Fermi delayed

[toyota]

I'm wating for the ATI 6 series, due out in November and 28 nm will be way better then Fermi, and they'll launch around the same time!

I highly doubt 28nm will be ready enough to have cards out the door in Nov. arent rumors suggesting 32nm for an ATI refresh?

 

[thilanliyan]

Why does nVidia rename quarters too? Q2 2011 (for nV) is NOT in reality May-July 2010!! 😛

 

[Genx87]

Why does nVidia rename quarters too? Q2 2011 is NOT in reality May-August 2010!! 😛

Financial calendar year. The United States federal fiscal calendar year starts in Oct. We were on the 2010 budget in Oct 2009.

 

[yh125d]

I highly doubt 28nm will be ready enough to have cards out the door in Nov. arent rumors suggesting 32nm for an ATI refresh?

I've heard both. If its nov/dec, 32nm is more likely. If feb or later, 28 probably


What would be really neat is if they did a 32nm 5890 this fall!

 

[cbn]

I highly doubt 28nm will be ready enough to have cards out the door in Nov. arent rumors suggesting 32nm for an ATI refresh?

I thought the refresh will be 40nm.

The N. Islands ATI GPUs (aka HD6xxxx) will be 32m.

 

[Schmide]

BSN says March 27th

 

[Keysplayr]

Anyone that doesn't "question" whether or not an unreleased product is superior to what's currently on the market is either a fan boy or delusional. You have an informed opinion based on past evidence that may end up being correct, but stating "there is no question" borders on blind faith. In a technology company? Really? Couple such devotion with the fact that they give you free stuff as part of what's basically a sponsored product review and your credibility is gone in my view.

You can spin however you want, the fact remains the same. Launch of Fermi is going to be 6 months later than the competition, at least, with mass availability between 8 and 10 months late, by NV's own admission. Whether the performance of the card can excuse or make up for the tardiness of the product certainly is in question, and remains to be seen.

I'm now confused. Which is it? Informed opinion or blind faith?

 

[v8envy]

"Our current-generation GPUs are fabulous and all the things that mainstream consumers would use their computer for." -- JHH

That pretty much says all you need to know about NV's upcoming lineup in the product segment most of us are interested in.

 

[thilanliyan]

Financial calendar year. The United States federal fiscal calendar year starts in Oct. We were on the 2010 budget in Oct 2009.

Hehe I know...it's still odd to me that it's done like that.

 

[mm2587]

I highly doubt 28nm will be ready enough to have cards out the door in Nov. arent rumors suggesting 32nm for an ATI refresh?

well considering there is no 32nm process to make gpu's on its going to be 28. (gloflo doesn't have one and TSCM canned their 32nm to work on fixing 40nm and readying 28nm. But yeah theres no way that's happening by November.

 

[evolucion8]

Wood screw one 🙁
I think they must have run out.

Probably they killed too much forests to create such beasts hehe. But is a pity that Fermy is delayed again. I hope AMD doesn't turn [green]evil[/green].

 

[Voo]

Anyone that doesn't "question" whether or not an unreleased product is superior to what's currently on the market is either a fan boy or delusional. You have an informed opinion based on past evidence that may end up being correct, but stating "there is no question" borders on blind faith. In a technology company? Really? Couple such devotion with the fact that they give you free stuff as part of what's basically a sponsored product review and your credibility is gone in my view.

No he's right that there's no doubt that fermi is a real innovation - out of order execution for GPUs is something completly new and really complicated to achieve. I wouldn't be too surprised if the added complexity by that alone could screw up their schedule quite a bit, after all they had really no experience with that stuff.

The problem with it is, if the advantages of OoO really are worth all the trouble for a gaming (!) GPU. No doubt it's useful for GPGPU stuff, but how many gamers are interested in fermis capabilities for GPGPU? A minority at best. I can't imagine that OoO is really helpful for gaming, but what do I know?


But from a solely technical standpoint Fermi is much more interesting than Cypress..