WSJ: IBM to Unveil a Powerful Chip for Home-Entertainmnt Mkt

Status
Not open for further replies.

FatherJohn

Newcomer
Don't know if you can read the full article without a WSJ subscription. Here's the good stuff. (The rest of the article is fairly light-weight speculation about how Cell will affect Intel, and also what new consumer applications might come from all this extra compute power. Pretty lame stuff like "Can reorient the video stream of the football game to show you how the game would look from the endzone", as if you couldn't do that trivially with any GPU. Xbox 2 is mentioned in passing as another product that's based on IBM Power technology.)

http://online.wsj.com/article/0,,SB110168960095385269,00.html?mod=home_whats_news_us

IBM to Unveil a Powerful Chip
For Home-Entertainment Market

By WILLIAM M. BULKELEY
Staff Reporter of THE WALL STREET JOURNAL
November 29, 2004; Page A3

IBM has been working with Japanese electronics giants Sony Corp. and Toshiba Corp. for four years on a chip design, code-named "Cell," and is expected today to announce the chip will go into limited production next year and will be used in several consumer-product applications starting in 2006.

The products include high-definition television sets from Sony and Toshiba and a "home server for broadband content" from Sony, the three companies are expected to announce jointly. Cell also is expected to be the brains of the next generation of Sony's popular PlayStation videogame system. Sony hasn't discussed timing of the product, but analysts don't expect it to appear before 2006.

IBM and Sony have invested $400 million in development of the chip. Sony has committed $325 million more for chip production at IBM's East Fishkill, N.Y., semiconductor plant.

Sony and IBM are expected to announce that next year they will start selling the first Cell-based product -- a high-performance workstation designed for use by videogame designers and Hollywood animation houses. Pricing and marketing plans haven't been determined. IBM said a version of the workstation mounted in a rack with multiple Cell processors will be able to perform 16 trillion mathematical operations a second. That speed would theoretically make it faster than all but a dozen of the world's supercomputers, although much of its power is dedicated to graphics processing rather than to general-purpose computing.
 
PC-Engine said:
Heh if the PC is reaching it's limits then Intel must be very afraid... :LOL:

Maybe that would explain their dramatic shifting of resources into parallel processing across their entire line-up?

  • EETimes.com said:
    If Kahle at IBM started with a nearly clean sheet, Pawlowski (Stephan Pawlowski, director of the microprocessor technology lab at Intel Corp.'s Hillsboro Facility) starts with an instruction set. "We are trying to establish a new paradigm to run those instructions," he said, adding that "it's a personal mission, it's a pride thing for us. Developing for these power-restrictive environments, with highly threaded apps, is a hard problem. Going from an eight-way core to a 32-processor core is a challenge, starting with how to do a compiler that could extract the parallelism."
 
Vince said:
PC-Engine said:
Heh if the PC is reaching it's limits then Intel must be very afraid... :LOL:

Maybe that would explain their dramatic shifting of resources into parallel processing across their entire line-up?

  • EETimes.com said:
    If Kahle at IBM started with a nearly clean sheet, Pawlowski (Stephan Pawlowski, director of the microprocessor technology lab at Intel Corp.'s Hillsboro Facility) starts with an instruction set. "We are trying to establish a new paradigm to run those instructions," he said, adding that "it's a personal mission, it's a pride thing for us. Developing for these power-restrictive environments, with highly threaded apps, is a hard problem. Going from an eight-way core to a 32-processor core is a challenge, starting with how to do a compiler that could extract the parallelism."

It's not like the PC architecture will die just because Intel is moving to a new architecture. Intel defines the PC architecture so it's not going to go away so there's no "PC architecture limit" since it always evolving.
 
Oh please PC-Engine.. is that twisting and clutching at straws really necessary.
Everybody understands what Kutaragi meant with that "PC architecture".
Agreed, it certainly was not the best choice of words, maybe he should've added "traditional PC architecture".
 
I think what he meant for PC architecture was not the Intel or AMD chips by themselves, but the... architecture, as in the bottlenecks, the North South West-End Bridge, the AGP/PCI express issues, the bus here and there, thedifferent kinds of memory, the bottlenecks associated with that........
An Intel chip on its own is not PC architecture. A whole PC is Pc Architecture.
Don't think that's too hard to grasp.
 
london-boy said:
I think what he meant for PC architecture was not the Intel or AMD chips by themselves, but the... architecture, as in the bottlenecks, the North South West-End Bridge, the AGP/PCI express issues, the bus here and there, thedifferent kinds of memory, the bottlenecks associated with that........
An Intel chip on its own is not PC architecture. A whole PC is Pc Architecture.
Don't think that's too hard to grasp.

Well that's like comparing apples and oranges. Why would you define a PC architecture to a console architecture to a supercomputing architecture? Regardless you are aware that Intel does indeed have a hand in defining the PC's architecture right?

McFly said:
PC-Engine said:
Intel defines the PC architecture

Actualy it's AMD right now that defines PC architecture (the 64bit CPU's). ;)

Fredi

Yeah and it's based on x86 with a license from Intel... :LOL:
 
PC-Engine said:
london-boy said:
I think what he meant for PC architecture was not the Intel or AMD chips by themselves, but the... architecture, as in the bottlenecks, the North South West-End Bridge, the AGP/PCI express issues, the bus here and there, thedifferent kinds of memory, the bottlenecks associated with that........
An Intel chip on its own is not PC architecture. A whole PC is Pc Architecture.
Don't think that's too hard to grasp.

Well that's like comparing apples and oranges. Why would you define a PC architecture to a console architecture to a supercomputing architecture? Regardless you are aware that Intel does indeed have a hand in defining the PC's architecture right?

Errm yeah? But you can't single out specific manufacturers, that's all.
 
PC-Engine said:
Yeah and it's based on x86 with a license from Intel... :LOL:

Oh yeah, I forgot, everything is based on the first x86 chip that came out 26 years ago. :rolleyes:

But if you look which company is defining the current x86 path, then it's not Intel, it's AMD.

Fredi
 
McFly said:
PC-Engine said:
Yeah and it's based on x86 with a license from Intel... :LOL:

Oh yeah, I forgot, everything is based on the first x86 chip that came out 26 years ago. :rolleyes:

But if you look which company is defining the current x86 path, then it's not Intel, it's AMD.

Fredi

Every cpu AMD has ever made for the PC is simply an x86 emulator chip that you can buy for a cheaper price than an Intel chip. If that's what you call defining the x86 path then so be it. :LOL:
 
:rolleyes: Good God give us strength...
I see where this is going (the usual PC-Engine versus everyone, cause he's right and everyone else is wrong), and i'm OUT.
 
PC-Engine said:
Every cpu AMD has ever made for the PC is simply an x86 emulator chip that you can buy for a cheaper price than an Intel chip. If that's what you call defining the x86 path then so be it. :LOL:

Intel says its new 64-bit x86 extensions will run the same 64-bit operating systems and almost all the same 64-bit application software as AMD’s 64-bit architecture. AMD says software compatibility should be no surprise, because Intel virtually reverse-engineered its 64-bit extensions from AMD64. An independent analysis by Microprocessor Report indicates both companies are correct. Except for a few minor differences, the two 64-bit architectures are identical.

http://www.mdronline.com/watch/watc...37&on=T&SourceID=00000377000000000000

Fredi
 
london-boy said:
:rolleyes: Good God give us strength...
I see where this is going (the usual PC-Engine versus everyone, cause he's right and everyone else is wrong), and i'm OUT.

Yep and it's the typical lb response...retreat with tail between legs with a comment in the spirit of I give up because I can't get the other party to agree. :rolleyes:

McFly said:
PC-Engine said:
Every cpu AMD has ever made for the PC is simply an x86 emulator chip that you can buy for a cheaper price than an Intel chip. If that's what you call defining the x86 path then so be it. :LOL:

Intel says its new 64-bit x86 extensions will run the same 64-bit operating systems and almost all the same 64-bit application software as AMD’s 64-bit architecture. AMD says software compatibility should be no surprise, because Intel virtually reverse-engineered its 64-bit extensions from AMD64. An independent analysis by Microprocessor Report indicates both companies are correct. Except for a few minor differences, the two 64-bit architectures are identical.

http://www.mdronline.com/watch/watc...37&on=T&SourceID=00000377000000000000

Fredi

Itanium is 64-bit and it came out long before A64. :LOL:
 
PC-Engine said:
Yep and it's the typical lb response...retreat with tail between legs with a comment in the spirit of I give up because I can't get the other party to agree. :rolleyes:

Yeah i retreat cause u're so right that i just can't win! Bye!!
Way to go PCEngine, another thread down the drain.
 
Correct me if I'm wrong, but didn't Intel have a particular 64 bit architecture that it is essentially abandoning due to MS' support of AMD's competing architecture, even if it is based off Intel's x86?
 
Status
Not open for further replies.
Back
Top