Starbreeze take on the Ps3 vs Xbox 360 (the Darkness Int)

[scooby_dooby]

The reception the chip has had, my knowledge of it, the comments of various developers and the fact that it's going to be in a closed box that will have a lot of developers working on it for 5+ years.

To turn the tables, what makes you think otherwise?.

The fact that multi-threaded game engines are still taking their baby steps, and CELL is a very complex design. And, sony's last console, after 6 years, was still nowhere near it's peak efficiency (wasn't one of the VU's hardly used?)

I dunno, I just see no evidence to believe that cell will have a high in-game efficiency. There's is just nothing to support that. Same goes for Xenon, but at least Xenon is a traditional symmetrical approach.

 

[Titanio]

The fact that multi-threaded game engines are still taking their baby steps, and CELL is a very complex design. And, sony's last console, after 6 years, was still nowhere near it's peak efficiency (wasn't one of the VU's hardly used?)

I dunno, I just see no evidence to believe that cell will have a high in-game efficiency. There's is just nothing to support that. Same goes for Xenon, but at least Xenon is a traditional symmetrical approach.

Meh, symmetry and potential in-game efficiency and performance have nothing to do with each other. It might make Xenon slightly easier to work with, but it needn't stand in the way of Cell's real-world performance at all.

PS2's "unused" VU - where do you see Cell's equivalent?

 

[Edge]

The fact that multi-threaded game engines are still taking their baby steps, and CELL is a very complex design. And, sony's last console, after 6 years, was still nowhere near it's peak efficiency (wasn't one of the VU's hardly used?)

That's true, hardly anyone used the second VU, but CELL is much easier to approach than the EE.

 

[scooby_dooby]

PS2's "unused" VU - where do you see Cell's equivalent?

I don't. It just pokes a large hole in your theory that just because something is closed box it will have a high realworld efficiency.

 

[Titanio]

I don't. It just pokes a large hole in your theory that just because something is closed box it will have a high realworld efficiency.

'Just'? I mentioned a number of factors, not 'just' that.

 

[scooby_dooby]

'Just'? I mentioned a number of factors, not 'just' that.

Well that was your strongest point IMO, the rest are all very 'iffy'

 

[predicate]

The fact that multi-threaded game engines are still taking their baby steps, and CELL is a very complex design. And, sony's last console, after 6 years, was still nowhere near it's peak efficiency (wasn't one of the VU's hardly used?)

I dunno, I just see no evidence to believe that cell will have a high in-game efficiency. There's is just nothing to support that. Same goes for Xenon, but at least Xenon is a traditional symmetrical approach.

Cell/EE comparisons are a little off. EE was a MIPS core with what, four or five coprocessors that were all different. Cell is a Power core with 7 other identical cores.

 

[SugarCoat]

That's more than long enough, especially since some 360 games at launch was bragged about, like Activision's Call of Duty 2 to be using three cores. Well, what about their PC game!!!

Have you taken a look at surveys lately? Out of gamers, about 5% are actually using Dual Core computers. Now you go ahead and write a letter to EA about how they should waste, excuse me i mean spend, thousands and thousands of dollars not to mention extra time to write the game with optimization code for dual core processors. *note, Xbox360 games do not easily port to PC and vice versa. They KNOW what Xbox360 owners have, and they'll use it. And they know what most computer owners have too. 1 Year is nothing when it comes to the full exploitation of cutting edge hardware adoption. Do they offer advantages? Yes they do thanks to drivers and the few games that use both cores. Does that mean that game developers should of had time machines 2, 3, 5 years ago (when they could of done something) to see the outcome/timeline of dual core processors in computers? No.

 

[Panajev2001a]

I don't. It just pokes a large hole in your theory that just because something is closed box it will have a high realworld efficiency.

He never said that, you are kind of twisting his words else I could stick PanaMPU(c) inside a closed box and make it render Toy Story in real time while it simulates the delicate stirring of Cappuccino.


P.S.: PanaMPU does not exist .

 

[scooby_dooby]

He never said that, you are kind of twisting his words else I could stick PanaMPU(c) inside a closed box and make it render Toy Story in real time while it simulates the delicate stirring of Cappuccino.


P.S.: PanaMPU does not exist .

No. The argument is you could throw a PanaMPU in a closed box and get extremely high efficiency.

 

[Edge]

Have you taken a look at surveys lately? Out of gamers, about 5% are actually using Dual Core computers. ...

Sounds to me then, they need to use all 360 cores to just match their PC version, since the PC version did not even need dual-cores to run just fine.

 

[overclocked]

I always get the feeling that many tends to critize people that think based on what they belive that Cell is the better Cpu. IMO this isnt something that the person should have to defend, its an opininion and fully entitled to it. Likewise with the xeCpu of course.

But somehow i cant help to notice that people are downplaying Cell while keeping full faith in xeCpu.
Now for me that is strange as we ultimately want better games.

Cell is/was a costly long project, it certainly has it flaws but if it turns out to be that good that atleast some developers have expressed i dont see the reason for this constant degrading of it and for myself i do belive that it will turn out good.

I think many on this boards with sceptism should look at the positive things instead of
being constant on the offense/defense when someone just express an opinion.

Btw i think i am the biggest AMD fan-boy on this forums!

 

[assen]

Sounds to me then, they need to use all 360 cores to just match their PC version, since the PC version did not even need dual-cores to run just fine.

If this is ever the case, it wouldn't be because the PC CPU has more FLOPS like the 360 cores, like the Cell has, but because the PC CPU has branch prediction, out-of-order execution and other goodies, which the Cell also lacks.

 

[nelg]

P.S.: PanaMPU does not exist .

Don't sell yourself short. It does exist. It runs on ATP and its performance is measured in postsynaptic potentials per second.

 

[Edge]

If this is ever the case, it wouldn't be because the PC CPU has more FLOPS like the 360 cores, like the Cell has, but because the PC CPU has branch prediction, out-of-order execution and other goodies, which the Cell also lacks.

Well what does that say for the 360 CPU! We been told many times that Xenon was a very powerful general purpose core, but I guess not?

 

[Asher]

Programming for any multi-core architecture when it comes to games is difficult, as look at the PC world were we have had dual-cores for a long time now, and do next to nothing for the frame-rate of games being released today. Today PC dual cores are more akin to Xenon, than CELL, so think about that.

Nice thead by the way, it's flogging a dead horse with all the same things being said before. I'm beginning to wonder if anything new can be said here about the different architectures.

It's easier/cheaper to get more efficiency out of Xenon than Cell.

For the developers who really take their time with optimization and solid design, Cell will outperform Xenon in many situations. But in the majority of cases where the budget and time constraints don't allow for it, performance will be either comparable or borderline in Xenon's favor, depending on the code...

 

[PSman]

he said theres a "slight" GPU advantage, but did not not say slight CPU advantage. Does this mean that there's a bigger difference between the the systems CPU's then its GPU's? .

 

[patsu]

For the developers who really take their time with optimization and solid design, Cell will outperform Xenon in many situations...

Then is it fair to assume that game companies and Sony will derive their own frameworks, engines, design patterns, middleware libraries, ... to enhance or even repeat these situations ? Or is the run-time so erratic that every situation is a one-off adventure ? I would assume it's the former right ?

To take things in positive light, we will see more great things coming from Cell, yes ?

 

[dukmahsik]

he said theres a "slight" GPU advantage, but did not not say slight CPU advantage. Does this mean that there's a bigger difference between the the systems CPU's then its GPU's? .

just like xbox and ps2?

 

[Titanio]

Then is it fair to assume that game companies and Sony will derive their own frameworks, engines, design patterns, middleware libraries, ...

Third party middleware will also optimise, and even the devs who really don't take their time with optimisation and solid design will benefit from that.

just like xbox and ps2?

No, I think you'll find there was a massive gulf in GPUs between those two systems, and a significant effective gulf between the CPUs. But it doesn't surprise me that you would again try to draw the parallel.