*Game Development Issues*

[T.B.]

What makes it about personal preference is your suggestion that leading on the PS3 is the only or best way to implement this and if the games don't push the platforms quite as hard then it's still all to the good because it's only multiplatform after all. I don't see it that way and I don't think it's healthy for the industry to cement the 3rd parties in such a subservient role.

Well, what you have here is a difference between theory and practice. In theory, I can have the same development process leading on the 360. To be honest, I would prefere that, as the built environment is just plain nicer. In practice, people will fall back to the old model, given the opportunity and an insane deadline. That is just the way programmers are, I guess. Path of least resistance and all that. Before anyone shouts "lazy developers!", please keep in mind that there are junior programmer positions in the industry. Some of them are even open.

On a side-note, one of the strongest argument I can give for leading on PS3 is actually GCC. It's slow, takes up huge amounts of memory and nobody really likes is. *But* it checks much more strictly than CL does. The number of bugs I found simply because GCC would complain about stuff that happily compiled on the PC and 360 is scary. I know of studios that actually GCC compile their Windows-only sourcecode just to get the warnings.

 

[Shifty Geezer]

Well, what you have here is a difference between theory and practice. In theory, I can have the same development process leading on the 360. To be honest, I would prefere that, as the built environment is just plain nicer. In practice, people will fall back to the old model, given the opportunity and an insane deadline. That is just the way programmers are, I guess. Path of least resistance and all that. Before anyone shouts "lazy developers!", please keep in mind that there are junior programmer positions in the industry. Some of them are even open.

It's not just programmers, but people.

I think a fair analogy here is is going metric. Consider a UK engineering firm that's worked in imperial measurements creating effective machinery for decades. A bunch of annoying upstarts start saying "go metric! It's easier and better for everyone in the long run!" The engineers have a look at metric but they've got deadlines to hit, and working out these base ten figures takes a lot longer than using the imperial that are second nature due to experience. But in absolute terms, metric is easier because it follows the same base ten standard that our number system uses and if they learnt metric they would find, once they had learnt it to the same standards as they currently know the imperial system, they would find calculations easier and save time. However, before they master metric, they are faced with the discomfort of an old system they know that works for them, versus a new system that is less effective. If you don't really believe the new system will improve things, there's no incentive to persevere and re-learn a whole load of stuff you have used successfully. Unless there arises a situation where imperial measurements don't work and the engineering firm has to learn and apply metric, chances are they won't adjust.

There are plenty of software companies, not just games companies, who run ineffective operations, who know how to improve things with properly structured operations, testing, feedback, etc., but for whom they're so busy chasing deadlines and fire-fighting, they never get to make the changes that would improve the overall function and efficiency of the company. The fact a company can get by with less than ideal systems doesn't mean it'd be better off without changes.

 

[joker454]

Ehm... It's ok conceptually, but I don't think I know anyone that thinks IBM did a good job with the console PPC cores (Xenos or Cell alike).

Thing is, for the health/success of our studio and everyone else's, these machines really needed to hit $199 as soon as possible so we can start getting good volume on games other than the GTA's and Halo's. An Intel cpu certainly wasn't gonna happen back in 2005 at any reasonable price, and a custom from scratch design would have cost too much. So they grabbed what was really cheap, tweaked it and shoved a few of them in there, namely the PPC's. They have their issues, but $199 pricing is finally just about here so I'm happy with their choice. Plus, I think with 3 of them in there they will have enough juice to go through this generation once stuff is re-written.

About the older programming methods, do you reference to a monolith single-threaded game engine just as an old-skool PC game engines were built?

Partly...but also throw in lots of nested loops, memory access all over the place, tons of branches, load hit store penalties, no vectorization, etc... I remember one of the first systems we got ported to the 360 years ago actualy ran slower than it did on the original Xbox 1 cpu! Fortunately, there were two more cores in there so we were able to port everything and make it all work on time even with that sub optimal code.

Its actually a suprising an early multi-platform games "made by lazy devs" even run on PS3 and delivered some sort of (playable) game experience. I mean as we have understood, Xbox360-copypaste-to-PS3 port should be a failure from the start. Not to mention a time constraints given to the multiplat devs, must have been a miserable long and stressfull days.

Yes.

Do we have any figures supporting this view, or some research? I know muliplatform issues make a lot of noise in the internet, but are sales really impacted by the knowledge one platform is inferior to another?

Parity is the new hotness, for better or worse everyone will have to get used to it. Bottom line, no negative press is always a good thing, and if both versions are identical then people have less bad stuff to talk about.

 

[Shifty Geezer]

Parity is the new hotness, for better or worse everyone will have to get used to it. Bottom line, no negative press is always a good thing, and if both versions are identical then people have less bad stuff to talk about.

That's true enough. I wasn't doubting people were leaning on devs for parity. Just wondering if the marketing guys were frightened of something real or imagined

 

[jandlecack]

@Shifty: Judging by EA's latest fiscal reports and up until Bad Company ports being inferior on the PS3, I think this is not reflecting in sales as much as nitpickers like us would imagine.

 

[nAo]

Which is why I said it the way I did.

The problem is that this is the ONLY thing you did, you made an argument based on personal preferences. How stupid is that?

How well they run on a given platform is a different story, but that's not really the point. Nothing I've seen you or anyone supporting your position say suggests how you proscribed solution (leading on the PS3) will lead to better games, just less controversy.

Nothing you will see or I will say will support my position and change yours, that's for sure.
You have not brought a single rational argument to support your view, I'll stop here. Adieu.

 

[obonicus]

@Shifty: Judging by EA's latest fiscal reports and up until Bad Company ports being inferior on the PS3, I think this is not reflecting in sales as much as nitpickers like us would imagine.

I think there's a threshold. By Eurogamer's comparisons, DMC4 was slightly worse on the PS3 but actually sold slightly better, IIRC. Bad Company has less detailed textures but runs at a similar framerate and resolution. R6V2 was a buggy mess, though.

 

[grandmaster]

Does that DMC4 sale include Japanese sales? As the majority of Western games barely make a dent on the Japanese market, it may well be that DMC4 is the exception rather than the rule. As it is, a mandatory installation and slightly blurrier graphics in fast motion are unlikely to have a massive impact on sales.

On the whole I find that many of the best crossplatform ports are the Japanese ones and I'm guessing that by virtue of 360's complete failure there, sub-optimal PS3 versions simply wouldn't be in any way acceptable. Koei's games may not be to many people's tastes, but they are uncannily similar between PS3 and 360.

 

[betan]

I think there's a threshold. By Eurogamer's comparisons, DMC4 was slightly worse on the PS3 but actually sold slightly better, IIRC. Bad Company has less detailed textures but runs at a similar framerate and resolution. R6V2 was a buggy mess, though.

I don't know how anyone can claim inferior ports sell less.
DMC4 SKUs were practically identical (besides the installation some people mind). On the other hand, PS3 AC was a screen tearing fest, GTAIV had orange tint yet they still sold better (maybe even more than on 360).

As for R6V2's online problems, I don't know, but pretty much all PS3 games launch with online problems, nothing new there. It sometimes takes two patches to get the experience to a reasonable level. CoD4, one of the best selling PS3 games, was similar too. The fact that the very late port R6V didn't sell on PS3, is probably better explanation for the sequel's sales.

People who read those "bad press" don't seem to be a significant portion of potential buyers. The people I play online don't even know b3d, gaf, eurogamer, etc.

As for word of mouth, I doubt hardcore PS3 fans that follow this stuff are eager to educate others.

 

[Shifty Geezer]

So as the thread doesn't get derailed completely, I'm going to put a stop to the 'do inferior ports sell less' discussion. It's inconsequential to the subject of how multiplatform development is handled and the issues developers face, which is the core of this thread's purpose. That particular subtopic is nicely summed up as 'Do inferior ports sell less? Maybe, maybe not, but the Powers That Be want identical games regardless!'

 

[Fafalada]

They have their issues, but $199 pricing is finally just about here so I'm happy with their choice.

We're all aware of economy issues, but other consoles(those I'm familiar with anyway) dealt with them without including CPU cores that are fundamentally broken - as you note yourself a bit further down in your post.
And OT, adding the fact that cost conscious market is pretty much already lost to both MS and Sony for foreseeable future, the economy factor is put into question as well, but that's not something they could have predicted of course.

Parity is the new hotness, for better or worse everyone will have to get used to it.

Perfect parity was 'never' a real issue of contention outside of people that buy multiplatfom releases for all platforms they own anyway, just so they can argue about them later. Software ultimately competes with offerings on platform where it's released, the other comparisons are largely unimportant outside of some major releases where everyone is gunning for pixel counting.
A good multiplatform title is competitive with software on all of supported platforms - and there's been a fair share of PS3 multiplatform releases that simply aren't. Not because 360 version is good, but because they look (and often play) like crap compared to other PS3 titles. And if you think that this kind of games would be better off if you "downgrade" their 360 versions, you're an idiot.

And if we want to take extremeties with "parity" argument - remember the plethora of PS1->DC ports, where DC versions were obviously at least a little enhanced, but they still looked like crap compared to most of DC library, because they were by definition, cheap shovelware.
Wii has got lots of this from PSP(sometimes PS2), PS2 received non-stop stream from DC,PC, XBox, XBox was on receiving end from PC etc.

 

[joker454]

We're all aware of economy issues, but other consoles(those I'm familiar with anyway) dealt with them without including CPU cores that are fundamentally broken - as you note yourself a bit further down in your post.

Does that mean you think the SNES's cpu was state of the art for the time, and that the Saturns cpu's were a dream arrangement? I'd say console history disagrees with you.

Regarding the 3 PPC setup, processing power was not the priority. Getting it cheap, getting fab rights, having it ready before their main competitor, and making it easily accessible to coders and/or pc friendly were the priorities. After that, they just needed it to have enough power to last ~6 years. For that, 3 PPC's will do just fine.

No...that doesn't mean I think they are without issue. But you have to look at the larger picture, and overall I'd say they succeeded. The 3 PPC's are driving some of the best games ever of any console generation, that doesn't sound like fundamentally broken to me. As a bonus, shutting out Intel completely from this generation sent them a strong message. I'll bet Intel will be willing to play ball next time around as a result.

And OT, adding the fact that cost conscious market is pretty much already lost to both MS and Sony for foreseeable future, the economy factor is put into question as well, but that's not something they could have predicted of course.

I couldn't disagree more. Until they hit < $200 this generation will remain elusive to most. You frequently encounter posts of people patiently waiting until these boxes become affordable, and that's even on a hardcore forums like this one. $400 + game + tax is too much. In fact, I'd wager many of the regulars on this forum still don't even own either a 360 or PS3 because they are still waiting for them to drop in price.

A good multiplatform title is competitive with software on all of supported platforms - and there's been a fair share of PS3 multiplatform releases that simply aren't. Not because 360 version is good, but because they look (and often play) like crap compared to other PS3 titles. And if you think that this kind of games would be better off if you "downgrade" their 360 versions, you're an idiot.

Sounds like you misunderstood me. Locking one version at 30fps even though they could have run it at 60fps with just a bit more work is a cost of 'parity'. Stripping down assets on a one version to make sure they fit into memory of the other is a cost of parity. Limiting one version because they had to shift a mass of coders away to the other version to get the darned thing working is a cost of parity. Having one version barely make the machine break a sweat, but holding it back to keep it similar to the other version is a cost of parity. Not being able to leverage that lovely fast ram on one box because you can't get that working on the other is a cost of parity.

It sucks when I know I can get this version to do more, but I'm not allowed because it must match this other version. You and others here know this happens all the time now and on some very high profile titles. That's what I meant with the comment "for better or worse", and not that I was suggesting pulling everything down to be at parity with a bad port. In the end though, take off the engineer hat and put on the business hat, and the 'pros' of parity ultimately outweigh the 'cons'.

 

[Nesh]

Wasnt the Saturn a bit problematic in CPU design and worked better with quads instead of polygons??

 

[Fafalada]

Regarding the 3 PPC setup, processing power was not the priority.

They could have gotten the same realworld utilization out of much lower clocked cores that didn't make some of the stupid concessions in current design(and in turn, better yields/cost efficiency). I'd say that processing power was very much a priority, at least for marketing staff.
I just think concessions made in the past were usually less paper spec oriented. On the positive side, at least we didn't get a clever/neat CPU design bottlenecked to hell with the memory busy like some other systems.

I couldn't disagree more.

Don't get me wrong I agree price point is an issue. I just don't think Wii is in its own little world, it's certainly affecting HD consoles now (some parts of world more then others) and will continue to do so, for better or worse.
But yea, more hw out there is a good thing for SW makers of course.

Sounds like you misunderstood me.

To me it sounds like you've been working on mismanaged projects both on tech and upper management side (if you're referring to real experiences in that post that is). That said, with nAo's words - if you're looking to make games that push specific hardware to its limits, try working for exclusive title developer.

I completely agree good multiplatform is hard, and only made harder by upper management influences, but there's teams that have shown or show that it can be done.
Frankly, having seen some of this gen multiplatform codebases I have little sympathy for people putting the blame on the hw (regardless of the platform). But I'll shut up here before I cross any line on what I can say

 

[ERP]

having seen some of this gen multiplatform codebases I have little sympathy for people putting the blame on the hw (regardless of the platform).

Actually IMO what your seeing has little to do with multiplatform and everything to do with development scale.

It's alway been true that you cannot equate how good a game looks, plays or sells with technical prowess. I stopped doing that 20 years ago when I first got to port product, porting some Amiga games I thought were technically great (before I saw the code) to the megadrive. Let's just say it radically changed my view of how code quality correlates to product quality.

My view on this has always been the same, let the Technical people make technical decisions, I don't think enforcing policy that dictates how a product is developed on a team is ever a very good idea. If the team underperforms, fire them.

There is a huge difference between building an original product from scratch and doing an incremental update to an existing code base, whether you are doing multiplatform development or not. There is a huge difference to starting with a team of 5 and starting with a team of 50, these things massively change how you can approach development at some level far more than whether you need to ship 1 SKU or 2.

 

[Heinrich4]

More information about multiplatform development issues between ps3 and x360:

http://www.computerandvideogames.com/article.php?id=194490

 

[-tkf-]

It sucks when I know I can get this version to do more, but I'm not allowed because it must match this other version.

How far do you have to go, are we talking 30 vs 60 fps with the same assets or is it just low res textures or down sampled movies/music/sounds?

 

[jandlecack]

@joker454: I'd really be interested in whether the version that could push more is on PS3 or 360 most of the time.

 

[holsty101]

You mentioned you've got a dedicated PlayStation team on board. What does this mean for platform-specific features on the console? There aren't many differences. They're basically a one-to-one parody but the only thing we did was add SixAxis support for planes, jet skis and helicopters. ...

Well, I guess that explains a lot ...

 

[Shifty Geezer]

@joker454: I'd really be interested in whether the version that could push more is on PS3 or 360 most of the time.

It'll be 360. Joker's experiences with the two machines have been valuable reading since his arrival on the board!