GT Sport will have damage

  • Thread starter Rob-F1-Fan
  • 121 comments
  • 16,057 views
The thing with collision physics is that it ought to account for deformation, temporary or otherwise. Then if you're doing that anyway, you can create a way of displaying the deformation that's already being determined.

Of course, the reality is that doing collision and deformation tests on the actual car's geometry is going to be prohibitively expensive (computationally), and so you're going to use proxies of some sort, but that rather takes away from the accuracy of the deformation itself (and hence the collisions).
 
The thing with collision physics is that it ought to account for deformation, temporary or otherwise. Then if you're doing that anyway, you can create a way of displaying the deformation that's already being determined.

Of course, the reality is that doing collision and deformation tests on the actual car's geometry is going to be prohibitively expensive (computationally), and so you're going to use proxies of some sort, but that rather takes away from the accuracy of the deformation itself (and hence the collisions).

It ought to account for deformation, but you can probably get away pretty nicely with just doing some sensible stuff with how energy is conserved and redistributed.
 
They've been showing the game off for over a year and it was supposed to be released last November. I think it's safe to say if there was going to be an amazing damage model we'd have seen it by now. This is exactly the same situation leading up to GT5 and GT6 with several features.

"These are placeholder sounds, i'm sure we'll hear the real sounds soon"
"We haven't seen the course maker yet but I'm sure it'll be in the game"
"The AI don't look too good here but I'm sure they'll be better, this is just the alpha"
"Those jaggy shadows look pretty bad but i'm sure they'll be sorted in the final game"

Sense a theme? Don't expect something that shows no signs of coming.
 
It ought to account for deformation, but you can probably get away pretty nicely with just doing some sensible stuff with how energy is conserved and redistributed.
Conservation is easy. But elastic collisions are generally incorrect for vehicles, they'd look goofy at least, and mid-corner punts would be catastrophic. The losses are typically in sound (cyclic deformations) and, uh, more permanent deformations - the tyres' slipping plays a part, too. You can fudge some of that, e.g. the larger the impact, the more inelastic it becomes, up to the limit of deformation, which varies by axis; and by car; and by prior deformation etc.

Conservation gets a little hard when bits decide to detach and fly off, spinning and such. But only really for the big bits.


Redistribution is not so easy, and is the aspect with the more substantial gameplay consequences for car-to-car impacts. Relative stiffness plays a huge role there, and that implies deformation, right down to the tyres. You can fudge that, too, partly by assuming suspension (and tyre) deformation dominates at low energies.

You still can't do any of it on the raw geometry (there's too much of it), you'd have to use proxies. One way is to build a complex damage model to collect data for a stiffness-loss lookup on, say, location/angle-energy values based on a proxy (e.g. a bounding box) per car, and return a different force for each car, or something. But then you don't get visual deformation without running a separate process that recreates some of the stuff in the original simulation, and you don't get any variation as the car deforms or as bits fall off, unless you include all of that in your lookup...


Sure, some of that is overkill, but all of it, even the basic stuff, pre-supposes that the cars react appropriately to those general forces applied to the body or chassis in the first place. I'm still waiting for a proper multi-body approach and the end to break-dancing cars in GT games, because currently (despite the progress) they only respond convincingly to forces applied at the tyres.
 
So, what we saw at the Copper Box multiplayer race, is pretty much it.

I'd guess there will be no damage online. Don't want people leaving a race because of a slight error and spearing off into a wall. Or causing a multicar collision.
 
It seems that creating an accurate damage model AND having fantastic graphics is a difficult task, and would prove a taxing operation on any gaming system. I'm not saying it can't be done, but it's an issue that needs to be dealt with, especially if the feature is promised day one. If we don't see any progress from PD in regards to damage, it will be a very bad look.
 
Why would it not be optional like in the two previous releases?
I don't know why it wouldn't be optional. We've only seen the "option" of it turned off so far. It's the degree of damage on, as promised, we haven't seen.
 
Conservation is easy. But elastic collisions are generally incorrect for vehicles, they'd look goofy at least, and mid-corner punts would be catastrophic. The losses are typically in sound (cyclic deformations) and, uh, more permanent deformations - the tyres' slipping plays a part, too. You can fudge some of that, e.g. the larger the impact, the more inelastic it becomes, up to the limit of deformation, which varies by axis; and by car; and by prior deformation etc.

Conservation gets a little hard when bits decide to detach and fly off, spinning and such. But only really for the big bits.

Redistribution is not so easy, and is the aspect with the more substantial gameplay consequences for car-to-car impacts. Relative stiffness plays a huge role there, and that implies deformation, right down to the tyres. You can fudge that, too, partly by assuming suspension (and tyre) deformation dominates at low energies.

You still can't do any of it on the raw geometry (there's too much of it), you'd have to use proxies. One way is to build a complex damage model to collect data for a stiffness-loss lookup on, say, location/angle-energy values based on a proxy (e.g. a bounding box) per car, and return a different force for each car, or something. But then you don't get visual deformation without running a separate process that recreates some of the stuff in the original simulation, and you don't get any variation as the car deforms or as bits fall off, unless you include all of that in your lookup...


Sure, some of that is overkill, but all of it, even the basic stuff, pre-supposes that the cars react appropriately to those general forces applied to the body or chassis in the first place. I'm still waiting for a proper multi-body approach and the end to break-dancing cars in GT games, because currently (despite the progress) they only respond convincingly to forces applied at the tyres.

I should point out that I was coming from the angle that energy is always conserved, it just depends in which form. Redistribution therefore meaning that kinetic energy becomes some other form of energy.

It seems to me that you could probably get a passable collision system by assuming that x amount of the kinetic energy at y speed gets turned into sound, deformation, or whatever else that is essentially a loss to the things that the physics system is considering. Vary it depending on angle and general area of contact, and you've probably got a decent enough collision system without necessarily needing any visual or calculated deformation.

It'd require a bit of tweaking to get the numbers just right, but that's what developers get paid the big bucks for.
 
I should point out that I was coming from the angle that energy is always conserved, it just depends in which form. Redistribution therefore meaning that kinetic energy becomes some other form of energy.

It seems to me that you could probably get a passable collision system by assuming that x amount of the kinetic energy at y speed gets turned into sound, deformation, or whatever else that is essentially a loss to the things that the physics system is considering. Vary it depending on angle and general area of contact, and you've probably got a decent enough collision system without necessarily needing any visual or calculated deformation.

It'd require a bit of tweaking to get the numbers just right, but that's what developers get paid the big bucks for.
I suspect that's close to how most of the systems work these days, but knowing the energy deposited in each car (as "loss") can help with a visual representation as well. You'd just have to develop that system, or rather the translation from energy to actual deformation (since PD probably have a few deformation methods available to them already by now).

In terms of the racing consequences (and not just hurtling into the scenery for giggles), suspension and chassis deformation should be considered for their believable impact on the physics, and a visual representation of each would be useful feedback as well.

I personally don't particularly like dice rolls and binary switches for damage, visual or otherwise! I like to see a bit of physicality to it all, especially in a "sim" :)
 
Highly doubt it anymore, they'll just keep that rubbish collision physics from GT5.
Probably for "GT7"...

http://www.polyphony.co.jp/recruit/jobs/974/
Physics Engine Development Engineer
・Development of a rigid body simulator
・Development of an elastic body simulator
・Development of a fluid body simulator
・Development of a rag doll simulator
・Development of other physics simulators
 
Probably for "GT7"...

http://www.polyphony.co.jp/recruit/jobs/974/
Physics Engine Development Engineer
・Development of a rigid body simulator
・Development of an elastic body simulator
・Development of a fluid body simulator
・Development of a rag doll simulator
・Development of other physics simulators

Yeah, WW3 will start before we get GT7.
 
It seems that creating an accurate damage model AND having fantastic graphics is a difficult task, and would prove a taxing operation on any gaming system. I'm not saying it can't be done, but it's an issue that needs to be dealt with, especially if the feature is promised day one. If we don't see any progress from PD in regards to damage, it will be a very bad look.

I think Project Cars did both pretty admirably and it didn't seem to tax the PS4 too much personal opinion, So that is the game PD need to beat in my eyes and I really can't imagine GT:S coming even close to matching it.
 
I think Project Cars did both pretty admirably and it didn't seem to tax the PS4 too much personal opinion, So that is the game PD need to beat in my eyes and I really can't imagine GT:S coming even close to matching it.
They do it quite well, yes.
 
I think Project Cars did both pretty admirably and it didn't seem to tax the PS4 too much personal opinion, So that is the game PD need to beat in my eyes and I really can't imagine GT:S coming even close to matching it.

Forza is another good example. If Turn 10 can manage competent damage and some of the best graphics on the market on Xbox One, I'd hope that Polyphony can get the same kind of performance out of PS4, which is a more powerful system.
 
Forza is another good example. If Turn 10 can manage competent damage and some of the best graphics on the market on Xbox One, I'd hope that Polyphony can get the same kind of performance out of PS4, which is a more powerful system.

Forza is a good example I have never played it however but I've heard good things, I think PD put too much focus on aspects that simply make no sense whatsoever this is the reason why turn10 and other devs appear to be leaving PD in the past.

I feel kind of sorry for PD and Kaz they simply seem to have lost their way and are still trying to catch up, Sadly.
 
Probably for "GT7"...

http://www.polyphony.co.jp/recruit/jobs/974/
Physics Engine Development Engineer
・Development of a rigid body simulator
・Development of an elastic body simulator
・Development of a fluid body simulator
・Development of a rag doll simulator
・Development of other physics simulators


http://www.polyphony.co.jp/recruit/jobs/988/

・ Construction of a 3D drawing system using low level graphics API

The Graphics Engine Engineer will be responsible for the development of a graphics system optimized for a GPU equipped on the PlayStation®4, PC's and mobile devices. To achieve the maximum performance, a system using low level graphics API will be developed from the ground up, so the applicant must have knowledge regarding the latest graphics technology for shader development, and a broad range of knowledge such as those required for optimizing GPU's and CPU'
 
http://www.polyphony.co.jp/recruit/jobs/988/

・ Construction of a 3D drawing system using low level graphics API

The Graphics Engine Engineer will be responsible for the development of a graphics system optimized for a GPU equipped on the PlayStation®4, PC's and mobile devices. To achieve the maximum performance, a system using low level graphics API will be developed from the ground up, so the applicant must have knowledge regarding the latest graphics technology for shader development, and a broad range of knowledge such as those required for optimizing GPU's and CPU'
Yep, it looks like the texture/shader part. I guess they want more than a simple geometry variation for collisions. Three years since GT6, I hope they have something acceptable for collisions.
 
I just find it purely hilarious that in 2017, yes, twenty seventeen, we have a racing game developer using the fact that their game will have damage, as news.

They were late with cockpit view, and even then it wasn't used on every car.
They were late with their first physical damage model, and it was abysmal.
They have been late with every damn release since. well. pretty much every game they've released.
They were late to the online scene, and it was ordinary at best.

How PD is still afloat is beyond me.
 
I just find it purely hilarious that in 2017, yes, twenty seventeen, we have a racing game developer using the fact that their game will have damage, as news.

They were late with cockpit view, and even then it wasn't used on every car.
They were late with their first physical damage model, and it was abysmal.
They have been late with every damn release since. well. pretty much every game they've released.
They were late to the online scene, and it was ordinary at best.

How PD is still afloat is beyond me.
To be fair, I'm not sure it can be said that PD are out there touting their damage model as a major feature, or even as news for that matter.

I could be wrong, but to my knowledge it's only been mentioned by Kaz in direct response to an interview question once, or maybe twice?
 
To be fair, I'm not sure it can be said that PD are out there touting their damage model as a major feature, or even as news for that matter.

I could be wrong, but to my knowledge it's only been mentioned by Kaz in direct response to an interview question once, or maybe twice?

Very true my friend. I just find it a bit laughable that he still has to mention the fact that the game will have damage.

If any other racing game took nearly 20 years to incorporate a decent damage model into their game, they wouldn't have a game because people would have stopped buying it. For some reason though, GT has been put up on such an undeserving pedestal and it grants them the ability to pull this crap off.

I jumped ship long ago, after the GT5 and GT6 fiasco and I have zero regrets, but i keep checking back every now and again with the fading hope that GT Sport will bring something amazing to the table. So far though, all I see is damage, futuristic, fake cars, and the signature 'bonk' when cars make impact. Nothing about this game has got me wanting to empty my wallet. In fact, I think each time I read news about GT sport, my wallet gets another stitch in it to keep it closed.

Its sad really.
 
How PD is still afloat is beyond me.

Brand name & support from parent, both fully exploited by Kaz (Double-edged sword imo)

To be honest, having damage or not does not matter at this point of time (even though it is still important). It is because there are more urgent matters (for e.g. getting the game out) for him and his team to solve......
 
Last edited:
Brand name & support from parent, both fully exploited by Kaz (Double-edged sword imo)

To be honest, having damage or not does not matter at this point of time (even though it is still important). It is because there are more urgent matters (for e.g. getting the game out) for him and his team to solve......
A world class eSports racing sim needs a proper damage model IMO. Running into walls and other cars, over revving the engine, locking up the wheels and much more, should have consequences for the game to be taken seriously.
 
proper damage model

Yup, I would want all these too. However, implementing these takes time - it is not a few months works, I believe that you know these better than me. In addition, Kaz is a perfectionist (seen from details he is putting inside cars). If he really wanted to go all out for damage and thus affecting his detailed graphics and other physics, how long will he and his team take? 1, 2 or more years? can everyone wait for that?

To each his own, but I think that sometimes gamers are asking too much from PD (not aiming at individuals and no hard feelings) given their past history and their work, like what @Samus said - Don't expect something that shows no signs of coming. Having expectations is ok but remember higher expectations can also lead to higher fall too. Solve the basic things (AI, sounds, online etc) first, then come back to these stuffs...

my 2 cents...put on flame suit

Edited: not saying damage is not important, but from what I noticed, the usual stuffs first asked whenever a new video is out, are - what car/track it is? how is the sounds/AI? what/how are the driving physics like for example...

From what was presented so far, A world class eSports racing sim, no. A world class pretty racing game, probably yes...
 
Last edited:
Yup, I would want all these too. However, implementing these takes time - it is not a few months works, I believe that you know these better than me. In addition, Kaz is a perfectionist (seen from details he is putting inside cars). If he really wanted to go all out for damage and thus affecting his detailed graphics and other physics, how long will he and his team take? 1, 2 or more years? can everyone wait for that?

Why is four years not enough to develop a game with 19 tracks, 140 cars and damage? I think you should tell T10 this. And Codemasters. And Kunos. And you get the idea.

Also, Kaz is not a perfectionist.
 
Back