Cross platform development and choice of 'Lead system' *Spinoff*

[ERP]

Port is really a bad word for what actually happens in development of a multiplatform game.
In general both versions are running all the way through development, and there are specialists doing work on both.
What lead SKU usually ammounts to is what the none specialists are running on, and often where new things are tried first.

The latter point is the key point, since the new things tend to be skewed towards the strengths of the lead SKU.

 

[wco81]

Well I don't care if there are no achievements or XBL features.

Just get me at least performance parity.

 

[Silent]

I find it intriging because the situations seems so differently on each continent with regards of sales and attachrates. It's a pitty that people tend to discuss with only the situation in the N/A as a startingpoint. Offcourse if you target only the N/A market that is a whole different story.

 

[AlphaWolf]

I find it intriging because the situations seems so differently on each continent with regards of sales and attachrates. It's a pitty that people tend to discuss with only the situation in the N/A as a startingpoint. Offcourse if you target only the N/A market that is a whole different story.

I don't think it unusual to expect that developers would generally target their home market first.

 

[AlNom]

Devs​

could also have leverage the extra few MB available in the 360 ram to implement fr10 hdr which is free on the computational side but augment frame buffer size.

Just to let you know... FP10 is a 32bpp format(10-10-10-2), just like RGBA 8888, so using FP10 does not entail a larger buffer than using RGBA 8888. The drawback to FP10 is that it is expanded to 16 bits per component, so it is essentially the speed of using FP16 (16-16-16-16).

Save on framebuffer size, get some MDRR, but lose on speed.

(On a side note, I wonder how a 21-21-21-1 format might affect the dynamic range )

 

[liolio]

I find it intriging because the situations seems so differently on each continent with regards of sales and attachrates. It's a pitty that people tend to discuss with only the situation in the N/A as a startingpoint. Offcourse if you target only the N/A market that is a whole different story.

I guess that your speak about Europe, right?
If we put nintendo aside.

Because if MS bombed in Japan, ~1 million in one year for Sony is pretty ugly, situation improves lately and they will end up with a healthy market share, but IMHO too late to change this gen figures (but important to reach profitability).

For Europe, the situation is more balanced, but I still wait for proper informations, IE in which countries Sony has succeed at match 360 user base.
I think Sony will have did it in France after this fall
May be there's others European market that really support Sony but which and what their size? there are also more balanced countries.

In Uk a growing market which is about a third of the european market, I'm pretty sure that sony is far from catching up, I feel more like Sony is fightning to maintain a statuquo in sale with MS. A huge + for MS.

Worse MS has make clear again that for 2008 Europe will be their battleground
http://www.gamesindustry.biz/content_page.php?aid=31694
I think they consider Us almost granted.

Ms make money now they will swallow the RroD costs, they have been lagging their pricing strategy I think, I think they will at last be aggressive with pricing, my bet is between march and June to clear the Chanel before the arrival of full 65nm 360.
My bet is that Sony won't react before fall 2008, more Ms will continue to provide attracting bundles no bare systems.

So I'm not sure the situation is so different for hardware in Europe or to a point it could change editors decisions, Sony may outsell MS in Europe,but significatively? I don't think so.

Worse both consoles are still too expensive to reach real mass market, I think MS will have an opportunity to break some of Sony momemtum in Europe during 2008 first semester.
(My guess)They will restaur the 100$/euros difference between the ps3 and 360, more due to older games library they could bundle one more game in the pro pack for cheap.

Don't get it wrong the situation is pretty tough for Sony even in Europe, in Japan the situation is pretty horrid but will improve due to almost no 360 penetration there, in Japan PS3 is mostly the only hardcore gamer system available (not that ps3 or 360 can attract casual gamers but the wii takes its shares ).

 

[liolio]

Just to let you know... FP10 is a 32bpp format(10-10-10-2), just like RGBA 8888, so using FP10 does not entail a larger buffer than using RGBA 8888. The drawback to FP10 is that it is expanded to 16 bits per component, so it is essentially the speed of using FP16 (16-16-16-16).

Save on framebuffer size, get some MDRR, but lose on speed.

(On a side note, I wonder how a 21-21-21-1 format might affect the dynamic range )

I used to think that it was the other way around, are you really sure?
I remember xenos support AA + some others stuffs in fp 8/10 but not in fp16?

EDIT I'm reading Dave's article... again...

EDIT from the xenos demystified thread:

The ROP's can handle several different formats, including a special FP10 mode. FP10 is a floating point precision mode in the format of 10-10-10-2 (bits for Red, Green, Blue, Alpha). The 10 bit colour storage has a 3 bit exponent and 7 bit mantissa, with an available range of -32.0 to 32.0. Whilst this mode does have some limitations it can offer HDR effects but at the same cost in performance and size as standard 32-bit (8-8-8-8) integer formats which will probably result in this format being used quite frequently on XBOX 360 titles

Still not clear for me... some devs gave explanation recently I can find out the post

 

[nAo]

What lead SKU usually ammounts to is what the none specialists are running on, and often where new things are tried first.

The latter point is the key point, since the new things tend to be skewed towards the strengths of the lead SKU.

100% spot on (according my own experience at least)
I'd like to add that the specialists that try new things on the lead SKU should posses a good grasp of both platforms architecture, so that they will be able to prototype new stuff from the start that can run well on all SKUs.
I know it sounds so obvious that I shouldn't even be mentioned, though most companies got it wrong.

 

[Brad Grenz]

To stay alive means to minimize risk. Leading with PS3 and hence resulting in a 360 game with less than it's full potential is risk. Worded another way, the sales gained from a better PS3 version do not outweigh the potential risk of losing 360 sales, not at this time.

Right, but the resulting "less than full potential" 360 game will still run really well. As a "risk" it's practically nonexistent. The tangible risk is a crappy PS3 port like we've seen time and time again. The Orange Box is a prime example. HL2, et all, are good games on other platforms. On PS3 it runs like shit and that's the word of mouth that's going around. It's not difficult to imagine sales on the PS3 for a quality version matching the Xbox360 sales relative to installed base much as COD4 and AS did, but instead it's barely moving. That is the real risk publishers are taking. As long as the 360 versions of PS3 lead games don't suck they will sell. And there's no reason to think being PS3 first will make them suck. The fact that you have to imagine some ludicrous set of clone game that can't differentiate themselves in any manner beyond slight technical markers is evidence of that. That doesn't happen. There's always a difference in style or story or marketing or first to market, etc.

I'm not arguing that bean counters don't just think of the 360 as safer. That's probably true. What I'm arguing is that if you've already decided to go multiplatform there are lots of good reasons to focus on the PS3 first which risk little for your 360 SKU. The bean counters probably don't think that way. But they should.

 

[assen]

The suits will argue all they want, normally in favour of not spending money, but if you want both your products to be the best possible, you are more likely to get closer to that ideal by going PS3 >> XB360 than vice versa, and you'll hit that target is you haven't independent design teams for both systems creating the engines to run the same assets.

Speaking of "the suits" as some evil entity that schemes and plans all day to bring down your product is childish and unfitting for a "Geezer", even if he's shifty ;-)

Seriously, business development concerns are just as important a part of game development as the choice of lighting model or a streaming solution. We're not talking demoscene development here, aiming to maximize hardware utilization. If "the suits" see a team too obsessed with technical excellence to notice market realities, they might pack up and leave (if they are a publisher overseeing an external developer) or simply shift marketing resources to another team (if they are part of the same company).

 

[one]

Like I mentioned earlier, they were able to get away with it because Burnout doesn't really have any competition to it. What if it did though? Burnout has its fans and will sell well regardless, but what me and you don't know is what would have happened if there was an alternative title available that was a bit better because it had the 360 as it's native platform. Do you really think sales of Burnout would not take a hit on 360 if a better game was available? Is so then thats where you and me differ. Given the high volume of good games on 360 this year, competition is fierce and it's unlikely a typical family will buy two Burnout style games, they will likely pick just one. Given the availability of a better alternative I myself would go for the better game, but maybe thats just me.

Now the assumption here is that a 360->PS3 port is always crap. Let's just say for the sake of argument that the same is true and this fictional alternate burnout style title runs with frame rate drops on PS3. Sure, they will lose sales on the PS3 version. But realistically, at this point in time, who cares? Putting bias and all that aside, just look at the numbers. 360 titles sell far more. Even top tier PS3 games sell far less than the 360 best sellers. Which do you think would worry a publisher more, risk selling 300k less copies of their PS3 version, or risk selling 800k less copies of their 360 version?

All of the assumptions I read here allude to the fact that a 360 game that is a PS3 port should be good enough, and sales should be all peachy because 360 owners buy anything. I think that's a risky assumption. It's certainly not a risk I'd be willing to take if I were financing the project.

Is it wrong to assume generally a 360 version is already 'good enough'? If you allow me to speak for CFO and bean counters, "if you have human resources available to make an already good 360 version 5% better, better spend the same resources for fixing a PS3 version to make it 50% better, or throw both of em and make 3 Wii games instead!"

 

[Barbarian]

if you have human resources available to make an already good 360 version 5% better, better spend the same resources for fixing a PS3 version to make it 50% better, or throw both of em and make 3 Wii games instead!"

Right, only if it was that easy. First it takes a lot of top engineers to get your PS3 "port" up to speed. On my project the PS3 sku was lagging few months behind the 360 sku in features and bugs. It didn't help that we had 30 xbox360 devkits and 6 ps3 devkits.
Setting up for ps3 development took a while, from messing around with SDKs and firmwares to dealing with GCC and toolchain issues. So by the time we realized all the limitations of RSX etc, we were too late in production and doing totally different solutions for both platforms was not an option. So the PS3 ended up always having to catch up to whatever clever tricks we were running on Xbox360. Eventually due to time constraints we had to run PS3 at 30fps, while Xbox360 runs at 60fps. Getting PS3 to run the same would have required massive refactoring of code, art and design.
For the next project we have a much better idea what works better on PS3 and things to keep an eye on, but we'll still lead on Xbox360 since it's platform is the easiest for development, not only for programmers but also for art and design, from profiling tools, to custom tool connections to host PC, debugging, automation etc etc.

 

[Shifty Geezer]

Speaking of "the suits" as some evil entity that schemes and plans all day to bring down your product is childish and unfitting for a "Geezer", even if he's shifty ;-)

You're reading an insult where there wasn't one. 'Suits' was just a nickname/shorthand for Management.

Although there is implied in the term a certain blinkered mentality because management on the whole in all industries is extremely cautious and lacks vision, though of course that doesn't apply to everyone. If they're doing their job well, great. If they're doing their job without actually understanding the underlying principles involved, which is a painfully common situation in all businesses where the management make dumb decisions that the actual company workforce know are really dumb decisions but have to implement them anyway because the boss orders it, then that's bad. In the context of this thread, if they throw more money at one SKU because that system gets higher sales to the detriment of the other SKU, without understanding that improved investment in the other SKU would yield better returns all round (not saying that's so, but that's the hypothesis presented here), then that would be a poor and short-sighted choice based on not understanding the technical points of game development and the hardware.

 

[novcze]

ok, so what we know, PS3 is not XB360 and XB360 is not PS3, you can`t do platform specific tricks on xb360 and expect great results on PS3, same in opposite direction.
I don`t see XB360 exclusive games be so ahead in graphic department and yes they have one year to the good.
Anyway, I don`t buy rushed crappy ports of XB360 games that performs poorly on PS3 just because developers don`t have enough time or enough experience.
I shall buy Burnout for example, because is it good PS3 GAME that runs on 60fps without tearing or framerate drops and i don`t care how xb360 version looks or what can they do with PS3 version if they were exclusive to platform. See that logic? I just buy good GAMES, not crappy ones.

 

[Acert93]

Yes but not ~50 mb less than 40 mb now. And I know _phil_ is a developer. Maybe, the 360 memory footprint decrease but I never heard anything about that.

There is also a difference in framebuffer footprint in addition to OS resources...

Is it wrong to assume generally a 360 version is already 'good enough'? If you allow me to speak for CFO and bean counters, "if you have human resources available to make an already good 360 version 5% better, better spend the same resources for fixing a PS3 version to make it 50% better, or throw both of em and make 3 Wii games instead!"

5% is a lot, notably when you are competing with other titles that aren't taking this approach. A title that scores 80% on the 360 may see a significant decline in potential sales compared to a title with an 85% score. Put another way, 2 developers are making "Modern Shooter" games. One goes for the "Push the 360" model and the second goes the "Same 360/PS3" model of development. The first scores an 85% on the 360 and 75% on the PS3. The second scores 80% on both platforms. In an ideal world everyone would have the resources of IW (who not only had a million seller in 2005 but also had a fairly modern engine running on the 360 at launch as well as strong publisher backing) to knock out a 95% quality game on 3 platforms (PC, 360, PS3).

But that isn't reality.

Any tradeoffs made in underutilizing a platform need to be made up by the other platform as well as cover the development costs on the 2nd (and 3rd and 4th) platforms.

Last gen saw this go two ways: The Xbox and GCN. PS2/Xbox titles were fairly plentiful and Xbox versions ran well, typically even upgraded. Why? While the install base wasn't huge, it was large enough in software sales volume to justify versions and had fast enough hardware to minimize development issues. The GCN had some performance gotchas that would result in either visual downgrades and/or expensive re-engineering and design. The install base invested heavily into 1st party software and not so much into 3rd party software, especially those on other platforms.

I don't remember a push for PS2 titles to underutilize the hardware to work well/look the same as the GCN builds--or to use GCN features that PS2 couldn't utilize.

A lot of this falls back to how people spend their money. The platform that consumers buy software on will get the most attention in general.

As for your other arguement that the 360 market is maxed and they should look to the PS3 as an expanding market, the sales indicate that the 360 market is growing faster in NA (where the big software sales are being generated). Further, does putting substantial effort in a small market the best tradeoff? Creating features and content for the larger market can make you more competitive and visible in the larger market.

There is always room for both. I know some developers have looked at the PS3 as an oppurtunity to distinguish themselves because of the competition on the 360. Big titles, like CoD4, have a huge financial benefit of being really good on all titles.

But I think each title is unique in this regards. Some will benefit more than others, and in some cases multiplatform development spreads development focus and resources. There are a lot of scenarios and factors to consider (even stuff like brand and audiance and market growth) but I think it is easy to visualize situations where being multiplatform, exclusive to the 360, exclusive to the PS3... even version on the Wii... can ultimately be beneficial to the publisher. Being multiplatform can be a "good" thing for consumers as well if it means more development time and resources... but bad as well as the resources afforded don't cover the cost of development and thin out the focus and quality.

On another note and not so ironically most of the posts on the forums about "I want this title to be exclusive because __________" or "I want this title to be multiplatform because ___________" revolve more around the console debate than the benefit/detriment to the title itself.

 

[Arwin]

In the end, just because a company that are self proclaimed Sony fanboys says that leading with PS3 is the way to go doesn't make it fact anymore than if Ford said that all car makers should build their cars off the Ford Fusion platform. If you want the correct financial answer then just look around at the game dev situation today and what most companies are doing. You may think they are all "silly" for doing business the way they are doing, but I'd wager their cfo and bean counters disagree with you.

You're making a lot of big assumptions here though, aren't you? I mean, Criterion basically started on PS3, and wrote the game making good use of modern technologies that emply data streaming, physics calculations and so on, and made their programming modules something that could fit well in the SPEs.

However, what if they managed to successfully transfer this general approach to the 360. What if these basics and general principles of game engine design translate very well to the 360? They may just as well have said that they designed a next-gen game engine that is well suited to both platforms instead of just the one. They happened to lead on the PS3, but they are very good coders and have done some impressive work on the 360 also.

Have you even seen both games running on both platforms, joker4545? I wouldn't want really to continue this discussion before you have (got the PS3 demo a while ago and a few nights ago also tried the 360 version).

Oh and on the 360, Flatout UC is competition for Burnout Paradise, methinks. I have yet to try that game though to see how it matches up against Burnout.

 

[Pozer]

It will be interesting how the Unreal Tournament 3 situation plays out over the coming months. Epic/Midway completely denied the 360 SKU for the holiday. Which probably would have been their bestselling by a far margin. Now we see signs that it isn't selling so hot on the PC. It will be interesting to see how the midway stockholders react and how the 360 owners react to what could be perceived as 6month old leftovers of a bad meal when it finally is released.

Personally im of the opinion that shipping ported SKU's later is a bad idea. Theres a certain level of hype you only achieve once. Especially these days the kids don't want to seem like they're behind the times or enjoying an experience there friends had 6 months ago. If you delay a SKU either you must have a really really great game or offer something very very substantial above the initial experience just to have 50% chance of the later sku succeeding. I really think epic is gonna have a hard sell on there hands because they offering a very shallow (yet fun) multiplayer expereince which may have been popular 10 years ago but has since come off as dated.

 

[wco81]

How would UT3 have done on the X360 with all the other hot-selling FPS being sold this Holiday season?

 

[assen]

How would UT3 have done on the X360 with all the other hot-selling FPS being sold this Holiday season?

Definitely better than 18k, that's for sure.
But it still probably wouldn't have paid for itself.
Now as it is, UT3 PS3 probably pays for itself and the PC version thanks to the Sony exclusivity contract.
It's all very muddy for a company like Epic, because it's hard to draw a line between engine development costs and game development costs. On thing is certain - if any other company started from scratch, or licensed UE3, and produced this much art for a game, it would be deep in red with this kind of sales.

 

[Pozer]

How would UT3 have done on the X360 with all the other hot-selling FPS being sold this Holiday season?

I think a December release would have sold great on the 360. December had no major releases on the 360 and live players are already looking past cod4.