News: Amd Ships First Athlons Following 65nm Tech Transition

[sdy284]

c|net Article

CRN Article

DailyTech Article

Actual AMD Press Release

OCC News Post

 

AMD chips made on the 65-nanometer process will consume about 30 percent less energy than the same ones produced on the 90-nanometer process when running at the same speed. In the first chips shipped on this process, virtually all of the benefit comes in reduced power consumption, said Nick Kepler, vice president of logic technology development at AMD.

 

Later, the company will balance energy conservation and performance gain, depending on what the designers want to achieve with various desktop, notebook and server chip designs, he said.

 

The improvements in part come because AMD is straining the silicon in the transistors inside its chips with a silicon germanium film to improve performance, a first for the company.

 

Straining silicon improves the performance of transistors because the larger germanium atoms slightly rearrange the silicon atoms and thereby allow electrical carriers to move more rapidly. In the N-channel transistors, which carry negative electrons, germanium spreads out the lattice of the silicon; electrons flow more freely, sort of like a deer running through a forest that's been thinned of trees. In P-channel transistors, which carry positive charges, the germanium compresses the silicon atoms.

 

"We planned it (silicon germanium) all along with 65-nanometer, but we were prepared to pull the plug on it if it didn't add up to a benefit," Kepler said. "Adding embedded silicon germanium is probably the biggest change."

[LoArmistead]

Very nice! High-five!

 

sort of like a deer running through a forest that's been thinned of trees.

 

I really wasn't expecting to see that kind of comparison in a tech article, so I got a chuckle out of that line

[sdy284]

more sources & processor breakdown added

[bigred]

is it me or did they screw with the FSB on these chips again?

[Verran]

Well I got my answer to a previous question from this DailyTech article:

Do note that Brisbane core processors are still K8 based and not K8L like the upcoming Stars core processors.

[romeo55]

Well I got my answer to a previous question from this DailyTech article:

 

meh... die shrink and that's all? not too exciting in my book It'll probably be cooler, use less power... and get a tad bit more OCing...

[Verran]

I think one of the smartest things AMD can do at this point is offer a 939 version of the 65nm chips. Yeah, it'd be great for me, so I'm kinda biased, but still. If people are forced to go AM2 for all their AMD 65nm needs, then that means new chip, ram, and mobo all at once. Why not just go with the crown-holder in that case?

 

If I'm given the option of a 939 65nm chip, I'll take it. I have no desire to upgrade my mobo and ram, as they both rock, and the "upgrades" wouldn't even provide a substantial performance increase with DDR2. If I'm not offered a 939 65nm, I'll more than likely go Conroe, and AMD will lose my business completely. At best, I expect this 65nm shrink to match where Intel is (especially with their 1066 chips coming), so why buy brand new buggy AMD 65nm when I can get the same or better performance from the tried and true Intel chips that have been out for half a year now?

 

The smart move would be to give current 939 users a reason to stay in the AMD ballpark. If they're forced to upgrade to AM2, AMD will probably lose them as customers all together.

 

That being said, I can find no evidence of 939 65nm chips, other than red's ES mysteries.

[DECwakeboarder]

They have a HUGE base in the 939 market, and forcing everyone to upgrade to AM2 just to get 65nm would be a huge mistake (read: an Intel mistake ) Granted they may have a few features that require AM2, but just going to 65nm would be a huge plus for us 939 guys. What is with the 2x512kb or cache though? That seems like quite a waste. Granted Intel's 2x4mb of cache is also a waste, but I'd like to see at least 2x1mb on something as new as this.

[fire_storm]

I wonder do these chips have better performance then the 90nm ones or do they preform about the same but the 65nm ones just make less heat then the 90nm ones with no performance advtage?

[bigred]

I wonder do these chips have better performance then the 90nm ones or do they preform about the same but the 65nm ones just make less heat then the 90nm ones with no performance advtage?

 

 

questions like that make me swear at my monitor! especially when the answer is sitting right there at the bottom of the first post.

 

 

they have the rating system for a really simple reason. a 5000+ performs in line with a 5000+. what a shock there. yet the cache size is now 1/2 what it used to be.

[CrackHead]

Does anyone know how these new AMD chips stand up to Intel's current ones?

[Crazystickfigure]

Does anyone know how these new AMD chips stand up to Intel's current ones?

 

Just a guess out in the blue here, but according to what bigred said about the chips being the equivalent of the current line, then they would preform just like the current line and size up to intel just like the current line. In other words, they will be far off the mark. Intel still owns the crown.