Suzuoki's CELL Patent application revisited....

D

DeadmeatGA

Banned
http://appft1.uspto.gov/netacgi/nph...&s1=20020138637&OS=20020138637&RS=20020138637

[0014] In accordance with this modular structure, the number of PEs employed by a member of the network is based upon the processing power required by that member. For example, a server may employ four PEs, a workstation may employ two PEs and a PDA may employ one PE. The number of APUs of a PE assigned to processing a particular software cell depends upon the complexity and magnitude of the programs and data within the cell.
Click to expand...

[0073] FIG. 7 illustrates a chip package for a BE 702 with two optical interfaces 704 and 706 for providing ultra high speed communications to the other members of network 104 (or other chip packages locally connected). BE 702 can function as, e.g., a server on network 104.

[0074] The chip package of FIG. 8 comprises two PEs 802 and 804 and two VSs 806 and 808. An I/O 810 provides an interface between the chip package and network 104. The output from the chip package is a video signal. This configuration may function as, e.g., a graphics work station.

[0075] FIG. 9 illustrates yet another configuration. This configuration contains one-half of the processing power of the configuration illustrated in FIG. 8. Instead of two PEs, one PE 902 is provided, and instead of two VSs, one VS 904 is provided. I/O 906 has one-half the bandwidth of the I/O illustrated in FIG. 8. Such a processor also may function, however, as a graphics work station.

[0076] A final configuration is shown in FIG. 10. This processor consists of only a single VS 1002 and an I/O 1004. This configuration may function as, e.g., a PDA.
Click to expand...
So let's summarize.

Server : 4 CELL cores + 4 GS3s
Workstation : 2 CELL cores + 2 GS3s
Low-end WorkStation : 1 CELL core + 1 GS3
PDA : 1 GS3

So why were people so excited by the block diagram of a "Server", which Sony has no intention of selling to you???

Sure, that 4 CELL core server chip might do 1 Teraflops in theory, but that's not what Sony plans to stick into your PSX3, Suzuoki says so in his patent application...

Damn, I was right all long...
 
...

Then again, you might get 1 TFLOPS on your PSX3 if you bought 4 of them and clustered them together via optical cable.

But as it stands, PSX3 tops out at 320 GFLOPS(256 GFLOPS from EE3 and 64 GFLOPS from GS3). GS3 will probably clock at half rate because of eDRAM.
 
Patent for <U>"Computer architecture and software cells for broadband networks"</U>, it is <U>NOT</U> the patent for "PS3 platform specification".

:!: From this point on, it's all speculation :!:
 
ChryZ said:
Patent for <U>"Computer architecture and software cells for broadband networks"</U>, it is <U>NOT</U> the patent for "PS3 platform specification".
Click to expand...

Just had a flashback to 1998, sorry.

PS. I thought we had already passed this Cell != PS3 debate and onto Deadmeat's slow, quicksand like trip to self inflicted death - but I guess not. &lt; shrug >.
 
What else... ? Xbox 2 will come out with 2 TFLOPS, developers will jump ship, Sony sucks, Kutaragi is a madman who D00M3D CELL with his ideas, SCE did not do anything with CELL they only begged IBM to license it to them ( both things you said and see how they link well to each other ? ), etc...

:rolleyes:
 
Vince said:
ChryZ said:
Patent for <U>"Computer architecture and software cells for broadband networks"</U>, it is <U>NOT</U> the patent for "PS3 platform specification".
Click to expand...

Just had a flashback to 1998, sorry.

PS. I thought we had already passed this Cell != PS3 debate and onto Deadmeat's slow, quicksand like trip to self inflicted death - but I guess not. &lt; shrug >.
Click to expand...
I never said "Cell != PS3", please read more careful. I just ask for, not to interpret too much into those EXAMPLES illustrated in the patent mentioned above, for your own good .... and don't lump me together with DM.
 
So why were people so excited by the block diagram of a "Server", which Sony has no intention of selling to you???
Click to expand...

PS3 supposedly to become the entertaintment server for your home.
 
Re: ...

DeadmeatGA said:
Then again, you might get 1 TFLOPS on your PSX3 if you bought 4 of them and clustered them together via optical cable.

But as it stands, PSX3 tops out at 320 GFLOPS(256 GFLOPS from EE3 and 64 GFLOPS from GS3). GS3 will probably clock at half rate because of eDRAM.
Click to expand...

3dfx had SLI with Voodoo 2

Sony could have PCC (Playstation Console Cluster) with the Playstation 3 :oops:


Also they could support up to three TV screens at one. :oops:


:p

I don't know about any of you, but if Sony made a totally scaleable console where the graphics got better and more detailed as consoles were linked together, I would have a stack of PS3's outputting into 3 TVs. Could you imagine Grand Turismo?

:devilish:
 
There is a new Kutaragi interview on Watch, I'll let DeatmeatGA post it, he is on a roll. :D
 
I don't know about any of you, but if Sony made a totally scaleable console where the graphics got better and more detailed as consoles were linked together, I would have a stack of PS3's outputting into 3 TVs. Could you imagine Grand Turismo?
Click to expand...

this has been my DREAM for ANY of the next gen consoles. hell, it was my dream for the current consoles. stacking/combining consoles to increase the detail. this would require some extra programming per game, but nothing all that far-fetched, it sort of happens with PC games, although not quite on the level that we're talking about here.

You know how SGI visualization systems. like Onyx2 with RealityEngine or InfiniteReality, you can add more Pipelines and more Raster Manager boards per pipeline to increase performance (vertex rate and fill rate) by at least 16 times if not more. It would be alot like this. just not as modular down to the pipeline level. you would just hook up more PS3s together for combined processing power.

As someone mentioned, Voodoo 2 SLI did this. It could also be done with Konami's Cobra arcade board. upto 5 boards for 5 million polygons/sec
NAOMI was also supposed to have the capability to scale to higher performance by adding more boards. 3Dfx Ramage would have been able to do this in its own way also.

With PS3, there should be 8-16 Processing Elements. (not 4) clocked to at least 4GHz, so that each PS3 has SEVERAL TeraFlops performance. I dont care whether these PEs are all on one die or not (we are still dreaming here) just as long as each $300 console has a few TFLOPs to play with. Then via optical cable, you can combine 8-32 PS3s in a network, and this combined power can be used in realtime to create graphics that are even MORE CGI-like than a single PS3,. we are not talking film-quality CGI but close to what you would see on television. tv show level CGI would be awesome. I would pay a few thousand dollars for an 8-PS3 setup.

Then this would go beyond gaming. the worstateion industry would transistion to PS3, by using several 32-PS3 setups to tackle major computation and rendering problems. PS3 would be used for content creation, CGI film making, military flight simulations. everything. Sony would eat into the market held mostly by SGI and Evans & Sutherland....

arcades to would just use 8-32 PS3s for their games.....

All PS3 consumer software is designed to run on a single PS3 unit, but can run in higher detail, higher resolution, higher framerates (120fps or more) with more PS3s...
 
Vince, read what ChryZ said more carefully.
I realize that some people on this board are taking the patent in question as the de-facto PS3 specification (DM most notably, but you and Pana are right up there too ;) ), but in opinion of some others the patent is just a nice reference/guideline about the architecture ideas and doesn't necesserily have a lot to do with the actual PS3 (IMO, without a doubt a lot less then DM is suggesting in this topic).
Of course, that doesn't make PS3 any less Cell based, just that all the nitpicking out of the small patent details like this topic looks quite pointless in that perspective :p
 
Fafalada said:
Vince, read what ChryZ said more carefully.
I realize that some people on this board are taking the patent in question as the de-facto PS3 specification (DM most notably, but you and Pana are right up there too ;) ), but in opinion of some others the patent is just a nice reference/guideline about the architecture ideas and doesn't necesserily have a lot to do with the actual PS3 (IMO, without a doubt a lot less then DM is suggesting in this topic).
Of course, that doesn't make PS3 any less Cell based, just that all the nitpicking out of the small patent details like this topic looks quite pointless in that perspective :p
Click to expand...
Exactly the point I tried to make. Fafalada, thanks for making it more clear!
 
¿Only 1 PE is a bad choice?

All of us know that the 4 PE option is better. But if we compare 1 PE specs with equal x86 in domestic market you can see that Cell with one PE is more powerful.
 
....

but in opinion of some others the patent is just a nice reference/guideline about the architecture ideas and doesn't necesserily have a lot to do with the actual PS3
Click to expand...
Numbers don't lie; you cannot fit 4 CELL cores into single die at consumer price range. This is why I know EE3 will have at most 2 CELL cores.
 
You must log in or register to reply here.

Similar threads

A
Embedded Memory & Novel Techniques for IQ and Rendering
Replies
3
Views
3K
XpiderMX
X
N
Deja Vu from the past :)
2
Replies
29
Views
5K
function
F
A
IBM Power7 Derivative: A Viable Console CPU?
2 3
Replies
52
Views
24K
Ninjaprime
N
A
Holy Wall of Text: My impressions post Playstation 4 (PS4) unveil
2 3 4
Replies
70
Views
20K
(((interference)))
I
T
  • Locked
New Gamebryo LightSpeed engine for Wii
Replies
3
Views
4K
Shifty Geezer
Shifty Geezer
Back
Top