blademaster
I'm moving to Brazil
+2,075|6909
Who came first? The chicken or the egg? This story continues. After Intel demonstrated its 32nm Westmere processors last month, AMD is back with some 32nm silicon fabrication technology update. AMD's CEO Dirk Meyer stated on Friday that the company intends to 'ramp up' 32nm processor production in mid-next year. AMD plans to start shipping 32nm processors by fourth quarter of 2010.

As mentioned during the processor roadmap announcement, AMD plans to finish testing 32nm processors by end of 2009. AMD is lagging behind Intel by 12 months in terms of die-shrink of processors. This gap is both good and bad for AMD. Good because AMD can learn from rival Intel's shortcomings in particular processor technology. Bad because if their processors come almost a year late while Intel's processors are happily flourishing the CPU arena.

Meyers talked about AMD's production with InformationWeek stating that company will 'ramp up' 32nm processor for chips production in the mid of next year. As mentioned in the roadmap announced last November, AMD will bring 32nm process CPUs dubbed Ontario, Llano and Orochi. Meyer said that AMD will roll out Yukon platform for ultra-thin laptops.

Being fab-less company, AMD's new manufacturing division Foundry Company will be fabricating these 32nm silicon chips.

However, their graphics deal with Taiwan Semiconductor Manufacturing Company and ATI deal remains unaffected in any case. It's uncertain whether AMD might be integrating major graphics in 32nm process through Foundry Company.

source
GR34
Member
+215|6808|ALBERTA> CANADA
amd should just skip 32nm tbh and start working on 22nm if that makes sence
Lawdy_
ahhhh dangit!
+15|6228|Australia
whats the fastest atm?
Little BaBy JESUS
m8
+394|6412|'straya

Lawdy_ wrote:

whats the fastest atm?
You mean fastest or smallest die-shrink
Cheez
Herman is a warmaphrodite
+1,027|6702|King Of The Islands

We heard you the first time LBJ.
My state was founded by Batman. Your opinion is invalid.
Freezer7Pro
I don't come here a lot anymore.
+1,447|6461|Winland

Cheez wrote:

We heard you the first time LBJ.
PROTIP: It's a bug.
The idea of any hi-fi system is to reproduce the source material as faithfully as possible, and to deliberately add distortion to everything you hear (due to amplifier deficiencies) because it sounds 'nice' is simply not high fidelity. If that is what you want to hear then there is no problem with that, but by adding so much additional material (by way of harmonics and intermodulation) you have a tailored sound system, not a hi-fi. - Rod Elliot, ESP
Cheez
Herman is a warmaphrodite
+1,027|6702|King Of The Islands

Freezer7Pro wrote:

Cheez wrote:

We heard you the first time LBJ.
PROTIP: It's a bug.
PROTIP: I know.
My state was founded by Batman. Your opinion is invalid.
Lawdy_
ahhhh dangit!
+15|6228|Australia

Little BaBy JESUS wrote:

Lawdy_ wrote:

whats the fastest atm?
You mean fastest or smallest die-shrink

Lawdy_ wrote:

Way to fail

Lawdy_ wrote:

Way to fail

Lawdy_ wrote:

Way to fail
Um well wouldnt the smaller it be the faster, so it means the same thing?
GC_PaNzerFIN
Work and study @ Technical Uni
+528|6678|Finland

eh no... that depends in 10000000 other things than die shrinks.
3930K | H100i | RIVF | 16GB DDR3 | GTX 480 | AX750 | 800D | 512GB SSD | 3TB HDD | Xonar DX | W8
Lawdy_
ahhhh dangit!
+15|6228|Australia

GC_PaNzerFIN wrote:

eh no... that depends in 10000000 other things than die shrinks.
Well, i wanted to know, which is the fastest at current time like the lowest nm
Little BaBy JESUS
m8
+394|6412|'straya
Gah i cant stop the bug. and if i delete one of the posts they all get deleted so why dont u quit ur bitching

Last edited by Little BaBy JESUS (2009-03-03 01:10:52)

GC_PaNzerFIN
Work and study @ Technical Uni
+528|6678|Finland

Lawdy_ wrote:

GC_PaNzerFIN wrote:

eh no... that depends in 10000000 other things than die shrinks.
Well, i wanted to know, which is the fastest at current time like the lowest nm
smaller isn't the faster always. but 45nm atm and 32nm soon.
3930K | H100i | RIVF | 16GB DDR3 | GTX 480 | AX750 | 800D | 512GB SSD | 3TB HDD | Xonar DX | W8
Lawdy_
ahhhh dangit!
+15|6228|Australia

GC_PaNzerFIN wrote:

Lawdy_ wrote:

GC_PaNzerFIN wrote:

eh no... that depends in 10000000 other things than die shrinks.
Well, i wanted to know, which is the fastest at current time like the lowest nm
smaller isn't the faster always. but 45nm atm and 32nm soon.
ah ok, thanks
Little BaBy JESUS
m8
+394|6412|'straya
Testing whether cheez's fixed my problem

EDIT: Woot it worked

Last edited by Little BaBy JESUS (2009-03-03 01:15:26)

Little BaBy JESUS
m8
+394|6412|'straya

Lawdy_ wrote:

GC_PaNzerFIN wrote:

Lawdy_ wrote:


Well, i wanted to know, which is the fastest at current time like the lowest nm
smaller isn't the faster always. but 45nm atm and 32nm soon.
ah ok, thanks
Generally doesnt a smaller die give less power consumption, cheaper to run and more (over)clocking headroom.

(im probably wrong lol)
CrazeD
Member
+368|6936|Maine

GC_PaNzerFIN wrote:

eh no... that depends in 10000000 other things than die shrinks.
In fact, faster isn't really a trait of smaller die at all.

Smaller die = less voltage = less heat.

As a technical overview, that's about it.
GC_PaNzerFIN
Work and study @ Technical Uni
+528|6678|Finland

Little BaBy JESUS wrote:

Lawdy_ wrote:

GC_PaNzerFIN wrote:

smaller isn't the faster always. but 45nm atm and 32nm soon.
ah ok, thanks
Generally doesnt a smaller die give less power consumption, cheaper to run and more (over)clocking headroom.

(im probably wrong lol)
It is cheaper to manufacture as you can fit more chips in same silicon wafer. Usually it lowers power consumption and gives more overclocking headroom... BUT

Many things can have exactly opposite effect too like:

-leaky process technology (like 40nm atm has poor overclocking until they fix the leakage issue and lots of partially not working stuff making it useless for high-end chips atm. bigger chip = higher chance that its not 100% perfect chip)

-possible architechture changes/revised silicon resulting in better/worse clockspeed scaling (A11 revision RV790 core with higher clockspeed scaling than older RV770 core etc.)

edit: yeah 65nm GTX 260 and 55nm GTX 260 perform identically at same clockspeeds for example

Last edited by GC_PaNzerFIN (2009-03-03 03:03:57)

3930K | H100i | RIVF | 16GB DDR3 | GTX 480 | AX750 | 800D | 512GB SSD | 3TB HDD | Xonar DX | W8
Doctor Strangelove
Real Battlefield Veterinarian.
+1,758|6732
It sucks but Intel is probably going to have 31nm Processors by 2010.
GC_PaNzerFIN
Work and study @ Technical Uni
+528|6678|Finland

Q4 2009 tbh

edit: 32nm

Last edited by GC_PaNzerFIN (2009-03-03 14:43:14)

3930K | H100i | RIVF | 16GB DDR3 | GTX 480 | AX750 | 800D | 512GB SSD | 3TB HDD | Xonar DX | W8

Board footer

Privacy Policy - © 2025 Jeff Minard