What is the true spec of PSX3 CELL???

..

PSP and HandheldEngine design references are valid because they serve as an indicator of SCEI's design and fabbing capability. And SCEI is not doing something extraordinary here...
 
Re: ..

DeadmeatGA said:
PSP and HandheldEngine design references are valid because they serve as an indicator of SCEI's design and fabbing capability. And SCEI is not doing something extraordinary here...

They have nothing to do. On the surface, Cell is being designed by IBM with Sony and Toshiba supporting, as opposed to PSP.

And this is totally forgetting that the PSP design literally blows everything else out there out of the water.
 
Just look at "16 pixel engine" GS that gets blown away by 4 pixel pipe GeForce2Ultra...
GF2s take more time to render just shadow volumes then GS for entire scene with all the effects in our game.
If that's the example of "blowing" away something in fillrate performance, then N64 blows XBox away too.
 
And if that was your insinuation then I apologize for misreading it, but your comment has some ambiguity and I'm forced to bring in flavors from other threads. ^_^;; )

yup.

As for the .1 micron maybe they were misquoted, but I remember reading that in sciam mag, and several other sources...

Because Sony couldn't fit more. Why only 12 MB eDRAM on PSP, a device that predates PSX3 only by a year? Because Sony couldn't fit more.

:rolleyes:
 
zidane1strife said:
Because Sony couldn't fit more. Why only 12 MB eDRAM on PSP, a device that predates PSX3 only by a year? Because Sony couldn't fit more.
:rolleyes:

Come to think of it. How many other 3D processors use ANY eDRAM? Who else is offering a SoC on par with PSP in it's entirety?
 
Vince said:
zidane1strife said:
Because Sony couldn't fit more. Why only 12 MB eDRAM on PSP, a device that predates PSX3 only by a year? Because Sony couldn't fit more.
:rolleyes:

Come to think of it. How many other 3D processors use ANY eDRAM? Who else is offering a SoC on par with PSP in it's entirety?

flaunting edram isn't something i'd be doing . Yes 12 megs in a handheld on a superfast connection to the cpu (since its part of the cpu) is better than external ram of the same amount of more. Of course in a desktop (for lack of a better word ) console . I'd go with tons of cheaper external ram like ddr.

a cell chip with 32 - 64 megs of ram on die will be very expensive. Which is why i'm doubting more htan 256 megs system ram in the ps3. Which would be more than enough . where as an xbox 2 would most likely have more than 512 megs. Simply because it needs more of it to make up for the speed issue.
 
where as an xbox 2 would most likely have more than 512 megs.

More than 512mb? Not a chance, too much money.

I would say expect 256mb main ram for PS3, 16-32mb eDRAM for Cell than 64mb for Rasterizer.

Xbox 2 with 512mb of unified.
 
Come to think of it. How many other 3D processors use ANY eDRAM? Who else is offering a SoC on par with PSP in it's entirety?


too bad about Rendition-Micron V4400E. :( that Verite would have probably rocked socks with its 12 MB eDRAM back in 1999.
 
I thought I remember reading that each Processing Element, made up of one PowerPC core and 8 APUs, would have 16 MB eDRAM.

16-32 MB eDRAM for the whole PS3 CPU seems a bit low, but perhaps they'll have to scale it back to that amount if we see PS3 in 2005.



you know, if PS3 gets delayed until 2007, they could use the .045 process and put 128 MB eDRAM on the CPU, another 128 MB eDRAM on the GPU plus have 1 GB of Yellowstone/XDR external ram. hehe. in my dreams. 8)
 
I thought I remember reading that each Processing Element, made up of one PowerPC core and 8 APUs, would have 16 MB eDRAM.

Oh I know, for 64mb total for Cell(BE) to share. However, everyone says 64mb will be too much so I just say 16mb.

Moral of the story: Always give in to peer pressure.
 
Well, have two CELL cores on a chip and you are already approaching 200 mm2. Does Sony intend to release the first gen PSX3 units as a multichip system then integrate later?? Because it is clearly unfeasible to pack all four cores into single die on 0.09 micron process.

Why not multichip system ? This thing isn't a PSP.

Sony wanted the Playstation to be home server since PS2.

Anyway, here is the spec from that old patent.

Broadband Engine (contains)
- 4 Processor Elements
- 64 MB eDRAM
- 1024 bit bus

Processor Element (contains)
- 1 Processor Unit (Hinted to be PowerPC base, or MIPS like all the other PS)
- DMAC
- 8 Attached Processor Units (APU)

Each APU (contains)
- 128 kb local memory (preferably SRAM)
- 256 bit bus
- 4 floating point units (32 GFLOPS)
- 4 integer units (4 GOPS)
- 128 128 bit registers

GPU
- 4 Visualisers
- 64 MB eDRAM
- 1024 bit bus

Visualiser
- 1 Processor Unit
- 4 APUs
- Pixel Engine (speed & configuration unspecified)
- Image cache (size & speed unspecified)
- CRTC
 
Paul said:
where as an xbox 2 would most likely have more than 512 megs.

More than 512mb? Not a chance, too much money.

I would say expect 256mb main ram for PS3, 16-32mb eDRAM for Cell than 64mb for Rasterizer.

Xbox 2 with 512mb of unified.

Last month i got 512 megs of ddr 3700 for 30$ usd. So i really odn't see a problem for them putting in more ram.
 
Last month i got 512 megs of ddr 3700 for 30$ usd. So i really odn't see a problem for them putting in more ram.

:LOL:

Sorry but if they put such slow ram like that into Xbox2 I'll laugh, DDR 3700 for a 2005 device?

Going by this logic Microsoft would have had no problems putting in 256mb of ram into Xbox, you know because it was so cheap.

MS WAS going to put 128mb in Xbox, they had a plan for it; but why didn't they? Cost.
 
Some thoughts arbitrarily popping into my mind while browsing through this thread.

1.) Comparison PS2 VU - PS3 FP APU: I don't expect these to differ as much in complexity as e.g. Deadmeat implies (for whatever reasons he has). First of, while the PS3 incarnation of a vector unit will feature more memory, logic for e.g. PS2's coprocessor mode can be ommitted. Also i'd imagine that quite a lot of what is a VIF responsibillity today might very well be relocated at the associated PE Core in PS3 (That's what i understand as their prime responisibility (naivly said)-- serving as a communication endpoint for software cells, "dismantling them", i.e. unpacking of data & code, performing various binder/loader functions for its apus, initiate code execution, etc.,etc.

2.)Two-chip solution: I am a bit sceptical of the two chip version that many are expecting. It would make quite a bit of sense to me to go with a single chip implementation where two PEs would feature special APUs for Set-up, Rasterization, Shading, etc, while the remaining two PEs handle the application model. Such an approach would have the added benefit of being quite flexible with regards to geometry transformation / procedural texturing, memory use, general load balancing, general post-processing of the frame buffer as on-chip bandwidth will be lots larger then in a more traditional seperated implementation. It Imo also sounds more feasible in regards to the traditional price point of console systems and supports the references to a SOC implementation in all official press releases (that i have read).

3.)developement costs: The 400 mio $ figure, while a lot of money, is not an exceedingly large amount of money, when compared to other high-performance, broad-scale, paradigm-introducing processor developement efforts. E.g. figures thrown around for the developement(ISA, Compiler(s) & Merced implementation (but nothing production related)) of IA-64 between Intel and HP range between high single digit - low dual digit billion $ sums. Nvidia once revealed to EETimes that the nv20->nv25 developement (along with masking costs though) costs where in the range of 170 mio $ (i am to lazy to dig up the links now, but browsing through the 3d technology forum, you'll probably find lots of references to this).

4.)4 GHz: The alledged 4 GHz design goal seem quite ambitous and a bit surprising to me. Attaining such short cycle times implies (IMO) quite lengthy pipelines for all of Cell's ALUs (even if fabbed at the 65nm node), which in return implies more complex control. To maintain decent IPC (and therefore decent performance relative to cells' theoretical specs, assuming a respectivly distributed application) ,i'd imagine, complex flow control logic (branch prediction, ooe-exec., etc.) might be required for general purpose "spaghetti code", though I expect them to just swallow these inefficiencies, as, if they manage to sustain a five-percent real world efficiency (50 Gops), it'll be quite powerfull for a 2005-2007 cpu.

I might be a little bit more sceptical about Cell then some others in this forum, but nevertheless i am really looking forward to see official info on its first implementations (and having it show up besides my tv-set someday ;) ...).
 
Nvidia once revealed to EETimes that the nv20->nv25 developement (along with masking costs though) costs where in the range of 170 mio $ (i am to lazy to dig up the links now, but browsing through the 3d technology forum, you'll probably find lots of references to this).

The NV30 generation, even prior to NV30's release (meaning it probably doesn't include the further developents such as NV35, 36 and 38) was quoted at being $400M.
 
XB2 will have 32mb embedded vram on 2x R5X VPU ea, 512mb ddr3 main + sound ram and 64mb of ddram! total 640mb! :LOL:

hush hush! 8)
 
Pinky,

The funding in the next three years given to SCE for R&D on the Cell project is going to be around $4 Billions ( of course this is going to include fab costs for part of Nagasaki #2 and part of Oita #2 ).
 
Re: ...

Vince said:
DeadmeatGA said:
If you forks really believe Sony can do things no others can, just look at GS and PSP.

Ok, lets.

Why only 4 MB eDRAM on original GS? Because Sony couldn't fit more.

Which is 4 more than any other graphic IC - and this was done on a 250nm process.

Why only 12 MB eDRAM on PSP, a device that predates PSX3 only by a year? Because Sony couldn't fit more

Lets forget that PSP is a portable device with power, thermal and size requirements.

Deadmeat, twicely 0WN3D... good job Vince :)
 
Paul said:
Last month i got 512 megs of ddr 3700 for 30$ usd. So i really odn't see a problem for them putting in more ram.

:LOL:

Sorry but if they put such slow ram like that into Xbox2 I'll laugh, DDR 3700 for a 2005 device?

Going by this logic Microsoft would have had no problems putting in 256mb of ram into Xbox, you know because it was so cheap.

MS WAS going to put 128mb in Xbox, they had a plan for it; but why didn't they? Cost.
Heh. I forgot who i was dealing with. Listen . I used that as an example . In 2005or 6 i will be able to get whatever the fastest ram is for around the same price if not cheap. If you haven't noticed every year ram gets faster and cheaper .

Btw sony wanted 8 megs of on die in the ps2 but that didn't happen . You know why ? they had to many problems with that much ram in the chip and it cost to much. Wonder if that wont happen with ps3
 
Quote Panajev
The funding in the next three years given to SCE for R&D on the Cell project is going to be around $4 Billions ( of course this is going to include fab costs for part of Nagasaki #2 and part of Oita #2 ).

AMD totaled its investments for bringing its 300mm fab in dresden online at approximatly 2.5 billion dollars. The quoted 400 mio figure is for isa / chip design developement as are all comparision figures.



[url]http://www.amd.com/us-en/Corporate/AboutAMD/0,,51_52_502_509,00.html
[/url]
 
Back
Top