nVidia Conference Call

[Dave Baumann]

I wonder how they reconcile their statements on IBM with reports such as these:

http://www.digitimes.com/NewsShow/Article.asp?datePublish=2004/02/13&pages=A1&seq=1

 

[jb]

Doh, nice find Dave

 

[martrox]

I wonder how they reconcile their statements on IBM with reports such as these:

http://www.digitimes.com/NewsShow/Article.asp?datePublish=2004/02/13&pages=A1&seq=1

And, in the year or so, can you honestly say you take anything nVidia says at face value?

It brings up something my father once told me:

Fool me once, shame on you; Fool me twice, shame on me!

 

[Sxotty]

I actually am curious though about what dave was saying... if things are really so bad at IBM why would they use them when TSMC is doing a good enough job?


I think something is wrong with both of those statements to be honest...

 

[Richthofen]

I wonder how they reconcile their statements on IBM with reports such as these:

http://www.digitimes.com/NewsShow/Article.asp?datePublish=2004/02/13&pages=A1&seq=1

dunno how this has any effect in the near term. I thought NV40 will be a 0.13 micron product.
Digitimes is talking about 0.11 here.

Nvidias margins in the 4th quater improved since NV36 launch (fabbed @IBM).
So i can understand that they are quite satisfied with IBM at the moment.

 

[digitalwanderer]

I was under the impression that they were going to introduce the NV40 at .13 to and migrate to .11 on a refresh when they made the move to PCI.

(I don't know exactly WHY I think that way, probably from reading threads here too much. )

 

[nelg]

I wonder how they reconcile their statements on IBM with reports such as these:

http://www.digitimes.com/NewsShow/Article.asp?datePublish=2004/02/13&pages=A1&seq=1

I thought vNividia had a deal with IBM that only pays per chip not per wafer.

 

[lost]

I wonder how they reconcile their statements on IBM with reports such as these:

http://www.digitimes.com/NewsShow/Article.asp?datePublish=2004/02/13&pages=A1&seq=1

Kind of odd that the .09 G5s are being put out of IBMs fabs supposedly with no problems and relatively low power consumption numbers, yet .13 and .11 micron products are somehow suffering.

 

[John Reynolds]

Kind of odd that the .09 G5s are being put out of IBMs fabs supposedly with no problems and relatively low power consumption numbers, yet .13 and .11 micron products are somehow suffering.

Look at the transistor count differences for the parts being discussed.

 

[Brandon]

I wouldn't be so quick to jump to this conclusion... PCI Express is certainly more than just "AGP 16x". The fast GPU->CPU transfers could open up some interesting new possibilities for game developers, like offloading collision detection, physics calculations, etc. to the GPU to do load balancing. It probably will have minimal impact on existing games, but this could present more performance optimization opportunities for future games regardless of how much memory is on the graphics card. The more programmable and powerful GPUs become, the more beneficial this could become.

Point taken. I'd be curious to know when we'll see these games. 2005?

 

[GraphixViolence]

That seems like a reasonable guess. It will probably be next year before there is a large enough installed base of PCI Express systems to make these kinds of optimizations worthwhile to implement.

 

[thegrommit]

So the GPU's are not limited to AGP 8X and they will recive a performance gain when a PCI-X bride is slapped on. At least the hi-end boards.

Just a nit-pick, but PCI-Express is not the same as PCI-X.

 

[rendezvous]

So the GPU's are not limited to AGP 8X and they will recive a performance gain when a PCI-X bride is slapped on. At least the hi-end boards.

Just a nit-pick, but PCI-Express is not the same as PCI-X.

Guess if i feel stupid now? What was i thinking when writing that post? Probably nothing all all.
Well, keep on with the nit-picking so that we could put an end to all this PCI-X / PCI-Excpress confusion.

 

[Mark]

I've seen it refered to as PCIe or, even better, PEG (which I assume stands for PCI-Express for Graphics). Far less confusion.