AMD 32 nm CPU Conquest to Begin in 2010

[btarunr]

After spinning off its manufacturing division to The Foundry Co., AMD is left with all the engineering resources it needs to make processors. Contradicting older roadmap slides by the company predicting it would start selling 32 nm processors in 2011, Dirk Meyer, CEO of AMD in an interview with Information Week said that the company is on-track to ship smaller, more powerful processors built on the 32 nm manufacturing process by 2010.

The new manufacturing process would allow the chip maker to step up transistor counts to add more features and computational power. Tomorrow, on March 2, the AMD is expected to close the deal with Advanced Technology Investment Company (ATIC) of Abu Dhabi to form The Foundry Company (TFC). This would render AMD fabless. The company will then, like other fabless market heavyweights such as NVIDIA and VIA, will focus on designing processors, while TFC, its largest foundry partner will manufacture the processors. Currently AMD is tied up with foundry companies such as TSMC and UMC for manufacturing products of its Graphics Products Group, products such as GPUs and chipset.

AMD will be an year behind larger market rival Intel in selling processors built on the 32 nm technology. Intel plans to roll out 32 nm processors by Q4 2009. AMD's designs will be ready by mid-2010, following which volume production of its 32 nm chips will commence by Q4 2010.

View at TechPowerUp Main Site

 

[Haytch]

Wow, could this mean AMD have caught up to Intel ? 6 month - 9 months is nothing compared to the lean Intel has maintained since the core2duo.

Focus seems to be a priority, which could mean a better performing and better quality end product. I hope so.

 

[AphexDreamer]

Noob question here but how does making the die size smaller allow for more stuff do be on it? In other words how does making the room smaller fit more stuff in it?

 

[ShadowFold]

Noob question here but how does making the die size smaller allow for more stuff do be on it? In other words how does making the room smaller fit more stuff in it?

I think it makes the transistors and stuff smaller. I'm not sure but that's what I've been believing for awhile now.

 

[btarunr]

Noob question here but how does making the die size smaller allow for more stuff do be on it? In other words how does making the room smaller fit more stuff in it?

Room gets smaller, and the things you put in it get smaller too.

Or

Room stays the same size, things get smaller, more number of things fit in the room.

Manufacturing technology has nothing to do with the die-area.

 

[oli_ramsay]

And it costs the same to make a wafer of 60nm dies or 32nm dies. Hence why technology gets faster and faster and cheaper and cheaper.

 

[Steevo]

At 6Ghz electrons only move 5.76mm roughly by my calculations, so the theoretical wall is only limited by switching speed and trace length, so by moving to a smaller mfg process the trace length can be shortened and the gate can switch faster, and less material carrying the voltage means less capacitive rolloff and cleaner signal.

 

[Tatty_Two]

32nm is sweet..... but wheteher it's AMD or Intel, it dont mean its faster than 45nm, cooler, yes most probably, lets hope we see some more serious horsepower from new architecture.

 

[crazy pyro]

I'd go for an AMD machine if they were performing the same as intel price for price including OCing when I come to needing a new machine (couple of years yet).

 

[Flibolito]

Well if the transistor shrink and you add more for any give surface you can do more with it. So @3GHz(which is just the carrier wave) you activate more transistors per pass hence increasing processing power ( more 1s & 0s ).
This is why 3GHz on a Intel chip is usually more powerful than on a AMD chip because the AMD has a lower Transistor count.

 

[PCpraiser100]

32nm is gonna be great with AMD's Athlon series for laptops. If they fit in a high-frequency dual-core (2.6GHz at least) in an energy-efficient laaptop that would be awesome.

 

[wolf2009]

this is too late for AMD, God knows what intel will have by then.

 

[eidairaman1]

Noob question here but how does making the die size smaller allow for more stuff do be on it? In other words how does making the room smaller fit more stuff in it?

its the transistor size that gets smaller, meaning the same die used on the 65nm node could have much higher transistor density.

 

[wolf]

lets hope it brings them back in the race, instead of playing second fiddle to intels monster chips.

so by the time they land in 2010, isn't that when intel want to land 32nm 6 core - hyper threaded chips?

 

[crazy pyro]

Yeah but these'll probably be 6 or even 8 threaded as well so they'll be able to compete (I hope otherwise we're all screwed).

 

[HolyCow02]

This is good news! AMD seems to be catching up again! Seems this Foundry Company idea was a good one!

 

[wolf]

"AMD 32 nm CPU Conquest to Begin in 2010"

Conquest you say?

 

[Flyordie]

lol, took you Intel monkeys (no pun intended, there are AMD monkeys too) that long to figure out? AMD expects 32nm by 3Q 2010.
Also with 32nm you can expect to see 20-24 core Processors from AMD since AMD designs their CPUs to be VERY small.

 

[pr0n Inspector]

hey how about this: dream on you Advanced Micro Devices FANBOIS!

 

[Flyordie]

hey how about this: dream on you Advanced Micro Devices FANBOIS!

.....
Last I checked, TFC had more $$ than Intel did... Since well... everytime a dude/woman fills up their tank with gas... thats a few dollars more that goes into TFC. So... telling AMD Fanboys to dream on is a very... sad thing to say.

 

[wolf]

lol, took you Intel monkeys (no pun intended, there are AMD monkeys too) that long to figure out? AMD expects 32nm by 3Q 2010.
Also with 32nm you can expect to see 20-24 core Processors from AMD since AMD designs their CPUs to be VERY small.

by your reasoning on 45nm fab we should already be seeing 8-12 CPU's or more... i do not.

small they may be, but a jump from 4 to 20 is a bit bogus man, they need to step up slowly just like everyone else.

telling AMD fanboys to dream on is a bit sad, but at this stage it is just a dream to come out on top of intel, as soon as they release a product, its already 1 step behind intel.

AM3 came to rival Core 2 marchitecture nicely, just 18 months too late.

 

[KBD]

A little confusion here, i thought the Foundry Co will be based on the plant in upstate NY, if thats the case how is AMD fabless now? They still have the Dresden fab and others. Or they tranferred all their fabs to Foundry co?

 

[spearman914]

Noob question here but how does making the die size smaller allow for more stuff do be on it? In other words how does making the room smaller fit more stuff in it?

The smaller the die size, the smaller the average size of the components. So it doesn't mean fitting more stuff in, it's just shrinking the size of it.

 

[Flibolito]

ok so more cores is not always the best way at all.
more cores = more complex software to actually take advantage.
and AMD has not lost any race at all.
Transistors@(X)GHz/price is the real race
for example 2 chips 1 has 500million @3GHz, the other 1billion@3GHz
if chip 1 is $200 and chip 2 is $65 then chip 2 wins even though chip 1 is faster
you have to understand most people are not enthusiast and dont know the difference very well but like the $65 much more than $200 especially since they never even tap into that power the chip 1 has.
so there are 2 races 1, the race for transistors/power/fabrication. race two is same as race 1 but the mighty $ factored in.

 

[Hayder_Master]

i want say to AMD better focus on platforms , no problem with with high performance cpu with 45NM , but the problem when the platform waste the cpu performance