backtop


Print 33 comment(s) - last by MrPoletski.. on Apr 21 at 9:52 AM

28nm technology promises 40% more performance than 45nm tech

The IBM Technology Alliance -- including IBM, Chartered Semiconductor Manufacturing Ltd., GLOBALFOUNDRIES, Infineon Technologies, Samsung Electronics, Co., Ltd., and STMicroelectronics -- have announced that they have jointly defined and are developing a 28nm, high-K metal gate (HKMG), low power CMOS process technology.

IBM reports that the 28nm technology can provide power-performance and time-to-market advantages for makers of a variety of power-sensitive and consumer electronics devices like MIDs and smartphones. The new technology creates improved leakage characteristics that will optimize battery life for next-gen mobile devices.

The alliance has outlined a migration path from the current 32nm process that is being used to the new 28nm technology that requires no costly and time-consuming redesign of the components according to IBM.

IBM's Gary Patton said in a statement, "Through this collaboration, IBM and its alliance partners are helping to accelerate development of next-generation technology to achieve high-performance, energy-efficient chips at the 28nm process level, maintaining our focus on technology leadership for our clients and partners."

IBM says that early work with some clients has shown that the 28nm technology can provide a 40% performance improvement while saving up to 20% in power compared to 45nm technology devices. The HKMG implementation also makes for one of the industry's smallest SRAM cells reports IBM at only 0.120 square microns.

ST-Ericsson's Jorgen Lantto said, "This statement of commitment to 28nm low-power technology by the IBM Joint Development Alliance is an important progression from 32nm high-k metal gate technology. Leaders in the mobile industry can utilize 28nm low-power technology to meet the increasingly aggressive demands for performance and improved battery life."

IBM recently walked away from purchase talks with Sun after Sun's board balked at IBM offer.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

RE: AMD, Intel, and IBM
By MrPoletski on 4/20/2009 11:34:57 AM , Rating: 2
quote:
The Phenom is a continuation of a failed design (the K7).


The phenom is a new architecture. To give you an idea a dual core K8 has 94m LOGIC (not cache) transistors vs 247m for the quad core phenom. Per core that's 31% more logic transistors.

quote:
The i7 is a continuation of an excellent design (the Pentium Pro).


Again, the i7 is a brand new design. Please learn about the subject you talk about before rattling off an ignorance fueled post.

quote:
The Pentium 4 is, luckily, a dead design. It was designed as a marketing processor, since people bought things based on clock speed, and weren't so shrewd about understanding performance.


The netburst design was ahead of it's time and din't deal with the unexepcted heat issues. Intel should have known better here. All the work they did trying to get the P4 working faster did wonders for the core 2 though. Core 2 didn't need the lga775 socket for signal integrity, or the plethora of power saving featrues the P4 spawned, nor the excellent heat spreader design and hs/f product base. Core 2 inhereited those though and overclocked like a crack addicted baboon.

quote:
Even so, it was competitive with the K7, generally outperforming it with the Northwood, and was competitive with the K8, but probably slightly slower overall. The damning thing was the size and power use. Kind of the same problem the Athlon had compared to the Pentium III.


you are just wrong, the P4 barely competed with anything. It only sold because it was Intel and it had big numbers next to the word 'speed'.

quote:
Luckily though, Intel got smacked around for the Pentium 4,


because it was far and away more of a failure than the K8.

quote:
and AMD is getting raped with its lousy processors.


well I was disappointed with the phenom, AMd needed it to be faster.

quote:
The market is smarter now, and that's a good thing. I'm still confused how AMD can still make such a bad processor for so long, unless they were foolish enough to believe the K8 was good before the Conroe exposed it as a power hungry, slow, inefficient design. Maybe the Pentium 4 did have a purpose, AMD actually thought they had a good design. Let's hope they know better by now and the Bulldozer gets it right.


AMD need some real magic to avoid falling back to what they were like in the K6 days.

quote:
Still in defense of AMD, even though they clearly have bad designs, competing with Intel is a nightmare these days.


You are talking about THOSE days, not these. the k7/k8 was a fabulous stab at intel and really smacked them.

quote:
The Atom is excellent,


ARM is better.

quote:
but with a bad chipset,


stop gap, they used an old chipset because the next gen atom will have it all on die.

quote:
and the i7 is just crazy good.


The i7 is very fast, I should know I run mine at 3.7 ghz. It is VERY VERY expensive though.

quote:
Maybe they should get some help from IBM in design too, they clearly are out of their league. The POWER chips are looking really good, by contrast.


The advantage intel has always had over intel is vastly superior manufacturing facilities and a lot more resoruces for custom gate deisgn (= higher clockspeeds out of the same logic)

quote:
One thing is for sure, the Bulldozer has to be good. If not, we're so screwed as customers, since Intel will command the market again, and we'll have to pay out the nose for processors. I wish IBM would buy AMD and provide real competition again.


IBM buying AMd would not solve anything, IBM are big enough to just start their own PC processor line.


RE: AMD, Intel, and IBM
By TA152H on 4/20/2009 1:03:04 PM , Rating: 2
The Phenom and i7 are both derivative designs based on their predecessors. Even AMD and Intel do not deny this. AMD does say the Bulldozer is new though, but, I have my doubts.

If you really do not believe the Phenom is the same, look at the resources they devote to x87. Since x87 is deprecated, and not even part of x86-64, no one would possibly want to do this. Also look at how the AGUs are part of the same ports as the ALUs. I don't think AMD would do that if they had to start over again, seeing how the Pentium III and its successors were more successful with a different design. In fact, if you think that the i7 is not a Penryn, look at the internals of the processor. They are essentially the same.

ARM isn't a processor, it's an instruction set that processors are based on. You're comparing apples and oranges. Atom can run a whole lot of software, without a whole lot of power. ARM based processors can't run the software. That's kind of a big deal.

I agree the chipset was a stop gap, but it should never have come to that. They new they were developing the Atom, and they should have been able to come up with something better.

The Pentium 4, even the last Presler, did beat the K8 at some benchmarks. The performance on games was poor, but it was competitive in overall performance. But, it was huge and used so much power, being competitive isn't really such a good thing. And it generally was a little slower. The interesting thing is, if they had moved it to 45nm, it probably would have finally showed some merits, since the power use was a lot lower. No doubt it would never have matched Conroe though.


RE: AMD, Intel, and IBM
By MrPoletski on 4/21/2009 7:40:14 AM , Rating: 2
quote:
The Phenom and i7 are both derivative designs based on their predecessors. Even AMD and Intel do not deny this. AMD does say the Bulldozer is new though, but, I have my doubts.


Umm, of course they are going to keep certain elements of the processor. Just like car manufactures tend to still use wheels on their cars instead of inventing something new. Doesn't mean a mustang is a copy of a toyota. Like I said though, to say the i7 is 'the same' as the penryn is like saying the 486 is 'the same' as the 8086.

quote:
If you really do not believe the Phenom is the same, look at the resources they devote to x87. Since x87 is deprecated, and not even part of x86-64, no one would possibly want to do this.


umm, backwards compatiability? the only thing that's really kept the x86 architecture alive this long?

quote:
Also look at how the AGUs are part of the same ports as the ALUs. I don't think AMD would do that if they had to start over again, seeing how the Pentium III and its successors were more successful with a different design.


A design that is not available to them and a design whos superiority is debatable. I mean why doesn't AMD just copy Intels cache specs instead of using its own blatantly inferior cache structure over the K6 to K8 days? because they couldn't.

quote:
In fact, if you think that the i7 is not a Penryn, look at the internals of the processor. They are essentially the same.


??

quote:
ARM isn't a processor, it's an instruction set that processors are based on. You're comparing apples and oranges.


well if your going to get all literal on me then ARM is a company, not an instruction set. They are a company that produces IP and you can use their processor designs if you like. The processors ARM are responsible for are better than atom when you consider the processors ARM make to compete with atom.

quote:
Atom can run a whole lot of software, without a whole lot of power. ARM based processors can't run the software. That's kind of a big deal.


Arm processors can run software, just not x86 software and they are much more power efficent than atom.

quote:
I agree the chipset was a stop gap, but it should never have come to that. They new they were developing the Atom, and they should have been able to come up with something better.


But it would have taken time, time they did not have if they wanted to grab the MASSIVE netbook market that is spawning right now.

quote:
The Pentium 4, even the last Presler, did beat the K8 at some benchmarks.


My casio watch beats the P4 in 'some benchmarks'.

quote:
The performance on games was poor, but it was competitive in overall performance.


But not PRICE.

quote:
But, it was huge and used so much power, being competitive isn't really such a good thing. And it generally was a little slower. The interesting thing is, if they had moved it to 45nm, it probably would have finally showed some merits, since the power use was a lot lower. No doubt it would never have matched Conroe though.


The P4 needed better branch prediction IMHO, a mis-predict was waaay too expensive.


"This is from the DailyTech.com. It's a science website." -- Rush Limbaugh

Related Articles
IBM Walks Away from Sun Negotiations
April 6, 2009, 11:20 AM













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki