GlobalFoundries outlines roadmap

[Idontcare]

Originally posted by: Viditor

Originally posted by: ilkhan

So figure early 2011 for AMD 32nm? That sounds about right from what I remember.

And because its relevant, isn't 22nm a "minor" shrink for intel, whereas 32nm is a "major" shrink for AMD? (forgive the terms, new equipment or something).

I believe that AMD is planning on shipping 32nm in mid 2010 still...

You figure that will be for server/opteron line in 2010 followed up by consumer (desktop/mobile) in Q1 2011 as they did with 45nm in 2008/2009?

Or you thinking clean-sweep 32nm launch (purchasable and shipping from newegg) in mid 2010?

Just based on history, I'm with ilkhan in terms of my expectations for 32nm desktop products from AMD.

Will be happy to see them in stock at newegg come Q1 2011, not disappointed if it takes till Q2 (they are allowed a little slippage given the non-zero productivity loss attributable to GF spinoff that no doubt impacted managers and worker bees alike) but it would be quite fantastic if they pulled off a 2010 availability of their 32nm consumer IC's.

 

[ilkhan]

Cool. Good info IDC.

Is bulldozer supposed to be first product on 32nm or is it later? BD is supposed to compete with sandy, which is supposed to be end of 2010. So that sounds right too.

 

[Viditor]

Originally posted by: Idontcare

Originally posted by: Viditor

Originally posted by: ilkhan

So figure early 2011 for AMD 32nm? That sounds about right from what I remember.

And because its relevant, isn't 22nm a "minor" shrink for intel, whereas 32nm is a "major" shrink for AMD? (forgive the terms, new equipment or something).

I believe that AMD is planning on shipping 32nm in mid 2010 still...

You figure that will be for server/opteron line in 2010 followed up by consumer (desktop/mobile) in Q1 2011 as they did with 45nm in 2008/2009?

Or you thinking clean-sweep 32nm launch (purchasable and shipping from newegg) in mid 2010?

Just based on history, I'm with ilkhan in terms of my expectations for 32nm desktop products from AMD.

Will be happy to see them in stock at newegg come Q1 2011, not disappointed if it takes till Q2 (they are allowed a little slippage given the non-zero productivity loss attributable to GF spinoff that no doubt impacted managers and worker bees alike) but it would be quite fantastic if they pulled off a 2010 availability of their 32nm consumer IC's.

I think your timings sound reasonable...

Is bulldozer supposed to be first product on 32nm or is it later?

Later, but not much...as AMD can now tell you from experience, releasing a new node with a new design is a recipe for disaster!

 

[ilkhan]

But do they really have the resources to design a 45nm version before shrinking it for 32nm? Or do you think they'll do a 32nm PIII before BD?

 

[drizek]

I think 32m phenom is a reasonable bet. They made 65nm K8s before killing off the arch. A die shrink of phenom will make it attractive for budget customers and low wattage PCs, much like the 65nm K8s. You take an architecture that is mature, compatible with current(ie. cheap) motherboards and ram, and is "good enough" for most people, increase the clocks, cut down the power consumption and slash the manufacturing costs and you breath new life into it. Use the small die size of phenom to iron out hte kinks in the manufacturing process and pave the way for bulldozer(joke: shouldn't the bulldozer be doing the paving?)

Edit: now that I think about it though, they might choose to shrink Athlon X4 instead. These CPUs aren't going to be going to demanding customers, so there is less need for the L3 cache. Now you have even smaller die sizes, better yields and even lower power consumption.

 

[VirtualLarry]

Originally posted by: KingstonU

Originally posted by: Fox5
They could probably do an atom competitor on that line. (scale the x2's down)

AMD is not going to try and compete in the Atom market, because there is no profit there. Intel has publicly stated that they regret making the Atom because of the very lower profit margins.

They regret supplying computer chips to the 3rd world?

 

[ilkhan]

drizek: oh I certainly agree that it'd be a good idea to make an earlier 32nm chip. My question is if they have the engineering resources to design such a chip for 32nm, WHILE finishing design work on 32nm BD chips.

 

[drizek]

I don't think they will have much choice. Bulldozer is going to be expensive on release, and 45nm phenom will be extremely obsolete. If they don't do 32nm K10 then they will basically surrender the budget market to intel. Thats never a wise thing to do, but especially so when you aren't completely sure of the competetiveness of your enthusiast parts.

 

[Idontcare]

Originally posted by: ilkhan
But do they really have the resources to design a 45nm version before shrinking it for 32nm? Or do you think they'll do a 32nm PIII before BD?

Yeah they wouldn't do a 45nm version of BD before shrinking it to 32nm. I fully expect a third generation Stars core on 32nm to be the leading product before BD debuts if for no other reason than for serving the mobile (laptop) space.

 

[Soleron]

Their server roadmap (which, unlike the client, is highly detailed through 2011) says 32nm for Bulldozer-based cores in 2011. Their 2010 plans are only the 45nm 6- and 12-core K10-based Opterons on Socket C32/G34.

So, unless client is ahead of server, it's still 2011.