmas]

Edge equivalent resolution (because it only makes a difference at edges, even for supersampling), a term coined by Matrox AFAIK.

 

[Moloch]

Just for the record that 8xS screenshot from the NV40 is showing a 4*8 EER; would you be willing to ignore to ignore supersampling performance penalties even more, 16x (4xRGMS + 4xOGSS) yields actually an 8*8 EER and in terms of EER exclusively it's equivalent to Voodoo5 6000's 8xRGSS.

Frankly I was hoping for a few more speculations about what those hypothetical new AA tidbits for G70 could stand for. Instead I read almost 3 pages of rechewed stuff from the past.

Yes there is a problem with NVIDIA's new angle-dependant AF algorithm, but it seems to be purely between the algorithm and applications not recognizing default "0" LOD values. I recall being amongst the first to officially document in a review said issue and speculated back then where the problem originates from and apparently I haven't been much off base either; call it beginner's luck if you please since I don't really need laurels either

It appears strictly with AF enabled and it doesn't appear due to filtering optimisations or goes away when you switch to high quality either. Furthermore even Supersampling won't cure it either, just push the problematic area by a -0.5 LOD shift further away (with 2xSSAA that is, with 4xSSAA by -1.0) and so on. If you colourize in an application MIPmaps you might notice that the forementioned problematic area is only a "band" of the size of one MIPmap and it usually appears before the first red MIPmap.

In most cases switching the negative LOD bias slider to "clamp" cures the problem; if not - and that in rare cases - a bit of extra positive LOD might be needed also. The question would be why it didn't turn up on former GeForces or why it doesn't happen on competitive products: I'll call it an "algorithmic bug" for simplicity's sake, otherwise an added driver switch for it wouldn't be really necessary.

One thing I agree with here, are Democoder's points about texture aliasing; I myself am extremely finicky with texture aliasing and way too many people out there consider "extra sharpness" as ideal. On a still screenshot and up to some degree probably yes, yet in real time motion I don't want any dancing meanders across my screen. And yes apart from any prementioned bugs (or call them whatever you want) the primary focus for hardware designers - especially with increasing shader lengths - for antialiasing should be more in the texture than on the polygon edge department. I won't deny that I'll be a sucker for a 8*8 or even higher EER, but if I get tons of texture aliasing my apetite for antialiasing deminishes quickly. And yes I can understand when Democoder says that it's often that bad that he'll rather swith off MSAA entirely.

Uhmmm wait I'm ranting again.....oooops

Indeed.
When I used my radeon 8500, I thought the sshots I took looked nice since they were so sharp, but during game play, there were several games that had bad textures aliasing/shimmering, like serious sam and never winter nights.
When I got my 6600GT I was plesently surprised.
One thing I don't like is in racing games, there's aliasing which seems related to filtering, as it's only in the distance, not right up close.
Notice the whole line.
Is that a result of angle dependant af?

 

[trinibwoy]

Edge equivalent resolution (because it only makes a difference at edges, even for supersampling), a term coined by Matrox AFAIK.

Thank you sir!

 

[Ailuros]

When I got my 6600GT I was plesently surprised.
One thing I don't like is in racing games, there's aliasing which seems related to filtering, as it's only in the distance, not right up close.
Notice the whole line.
Is that a result of angle dependant af?

You mean the white line on the road? Nope it'll just need better antialiasing to smoothen it even further from what it looks like.

That's from a Nascar game right? It uses the same engine as F1 2001/2 did if I'm not mistaken. If there's any sort of texture aliasing present (especially in the far clipping distance and on the left and right side of the tracks, where the crowds are f.e.), then it's more like the game's fault. I still play a couple of rounds in F1 2002 occassionally and I can't stand anything lower than 4xS in 1600 and yes it can get that bad.

Many games out there present either lacklustering MIPmapping, Z- or texture aliasing or Lord knows what else. I wish developers would be not only more careful with details like that, but wouldn't treat antialiasing these days as a a luxury feature.

 

[Moloch]

When I got my 6600GT I was plesently surprised.
One thing I don't like is in racing games, there's aliasing which seems related to filtering, as it's only in the distance, not right up close.
Notice the whole line.
Is that a result of angle dependant af?

You mean the white line on the road? Nope it'll just need better antialiasing to smoothen it even further from what it looks like.

That's from a Nascar game right? It uses the same engine as F1 2001/2 did if I'm not mistaken. If there's any sort of texture aliasing present (especially in the far clipping distance and on the left and right side of the tracks, where the crowds are f.e.), then it's more like the game's fault. I still play a couple of rounds in F1 2002 occassionally and I can't stand anything lower than 4xS in 1600 and yes it can get that bad.

Many games out there present either lacklustering MIPmapping, Z- or texture aliasing or Lord knows what else. I wish developers would be not only more careful with details like that, but wouldn't treat antialiasing these days as a a luxury feature.

Ya, nascar simracing.
Hmm.

 

[Reverend]

By comparison, the differences in FSAA are enough for me to buy one card over another.

I doubt that's the majority's thinking/criteria.

Otherwise, 3dfx may still be around.

 

[Kombatant]

nVidia's 8x mode is indeed very comparable to ATI's 6x MSAA, I hardly remember any differences (if any). It's just that, due to the SSAA that takes place in nVidia's mode, the performance drop is larger.

 

[PeterAce]

Well I'm sensitive to texture aliasing and I was one of the ones who could notice a slight 'shimmer' increase between my 9800 Pro (Trilinear) and my X800 XT ('Fast Trilinear' AI=ON) but it was only certain very hi-frequency textures that showed up this case.

Also my 6800 GT I noticed more texture aliasing until we changed the Forceware driver settings to:

AF mipmap optimization = Off
AF sample optimization = Off
Trilinear optimization = On.

'Brilinear' (the Trilinear optimization) seams quite tweaked to be acceptable now on NV40 (as long games arn't setting some negative LOD).

The 'AF sample optimization' significantly increased the texture 'shimmer' while set to On.

And I completely agree with DemoCoder and Ailuros - that shader aliasing or 'in-surface aliasing' on titles like Farcry (bump-mapped walls in caves etc) and Chronicles of Riddick (on many surfaces) cause me such a drop in realism due to shimmering and twinkling of the shaded pixels.

 

[KimB]

It'd be really nice if future games allowed quality options that allow the enabling of shaders that decrease aliasing from bump mapping and the like.

 

[Blazkowicz]

I'm not really satisfied with texture filtering on my ti4200 too (8x AF), even if it does true trilinear filtering and angle independant AF; the way textures behave when moving buggers me sometimes in HL1 mods, that's especially visible in Day of Defeat, on the ground texture of dod_avalanche.

Strangely I tried to set a positive LOD in rivatuner and didn't see a difference. maybe there's some noise ultimately coming from the textures themselves, I don't know, or the LOD setting did not work.

I'd like to see perfect texture filtering on HL1 games (as for good anti-aliasing I'll wait for geforce 7600, the gf4 is so bad I stopped using AA; I only recently upgraded to it, and I upgrade the GPU every 1.5 year or so)

 

[KimB]

The Ti 4200 has about the best texture filtering available anywhere. Your problems are most likely related to alpha tests, which no texture filtering in the world can take care of.

 

[Ailuros]

The Ti 4200 has about the best texture filtering available anywhere. Your problems are most likely related to alpha tests, which no texture filtering in the world can take care of.

Supersampling does. While not an absolute cure either (due to low sample density) if he's playing in a not too high resolution, 4xS might be an option.

 

[KimB]

Well, yeah, but that's not texture filtering.

 

[Ailuros]

Well, yeah, but that's not texture filtering.

Supersampling also filters textures as you very well know.
Under that sense:

....which no texture filtering in the world can take care of.

Texture filtering within Supersampling can.

 

[Ostsol]

Well, yeah, but that's not texture filtering.

Supersampling also filters textures as you very well know.

Supersampling does not filter textures. It simply has a positive effect on the quality of rendered textures.