Southern Islands ... why?

[OCGuy]

Wonder if they'll make an SI to compete with the 5970. i'd get one then - atleast by the holidays if that's when they release it

I highly doubt a half-gen release is going to double performance. If it did, I would be more than impressed....

 

[faxon]

cause the northern islands are to cold that time of the year ^_^

 

[at80eighty]

I highly doubt a half-gen release is going to double performance. If it did, I would be more than impressed....

What I had in mind was hopefully some double GPU card that competed at par with the 5970 but has atleast more manageable physical dimensions and (hopefully) runs even a little bit cooler. maybe wishful thinking

 

[MarcVenice]

It's about the money Lebowski.

AMD currently has a 40nm 5xxx cash cow on their hands at all market segments/price points. Economically, why change that up anytime soon except to ...

a) move to a smaller, more cost effective node.
b) respond to competitive pressures from Nvidia.

If S.I. is on the 40nm node, that would nullify a).

That leaves what Nvidia will be bringing to the table ... after all, if AMD had reason to believe Nvidia wasn't going to bring much heat for the rest of 2010, they would have little reason to do much at all ... maybe do a lower cost 'hybrid' 5790 part to replace the 5830 (or shove it up to a higher price point) and a 5890~5950(5790x2?) to hit the $500 price point and take back the single gpu crown.

It would be very time/resource economical to tape out only one 'hybrid' chip if that is all that was really needed. It would also give them working knowledge and experience of most of N.I.'s architecture at 40nm, paving the road and freeing up resources to concentrate on implementing the full N.I. on GF's 28nm.

This would also keep product line confusion to a minimum ... bringing out hybrid S.I. cards that compete with existing 5xxx cards ... why? what would they be called? Better to wait and roll out an entire 6xxx product line over a few months like they did with the 5xxx line.

At best, Nvidia is unlikely to BETTER AMD's cost/performance in the low end and mainstream graphics segments, and more likely will be worse. Timeline wise all I've seen so far for Nvidia's GF104/108 is 'june at the earliest' ... meaning what? ... sometime in august or september?

Only reason I can see for AMD to do a complete (6xxx) S.I. refresh at 40nm is if the progress of GF's 28nm bulk process is such that N.I. will be delayed into 2H of 2011.

I think you're assuming a bit to much. Who sais the 5000-series is a cashcow? Sure, the 57x0 and lower parts are probably yielding nicely, but who knows how 58x0 yields are?

Also, TSMC can only produce a limited amount of 40nm wafers. You have to order wafers in advance. What if AMD isn't getting enough wafers? Then they can't really make as much as they wanted to, and they can't take back marketshare from Nvidia.

And as others pointed out, doing a hybrid is like taking a page out of Intels tick-tock book.

 

[MrK6]

I think it's about not letting your guard down. As others have mentioned, AMD is in a good position right now, they have a nice, full line up that is probably bringing in some good money. They should use that advantage to work on some R&D and future products. AMD knows not to try out new architectures on a brand new process - the 4770 gave them a heads up on 40nm for the 5xxx series. If they can introduce some of the new architectural elements of Northern Islands while still on a process they have experience with, I think it will give them a good heads up and benchmark on how to finalize their Northern Islands chip for 28nm. If the Fermi debacle has taught anyone anything, it's not to introduce a new architecture on a new process.

 

[Lonyo]

Well for good or bad I can't imagine them "leaping" without a die shrink. I wouldn't hold my breath for anything better than 20% mm for mm. Not to say they won't get more than that is they make something huge.. but I doubt they'd drop their strategy that quickly... Expect new features, better tessellation, and a touch more performance with NI giving us a whole new everything.

Does anyone know of historical examples of in between architectures like this?

http://www.anandtech.com/show/2376/7

On average, the Radeon HD 3870 gives us a 6 - 10% increase in performance

Might not be exactly the same (difference process, obviously), but they have managed to reduce transistor counts but increase performance in the past, so if they stuck with the same count or increased it slightly, they might be able to get more performance, or make up for shortfalls in feature performance (DX11 features), e.g. with a better tessellation unit.

 

[happy medium]

Southern Islands why?

Easy, because no one wants to be second best.
End of thread.

 

[extra]

This has nothing to do with Nvidia or trying to beat Nvidia's performance. ATI has their own internal thing they use. They aren't designing parts any more with the performance target of "it must be able to beat xxx Nvidia part". This is about ATI executing nice solid on time releases no matter what. In this case they probably had to switch to a "plan b" or "plan c" thing when the process they wanted to use got cancelled/pushed back, etc. I'm sure Nvidia has something like that going on as well, lot of smart people at both companies. Overall everyone is getting screwed by TSMC.

 

[badb0y]

Basically they have a schedule to follow regardless of what the competition is doing.