NVIDIA GF100 & Friends speculation

So the entire working set of chips from thousands of wafers on two or three spins over many months, some chips bigger than others, with hundreds of watts of board power variation, all in the hands of websites and journalists and thus none available to the public whatsoever until July. Or is that June? Personally, I'm going for B2 in August.

Hmm.. What? Was it Rys who typed that?
 
I think some people are trying to blurry things. I remenber reading this:

On the flip side of this, many hardware enthusiasts are asking, where is NVIDIA? The fact is that NVIDIA has no products to immediately answer AMD’s 5800 series and it is our understanding that it will be late February at the earliest before we actually see a next-gen GPU show up from NVIDIA in the retail channel.

It was in the 5850 review around september last year from Hardocp/kyle. if it was known already, it seems to me that most of the people saying Q4 09 or close were tools for nvidia PR :p
 
It was in the 5850 review around september last year from Hardocp/kyle. if it was known already, it seems to me that most of the people saying Q4 09 or close were tools for nvidia PR :p

I honestly believe the "best case" launch would be around the Win7 launch, judging by marketing documents from NV and the time period they covered and pointed at October 26th.

A post CeBit launch was communicated to just about everyone back in Early December, when A3 went into production.

I believe Kyle since I think he gets his information from the same source.
 
I'm embarrased to admit it but I read the sticker on the back of the card at 400 magnification. Picture ^7 as supplied by Carstens ...150 post ago.

The one that has "..100 375 A3 ^51D" printed on it.

What does 375 stand for?
 
Not to be unpolite, but what is finally the explanation as to why some posters here claimed the card was going to launch in 2009?
 
It was in the 5850 review around september last year from Hardocp/kyle. if it was known already, it seems to me that most of the people saying Q4 09 or close were tools for nvidia PR
I'm hoping someone like Fuad who just happily promised ever shifting release dates comes clean some day with how this went down.
 
It was in the 5850 review around september last year from Hardocp/kyle. if it was known already, it seems to me that most of the people saying Q4 09 or close were tools for nvidia PR :p

Nvidia kept promising launch/availability based on their idea that the next respin would be great, fixing yields, clockspeeds, temp, power, etc. Of course that schedule kept getting put back as the next spin turned out to be not as wildly successful as Nvidia hoped. Nvidia kept making empty promises while they ran to fix their problems, unable to tell the truth of the situation for fear of damaging their share price and reputation, and fear of ceding a year's market to AMD. Nvidia kept promising release dates without knowing if they had a viable product.

Pure hubris, lies or incompetence? Probably a bit of all three. I expect there to be very few Fermi cards around, even fewer at the very top of the range, while Nvidia scrabbles around at another respin. Fermi availability before August/September will make AMD look like they've been raining 5xxx cards on everyone for the last six months by comparison.


Not to be unpolite, but what is finally the explanation as to why some posters here claimed the card was going to launch in 2009?

Because that's what Nvidia was insisting on, and some people believed them at face value. Either Nvidia was using/lying to them, or Nvidia just kept promising things they had no clue if they could deliver. Better that than admit they were going to be several quarters behind, with no idea if they can build a workable product or one that could reach yields, heat, power or noise targets. You would think people would have wised up after the first few trumpeted dates were missed.
 
lol..Sarcasm?

No, it's just that 9-11 (no pun intended) month delay is beyond my wildest dreams.

But then, truth is stranger than fiction. :???:

[All of the following assumes that Bx spin will be needed, which will materialize in June/July time frame. This assumption may or may not turn out to be true.]

That is probably enough time for AMD to start pulling their next gen stuff out. In all probability, not all of it. I am pretty sure it will have to go up right against whatever AMD has in store for the back-to-school season. As for midrange, my guess is that they'll receive a post launch silicon-respin as well.

If GT200/GF100 are any indication, building giant sized GPUs has hit not just power and heat wall, but also an engineering wall. These things need too much time to just get out the door, not to mention their kids. I shudder to imagine what will happen if GTX580 suffers from similar delays.
 

More pics here:

http://ic.tweakimg.net/ext/i/imagelarge/1267547764.jpeg

http://tweakers.net/nieuws/65933/cebit-nvidia-gtx-480-op-de-gevoelige-plaat-gevangen.html

Quite a "beast".... who the heck buys this if the focus on games shift to console games.

I now buy games for my PS3 these days. Got sick and tired of the endless BSODs and driver issues when a game launches.
 
No, it's just that 9-11 (no pun intended) month delay is beyond my wildest dreams.

But then, truth is stranger than fiction. :???:

[All of the following assumes that Bx spin will be needed, which will materialize in June/July time frame. This assumption may or may not turn out to be true.]

That is probably enough time for AMD to start pulling their next gen stuff out. In all probability, not all of it. I am pretty sure it will have to go up right against whatever AMD has in store for the back-to-school season. As for midrange, my guess is that they'll receive a post launch silicon-respin as well.

If GT200/GF100 are any indication, building giant sized GPUs has hit not just power and heat wall, but also an engineering wall. These things need too much time to just get out the door, not to mention their kids. I shudder to imagine what will happen if GTX580 suffers from similar delays.

If real avaiblity isn't here till june/july , ati will most likely have hecto whatever chip out by sept. So even if fermi is faster than cypress they may not be able to sell enough of it anyway.


What i'm really wondering is how far along GF's 28nm is. Ati can bring out Hecto and shrink cypress and use cypress to replace the 57x0 chips and the 57x0 series shrunk should be able to replace chips as far down as the 54x0 chips.

I'm not sure where this would leave nvidia. They might have the workstation market but if ati chooses to had back in with firegl and a 28nm chip vs the 40nm fermi that might take a chunk out of nvidia's sales there also.


But ati really needs to get the ball rolling with software and advertising. I know to many people who don't want me installing ati cards in thier machines because of all the twimtp and physx advertising nvidia throws around.
 
If real avaiblity isn't here till june/july , ati will most likely have hecto whatever chip out by sept. So even if fermi is faster than cypress they may not be able to sell enough of it anyway.


What i'm really wondering is how far along GF's 28nm is. Ati can bring out Hecto and shrink cypress and use cypress to replace the 57x0 chips and the 57x0 series shrunk should be able to replace chips as far down as the 54x0 chips.

I'm not sure where this would leave nvidia. They might have the workstation market but if ati chooses to had back in with firegl and a 28nm chip vs the 40nm fermi that might take a chunk out of nvidia's sales there also.


But ati really needs to get the ball rolling with software and advertising. I know to many people who don't want me installing ati cards in thier machines because of all the twimtp and physx advertising nvidia throws around.

NI is @ TSMC.
NV will chase 28 nm just as badly.
AMD needs to make money first to before investing in sw/ads. They are living hand to mouth ATM.
 
Back
Top