NVIDIA shows signs ... [2008 - 2017]

Status
Not open for further replies.
It makes me wonder how far in advance the fab orders are put in. Because any company that thought last fall's economic downturn wasn't going to hit them were being delusional (not pointing that finger at NV in particular or anything like that). Intel delaying Lynnfield (mainstream Core i7) is probably the same problem: excess inventory of existing products they need to move through the retail channels first.
 
In the current environment, there's no way they'll get that. So yeah, looks like jobs are on the chopping block big-time; can't help but wonder what they'll be able to cut. They can't compromise on GT21x in the very short-term and they certainly can't do anything that might cause a GT300 delay. And they already cut plenty of MCP jobs in the last layoffs.
Don't think they will touch MCP at the moment either, is too needed for ion and the apple deals. I just read a Dean Takahashi interview with Jen-Hsun here:

Q: You said the industry would look a lot different after the recession. Who will be gone?

A: That’s harder to predict. I’m going to invest a lot more in R&D this year than I did last year — by a lot. I will be unique that way. I believe in the work we are doing with GPU computing, digital computing and Ion and mobile devices. The reason to have cash in the bank is to weather storms. This is an extraordinary storm. But we will invest in those things. For a lot of companies, the concern is that maybe all anybody wants after the recession is something like Ion. I am prepared for that outcome. Not every company will be happy about that. The CPU sells for less than $20 in this platform. But it’s a realistic world. If you know a $99 DVD player works fine, you will not buy a $499 DVD player.

Am not so sure GT21X or GT300 and similar products are safe. Except for the gpu computing aspect does not appear to be much of a place for such large powerful chips in the miasma the ceo has outlined. Lucky the money for them has likely already been commited, so there is still a chance they are going ahead, damn the torpedoes ;)

I suspect they'll try to cut mostly longer-term projects, including things like early work on 28nm, arch revisions to G300, DMI and/or GT3xx IGPs, a few derivatives, etc. - and if they ever at any point considered doing anything on TSMC's 32G process instead of going straight to 28HP, then obviously that's out of the window right now.
TSMC seems to be having ordeals of its own, its likely their 32G and 28HP will be pushed back by them anyhow. No sense nvidia spending money on future products that they cant actually physically make.

And Tegra/Tesla aren't going to escape from it this time around; in Tegra's case, I would suspect especially SW teams in India and elsewhere are most at risk rather than HW.
There's is apparently a tegra deal with HTC later in the year that's according to Doug Freedman is worth about $100m.

Here's hoping Charlie is wrong and GT216 isn't delayed; that chip is really exactly what they need right now, and it'd be truly catastrophic if it missed Back-to-School.

I heard that GT216 was pushed back too, but thought maybe they were confusing it one of the other GT21X chips, probably the GT212.

I'm willing to bet it has hit quite hard also; remember capital investments are going down the shitter worldwide, and car manufacturers were also one of their big customers there...

Saw when one of the quadro products was released earlier this year, at the commodity prices were peaking they used resource exploration as example to demonstrate the power of their products. Looking at this IMF graph, don't think exploration companies will be able to afford tesla's anymore either. Anyone got anything of how hollywood and tv production is going...offhand would of thought the advertising would be one of the first things to be cut back.
 
Well, you definitely can't say Jen-Hsun isn't a gambling man...

I’m going to invest a lot more in R&D this year than I did last year — by a lot.

That's a rather large gamble that a huge R&D investment now will lead to great rewards if things turn around in time.

On the plus side, it "should" guarantee them the technology advantage and if things fall into place and the global recession ends relatively soon (6-12 months) it could pay huge dividends as R&D tends to build upon itself.

On the minus side, that's a huge drain on your cash position if things don't turn around soon. If sales continue to decline you'll have a large cash drain on top of a large cash drain. On top of that if sales decline to the point where you can't recoup your R&D for those projects... Things will really start to hurt.

Has Nvidia released their 4th quarter numbers yet?

I think most companies right now are just buttoning down the ship, reducing cash drains as much as feasable and weathering the storm. At least those with the cash reserves to do so. Everyone else is just wondering when their ship is going to sink.

But one thing Jen-Hsun sure doesn't lack is a big pair of brass balls. :)

Regards,
SB
 
It makes me wonder how far in advance the fab orders are put in. Because any company that thought last fall's economic downturn wasn't going to hit them were being delusional (not pointing that finger at NV in particular or anything like that). Intel delaying Lynnfield (mainstream Core i7) is probably the same problem: excess inventory of existing products they need to move through the retail channels first.

Which says quite a bit. As idle fabs are a constant drain of cash. So it's saying quite a bit that Intel would rather take the compounding losses from idling fabs than writing off existing stock.

In other words, there must be a relatively huge pile of existing stock they have to move in order to make idling a fab (or 2 or 3 or...) the lesser of two evils.

I wonder if Intel will be taking in outside orders to keep their fabs running?

Regards
 
That's a rather large gamble that a huge R&D investment now will lead to great rewards if things turn around in time.

On the plus side, it "should" guarantee them the technology advantage and if things fall into place and the global recession ends relatively soon (6-12 months) it could pay huge dividends as R&D tends to build upon itself.

On the minus side, that's a huge drain on your cash position if things don't turn around soon. If sales continue to decline you'll have a large cash drain on top of a large cash drain. On top of that if sales decline to the point where you can't recoup your R&D for those projects... Things will really start to hurt.

I wouldn't read too much into "by a lot," but as to the R&D increase itself, well... it seems with industry trends that the graphics picture of 2011 may be significantly shifted from that of 2008, so for NV I think it's less a matter of holding on to cash vs ensuring they are positioned to retain their 'preeminence' come the/a recovery.
 
I wouldn't read too much into "by a lot," but as to the R&D increase itself, well... it seems with industry trends that the graphics picture of 2011 may be significantly shifted from that of 2008, so for NV I think it's less a matter of holding on to cash vs ensuring they are positioned to retain their 'preeminence' come the/a recovery.

Aye, which is still a gamble, as there are no solid indicators of when the recession will end. And no indicators as to whether it will get worse before it gets better.

At least one good sign is that the new president has postponed implementing new taxes that would increase the burden on already struggling businesses.

Here's to hoping he does the right thing to get make it easier for the economy to recover.

Regards,
SB
 
That's a rather large gamble that a huge R&D investment now will lead to great rewards if things turn around in time.

It's a gamble but it's not that hard of a call to make. Remember they're competing against Intel as well and Intel is pushing up into their space. So sitting back now is just as or even more risky than forging ahead. A strong cash position during a downturn is a great asset since you can continue to invest while the competition has to focus completely on keeping the shop open. But if you just sit on it out of fear then it does you no good. It will be interesting to see what Nvidia's Q4 cash flow looks like though.
 
Also note that he said he was ready for the outcome of people only wanting something like Ion. Putting R&D into low-cost solutions like that has a rather high chance of paying off. Especially if they can get a modular system somewhat along the lines of what ATI is currently offering.
 
Nvidia thought they were onto a good thing, putting both 9400M and 9600M GT into a MacBook Pro...

http://www.engadget.com/2009/03/05/nvidias-geforce-9600m-causing-issues-in-17-inch-macbook-pro/

Jawed

If I were the superstitious sort, I would attribute this as karma for the recent underhanded bilking of otherwise unknowing consumers by calling G92b anything from GTX 250 to Mobility GTX 260 or Mobility GTX 280.

But I'm not, so I wonder if this is yet another hardware issue or whether it's something that can be resolved by firmware? And whether Apple or Nvidia will take responsibility?

Regards,
SB
 
Bad karma for taking an existing card (ie. 9800 GTX+), reducing price by 25%, reducing power consumption, reducing physical size, and offering more memory options? Bad karma should be reserved for a company that is screwing over it's customers. Last time I checked, substantially lower prices, lower power consumption, smaller physical size, and more memory options is actually a good thing for customers. In fact, even the new name is more appropriate than the old one, so that the card is clearly positioned below the 260 (in the sense that it offers a bit lower performance while still having a very similar feature set with PhysX/CUDA/SLI/etc).

Above I am speaking about GTS 250 of course. I do agree that the naming scheme for the notebook GeForce GTX 260 and 280 is a bit wacky. But that is ultimately just semantics, because it is not important how one desktop GPU compares to one notebook GPU, but rather how one notebook GPU compares to another notebook GPU. Anyway, if a company was selling cards that were way overpriced and underperforming vs the competition, then they are the ones who deserve bad karma, but that is not really the case here. At the end of the day, the most important thing is good vs bad engineering decisions, relative to that of the competition ;)
 
Last edited by a moderator:
Bad karma for taking an existing card (ie. 9800 GTX+), reducing price by 25%, reducing power consumption, reducing physical size, and offering more memory options? Bad karma should be reserved for a company that is screwing over it's customers. Last time I checked, substantially lower prices, lower power consumption, smaller physical size, and more memory options is actually a good thing for customers. In fact, even the new name is more appropriate than the old one, so that the card is clearly positioned below the 260 (in the sense that it offers a bit lower performance while still having a very similar feature set with PhysX/CUDA/SLI/etc).

Hum really?
8800GTS.........G92 core, 128sp, 256bit
9800GTX.........G92 core, 128sp, 256bit
9800GTX+.......G92 core, 128sp, 256bit
GTS250..........G92 core, 128sp, 256bit

Wow I didn´t know that GTS250 or 9800GTX+ or 9800GTX or 8800GTS where so revolutionary between them that this renaming madness was awesome.


jimmyjames123 said:
Above I am speaking about GTS 250 of course. I do agree that the naming scheme for the mobile 260 and 280 is a bit wacky. But that is ultimately just semantics. If a company was selling cards that were way overpriced and underperforming vs the competition, then they are the ones who deserve bad karma, but that is not really the case here. At the end of the day, the most important thing is good vs bad engineering decisions, relative to that of the competition ;)

And what is the excuse for this?:
CeBIT: ASUS shows nForce 980a board M4N82

Just a rebranded nForce 780a.

Hardware-Aktuell - News - CeBIT: ASUS mit nForce 980a-Board




And that´s not even AM3 compatible =/
 
Hum really?

What don't you understand here? The reason to "rebrand" from 9800 GTX+ to GTS 250 is simple: higher performance at a lower price, without requiring additional R&D cost and additional time to market in order to bring a GT200-based card to the market at that price point. What's the point when NVIDIA can focus those additional resources on GT300? The consumer is hardly getting screwed by having 9800 GTX+ performance, albeit with smaller physical size and lower power consumption, at the lower price of the GTS 250. In fact, the naming progression of GTS 250 --> GTX 260 --> GTX 280 makes a lot more sense than 9800 GTX+ --> GTX 260 --> GTX 280, considering that the GTS 250 has largely the same feature set as the 260 and 280.

8800GTS.........G92 core, 128sp, 256bit
9800GTX.........G92 core, 128sp, 256bit
9800GTX+.......G92 core, 128sp, 256bit
GTS250..........G92 core, 128sp, 256bit

Looks like you meant 8800 GTS 512 MB. Were there really no differences between 8800 GTS 512MB and 9800 GTX (such as physical size, power consumption, operating frequencies, etc)? Even so, that type of rename was a poor choice, because it implies that the 9800 GTX is a superior performer to 8800 GTS 512MB. On the other hand, renaming from 9800 GTX+ to mid-level GTS 250 name makes much more sense. One would have to be a real bonehead to think that the GTS 250 has higher performance than 9800 GTX+ based on the mid-level name and the lower price of the 250!
 
Last edited by a moderator:
What don't you understand here? The reason to "rebrand" from 9800 GTX+ to GTS 250 is simple: higher performance at a lower price,
You missed something here. Same performance.
And yet we are talking about 4 renamings for the same thing.

Also do you want to talk about motherboard renamings, mobile GPU renamings and all desktop lineup multiple renamings? Hope not....

Nvidia is damaging their reputation. I had never saw so much bad publicity around the Web. Every forum, every review every site is hammering on Nvidia.
Join this with mobile GPU failure rate in Q3 2008 and you have one big explosing box.
I'm not even going to talk in the woooooopppp asss.

Time will tell but Nvidia is losing reputation every week.
On the other side ATI is kicking in 1 month RV740, RV790. 40nm to mobile and desktop and at the same time the wave of hd 4000 mobile into the notebooks. I smell hard times for Nvidia in the next 3/6 months.
 
What don't you understand here? The reason to "rebrand" from 9800 GTX+ to GTS 250 is simple: higher performance at a lower price, without requiring additional R&D cost and additional time to market

So you're saying ATI should have "rebranded" it's HD 4870 at least 4 times by now, as they can be now be bought for around $170? :rolleyes:
 
Actually the implication is either that...

1. Nvidia doesn't innovate in anything except in the enthusiast sector. Therefore it's best to just rebrand for mid/budget sectors.

or...

2. ATI are obviously idiots for spending money to R&D chips for the midrange/budget sectors when it's such a good idea to just rebrand the same chip for multiple generations and bilk the unsuspecting consumer for all they're worth.

Then again I might be biased. I hated this with a passion when ATI did it, and I hate it with a passion when Nvidia does it.

Regards,
SB
 
Status
Not open for further replies.
Back
Top