Vision Gran Turismo discussion- M-B AMG Vision unveiled

[LMSCorvetteGT2]

Well I think it depends on what ever contract PD has with the car manufacturer.The cad model could be extremely high poly or incredibly tessellated or have no interior,all of which would make the car itself not ready to be put in game .If they do use cad models direct from the manufacturer, they more than likely use it as a reference if it still takes that long to model a car.As for the length, I personally think ( and hope ) its an estimate lol.Or atleast not only include the modeling process but also the creating and assigning of the joints ( animations ), car research and programing.I can see it taking close to 6 months for some cars though.But 6 months for say a gordon murray rocket car is insanely slow lol.

Even from scratch it shouldn't take 6 months it's not like the CAD models they get are from decade old programs. Also I'm sure systems like Catia or Solid Work cad programs are probably used more by european car makes and Japanese. Maybe if we had a full understanding of what a finished model is by their terms...

 

[Griffith500]

CAD files don't use polygons, though. At least, not in the same way. You still have to create a detailed mesh from those parametric shapes.

I think the 6 month figure includes everything: mesh, textures, physics, mods, tuning, sound (that we've not heard yet) etc.

 

[LMSCorvetteGT2]

CAD files don't use polygons, though. At least, not in the same way. You still have to create a detailed mesh from those parametric shapes.

I think the 6 month figure includes everything: mesh, textures, physics, mods, tuning, sound (that we've not heard yet) etc.

Yeah I would think the same thing, as far as what is considered a complete model to them. To some of us I think it is just the actual car exterior and interior as it sits, maybe not though. To do the building of the car itself in CAD, doesn't take six months though

 

[ildd]

I think the new concepts will (mostly) resemble the GT by Citroën concept. I think that 13/18 will have an actual possibility of making it to the streets... and by streets I mean a prototype with a lesser engine and will never actually go on a street

 

[Griffith500]

Yeah I would think the same thing, as far as what is considered a complete model to them. To some of us I think it is just the actual car exterior and interior as it sits, maybe not though. To do the building of the car itself in CAD, doesn't take six months though

In CAD? No, realistically, for a modern car, that'd take many man years once you factor in R&D etc.

For one of these concept things, you could knock something up in a day or two (each for the interior and exterior) - refine it over a few weeks (each), maybe. Probably longer for something of real quality.

PD still then have to take that and convert it to their format, and that won't be changed much by the relative lack of time spent by the manufacturer compared with "real" cars.

 

[GOTMAXPOWER]

These Gran Turismo concepts developed with manufactures kind of remind me of WRC Extreme on PS2, anybody else remember these kind of beauties?

http://www.igcd.net/vehicle.php?id=42643&PHPSESSID=c18ff1...&width=1366

 

[LMSCorvetteGT2]

In CAD? No, realistically, for a modern car, that'd take many man years once you factor in R&D etc.

But this isn't a car in an R&D phase it's current cars that have already hit production and have been made, and the vision ones are being worked on by manufactures and PD. So I don't see how, I've done tons of CAD and why would colleges make you work in a team to recreate a CAD of an existing model with a group if it was a daunting as "many man years".

For one of these concept things, you could knock something up in a day or two (each for the interior and exterior) - refine it over a few weeks (each), maybe. Probably longer for something of real quality.

PD still then have to take that and convert it to their format, and that won't be changed much by the relative lack of time spent by the manufacturer compared with "real" cars.

Exactly my point realistically with the ability and power PD along with major automotive groups has, I see it only taking a third or less of the six months to build such cars. Also you'd think after working on the PS3 with two GT games PD would have found a faster way to do this process in the almost eight years of the PS3 life span so far.

 

[trustjab]

Maybe if you look at it from another point of view, these cars don't seem like a waste of time but rather a marketing tool that allows PD to get these fruitful relationships with these auto makers. By including things like the BMW concept, this pleases BMW and solidifies a working relationship between the two. This could mean future models or new models of BMW's appearing in GT6 thanks to the partnership. Theres always two sides to the story. I believe this is showing the willingness of cooperation with car manufacturers and even boutique companies in the auto industry and PD. This can only be a good thing, even if you don't like the Nikemobile or the Air Jordan shoe-car.

 

[Inferno251]

It's interesting how they've coined their 'Vision Gran Turismo' term-now and previously. The original 'Vision Gran Turismo' was about turning the GT series into a visual and physically realistic game. Now, with that being more of the case with GT6, the new 'Vision Gran Turismo' seems to be about using GT as a means to implement concepts from the virtual world to the real world.

I'm curious on how the Nike Two will have any relation to the Nike One from GT4. The driver seems to be wearing a odd looking helmet give the idea of the "Spark Suit" being used like the One was.

I really hope they do bring back the Nike One, and make it compete against other cars in GT mode this time, because that was a great car to drive. If the Nike Two doesn't make it upon release, PD could do a 'Nike Pack' with both cars maybe?

 

[nascarlemanslmp]

I hope they do bring back the nike car. Why did they remove that? Also the two old cars
from gt4.

 

[Shirrako]

Most exciting news from GT6 for me, always wanted those far in the future concepts to be in a GT game, can't wait to try out the future of cars in GT6.

 

[MrYasser]

I think some of them are already finished

 

[Duphman]

http://www.youtube.com/watch?v=zV1abWVXGn4

Just watched the Gran Turismo conference headed by Kaz.

The 6C Biposto seems to be already finished, was designed by the same team behind the 4C and will be position performance wise between the 4C and the 8C.

The Aston Martin and Alpine have not been completely finished.

The Audi won't just be a exterior design test, but will use some kind of mechanism Audi hasn't used on their cars before.

The BMW Vision Gran Turismo is supposed to be an M Power halo car.

GM Design has multiple design teams working in their Vision Gran Turismo car, which could imply that it will be the first car to actually have a physical model to be shown at a car show (possibly Detroit).

Honda is working with Gran Turismo with both the design of the NSX NA3 (ala, the GT-R CBA-R35) and will be providing a separate "more extreme" model for Vision Gran Turismo.

Tinker Hatfield is in charge of the design behind the JORDAN Vision Gran Turismo.

Several hundred designers have worked on the Mercedes-Benz AMG Vision Gran Turismo's design. A lot of man hours have been put into it.

Yamauchi says the Nissan Vision Gran Turismo will "possibly" be a future vision of the Nissan GT-R.

Infiniti is designing their own separate model, with no input from Nissan.

He also said the Peugeot Vision Gran Turismo's side silhouette is subject to change since it is still in the design process.

The Volkswagen GTI Vision Gran Turismo will be a future interpretation of what the GTI is.

All of the other design studios gave very limited information about what they are creating. These cars will be unveiled throughout the year as they are finished.

When asked about liveries Yamauchi specifically stated that "some additional features" will be shown at the Tokyo Game Show.

 

[TokoTurismo]

^ Wow can't wait to see them, I feel all those cars are gonna rock. 👍 It's almost like the car makers are doing a school project for Kaz and Gran Turismo.

 

[Imari]

http://www.youtube.com/watch?v=zV1abWVXGn4

Just watched the Gran Turismo conference headed by Kaz.

The things I noticed from that conference were:

-The Vision GT cars will be released over the course of a year as DLC, one or two a month.
-They haven't decided whether they're going to charge money for them yet.

Watch from about 13 minutes to 15 minutes for these parts.

 

[symustafa1996]

Maybe GT is no longer about driving real cars against each other just for the sake of driving them, but kind of a vision of what the automobile is and what it could be.

 

[Griffith500]

But this isn't a car in an R&D phase it's current cars that have already hit production and have been made, and the vision ones are being worked on by manufactures and PD. So I don't see how, I've done tons of CAD and why would colleges make you work in a team to recreate a CAD of an existing model with a group if it was a daunting as "many man years".



Exactly my point realistically with the ability and power PD along with major automotive groups has, I see it only taking a third or less of the six months to build such cars. Also you'd think after working on the PS3 with two GT games PD would have found a faster way to do this process in the almost eight years of the PS3 life span so far.

I think you're missing the point. PD doesn't make the CAD data. The manufacturer does. Regardless of what data is in that file, or where it comes from, or whether it has been rigorously tested in simulations and the real world, it doesn't change that PD then have to convert it to their own format and collate and input the right data for that car. PD don't say that cars take 4 years to put into the game, i.e. they don't add the manufacturer's work onto their own.

If anything, that's likely to be harder with these concepts than with a real car, because you can't measure a lot of it (although "simulations" will give you numbers that are close). That measurement step is important, because no physics engine works from absolute first principles; there is a great deal of aggregation of real-world effects bundled up into a single number e.g. "lateral rigidity" of the chassis, something like that. What you'll get might as well be guesses in some cases - remember the RX-8 in GT3?

In that sense, these concepts would take just as long as real cars for PD to put into the game. I have no idea where this factor of a third comes from, at all.

(I've done a little bit of CAD, but only in a simulation framework. The iteration required is staggering. It wouldn't be daunting for a car manufacturer because that work is spread across a very large team, so that man years become months in real time, but I'm sure that's obvious. There's also a difference between design, engineering and research uses of CAD, too.)

By the way, simply pulling a mesh out of the CAD data to show off isn't difficult. Making it to a specific spec (performance, cleanness etc.) and adding all the GT game features don't have to have been done to show them off.
Then again, the RX-8(s) in GT3 were really rough around the edges, so I suppose they don't actually have to put the time in at all.

 

[Panoz]

Reading about the Mercedes, this project must be pretty big for M-B with the amount designers used on the car.

 

[diegorborges]

I think you're missing the point. PD doesn't make the CAD data. The manufacturer does. Regardless of what data is in that file, or where it comes from, or whether it has been rigorously tested in simulations and the real world, it doesn't change that PD then have to convert it to their own format and collate and input the right data for that car. PD don't say that cars take 4 years to put into the game, i.e. they don't add the manufacturer's work onto their own.

If anything, that's likely to be harder with these concepts than with a real car, because you can't measure a lot of it (although "simulations" will give you numbers that are close). That measurement step is important, because no physics engine works from absolute first principles; there is a great deal of aggregation of real-world effects bundled up into a single number e.g. "lateral rigidity" of the chassis, something like that. What you'll get might as well be guesses in some cases - remember the RX-8 in GT3?

In that sense, these concepts would take just as long as real cars for PD to put into the game. I have no idea where this factor of a third comes from, at all.

(I've done a little bit of CAD, but only in a simulation framework. The iteration required is staggering. It wouldn't be daunting for a car manufacturer because that work is spread across a very large team, so that man years become months in real time, but I'm sure that's obvious. There's also a difference between design, engineering and research uses of CAD, too.)

By the way, simply pulling a mesh out of the CAD data to show off isn't difficult. Making it to a specific spec (performance, cleanness etc.) and adding all the GT game features don't have to have been done to show them off.
Then again, the RX-8(s) in GT3 were really rough around the edges, so I suppose they don't actually have to put the time in at all.

And just to add some technical bits from my short experience with 3d modelling and the stuff about car design process I usually read (I'm a car design nut), manufacturers usually produce very complex 3d CAD data, very detailed, and they usually are in different format (splines, nurbs curves) which are different than what the game industry usually works (polygons), so it's a huge work to make a proper conversion of those meshes, then just after that, there's cleaning and optimization to run on an old system such as the PS3, then there's the LOD's (lower level of detail versions of the same cars) and/or adaptive tesselation tests and more and more optimization.

All those with details, textures, interiors, etc.

One simple example: I'm used to handle 3d files from Forza because I mod some games on PC, some guys have made tools to export those meshes. After importing, you need to clean, assign textures, adjust lighting, etc for each game. They come in 5-6 different LOD (level of detail meshes), with different sets of textures. It's a lot of work and that's already from stuff that is READY in another game. And game modders doesn't have to respect any standards or quality control. Imagine all that done from a japanese company known by it's quality. You can say everything about Gran Turismo models, but you'll never found something out of the right proportions (like some other games do sometimes)

I can see where all those man hours may go. It's not just model and throw them ingame.

Even if they develop a "mesh converter" (already exist for several formats) and tools for the process, it is never perfect and always need serious cleaning and optimization.

Edit: just seen that Griffith already told a lot of stuff I wrote...

 

[Johnnypenso]

There is definitely a trade off between adding concept cars to the game and real cars and I'm pretty certain it's a 1:1 tradeoff. Like it or not, PD is really pumping up the whole cross promotion thing with auto manufacturers and in the larger scheme of things, even with the tradeoff against having more real cars in the game, I think that level of collaboration is good for the series in the bigger picture. Hopefully in the future it translates into more access to new models not even released to the public, more access to classics held in corporate or other museums, better access to sound sampling data etc. etc. etc.

 

[LMSCorvetteGT2]

I think you're missing the point. PD doesn't make the CAD data. The manufacturer does. Regardless of what data is in that file, or where it comes from, or whether it has been rigorously tested in simulations and the real world, it doesn't change that PD then have to convert it to their own format and collate and input the right data for that car. PD don't say that cars take 4 years to put into the game, i.e. they don't add the manufacturer's work onto their own.

If anything, that's likely to be harder with these concepts than with a real car, because you can't measure a lot of it (although "simulations" will give you numbers that are close). That measurement step is important, because no physics engine works from absolute first principles; there is a great deal of aggregation of real-world effects bundled up into a single number e.g. "lateral rigidity" of the chassis, something like that. What you'll get might as well be guesses in some cases - remember the RX-8 in GT3?

In that sense, these concepts would take just as long as real cars for PD to put into the game. I have no idea where this factor of a third comes from, at all.

(I've done a little bit of CAD, but only in a simulation framework. The iteration required is staggering. It wouldn't be daunting for a car manufacturer because that work is spread across a very large team, so that man years become months in real time, but I'm sure that's obvious. There's also a difference between design, engineering and research uses of CAD, too.)

By the way, simply pulling a mesh out of the CAD data to show off isn't difficult. Making it to a specific spec (performance, cleanness etc.) and adding all the GT game features don't have to have been done to show them off.
Then again, the RX-8(s) in GT3 were really rough around the edges, so I suppose they don't actually have to put the time in at all.

That is what I'm saying to make the mesh and textures of just a siting car that you'd see rotating in GT garage shouldn't take that much time and that is where the building of the car and details come from. Now the physics engine portion I agree with you on is obviously going to take much longer and be the difficult aspect unless the manufactures provide their own simulations and so on that are rough but can be built on.

In my last post I was not making any regards to physics I think you weren't getting where I was coming from. The general idea is building a car from scratch or detailing similar models to it shouldn't take six months, especially when you similar physics models that get plugged in to it.

 

[Jahgee]

Jordan car looks like a light cycle

Nike car looks like It

 

[Griffith500]

That is what I'm saying to make the mesh and textures of just a siting car that you'd see rotating in GT garage shouldn't take that much time and that is where the building of the car and details come from. Now the physics engine portion I agree with you on is obviously going to take much longer and be the difficult aspect unless the manufactures provide their own simulations and so on that are rough but can be built on.

In my last post I was not making any regards to physics I think you weren't getting where I was coming from. The general idea is building a car from scratch or detailing similar models to it shouldn't take six months, especially when you similar physics models that get plugged in to it.

But you previously agreed that the six month figure was for everything, not just modeling. I actually don't know what you're saying.

Regardless of whether the car is real or not, to make it to PD's spec in their own medium takes the same amount of time, in principle - it's the exact same process. You get the data (CAD, drawings, photographs - or renders), and you convert it to a polygon mesh with multiple textured layers mapped on top. Obviously the time taken would vary from car to car, and I'd bet there are some real cars that were / would be quicker than some of those concepts, at least in the modeling (mesh) phase, and certainly in the texturing phase (liveries).

Presumably, these cars are going to be driveable in the game (and they all use the same model, just different input data), and the only reason I mentioned pulling a mesh out of the CAD data for display purposes (covered by a cloth simulation, for the most part) was simply to offer a counterpoint to those thinking these cars are ready and will be held back for DLC (which might possibly be free anyway...)

 

[oink83]

No, they don't have the luxury. As noted, they're a small team. If they put in 15 of these concepts, that's 15 other cars that we're not seeing.

In theory, I approve of them adding new and unique cars from the cutting edge of the automotive industry. If they weren't so strapped for time that two thirds of the cars in their game weren't low quality PS2 models, and if they'd actually managed to get a reasonable selection of the interesting cars from the ten years since GT4 came out.

It's about having a decent foundation of real cars in the game first, before getting all excited about concepts. GT has a really good selection of cars up to 2004, and then it sort of falls apart.

Of course, the other 185 new cars may be perfectly selected to fill that void. But their approach so far doesn't give much hope for that (*cough*Schwimmwagen*cough*).

But PD's mission statement has always been to bring the feeling and emotion of driving experiences to people that otherwise don't get the chance. It's what sets GT games apart from every other racing game. They aim to bring diverse and different driving experiences that most people will never get to experience in real life, rather than just the exact same lineup of cars that every other driving game offers.

These concept cars (along with examples like the schwimmwagen as you pointed out) are an integral part of the Gran Turismo experience, and is part of what makes it a GT game.

 

[Raggadish]

They aim to bring diverse and different driving experiences that most people will never get to experience in real life, rather than just the exact same lineup of cars that every other driving game offers.

These concept cars (along with examples like the schwimmwagen as you pointed out) are an integral part of the Gran Turismo experience, and is part of what makes it a GT game.

Well said Mr. Oink! 👍

 

[Imari]

But PD's mission statement has always been to bring the feeling and emotion of driving experiences to people that otherwise don't get the chance.

Show me that this mission statement is not just something that you feel about the game personally.

Frankly, I'd have thought that every simulation based game would be trying to bring the feeling and emotion of driving experiences. That's the whole point. See the NFS:Shift and Shift 2 marketing for examples of a company taking "feeling" to the nth degree.

 

[another_jakhole]

Show me that this mission statement is not just something that you feel about the game personally.

Frankly, I'd have thought that every simulation based game would be trying to bring the feeling and emotion of driving experiences. That's the whole point. See the NFS:Shift and Shift 2 marketing for examples of a company taking "feeling" to the nth degree.

It's not written down in stone, but Kaz might have said some form of it before.

These concept cars (along with examples like the schwimmwagen as you pointed out) are an integral part of the Gran Turismo experience, and is part of what makes it a GT game.

Tank car, is it really a car that can be seriously raced? Yet, people do it, just like the WWII cars. The "Sambabus" is another one and was a part of the TGTT Special Event. The NIKE One from GT4,
the Suzuki

the Citroen

the Red Bull X2k's.

Why add these cars if car manufacturers really won't gain from it? I suppose you think most/all of the cars I named off are pointless? Kaz and PD are just unique in that way. Oh yeah, all those race cars that will never see another track/race day, all those classic cars that we will likely never get to ever see in person. I mean, one can draw their own conclusions from them. It may very well not be PD's "official mission statement".

Oink83 went on to explain what he meant "mission statement" means.

They aim to bring diverse and different driving experiences that most people will never get to experience in real life, rather than just the exact same lineup of cars that every other driving game offers.

These concept cars (along with examples like the schwimmwagen as you pointed out) are an integral part of the Gran Turismo experience, and is part of what makes it a GT game.

Show me that this mission statement is not just something that you feel about the game personally.

Frankly, I'd have thought that every simulation based game would be trying to bring the feeling and emotion of driving experiences. That's the whole point. See the NFS:Shift and Shift 2 marketing for examples of a company taking "feeling" to the nth degree.

Is his opinion wrong, or are you trying to make him look dumb by asking to see proof that it's Kaz's/PD's "mission statement"? I think you're just taking the definition of "mission statement" literally, then dismissing what oink83 described what it means to him and no doubt, to others too. Shift isn't what oink83 was describing.

 

[fporto_br]

The VW design reminds me so much of the Design Vision GTI presented at Wörthersee...

 

[Stewie Griffin]

The VW design reminds me so much of the Design Vision GTI presented at Wörthersee...

That's a nice car

 

[Imari]

Is his opinion wrong, or are you trying to make him look dumb by asking to see proof that it's Kaz's/PD's "mission statement"? I think you're just taking the definition of "mission statement" literally, then dismissing what oink83 described what it means to him and no doubt, to others too. Shift isn't what oink83 was describing.

I wanted to see if it was actually a concrete thing, or just a general opinion.

As a general opinion, it seriously applies to every simulation game. All devs will talk about how they want to bring the player the real experience, get them to feel the excitement the real drivers feel, and so on. It's not unique to GT.

And Shift is exactly what he was describing. It was badly, awfully, terribly done. But the game was designed with a bunch of features whose primary function was to give greater immersion and feeling to the driving. The tunnel vision and rattles at speed. The black-and-white wobble-vision when you hit something hard. The helmet cam that looks through the apex of the corner like a real driver would. These things didn't work, but the idea was to give people a better feeling of what it was like to actually be in a car giving it 100%.

Again, Shift was badly done. It's a bad game. But it's a perfect example of another game that tries to "bring the feeling and emotion of driving experiences to people that otherwise don't get the chance".