NVIDIA GeForce 6800 Ultra: The Next Step Forward
by Derek Wilson on April 14, 2004 8:42 AM EST- Posted in
- GPUs
Anisotropic, Trilinear, and Antialiasing
There was a great deal of controversy last year over some of the "optimizations" NVIDIA included in some of their drivers. We have
NVIDIA's new driver defaults to the same adaptive anisotropic filtering and trilinear filtering optimizations they are currently using in the 50 series drivers, but users are now able to disable these features. Trilinear filtering optimizations can be turned off (doing full trilinear all the time), and a new "High Quality" rendering mode turns off adaptive anisotropic filtering. What this means is that if someone wants (or needs) to have accurate trilinear and anisotropic filtering they can. The disabling of trilinear optimizations is currently available in the 56.72
Unfortunately, it seems like NVIDIA will be switching to a method of calculating anisotropic filtering based on a weighted Manhattan distance calculation. We appreciated the fact that NVIDIA's previous implementation of anisotropic filtering employed a Euclidean distance calculation which is less sensitive to the orientation of a surface than a weighted Manhattan calculation.
This is how NVIDIA used to do Anisotropic filtering
This is Anisotropic under the 60.72 driver.
This is how ATI does Anisotropic Filtering.
The advantage is that NVIDIA now has a lower impact when enabling anisotropic filtering, and we will also be doing a more apples to apples comparison when it comes to anisotropic filtering (ATI also makes use of a weighted Manhattan scheme for distance calculations). In games where angled, textured, surfaces rotate around the z-axis (the axis that comes "out" of the monitor) in a 3d world, both ATI and NVIDIA will show the same fluctuations in anisotropic rendering quality. We would have liked to see ATI alter their implementation rather than NVIDIA, but there is something to be said for both companies doing the same thing.
We had a little time to play with the D3D AF Tester that we used in last years image quality article. We can confirm that turning off the trilinear filtering optimizations results in full trilinear being performed all the time. Previously, neither ATI nor NVIDIA did this much trilinear filtering, but check out the screenshots.
Trilinear optimizations enabled.
Trilinear optimizations disabled.
When comparing "Quality" mode to "High Quality" mode we didn't observe any difference in the anisotropic rendering fidelity. Of course, this is still a beta driver, so everything might not be doing what it's supposed to be doing yet. We'll definitely keep on checking this as the driver matures. For now, take a look.
Quality Mode.
High Quailty Mode.
On a very positive note, NVIDIA has finally adopted a rotated grid antialiasing scheme. Here we can take a glimpse at what the new method does for their rendering quailty in Jedi Knight: Jedi Academy.
Jedi Knight without AA
Jedi Knight with 4x AA
Its nice to finally see such smooth near vertical and horizontal lines from a graphics company other than ATI. Of course, ATI does have yet to throw its offering into the ring, and it is very possible that they've raised their own bar for filtering quality.
Facebook
Twitter
RSS
77 Comments
View All Comments
Marsumane - Wednesday, April 14, 2004 - link
This card owns... Anyone know when it ships to retail stores? Guesses even?SpaceRanger - Wednesday, April 14, 2004 - link
I'd like to see what ATI comes up with before I make my decision. I rushed to judgement back when the GF4 TI4600 came out, and regretted making the quick call to buy. If I don't have to get a new PSU for the ATI solution, I'll consider it, even if performance is 5-10FPS slower. Adding 100 bucks to the already costly 500 for the card doesn't justify the expenditure.gordon151 - Wednesday, April 14, 2004 - link
AtaStrumf is so right. More than likely you'll be able to buy the X800s before you can buy this.Shinei - Wednesday, April 14, 2004 - link
Well, I'm sold. Yeah, that sounds fanboyish, but this thing is a solid performer and doesn't require me to completely replace my display drivers... Even if ATI wins by five FPS and has a lens flare in a forgotten corner of a screenshot that you have to stare at for ten minutes to spot, my money is going to NV40--assuming the prices come down a little. ;)Speaking of DX9/PS2.0, what about a Max Payne 2 benchmark? I'm curious what NV40 can do on that game with maxed out everything... :)
skiboysteve - Wednesday, April 14, 2004 - link
i love anandtech's deep technical reviews but yall did no where near enough testing, the xbit article does a hell of allot more testing, 48 pages!http://www.xbitlabs.com/articles/video/display/nv4...
the card fucking rapes everything.
the anand tests dont show nearly the rape the xbit ones do...
AtaStrumf - Wednesday, April 14, 2004 - link
I find it really funny when people say that they will wait until ATi releases their X800 to make up their buying decisions.It's not you can run out and BUY this card right now or tomorrow. Of yourse you will wait. You don't really have a choice :)
ChronoReverse - Wednesday, April 14, 2004 - link
The Techreport tested out the total power draw of this thing and it only drew slightly higher than the 5950 (both of which draws more than the 9800XT).So it seems the recommendation isn't actually necessary (and my Enermax enhanced 12V lines will take it easily).
Pete - Wednesday, April 14, 2004 - link
mkruer #27, all the reviews I've read mention $500 for the 6800U, and $299 for a 12-pipe 128MB 6800.DerekWilson - Wednesday, April 14, 2004 - link
#27,The 6800 Ultra (which we tested) will be priced at $500
The 6800 (with 12 pipes rather than 16) will be priced at $300
Pete - Wednesday, April 14, 2004 - link
quikah #26: FarCry comparison screens are at HOCP.http://hardocp.com/article.html?art=NjA2LDU=
Apparently PS3 wasn't enabled, but the 6800U looks better than the 5950U running PS2. It's still uglier than the 9800XT, sadly. Banding abounds, both here and in FiringSquad's Lock-On screens. Puzzling, really. If the 6800U really runs FP32 as fast as FP16 within memory limits, I wonder if all it will take to get IQ on a level with ATi is forcing the 6800U to run the ATi path or removing the NV3x path's _pp hints.