Why I'm very, very worried about GT5

[H20HYBRID_GT]

I'm worried about Gran Turismo, because I feel it lost it's soul...
Today I played GT4 on my old PS2 for a couple of hours and the driving, the menues, the sounds, everything seemed perrrrfect! (even loading times seemed faster than GT5...?).
The only things feeling odd in GT4 is, if you bounce a wall or drive on grass, but other than that it just actually feels like driving a real car... simply PERFECTION!
Now, GT5 or even TT demo in comparison feels like driving against a wall all the time!

Yes, this is me being a bit sarcastic, but it's also my opinion, I think that's what should make everyone (who cares) worrying about Gran Turismo 5... if you ask me, that's the reason for any further delays also, because KY feels likely the same way and he's certainly not stupid, he may be a liar, but he's not stupid. ( I hope )

 

[Dravonic]

Damage was planned for some cars since way back when.

The damage to production cars being an addition seems to fit in very well in the timeline of F3 hitting apparently strong, press and fan response to GC demo without damage...

Especially considering insider docs and all the other stuff GTP had posted to news that pointed solidly to an xmas 09 release it all fits together very well as a hoped 09 xmas release, a final testing of the waters which made KY and PD reacess their situation and push back release to add/polish features.

You're assuming that not all cars would have damage initially but we don't know that do we? Just because the TGS demo worked that way it doesn't mean that the final game was planned to work that way too.

 

[Devedander]

You're assuming that not all cars would have damage initially but we don't know that do we? Just because the TGS demo worked that way it doesn't mean that the final game was planned to work that way too.

https://www.gtplanet.net/yamauchi-clarifies-questions-about-damage/

The next detail Kazunori touched on was regarding damage models. I can confirm that not every car in Gran Turismo 5 will sport damage models — it will only be select sets

That’s consistent with Yamauchi’s previous comments, when he mentioned that only “race cars” could be damaged in the game.

I went into this in detail a long time ago on here, but basically unless we see a "history of GT5" movie or something that clearly lays it out, my best guess is:

Originally GT5 was planned to have damage to race cars. PD was rushing to hit an xmas 09 launch to hit with the PS3 slim and put up a solid competition against F3. At GC PD tested the waters with their demo of a pretty close to final product. First public and press opinion resounded that it's time GT has damage and it's not acceptable it doesn't. F3 had a strong showing around that time.

KY and PD regroup, decide that damage must be implimented and rush to put together a damage engine. It's decided this can't be done right before xmas 09 and considering F3 is showing storng, it's a bad idea to launch GT5 after this much time in the condition it's in. Chances are many will complain about how how GT5 STILL doesn't have damage after all these years (amongst any other issues people think GT should have by now) and F3 will benefit from this direct comparison.

KY and PD tell Sony sorry, we are out for xmas 09 and it will take until mid 2010 to get this released. That's an agressive timeline but they have to give Sony something decent to keep from Sony Execs screaming their heads off over one of the PS's greatest exclusive games missing such an opportunity. That's where I think we are now and I think probably KY and PD really want to push it till xmas 2010 and said mid 2010 knownig a 6 month push back would sound better to marketing than a 1 year push back and also knowing once they were sold on 6 more months, another 6 wouldn't be so terrible especially considering another 6 puts them on track to launch at xmas again giving good options for tie ins. That's why I think we are hearing talk of late 2010 release... KY has played the cards he had to play to get the time he wants to do with the game what he wants.

That's how I see it at least. This is based on how things unfolded and a lot of GTP news posts and what not...

If there is damage, i want it to actually affect performance. This is just getting Stupid. Release the bloody thing already. lol

I assume that mechanical damage will be included for all cars. I think this was even alluded to directly and I recall Kaz saying something like people will turn it off because it's too hard, which doesn't make sense unless it's mechanical damage. Also he said they could set damage so agressive that a sinngle crash would end your race.

So I am pretty sure mechanical damage is in.

 

[eSZee]

he also recently said that only some will have interior damage or higher levels of damage.

 

[Devedander]

he also recently said that only some will have interior damage or higher levels of damage.

Yes and I am not sure if that ever got cleared up but I think interior damage either meant the engine bays and parts you would see only if parts fall off the car or break open (as race cars will be having full damage while production will have scratch and dent only) or if the interior cabins will actually show damage which I am not sure really what value that would have... like a cracked speed guage or the glove box popping open?

Anyhow, it seems likely that the production cars will feature damage modeling simmilar to Forza with no real deformation or minimal deformation and mostly just textures reflecting damage and that race cars will have fully modeled damage like we saw in TOCA or RalliSport Challenge (although I think it will be a step up from both those games).

 

[Dravonic]

https://www.gtplanet.net/yamauchi-clarifies-questions-about-damage/

I went into this in detail a long time ago on here, but basically unless we see a "history of GT5" movie or something that clearly lays it out, my best guess is:

Originally GT5 was planned to have damage to race cars. PD was rushing to hit an xmas 09 launch to hit with the PS3 slim and put up a solid competition against F3. At GC PD tested the waters with their demo of a pretty close to final product. First public and press opinion resounded that it's time GT has damage and it's not acceptable it doesn't. F3 had a strong showing around that time.

KY and PD regroup, decide that damage must be implimented and rush to put together a damage engine. It's decided this can't be done right before xmas 09 and considering F3 is showing storng, it's a bad idea to launch GT5 after this much time in the condition it's in. Chances are many will complain about how how GT5 STILL doesn't have damage after all these years (amongst any other issues people think GT should have by now) and F3 will benefit from this direct comparison.

KY and PD tell Sony sorry, we are out for xmas 09 and it will take until mid 2010 to get this released. That's an agressive timeline but they have to give Sony something decent to keep from Sony Execs screaming their heads off over one of the PS's greatest exclusive games missing such an opportunity. That's where I think we are now and I think probably KY and PD really want to push it till xmas 2010 and said mid 2010 knownig a 6 month push back would sound better to marketing than a 1 year push back and also knowing once they were sold on 6 more months, another 6 wouldn't be so terrible especially considering another 6 puts them on track to launch at xmas again giving good options for tie ins. That's why I think we are hearing talk of late 2010 release... KY has played the cards he had to play to get the time he wants to do with the game what he wants.

That's how I see it at least. This is based on how things unfolded and a lot of GTP news posts and what not...

Yeah it's true, I remember it now. He did say not all cars would have damage. I'm not sure what he meant though. He could be talking about how only the race cars will have deformation and proper damage. It's possible that production cars were never meant to have any damage but I don't think so. I mean, look what he said: "damage models". Damage models probably refers to the car models with the extra work like modeling under the bonnet and the sort (that he also mentioned), so pieces of the car can fall off like we saw with the subaru. On top of that, the mechanical damage engine should work with every car. There's no reason for it not to. The paint scrapes too. If they have mechanical damage and scrapes on the race cars, and both of those things can easily work with the production cars, why not use them on every car?

 

[Devedander]

Yeah it's true, I remember it now. He did say not all cars would have damage. I'm not sure what he meant though. He could be talking about how only the race cars will have deformation and proper damage. It's possible that production cars were never meant to have any damage but I don't think so. I mean, look what he said: "damage models". Damage models probably refers to the car models with the extra work like modeling under the bonnet and the sort (that he also mentioned), so pieces of the car can fall off like we saw with the subaru. On top of that, the mechanical damage engine should work with every car. There's no reason for it not to. The paint scrapes too. If they have mechanical damage and scrapes on the race cars, and both of those things can easily work with the production cars, why not use them on every car?

I dunno... damage modeling in the game world usually refers to modeling any kind of damage. If something doesn't have a damage model, it's pretty much the same as not having damage.

Also he did say only race cars could be damaged at one point (quote above).

People have constantly found ways to see what meaning they want to see in what Kaz has said, but realistically I think without having to reach for something and just taking things the most simple and straight forward way it always shapes up to being damage on production cars was not something originally planed.

I mean saying only race cars can be damaged and some cars will not have damage models is about as close to solid as anything KY has ever said (he isn't notoriusly specific but those seemed awful specific) and along side the way things played out, short of desire to believe damage to all cars was always in there really isn't much to support that theory...

 

[Dravonic]

I dunno... damage modeling in the game world usually refers to modeling any kind of damage. If something doesn't have a damage model, it's pretty much the same as not having damage.

Also he did say only race cars could be damaged at one point (quote above).

People have constantly found ways to see what meaning they want to see in what Kaz has said, but realistically I think without having to reach for something and just taking things the most simple and straight forward way it always shapes up to being damage on production cars was not something originally planed.

I mean saying only race cars can be damaged and some cars will not have damage models is about as close to solid as anything KY has ever said (he isn't notoriusly specific but those seemed awful specific) and along side the way things played out, short of desire to believe damage to all cars was always in there really isn't much to support that theory...

I agree. Scrutinizing every little word is not the way to go but the only reason I'm doing it is because it makes no sense not to include damage in production cars. PD worked on a deformation engine, scrapes and mechanical damage all of which can work with production cars with little to no adaptation. The only thing production cars don't have is the special modeling that allow the deformation engine to go further to deliver (hopefully) amazing results.

Even if they went with the odd decision of damaging only race cars, applying the limited damage to production cars should be easy since everything they need is already done. Delaying the game for a year (hopefully not but it looks like it will), just because of it? Only makes sense if they are applying "race car" damage on every car.

 

[Devedander]

I agree. Scrutinizing every little word is not the way to go but the only reason I'm doing it is because it makes no sense not to include damage in production cars. PD worked on a deformation engine, scrapes and mechanical damage all of which can work with production cars with little to no adaptation. The only thing production cars don't have is the special modeling that allow the deformation engine to go further to deliver (hopefully) amazing results.

Even if they went with the odd decision of damaging only race cars, applying the limited damage to production cars should be easy since everything they need is already done. Delaying the game for a year (hopefully not but it looks like it will), just because of it? Only makes sense if they are applying "race car" damage on every car.

While they did make the engine, I don't think it's as simple as just saying it makes no sense to use it on production cars...

First off there is the licensing issue, which for that many cars may have come to a pretty penny if possible at all.

And just because there is a damage engine doesn't mean it can just be easily applied to a car. You probably have to do some tayloring of the car model with how and where the parts attach (so that things don't break in unbelievable ways for their material and part). I have to imagine a simple invincible shell type model is much easier and less resource intensive than one that is programmed with different materials and has it's panels broken down so as to be deformable.

And look at the production car damage we are getting... it's basically scratch and dent as it is... it looks currently like most of the damage is done via textures and very little if any actual deformation. So even now we aren't seeing the procedural damage model applied to those cars. In fact if damage was planned all along why the sever skew in work done in the damage department for racing cars vs production cars? Wouldn't it make more sense that they were closer in terms of damge modeling than they are?

It all seems to strongly suggest those production cars being planned and built as GT4 style invincible cars up until recently and then a bone being tossed their way in terms of damage to appease the demanding crowds. On the surface it seems simple to conclude that simply having a damage engine would mean it should be applied across the board, but there are a lot of logical reasons not to.

 

[Dravonic]

While they did make the engine, I don't think it's as simple as just saying it makes no sense to use it on production cars...

First off there is the licensing issue, which for that many cars may have come to a pretty penny if possible at all.

And just because there is a damage engine doesn't mean it can just be easily applied to a car. You probably have to do some tayloring of the car model with how and where the parts attach (so that things don't break in unbelievable ways for their material and part). I have to imagine a simple invincible shell type model is much easier and less resource intensive than one that is programmed with different materials and has it's panels broken down so as to be deformable.

And look at the production car damage we are getting... it's basically scratch and dent as it is... it looks currently like most of the damage is done via textures and very little if any actual deformation. So even now we aren't seeing the procedural damage model applied to those cars. In fact if damage was planned all along why the sever skew in work done in the damage department for racing cars vs production cars? Wouldn't it make more sense that they were closer in terms of damge modeling than they are?

It all seems to strongly suggest those production cars being planned and built as GT4 style invincible cars up until recently and then a bone being tossed their way in terms of damage to appease the demanding crowds. On the surface it seems simple to conclude that simply having a damage engine would mean it should be applied across the board, but there are a lot of logical reasons not to.

Yeah I made it sound too simple. What I meant is that delaying the game form 12/09 to 03/10 would be understandable but any further seems far fetched. As you said, it even looks like the deformation won't happen in the production cars and it surely is the most complicated of the three. Three months to include mechanical damage and scrapes on production cars seems more than enough to me.

I think you're confusing PDs slow rate of publicizing game features with the lack of game features. I think they were just holding the info that all the cars would have at least some form of damage.

 

[Devedander]

Yeah I made it sound too simple. What I meant is that delaying the game form 12/09 to 03/10 would be understandable but any further seems far fetched. As you said, it even looks like the deformation won't happen in the production cars and it surely is the most complicated of the three. Three months to include mechanical damage and scrapes on production cars seems more than enough to me.

I think you're confusing PDs slow rate of publicizing game features with the lack of game features. I think they were just holding the info that all the cars would have at least some form of damage.

Well as I said I think they not only delayed it to do damage but also took the opportunity to stick other stuff in to make sure they were ahead of the curve (ie day/night racing, weather etc). I wouldn't be at all surprised if KY knows the value of an xmas release and basically figured "if we miss xmas 09 we might as well take a full year and pack in the goods for xmas 10".

Also doing texture based damage on cars isn't necessary a fast job either... you have to actually design the textures for every car at every leve of damage you want to portray... how long that takes obviously depends on your team and how far you want to take it but I could see it being a pretty lengthy process also...

I think I have a pretty good feel for how PD confirms features (remember PD also has a habbit of publicizing features a lot and early but then leaving them out). When you read Kaz's interviews the wording, the timing and coordination with what we are seeing along with just general marketing common sense seems to paint a very believable and simple story.

It's of course entirely possible it's some complex twisted behind the scenes crazy fest, but when it all fits together so easily and simply as my timeline does above, it's pretty hard to not believe that's just what it is.

Either way, I still say all actual evidence points storngly in favor of production car damage being a recent imlimentation and not something being worked on from the start.

 

[Stonemonkey]

just personal opinion but I have to disagree, I don't see any advantage for a game like GT to have a christmas release, any marketing campaign would be far more effective not being lost in among all the other chrismas crap and not having to compete with everything else from peoples wallets. GT5 will sell anytime and it'll sell PS3s and wheels and whatever else goes with it.

Another thing I'm not so sure about is your idea of the damage, if it's texture based then it's probably quite likely it'll use some algorithm to apply the scratches/damage to the textures already wrapped onto the cars rather than being any pre-made scratches for every car.

 

[P3nT4gR4m]

Well as I said I think they not only delayed it to do damage but also took the opportunity to stick other stuff in to make sure they were ahead of the curve (ie day/night racing, weather etc). I wouldn't be at all surprised if KY knows the value of an xmas release and basically figured "if we miss xmas 09 we might as well take a full year and pack in the goods for xmas 10".

Also doing texture based damage on cars isn't necessary a fast job either... you have to actually design the textures for every car at every leve of damage you want to portray... how long that takes obviously depends on your team and how far you want to take it but I could see it being a pretty lengthy process also...

I think I have a pretty good feel for how PD confirms features (remember PD also has a habbit of publicizing features a lot and early but then leaving them out). When you read Kaz's interviews the wording, the timing and coordination with what we are seeing along with just general marketing common sense seems to paint a very believable and simple story.

It's of course entirely possible it's some complex twisted behind the scenes crazy fest, but when it all fits together so easily and simply as my timeline does above, it's pretty hard to not believe that's just what it is.

Either way, I still say all actual evidence points storngly in favor of production car damage being a recent imlimentation and not something being worked on from the start.

This is, IMO, the real reason to be worried about this release. GT has always enjoyed a position of being the leader in console driving sims. Development for previous games has always been (I'd imagine) KY comes up with a plan and a timescale and, other than inevitable hiccups along the way the game comes out.

Not anymore. With Forza2 and 3 both bringing serious competition to the table, slap bang in the middle of the development, for the first time PD is on the back foot and trying to catch up with what Turn 10 have put out. This means that features such as damage and livery editing have been forced upon them andthat's a much weaker position to be in than if it'd been planned from the start.

Deadline is approaching - there will come a time when the suits order the game to be released whether it's working or not and we could end up with a bug-ridden dog on our hands, full of glitchy half-arsed implementations of this that and the next thing. There is a very real danger that GT5 will end up being ranked "an inferior Forza clone" and if it is they've done it to themselves.

Should have stuck to the roadmap, ignored the competition and competed on quality - an area where they have a proven track record. Instead they're trying to compete with Forza in areas where Turn10 have a head start.

 

[RedBaron]

Achilles' heel of GT is focusing too much on car models and everything else seems underdeveloped in direct comparison, there are even many standard features of competition missing. Track graphics with muddy textures from angle (compare the game with latest FT-86 screenshots in term of pavement detail), cardboard objects, weak and somewhere not even appropriate sound, stupid AI mashing everything in their way and running out after colission like it never was, even from latest TT demo physics I haven't good impression in term of oversteer and comparison to real life drifting. Now they have time to set it right. I will swallow that bitter pill, wait for them to upgrade everything not to stay only biggest car showcase in the world and believe this is the last release date fiasco.

 

[Dravonic]

Well as I said I think they not only delayed it to do damage but also took the opportunity to stick other stuff in to make sure they were ahead of the curve (ie day/night racing, weather etc). I wouldn't be at all surprised if KY knows the value of an xmas release and basically figured "if we miss xmas 09 we might as well take a full year and pack in the goods for xmas 10".

Also doing texture based damage on cars isn't necessary a fast job either... you have to actually design the textures for every car at every leve of damage you want to portray... how long that takes obviously depends on your team and how far you want to take it but I could see it being a pretty lengthy process also...

I think I have a pretty good feel for how PD confirms features (remember PD also has a habbit of publicizing features a lot and early but then leaving them out). When you read Kaz's interviews the wording, the timing and coordination with what we are seeing along with just general marketing common sense seems to paint a very believable and simple story.

It's of course entirely possible it's some complex twisted behind the scenes crazy fest, but when it all fits together so easily and simply as my timeline does above, it's pretty hard to not believe that's just what it is.

Either way, I still say all actual evidence points storngly in favor of production car damage being a recent imlimentation and not something being worked on from the start.

The scrapes, or texture based damage, is using an algorithm to determine where the car hit something and apply the textures there. The only work that has to be done on every car is telling the scrape engine what kind of texture it should apply where. So you don't apply metallic textures in a car with a plastic bumper for example. It doesn't require pre-texturing of the cars.

Anyway, that's not the issue. The main problem with your argument is that it is speculation. I do think it makes some sense and is possible, but I can come up with 10 other things that are possible too.

 

[Devedander]

The scrapes, or texture based damage, is using an algorithm to determine where the car hit something and apply the textures there. The only work that has to be done on every car is telling the scrape engine what kind of texture it should apply where. So you don't apply metallic textures in a car with a plastic bumper for example. It doesn't require pre-texturing of the cars.

Anyway, that's not the issue. The main problem with your argument is that it is speculation. I do think it makes some sense and is possible, but I can come up with 10 other things that are possible too.

I think when they build the textures for damage via texture, they really do customize the tetures to each car or at least sets of cars... for instance the shape of a body panel needs to be brought into consideration when the damage is drawn on, if it doens't bend or countour properly it could look really bad. If you look at the existing scrape damage images they look very fitting fo the car... complex shaped bumpers look decent etc... I don't think it's like the computer just measures where they touch and apply a scarp in that contact area, mainly because that would look odd and if you look at the scrape damage, it has that same look as Forza where there is canned damage for a hit in a certain area even when the hit doesn't make sense for that level of damage (ie bumper to bumper results in scratch all the way up onto part of hood when nothing touched the top of the hood).

As for speculation, I think it's actually the other way.

I have KY quoted as saying only race cars can be damaged and then also saying damage only to some sets, other cars can't be damaged.

That's pretty clear.

Your take on it is that it seems like if they have a deformation engine, why not apply it to all cars instead of just some?

That seems to be the speculation to me. You are speculating a why not in the face of a quote from KY himself saying quite planely (at the time) it is this way.

As for why not, I speculated a lot of plasuible reasons why not (none of which I know to be fact but they are reasonable). I also speculated ways to support what KY said and how it fits into the big picture, but those are superfluos and unecesasry in the face of a direct quote. They were more just expounding on the subject.

So yeah... I think the speculation on this subject is from your side in the form of a why not and I am going off the direct quote which isn't really speculating.

So we can both come up with things that are possible, but there are two pretty clear quotes from KY on the subject which are about the only things that aren't speculation.

I mean if the coach of the SF 49ers said "We only train our linebackers in our gym" and then said "No players other than linebackers train in our gym" you might speculate "well that doesn't make sense, they have a whole gym, there is no reason not to train the whole team in that gym so that means they must train the whole team in the gym" but that's just speculation. It's already been laid out pretty directly they just don't.

 

[Dravonic]

I think when they build the textures for damage via texture, they really do customize the tetures to each car or at least sets of cars... for instance the shape of a body panel needs to be brought into consideration when the damage is drawn on, if it doens't bend or countour properly it could look really bad. If you look at the existing scrape damage images they look very fitting fo the car... complex shaped bumpers look decent etc... I don't think it's like the computer just measures where they touch and apply a scarp in that contact area, mainly because that would look odd and if you look at the scrape damage, it has that same look as Forza where there is canned damage for a hit in a certain area even when the hit doesn't make sense for that level of damage (ie bumper to bumper results in scratch all the way up onto part of hood when nothing touched the top of the hood).

As for speculation, I think it's actually the other way.

I have KY quoted as saying only race cars can be damaged and then also saying damage only to some sets, other cars can't be damaged.

That's pretty clear.

Your take on it is that it seems like if they have a deformation engine, why not apply it to all cars instead of just some?

That seems to be the speculation to me. You are speculating a why not in the face of a quote from KY himself saying quite planely (at the time) it is this way.

As for why not, I speculated a lot of plasuible reasons why not (none of which I know to be fact but they are reasonable). I also speculated ways to support what KY said and how it fits into the big picture, but those are superfluos and unecesasry in the face of a direct quote. They were more just expounding on the subject.

So yeah... I think the speculation on this subject is from your side in the form of a why not and I am going off the direct quote which isn't really speculating.

So we can both come up with things that are possible, but there are two pretty clear quotes from KY on the subject which are about the only things that aren't speculation.

I mean if the coach of the SF 49ers said "We only train our linebackers in our gym" and then said "No players other than linebackers train in our gym" you might speculate "well that doesn't make sense, they have a whole gym, there is no reason not to train the whole team in that gym so that means they must train the whole team in the gym" but that's just speculation. It's already been laid out pretty directly they just don't.

The scrape damage is certainly detecting where the car hits something and applying a texture on it. It's safe to say so because if you look at the videos you can see that scrapes only appear where the car surface hit something. It's a technique very common on video games for years. For example, when you shoot the wall in a FPS game a "bullet hole texture" is applied to the wall where the bullet hit it.

But that's not important really. My main argument is, if they have mechanical damage and scrape damage that can be easily applied to every single car, it makes no sense to refrain from using it. They need to be stupid not to and stupid is the last thing PD is. And as it turns out, they will use it on all cars. Also, there are old tokyo showroom demos with scrapes and mechanical damage on cars like the 458 italia and the SLS AMG so it's not like I'm taking a wild guess here.


http://ps3.ign.com/articles/102/1028707p1.html:

Now, even though the Subaru looked like death and the Benz/Ferrari looked scratched up, all three drove like crap after the accidents. Whereas I could fly around in the rides beforehand, now I felt tension in the wheel when I'd be driving down a straightaway and cornering was tougher than ever.

From September 26, 2009.

As I said before, in your quote from Kaz he must have been referring to the special deformation damage that apparently only the race cars will have.

 

[Devedander]

The scrape damage is certainly detecting where the car hits something and applying a texture on it. It's safe to say so because if you look at the videos you can see that scrapes only appear where the car surface hit something. It's a technique very common on video games for years. For example, when you shoot the wall in a FPS game a "bullet hole texture" is applied to the wall where the bullet hit it.

Yes it detects where, but I think it's done using hitboxes and only generalized localization. Forza does this. Damage is applied to the appropriate area after a collision is detected. However the accuracy is only rough. For instance a small impact to the rear right side of a car can result in damage to the entire right rear quarter panel. While the damage roughly corresponds to the contact, how it looks is correct for the car showing that it's been custom drawn for that cars body.

I think we have all seen that Forza video where two cars don't even touch and magically damage pops up on both cars... that's an example of collision detection using hitboxes and not using the actual car model. The damage you see is a predrawn texture to show damage on that particular part of that car. These are not dynamically generated, they are actually done one off for each car (or maybe shared between similar cars but you get the idea). What I have seen in GT so far looks very much like they are doing the same.

If you look at the production car damage we have seen so far it looks good but it doesn't necessaily look accurate for the accident the car was in. It seems far more likely the damage location is roughly calculated as is the norm for collision detection. It's much less processor intensive to detect collisions between to boxes than two multi hundred thousand poly car bodies. I would expect better than just plane boxes but you get the idea.

But that's not important really. My main argument is, if they have mechanical damage and scrape damage that can be easily applied to every single car, it makes no sense to refrain from using it They need to be stupid not to and stupid is the last thing PD is.

Correction, they need to be stupid, or have othe reasons not to.

This brings us back into speculation as to why again but if you consider it only from a programing perspective, it's easy to say, if you have it, use it everywhere. But as I pointed out before there are many potential reasons not to outside of just programing (licensing, time conumption etc).

And as it turns out, they will use it on all cars. Also, there are old tokyo showroom demos with scrapes and mechanical damage on cars like the 458 italia and the SLS AMG so it's not like I'm taking a wild guess here.


http://ps3.ign.com/articles/102/1028707p1.html:



From September 26, 2009.

As I said before, in your quote from Kaz he must have been referring to the special deformation damage that apparently only the race cars will have.

Yes and my quotes were from early August... things changed. Damage is now being had across all cars but earlier on it was clearly said there wouldn't be, hence the conclusion that the original plan was exactly as kaz said, damage to some cars, not to others, then later changed to damage to all cars. I can't really see the flexibility necessary to conclude he was talking about special damage when he clearly said no damage twice in two different quotes.

I believe the quote you posted above is referring to the last outing at GC when GT was debuted with no damage. Now they are at TGS and damage is in.

This also fits the timeline and KYs quotes:

Before and including GC no damage for production cars, KY says as much. GC results in very negative press about no damage. By TGS KY and company are responding to the criticizm and working damage in.

Memory is kind of fuzzy now but I think that's more or less how it played out.

 

[Dravonic]

Yes it detects where, but I think it's done using hitboxes and only generalized localization. Forza does this. Damage is applied to the appropriate area after a collision is detected. However the accuracy is only rough. For instance a small impact to the rear right side of a car can result in damage to the entire right rear quarter panel. While the damage roughly corresponds to the contact, how it looks is correct for the car showing that it's been custom drawn for that cars body.

I think we have all seen that Forza video where two cars don't even touch and magically damage pops up on both cars... that's an example of collision detection using hitboxes and not using the actual car model. The damage you see is a predrawn texture to show damage on that particular part of that car. These are not dynamically generated, they are actually done one off for each car (or maybe shared between similar cars but you get the idea). What I have seen in GT so far looks very much like they are doing the same.

If you look at the production car damage we have seen so far it looks good but it doesn't necessaily look accurate for the accident the car was in. It seems far more likely the damage location is roughly calculated as is the norm for collision detection. It's much less processor intensive to detect collisions between to boxes than two multi hundred thousand poly car bodies. I would expect better than just plane boxes but you get the idea.



Correction, they need to be stupid, or have othe reasons not to.

This brings us back into speculation as to why again but if you consider it only from a programing perspective, it's easy to say, if you have it, use it everywhere. But as I pointed out before there are many potential reasons not to outside of just programing (licensing, time conumption etc).



Yes and my quotes were from early August... things changed. Damage is now being had across all cars but earlier on it was clearly said there wouldn't be, hence the conclusion that the original plan was exactly as kaz said, damage to some cars, not to others, then later changed to damage to all cars. I can't really see the flexibility necessary to conclude he was talking about special damage when he clearly said no damage twice in two different quotes.

I really doubt it's using hitboxes as crude as forza is. But that's not important so let's move on.

Let me see if I got this right. You were telling me that including damage on all cars would be a lengthy process, but in one month time PD got the negative damage feedback, decided they should do something about it, developed the damage you consider complicated, and included it in the demo? In one month? Now you're contradicting yourself.

 

[Devedander]

I really doubt it's using hitboxes as crude as forza is. But that's not important so let's move on.

I think the hitboxes will be more refined than F3s (I was surprised how gross those hitboxes were, reminds me of old FPS where the hitbox for the head was almost as big as the chest appeared to be).

Let me see if I got this right. You were telling me that including damage on all cars would be a lengthy process, but in one month time PD got the negative damage feedback, decided they should do something about it, developed the damage you consider complicated, and included it in the demo? In one month? Now you're contradicting yourself.

Not quite...

Here's how I figure it played out:

KY wants to put damage on the rally cars for sure and wants to do some pretty impressive damage. Knowing it's a pretty big undertaking he decides to not do it for all 1000+ cars but rather limit it somewhere. Also he has to consider licensing - race cars being smashed up is probably easier to negotiate than production cars. Also with production cars you may have different manufacturers giving different levels of permission, we know KY doesn't like to have things sloppy so he figures no damage for production cars, lets do damage on race cars and do it right.

The level of damage on race cars is pretty impressive and it's believable that doing it to 200 ish cars is much less time consuming that doing it to 1000+ cars baring other issues like licensing.

Later one it seems he is gonna get a lot of flack for releasing no damage on production cars so he says, fine lets do it. Quite possibly again they run up against license issues or something and they do cursory damage. Usually companies don't like their production cars SHOWN too badly beaten up but mechanical damage is probably not a hard one to get in. So we get mechanical damage on production cars too (this might have always been in) and a basic level of visual damage. This satisfies getting damage on the other 800ish cars done in a timely manner and also avoids licensing issues since the cars don't actually deform or have pieces falling off noteably so manufactures shouldn't gripe too much.

So I am not simultaneously saying that damage is both easy and fast and also time consuming and hard, I am saying there are different kinds of damage some of which are more time consuming than others and even the simpler one is not necessarily fast however is probably still faster than the more advanced one (again possible licensing or other issues I haven't touched on involved).

Basically this is what I figure is a rough timeline: around August 09 it was still damage to race cars only, again Kaz directly says this. Gamescon shows damage to the Subi (a race car).

Response to the lack of damage on production cars spurns KY and PD to reconsider their game. I expect this is when KY and PD make the call to take more time and add more features including damage.

By the time TGS comes rollilng around in late Sept KY is ready to announce more damage to satisfy the demands and about two weeks worth of new damage engine work is put into the TGS demo (I believe KY mentions this in an interview somewhere aroudn this time) and the quote you linked to above is the response to the difference between the previous GC build and the TGS build.

The games expected release date of xmas 09 (https://www.gtplanet.net/gran-turismo-5-march-2010-in-japan/ there is a recap of why that was expected) becomes march 2010 in Sept. I think up until now no release date had been officially announced as KY new he was testing the water with GC, if response was good, 09xmas releaes, if not 2010 release.

 

[Mr P510]

Yes....PD focusing too much on car models.
I fear this could be their downfall.

 

[Devedander]

Yes....PD focusing too much on car models.
I fear this could be their downfall.

Could be... eye candy is important but it does seem a balance is necessary too...

 

[TrievelA7X]

Has anyone said yet, "who cares, you're assuming it will ever be released"? If not, it's about that time.

 

[Dravonic]

I think the hitboxes will be more refined than F3s (I was surprised how gross those hitboxes were, reminds me of old FPS where the hitbox for the head was almost as big as the chest appeared to be).



Not quite...

Here's how I figure it played out:

KY wants to put damage on the rally cars for sure and wants to do some pretty impressive damage. Knowing it's a pretty big undertaking he decides to not do it for all 1000+ cars but rather limit it somewhere. Also he has to consider licensing - race cars being smashed up is probably easier to negotiate than production cars. Also with production cars you may have different manufacturers giving different levels of permission, we know KY doesn't like to have things sloppy so he figures no damage for production cars, lets do damage on race cars and do it right.

The level of damage on race cars is pretty impressive and it's believable that doing it to 200 ish cars is much less time consuming that doing it to 1000+ cars baring other issues like licensing.

Later one it seems he is gonna get a lot of flack for releasing no damage on production cars so he says, fine lets do it. Quite possibly again they run up against license issues or something and they do cursory damage. Usually companies don't like their production cars SHOWN too badly beaten up but mechanical damage is probably not a hard one to get in. So we get mechanical damage on production cars too (this might have always been in) and a basic level of visual damage. This satisfies getting damage on the other 800ish cars done in a timely manner and also avoids licensing issues since the cars don't actually deform or have pieces falling off noteably so manufactures shouldn't gripe too much.

So I am not simultaneously saying that damage is both easy and fast and also time consuming and hard, I am saying there are different kinds of damage some of which are more time consuming than others and even the simpler one is not necessarily fast however is probably still faster than the more advanced one (again possible licensing or other issues I haven't touched on involved).

Basically this is what I figure is a rough timeline: around August 09 it was still damage to race cars only, again Kaz directly says this. Gamescon shows damage to the Subi (a race car).

Response to the lack of damage on production cars spurns KY and PD to reconsider their game. I expect this is when KY and PD make the call to take more time and add more features including damage.

By the time TGS comes rollilng around in late Sept KY is ready to announce more damage to satisfy the demands and about two weeks worth of new damage engine work is put into the TGS demo (I believe KY mentions this in an interview somewhere aroudn this time) and the quote you linked to above is the response to the difference between the previous GC build and the TGS build.

The games expected release date of xmas 09 (https://www.gtplanet.net/gran-turismo-5-march-2010-in-japan/ there is a recap of why that was expected) becomes march 2010 in Sept. I think up until now no release date had been officially announced as KY new he was testing the water with GC, if response was good, 09xmas releaes, if not 2010 release.

Trust me, PD came up with an engine capable of deforming a car depending on direction and force of impact. Playing with hitboxes is childplay for these guys. The scrapes are not based on pre-textured hitboxes as you're suggesting. Even GTAIV doesn't use such a thing to scrape the cars. GT certainly won't.

This is the whole paragraph from your quote:

The next detail Kazunori touched on was regarding damage models. I can confirm that not every car in Gran Turismo 5 will sport damage models -- it will only be select sets. However, Kazunori has yet to decide on a finalized list of featured car types for damage support.

The last sentence shows how what we saw at GC was just a work in progress. It wasn't decided yet which cars would have damage. At some point, for whatever reason, Kazunori decided to include at least some kind of damage on all cars. According to the quote it was still in time for the decision but I can see how "damage on all cars" could be unexpected and push the game to a march release. But any further? As you can see they got it working in less than a month. And FM3 having any influence on his decision? That's just a supposition.

Something happened to push GT5 off the shelves for such a long time, but it doesn't look like production car damage is playing an important role here.

 

[shadynader]

Deformation will be on all cars, scratches will be on all cars, loss of doors and damagable cockpits wil be exclusive to select (race?) cars. Loss of other body parts like bumpers and wings on all cars is still unknown.

 

[Spagetti69]

Most people will turn it off anyway.

 

[Devedander]

Trust me, PD came up with an engine capable of deforming a car depending on direction and force of impact. Playing with hitboxes is childplay for these guys. The scrapes are not based on pre-textured hitboxes as you're suggesting. Even GTAIV doesn't use such a thing to scrape the cars. GT certainly won't.

I would have to put GTA IV back in for a closer look, but I think GTA does use hitboxes and some rather genericed damage. It's been a while since I played (sadly GTA IV was pretty boring to me) but I do seem to recall seeing the exact same damage on different occassions (pretty much impossible if you are really deforming based on true collision and not just generic camage based on general area of impact). Also I am pretty sure they do a hitbox style collision detection becuase I have had a physical effect on something by hitting it without actually visually touching it. (watch this video around 55 seconds, the cop car turns left then deflects off a pole but never actually touches it http://www.youtube.com/watch?v=qj0yu6Rgp5k there are probably better examples but I didn't have time to dig them up)

Even if GTA did use true model collision detection, their deformation engine is pretty unrealistic and the models are so much lower poly count that it's not the same (detecting collision for every polly in 2 400k poly cars is much more intesnive than hitbox collision with the hitboxes only having say 100 polys - however in GTA the cars are probably only a few hundred polys anyway so to make a lower poly box for collision might not save as much anyway).

The last sentence shows how what we saw at GC was just a work in progress. It wasn't decided yet which cars would have damage. At some point, for whatever reason, Kazunori decided to include at least some kind of damage on all cars.

Which is exactly what I have been saying all along... originally damage was only meant for some cars, then at some point (pretty much between GC and TGS) it was decided to broaden this to all cars.

According to the quote it was still in time for the decision but I can see how "damage on all cars" could be unexpected and push the game to a march release. But any further? As you can see they got it working in less than a month. And FM3 having any influence on his decision? That's just a supposition.

They have gotten something working in that short of a time (as I said I believe KY mentioned what we saw was only a few weeks worth of work and they were continuing to work on it), however I wouldn't be surprised if doing it well enough to satisfy KY takes longer than a few months. I don't know how any of us can really judge what is reasonable for devs in a certain time frame... it depends on who is doing it, how they are doing it, how well they want it done and how well they can fit it into existing code. I mean the entirety of GT5s dev time is a testimant to how you can't really guess someones dev time accuarately.

Something happened to push GT5 off the shelves for such a long time, but it doesn't look like production car damage is playing an important role here.

I have to disagree... I think it was a large part of why but I also said that they probably chose to do a lot of other things also which are taking some of the time. I would say it's analogous to pulling your teams car from the first race of the season to tweak the suspension. Before the second race you work on the suspension but you don't just leave the rest of your team sitting idely by so you get the aero, engine teams working on stuff too.

BTW https://www.gtplanet.net/brand-new-real-time-damage-coming-to-gran-turismo-5/ I am not the only one who thinks that damage is a large part of the push to 2010

reinforcing my personal suspicion that negative reactions to the damage shown off at GamesCom was the primary decision for a 2010 release.

There was a lot of negative talk after GC about no damage to production cars both from gamers and from magazines/game sites. Then shortly after damage is in for all cars, then in december KY talks about how damage is in mostly due to the demans of American gamers.

Again, you have to do a little reading between the lines and putting together of quotes, but it paints a pretty solid story I think...

 

[craste]

Listen guys KY has said before that he doesn't play other peoples car games so when he created the damage model we saw at gamescon he thought GT5 was the first to show damage and that we would think it was amazing - but no, we saw it for was it was - A step backwards.

That has got to be the reason it was so dire.

Tip to KY - NEVER take your eye off the competition.

 

[Devedander]

Listen guys KY has said before that he doesn't play other peoples car games so when he created the damage model we saw at gamescon he thought GT5 was the first to show damage and that we would think it was amazing - but no, we saw it for was it was - A step backwards.

That has got to be the reason it was so dire.

Tip to KY - NEVER take your eye off the competition.

I don't think he doesn't play them entirely, but I think he claimed he doesn't concern himself with them very much.

KY: To tell you the truth, when I’m interested in other games, it’s really difficult to not look at them with the creator’s eyes. I notice some details and I think to myself: “oh, they must have really hard worked on this point” or “hmm, they must have lacked time for this very aspect”. I can’t prevent myself from putting me in their place.

However I can't really beleive he doesn't keep an eye on the competition... he probably doesn't spend hours playtesting himself, but I think he is kept well aware of what's going on out there by his team.

It's possible he is in some kind of bubble himself though, and if so, I agree that's not a good way to go about things... he isn't the only chair maker anymore

 

[P3nT4gR4m]

I never put to much stock in that quote. Struck me as bluster and bravado. If he does ignore the competition then he's an idiot and he doesn't strike me as being that. A bit tunnel visioned perhaps but not an idiot.