I suppose you have a link to where Bethesda discussed how long they worked on each system and with what kits? Else everything you are saying is pure speculation and spin.
Am. Football games 2008 PS3/X360
- Thread starter Tap In
- Start date
I suppose you have a link to where Bethesda discussed how long they worked on each system and with what kits? Else everything you are saying is pure speculation and spin.
Actually they said both ports took same amount of time, but PS3 was more difficult. Take that as you will.
Mintmaster
Veteran
Software will nearly always be the limiting factor between these two consoles. If you were comparing Wii and PS3 then yeah, hardware is the main factor. Between PS3 and 360, though, it's all about software. A factor of 2 is visible in graphics (be it with RAM or speed), but very hard to identify with the CPU.
Take joker's example of cloth simulation in NBA. If you want to do it right without it looking like a springy mess, you need an implicit solver using conjugate gradient, which generally scales like n^1.5 (n is the total number of vertices). Now, you look along the edge of a player's shorts and see 10 polygon edges on 360, and 13 on PS3. The total # verts in the shorts is then 1.69x greater on PS3. The processing time is 2.2x greater. So assuming Cell can do this algorithm this much faster on top of the extra load of culling for RSX assistance, how much of a difference is that? Better yet, what if 360 had the same number of visible vertices, but had fewer control points and simply interpolated? How visible is this double computation?
GT4 is another example. Simple lighting model, but perfectly tuned. Another dev team does twice the math per pixel but still can't get the cars to look as realistic.
For different games, you cannot tell which hardware is more powerful because it's all about the software. For the same games released at different times, simple changes again can make a huge difference. Enable AF and disable normal mapping (!), and voila! 9/10 people think Oblivion's ground looks better on PS3 because it's more capable or powerful. Oblivion is one of those games that really needed amazing hardware to run well, but it really didn't look much better than other games. They really didn't put much thought into incremental cost/benefit analysis before the 360 release, again proving my point of how judging hardware from game output (and thus vice versa) is silly. Even simultaneous releases are usually done by different dev teams, so again we have little idea of which is more powerful.
That's an assumption that's ignoring everything we know. RSX is based on the same graphics technology we've seen for many years, and Xenos is the radical one. Why should the former be harder to work with? Most of the differences we're seeing are graphically related. Available RAM affecting textures, vertex setup and shading, render BW, etc. Unfamiliarity and "peaking" are not huge issues here.
We're not going to see RSX suddenly become twice as good as Xenos regardless of backface culling. Even for fine-grained frustum culling, quadrupling your packet size will only slightly increase GPU load while making CPU load one quarter. Thus it's fully applicable to the 360.
I maintain that CPU power alone has little effect on the quality of a game if a developer is clever about it (unless we're talking about 5-10x). You can't see what's going on behind the scenes, so if you tailor your load to your system then the perceptible impact is minimal, especially when your graphics load must be the same or less.
Sony will have some fantastic games for the PS3, but they will be great for the same reason they had great games on the PS2. It's all about the software and art.
Mintmaster
Veteran
The differences in Fight Night had nothing to do with optimizing. It had new art. They changed the shaders a lot. They also drastically dumbed down the environment, remember?
It says little about the capabilities of the machines.
Even the main topic of this thread is hardly evidence of one platform's superiority. It EA for god's sakes! Maybe some effort in the PS3 version got it up to 40-50 fps, and they just said "screw it", and "locked 30 fps" is a lot easier than going up to 60 fps. For the 360, it could be 55+ fps most of the time, so it's over 60fps often enough to make it useful. So the real difference could be as little as 20-30%.
Look at the target market. Not only is the PS3 userbase small in NA, but of those that do know of this issue, how many would have bought the game but will avoid it for this reason? Probably next to nobody. IMO, definately a lot fewer than the number of XB360 owners who'd buy it knowing it was 60 fps.
If you guys want to talk about overall execution of Microsoft and Sony, fine, but let's stop this "hardware A is better than B because game X is better than game Y". Nobody on the planet can, with any confidence, judge a console's hardware by solely looking at the games.
Fair point, and a very good post. But do you think it may be easier/more valid to identify the differences between the two system by comparing 1st-party games that use engines built from the ground up, say in two or three years time?
I agree with you. I'm one of those people who believe that you have to have both excellent graphics and excellent gameplay to give gamers the best experience possible. In most cases when a game is introduced to us, through the web or in print, we can't play the game so it better have good graphics to get my attention. First impression is everything.
I can't really see good AI in screenshots. Physics might be there but can I tell if they are good from a screenshot? Even with video it is sometimes hard to tell if AI is good, physics does gets the nod though. If Gears had looked like garbage at E3 2005 or 2006 I doubt many would even cared enough to find out how well game plays. Same thing goes for all of these first and second party PS3 games. If they had looked like much of the 3rd party software being released now I doubt people would be talking about them as much as they are now.
OT:I was kind of puzzled by Microsoft releasing images of the Halo 3 alpha to the public. While it is true that the Halo 3 trailer left a good impression most people have probably forgotten how it looks like and veiw the screens from the alpha and beta of the multiplayer as indicative of what the single player will look like dispite the fact that they aren't finished with it yet. Were they trying to lower expectations?
iceberg187
Regular
Bungie has always been pretty transparent. Frankie is the best community manager I know of in gaming and with Luke there, they have a pretty formidable team at Bungie. Granted, all of the fans don't read the updates, but us hardcore Halo fans do. They threw out the trailer last year, saying that those assets are in-game assets. Now it's up to them to reach that bar in a real world scenario in regards to gameplay.
FYI, I've been told that FIFA is supposed to follow this 60/30 fps situation as well. We'll find out soon enough.
For me to be honest it has been AI, physics, good story,... .
I'm really tired of playing always the same game with the only big improvements are on the graphical side. Maybe i'm becomming to old (24) but thats why i'm really looking forward to games like a heavenly sword (seems very cinematic) or a little big planet.
I even think that buying a console for the graphics seems rather strange. I didn't buy a PS2 because of the graphics for example.
Mintmaster
Veteran
If you look at my post replying to Shifty, then you'd realize my answer is a resounding "hell no".
If a dev studio is acquired to become first party, it's usually because it has very valuable video game franchises. If not, there's not much we can say about them in terms of talent. Either way, there is no guarantee that they will be able to wring more out of a system than anyone else. Their job is to make a game that earns their owner money. Graphics and CPU dependent effects only have a very small role in that job. Art, content, gameplay, marketing, appeal, etc. are all bigger factors.
To suggest first different different first party dev teams for different consoles will put out titles that are exactly equal in all those latter aspects is laughable (not that I'm saying you are suggesting this). There's really no reason, then, to assume graphics and CPU dependent effects will be as good as they can be in first party titles. Even if by some miracle they were, it would still be next to impossible to judge the hardware when they're so close.
The only way to judge hardware is through apples to apples comparisons. That will never happen on a console. The closest you can get is situations like that of this thread, and even then it is far more a statement of overall execution (hardware, tools, dev support, timing, etc.) than just hardware alone.
Mintmaster
Veteran
The thing is that a CPU doesn't make a whole lot of difference here unless its a lot faster (say 5 times or more). Scaling the workload down to fit a half speed CPU makes such a marginal difference.
AI is like 95% software. Play a chess game on a half speed CPU and you barely notice the difference. Do path searching on a half speed CPU and the depth is only slightly shorter. The examples are endless. Use different chess programs, though, and the difference is enormous. Compare heuristic based AI in various scenarios and the relevance of CPU vs. software is even less.
Physics is another thing where the CPU-related quality has minimal impact on user experience. Little Big Planet devs could have the CPU time used by physics and everyone would still be screaming about how the power of PS3 is what makes it possible. Even if the same physics engine is used (e.g. Havok), good physics is about designing the gameplay to use it well, having content that uses it properly, having software that keeps track of enough objects (RAM!), etc.
There's no excuse for crap AI or physics on slower CPUs. My favourite racing AI is for a game called Viper Racing, which I played on an original Pentium. I saw pretty cool physics with Carmageddon on slowass CPUs. It is natural for people to think better AI in more recent games is due to faster processors. However, it is nearly always due to smarter code. I have never seen an algorithm that is unusable for one CPU but usable for a CPU that's twice as fast. You only see this for fixed load, cross-platform programs. That doesn't happen in the console world.
I too agree that gameplay is more important than graphics. However, the parts of a game that would have to altered for a half speed CPU are even less relevent than graphics.
I'm not sure if this is what you're saying, but I don't think it's just a matter of taste. It's a matter of discernability. If you played a game, would you be able to tell whether its physics were implemented as fast and extensive as it could be on the platform? I really doubt it. Certainly less so than with graphics.
MonkeyLicker
Newcomer
I think one of the biggest problems is with people swallowing every piece of PR they're fed, so long as it's positive to their console. One things Sony is good at is hype, so they used that to their advantage. That's why you still have people expecting Cell to make a major difference in pushing the PS3 well past the 360. People tend to believe whatever makes them feel good.
Well considering what we have seen Cell do already id say thats a very high possibility
Not really, you've only got to look at what Cell has added to Heavenly Sword, Uncharted, LAIR and other games to know that Cell will add alot of extra oomph to RSX. Maybe well get a better glimse of Cell+RSX working together next week.
Yeah, you need to look at Heavenly Sword, Uncharted and LAIR.
Not at their screenshots. You talk as if you've played these games for months - as opposed to swallowing PR about them.
There are upcoming games on 360 or multiplatform that are as impressive as the games you listed. And I dare to say we'll get a better glimpse on Xenon+Xenos working together next week.
I think you guys are still missing Mint's point. Uncharted, Lair, HS, which ever games you want to pick, whether you want to look at screenshots, gameplay video, or the actual games, impressed or not, it doesn't say anything about the hardware relative to the 360 because their performance difference is too small. Even 2x more CPU power will only allow for very small perceived improvements as he has thoroughly explained.
What those titles do speak to, is the talent and ability of the programmers and artists which worked on them. That skill is a greater differentiator than the hardware could ever be. It's extremely difficult to do apples to apples comparisons because the developer/publisher goals, budgets, timelines, tools, analysis, problems encountered, etc. all vary between projects.
What those titles do speak to, is the talent and ability of the programmers and artists which worked on them. That skill is a greater differentiator than the hardware could ever be. It's extremely difficult to do apples to apples comparisons because the developer/publisher goals, budgets, timelines, tools, analysis, problems encountered, etc. all vary between projects.
And that would be? I cannot see anything that comes even close to the games he mentioned.
