Sign in with
Sign up | Sign in
Your question

IBM, AMD Announces 45nm Chip High-K Gate Process

Last response: in CPUs
Share
January 29, 2007 3:57:55 PM

I'm sure this is old news, but I thought it was interesting enough to deserve its own thread.

(link)

IBM plans on producing 45nm High-K chips in early 2008. The article states that this is a development made jointly with AMD. If AMD capitalizes on this, I wonder if they can start producing 45nm chips in early 2008. This would surely close the gap further between AMD and Intel in terms of mfr processes.

One benefit to being a year behind your competition is that you can learn from their mistakes (though someone probably should've told Sony that re: PS3 ;) )

I'm excited to see how this shapes up. I originally thought that the excitement behind release of Conroe was going to last another year or so, but it seems Intel's got a lot up their sleeves. I'm on the edge of my seat.

On a side note, with chip technologies moving ahead so fast, you'd think the high-end sales would suffer (being replaced so quickly and what-not. Especially with the EE965 getting slaughtered so badly < 6 months after its release). This doesn't seem to be the case, though. There seem to be a lot of QX and X-series conroes about.

Very interesting times....
January 29, 2007 4:04:48 PM

*add*

Thinking back, it was less than 4 years ago when we were excited about the 0.13 (130nm) process.

A chip manufactured on the 45nm is 11% the size of the same chip fab'bed on the 130nm process....
January 29, 2007 5:13:28 PM

Quote:
Even funnier is that,They only announced it after Intel did.Who is really pushing the trends?The jedi band or the empire? :lol:  I wonder if AMD will finally use SiGe on K8L.


Was after? I read that was almost at same time but Intel claimed better the tech advance than IBM!! Crazy thing.

Jedi band? Empire? hehehe... that's good!
Related resources
Can't find your answer ? Ask !
January 29, 2007 5:38:09 PM

That article makes me think that AMD isn't as out of the game as I thought they were, or as much as people perceive them to be.
January 29, 2007 6:18:14 PM

Quote:
That article makes me think that AMD isn't as out of the game as I thought they were, or as much as people perceive them to be.


They are far from out of the game;they are just out this round,and its debateable as to the actual reason beyond the AMD backpeddling.

Intel 65nm debuted near mid late mid 05,core duo mobile in dec 05.Here we are over a year later with the new york times calling it 9 months.
Something is dreadfully wrong with this.this is misinformation on a vast scale.

I think it's quite clear as to what is happening with AMD. No big mystery. Look at it this way:

Charlize Theron invited AMD and Intel for a threesome and Intel stole AMD's Viagra.

Droooooooooooooooooooop... :twisted:
January 29, 2007 8:35:57 PM

Quote:
Even funnier is that,They only announced it after Intel did.Who is really pushing the trends?The jedi band or the empire? :lol:  I wonder if AMD will finally use SiGe on K8L.


Was after? I read that was almost at same time but Intel claimed better the tech advance than IBM!! Crazy thing.

Jedi band? Empire? hehehe... that's good!

I dont doubt they were on it at the same time at all;what id like to point out is intels putting it on 45nm and The jedi following suit.
It wouldnt matter either way to intel;what process got it,I think maybe they did it to draw out IBM"s hand.

Which insinuates that Ibm /amd etc were dragging their heels on implementation as a scale improvement at 32nm.Meaning They wanted to draw out the length of the current process :wink:

It could very well insinuate such a thing. :lol: 

Check it out:
http://www.theregister.co.uk/2007/01/28/intel_ibm_highk...
January 29, 2007 9:07:05 PM

Quote:
Even funnier is that,They only announced it after Intel did.Who is really pushing the trends?The jedi band or the empire? :lol:  I wonder if AMD will finally use SiGe on K8L.


Was after? I read that was almost at same time but Intel claimed better the tech advance than IBM!! Crazy thing.

Jedi band? Empire? hehehe... that's good!

I dont doubt they were on it at the same time at all;what id like to point out is intels putting it on 45nm and The jedi following suit.
It wouldnt matter either way to intel;what process got it,I think maybe they did it to draw out IBM"s hand.

Which insinuates that Ibm /amd etc were dragging their heels on implementation as a scale improvement at 32nm.Meaning They wanted to draw out the length of the current process :wink:

It could very well insinuate such a thing. :lol: 

Check it out:
http://www.theregister.co.uk/2007/01/28/intel_ibm_highk...

Great find! I'm curious what JJ would say about this as there and PDF at the end of the article which presents IBM technology.
January 29, 2007 10:50:11 PM

Quote:
hmm really?theres testing and then deposition which takes time.Just when you though High k was interesting here comes a process named after an italian pasta. :wink:


A bit of background. Fusi was first discovered in Naples when a Camorrista Mafioso was tasered in a Trattoria by the Carabinieri. The electic current ran through his body and onto his plate of Fusilli and the resultant deep plasma nitridation melted it together into a single large Fusi with remarkable electronic properties against parasitic leakages which are also common in Naples due to tapeworms. Since then, the application of the Parmigiano electrode on HfSiON and its resultant effect on NiSi sauce has advanced the doping to a point where it is no longer necessary to phase control the sillycide. The sillies just terminate their own existences even without dope.

Got that? Good. There will be a test later.
January 30, 2007 12:26:04 AM

Wow... no one here understands satire?
January 30, 2007 5:39:30 AM

Quote:

:lol:  :lol:  :lol:  history lesson was great,the terminology was better.


No prob. Glad to help clear it up for you. Just ask me anytime you need a technical dissertation.

8)
January 30, 2007 6:12:44 AM

I think after getting my Core 2 Duo machine, i'm not that impress with this latest from Intel. You would ask why" dudes i have a reason for this" After seeing this you would definitely change your mind.
January 30, 2007 6:44:54 AM

I will wait till i save enough money for that cooling kit they are using. After that surely i will make the move to atleast 6GHZ :lol:  :lol:  :lol:  8O 8O :oops:  :oops:  :p  :p  :p 
January 30, 2007 7:02:15 AM

Quote:
I think after getting my Core 2 Duo machine, i'm not that impress with this latest from Intel. You would ask why" dudes i have a reason for this" After seeing this you would definitely change your mind.


Just leave it to the Italians! They can't run a government, keep the streets clean or make the trains run on time but when it comes to making things go FAST... VROOOOOOOOOOOOOOOOOMMMM...
January 30, 2007 7:13:58 AM

Quote:

Just leave it to the Italians! They can't run a government, keep the streets clean or make the trains run on time but when it comes to making things go FAST... VROOOOOOOOOOOOOOOOOMMMM...


Ha ha :p  :p  :p  i will do my personal thingy
January 30, 2007 7:26:50 AM

Quote:

Just leave it to the Italians! They can't run a government, keep the streets clean or make the trains run on time but when it comes to making things go FAST... VROOOOOOOOOOOOOOOOOMMMM...


Ha ha :p  :p  :p  i will do my personal thingy

Dude, I have no desire to watch you do your personal thingy. Go do your thingy somewhere else. :lol: 
January 30, 2007 7:33:30 AM

Quote:


Dude, I have no desire to watch you do your personal thingy. Go do your thingy somewhere else. :lol: 
Thats nice me 2 don't wanna fight over a CPU.
January 30, 2007 7:36:26 AM

If Intel is one year in advance in technology, why did AMD hold the crown for approx 3 years with outdated technology? anything to brag about being the first i suppose, doesnt matter if it run's like crap.
January 30, 2007 7:40:32 AM

Quote:

Dude, I have no desire to watch you do your personal thingy. Go do your thingy somewhere else. :lol: 
Thats nice me 2 don't wanna fight over a CPU.

CPU? Is that what you call it? I thought you were talking about your willy! :lol: 

Quote:
If Intel is one year in advance in technology, why did AMD hold the crown for approx 3 years with outdated technology? anything to brag about being the first i suppose, doesnt matter if it run's like crap.


At the time AMD held the performance crown, they had cutting edge technology and Intel was sucking hind tit. However, time has a way of passing and Intel has just trumped AMD into the weeds. That may change again in the future or then again it might not. We'll just have to wait and see.
January 30, 2007 7:41:52 AM

Quote:
If Intel is one year in advance in technology, why did AMD hold the crown for approx 3 years with outdated technology? anything to brag about being the first i suppose, doesnt matter if it run's like crap.


Its up to user's preference, I too used to have a AMD, 2 years back but that PC is now used by my Sister. I'm upset at this PC thing, jsut cos a new game needs DX9, 10 doesn't means i have 2 upgrade. Thats why i'm hoping to get a XBOX 360 version 2 when HALO 3 gets release. I'm shit :oops:  angery at yearly upgrades, can't ask DAD to pay my bills every now and then. I will get a 360 and hope that last for quite a while.
January 30, 2007 7:44:46 AM

Quote:
Its up to user's preference, I too used to have a AMD, 2 years back but that PC is now used by my Sister. I'm upset at this PC thing, jsut cos a new game needs DX9, 10 doesn't means i have 2 upgrade. Thats why i'm hoping to get a XBOX 360 version 2 when HALO 3 gets release. I'm **** :oops:  angery at yearly upgrades, can't ask DAD to pay my bills every now and then. I will get a 360 and hope that last for quite a while.


I'm writing this on a San Diego and I've been on AMD for years. This summer when I do my massive upgrade I'll go with AMD or Intel. Don't care which one. Whatever is on the market at the time that offers the best price/performance ratio. Couldn't give a rat's cornhole what fanboys say. The bottom line is always where you can get the best bang for your buck.

I tried asking my dad to pay my bills once when I was young. He replied by throwing me outta the house! :D 
January 30, 2007 7:48:50 AM

Quote:


I tried asking my dad to pay my bills once when I was young. He replied by throwing me outta the house! :D 


Ha ha i will do some more :roll: :roll: blogging and earn some money so tha t this time i BUY THAT 350$ XBOX 360
January 30, 2007 7:50:51 AM

Quote:
Quote:


I tried asking my dad to pay my bills once when I was young. He replied by throwing me outta the house! :D 


Ha ha i will do some more :roll: :roll: blogging and earn some money so tha t this time i BUY THAT 350$ XBOX 360


Dude, you'd have a better shot at getting people to pay to watch you play with your thingy than you would in blogging...
January 30, 2007 7:55:32 AM

This would we very risky at the moment. Cos if somehow even after taking all precautions, i melt my PC, you know my DAD's gono make a million pieces of mine. I will wait till i get the 360 this year, then i will experiment :roll: :roll: :roll: :roll: :roll: :roll: :roll:
January 30, 2007 8:03:53 AM

Thingy=Willy=D**k... never mind... :D 
January 30, 2007 8:07:07 AM

well i'm a loser i gono sleep now what can i do :(  :(  :(  :(  :(  :(  :(  :( 
January 30, 2007 8:08:36 AM

Quote:
Thingy=Willy=D**k... never mind... :D 
That went over his head. :wink:
January 30, 2007 8:27:54 AM

Quote:
Thingy=Willy=D**k... never mind... :D 
That went over his head. :wink:

And I'm the one who has the reputation around here for having a dirty mind... :lol: 
January 30, 2007 9:08:01 AM

Quote:
so let me get this straight intel is making chips with this process and has demoed them already working with benches and stuff. and IBM is saying they can do this too but they have no working chips or anything?

is this correct?


No. Both are still working on this technology but both wanted at this time announce it first to the public. The funny thing is that it was almost the sametime, intel slightly first and with better quality message to the audience!

What intel says is that it will use this technology in the 45nm process to come. IBM says it will use this technology on the POWER risc cpus first.
And we get all happy!! :-)
January 30, 2007 10:50:59 AM

Quote:
IBM says it will use this technology on the POWER risc cpus first.
And we get all happy!! :-)


Wouldn't it be fun if a POWER RISC system could be used for conventional PC uses, the way the Mac once had? If the sw was optimized for the RISC instruction set, it would 'effin fly!
January 30, 2007 11:07:53 AM

Quote:
IBM says it will use this technology on the POWER risc cpus first.
And we get all happy!! :-)


Wouldn't it be fun if a POWER RISC system could be used for conventional PC uses, the way the Mac once had? If the sw was optimized for the RISC instruction set, it would 'effin fly!

The old WinNT 3.51 for old Digital workstations (DEC) was a RISC version?
January 30, 2007 12:44:21 PM

Quote:
Short channel HK/MG devices were fabricated from <20Å HfO2 with
thermally stable BE metal gates in a gate-first approach where
conventional poly-Si is deposited over the metal gates. Following a
lithographic patterning and gate stack etch process, a conventional selfaligned
implant process flow with a final S/D spike RTA (T>1080°C) +
advanced annealing (AA) and dual stress liner (DSL) with conventional
MOL and BEOL was used. Fig. 3 shows a TEM image of the nFET
HK/MG stack after full build showing a physical gate length of 33nm.Short channel device properties of the HK/MG nFET devices exhibit
superior drive current and short channel control as compared to state-ofthe-
art 65nm node 11Å SiON/Poly-Si gate stacks. DIBL improvement
of 15mV at fixed Lgate as shown in Fig. 4 supports a 5nm reduction in
channel length as predicted from the simulations in Fig 1. The
improved short channel control is due to high-!/metal gate process
improvements that enable Tinv <12Å as seen in Fig 5 by preventing regrowth
that plagues gate-first processing. Fig. 6 shows a DC Idsat of
1240uA/um @ Ioff = 200nA/um with a corresponding AC Idsat
1364uA/um (non-self-heated) in the nFET HK/MG devices at 1.0V.Well controlled Vtlin / Vtsat roll-off to an Lgate of 25nm is shown in
Fig. 7, demonstrating excellent short channel control. Fig. 8 shows that
Idsat is maintained as Lgate is scaled compared to an 11Å SiON at fixed
Ioff and optimal Cov. The flatness of the HK/MG vs. SiON curve again
demonstrates the improved short channel control in these HK/MG


cool they actually mention the compound :wink:

http://periodic.lanl.gov/elements/4.html beryllium metal gates :wink:

BE stands for Band Edge, not Beryllium. If it were Beryllium gates it would have been labeled Be.
January 30, 2007 1:58:49 PM

Quote:
IBM says it will use this technology on the POWER risc cpus first.
And we get all happy!! :-)


Wouldn't it be fun if a POWER RISC system could be used for conventional PC uses, the way the Mac once had? If the sw was optimized for the RISC instruction set, it would 'effin fly!

The old WinNT 3.51 for old Digital workstations (DEC) was a RISC version?

Yup. The only one they ever did. Then they dumped it. Now apparently the current dual cores run some mix of CISC and RISC but I'll be damned if I have a clue as to how that works.
January 30, 2007 6:24:12 PM

Quote:
Bandage metal gates :?

ok ;one question;why the hell is it so friggen hard to find the gate materials?and jack should have corrected that;I guess i am not liked any more,if he chooses not to correct my falsely reading something. :( 

(joking)

what sense does band edge metal gate mean?the gate is in a trench deposited in vapor form,i am not seeing a band or an edge significance.
I think diaper-duty trumps forum-duty...and rightly so. :wink:
!