Game Disappointments in the Graphics Realm

[MrFox]

I suppose the reason to build an engine from the start with VR in mind is because it requires a reliable dynamic LOD to avoid dropped frames. But then, since the LOD is dynamic, I don't understand why the non-VR experience would suffer. What are the compromises?

Also, Project Cats on PS4 doesn't look that much better than GT Sport if you disable enough options to get a stable 60, and it will receive a patch for psvr so they'll disable most post-effects there. Why wouldn't it be the same for GT Sport?

 

[Shortbread]

If the rendering requirements are different then they are different products.
These have the luxury of targeting a visual style and letting the users choose to pay for hardware that supports it or not. On PS4, users don't have that choice meaning the dev has to picmk a visual style. Will they pick the best possible for TV users (1080p60) or the best possible for VR (2160p60 up to 2160p120)? If PD have picked the latter and designed a game to look good on PSVR, they are compromising the experience for their TV users.

There absolutely should be a TV SKU or mode and a VR product if Sony still want to sell to TV users. Or, they have the option to sell a sub-par TV experience to TV users and hope they still buy in, but they shouldn't be surprised if the response isn't too enthusiastic when other devs are producing racing games targeting the TV users and pro viding a tailored experience.

Exactly.

PC gamers have the luxury of having multitudes of hardware/software configurations on tailoring their gaming experiences with or without VR. The developers can provide eye-candy across the board (VR users and non-users) without hampering one camps preferences. Console gamers have no luxury in that respect - other than treating each device (VR and game console) as a separate platform. Which Sony and other game console developers should be doing.

In the end, I guess what I'm saying is: Keep the platforms separate (games) and provide the best experience for both platform holders.

 

[bunge]

Also, Project [Cars] on PS4 doesn't look that much better than GT Sport if you disable enough options to get a stable 60, and it will receive a patch for psvr so they'll disable most post-effects there. Why wouldn't it be the same for GT Sport?

I think you're in agreement but saying what others are saying from the opposite direction. IF Gran Turismo Sport has a high end PS4 version and a PSVR version, then I don't think anyone (master race notwithstanding) would have a complaint. If all people get is a "gimped" version then this game will be missing some of the PS4 specific features of Project Cars.

 

[milk]

In the PS3 days the optimization process was slower with the complicated architecture, but this time the room for optimization looks smaller and games like Uncharted 4 suggest the potential of the hardware is 99% tapped already, from here on they are at the point they need vectors of evolution other than simply better graphics on a flat screen TV.

I doubt ND isn't full of ideas of how to improve their engine, tech and art for their next game that will make it look even better. The same aplies to all other devs who just now are releasing, or have just released, their first ture next gen games.
The future will tell us how substantial the diference will be, but I'm betting it will be considerable.

 

[Silent_Buddha]

If the rendering requirements are different then they are different products.
These have the luxury of targeting a visual style and letting the users choose to pay for hardware that supports it or not. On PS4, users don't have that choice meaning the dev has to picmk a visual style. Will they pick the best possible for TV users (1080p60) or the best possible for VR (2160p60 up to 2160p120)? If PD have picked the latter and designed a game to look good on PSVR, they are compromising the experience for their TV users.

There absolutely should be a TV SKU or mode and a VR product if Sony still want to sell to TV users. Or, they have the option to sell a sub-par TV experience to TV users and hope they still buy in, but they shouldn't be surprised if the response isn't too enthusiastic when other devs are producing racing games targeting the TV users and pro viding a tailored experience.

Not only can you choose hardware to match the graphics quality you wish, you also have the option to choose the quality of effects to match the hardware that you own. A level of flexibility that is required for that platform due to the myriad hardware and software combinations available.

Consoles don't typically have that flexibility. Very few games released on console give any options to adjust graphics settings (why when there is only one hardware target?) and there is definitely no way to increase hardware capability on most current and previous generations of consoles. There were some exceptions like the enhanced carts for N64.

PSVR is introducing a wrinkle. It's basically introducing another hardware configuration for games. Some may equate this to being similar to just a difference in the display that is being connected (1080p versus 720p versus 480p etc.), except it isn't. You don't have the much if any options for reduced resolution or reduced frame rates. You also have an added required for not dropping any frames if at all possible.

What is means is that now console developers have to either,

1. Support multiple graphics quality options similar to PC, but with less hardware permutations.
2. Treat PSVR as a different platform than PS4. You either target PSVR customers or you target the much larger PS4 customer pool.

If you choose option 2. Then you have some consequences.

1. PS4 consumers with PSVR obviously won't be able to play some PSVR games and won't buy them.
2. PSVR games that can be played without PSVR are going to be significantly graphically compromised. In some cases (like most, but not all, Indie developed games without AAA budget for graphics) this won't be an issue.

So you either target PSVR and potentially significantly reduce the profit potential of your game title. Or you support both and offer a subpar experience for users without PSVR, potentially reducing their enjoyment of the game potentially getting more sales. Or you support PS4 only and have the largest profit potential.

Some one mentioned this being a larger impact to racing games than steering wheels. True, but Sony hopes it won't be as niche as steering wheels. Adding support for steering wheels also doesn't significantly increase the rendering requirements of a game, which in turn basically means it's a different gaming platform in relation to the base console required to run it.

Obviously supporting multiple graphics presets or options would be ideal. And I'm sure PD would do that if they had the luxury. But PD are literally being rushed to have this ready in time for PSVR. They can relatively easily put in graphics quality presets in the time they have available. But what's the point if they don't have the development time available to them to put in the assets and work to have significantly better graphics quality?

BTW - PS4k won't get rid of this problem. It just introduces 2 more gaming platform targets. To the base PS4 and PS4 VR gaming platform targets, they'll be adding PS4+ and PS4+ VR gaming targets. Meaning going forward it would greatly behoove console game developers to start developing as if they were developing on PC to some extent by having multiple graphics presets or just offering graphics rendering options to the player.

TL: DR - Polyphone Digital doesn't have the development time to do what is best. So they are doing what's best for Sony. IE - Push PSVR.

Regards,
SB

 

[Arwin]

GT has historically been pretty unique in thinks like having in-game options to switch resolution, aliasing/sharpening, support for 3D displays, even head tracking. All that has been in there - it's almost nothing new to add VR support. GT6 also added true, albeit Cell based, tesselation.

I am pretty convinced they will leave little on the table graphics wise in either mode. Their biggest challenge no doubt has been porting all that Cell code. Tesselation isn't easy to do efficiently on 'PC' hardware either - several programmers here have suggested that while there is 'native' support, it is too expensive.

 

[iroboto]

Hmmm a lot of assumptions flying about here. Not sure if these assumptions are fair to other developers.

Right now from how I read this thread, it's GTS sucks (debatable) looking for a reason to defend their quality of work. PSVR yup that's why. Let's shift all the blame there. Indirectly: we've secured PDs position of being leaders of car graphics again, but at the same time I don't like it because no one will see it that way. I expect graphics quality to be 11/10. Somehow... On a 60fps simulation racer where we would expect heavier memory contention and much more pressure on the CPU.

Just saying this is how it's reading. Whether or not it's true is something else entirely. I don't think Sport looks bad and it's quite possible that they are using a lot of placeholder textures still while they wait for X or Y track to be laser scanned more or what not.

We know very little of the impacts designing a game for VR has on PS4 titles except that we know it should be more, we know nothing about their actual game and where they are in progress for it except that we know it's due out later this year.

I'm not defending PD but it should be worth pointing out that it comes off as insulting to other developers as though they are incapable of surpassing the bar called PD.


Sent from my iPhone using Tapatalk

 

[Globalisateur]

New thread title ideas:

Gran Turismo Sucks

Gran Turismo Six remastered

 

[MrFox]

I'm not defending PD but it should be worth pointing out that it comes off as insulting to other developers as though they are incapable of surpassing the bar called PD.

Insulting is probably the wrong word. Nobody should refrain from expressing an honest opinion about what we see (it's visually under expectations), and contrast it with an expectation based on past titles.

I'd like to point out GT5 has an 84% metacritic, but somehow there are claims of it being shit, fall from grace, etc.... This is much less rational than simply hoping PD delivers again upon the expectation of gran turismo fans. GT6 was visually impressive, and the driving physics were the best of the generation. We were promised better sounds (seems to be much better in GTS) and we are told by Kaz the quality is currently only 50% of their goal for launch. So maybe expecting something better than what we see now is reasonable?

It wasn't insulting to other devs that fans expected U4 to be amazing, just like all previous games from ND. Sometimes hype becomes a liability, it just happened that ND delivered.

 

[Shortbread]

I'm not defending PD but it should be worth pointing out that it comes off as insulting to other developers as though they are incapable of surpassing the bar called PD.


Sent from my iPhone using Tapatalk

No one's insulting any developers or their prowess on surpassing PD's skillset. This boils down to old fashion bitching about what was expected versus what was shown. And what was shown wasn't something that one would have expected from PD's triple-A franchise first showing on PS4. And anyone with VR hardware (PC wise) can vouch that it requires some beefy specs on running something that looks appeasing (of a modern day game), with good performance. So, more than likely PSVR is very taxing on the PS4 GPU/CPU... hence the PS3 esque like graphics (just sharper and more refined).

 

[Silent_Buddha]

I'm not defending PD but it should be worth pointing out that it comes off as insulting to other developers as though they are incapable of surpassing the bar called PD.

A lot of developers also didn't have the luxury of having 6 years to develop their game and engine as PD did last generation (GT4 -> GT5) during a console generation shift. A lot of developer's didn't have 3 years to create what was basically just an iteration of the previous entry on the same engine (GT5 -> GT6).

Compare that with Turn 10 where they turn out a new entry in the series every 2 years.

It's not hard to see why GT often looks better than the competition regardless of the quality of their engineers compared to the engineers are other development studios.

Assuming relatively equal development talent, greater development time will lead to greater results, especially when it comes to asset creation. Look how long it took to develop GTA V and how that resulted in a gorgeous looking game. Especially when one considers it is also an open world game.

Is PD better than Turn 10? Probably not. Is Turn 10 better than PD? Probably not. But PD gets to have significantly more time between iterations of their games than Turn 10 and most other studios. Of course, time can also be a detriment if development isn't properly managed (DNF and Daikatana, for example), but I'm going to assume that PD are properly managed.

There's a variety of factors that is affecting GT Sport. Short development time by PD standards for a new engine for a modern console driven by the requirement that it be out in time for PSVR. GT5, as an example, wasn't required to be out in time for PS4 launch. VR significantly impacts the performance of any game developed for it. Combined with that it has very strict requirements in regards to performance which further impacts any game created for VR.

Cardboard trees for example could be an effort to reduce vertex data which will have benefits for VR rendering. In other words, you sacrifice some things that you normally could use in a non-VR game (assuming you are programming to the same base hardware) in order to make it work well in VR.

PD can certainly be surpassed depending on the metric used. They are one of the top developers but I don't think they stand out among the other top developers given the development time they are generally allowed. And thus we shouldn't expect miracles from them.

Regards,
SB

 

[Recop]

A lot of developers also didn't have the luxury of having 6 years to develop their game and engine as PD did last generation (GT4 -> GT5) during a console generation shift. A lot of developer's didn't have 3 years to create what was basically just an iteration of the previous entry on the same engine (GT5 -> GT6).

Maybe but they worked on a more complex hardware. The PS3 wasn't developer friendly.

Compare that with Turn 10 where they turn out a new entry in the series every 2 years.

PD and Sony might use a different business model... actually, you can't know the real reason.

If the next Half-life comes this years, it doesn't necessarily mean that Valve was unable to produce a game sooner or that they took 12 years to develop the game.

Not all developers have the same priorities.

 

[Silent_Buddha]

If the next Half-life comes this years, it doesn't necessarily mean that Valve was unable to produce a game sooner or that they took 12 years to develop the game.

Not all developers have the same priorities.

Is there anything that PD makes other than Gran Turismo? If they aren't working on that what else are they working on? As for Valve, how many people actually consider them a developer anymore. I'm not sure they even have the personnel to develop games at Valve anymore. I'd be extremely surprised if another Half-Life game ever comes out from Valve. It's probably more likely that Sony will decide to give up on anything related to electronics and instead decide to start producing petroleum or maybe go whale hunting to make kerosine.

And just by the fact that I mentioned some developers have a far quicker turn around time from game to game showcases that different developers have different priorities. I'm not sure that was ever in question. In general, however, developers are pushed by their publishers as both need to make a profit.

Long development times mean high cost of development and greater risk of not being able to recoup your investment in the game. Sony developers in general have been fortunate to have much more leeway with regards to delivering a game than non-Sony developers. If the game doesn't recoup its investment Sony can still look at it as boosting the visibility of their hardware platform. That's not a benefit that other publishers enjoy other than Nintendo and Microsoft. And Microsoft is far more dependent on having to make sure their stockholders are happy when those stockholders don't view a gaming console as a core business department.

Sony's priority currently is doing what it can to make PSVR a success. A key part of that is pushing PD to make a GT game for PSVR. Hence the rushed development and having to focus on bringing the best experience they can to PSVR even if that provides a sub par experience to PS4 non-PSVR users. Something I doubt very heavily would happen if PD had the time they wanted in order to do a proper job on GT for both PS4 VR users and non VR users. I'm sure whenever Gran Turismo 7 is announced it'll be much better than Gran Turismo Sport as they'll likely have had the time to do the job they wanted on it.

Thinking about wanting a racing game for the launch of PSVR makes me think of Ridge Racer. In the past it was used as a racing launch title for every version of the PlayStation console. Even the PSP and PS Vita featured a Ridge Racer game as a launch title. It's a shame that's not really a thing anymore.

Regards,
SB

 

[bunge]

Would 3-D trees really make everyone happy?

 

[MrFox]

Would 3-D trees really make everyone happy?

Absolutely not. It will be something else then. We just haven't decided what yet.

Maybe it's sony who forced them to support VR against their will, despite all major upcoming racing titles supporting VR on all reasonably powerful platforms. It's a conspiracy against the xbox. Something.

I can't follow what this thread is supposed to be about, it almost was PC vs consoles, maybe we shouldn't delay the inevitable.

 

[Globalisateur]

Absolutely not. It will be something else then. We just haven't decided what yet.

Maybe it's sony who forced them to support VR against their will, despite all major upcoming racing titles supporting VR on all reasonably powerful platforms. It's a conspiracy against the xbox. Something.

I can't follow what this thread is supposed to be about, it almost was PC vs consoles, maybe we shouldn't delay the inevitable.

For me this thread is about the fact that after ~3 years of development the main trailer about GTS running on https://www.youtube.com/user/playstation displays cardboard trees, PS3 textures, and no anti-aliasing at all on many screens (notably car selection).

Obviously they are improving things and I have no doubt the cars will control great, like previous GT games. But this official trailer is really overwhelmingly unpolished on vital aspects of the game (the visual immersion). Is that their "vision" about the final game and how a "next gen" game should look?

 

[Nesh]

One other thing I dont like in GTS are the interior shadows. They are too soft.

 

[Shifty Geezer]

Isn't this also the first GT trailer in the history of GT that didn't wow people?

 

[Nesh]

Isn't this also the first GT trailer in the history of GT that didn't wow people?

Yes

 

[Arwin]

Really? I don't think GT5 and GT6 got so much universal praise either. But for us graphical technuts, at least GT6 highlighted stuff like tesselation.

But perhaps it is true that the trailer is relatively meh, as some of the actual gameplay shown (Brand's Hatch and that Oval) displays better graphics. But right now we know very little about the game in terms of day-night cycles, weather effects, damage, etc.

The GT Academy stuff has a yearly deadline and the FIA stuff had to be announced, so I'm guessing there's a different clock there.

The really interesting bit for me will also be if they managed to solve the Logitech driver issue.