PS3 AC/DC Adapter?

All I know is judging by 360, PS3 is gonna use huge amounts of power..

So they better take some measures!
 
According to IBM's own diagrams, Cell itself @ 3.2GHz doesn't actually draw all that much power. Comparatively speaking, of course. How much is needed to run the other components remains to be seen, but it doesn't neccessarily have to be any more than x360's rather liberal peak of a smidgen over 250W.
 
Cell should use at least as much as Xcpu. Actually more, but it's certainly not an advantage.

But the real difference will be GPU. Somehow Xenos uses only 35 watts, whereas a 7800 uses around 110.

I expect Sony to get that down, but not that much..
 
In power circuit of the next generation game machine "PS3"

Technology of American Vicor corporation to adoption (2005/05/16)

Translated:
The right technology and the left actualizes until recently with FPA technology, It is power source module. FPA technology is adopted With 繧・thing, substantial miniaturization was actualized. Copying

As for the truth, something related to power source which is held in the middle of April
From exhibition of Vicor corporation in exhibition.

The SONY computer entertainment next generation game machine "place t Shaun 3 where the same company is advancing development (PS3)" FPA which the American Vicor corporation developed (factorized POWER architecture) adopts technology for the power circuit. As for this, those which from testimony of the group of power source industry authorized personnel become clear. Concretely, you say that "Cell (the cell)" FPA technology is introduced into the power circuit which drives the processor.

However it is not the case that you adopt the power source module "V ・･ IChip where" the Vicor corporation already has started sale, adopts FPA technology. SONY & the group adopted FPA technology to develop the new tip/chip, it seems that loads this onto PS3. Furthermore, SONY and the Vicor corporation on 2004 June 30th, have announced that it concludes V ・･ IChip and the license agreement regarding FPA technology, (news release). Utilizing this contract, it is the pattern which develops the new tip/chip.

FPA technology at one of DC-DC converter technology, is a feature in conversion method of electric power. The high input potential as for the general DC-DC converter, such as 48V and 100V the voltage first such as 7V and 12V (intermediate bus voltage) using trance, it converts, the processor and the like which supplies electric power being soon, it controls the stabilization of voltage. But FPA technology is different. Administering the control of stabilization to high input potential, the processor being soon and using trance it converts to low voltage. If this method is used, not only being able to raise conversion efficiency, in comparison with method until recently, we have assumed that it can actualize also the miniaturization of power source module. According to a certain power source module manufacturer, "as for the power circuit of the Cell processor, supplying the electric current of maximum 100A with 1V is required. Because it corresponds to such difficult request, it seems that adopts FPA technology ", that you say. (Yamashita victory oneself)
Click to expand...
 
"as for the power circuit of the Cell processor, supplying the electric current of maximum 100A with 1V is required. Because it corresponds to such difficult request, it seems that adopts FPA technology "
Click to expand...

100 amps?! :oops: That's a whole lotta current! That's like up there with starter motors to crank your car engine up (albeit, not the same voltage, though). 100 A @ 1 V does work out to 100 W, fwiw. Sounds "low", so maybe the figures given in the quote were more for example than indicative of the real part.

From what little tidbits I could pull out, the excerpt speaks of a dc-dc power converter (high-frequency switching, I'm guessing, and likely some sophisticated d-class/h-class duty cycle-to-load matching design to maximize efficiency, minimize bulk, and minimize waste heat).
 
Last edited by a moderator:
 
randycat99 said:
100 amps?! :oops: That's a whole lotta current! That's like up there with starter motors to crank your car engine up (albeit, not the same voltage, though). 100 A @ 1 V does work out to 100 W, fwiw. Sounds "low", so maybe the figures given in the quote were more for example than indicative of the real part.
Click to expand...

Well, cell at 3.2ghz really shouldn't draw that much power if the SPE wattage numbers that IBM gave are to be believed (2-4 watts a piece) -- if we also believe the ~85 watt for the Xenon then we can conclude that a PPE is probably ~30 watts a piece (probably a little less). 30 watts + 7 x 4 watts is only ~60 watts.

So, 100 amp max on the current Cell revision seems rather high. Similar to the Athlon 64 (their power envelope I don't think is reached even on their fastest CPUs), I don't think the 100watt Cell power envelope will be touched (not even close).

Who knows though.
 
Bobbler said:
Well, cell at 3.2ghz really shouldn't draw that much power if the SPE wattage numbers that IBM gave are to be believed (2-4 watts a piece) -- if we also believe the ~85 watt for the Xenon then we can conclude that a PPE is probably ~30 watts a piece (probably a little less). 30 watts + 7 x 4 watts is only ~60 watts.
Click to expand...

ISSCC presentations tend to put things into a better light than actual production parts. So say 5 Watts peak per SPU. Then another 10 watts or so for the ring/interconnect/misc and another 10 watts or so for the I/O interfaces + 30 watts for the PPE puts it at about 85-90 watts total power for the CELL. Which matches up quite well with a 100A 1V supply.

Aaron Spink
speaking for myself inc.
 
Hey, it's a compelling case, and I would be quite delighted to see PS3 at only 100 W. :D As a hunch, I simply expected it would be closer to X360's 240 W rating. It would really surprise me (not saying it can't be true, of course) if for all that Sony can pack in a box and MS can pack in a box (brute technology and performance-wise), one should come out using less than half the electrical power than the other. Excellent job on the design, if it is true... ;)
 
randycat99 said:
Hey, it's a compelling case, and I would be quite delighted to see PS3 at only 100 W. :D As a hunch, I simply expected it would be closer to X360's 240 W rating. It would really surprise me (not saying it can't be true, of course) if for all that Sony can pack in a box and MS can pack in a box (brute technology and performance-wise), one should come out using less than half the electrical power than the other. Excellent job on the design, if it is true... ;)
Click to expand...

I seriously doubt the PS3 will get below 100W. Realistically you have to budget ~100W for Cell plus XDRAM, another 100W for RSX plus DRAM, another 10W for the HD, and another 10W for the BD drive, for a reasonable floor of 220W.

The issue is that the vendors are going to be pushing frequencies and hence voltages to the max in order to both get a reasonable launch yield and to hit performance targets. Generally with something like a console, the initial design will be focused on the FF corners of the processes so that performance which will be easy to reach later in the console manufacturing life isn't left on the floor at the beginning. It wouldn't supprise me if the next manufacturing revs of both the PS3 and the X360 drop power budgets on the order of 30-45%, or ~100W. Both MS and Sony will be pushing to get into 65 nM both to increase manufacturing yields, lower cost, and decrease power.

Aaron Spink
speaking for myself inc.
 
randycat99 said:
Hey, it's a compelling case, and I would be quite delighted to see PS3 at only 100 W. :D As a hunch, I simply expected it would be closer to X360's 240 W rating. It would really surprise me (not saying it can't be true, of course) if for all that Sony can pack in a box and MS can pack in a box (brute technology and performance-wise), one should come out using less than half the electrical power than the other. Excellent job on the design, if it is true... ;)
Click to expand...


It should be far above the X360's rating.

Unless they seriously cut down the GPU..

Of course, the 35 watts for Xenos doesn't include the 256 of high speed RAM on a graphics card. Then again I would guess that to be only 10 or 20 watts.
 
You keep repeating what PS3/Cell "should" be, but you don't add any reasons why.

That's because you don't have any, and you're just speculating out of your nether regions, isn't that so, Bill?
 
holy moly I hope PS3 doesn't have a huge external power plant reactor brick like X360. I was hoping that the apparent large size of PS3 meant the power supply and everything would be in the console.
 
Belmontvedere said:
holy moly I hope PS3 doesn't have a huge external power plant reactor brick like X360. I was hoping that the apparent large size of PS3 meant the power supply and everything would be in the console.
Click to expand...

PS3 looks to be the same size as Xbox360 so the P/S will likely be external. Whether it will be the same size or smaller than Xbox360's P/S is anyone's guess.
 
Why is 360 PSU so hugh? Could it have been smaller? Or it had to be that big with the amount of power 360 guzzles?
 
You must log in or register to reply here.

Similar threads

B
Microsoft is reportedly subsidizing $100-200 per Xbox while Sony started breaking even more than a year ago, help me understand
2 3
Replies
54
Views
6K
Nesh
N
Rys
Reopening Architecture and Products + rules and guidelines changes
3 4 5
Replies
90
Views
15K
Shifty Geezer
Shifty Geezer
E
Assassins Creed 4: Black Flag (PS3, X360, PS4, XBOne)
11 12 13
Replies
244
Views
35K
benroveno
B
A
Dynamic framebuffer resolution
2 3
Replies
42
Views
16K
chris1515
C
Cyan
Wholly unique/special ways to play in the next gen! Surprises inside
2
Replies
30
Views
8K
pjbliverpool
P
Back
Top