Why choose cloth physics among all physics, is the 2x figure based on any tests?
It is only one example, rhetorical as an example as Shifty picked up from the context (i.e. fill in the blank, and example for those without imaginations and with some generic performance difference based on some of the "2x faster" rhetoric). Btw, before the 360 launched cloth physics was something quite a few posters harped on with the 360 and the NBA 2K media was deemed fake by some and yet 2K was able to deliver cloth physics on 10 players at a time. So it went from theoretically far beyind Cell to achievable in a real game. Hence all the, "It isn't possible" talk should be more tempered imo until we see some clear examples deployed and then developers demonstrate an inability to emulate such.
Anyhow, Shifty's response is more of what you are after:
Rigid body physics is more apparent and it affects gameplay.
http://www.gametrailers.com/player.php?id=17744&type=wmv&pl=game
It was just a rhetorical point. Cloth dynamics isn't unique to PS3, though it may be better. We can say the same of most effects. Thus, perhaps PS3 can handle 1000 rigid bodies versus XB360's 250. Is that going to make a game totally impossible on XB360, or will it be possible in downgraded form? How about if PS3 can handle 1000 rigid bodies and XB360 can only handle 10? Will that enable unique experiences on PS3? For sure.
Only if it's impossible will new gameplay be the reserve of PS3, and that seems improbable. The only other aspect is better, rather than unique. Will PS3 games look better? Perhaps. But if the XB360 doesn't look bad, and the difference isn't enormous, the benefit of PS3's SPE's will go missed by most.
That Sony hating USA has bought 40 million PS2s, 40 million PS1s, and 10 million PSPs. Wish I got that sort of hate!
I'd say it's a $500 pricetag hating US, rather than Sony hating.
one said:
As for game play, Little Big Planet is one of the examples of physics-based game play where all user-created objects have physical properties.
Are you suggesting that the 360 or PC could not create a game where user created objects cannot have physical properties?
one said:
Which port do you mean exactly? If you meant DMC4 that'd be a surprise for me.
Not DMC4 at all. I was more thinking of the history of gaming in general where we see new games end up on older platforms.
When I said 'gameplay not possible on the 360,' it wasn't meant in the absolute sense like, "360 is incapable of physics," but it was in the sense that 360 is not capable of physics-based operations to the extent that a game premised entirely on this attribute would reach the same level of achievement on the 360 as the PS3.
...
When I say that this sort of game would not be possible on the 360, it is not to say that 360 could not also run a game based exclusively on fluid dynamics, but it *is* to say that one will so vastly outclass the other if that were to be the primary computational operation, that one might seem a generation removed.
"Gameplay not possible on the 360" at face value does appear to be a constrast from your explaination, so thanks for elaborating. Level of achievement (fidelity) is different from gameplay. I would agree that Cell could and should surpass Xenon in fidelity in some scenarios (give-take overall as systems IMO). I was
only disagreeing that the SPEs can create new gameplay that is impossible on the 360 which was seemed to be implied. Thanks for the clarification
You do note that very few games, namely non-standard games, take the route of using a design that a single computational program dominates the design and process cycles. This ventures close to the concept of tech demo imo, although tech demos can be fun and enjoyable. And we would agree that a larger/robust game has a lot to balance and as developers have noted that any single gain can be offset by other losses--and sometimes our opinions on how that comes out can be skewed (e.g. Xenon has been vindicated to a minor degree to being better in the end of things in some games than had been hinted at by some).
It will be curious to see if the results of any such games/demos do appear a generational leap ahead. It would be even more interesting if we saw a similar project on other platforms take a similar approach of single operation emphasis and leverage other systems and stream processors. Physics isn't isolated to CPUs these days
Carl said:
For example, word out of High Moon after the 'Cell Jam Session' in Carlsbad is that they are working on some edgier EDI games centered on taking advanatage of Cell; one mentioned was a game revolving entirely around fluid dynamics.
Are they stating this isn't possible on other platforms from a core gameplay perspective? Obviously we would have to wait to see the end results, and it should be impressive. I wonder how their research could port to other streaming processors.
Carl said:
LBP is a larger-scale, less extreme demonstration of this... but it is a demonstration all the same of the level to which PS3 is already a step (or two) ahead of 360 on something such as physics operations.
Really? How so? Where do you get the quantitative measurement for such? I ask because this is a pretty bold claim.
FWIW, Forza Motorsport 2 has 8 cars with 9,000 car parts simulated 360 times per second for accurate, real world physics that interact with eachother. At face value one could argue that sort of number and detail is substantially ahead of other PR numbers... but looking at actually released games one could start making counter arguements that based on released software doing certain types of physics the 360 is a step or two ahead of the PS3. Something more certain than a specific game using a specific technique not yet shown seems often sparse ground to being making claims (else we will begin seeing posters claiming Motorstorm style destruction cannot be done on the 360, Forza style simulation cannot be done on the PS3, BFBC style world destruction cannot be done...) So narrowing the sort of physics operations in question and a benchmark other than a game demo versus the lack of such a game demo as a means to say one is a couple steps ahead would seem less... arbitrary?
Btw, and we both know this, but physics is broader than flops and many
robust past solutions haven't mapped near linearly to multiprocessing (emphasis on robust and past... e.g. GPUs are killer at certain physics tasks but suck at others).
Carl said:
And we needn't even look to unreleased/announced titles to see the spread - a game such as Motorstorm incorporates a level of physics-simulation as yet unseen on any 360 game to my knowledge. This is partly due to the optimization that Havok has done for the Cell and the SPEs... but that again itself is a product of the SPEs having a much higher ceiling than the XeCPU in that regard.
Game design anyone? Quick, name a car game on the 360 that has the primary goal of vehicle destruction that wasn't a PS2 port...
I think your logic is extremely flawed here Carl. Because developers haven't gotten around to a specific game design at this point doesn't mean it is impossible or even significantly hampered as you indicate.
At this point all I can do is refer, tongue firmly in cheek, a game such as Forza Motorsport 2 incorporates a level of physics-simulation as yet unseen on any PS3 game to my knowledge. This is partly due to the tools on the 360... but that again itself is a product of the 360 having a much higher ceiling than the PS3 in that regard.
As for Havok, maybe developers are talking to you on this, but I wasn't aware that that they had released details on Havok versus platforms, but namely the results over the past iterations on the PS3 as well as an unamed processor. Do those improvements somehow dictate that a custom, system specific physics solution (like FM2) cannot be leveraging their system in a way middleware would not?
Though I agree the PS3 could well show as much as an order of magnitude better physics, I don't know that LBP is a good example. I'm sure LBP's physics can be managed on XB360. At least what's shown so far. Perhaps the scalability will allow huge towers of moveable blocks and stuff which XB360 couldn't handle, but the level shown to date wasn't hugely taxing that I saw.
That is what I thought. The 360 has a number of games that show physics interaction and techniques not shown currently on the PS3 (and this from even some Xbox ports), it doesn't mean it is impossible. I actually found the collision detection on LBP to be less than acceptible considering the detail.
Are the reasons for greater physics showings on PS3 (if there are) because it alone can do physics well, or because other developers on XB360 haven't considered it worthwhile?
I agree, if we are going to begin using techniques demonstrated in a game but lacking in the others current portfolio, something I warned about over a year ago (!), there is no end to this sort of debate and it is based on arbitrary assumptions.
I have no problem with being put on record for saying this:
The PS3 will outsell the other consoles, it will be tough but it will get there, Europe and Japan is in the bag, Sony "hating" USA will be tough but that is mostly question of price.
There will be PS3 games that will outshine the games from the Competition, thinking that the consoles are "about equal" in power is just plain wrong. Will it be every title? no, as it can be clearly seen from these forums, some developers just can´t be bothered or just isn´t up to learning something else than their primary platform.
1. Japan is in the bag? I thought Wii was showing quite strongly there.
2. The majority thought Sony wouldn't have issues selling their first 6M units in the US and Japan. Japan, after all, was Sony territory that they had OWNED for over 10 years. And yet sticker shock and current game portfolio, as well as competition, made a certainty (6M sales easily) much more difficult than imagined. And that in the face of strong brand name and loyality and talk of "Japan being Sony territory".
3. The US doesn't hate Sony. What a silly comment. Have you even been paying attention to PS2 or PSP sales? Or the last 10 years... sheeeeesh!!
4. Of course the PS3 will have games that outshine the competition--and by a significant marging! The problem is you are one tracked and don't seem to conceed the reverse ever happening...
5. The
consoles (PS3/360) being "about equal" in power being wrong... actually I have heard of scenarios where the 360 a lot faster.
You do realize they are systems and that power is a derivative of developer skill and resources, tools across the board and system resource accessibility, and system performance across many metrics and how that dovetails with a specific game design?
