Facts you should know about PS3

  • Thread starter Thread starter tha_con
  • 82 comments
  • 3,695 views
Jedi2016
Just because a better GPU comes along doesn't mean it's going to make PC's the equal of the consoles. That's never happened, if you actually look at the last few generations of consoles. PC architecture is just too different. It always takes them several years to catch up to the level of console games, and the PCs that do it are statistically far superior to their console counterparts.

Case in point:

Build a computer to the following specifications:

300MHz CPU
32MB RAM
4MB graphics card running at approx 147MHz.

Then I want you to run Gran Turismo 4 on it at 60fps.

As for comparisons for next-gen.. it'll work the same way. PC's may be "more powerful" fairly quickly, but it will take them no small amount of time to catch up to what the consoles are doing. As for the dev kits, don't think for a minute those things are PC's. They may look like PCs, but they ain't. They sure as hell aren't running Windows, that's for sure. Their operation is far closer to that of a console than a PC. Which means they're FAR more powerful than any "comparable" PC.

What are you a cave man? Plenty of Consumer PCs run 3+Ghz

Also, Alienware is a "consumer PC"

http://www.go-l.com/#
5Ghz available to any "consumer"
 
*sigh*

I never said PCs don't run at 3.2GHz. I said that a 3.2GHz PC is NOT capable of pushing a PS3 game, AS A PC. By the time PC's start running games equivelant to next-gen consoles, they'll be many times as "powerful" as those consoles.

Did you miss the GT4 thing? Those "computer" specs I listed are for a PS2, feel free to look them up. A system like that can push GT4. But no PC with those specs could even dream of it.

Comparisons between consoles and PCs of equal power is utterly worthless.
 
If you think about it a pc has alot more to do than just run games, it has all sorts of operations that take cpu time and RAM, but the console just has to concentrate on the running of a bit of software.

Sure eventually pcs will have the same power but the console will have smoother game, one of the things pc games dont have to console is the fact that everyone using a pc has different spcs and what not, but evryone has the same console.

but the thing pcs have over consoles is the development time, a console comes out once every five years, but pcs have hardware development every six months and is on going. so things like normal mapping would have come out on pc first if the ps3 was later or earlier, it just happens that they are coming out at the right time which is good.

Both consols will be very good in fact i would go as far to say that the Game revolution will also be good, games will look awsome even if they dont take all the power the consols have to offer, and im sure pc's will eventually catch up but for once the consolse have a better hand to play on the pc 's.

As for consumer pc's, the one the public buy in bulk (asuming these are the ones you talk about - Alienware, Dell, PC world ect.) these pc's have come a long way in the last few years due to price drops in technology and these will grow in power and drop in price.
 
cobragt
"Xbox 360 has 1MB of L2 cache, and they mention this as a strong point
because PS3 has only 512KB of L2 cache. However, they forget to mention
that each SPE (8 of them) is loaded with 256KB of it's own cache, and
if you count them all together, including the L2, that's 3.5MB of total
processor cache. Go figure."

That is not how cache works...it's like you saying adding all the cores on the 360= 9+ghz since they all run at 3.2ghz. You cannot add them up they will or diffrently. Only 1 core is used for gaming.
hold on
you don't add like that...

the the PPC core on the cell has cache
and each SPE has its scratch ram... (not cache)
 
simontemplar09
Cobragt is right each SPE has it's own local memory that operates a little differently then conventional cache.

for a good explanation on it:
http://www.blachford.info/computer/Cells/Cell1.html

The cache is discussed at the bottom of that page. The rest of the article is a very good write up on the cell architecture, and it's possibilities.

That article is full of errors and IMO is not a viable source of information.

However, to clarify, I need to correct both myself and cobra.

The 256KB of memory on each SPE is essentially a cache, however it has characteristics that seperate it from a conventional cache.

L1 and L2 cache are highly automated memory systems, which makes them simple to program for.

The only difference between L1 and L2 cache and the memory found on each SPE is that developers can dictate the way they wish the memory to be used on each SPE, increasing efficiency, however it operates much in the same manner as a cache would.

The report has many MANY errors in it, leading me to believe either it was written before the spec's of the Cell were released, or they are just using guess work.

For instance he states that the Cell has no cache, and only individual memory for each "APU" (even though they are referred to as SPE's).

He then states that each has it's own 128KB memory, when in reality it is 256KB.

Each SPE has 256KB of on-die memory allotted only to it, but instead of being used as conventional cache memory, this small area of high-speed storage is actually addressed almost like typical system RAM. The L1 cache memory found in conventional processors is highly automated, making it simple to program for but adding overhead. With the Cell processor, programmers can dictate exactly how they wish their software to use the 256KB allotment available to each SPE. This allows execution efficiency to be increased with good software design. It also allows better memory management and security from buffer overflows and other exploits.

Individual SPEs should be able to pass data to each other by storing it in specific areas of the system memory, forming a chain of processing units each performing a different operation on the data. Obviously this requires an extremely fast interface between the system RAM and the SPEs, which the Cell has in spades… More on this later.

Read that. It states that each SPE has 256KB of on-die memory. But it's not USED as "conventional cache", as it's much more flexible, however it serves the same purpose as a cache, considering it is possible for each SPE to communicate with eachother. The way it does this is by passing information along via the main L2 cache. Thus, since all of the memory works together, you can consider it a "total cache" since each SPE can pass along information to be used by eachother via system memory.

:)
 
Yes, the article was original written from the initial patent files. This accounts for the change in acronyms and increase in memory size, as clarified in the articles updates section. I do believe it to be agreeable with what you have said on how it address this memory, and had hoped it would help clear up the cache discussion.
 
Kudos guys 👍 you really know your games.

I just wanted to add that PS3 is being presented as an “Entertainment Supercomputer” by Sony’s Ken Kuratagi and they expect a lot more from the machine (you would hardly see the word console in any sony press release) than playing games. They mention that people will be able to play games AND view HD movies at the same time or may be edit video/pictures. That’s why it’ll have two HD ports plus one conventional video output. My point is that maybe, they’re not expecting developers to use all 7 or 8 SPEs right away and that while the PS3 maybe more powerful than the Xbox360 not all its capabilities will be used by games. I think that the PS3 will be bundled with a whole bunch of utility software including video/photo editing/printing. They’ve mentioned that Cell supports several OS at the same time too. As developers learn to get more and more from the multiple core architecture less will be available for other process and then we’ll have the PS4. 💡

Just my two cents
 
dolande
Kudos guys 👍 you really know your games.

I just wanted to add that PS3 is being presented as an “Entertainment Supercomputer” by Sony’s Ken Kuratagi and they expect a lot more from the machine (you would hardly see the word console in any sony press release) than playing games. They mention that people will be able to play games AND view HD movies at the same time or may be edit video/pictures. That’s why it’ll have two HD ports plus one conventional video output. My point is that maybe, they’re not expecting developers to use all 7 or 8 SPEs right away and that while the PS3 maybe more powerful than the Xbox360 not all its capabilities will be used by games. I think that the PS3 will be bundled with a whole bunch of utility software including video/photo editing/printing. They’ve mentioned that Cell supports several OS at the same time too. As developers learn to get more and more from the multiple core architecture less will be available for other process and then we’ll have the PS4. 💡

Just my two cents

That was one of the most insightful, easy to read posts in a few pages. Well said.
 
I still think it's hilarious that in the same breath the Sony camp can point out how lower tech specs can still yield an excellent machine, but then turn around and tout all the tech benefits of their hardware (this is not to say the MS camp doesn't do close to the same but inverse).

I have a feeling this will turn out similar to the PS2 and PS1... Sony claims the sky as the limit, turns out to be nowhere near that in real life, then slowly they find tricks to squeeze every bit out of the system resulting in some pretty amazing feats...
 
Devedander
I still think it's hilarious that in the same breath the Sony camp can point out how lower tech specs can still yield an excellent machine, but then turn around and tout all the tech benefits of their hardware (this is not to say the MS camp doesn't do close to the same but inverse).

I have a feeling this will turn out similar to the PS2 and PS1... Sony claims the sky as the limit, turns out to be nowhere near that in real life, then slowly they find tricks to squeeze every bit out of the system resulting in some pretty amazing feats...

I think it's even more hilarious when people don't understand what they are looking at then make foolish assumptions...as does 99% of the media. Sony rarely promises anything, they give specs, tech demo's and then let the videos speak for themselves, the meida is responbile for ALL hype, and all let downs, yet you don't hold them nearly as accoutable.

Don't you think it's understandable that a company that DEVELOPED the hardware would be able to use it better than the companies who are using it for the first time to produce games? Don't be clouded in the darkenss, think about what you say and apply common sense.
 
Devedander
I have a feeling this will turn out similar to the PS2 and PS1... Sony claims the sky as the limit, turns out to be nowhere near that in real life, then slowly they find tricks to squeeze every bit out of the system resulting in some pretty amazing feats...

They say Cell is 10 times more powerful than a Pentium 4 3.2ghz. If the true number was 2 and not ten that is still amazing that this is in a console.
 
Here's a couple of points I wanted to mention after reading over this thread, to see what other people thought. First, regarding the cache (or lack thereof) of the PS3. From what I understand, tha_con is correct, the 256k high speed local memory of each cell is like a programmer controlled cache. There is no cache hardware to keep it synchronized with main memory, it's a scratchpad that the programmer controls. It's better than regular cache.

Regarding the cells themselves, they are not really what we commonly think of as general purpose microprocessors. The main PPC processor is, and it is very similar to the 3 PPC cores that are in the X-Box 360. The cells themselves are more like vector or signal processing units, special purpose processors that can be a little tricky to program for, or at least to get maximal performance out of.

This design is part of a larger trend resulting from current processor design hitting a performance wall. Todays processors have many, many functional elements on them, and a great heap of circuitry designed to make sure all those functional elements are used. Your linear sequence of instructions won't get processed one at a time, or even necesarily in the same order you programmed them in. But there is only so much the processor itself can do to figure out how break a single thread down into something like a parallel problem. And doing so takes a lot of transistors.

Future processors will leave the parallelization (is that a word?) to the pros, i.e. the human programmers themselves, or at the very least the compiler, which can take a long time looking over the whole program and deciding what can be done in parallel. The cell processor is such a beast. It is trickier to program for such a processor, but theoretically you can wring far more performance from the same number of transistors, especially when dealing with inherantly parallel problems like rendering a frame.

Anyway, I just wanted to point out that saying the PS3 has 8 processors and the X-Box 360 has 3 is not totally correct as I understand it. The PS3 has one processor and eight (one being redundant, not for calculating redundancies, but for if one of the other seven goes kaput, as I understand it) vector processing units while the X-Box has 3 general purpose processors. Apples and oranges, really.

But the PS3 is still gonna kick the X-Box inna rear end according to the stats I've seen. It will take some time for programmers to really learn how to unlock it's power, but when they do, watch out!
 
Lets talk about sound proscessing, This will be handled by the Cell which is very amazing. Giving an insane number or audio channels and eliminating the need for a dsp since the cell is doing it. I'm guessing the number sound channels are limited to ram.
 
LaBounti
Lets talk about sound proscessing, This will be handled by the Cell which is very amazing. Giving an insane number or audio channels and eliminating the need for a dsp since the cell is doing it. I'm guessing the number sound channels are limited to ram.

Depends on how you refer to "sound channels". The system itself will only export 5.1 to begin with, although that may be expandable to 7.1 or higher since it seems to be entirely software-driven.

If you're referring to individual sounds in-game, I think the number is virtually limitless. Harrison described the possibility of having hundreds of thousands of individual sounds playing at the same time (although this number would probably be significantly less in a game environment that's doing more than blowing leaves around). But, considering that even big Hollywood pictures rarely have more than a couple dozen soundtracks going at a time, this shouldn't be an issue.

I imagine the sound will be no more or less than the developers put in. If it's not quite immersive enough, it'll be the developers who are to blame, not the system.
 
cobragt
Didn't the leaves in the leaves demo have individual sounds? there were more than 1000 sounds on screen!

I think some of them did. His exact quote was:

"In fact, the Cell has so much power, that every one of the leaves could have it's own audio channel, if you wanted to hear them rustle."

Then he pauses for a moment, so the audience can hear the leaves rustle.
 
Oh, yes, definitely.

It would also allow for background noise (which is noticably lacking even from GT4), as well as other incidental noises, like squeaky suspension, and tire squeal from individual tires. All in discrete Dolby Digital 5.1. :)

I really hope Polyphony uses the opportunity to revamp their sound system. GT5 is going to look almost photo-real... I want it to sound real, too.
 
I know were not suppose to go of on a GT5 tangent in this thread but after reading all this I wonder how long PD have had the PS3 Dev tools for? Longer than most other developers I guess.
 
Possibly, but still not very long. Hideo Kojima of Metal Gear Solid fame said he's only recently recieved a proper dev kit for the PS3.
 
Back