Profit of NV40 parts

[Sabastian]

Ok for the most part the analyst asked questions that were par for the course really. None were too hard hitting until the last fellow. He knows bloody well that NV management is basically marketing the PS3.0 feature to the hilt. Even they made admissions that there really won't be much in the way of effect in terms of visual disparities between PS2.0. The last guy hit the nail on the head so too speak in that NV is capitalizing on the PS3.0 moniker in that it is a marketing gimmick... at least for now. I thought it was kind of funny how the antagonist pushed the fact that while they had PS3.0 as a marketing feature that NV does not have PCIe to flaunt. Then he basically turned the feature argument around on management ..

But I can't be more surprised by the fact that they basically admitted that the NV4* won't be on shelves any time in the near future. They did not give an affirmative answer to the question as to if the GF6800 will be available within the next couple of months!! I can't imagine why they would not tell analyst and stock owners when they are going to have these new high end/high margin products available outside of the stark reality that it simply will not be available. I don't see these cards available in any serious numbers in the next three months possibly even longer..

EDIT: Dear god, I need to get my ears checked. I guess they were referring to "derivatives" of the NV40.. or something. Is the NV40 shipping now or what? When can we buy it?

 

[nelg]

Funny stuff...

6800U designed specifically for enthusiasts, has lots of "frequency headroom", and comes with an optional second power connector, required only for overclocking

I guess this confirms that the 6800U is actually overclocked then.

Hmmm, I am surprised that Nv is officially sanctioning overclocking. Comments like that may open up some legal issues.

 

[Bjorn]

Just as I'm listening:

- Margins to improve throughout the year, far beyond NV3x levels
- Profits to be flat to slightly up next quarter, with XBox finally rising after 2 declining quarters in a row
- NV4x biggest leap NV have ever taken
- Shader 3.0 "dramatically extends programmability"
- A dozen games and big engines are going to support 3.0
- 6800U designed specifically for enthusiasts, has lots of "frequency headroom", and comes with an optional second power connector, required only for overclocking
- The GPU is affected heavily by the compiler, which is shipping at 1.0 and is very early. They expect big performance boosts in later drivers
- Quadro FX4000 is showing 38x performance in a specific app (didn't pick up which one)
- 3% Market share gain came from mid to low end boards, as opposed to the enthusiast segment.
- Shader 3.0, Superscaler, and programmable video processor made up the transistor count over and above ATi.
- 40% more transistors, yet only 9% bigger die. Costs 10-15% lower, with better process capacity.
- Redundancy and re-configurability in the NV4x, and "immunity from defects", to achieve great yields moving forward.
- Multiple NV4x chips taped out, a couple of them are back from an unspecified fab, and by Q4 everything will be based on NV4x. They refused to be any more specific than "by the end of the year".
- Talked about R420 being effectively a 3 year old architecture, and customers will pick the NV4x over that based on features such as Shader 3.0, since theirs is the only true next gen GPU
- Game Developers "clamouring" for their Shader 3.0 and FP Filtering tech, building lots of boards specifically for devs. Easier to write, with better performance. Conditional branches are heavily touted.
- ATi have hand picked boards to be sent to reviewers, each at a different speed (?), potentially confusing consumers. NVIDIA don't do this, and their boards are much better overclockers that the XT (due to the hand picking mentioned previously, apparently).
- PS2.0 vs PS3.0 will be a "glaring" difference, but HSI vs. native PCI-E is identical (including bi-directional bandwidth). Competitors are doing much the same, so it's costing them money as well.
- Competitors are going to hurt badly with PCI-E, since they'll have to build 4 or 5 extra boards.
- NV4x can operate in 3 modes: AGP8x, PCI-E, or PCI-E with HSI.
- Started arguing with one caller, who pushed them about whether PS3.0 will actually make any difference NOW. They went back and forth several times, and it got reasonably heated and amusing. They said they'd call the guy back after the conference call

I'd say that Nvidia has a lot to live up now

Should be interesting to follow the availability of the NV4X. And when it's available, by which quantities and how long will it take from that point until it's available in mass quantities ?

And it'll be very interesting to see the tests with bi-directional bandwidth of the AGP/PCI-Express bridge chip f.e.

Other then that, some FUD, marketing tactics and some interesting information thrown in between.

 

[Evildeus]

The part comparing the Nv die to the R420 is interesting. Anyone can give some more insight on this?

 

[101]

My favorite part is during the morgan stanley guys question when Jen-Hsun says "defect prone, defect sensitive, err defect immune" three obviously synonymous terms

 

[AlphaWolf]

Other then that, some FUD, marketing tactics and some interesting information thrown in between.

Now to figure out which is which.

 

[Scarlet]

The part comparing the Nv die to the R420 is interesting. Anyone can give some more insight on this?

Sure. If you look at the various Intel events of the last several months, it's pretty clear ATI has a true native PCIE interface and nV does not.

I will say it out loud:
Jen-Hsun Huang is an honorless bald-faced liar who will state anything to get what he wants.

What he said about the ATI product is materially false and made in front of analysts, not an off-hand comment made in a materially non-relevant setting. If ATI had half the mind too, they could make a legal issue of it. I believe the legal term in comercial misrepresentation - http://www.cojk.com/resources/article-detail.php?id=5

He didn't get what he wanted with NV40, which was uncontested performance leadership, and now he is resorting to out-right fabrication.

Who is smoking hallucinogenic substances Mr. Huang?

 

[Evildeus]

Well, fisrt could you pleased answer my question rather than you blatant anti-Nv flame war?

Second, he can't say things like that without being true, because if it isn't, legal issues, as you said, could be much worse that little flame war like you seems to like.

So make your answer coherent please.

 

[Ylandro]

What NVidia really said:

- Margins to improve throughout the year, far beyond NV3x levels

we need to lower margins again to bead ATI... Hopefully not as badly as last year.

- Profits to be flat to slightly up next quarter, with XBox finally rising after 2 declining quarters in a row

Hopefully that fucking Xbox will finally produce some profits...

- NV4x biggest leap NV have ever taken

We've never been so desperate.

- Shader 3.0 "dramatically extends programmability"

Developers will have to program for yet another codepath...

- A dozen games and big engines are going to support 3.0

They don't yet... but we want them to!

- 6800U designed specifically for enthusiasts, has lots of "frequency headroom", and comes with an optional second power connector, required only for overclocking

Shit... we're running at way lower frequency, and require much more power.... How to sell that to the public?

- The GPU is affected heavily by the compiler, which is shipping at 1.0 and is very early. They expect big performance boosts in later drivers

We haven't had enough time yet to fully implement cheats for every benchmark game.

- Quadro FX4000 is showing 38x performance in a specific app (didn't pick up which one)

Notepad is running blindingly fast!

- Shader 3.0, Superscaler, and programmable video processor made up the transistor count over and above ATi.

Can't say ATI's design is more efficient...

- 40% more transistors, yet only 9% bigger die. Costs 10-15% lower, with better process capacity.

Nobody know cost and capacity, so we can please our shareholders here.

- Redundancy and re-configurability in the NV4x, and "immunity from defects", to achieve great yields moving forward.

We have big problems in using such a big die.

- Multiple NV4x chips taped out, a couple of them are back from an unspecified fab, and by Q4 everything will be based on NV4x. They refused to be any more specific than "by the end of the year".

Our yields suck. We need to find a new lab, but haven't yet. We'll call it unspecified for now. The NV3x stinks, X800 is out, so we desperately need to get the NV4x out. Hopefully we have a handfull at the end of the year.

- Talked about R420 being effectively a 3 year old architecture, and customers will pick the NV4x over that based on features such as Shader 3.0, since theirs is the only true next gen GPU

The R420 is way faster, so we desperately need to focus on our features, even though they are useless for the first three years.

- Game Developers "clamouring" for their Shader 3.0 and FP Filtering tech, building lots of boards specifically for devs. Easier to write, with better performance. Conditional branches are heavily touted.

That's what they should do.

- ATi have hand picked boards to be sent to reviewers, each at a different speed (?), potentially confusing consumers. NVIDIA don't do this, and their boards are much better overclockers that the XT (due to the hand picking mentioned previously, apparently).

Yeah, we know it's the other way around.

- PS2.0 vs PS3.0 will be a "glaring" difference, but HSI vs. native PCI-E is identical (including bi-directional bandwidth). Competitors are doing much the same, so it's costing them money as well.

Whoops... we missed a feature. Quick, downplay it's importance

- Competitors are going to hurt badly with PCI-E, since they'll have to build 4 or 5 extra boards.

They have better performance, hopefully it will be more costly for them.

- Started arguing with one caller, who pushed them about whether PS3.0 will actually make any difference NOW. They went back and forth several times, and it got reasonably heated and amusing. They said they'd call the guy back after the conference call

Shit... someone discouverd PS3.0 won't mean shit for the consumer...
PANIC!! Make him shut up... don't let it get out. Get his adress, and send a hitman after him!

 

[Scarlet]

Well, fisrt could you pleased answer my question rather than you blatant anti-Nv flame war?

Second, he can't say things like that without being true, because if it isn't, legal issues, as you said, could be much worse that little flame war like you seems to like.

So make your answer coherent please.

Oh yes he can and did.

Jen-Hsun claimed the ATI products were no better than NV40 on PCIE because the ATi products did not have a native bridge, that ATI would use an external bridge as does NV40.

At the Intel Developer's Forums it was quite clear - Intel noted the method of PCIE implementation of both products publicly. I don't have die photos of either product, but I rather expect Intel knows exactly what it is talking about. Jen-Hsun knows perfectly well the public position ATI has taken (native bridges only) and that Intel has endorsed. Furthermore, by making his statements publicly he is all but accusing ATI of misrepresenting what its products do.

So when he doesn't get what he wants (I seem to recall substantial boasting to analysts the last conference call how ATI wasn't going to lead this time around), what does he do?

As far as me making/starting/liking a flame war I could care less. If anyone feels the need to defend a CEO whose company has reflected the CEO's ethics the past two years, go right ahead.

 

[Dave Baumann]

Should be interesting to follow the availability of the NV4X. And when it's available, by which quantities and how long will it take from that point until it's available in mass quantities ?

It appears to have slipped by a quarter since the last CC and this. He was happy to talk about Q3 for the rest of the NV4x line, but now it appears to be year end for this one.

 

[Evildeus]

Scarlet,
We are not talking of the same things. Marketing words to say that NV4* won't suffer from the bridge in a developper conference is not the same things as saying that fact 1 and fact 2 on a cc with plenty of analyst and shareholders ready to go to court if anything is false. And then why are you talking of PCI-E? I don't think my primery question was about that, or i need to change my brain

 

[PatrickL]

- 40% more transistors, yet only 9% bigger die. Costs 10-15% lower, with better process capacity

It s nice trick here. He is talking about die and the following sentence he is talking about wafers.

i can trust each wafer cost 10 or 15 % more due to low-k process, but that tells nothing about the cost per die despite the obvious attempt of confusing things

 

[Evildeus]

- ATi have hand picked boards to be sent to reviewers, each at a different speed (?), potentially confusing consumers. NVIDIA don't do this, and their boards are much better overclockers that the XT (due to the hand picking mentioned previously, apparently).

Methinks he has ATI and nVidia completely confused.

I think he is talking about that:

- AnandTech : 500/550MHz 
- CHIP : 520/560 MHz 
- HardOCP : 520/560 MHz 
- Toms Hardware Guide : 525/575 MHz 
- Extreme Tech : 520/600 MHz 
- Beyond3D : 520/550 MHz

 

[Mariner]

Some interesting stuff in there but throwing FUD at your competitor's products isn't a good thing to do, IMO.

I find it quite amusing that he is criticising ATI for sending out cards with slightly differing specs when NV themselves sent out the as yet unannounced "GeForce 6850 Ultra" and "6800 GT" cards to compare to R420. In fact, to play Devil's advocate, how do we know that these cards will actually be released? Nothing on NV's web site indicates any such product exists - no press releases or anything like that.

At least ATI have a press release stating final clock speeds of the two announced cards.

 

[Bjorn]

Some interesting stuff in there but throwing FUD at your competitor's products isn't a good thing to do, IMO.

That's true. Ati isn't exactly innocent when it comes to stuff like that though.

I find it quite amusing that he is criticising ATI for sending out cards with slightly differing specs when NV themselves sent out the as yet unannounced "GeForce 6850 Ultra" and "6800 GT" cards to compare to R420. In fact, to play Devil's advocate, how do we know that these cards will actually be released? Nothing on NV's web site indicates any such product exists - no press releases or anything like that.

While the criticism isn't exacly coming from the right person, it's still valid criticism and it would be interesting to know why it happened.

 

[Dave Baumann]

One of the reviewers trashed a board when trying to flash to the retail BIOS.

 

[Evildeus]

One of the reviewers trashed a board when trying to flash to the retail BIOS.

Really?

 

[Dave Baumann]

- ATi have hand picked boards to be sent to reviewers, each at a different speed (?), potentially confusing consumers. NVIDIA don't do this, and their boards are much better overclockers that the XT (due to the hand picking mentioned previously, apparently).

I wonder if its just coincidence that ATI put out this release today.

On the die sizes - as has been noted several times already they are not necessarily counting transistors in the same way, the die size also has a surface area difference of 9%. I also have information (although not directly from ATI) that suggests they are actually using 12" fabs for R420.

They also quoted the "internal bridge" for "competitors PCI-Express", which has been debunked by ATI arleady. They also said they would be interested in hearing about ATI's solution for plugging native PCI-Express boards into AGP sockets, however ATI have already been on record saying they will use a bridge for that.

 

[Florin]

Funny stuff...

6800U designed specifically for enthusiasts, has lots of "frequency headroom", and comes with an optional second power connector, required only for overclocking

I guess this confirms that the 6800U is actually overclocked then.

Provided the transcription is correct, how so? Does the U require the second connector? Regardlessly, I'm sure you're aware that it is not overclocking as long as Nvidia guarantees the U core will run at the speed it does..