A promise is a promise :) there you go!

Right, cuz the edges are so much better.

Sure you've admitted that ATI's MSAA is better than NVidia's, but you've never been excited about it, not the way you are about this inferior antialiasing.

Seriously, I had so much more respect for you before this NV40 business came about. You had some legitimate gripes about ATI's AF, but I can't see you as much more than a fan_boy now.

For everybody else, I promise, this is my last post on the subject.

 

nVidia at Gdc....

http://www.nvidia.com/object/IO_12112.html

 

ROFL, now that's funny.

 

and the nVidia marketing machine begins.

 

So what?

 

Just thought I would try something out.

Mystery Card benchmarks:
UT2004
All tests @ 1600x1200 no AA/AF

Torlan - 61fps
Convoy - 54fps
Bridge of Fate - 106fps
Collossus - 90fps
Rankin - 88fps

Maybe it's too easy, but you never know .

 

Let me guess: 9800XT

 

You'll never guess what the Inq just put up a link to:

Purported NV40 Benchmarks-- you can take a look at them here.

They look very familiar indeed.

 

Some people must just think poor lil EB is made of bandwidth.

 

I wasn't disagreeing that NVIDIA PR may give insight to shader performance. However, I was disputing that they would tell us anything useful about FSAA/AF.

One thing I forgot to point out was NV30 marketing claims of "free FSAA" and "eliminating jaggies forever." Two things that weren't exactly true. Sure 2x and QCA had "only" a 10% performance hit, but that still wasn't free, and it only applied to the 5800Ultra. Also, good luck in eliminating jaggies forever with NVIDIA's FSAA modes on NV30. I still remember the big stink in forums when those claims were shown to be untrue.

 

I actually don't remember those claims. About what was the timeframe? Was it during the original "paper" launch? Or a while before?

 

They're still present on Nvidia's site. http://www.nvidia.com/object/feature_intellisample.html

________
Marijuana pictures

 

If that's what you are reffering to then the document in question is http://www.nvidia.com/object/intellisample_tb.html - gives a pdf file and at the bottom of the first page . document dated 2/10/2003.

I've got to say the marketing hype does very much resemble the hype going on before the NV30 launch - in some cases it's almost word for word , just replace a few "shader model 2.0" with "shader model 3.0" and a few "nv30" with "NV40" and it could almost be exactly the same pre-launch marketing material .

http://www.nvidia.com/page/fx_5800.html

Mark

EDit SsP45 beat me to it

 

http://www.guru3d.com/article/article/125/9/

Anyone at the Cebit happens to witness the demo of MSI?

I know it would be under NDA but could at least say whether it is disappointing or promising?

 

Old news

 

Really? Nvidia has been maken that claim since the Geforce 3, well the Geforce 3 was 4x Speeds @ 2x performance.


Geforce 4 was free 2x and free quincunx with 4x Quality.


And Geforce FX was free 4x

 

And all those claims were untrue

Yup, either Derek Perez or Brian Burke was going around before the NV30 launch saying that NV30 was "built for full speed execution of 4x FSAA" implying that it was free. It sure made the fanboys happy. Until reviews came out that showed the R300 annhilating NV30 with 4x FSAA and 8x AF.

 

Yeah, that's pretty bad. But I changed your emphasis a little bit (italics mine). It's not an outright lie, but it is close.

From the legal information:

They really should take the statement you quoted down, even if it is technically not lying to have it up (it is still misleading).

 

Ya.. currently I think the closest we are coming to free AA is 2x/quincunx for Nvidia and 2x for ATI.. But its still not free, ;\