If this holds true then we can consider a PU as a mulithreaded processor ( 2+8 )ccording to papers to be presented at the ISSCC, the initial Cell chip has a single processing unit that can pass computing tasks out to as many as 8 other processors. Thus, working in tandem, it can process up to 10 sequences of instructions simultaneously.
That's niceSony are due to present the full technical details of the machine in Tokyo next month.
Well..if Xenon is aimed to our living rooms as the PS3 then MS itself is trying to break Intel dominance going with IBM processors. To good to be true..The WSJ outlines the purpose of the Cell processor, mentioning that the aim of the chip is to help break the Microsoft / Intel dominance of computing architectures, and to prevent it from extending into the living room.
nAo said:From The Inquirer who's quoting Wall Streen Journal:
If this holds true then we can consider a PU a mulithreaded processor ( 2+8 )ccording to papers to be presented at the ISSCC, the initial Cell chip has a single processing unit that can pass computing tasks out to as many as 8 other processors. Thus, working in tandem, it can process up to 10 sequences of instructions simultaneously.
IIRC it's not the first time we heard something along the same line.
...
"Sony are due to present the full technical details of the machine in Tokyo next month."
Xenon cores are expected to have powerful SIMD units and I don't think (CELL) PUs will have anything like that.Jaws said:I swear the PUs and Xenons cores are the same!
nAo said:Xenon cores are expected to have powerful SIMD units and I don't think (CELL) PUs will have anything like that.Jaws said:I swear the PUs and Xenons cores are the same!
But well..maybe both of them (Xenon cores and CELL PUs..) are derived from the same base architecture. I bet we'll discover that soon.
One of the CELL leading architects (Dr. Gschwind) has been working on CELL and Xenon
ciao,
Marco
18th of March? Weird date, 1 week after the GDC, 8 weeks before E3. Isn't Sony's fiscal year ending near that date, 31st of March, no? *puzzled*a bird under NDA told me he had huge news and was saying to me that it would point itself out 18 March . the sucker didn't want to tell me anything .. :? like I would tell someone..
google said:Results 1 - 24 of 24 related articles. Search took 0.20 seconds
The really scary part is that STI hasn't really done a lot of serious PR by themselves yet. Oh and as I submit this post, we're about 8 hours from the first official presentation. I suppose we can expect the first 'resonably good' details in less than 12 hours.It became 24 in the last 10 min
nAo said:Xenon cores are expected to have powerful SIMD units and I don't think (CELL) PUs will have anything like that.Jaws said:I swear the PUs and Xenons cores are the same!
But well..maybe both of them (Xenon cores and CELL PUs..) are derived from the same base architecture. I bet we'll discover that soon.
One of the CELL leading architects (Dr. Gschwind) has been working on CELL and Xenon
ciao,
Marco
bad news, hm my brother just found out , they are available for download only AFTER the exhibition . his collegues will return with the book they had with all the details next week, he will copy the relevant pages for me. of maybe i can let him download it .. but then again, the news isn't that HOT anymore
The Cell architecture project has been funded by Sony, Toshiba and IBM.pc999 said:Well Sony had the patent for the Cell (e.g.software package like), but who hold the PUs,PEs and APUs tech is IBM right? So Cell and XeCPU could share a lot of tech, and from there we got the rumors from Cell=XeCPU and a (virtually) unexistent performance gap from a tech point, once I dont see why the network distribuition is need for PS3 (while in others sectors should be great).
In conclusion nothing holds MS and N to have a CPU as powerfull as Cell, RIGHT?
Correct me if I am wrong please.
pc999 said:Why cant MS or N use those S|APUs , if we do not belive in the leaks, these tech are from IBM, Right?