Official PS3 Thread

Status
Not open for further replies.
Wether the patent has anything to do with PS3 is an open issue ... but if you read the article I linked and still maintain the patent defines Cell you need to work on your deductive reasoning.

Cell's high level architecture was designed together with IBM (well it would probably be more accurate to say "largely by IBM") long after that patent was filed.
 
Than what is the patent MfA? What does that architecture describe? It's a 1tflop total piece as well as a Rasterizer that has 4 pixel engines.

Suzuoki, Masakazu supposidly was the one who filed this patent. He was involved in the PS2, and now something is filed that has 4 pixel engines, image caches, and a 1tflop cpu is filed under his name. It also mentions Dram amoung other things. Hmm what can it be? I wonder..

If the patent isn't ps3, then tell me what is it?

ps3.jpg


What is that? It has all the ingrediants of the ps3, if it's not ps3 then by all means tell me what it is.


Sony said that ps3's chip would be able to do 1tflop, and here it is, a patent filed by someone 100% certain is heavily involved with the ps3 that describes a system with 1tflop power.


Cell supposidly does 1tflop correct? And this patent is it. Or maybe a modified one to work inside of a game console, but the power is still there, which is 1tflop.[/img]
 
I think Sony expected PS2 to be 1000X PS1, but ended up to be only 300X. I wont be surprise if PS3 will not hit 1Tflops/1000X PS2. Its all their words and their XXXX. Anything is possible. :D
 
"I think Sony expected PS2 to be 1000X PS1"

And what is this based off of? Nothing, it's what you "think" which doesn't mean much.
 
Yes, you cannot take Sony serious with EVERYTHING they say. Although this time the 1000X actually has weight.. you see we have the patent.
 
Paul said:
How can people still say that PS3 won't hit 1tflop when Sony said it, then it was later proved in the patent? Please don't give me this 500gflops BS jvd, take a look at the patent some time eh?

And Cell WILL be used in PS3, are we forgetting what Okamoto said at gdc 2002? Sony wants 1000X, and it's not going to happen with a 500gflops CPU, stop getting yourself excited with nonsense that ps3's cpu won't hit 1tflop jvd. As there is no evidence to support your claim, only evidence to deny it.
Click to expand...

We have a patent not chips. We have a patent for a chip designed with a certian target for its mhz. Untill they produce it giving 1 tflop performance all we have is hopes. Trust me i'd love to see a 1tflop chip. It can happen but it has just as much of a chance not happening too. Also a cell will be used in ps3 there are many diffrent types of cell chips. thats the purpose of the cell chip. To scale. Scaling to meet demands and price points and what the micron process can give us.
 
And from the patent we have a version of the Cell which gives 1tflop. Sony said ps3 will be able to do 1tflop, the patent proves it, so for now, PS3 can do 1tflop.
 
Paul said:
And from the patent we have a version of the Cell which gives 1tflop. Sony said ps3 will be able to do 1tflop, the patent proves it, so for now, PS3 can do 1tflop.
Click to expand...

Can u give me a link to where sony said the ps3 will be able to do 1tflop please ? I'm not doubting u i just don't remember ever reading it .
 
How does the patent prove the Playstation 3 can do 1 TFLOPS? I'm a tad bit curious how you come to this conclusion. I don't recall ever seeing a CELL chip demonstrated to the public that runs a 1 TFLOPS. Just because something is in a patent doesn't prove what it can do. As soon as we see working silicon then that will be our proof. The patent only gives the owners rights to the ideas and no one else being able to copy those ideas. We can't go by everything said in this patent, as I'm sure the actual PS3 will differ a little or by a lot. This patent doesn't set everything in stone.
 
The patent proves that the CELL will run at 1tflop, just like everyone working on cell says cell will do 1tflop. And Cell is going to be inside ps3, so as of now ps3 has a 1tflop cpu. As demonstrated in the patent.
 
Paul said:
The patent proves that the CELL run at 1tflop, just like everyone working on cell says cell will do 1tflop. And Cell is going to be inside ps3, so as of now ps3 has a 1tflop cpu. As demonstrated in the patent.
Click to expand...

Cell is just a name for the tech. Just like hammer. We all know there are diffrent versions of hammer , desktop version , server version , versions with more on die ram , ect... So sony has said cell is capable of 1 tflop , they said ps3 will use the cell chip. They never said ps3 will be 1 tflop of performance . For cost reasons they may have to use a lesser version of the cell chip. Don't put words in sonys mouth u will get burned
 
Explain the patent to me jvd? Why does it show an architecture of 1tflop? And it's common knowledge this is ps3, the patent.
 
Oh and, Sony has said NOTHING about lower version of Cell have they? No, as of now it's just one chip, something that does 1tflop.
 
All stuffs are just speculations from people reading too much into the patent documents, no one has officially announced anything other then marketing materials.

But it can be your belief that the patent will be realized, but it is simply not a proof unless PS3 is released or the PS3 CPU is successfully fabricated and doing 1TFlops.
 
Paul said:
Oh and, Sony has said NOTHING about lower version of Cell have they? No, as of now it's just one chip, something that does 1tflop.
Click to expand...

Actually ibm has talked about other versions of the chip. Its a scalable chip thats the major thing about it .
 
"Actually ibm has talked about other versions of the chip. Its a scalable chip thats the major thing about it"

Yes, as far as I know different versions for doing different things. Like one that would be in a server would be different than one that would go into lets say a TV. However, I haven't heard anything different power wise.

Before this thread goes nuts, let's agree on one thing. The patent picture above is ps3, and THAT particular architecture calls for 1024GFLOPS. Yes, something can happen and sony may not be able to do it, but that's a slim chance happening. The architecture in the patent, whatever it is, call's for 1tflop im sure we can both agree on that.

Cell is going to be something that will be hard to do, but I think it can be done.
 
The point is that Cell is an architecture who's high level design was hammered out with IBM long after the patent's filing date (if you read the article and deny this there is no point in any further discussion). What the patent is, is inconsequential in relationship to Cell.

Im guessing it is part landgrab, part aborted project. A patent does not have to describe workable technology.

Those are rather poor links to show Sony said Cell would do a TFLOP/s. "Accepted" facts are copy pasted all the time, right or wrong. Of course there are pages where Kutaragi actually is actually quoted as saying Cell will do 1 TFLOP/s ... so it doesnt matter much, although I still say he was full of it at the time. Id like to see if someone can get him to repeat it now, now there actually is a real design.
 
What is that? It has all the ingrediants of the ps3, if it's not ps3 then by all means tell me what it is.
Click to expand...

It is probably a significantly weaker version of what will be in ps3, a way to throw off competitors into making h/w to compete with a far weaker version of what will really be out there... no good player shows his true hand before playing it...
 
Status
Not open for further replies.

Similar threads

M
Playstation portable gaming device at PS5 Pro price point
Replies
13
Views
1K
Proelite
P
V
Requirements to declare a game fully Path-Traced
2
Replies
26
Views
2K
jlippo
J
Shifty Geezer
  • Locked
Digital Foundry Article Technical Discussion [2024]
169 170 171
Replies
3K
Views
302K
BRiT
BRiT
B
The Lack Of Interactivity In Games Is A Big Problem
2
Replies
37
Views
2K
Boss
B
R
State of GDeflate *spawn
2
Replies
25
Views
2K
Remij
R
Back
Top