The Middle Ground Approach: PC, PS4, XB1

[Billy Idol]

I do wonder the following: I read several times here at B3D that assets are adjusted to the expected resolution of the renderer. So there is something like '1080p' assets and so on.

When one version uses 900p and another uses 1080p as a resolution, is there even a difference in the asset quality? Is there a difference to PC versions, which allow basically arbitrary resolution? If I see it right, PS4 and X1 games are about the same size...so I do not expect that the asset changed, same goes for the PC game..right?

 

[Goodtwin]

If it makes life easier for the developers, then yea, I think it makes sense to aim for parity. The truth is that players ultimately judge a game on its content, the direct comparisons only mean a lot to the fanboys. COD was generally superior on the 360, but that didn't stop it from selling tons of units on PS3. From a gameplay standpoint, there often isn't much that can be done on one console compared to the other, its just a prettier coat of paint that the more powerful hardware allows. The percentage of people who would have bought a game like Destiny on PS4 only if it was really pushing the hardware is a pretty limited percentage of consumer. At some point the selling point of your game should be the "game", not that its a technical achievement. Fanboys love that, and a lot of PC gamers like games that melt their PC's, but ultimately the most popular games in the world don't rely on being a technical marvel, so as a developer, why stress over trying to push the more powerful platform?

 

[Shifty Geezer]

I can't imagine Infinity Ward or DICE using that approach in the long term, for example. Not after the audience sees how Uncharted 4 and The Order look like.

I disagree. I don't think gamers are going to effectively boycott third party titles and only play the limited range of first party titles because only they look good enough. First and foremost is the game. Someone's not going to buy UC4 over COD even if it looks much prettier if they don't like action adventure shooters as much as hard-core multiplayer shooters. I'm not going to avoid FIFA and buy The Order when I want to play a football game even if The Order is gorgeous and FIFA is only using 50% of the consoles power or whatever (figure chosen as extreme example of performance deficit, and not intended in any way to be representative of performance differential between PS4 and X1 or any other machine). A game that's going toe-to-toe with first parties, such as a racer (Project Cars) versus GT or DC or Tomb Raider vs Uncharted, will need to, but otherwise the game is sold on its merits. Is it fun enough and pretty enough? That's good enough, stop there.

I do wonder the following: I read several times here at B3D that assets are adjusted to the expected resolution of the renderer. So there is something like '1080p' assets and so on.

When one version uses 900p and another uses 1080p as a resolution, is there even a difference in the asset quality?

Not with that small a difference. Where you have a significant disparity in res, you won't need such high resolution textures to achieve relative clarity on the lower resolution screen. Generally choice of assets is by hardware power though. Less RAM means less detail, whether geometric or texture. Less geometry power means a need for fewer objects and/or simpler ones. Less shading power means a need for simpler shaders. Less BW or ROP power means less drawing. Two identical machines would be best served with identical assets regardless of rendering resolution.

 

[Scott_Arm]

Because studios iterate from one version of a game to the next, or reuse tools and engines moving from one game to a new game, you'll probably see the multiplatform games progressively stress the hardware more just because they can. I just don't know if there will be the push to keep the hardware maxed out, especially now that people are demanding games run without dips in framerate. At least this time around you'll probably get even better performance out of middleware since Xbox One and PS4 will likely share a lot of optimizations that could also benefit the PC as well. Still, my expectation is that you will not see vastly different features on one platform vs another. The games will look relatively the same but with changes to resolution, AA being the easiest ways to scale.

 

[Mobius1aic]

Because studios iterate from one version of a game to the next, or reuse tools and engines moving from one game to a new game, you'll probably see the multiplatform games progressively stress the hardware more just because they can. I just don't know if there will be the push to keep the hardware maxed out, especially now that people are demanding games run without dips in framerate. At least this time around you'll probably get even better performance out of middleware since Xbox One and PS4 will likely share a lot of optimizations that could also benefit the PC as well. Still, my expectation is that you will not see vastly different features on one platform vs another. The games will look relatively the same but with changes to resolution, AA being the easiest ways to scale.

This.

The performance table of a hypothetical multiplatform game could/should look like this:

Xbone: 900p @ 30 FPS
PS4: 1080p @ 30 FPS
PC: 1080p @ 60 FPS

Honestly I'm fine with this. I'd expect the raw IQ to look the same close up, though I'd still hope for more entities, better LODs, AA, AF of course, which on the PC version can go a long way.

 

[Deleted member 13524]

I disagree. I don't think gamers are going to effectively boycott third party titles and only play the limited range of first party titles because only they look good enough. First and foremost is the game. Someone's not going to buy UC4 over COD even if it looks much prettier if they don't like action adventure shooters as much as hard-core multiplayer shooters. I'm not going to avoid FIFA and buy The Order when I want to play a football game even if The Order is gorgeous and FIFA is only using 50% of the consoles power or whatever (figure chosen as extreme example of performance deficit, and not intended in any way to be representative of performance differential between PS4 and X1 or any other machine). A game that's going toe-to-toe with first parties, such as a racer (Project Cars) versus GT or DC or Tomb Raider vs Uncharted, will need to, but otherwise the game is sold on its merits. Is it fun enough and pretty enough? That's good enough, stop there.

I meant comparisons between games of the same genre, of course.
Gamers obviously won't boycott third parties for not looking as good, but IQ is just another factor that weighs in (more or less depending on the person, but it does, nonetheless) over the decision to eventually choose one game over another. Money isn't infinite for many gamers and those may have to choose between DriveClub and a future Need for Speed, between a future Killzone/Resistance and Doom, Order 1886 and The Division, etc..


Again: I'm talking AAA games, where many of them need to sell over 5million units just to break even, so they must try to turn every single point in their favor.
I don't think "pretty enough" is sufficient in this market.

Besides, try to see it from the perspective of a publisher/investor. If you're spending say $15 million on the IQ department alone, they'd better be reproducing close to the best graphics in the industry.

 

[Jay]

@Mobius1aic
Gimp XB1, just so it can prove hypothetical performance and win DF face off's and keep the warriors happy?
Personally I think the studio should decide.
If the games art style and engine looks suffers bellow 1080p, then choose 1080p
If the games art and engine isn't affected as much by a lower resolution, drop it to 900p if you cant get all the effects and framerate done at 1080p.

I would like if every platform was pushed as much as possible, but I also understand that there's a lot more to the business of making and selling games than that.

The meltdown that's going to happen on the net when the XB1 wins a face of is going to be amazing.
Could have parity but have better exclusive content etc. Which was one of the main reasons for the win in Destiny.

 

[Jay]

Again: I'm talking AAA games, where many of them need to sell over 5million units just to break even, so they must try to turn every single point in their favor.
I don't think "pretty enough" is sufficient in this market.

Besides, try to see it from the perspective of a publisher/investor. If you're spending say $15 million on the IQ department alone, they'd better be reproducing close to the best graphics in the industry.

Pretty enough = best we could do given budget and time etc
Guess in this scenario it may be best not to make it for that platform in the first place (as it wont sell well enough), as the cost to improve the game to match the exclusives may be too much.
Could use that money to make the game better for the remaining platforms, or just save that money so game had a lower budget.

Just turning up the AA, or AF if possible isn't going to make the game look so much better, it would take a lot more than that.

 

[Scott_Arm]

I meant comparisons between games of the same genre, of course.
Gamers obviously won't boycott third parties for not looking as good, but IQ is just another factor that weighs in (more or less depending on the person, but it does, nonetheless) over the decision to eventually choose one game over another. Money isn't infinite for many gamers and those may have to choose between DriveClub and a future Need for Speed, between a future Killzone/Resistance and Doom, Order 1886 and The Division, etc..


Again: I'm talking AAA games, where many of them need to sell over 5million units just to break even, so they must try to turn every single point in their favor.
I don't think "pretty enough" is sufficient in this market.

Besides, try to see it from the perspective of a publisher/investor. If you're spending say $15 million on the IQ department alone, they'd better be reproducing close to the best graphics in the industry.

I don't know. Gameplay does seem to sell over graphics.

 

[Shifty Geezer]

Again: I'm talking AAA games, where many of them need to sell over 5million units just to break even, so they must try to turn every single point in their favor.
I don't think "pretty enough" is sufficient in this market.

[Well, 'pretty enough' by definition means 'pretty enough to get the sales', but I agree that AAA tends to chase the higher levels of attainment. This couples with things like Metacritic scores. You don't want to drop points for graphics if that'll lower your score and lose your bonus. However, AAA blockbusters hoping for GotY awards are a small part of the games market, and talking more generally about games developers, pushing a high end machine doesn't make economic sense.

Besides, try to see it from the perspective of a publisher/investor. If you're spending say $15 million on the IQ department alone, they'd better be reproducing close to the best graphics in the industry.

The most sensible perspective is to look at how much your willing to spend on graphics and how many more sales that'll get you. If you spend $20 million getting XB1 looking gorgeous, and it'll cost another $5 million to get PS4 looking notably more gorgeous, is that extra 5 mil going to result in significantly more than 5 mil extra profits from PS4? Whereas just dialling up the quality settings of a middleware platform, you can make the PS4 version better for pretty much free.

 

[FarticusMaximus]

I'm sure at the start while most games are enhanced ports of last gen engines and games it will be easier to reach parity but I bet in year 3 we will see the xbox one fall back to 900p or even lower and the ps4 stay at 1080p.

Based on evidence, exactly the opposite appears to be true. It's no different to what happened at the start of last gen, where PS3 engines needed significant Cell-specific tuning to achieve parity with the more straightforward 360 engines and for a long time the result was many multiplatform games ran at lower resolutions on PS3.

Ports of last gen games have been using engines that are not optimised for ESRAM dependency. These engines perform much, much better on the PS4's more straightforward architecture.

As newer engines emerge which are better optimised for ESRAM usage, evidence shows that more and more games are achieving 1080p on XOne and more multiplatform games are achieving parity with the PS4 versions also, or at least the delta is reducing considerably.

Historically, multiplatform developers tend toward parity. There's no reason to think it won't be the case this gen.

 

[FarticusMaximus]

Gimp XB1, just so it can prove hypothetical performance and win DF face off's and keep the warriors happy?
Personally I think the studio should decide.
If the games art style and engine looks suffers bellow 1080p, then choose 1080p
If the games art and engine isn't affected as much by a lower resolution, drop it to 900p if you cant get all the effects and framerate done at 1080p.

I would like if every platform was pushed as much as possible, but I also understand that there's a lot more to the business of making and selling games than that.

DF and the platform warriors. They represent a tiny percentage of the gaming community. Very vocal on forums, but pretty much irrelevant in terms of the gaming market as a whole.

Expecting developers to listen to this noisy minority is ludicrous and destructive. If developers really did pander to this kind of mentality, and spend more and more on pushing each platform to it's limit, do people think the outcome would be:

A) Significantly more sales, offsetting the increased development cost (hint: no)

or

B) Less chance of time and cost budgets being met, no significant increase in revenue, more chance of delays pushing release into an undesirable time-frame, more chance of developer going broke/being cut loose from a publisher. (hint: yes)

The most this action could realistically accomplish is making sales on one platform slightly higher than the other. It's very unlikely to increase overall sales.

There's really very little choice. If you want to stay in business you need to set a baseline (time, budget, image quality, content and features) and stick to it.

'Could squeeze out a few more effects', 'should use 100% of platforms power', and all these other enthusiast wishes are irrelevant when it comes down to a simple man-hours spent vs. revenue equation. Here, in the real world, that equation is pretty much the be-all and end-all.

 

[Cjail]

Historically, multiplatform developers tend toward parity.

Or rather toward what is more convenient.

 

[function]

I think what Bungie did with Destiny was sensible.

They set themselves a fair target for both platforms, and if the Xbox One wasn't able to meet it, then, well, it wasn't going to meet it. The point was that they were able to put all their efforts into a single, core experience and optimise around that.

The "middle ground approach" is interesting as it allows developers to work around a single core experience and single set of technical characteristics, but with one or two easily compartmentalised, art independent elements that can diffuse fanboy rage.

As the "core" gamers begin to represent a smaller proportion of the market and mainstream a larger one, such tactics should become less necessary, and developers will do away with any of the modular "extra" components that require anything other than a very basic amount of work to include.

 

[function]

Or rather toward what is more convenient.

Think in terms of cost and risk rather than convenience (the word's a little too close to 'lazy' for my comfort!).

 

[Prophecy2k]

Historically, multiplatform developers tend toward parity. There's no reason to think it won't be the case this gen.

When your trend consists of a single datapoint (i.e. only one single gen have MP games tended towards parity, and only for about half of that gen), I'm not sure that allows it to qualify as a rule.

In the PS2/XB/GC era, MP games looked completely different on each platform, with each game dev'd to the strengths of the individual platforms. You even saw alot of that in the PS360 gen.

So your statement here is categorically false in my mind.

On the other hand, I could still see many developers this gen simply pushing out the same game in terms of assets on the XB1 and PS4, whilst just running the PS4 version at a higher resolution, or using a better AA algorithm.

It will still provide a clear a perceptual difference for the hardcore PS4 fanboys, and yet be similar enough that XB1 gamers don't feel as if they are getting the shaft.

 

[Scott_Arm]

...

In the PS2/XB/GC era, MP games looked completely different on each platform, with each game dev'd to the strengths of the individual platforms. You even saw alot of that in the PS360 gen.

...

I would disagree that there was a lot of that last gen. If anything, if you look over the last two gens, the trend has been towards partiy. Last gen, the multiplatform games had some differences, but they tended to be resolution, AA, better shadow filtering, LOD transitions or something like that. Nothing game changing. The look of the games was generally the same.

 

[Shifty Geezer]

In the PS2/XB/GC era,...

Those consoles were so disparate that aiming for the common denominator would have produced a dog of a game. "Okay, leave out shader effects and high poly counts on the XB game because the PS2 version can't handle it. And leave out the awesome particle effects on the PS2 version because the XB can't handle it. What have we got? Oh, a bunch of low-res, flatly lit poly assets. Maybe we should rethink this parity thing?"

They had to write an engine for each version of the game, tailored to the hardware, so tweaking the game to make the most of that engine was pretty much necessary to get something produced, I think.

 

[FarticusMaximus]

When your trend consists of a single datapoint

It doesn't. The hardware generation has no bearing on the issue. I became a multi-console owner for the first time in 2001 and noticed the multiplatform/parity/power difference then, so that's 13 years worth of data.

In the PS2/XB/GC era, MP games looked completely different on each platform, with each game dev'd to the strengths of the individual platforms.

My memory (while a little hazy) of Burnout 3, SSX 3, Mercenaries and Soul Calibur 2 among many others is of games that looked incredibly similar across platforms.

Were they really 'completely different' looking? Were there DF-style face-offs back then?

 

[dobwal]

When your trend consists of a single datapoint (i.e. only one single gen have MP games tended towards parity, and only for about half of that gen), I'm not sure that allows it to qualify as a rule.

In the PS2/XB/GC era, MP games looked completely different on each platform, with each game dev'd to the strengths of the individual platforms. You even saw alot of that in the PS360 gen.

So your statement here is categorically false in my mind.

On the other hand, I could still see many developers this gen simply pushing out the same game in terms of assets on the XB1 and PS4, whilst just running the PS4 version at a higher resolution, or using a better AA algorithm.

It will still provide a clear a perceptual difference for the hardcore PS4 fanboys, and yet be similar enough that XB1 gamers don't feel as if they are getting the shaft.

In many those cases it was probably a result of cheap porting.

It wasn't a case of devs regularly maxing out the Xbox OG and making PS2/GC ports look bad or the other way around.

The market was less mature in those days. As I remember it, the outsourcing of ports wasn't usually handled by the most respected devs in the business. The devs that handled outsourced porting as their primary business seemed to be the least respected dev by gamers. Someone was always pointing out the "shitty" job they did on porting a past title when that that type of dev was associated with a yet unreleased title.

You don't even see much of that these days. Now there is more talent. The market is much bigger allowing a lot of these devs to work as primaries. Plus the level of investment in porting is greater as pubs aren't as cheap as they use to be.