Does Cell Have Any Other Advantages Over XCPU Other Than FLOPS?

AlphaWolf said:
It's not like we're talking about replacing a future product they would have to R&D, we're talking about replacing a product that is already in production in the millions of units. AFAIK decoders that are in use now don't cost hundreds of dollars so I question how much there is to save.

Exactly and at the end of the day the question is still what benefit to consumers will CELL have inside of a tv vs another chip? Nobody seems to be able to answer this that's why I see it as a solution looking for a problem. SONY is claiming it's a cheaper solution for their CE products which may or may not be true. It could simply be a marketing move to get the CELL name out there. Question is how do I benefit? Why should I care?

Kinda of funny for a guy who is such a huge Nintendo fan, can't seem to see it's Microsoft that is slowly but surely killing Nintendo as a hardware player.

Yeah that's why MS has a handheld called PSP right? :LOL:

MS's target is SONY not Nintendo.
 
AlphaWolf said:
It's not like we're talking about replacing a future product they would have to R&D, we're talking about replacing a product that is already in production in the millions of units. AFAIK decoders that are in use now don't cost hundreds of dollars so I question how much there is to save.

True. I guess I forgot that issue. Well, I suppose we'll save a couple bucks. :D
 
PC-Engine said:
Exactly and at the end of the day the question is still what benefit to consumers will CELL have inside of a tv vs another chip? Nobody seems to be able to answer this that's why I see it as a solution looking for a problem. SONY is claiming it's a cheaper solution for their CE products which may or may not be true. It could simply be a marketing move to get the CELL name out there. Question is how do I benefit? Why should I care?



Yeah that's why MS has a handheld called PSP right? :LOL:

MS's target is SONY not Nintendo.

Sony has already humiliated Nintendo with their console, now they want to do the same with their handheld. :LOL:

For consoles, though, Microsoft is the one killing Nintendo, which used to be the 'secondary' choice if you don't like PlayStation. It seems Microsoft is taking this spot.
 
dodo3 said:
Sony has already humiliated Nintendo with their console, now they want to do the same with their handheld. :LOL:

For consoles, though, Microsoft is the one killing Nintendo, which used to be the 'secondary' choice if you don't like PlayStation. It seems Microsoft is taking this spot.

For a console that came out 18 months before GCN with massive software support, they should be selling more. It's nice to live in a world where PS3 will be released 18 months before Xbox360 and Revolution isn't it? ;)

Too bad they're the one being humiliated by a cheaper inferior handheld that makes money on the actual hardware. :LOL:

MS taking 2nd place doesn't mean they aren't targeting SONY. You usually need to pass 3rd and 2nd to get to 1st.:LOL:
 
In an interview about the Xbox at E3 2002, Dan Vivoli was asked about nVidia's outlook for designing another console processor for a next generation Microsoft system. He answered with the interjection that they'd like to be in multiple next generation consoles in a tone that was conspicuously suggestive rather than just hopeful.
 
Mobile phones are the highest volume market for processors, closing in on a billion devices a year and with a lot of China still left to tap. For the next several years, ARM and SuperH are already positioned to dominate the CPU side, and MBX/SGX have it for the co-processor, projected to power hundreds of millions of units annually (while some people apparently still wait for PowerVR to make a "comeback".)

Other cost sensitive embedded markets like the millions of cars which will eventually ship with info/nav systems installed and the millions of digital TVs and consumer electonics which will need video processing are also already on their way to going to ARM, SuperH, and PowerVR in their latest models.
 
Last edited by a moderator:
Lazy8s said:
Mobile phones are the highest volume market for processors, closing in on a billion devices a year and with a lot of China still left to tap. For the next several years, ARM and SuperH are already positioned to dominate the CPU side, and MBX/SGX have it for the co-processor, projected to power hundreds of millions of units annually (while some apparently still wait for PowerVR to make a "comeback".)

Other cost sensitive embedded markets like the millions of cars which will eventually ship with info/nav systems installed and the millions of digital TVs and consumer electonics which will need video processing are also already on their way to going to ARM, SuperH, and PowerVR in their latest models.

In other words a 1PPE+1SPU CELL doesn't stand a chance in those markets. Heck a 1 PPE+1SPU CELL will have a die area many times that of an ARM11 core.
 
Neither Microsoft nor Sony have had much effect on Nintendo's business. Nintendo still makes their obscene profits in gaming that neither competitor can touch. The two haven't ever been able to reach much of Nintendo's core, sizeable niche.
 
nAo said:
Quote of the year, period.
btw..my friend Elvis wants to meet your pusher.

nAo, then explain to me how an SPE does anything without the EIB hardware or the associated coherence logic to access memory? The answer is it doesn't. Comparing the whole of the Xenos die with its performance vs solely an SPE and its performance is an illogical and inaccurate comparison.

Aaron Spink
speaking for myself inc.
 
The 'why will Cell be used in CE goods' argument has already been had. No need to raise the same old arguments when there's no new worthwhile information to back them up or discredit them.
 
Shifty Geezer said:
The 'why will Cell be used in CE goods' argument has already been had. No need to raise the same old arguments when there's no new worthwhile information to back them up or discredit them.

You're right their is no new information, but the information that we do have support our claim that Sony and Toshiba want to use CELL in CE products. The non-believers are the ones that need to prove us wrong.
 
mckmas8808 said:
You're right their is no new information, but the information that we do have support our claim that Sony and Toshiba want to use CELL in CE products. The non-believers are the ones that need to prove us wrong.

Nonbelievers? Nobody said Toshiba and SONY isn't planning to use CELL in some CE devices. What we're saying is what benefit will the consumer see. Please learn how to read before wasting board space/bandwidth.
 
PC-Engine said:
Nonbelievers? Nobody said Toshiba and SONY isn't planning to use CELL in some CE devices. What we're saying is what benefit will the consumer see. Please learn how to read before wasting board space/bandwidth.

You of all people. Yes that is a great question. But to assume that there will be no benefit what so ever seems a bit naive imo. Looking at the benchmarks of the CELL in the other thread it seems to me that CELL can help HDTVs look better. Perhaps it can make their Bravia line even better than it is.

I wish others hear could make another thread on it and discuss.
 
This discussion of what cell brings to CE goods has already been had, including discussion on unit costs and everything else. These released benchmarks don't change the point's made there - there wasn't any doubt it was a capable chip in that regard anyway.
 
Shifty Geezer said:
there wasn't any doubt it was a capable chip in that regard anyway.

That's what I know. It's just another person wants to know the benefits for using a CELL chip in CE products. I guess that's what we will find out in the future.
 
mckmas8808 said:
That's what I know. It's just another person wants to know the benefits for using a CELL chip in CE products. I guess that's what we will find out in the future.

Wasn't on of the initial goals was to have common code/threads running on a cell networked infrastructure where varying tasks can leverage off your CE products?
 
Unfortunate

aaronspink said:
Apparently you enjoy being a tard.

It is better to make sensible statements when possible my friend than childish comments and insults. It is you who are hurt by such comments..

I was pretty specific. If I'd known you where incapabable of being intelligent I would of exactly delineated my suggestion to the computational complex of the Xenos design, but that should have been pretty f*cking obvious to anyone with half a brain. But I forgot myself and didn't realize I was conversing with the mentally challenged.

Again you resort to childing insults. This is a bad sign of your character and it is wise to change this.

As for Xenos, as I said in previous post, picking out only portion of Xenos means that Xenos is no longer Xenos but new "hack job" chip of your imagination. So it seems you are unsatisfied with Xenos as companion processor of PPE and prefer new chip using select components of Xenos. It is ok to change your mind but it is better to change mind with apology instead of insults.

A single SPE has no method of getting either data or instruction into or out of the SPE. In fact, it has no way of allowing any control what so ever. The SPE relies of a large collection of logic on die in order to move data around. Without this large collection of logic the SPE is useless.

This is very silly and wrong statement. You should not embarrass yourself like this so I ask you to read SPE documentation more closely next time.

You seem to want to make comparisons where you only include what is labelled as an SPE while leaving out the large amount of support logic that is required for the SPE to function.

I have already provided comparison with entire contents of CELL chip including areas labeled mic, xio, flexio, bei, and eib, and all 8 SPE, including "dead" SPE. Even with extra components of entire CELL and "dead" SPE, it is shown that Xenos, what you call more specialized than SPE, is still inferior choice from performance stand point and also from flexibility and capability standpoint. Please read carefully before making foolish statement(s).

We might as well add the 4th SIMD array from the Xenos design then.

Current implementations of CELL is of two versions with 8 active SPE and 7 active SPE therefore I provided numbers for both. However once again you seem unsatisfied with current implementation of Xenos and propose changes to your original proposal and to Xenos. I am curious what name will you give to this new chip of your design. I propose Spinkos. Then everyone will know who performed "hack job" of Xenos and had silly idea of using certain non-independent components of Xenos in non-functioning coprocessor to PPE.

And one again you have chosen to do a flawed comparison. Apparently you brain is incapable of understanding the issues. If you really want to make this comparison in the fashion you are trying, why don't you use as your area estimate 1/2 of the Xenos Die area, because that IS the area for the flops that are in question, that is the computational core. And that computational core also includes 33% redundancy.

So once again it seems you no longer want comparison of Xenos as companion of PPE instead of SPEs but "hack job" version of Xenos without important parts. If you only want computational part of Xenos in comparison with SPE, then we can also remove non-computation parts of SPE such as storage and control logic. This only doubles performance for die-area for both "hack job" SPE and "hack job" Xenos (so SPE still has very large performance advantage) but prevents both from actually functioning so in the end it is meaningless to make such comparisons. You must be consistent with your statements and be careful thinking before making your proposals or sensible conversation is not possible.

It would be a discourse if you could understand what is actually going on.

No. When one participant is repeatedly confused, changing minds about what to compare, does not read carefully before responding, and uses insults when unable to proceed with logic and be consistent with previous statements, then discourse is not possible regardless of how well other participant understands or not.

You make many unfortunate mistakes and it is apparent you are avoiding true reality which shows your proposal of Xenos instead of SPE was not sensible. It is ok to make mistakes, as everyone makes mistakes, but it is wise to admit mistakes than make insulting comments (I notice you have done this to others as well) and change what you said and hope no one notices. With course of action you have chosen it is you who loses because you do not learn and hence fail to grow. Do not let your pride hurt you my friend.
 
Cell

PC-Engine said:
Niche market. ;)

We've been over this dude. CELL is a solution looking for a problem. Why do you NEED CELL in CE equipment??? My tv and DVD player works just fine without CELL as do my cellphone and PC..:LOL:


My friend, if CELL is nothing but PS3 chip it is very successful design. But because of high floating point processing density, it is enabler of new applications, including niche applications for which no such chip existed in the past.
 
At 65nm and 45nm, a 1:8 Cell won't be all that large or power-hungry. The chip will scale just fine, and as long as they are making PS3s, there will be the redundancy factor built in. So that 1:1 Cell may have the same die size as a 1:8, but at 45nm or 65nm, it's not a big deal. And with only a PPE and SPE active, the power drain will be small. PEACE.
 
Back
Top