BEHOLD, A CELL Q&A website!!!!!! (CELL to cost $100?!?!)

KK say on conference: "this is BASIC ps3 specs" :)
do you remember ps2 speedup 200->300 Mhz ?

and RSX(550mhz) and cell(3200mhz) run asynchronous -> false specs

facilities maybe:
500/4000
550/4400
600/4800
 
jvd said:
version said:
KK say on conference: "this is BASIC ps3 specs" :)
do you remember ps2 speedup 200->300 Mhz ?
:rolleyes: okiay then .

jvd, you don´t know either, there´s no need to act like a prick. God, how did you became a mod is one of the biggest misteries around.

As for me, I don´t see a clockspeed bump unreasonable. Anything that will help devs reach Sony´s high bar for next gen is welcome.
 
heh , all i said was okay then. I dropped the arguement because there is no need to continue .
 
Actually if this information is true (TBD) then it does worry me a little about the price of the PS3. If this does turn out to indeed be true then the speculation of a $400 -$500 cost could definately be possible. This with Blue-Ray and other components does seem to point that way :cry:

AS much as I would like to purchase the PS3, I might end up actually waiting for cost to lower (more games would be released as well) to purchase it. That with the possability of $60 games runs that first day buying tab up quick but we will just have to wait and see :rolleyes:
 
JasonLD said:
250 to 300..

Yes and I don't remeber that 250 ever been said to be the version PS2 will get, IIRC Emotion Engine was introduced with that speed, but nothing was said about PS2, just like Cell was introduced at 4.6 and nothing was said about PS3, only this time the speed will go down.
 
jvd said:
This isn't like a server chip. They can't make a 1x8 at 4ghz with really bad yields and use the rest as 3.5 ghz ones and sell them cheaper u know what i mean ?
Sony do potentially have this option though. If they're creating content creation workstations for the movie biz etc., they can set to work manufacturing 4GHz 1:8s, sell those at a premium, and use those that missed the mark (1 SPE knackered and not so highly clockable) as in their console.

Don't know if that would be more economical then two different production runs, one for Workstation Cells and one for PS3 Cells, but I'd have thought so.
 
well shifty i highly doubt they will sell at the volumes sony would need to break even. I agree mabye a few hundred in its first year may go to that , but i doubt more than that and i doubt it would make a diffrence to the bottom line
 
I highly doubt that Sony will raise the clock on CELL. Come on, they're forced to disable one of the 8 SPEs to improve their yields!
 
Yeah but that disabling is presumably due to die defects appearing across the wafer on these gigantic chips.
 
and RSX(550mhz) and cell(3200mhz) run asynchronous -> false specs
facilities maybe:
500/4000
550/4400
600/4800
Why would it matter so much that the GPU and CPU are not at synchronous speeds? The GPU and VRAM are not synchronous either (550 & 700)... for that matter, the GPU and VRAM on PC video cards are not always evenly divisible clocks either. GPU and CPU are not synchronous for X360's layout either.

If anything, the only things that really need to be synchronous would be the main memory and CPU. And when you look at it, 3.2 GHz CPU to 3.2 GT/s RAM... sounds like a good deal there. Also, assuming that 3.2 GT/s means 400 MHz DRAM clock for the XDR, and that the CPU and GPU share this 400 MHz clock signal, the CPU would only be able to step in 400 MHz increments (also assuming no half-integer multipliers).

Besides which, without a corresponding bandwidth increase throughout, I think there's no major practical gain to be seen by increasing the clock speed of CELL from 3.2.
 
seems like the limiting factor is the ppe. that may be why the ppc core on both x360 and ps3 are clocked the same.
 
Laa-Yosh said:
I highly doubt that Sony will raise the clock on CELL. Come on, they're forced to disable one of the 8 SPEs to improve their yields!

I thought the same, though mass production is not due to start for a long time and even 6 months is a long time to get things right.
Not raising my expectations, but it's definately not impossible.
 
Is the 8th spu really disabled, as in sitting there doing absolutely nothing.
In the slides it was marked as "reserved for redundancy functions" or something like that. Which to me sounds like it would have some fixed function, not being programmable.
Has there been any confirmation on the role of the 8th spu?
 
rabidrabbit said:
Is the 8th spu really disabled, as in sitting there doing absolutely nothing.
In the slides it was marked as "reserved for redundancy functions" or something like that. Which to me sounds like it would have some fixed function, not being programmable.
Has there been any confirmation on the role of the 8th spu?
SCE said it's due to beter yield. So yes, it does nothing (or can't do anything, if you get a Cell with 1 defective SPE).
 
rabidrabbit said:
Is the 8th spu really disabled, as in sitting there doing absolutely nothing.
In the slides it was marked as "reserved for redundancy functions" or something like that. Which to me sounds like it would have some fixed function, not being programmable.
Has there been any confirmation on the role of the 8th spu?

for reduce the production cost
 
rabidrabbit said:
Is the 8th spu really disabled, as in sitting there doing absolutely nothing.
In the slides it was marked as "reserved for redundancy functions" or something like that. Which to me sounds like it would have some fixed function, not being programmable.
Has there been any confirmation on the role of the 8th spu?

We talked about this before. If it's deactivated it sits there twisting his thumbs.
 
Back
Top