AMD 45nm slips to second half 2008

[Phynaz]

In an interview Tuesday (Dec. 11) during the International Electron Device Meeting (IEDM) here, John Pellerin, AMD's director of logic technology development and project leader on a joint development effort with IBM Corp., said the company expects to start shipping the new processors in the second half of 2008.

 

[exar333]

Isnt the first variant of AMD 45nm SOI and not Hi-K like Intel? Will that mean higher power consumption for AMD?

 

[Nemesis 1]

I don't have a clue what AMD 45 is going to be . First I heard was lowk using zram .

I believe Zram was projected for the 45nm process. Which requires soi. But After Intel announced that It was using High K and metal gates. IBM jumped in and said ya ME and AMD have high K on metal gates @ 45nm . Also . Even tho they always said 32nm.

So Who knows? I know nothing AMD says can be believed but I thought IBM might be differant . Who knows?

But If AMD comes with High K and metal gates @ 45nm you can bet. It will be more effcient than intels single gate . Compare to AMDs finfet gate . The problem with finfet at 45nm is the gate alinment. Now Intel is getting a little heat over using gate last . But at the time its all they can do. That brings me to this. Intel had to go gate last because the metal gate melted during the annealing process. If AMD/IBM go gate first with finfet they could have an alinment problem after annealing . So I am betting IBM will also be gate last.

 

[GundamF91]

The tech race is going, and that's a good thing for innovation and pricing. I had hoped AMD wouldn't lag behind too much, but as long as they hang in there, it'll drive Intel to do better as well.

 

[Phynaz]

IBM is going gate first.

IBM and Advanced Micro Devices (AMD, Sunnyvale, Calif.) separately are developing a silicon-on-insulator (SOI) 32 nm technology, based on the HKMG gate-first solution developed for the 45 nm generation. Patton said IBM will use the SOI 45 nm process for IBM?s proprietary products, and will offer a non-high-k 45 nm process to its foundry and OEM customers.


Text

 

[Miramonti]

"2nd half"

Kind of scary imo they aren't pinning it down to a 'quarter'.

 

[nonameo]

Originally posted by: jjsole
"2nd half"

Kind of scary imo they aren't pinning it down to a 'quarter'.

LOL, any time a half is specified, it means late late late 2q or late late late 4q.

 

[sonoran]

I'm betting late 2nd half...anyone want to place bets?

Has anyone seen a working 45nm chip demonstrated by IBM or AMD yet?

 

[Idontcare]

Originally posted by: sonoran
I'm betting late 2nd half...anyone want to place bets?

Has anyone seen a working 45nm chip demonstrated by IBM or AMD yet?

I doubt you'll be able to buy an AMD 45nm CPU from Newegg anytime in 2008.

This is AMD we are talking about. That means sometime in April they'll claim their 45nm is 40% better than Intel's, claim it is shipping in volume in Aug and folks will be looking for them still come December.

The only thing I expect to find entertaining about AMD's 45nm node is going to be the process of seeing if they continue this newest trend of their's where successive technology nodes result in ever decreasing maximum speed-bin parts. (inverse Moore's law? Should we call it Hector's Law?)

Does the 90nm > 65nm > 45nm trend continue? Will AMD "break" the 1GHz barrier (only in reverse) by the time they get to 32nm? Shareholders are holding their breath in utter expectation!

 

[myocardia]

Originally posted by: Idontcare
I doubt you'll be able to buy an AMD 45nm CPU from Newegg anytime in 2008.

This is AMD we are talking about. That means sometime in April they'll claim their 45nm is 40% better than Intel's, claim it is shipping in volume in Aug and folks will be looking for them still come December.

The only thing I expect to find entertaining about AMD's 45nm node is going to be the process of seeing if they continue this newest trend of their's where successive technology nodes result in ever decreasing maximum speed-bin parts. (inverse Moore's law? Should we call it Hector's Law?)

Does the 90nm > 65nm > 45nm trend continue? Will AMD "break" the 1GHz barrier (only in reverse) by the time they get to 32nm? Shareholders are holding their breath in utter expectation!

Haha, I know you've only got ~700 posts, but this is by far the best of them. Hector's Law, I love it.

 

[n7]

Originally posted:laugh: by: Idontcare

Originally posted by: sonoran
I'm betting late 2nd half...anyone want to place bets?

Has anyone seen a working 45nm chip demonstrated by IBM or AMD yet?

I doubt you'll be able to buy an AMD 45nm CPU from Newegg anytime in 2008.

This is AMD we are talking about. That means sometime in April they'll claim their 45nm is 40% better than Intel's, claim it is shipping in volume in Aug and folks will be looking for them still come December.

The only thing I expect to find entertaining about AMD's 45nm node is going to be the process of seeing if they continue this newest trend of their's where successive technology nodes result in ever decreasing maximum speed-bin parts. (inverse Moore's law? Should we call it Hector's Law?)

Does the 90nm > 65nm > 45nm trend continue? Will AMD "break" the 1GHz barrier (only in reverse) by the time they get to 32nm? Shareholders are holding their breath in utter expectation!

:laugh:

So funny & yet so sad...

 

[Roy2001]

Originally posted by: Idontcare
I doubt you'll be able to buy an AMD 45nm CPU from Newegg anytime in 2008.

This is AMD we are talking about. That means sometime in April they'll claim their 45nm is 40% better than Intel's, claim it is shipping in volume in Aug and folks will be looking for them still come December.

The only thing I expect to find entertaining about AMD's 45nm node is going to be the process of seeing if they continue this newest trend of their's where successive technology nodes result in ever decreasing maximum speed-bin parts. (inverse Moore's law? Should we call it Hector's Law?)

Does the 90nm > 65nm > 45nm trend continue? Will AMD "break" the 1GHz barrier (only in reverse) by the time they get to 32nm? Shareholders are holding their breath in utter expectation!

Wow, Hector's law. You are genius

So in order to push Phenom's frequency up AMD need to do a die-enlarge, make a 450mm^2 mono-die real quad-core chip with 90nm process. Fantastic.

 

[Phynaz]

This is AMD we are talking about. That means sometime in April they'll claim their 45nm is 40% better than Intel's

They already did.

In the interview here, Mario Rivas, Executive Vice President of the Computing Products Group said "our 45nm will be as good as the other guy's 32nm"

 

[taltamir]

Everywhere I read on the forums people were talking like the TLB errata was due to a process issue or some such causing a part of the CPU to overheat...

But I read the explanation by AMD and according to them they messed up the order in which L2 and L3 cache are updated so that you could end up with different data on L2 and L3 cache (With the L3 data being wrong). If the calculation finished with the L2 cache then nothing happened, but if ANOTHER process was intensive enough to cause the first one to drop out of L2 cache it will then later get a copy of what it dropped from L3, and THAT copy was wrong, causing the crash.

This means that any time a sufficiently intensive operation occurs the chip will crash. Above 2.4ghz almost every program is sufficiently intensive to deplete L2 cache causing the crash (not the error mind you, the crash because of the error). But even at 2.3 ghz certain programs (like photoshop for example) will cause it quite often. That is why the whole shebang has to be disabled by the bios... but with it disabled you loose 10-20% in performance...

To fix such a problem they will have to update the L2 and L3 cache in a slower process, resulting in a speed DECREASE not an increase. OR use a much more complicated logic design (the ability to somehow update BOTH at once is one example, another example given was locking the data against change temporarily...).

I am guessing they probably went with the "more advanced circuit design" to fix it, which means they are taking extra time and hoping to end up with something faster, rather then slower (but not as slow as 10-20% slower). But we will know when the xx50 versions show up.

EDIT: Doh, phenom is 65nm not 45.... I was talking about some unrelated stuff the whole time.

BTW, due to the nature of the problem with phenom it should provide NO delays whatsoever in AMD's transition to 45nm. Since it had nothing to do with manufacturing and was due to an architectural flaw in the design of the chip.

 

[bryanW1995]

Originally posted by: Idontcare

Originally posted by: sonoran
I'm betting late 2nd half...anyone want to place bets?

Has anyone seen a working 45nm chip demonstrated by IBM or AMD yet?

I doubt you'll be able to buy an AMD 45nm CPU from Newegg anytime in 2008.

This is AMD we are talking about. That means sometime in April they'll claim their 45nm is 40% better than Intel's, claim it is shipping in volume in Aug and folks will be looking for them still come December.

The only thing I expect to find entertaining about AMD's 45nm node is going to be the process of seeing if they continue this newest trend of their's where successive technology nodes result in ever decreasing maximum speed-bin parts. (inverse Moore's law? Should we call it Hector's Law?)

Does the 90nm > 65nm > 45nm trend continue? Will AMD "break" the 1GHz barrier (only in reverse) by the time they get to 32nm? Shareholders are holding their breath in utter expectation!

QFT! LOL!

 

[Idontcare]

Originally posted by: Phynaz

This is AMD we are talking about. That means sometime in April they'll claim their 45nm is 40% better than Intel's

They already did.

In the interview here, Mario Rivas, Executive Vice President of the Computing Products Group said "our 45nm will be as good as the other guy's 32nm"

I died a little on the inside when I read that article. Thanks for the link even still.

It's almost like AMD looked at Steve Jobs and the Apple reality distortion field and decided they'd give it try for themselves this year.

 

[imported_smilingcrow]

Originally posted by: taltamirBTW, due to the nature of the problem with phenom it should provide NO delays whatsoever in AMD's transition to 45nm. Since it had nothing to do with manufacturing and was due to an architectural flaw in the design of the chip.

I read the opposite of this somewhere. AMD will be currently working towards taping out at 45nm the initial draft of the silicon and if that design has the bug which it likely does it means that they have to fix it. It?s hard to say how much time they will lose over this.

 

[CTho9305]

taltamir, have you made that exact same post in 3 different threads now?

 

[Idontcare]

Originally posted by: CTho9305
taltamir, have you made that exact same post in 3 different threads now?

Just Great ...now he'll be flattered that someone is paying attention to him and we'll see the same misinformation in a fourth thread soon.

 

[CTho9305]

Originally posted by: Idontcare

Originally posted by: CTho9305
taltamir, have you made that exact same post in 3 different threads now?

Just Great ...now he'll be flattered that someone is paying attention to him and we'll see the same misinformation in a fourth thread soon.

Anandtech Moderator could put an end to that if it continues.

 

[taltamir]

Don't be a jerk. I made a thread about this... was pointed at THIS thread. and decided it made more sense to post here... so i posted it here as well. If I could I would have deleted that original thread, but deleting posts in not an option in this forum.

And what misinformation exactly? do you have anything to saw about what I SAID or only the fact I said it twice in a relevant manner?

 

[Idontcare]

Originally posted by: taltamir
Don't be a jerk. I made a thread about this... was pointed at THIS thread. and decided it made more sense to post here... so i posted it here as well. If I could I would have deleted that original thread, but deleting posts in not an option in this forum.

And what misinformation exactly? do you have anything to saw about what I SAID or only the fact I said it twice in a relevant manner?

No major offense intended taltamir, just joshing with you tongue-in-cheek like.

But seriously I was referrring to what you have edited your post to now state as well:

Originally posted by: taltamir
EDIT: Doh, phenom is 65nm not 45.... I was talking about some unrelated stuff the whole time.

Thanks for clearing up the misinformation in your own post. And please don't be so offended next time some others call it like they read it.