Our goal with the Cell is to be an order of magnitude faster

V3 said:
Sure but it's still not tiny...150^2mm for one PE on 90nm process is not tiny.

Its about the size of 180nm Flipper at 120 mm2. But what you expect, the PE has 9 cores, one of them 64 bit Power.

I wouldn't call the APUs cores. It's more like SIMD units. Anyway Flipper is a GPU, sound chip, North bridge, and contains 3MB of eDRAM.
 
Re: Our goal with the Cell is to be an order of magnitude fa

"Our goal with the Cell is to be an order of magnitude faster," said Lisa Su, an I.B.M. executive in charge of technology development and licenses.
10x faster than what? They segued from Sony's remark to a cut-short IBM quote. Conveniently structured news article, I say.

Cell will not be an order of magnitude faster than anything, because it's a template. Templates are not faster or slower than other templates because the blanks have not been filled in.
 
Anyway Flipper is a GPU, sound chip, North bridge, and contains 3MB of eDRAM.

That Cell supposedly contained the northbridge too, and probably at least 1 MB of SRAM in there. :)
 
If the APUs are cores then why would you also need a PPC core?

So IBM can have their Power PR in Cell :LOL:

Its written in the contract, Sony and Toshiba gets a cheaper deal, if Power name is in there somewhere :LOL:
 
one said:
The Cell has some components that in the lab switch at 5.6 GHz,

Well, I missed too many things from the article when I first read it in the other thread as the article didn't sound very trustable then.

Yeah, I'm wondering what that's referring to. We already have a clockspeed and SRAM speed, though there's been a lot of speculation about what the 4.6Ghz clockspeed actually refers to.
 
PC-Engine said:
Titanio said:
PC-Engine said:
The 4.6GHz is for the SRAM.

I thought that was 4.8Ghz? 4.8Ghz SRAM and 4.6Ghz "clockspeed"?

Sorry I meant 4.8GHz. I don't really know where the 4.6GHz figure came from...

It was on one of the ISSCC preview slides. No one seems to be quite sure what it was referring to, the 5.6Ghz info makes things even more obscure ;) I guess we'll find out in the next couple of days..
 
lol @ CVG:

Designed for high-end graphical workstations, the Cell chip is alleged to feature four power cores, is supposed to run at 4.6GHz and will apparently be capable of 16 trillion (that's a one with a lot of noughts behind it) calculations per second.

http://www.computerandvideogames.com/r/?page=http://www.computerandvideogames.com/news/news_story.php(que)id=114631

;)

Fredi
 
McFly said:
lol @ CVG:

Designed for high-end graphical workstations, the Cell chip is alleged to feature four power cores, is supposed to run at 4.6GHz and will apparently be capable of 16 trillion (that's a one with a lot of noughts behind it) calculations per second.

http://www.computerandvideogames.com/r/?page=http://www.computerandvideogames.com/news/news_story.php(que)id=114631

;)

Fredi

That's where misinformation starts and people start complaining about Sony hype and misrepresentation :( Boooo CVG!
 
PC-Engine said:
Then why not just say they're 9 PPC cores?
Just think of the PU as a big dispatcher/scheduler, that could also deal with others PU (If there's others).
 
Vysez said:
PC-Engine said:
Then why not just say they're 9 PPC cores?
Just think of the PU as a big dispatcher/scheduler, that could also deal with others PU (If there's others).

Ok if we were to compare a APU to a known cpu core, what would it be equivalent to? A i486? Can a APU be used as a standalone cpu without the help of a host processor core? If it can't operate as a standalone microprocessor then it's not really a core, but more like a coprocessor.
 
Back
Top