*Game Tech*
- Thread starter zaftxx
- Start date
- Status
woundingchaney
Regular
The problem with this comparison is that one version they chose to v-cap the other they didnt. If they didnt v-cap the 360 version it could very well have higher avg. fps than the PS3 version. With a suggested difference of .3, Im fairly sure ME would/could have avg. fps be higher on the 360 without a v-cap. Though if we are speaking totally about parity then kudos to Dice and ME.
I completely agree that writing off worse performing cross platform titles on the PS3 due to developer negligence is complete ignorance. It is as if console loyalist believe the sky is the limit with the system and there is no reason for these compromises to be made. Additionally I agree with Joker where we see people brushing off issues with PS3 exclusives and then hanging cross platform titles over the same issues.
Last edited by a moderator:
HT
0.3 frame per second... + granmaster hints at bigger dip in frame rate on the ps3...
I wonder to which extend this values are useful as people reads whatever they want to read in those measurements...
Between I can't stand attacks against our deer devs as they're the one providing useful infos that may not be public otherwise, some people just really have to cool down...
/HT
Between ShootMyMonkey made some really interesting comments in regard to multi platform development in the tomb raider thread.
0.3 frame per second... + granmaster hints at bigger dip in frame rate on the ps3...
I wonder to which extend this values are useful as people reads whatever they want to read in those measurements...
Between I can't stand attacks against our deer devs as they're the one providing useful infos that may not be public otherwise, some people just really have to cool down...
/HT
Between ShootMyMonkey made some really interesting comments in regard to multi platform development in the tomb raider thread.
Last edited by a moderator:
Assuming the reports of sporadic slowdown on 360 and frequent tearing on PS3 are right (i didn't notice much tearing on the PS3 demo), then average fps would 100% surely be higher on the 360 if both were capped. Basically, everytime you see tearing on the PS3, if it were capped the fps would drop to 20 or less. Everytime you see slowdown on the 360, it would tear if not capped.
DICE chose some tearing over slowdown (correctly imho) on the PS3 because it would have been much more frequent than on the 360. Perhaps they optimized it a bit more since the demo build and were able to cap, we'll see.
I wonder if ME2 (if there even is a sequel) will use the Frostbite engine, rather than UE3. BF:BC was essentially at parity.
Um... One is v-synced the other is not. So thats kind of a ridiculous comparison. framerate on the X360 would have been higher than the PS3 if it wasn't vsynced
The 360 is only capped above 30fps, it tears when it (rarely) drops below it and quite noticably, whereas the PS3 seems to tear on the edge of the image so the effects are not nearly so pronounced - I certainly didn't notice any in the demo yet.
Well, it's not my comparison anyway (it's from the link), but ifs and buts are equally ridiculous without considering all the consequences. If the 360 version wasn't vsynced above 30fps then yes it would achieve a higher average framerate, but it would tear above 30fps too, and the 360 version seems to tear in a noticable area of the screen whereas the PS3 version tears at the edge of the image in the overscan area. Why that's the case, I don't know. They've enabled vsync above 30fps on 360 for a reason, and it's not as if a 29.6fps average needs defending. They made the right choice.
But despite all of that, my original point was that both versions of the game are incredibly similar in IQ and performance - which seems to have gone unnoticed in the nitpick-fest that B3D has become.
In the PS3 version I noticed it only at the beginning of the tutorial in the demo I think, in the flythrough, and only for a very brief moment. Perhaps there was more during gameplay but I was too engrossed in it.
Regarding the 360 version, I thought it was truly v-synced at 30, I recall someone mentioning slowdown. Not sure though, since I haven't had the chance to try it yet.
The thing is the "lazy coder" argument only applies to a minority of the PS3 gamers. I don't see why developers need to get worked up over them.
I have my own misgivings about "parity ports" (I think they are red herrings and would rather developers maximize to each consoles' strengths).
What people do want to see is quality work. Mirror's Edge did a splendid job. I think DMC is fine too. GTA4 would have done better if not for their choice of (mandatory) blur filter. BioShock would have done well if not for the low res textures they missed (but the hardware could deliver). EA's 30fps sports games last year were "unfortunate" (when the competition or even EA themselves this year could deliver 60fps).
From the users' point of view, there are certainly room for improvement in some of these cases.
As for "issues" like sub-HD, pre-installs in exclusive games, they had their fair share of complains (e.g., MGS4's installs, Halo 3's resolution). The great titles survived because they typically exceeded the users' expectation in one or two areas. The not-so-good ones (like Haze) will disappear from the scene. So I don't think exclusive games are getting as much free pass as you think.
Those last few posts were not at all the sort of thing we want. Discussions about 'the wider PS3 community' or 'the intarwebs' or various contingents of fanboys are not a valid topic for the console forum. Maybe the RSPCA forum, but this isn't a sociology thread. The topic is mulitplatform development issues, not fanboy attitudes to poor ports, and whatever the rest of the world is saying is of no consequence here.
This glorious technology forum of ours categorically identifies most XB360 games as better. This thread is to discuss what are the reasons for that. TBH I think the thread's run its course. There's nothing really new to add, not many subjects to hit on. PS3 has less RAM available and is harder to work with, and economically offers less returns for the investment. Each new cross-platform title adds nothing to this understanding. I guess it's worth keeping open for any games that reach parity, or favour the PS3, so we can consider the improvements in technique and developer understanding or whatever.
Complaints of 'lazy devs' are, however, incongruous. Although in principle a valid discussion point, it's been done to death and can only irk the long-hour-working devs who create your games. Though I'm loathe to censor such attitudes, they aren't welcome. If you do want to counter an argument for 'lazy devs' instead of ignoring the poster, please stay civil and intelligent. Thanks.
This glorious technology forum of ours categorically identifies most XB360 games as better. This thread is to discuss what are the reasons for that. TBH I think the thread's run its course. There's nothing really new to add, not many subjects to hit on. PS3 has less RAM available and is harder to work with, and economically offers less returns for the investment. Each new cross-platform title adds nothing to this understanding. I guess it's worth keeping open for any games that reach parity, or favour the PS3, so we can consider the improvements in technique and developer understanding or whatever.
Complaints of 'lazy devs' are, however, incongruous. Although in principle a valid discussion point, it's been done to death and can only irk the long-hour-working devs who create your games. Though I'm loathe to censor such attitudes, they aren't welcome. If you do want to counter an argument for 'lazy devs' instead of ignoring the poster, please stay civil and intelligent. Thanks.
It may very well do that. Edram like anything else has finite bandwidth, it's possible to exhaust it. I've already worked on one title that had to use a 1/2 size particle buffer on the 360 version, and Lost Planet, being an orgy of overdraw, possibly had to do the same. Now if a game can exhaust edram, you can imagine the toll it extracts on non-edram memory! To keep the same performance I had to use a 1/16th size particle buffer on the PS3 version of the title I worked on, but I did add an extra post process step on the PS3 version to help reduce particle artifacts. Motorstorm 2 would not need a downsized particle buffer as their transparent overdraw is minimal.
Yes, exactly! If you try implementing a very high vertex game with 4xmsaa and no predicated tiling on 360, you will have a problem. If you try making Grand Theft Post Processing on 360 where you have 25 different post process passes each one needing a resolve, you will have a problem. Both machines have problems, and in some cases very similar problems. But...the automatic "lazy port" claim when for example Fallout 3 on 360 has msaa and the PS3 version doesn't is just silly. The "Lazy Port" flowchart has literally become:
1) Locate game with performance problems and/or design compromises.
2) If there is no 360 equivalent of said game, then goto step 10
3) If performance/design problems are the same as the 360 version, goto step 20.
4) Goto step 30.
10) Great job! All performance consequences are deemed ok and/or not noticeable. END
20) Good job, but clearly the game could have been better if the PS3 was actually used. Maybe the coders won't be quite so lazy next time. END.
30) What a crap port, damn coders must be at the beach drinking mai tai's. END.
You won't notice it on the pic you provided, it's easier to spot on large area particles and while in motion because you will see shimmering. We were arguing just yesterday if the typical user would even notice. Incidentally, I think they made the correct choice to leave them pixelated on Motorstorm 2. 60fps is more important than a particle artifact which many won't spot anyways. Likewise, there's ways to mitigate the artifacts but on a 60fps game it's not worth spending the extra cycles. So yeah they totally made the correct design choice. Of course....if there were a 360 version that didn't exhibit those artifacts, you'd probably see a thread here with 8000% magnified screenshots and people crying foul that the team didn't spend any time "properly optimizing" the PS3 version.
EDIT: Posted this before seeing your comment Shifty. Feel free to use your modhat of +30 pain to delete stuff.
MazingerDUDE
Regular
I spent some time comparing PS3 & 360 version of Fallout 3, and the findings are rather interesting.
Theses are some of the shots I took with digital camera, the actual difference is even bigger. All shots are taken at the exact same time period.
http://www3.telus.net/public/dhwag/FO1360.jpg
http://www3.telus.net/public/dhwag/FO1PS3.jpg
metal parts are glossy on the PS3, it's completely flat on the 360
http://www3.telus.net/public/dhwag/FO2360.jpg
http://www3.telus.net/public/dhwag/FO2PS3.jpg
this one's the much talked 'access panel', unlike what people have suspected, the base textrue is identical. only difference is that the PS3 version's got normal map, the 360 got none looking totally flat.
http://www3.telus.net/public/dhwag/FO3360.jpg
http://www3.telus.net/public/dhwag/FO3PS3.jpg
the ps3 version of roaches look just beautiful with detailed normal map, 360 version is just flat textured. the giant ants & scorpions are similar, the ps3 version being realistic with high-res normal map.
http://www3.telus.net/public/dhwag/FO4360.jpg
http://www3.telus.net/public/dhwag/FO4PS3.jpg
even rocks are different. the 360 version's got normal map, but it's in much lower res. it's just more detailed and sharp on the ps3
http://www3.telus.net/public/dhwag/FO5360.jpg
http://www3.telus.net/public/dhwag/FO5PS3.jpg
the last but not least. the 360 version look terrible with flat low res texture, where the ps3 version is just beautiful blonde hair with so much detail.
Could this be the first multi platform game ever to have better textures on the ps3 version?
It seems like trade off for 4 x AA is rather severe.
As for the frame rate, the 360 is rock solid 30 fps. The PS3 is good at 30, but it just slightly stutters.
When things get heavy, you get screen tearing for the 360 version, where the PS3 remains V-locked with drop in frame rate.
After all, the PS3 version wasn't just another inferior port.
It's just AA vs Better Texture or Screen Tearing vs Frame drop
Each system had to make compromises, probably based on their strength and weakness.
This could very well be a game that's truly handled by 2 seperate teams!! The BS we hear so often as a part of PR
Theses are some of the shots I took with digital camera, the actual difference is even bigger. All shots are taken at the exact same time period.
http://www3.telus.net/public/dhwag/FO1360.jpg
http://www3.telus.net/public/dhwag/FO1PS3.jpg
metal parts are glossy on the PS3, it's completely flat on the 360
http://www3.telus.net/public/dhwag/FO2360.jpg
http://www3.telus.net/public/dhwag/FO2PS3.jpg
this one's the much talked 'access panel', unlike what people have suspected, the base textrue is identical. only difference is that the PS3 version's got normal map, the 360 got none looking totally flat.
http://www3.telus.net/public/dhwag/FO3360.jpg
http://www3.telus.net/public/dhwag/FO3PS3.jpg
the ps3 version of roaches look just beautiful with detailed normal map, 360 version is just flat textured. the giant ants & scorpions are similar, the ps3 version being realistic with high-res normal map.
http://www3.telus.net/public/dhwag/FO4360.jpg
http://www3.telus.net/public/dhwag/FO4PS3.jpg
even rocks are different. the 360 version's got normal map, but it's in much lower res. it's just more detailed and sharp on the ps3
http://www3.telus.net/public/dhwag/FO5360.jpg
http://www3.telus.net/public/dhwag/FO5PS3.jpg
the last but not least. the 360 version look terrible with flat low res texture, where the ps3 version is just beautiful blonde hair with so much detail.
Could this be the first multi platform game ever to have better textures on the ps3 version?
It seems like trade off for 4 x AA is rather severe.
As for the frame rate, the 360 is rock solid 30 fps. The PS3 is good at 30, but it just slightly stutters.
When things get heavy, you get screen tearing for the 360 version, where the PS3 remains V-locked with drop in frame rate.
After all, the PS3 version wasn't just another inferior port.
It's just AA vs Better Texture or Screen Tearing vs Frame drop
Each system had to make compromises, probably based on their strength and weakness.
This could very well be a game that's truly handled by 2 seperate teams!! The BS we hear so often as a part of PR
djskribbles
Legend
Thanks for taking the time to post comparison shots.
This game looks quite different on all platforms. But from what i've seen, the PS3 version isn't so bad after all.
This game looks quite different on all platforms. But from what i've seen, the PS3 version isn't so bad after all.
grandmaster
Veteran
Some excellent work there and a lot of it is tallying with my results too. However, the further you progress into the game, the more you will probably be re-assessing the 'slightly stutters' part of your report.
Betanumerical
Veteran
It would be great to get some proper sized grabs of these areas to really see the difference in its fullest.
Mintmaster
Veteran
It almost looks like there's some sort of shader bug for the specular shine. There's definately some non-diffuse light influence in the 360 ant, but it's behaving incorrectly. I think the hair is a similar issue. You're definately wrong in suggesting there's no normal map in the 360 version, because I can see the lighting on the access panel.
You also forgot to mention the craptacular pixelated shadows in the PS3 verison vs. the decent shadows in 360. Look at that roach comparison.
Here you are right. Not what you'd expect, judging from other games.
You're jumping to conclusions here. Fixing the specularity is not going to impact performance much at all, even if they have to add some math ops. There's no tradeoff there, just a bug.
It seems like devs scrutinize their PS3 graphics more than 360, perhaps because they're worried about backlash. First there was that GTA4 texturing bug and now FO3 has this specularity issue.
This is truly a breakthrough comparison (despite the fact it's comprised of low-quality off-screen camera shots). The conventional hype recently has been how awful and inferior the PS3 version is; the screens somewhat indicate the contrary. More detail in the PS3 version's textures; as well as better (or just there) specularity effects (gloss).
On the other hand, the PS3 version seems to have a complete lack of filtering on its shadows, which is a little strange to me. I don't ever recall seeing shadows that pixellated and unfiltered in a current-gen game before.
I don't think it's necessarily a bug/glitch. The lack of prominent specular reflectivity could very be a deliberate omission in the 360 version for performance considerations. You mention the presence of "some" very faint/soft specularity in the 360 version, but that could just be the way it's supposed to be, as opposed to a glitch or oversight. Some games have used this style of soft "gloss mapping" (instead of regular specular lighting) before, such as DOOM 3 for instance.
Last edited by a moderator:
- Status
