Another Anand tease...

[McElvis]

Just spotted this news post on AnandTech about UT2K3 CPU scaling article. Here's a quote at the end -

In fact, one of the companies in today's article is stopping by our labs to show off something special later today. If all goes well we should also have a few guys from Epic over here to benchmark it as well, too bad you won't see the results of that for a little while

What do you think he is talking about... R300/NV30/P10??

http://www.anandtech.com/news/shownews.html?i=3422&t=sn

 

[RussSchultz]

P10 isn't mentioned in the article.

 

[McElvis]

Yeh, your right sorry

 

[nAo]

NV30?
I believe Epic guys already have a R300, so that would be their occasion to test their tech on the new part from nvidia.

ciao,
Marco

 

[mboeller]

NV30?
I believe Epic guys already have a R300, so that would be their occasion to test their tech on the new part from nvidia.

ciao,
Marco

If true then we would see an "paper" launch of the NV30 when the R300 comes out. WOW two cards at the same time. Will be nice to see Nvidiots and ATi-fans (how are they called?) to fight.

 

[phynicle]

i can't take this....r300 is on the horizon and so much secrecy...i belive in open information....one person in this whole forum should have the indsider scoop...whu eva u are spill da beans already

 

[jb]

FanATIc works

 

[Doomtrooper]

I wouldn't assume a Nv30, if so it must be a .15 micron version...I can almost guarantee it is 3Dlabs with the P10. The cards are far more along in development and make much more sense.

 

[Randell]

NV28?

 

[RussSchultz]

I wouldn't assume a Nv30, if so it must be a .15 micron version...I can almost guarantee it is 3Dlabs with the P10. The cards are far more along in development and make much more sense.

Why MUST it be a .15 version?

 

[Joe DeFuria]

Obviousloy, it's the Kyro III...or then again, maybe the R-300. Or it could be the NV-30, or maybe new drivers for the Parhelia. On the other hand, the P10 sounds like it would be the one. Although there is the possibility that it's the BitBoys prototype.

Now that I think about it, it's probably the S3's Columbia, or a prototype of the new SiS chip...

Erhm...did I miss anyone?

 

[Doomtrooper]

Mass production .13 Micron as proven by press releases from lots of companies is not viable at this time , in fact the only one that has made the transition well with their own fabs is Intel ($$$$)which has 7 fabs I believe. AMD struggled but is ramping up now... TSMC is already producing Via's C3 and Transmeta's Crusoe at 0.13 as well as DRAM, the line up is getting longer and TSMC has limited .13 micron resources...not to mention a little birdie told me NV30 came back @ .15

http://eedesign.com/news/OEG20020614S0123

I do find it strange that everyone assumes Nv30 when 3Dlabs has working cards now with proven drivers (Carmack stated drivers were very good). In reality people what card is more than likley ready, a GPU that just got taped out and came back @ .15 or a card that is already been reviewed working

 

[Evildeus]

Doomtrooper,

What's the point talking of mass production and the relatives problems, when we are considering sample alpha cards? It's not because now there're problems with the 0.13 process that in 4-5 month the issues will stay. And it's not because right now you can't produce in mass, that you can't provide some chips.

I don't know if it's R300, NV30 or P10, but i would say that it could be either chip.

 

[Joe DeFuria]

Just because mass production 0.13 may not be ready, that doesn't mean nvidia can't have samples of parts at 0.13.

I think the main reason why people are not assuming P10, is because Anand said it was "one of the companies" in the article. 3D Labs is not in that article...only ATI, nVidia, Matrox, and STMicro.

 

[Doomtrooper]

Just because mass production 0.13 may not be ready, that doesn't mean nvidia can't have samples of parts at 0.13.

I think the main reason why people are not assuming P10, is because Anand said it was "one of the companies" in the article. 3D Labs is not in that article...only ATI, nVidia, Matrox, and STMicro.

EDIT:

Joe ok fine I read the quote again,strange though for no showing yet for the P10 , so I assume it is a R300 since were are a month away from paper launch ....I trust my birdie

 

[Typedef Enum]

n fact, one of the companies in today's article is stopping by our labs to show off something special later today.

This statement alone tells you who the candidates are...

1. ATI
2. nVidia
3. ST Micro
4. Matrox

It's safe to scratch Matrox off the list. And while you're at it, eliminate ST Micro as well. That leaves 2.

The fact that somebody would travel all the way to Anand's location, just to see how Unreal Tournament performs?...I'm going to say it was ATI.

Sounds uncharacteristic of nVidia...I'm sticking with R300.

 

[Doomtrooper]

Always seems to come to down to these two doesn't it, well if its a NV30 I will eat my work boot.

 

[Joe DeFuria]

I agree. Probably R-300 or NV-30. It may be the first "non-alpha" R-300 hardware / drivers, or the first Prototype R-300. (EDIT...OOPS, I meant first prototype NV-30)

Edit: Oh, and I don't it's really strange at all that there's no P10 showing yet, because we aren't expecting any consumer based P10 - boards until late this year at the earliest. And while Carmack has commended the P10's OpenGL drivers, (which should actually be expected considering the intial professional 3D market for P10) we don't know about the Direct3D driver situation. So we have no idea how P10 would handle UT-2003.

 

[Dave Baumann]

Given Anand's previous statements relating to ATi my assumption would be that he's already seen the performance of R300 under these tests so I don't think it would be that much of a surprise to him.

A Kyro based product isn't entirely out of the window as Hercules are soon to be producing the SE's - however interest in this may not be that great (although a comment like Anands could just be to generate some hits on a relatively uninteresting product through speculation).

However, my guess would be an NV based item (may not even necessarily be specifically a GFX chip).

 

[Dave Baumann]

And while Carmack has commended the P10's OpenGL drivers, (which should actually be expected considering the intial professional 3D market for P10) we don't know about the Direct3D driver situation. So we have no idea how P10 would handle UT-2003.

DX8 drivers are fully operational and running DX games fine. I don't know if they have taken out any serious compatibility testing yet (I do know that many of them were testing games on P10's when I accidentally saw the Alpha P10's in their ofice) and I don't know if they are doing any pre-release testing either.