GlobalFoundries Details Plans For 2011 And Beyond - HotHardware

GlobalFoundries Details Plans For 2011 And Beyond

0 thumbs up
Now that we've discussed GlobalFoundries' plans in general, let's take a gander at what we can expect from the company's 32nm SOI and 28nm bulk silicon processes. Both processes will use what's known as high-k metal gate technology. High-k can be built either through the use of a gate-first or a gate-last approach. Intel—the only company to date that's shipped high-performance / high volume products using high-k—favors the gate-last approach. GlobalFoundries, IBM, and Samsung are championing gate-first technology.



Here's how the nerdfight breaks down. Proponents of gate-first technology argue that it allows for less-restrictive designs, smaller dies, and is easier to build. It's also cheaper—GF expects to save ~$75 million over four years by opting for gate-first rather than gate-last. Gate-last proponents (Intel, TSMC) claim that gate-first technology isn't a long-term solution and can't deliver all the benefits of the gate-last approach. We won't actually know which approach makes more sense until we see shipping silicon.

What we do know is that the shift to high-k metal gate manufacturing has significant implications for Bulldozer and, by extension, Bobcat. If you remember Intel's Prescott, you should also remember that the CPU's high temperatures and significant power consumption made it a lose/lose proposition for enthusiasts almost across the board. One of the reasons for this was the amount of current that 'leaked' out of the CPU—at just 1.2nm thick on a 90nm process, the gate dielectric material was unable to prevent electron tunneling.  

Prior to 90nm, both Intel and AMD shrank gate thickness by ~0.7x per process, right up to the point where gate leakage, which was already present at 130nm, skyrocketed. Intel didn't change gate thickness at 65nm, but was able to do so again at 45nm because switching from Poly/SiON to high-k reduced the amount of gate leakage exponentially. AMD's processors up to and including the Phenom II, have yet to make this switch. Here's GlobalFoundries own prediction for gate scaling at 32nm and below.



AMD's curve looks much the same as Intel's, with the exception that AMD held gate lengths constant through 90nm, 65nm, and 45nm. At 32/28nm, however, power leakage will drop while gates again scale downwards. Over the past few weeks, we've wondered how Bulldozer's power consumption and performance will compare to Sandy Bridge. We're nowhere near filling in that particular puzzle, but we do know, at least, that Bulldozer will be fighting with an advantage previous generations of AMD chips haven't had.

When Will Bobcat Make The Leap?

AMD's first generation of Ontario processors (dual-core Bobcat + GPU) will be built by TSMC on that company's 40G (high performance) process. Left unanswered is the question of whether or not AMD will stick to TSMC at the 28nm node or move to GlobalFoundries. The peculiarities of 28nm technology make this a more important question than it might have been at previous nodes. TSMC currently plans to launch two 28nm processes. One will use high-k metal gate and a gate-last approach while the other is built around conventional Poly/SiON. There are questions regarding the suitability of Poly/SiON at 28nm and TSMC has, as we've noted, opted for the more difficult gate-last approach.

Even if we assume that TSMC executes both approaches perfectly, the inherent differences between gate-first and gate-last mean that there's no simple way to jump from one to the other. This difference will also apply to Radeon chips; switching from one foundry to the other at or below 28nm will necessitate significant and time consuming design/material changes.

AMD may still be keeping its options open and watching to see how the two approaches play out in the real world, but we'd be surprised if the company doesn't opt for one or the other before the end of the year.

Article Index:

0
+ -

Very good article 1st off. Global Foundries will be a nice addition to the chip manufacturing industry. Again the problems that TSMC had as far as delays on the 40nm ATI chips for the 4 and 5 series shows we need a new player in this arena.

 

0
+ -

lonewolf:
Very good article

I agree. Sounds like they've got it together at Global Foundries. They may make their mark in this industry yet, but don't expect Intel to just sit around waiting for it to happen either. They most likely have something up their sleeves, or somewhere hidden, just waiting to spring it on an unsuspecting world.

0
+ -

Yea you can bet Intel is not sitting on their hands, I am sure they are brewing up something in Santa Clara. 

Again though until they get a decent GPU company or Larabee really comes through they are losing ground.

 

0
+ -

Lonewolf:

Larrabee is dead. Dead dead. Intel has no plans to enter the graphics market at this time with a Larrabee product. In any event, Larrabee would never be a competitor with Bobcat or Bulldozer.

Intel already offers CULV products based on Arrandale (32nm, Core i3, dual-core). If Bobcat looks to be gaining ground on Atom, it's likely that Intel will respond in one or both of two ways: First, it could nudge Atom's speed upwards to 2-2.5GHz. That's 33-66 percent faster than current dual-core Atom's and it would certainly improve performance.

Second, Intel could revise its existing CULV line in order to hit price/performance targets that keep it competitive vs. Bobcat. We don't know which approach the company will choose, but one of these two make the most sense. They also don't require Intel to be sitting on an OMGSECRETWEAPON. Intel will absolutely fight back if Bobcat gains traction, but given AMD's current tiny share of the mobile market there's almost nowhere for the company to go but up.

0
+ -

@ Joel H that is precisely my point Intel IMO is in trouble. I have been saying they "must" buy out Nvidia.

With the merger of CPU/GPU, AMD is in the drivers seat going forward.

As you correctly stated Larabee is dead and I agree so what options does Intel have to fight Bulldozer I don't know.

0
+ -

lonewolf:
I have been saying they "must" buy out Nvidia. With the merger of CPU/GPU, AMD is in the drivers seat going forward.

I disagree. The acquisition of ATI didn't provide any Silver Bullet for AMD to nail Intel with. Not being a part of the High-End GPU market doesn't mean that Intel's floundering.  Look at the low end segment for a moment,...how many Intel based on-board video graphics solutions are in use right now? Don't they OWN that segment?

Don't they have a crapload of money laying around just waiting to snap up new technology companies?

Intel has better/smaller 32nm fab processes already in place and producing product as we speak. AMD doesn't yet,......

Intel says that they're on track to further shrink their process again in very little time. I believe them. It will be significant too.

Don't get me wrong,...I want to see AMD succeed. I like to see more than one CPU manufacturer. AMD being here is (what I think) keeping prices out of the Stratosphere. Their CPU's may not be the fastest, but they are usually the best bang for the buck for us low budget gamers and they work well too.

Now,....I'm not counting on winning this contest and getting the Mainboard and Video cards, but if 'it happens', (yeah right!) then I'll build an Intel box right after Christmas. But I expect that it's gonna be a Phenom Hex-Core Black with the same video cards in it because that's what I can afford to do.


0
+ -

Lonewolf,

I agree with RealNeil. Frankly, the latest GPU inside Arrandale provides the hardware decode features that previously weren't included on most Intel G45 chipsets. True, if you're a gamer, Intel GPUs are still substandard, but for everyone else they'll suit--the days of a crawling Vista/Win7 GUI and no HD offload are gone for everything but Atom.

AMD continues to make noise about using the GPU for processing, but noise, thus far, is all it is. The company claims it'll eventually develop a more tightly-integrated version of the CPU/GPU than what we'll see with Llano/Bobcat, but thus far the only company that's heavily invested in GPU offload is the one company without a processor to marry it to. :P

Login or Register to Comment
Post a Comment
Username:   Password: