Promised Functionality vs Realistic Timelines in GT6

  • Thread starter Squanto
  • 51 comments
  • 2,887 views
394
Netherlands
NL
Morphisorius
PD does generally promise things a bit too optimistically, but they do tend to communicate some sort of feature plans every odd month, like we saw last with the Zahara track teaser. That there has been no news about it since, is telling. It tells that there is something going on that's interfering severely.

The thing is, there are a few major hitches in the game that is GT6, which were not anticipated by PD. Yet it's absolutely essential that they get fixed before the more complex online-related features get released.

I am, of course, talking about both the tire/pit stop bug and the hacking. With the tire/ pit stop bug as it is, any plans of endurance racing and more complex community features would be severely affected.

The current situation with hacking is even worse, especially when there is a very real risk that hacked assets become easily accessible to the general public.


My personal guess is, that both these problems relate to the very core of the game. This means that it would be VERY bad practise to just ignore them at first and work on more complex additions based on that very same core. It could create much more problems, which would be harder to find later on.

There is not a developer in the world that likes to lie to its customers on purpose, but major problems can and do occur; this is certainly not limited to PD. I still have faith that these things will eventually be resolved. In part, because any other outcome would not speak well for PD at all.
 
Last edited:
Zahara isn't released because it lacks premium water and according to PD perfection standards that would not look good enough.

Joking aside. PS3 is not the easiest plattform to write code for. Almost all PS3 games use 720p resolution. Not many run at 60fps or close. Not that many have as detailed models. Not many have as good light model. Compare driving physics and no one comes close on the ps3 (imo).

So what would set aside a GT game from any other game. One word that comes to mind.. optimization (endless code optimizing). Many would settle for good enough, but what if that isn't good enough. Well that's where optimization comes into play. To optimize something that would already be considered perfect takes time.. sometimes considerable amounts.

So even if it maybe doesn't make sense why everything is going at slow pace... thinking about it... it does make a little sense.
 

Squanto

GT1-2-3-4-5-6 Enthusiast
Premium
641
Netherlands
Dordrecht
Squanto666
PD does generally promise things a bit too optimistically, but they do tend to communicate some sort of feature plans every odd month, like we saw last with the Zahara track teaser. That there has been no news about it since, is telling. It tells that there is something going on that's interfering severely.

The thing is, there are a few major hitches in the game that is GT6, which were not anticipated by PD. Yet it's absolutely essential that they get fixed before the more complex online-related features get released.

I am, of course, talking about both the tire/pit stop bug and the hacking. With the tire/ pit stop bug as it is, any plans of endurance racing and more complex community features would be severely affected.

The current situation with hacking is even worse, especially when there is a very real risk that hacked assets become easily accessible to the general public.


My personal guess is, that both these problems relate to the very core of the game. This means that it would be VERY bad practise to just ignore them at first and work on more complex additions based on that very same core. It could create much more problems, which would be harder to find later on.

There is not a developer in the world that likes lie to its customers on purpose, but major problems can and do occur; this is certainly not limited to PD. I still have faith that these things will eventually be resolved. In part, because any other outcome would not speak well for PD at all.

I'm beginning to see the outlines of this problem .... just imagine the effect it will have if they communicate about these issues ... It would fuel the power of the hackers ... who get feedback on their actions, which is something we all shouldn't want ... I like a free of hacks/cheats game ... So its good they take their time to really address the problems.

Also I still get the feeling that the "PSN" is introducing new cross-game community features, so I think they are also delayed by these efforts which are from "inside". Gradually I see things change through the PS3, PSVita, mobile app and PS4 environment ... Still find it strange to not see my friends on PS3 online while I do see them when I play on my PS4 ... depending on which platform they play on ... they recently extended the use of community features on the Vita ... and also on PS4 functionality is extending. What has my concern is: do they think its fruitful to extend the community features on the PS3 platform or will they just leave it up for the game developers here .... and not regulate a cross PS-platform standard community ...
They announced for may the shut down of community services .. this could well be the starting point for new community features ... but in which form is still unclear .... maybe someone knows more facts? ...
I would love to get a better picture of what is going on in this part of the game ...
 
130
Netherlands
Rainbow Bar&Grill
Rrralphster
"I think a lot of us have false expectations of the software (GT) development process because PD clearly doesn't bring any clear time lines."


I don't want strict timelines (impossible in the world of programming).

I would like to see/read/hear some communication though.

"For next patch, we are working to fix ABC, and we will try to add XYZ"

Next comms:

"Sorry guys, in the last update we talked about working to fix ABC, but we are having troubles getting C to work.
Progress on XYZ is still going strong"


Pretty simple....




Also, still no word about the status of some the current features (like improved AI... (my biggest disappointment for GT6))


TALK TO US PD....
 
155
Germany
Jurrmunney
talkin_2_me
I would like to see/read/hear some communication though.

"For next patch, we are working to fix ABC, and we will try to add XYZ"

Next comms:

"Sorry guys, in the last update we talked about working to fix ABC, but we are having troubles getting C to work.
Progress on XYZ is still going strong"
As much as I would like that sort of communication, I fear at this point it would be too late. There would still be a ********* because a lot of people (me included) already think PD is out of touch with its community and focusing to much on the wrong stuff.
Reactions would probably read a lot like:

"WTF are they working to fix ABC and add XYZ when they should focus on fixing DEF and adding UVW!"
 

Squanto

GT1-2-3-4-5-6 Enthusiast
Premium
641
Netherlands
Dordrecht
Squanto666
As much as I would like that sort of communication, I fear at this point it would be too late. There would still be a ********* because a lot of people (me included) already think PD is out of touch with its community and focusing to much on the wrong stuff.
Reactions would probably read a lot like:

"WTF are they working to fix ABC and add XYZ when they should focus on fixing DEF and adding UVW!"
Maybe I have to agree with that ... Internet nowadays is flooded with people who only tend to give negative comments ... If they have something positive to say they are silent ... but negative thoughts are scattered without any limitations ... I find that sad really ... sure I do my share of complaining but still try to address the positive sides of things as well ...
 

MustangManiac

TeamJaguarUSA
Premium
2,121
United States
Chicago
mustang622006
Joking aside. PS3 is not the easiest plattform to write code for.

Sorry, but I have to call bull on this, I am so tired of this excuse. The PS3 came out 8 years ago which means the dev kits were probably out at least 9 years ago. Lets not forget that Sony was the primary developer of the cell chip and that PD is a Sony developer either. If they haven't figured out how to code for it yet they never will.
 

Johnnypenso

Well known double poster
Premium
28,466
Canada
Windsor, Ontario, Canada
Johnnypenso
I'm sorry, I don't get the link between hybriding and things like the pit stop bug, camber bug or whatever, and not being able to release the Community Features or any other updates. Online is practically dead already, I fail to see how adding features to the game that allow us to create race nights and series and promote them directly to players will negatively impact the game because an infinitesimally small number of players have made some hybrids. Most of the players interested in sim racing have already left the game anyway judging by the quantity and types of lobbies online, so it's sort of too little too late no matter what happens now.

On top of that, hybrids have been in the GT series forever. If that really is the issue, don't you think they would have solved it a long time ago? Won't a relatively simple checksumm algorithm solve this anyway? Didn't GT5 eventually eliminate hybrids online that were created through game save hacking?

Withholding features because of what a tiny minority of people do would be incredibly short-sighted. Hybriding in GT6 will never be nearly as prevelant as it was in GT5 because the trading feature is gone, and currently hybriding is done through hex-editing with a jailbroken PS3. If there were more than 100 hybriders in the entire world I'd be very surprised.
 

Squanto

GT1-2-3-4-5-6 Enthusiast
Premium
641
Netherlands
Dordrecht
Squanto666
I'm sorry, I don't get the link between hybriding and things like the pit stop bug, camber bug or whatever, and not being able to release the Community Features or any other updates. Online is practically dead already, I fail to see how adding features to the game that allow us to create race nights and series and promote them directly to players will negatively impact the game because an infinitesimally small number of players have made some hybrids. Most of the players interested in sim racing have already left the game anyway judging by the quantity and types of lobbies online, so it's sort of too little too late no matter what happens now.

On top of that, hybrids have been in the GT series forever. If that really is the issue, don't you think they would have solved it a long time ago? Won't a relatively simple checksumm algorithm solve this anyway? Didn't GT5 eventually eliminate hybrids online that were created through game save hacking?

Withholding features because of what a tiny minority of people do would be incredibly short-sighted. Hybriding in GT6 will never be nearly as prevelant as it was in GT5 because the trading feature is gone, and currently hybriding is done through hex-editing with a jailbroken PS3. If there were more than 100 hybriders in the entire world I'd be very surprised.
I like to understand but I just don't know enough about this subject, I guess you mean the hybrids are the main issue in regard to the hacking? Or am I now totally on the wrong track?
 

Johnnypenso

Well known double poster
Premium
28,466
Canada
Windsor, Ontario, Canada
Johnnypenso
I like to understand but I just don't know enough about this subject, I guess you mean the hybrids are the main issue in regard to the hacking? Or am I now totally on the wrong track?
People use "hybrid" and "hacking" to denote the same thing on GTP. Technically "hacking" is hacking into the game save or hex coding of the game where you can do all kinds of stuff, one of which is create "hybrids" by swapping parts around, multiplying stuff etc. Hybrids are essentially cars that are no longer within the programmed limits of the game for any reason. In GT5 you could do this with a piece of software someone created and edit the gamesave, in GT6 you must hex edit which is modifying the actual game code directly and have a jailbroken PS3 to do it, which means it's far less widespread, especially since there is no trading in GT6.

While I don't want to see any TT's ruined by hybrids and it is something PD has to take care of, I think the vast majority of us could play online on a regular basis in circuit racing lobbies and never see a hybrid. I know I never have. I just don't believe that hybrids have anything to do with features not being released for the game, I think it's just a convenient scapegoat.
 
11,231
United States
Chicago, iLL
GTP_Nismo
Community features should've been the very first big update after release.
I don't play online so it doesn't really concern me that much but that feature should've been out before 2014 hit.
To me the most important promised aspects were RACING MODS(some might say questionable) and deeper customization. Neither were delivered and my hope for the game has been diminishing ever since.
January rolled around and we were promised a new track. That never happened and we haven't heard a peep from PD as to why that is. Same goes for the VGT BMW that was supposed to happen at Geneva.
In my opinion the future updates/features PD promised don't seem too ambitious so why can't they deliver?
 

Squanto

GT1-2-3-4-5-6 Enthusiast
Premium
641
Netherlands
Dordrecht
Squanto666
People use "hybrid" and "hacking" to denote the same thing on GTP. Technically "hacking" is hacking into the game save or hex coding of the game where you can do all kinds of stuff, one of which is create "hybrids" by swapping parts around, multiplying stuff etc. Hybrids are essentially cars that are no longer within the programmed limits of the game for any reason. In GT5 you could do this with a piece of software someone created and edit the gamesave, in GT6 you must hex edit which is modifying the actual game code directly and have a jailbroken PS3 to do it, which means it's far less widespread, especially since there is no trading in GT6.

While I don't want to see any TT's ruined by hybrids and it is something PD has to take care of, I think the vast majority of us could play online on a regular basis in circuit racing lobbies and never see a hybrid. I know I never have. I just don't believe that hybrids have anything to do with features not being released for the game, I think it's just a convenient scapegoat.
Thanks for explaining I misunderstood this completely 👍
 
Sorry, but I have to call bull on this, I am so tired of this excuse. The PS3 came out 8 years ago which means the dev kits were probably out at least 9 years ago. Lets not forget that Sony was the primary developer of the cell chip and that PD is a Sony developer either. If they haven't figured out how to code for it yet they never will.

If there's is any developer that would squeeze every bit of juice out of the PS3 that would be PD. Like I said there are not many that can provide almost 60fps at almost full hd resolution for the PS3. How would you be able to accomplish that if you don't know how to write code? Maybe you have heard of many developers that say that it is very easy to code PS3 games.. I on the contrary have not.

I don't know why many see GT6 as GT 5.5. GT5 was a technical masterpiece, yes, but they have improved a lot of core features (driving physics, hdr lightning, resolution, tessellation and so on) since then. How do you implement new features while still maintaining the processing power - ps3 has limited power - and GT5 was already using up all of it. You have to free resources to be able to, which is not an easy task and takes time.

Yes, everybody is complaining about missing content and some core bugs. But how many are complaining about the driving physics of GT6. Not the majority anyway. Waiting for new updates is tiresome and boring so I play other games in the meantime till PD decide to release something that interests me, probably around may.
 
12,261
Australia
Adelaide
GTP_Imari
If there's is any developer that would squeeze every bit of juice out of the PS3 that would be PD. Like I said there are not many that can provide almost 60fps at almost full hd resolution for the PS3. How would you be able to accomplish that if you don't know how to write code? Maybe you have heard of many developers that say that it is very easy to code PS3 games.. I on the contrary have not.

Whether it's hard or not, you'd think after 8 years of combined development PD would have a reasonable handle on it. They're professional developers.

I don't know why many see GT6 as GT 5.5. GT5 was a technical masterpiece, yes, but they have improved a lot of core features (driving physics, hdr lightning, resolution, tessellation and so on) since then. How do you implement new features while still maintaining the processing power - ps3 has limited power - and GT5 was already using up all of it. You have to free resources to be able to, which is not an easy task and takes time.

GT5 was a technical achievement, perhaps. I wouldn't call something that couldn't hold a steady frame rate, had huge issues with pop in, had trackside objects that looked like they came from PS1, and sounds from not much later a technical masterpiece.

Instead of jamming GT6 full of new features (or not, we're still waiting for those) they could polish some of the ones they already had, which was GT5's biggest failing. GT5 had all the elements there for an astounding racing game, but they were buggy, badly implemented and generally mismanaged. Licenses like NASCAR and WRC were misused. Online never really became all it could have. Single player hamstrung itself at every step.

Think of how good a game you could make if you were given all the things that GT5 was capable of, and told to arrange them as you saw fit. That's what GT6 should have been, plus a few new cars and tracks.

Yes, everybody is complaining about missing content and some core bugs. But how many are complaining about the driving physics of GT6. Not the majority anyway. Waiting for new updates is tiresome and boring so I play other games in the meantime till PD decide to release something that interests me, probably around may.

Camber and ride height bugs still exist. The physics feel good, but they're not simulating reality in any fashion, they're just some numbers that happen to feel good.

But physics is the good part of GT6. Even with GT5, the physics were more than acceptable for what the game was trying to do. It's all the other stuff that supports the physics that was pretty poor in GT5, and then they stripped out a lot of the more fun parts for GT6. Ostensibly those parts were to be replaced, like the course maker and online features, but we're still waiting four months later.

There are things that GT5 does that GT6 simply can't, and that's not right. Were we going to GT7 on PS4 I could understand cutting some features, since everything has to be redone for the new hardware anyway. But on the same system, as an evolution of the previous game? That's just silly. Let people keep their dodgy course creator until the new one is ready. Let them play shuffle, let them have seasonal races, let them have endurance races.

GT6 has lots of cars and decent physics, and very little else. Once upon a time, that would have been fine (Gran Turismo 1). In 2014, it's absolutely not acceptable from a AAA first party developer.
 
Whether it's hard or not, you'd think after 8 years of combined development PD would have a reasonable handle on it. They're professional developers.

They have a handle on it. Those features that were implemented in GT6 are not something that every developer can do, like for example tessellation.

GT5 was a technical achievement, perhaps. I wouldn't call something that couldn't hold a steady frame rate, had huge issues with pop in, had trackside objects that looked like they came from PS1, and sounds from not much later a technical masterpiece.

At 2010 not many driving games (if any) were doing what GT5 did - even with those issues. By 2010 standards on the PS3 I consider it technical masterpiece. But that's just my opionion.

Instead of jamming GT6 full of new features (or not, we're still waiting for those) they could polish some of the ones they already had, which was GT5's biggest failing. GT5 had all the elements there for an astounding racing game, but they were buggy, badly implemented and generally mismanaged. Licenses like NASCAR and WRC were misused. Online never really became all it could have. Single player hamstrung itself at every step.

Think of how good a game you could make if you were given all the things that GT5 was capable of, and told to arrange them as you saw fit. That's what GT6 should have been, plus a few new cars and tracks.

Mostly agree. Personally I didn't finish GT5 (think I was LVL 30 something) and have lots of playing time left in there. I played GT5 up until the day GT6 was released. From the first moment playing GT6 I thought it would be difficult to go back to GT5. Most things that bothered me with GT5 were in some way addressed in GT6 (didn't like the loading times and UI but there is a lot more). But my main reason for liking GT6 more than GT5 is the driving physics. Don't know why but they feel just so much better (new tire and suspension model does something to the experience). Content wise, online features and so on, yes, GT5 is better (so far).


Camber and ride height bugs still exist. The physics feel good, but they're not simulating reality in any fashion, they're just some numbers that happen to feel good.

But physics is the good part of GT6. Even with GT5, the physics were more than acceptable for what the game was trying to do. It's all the other stuff that supports the physics that was pretty poor in GT5, and then they stripped out a lot of the more fun parts for GT6. Ostensibly those parts were to be replaced, like the course maker and online features, but we're still waiting four months later.

There are things that GT5 does that GT6 simply can't, and that's not right. Were we going to GT7 on PS4 I could understand cutting some features, since everything has to be redone for the new hardware anyway. But on the same system, as an evolution of the previous game? That's just silly. Let people keep their dodgy course creator until the new one is ready. Let them play shuffle, let them have seasonal races, let them have endurance races.

Agreed. I think all "online" releated special features will turn up after may 20 when the GT5 servers shutdown for good.

GT6 has lots of cars and decent physics, and very little else. Once upon a time, that would have been fine (Gran Turismo 1). In 2014, it's absolutely not acceptable from a AAA first party developer.

Even if it's 2014 it's still the same old hardware. Not so much you can do with it without sacrificing/rewriting something. Anyway I don't doubt that all the features mentioned will turn up. New sounds and better AI I'm not sure they can pull of with GT6. Still hope so.
 
12,261
Australia
Adelaide
GTP_Imari
Even if it's 2014 it's still the same old hardware. Not so much you can do with it without sacrificing/rewriting something.

That was my point though. There's heaps you can do to GT5 without using a single extra processing cycle.

-A career mode that isn't dreary.
-Simple invisible power and speed boosts to the AI so that they can keep up with the player. Same AI algorithm, just tweak the AI car parameters. Player selectable difficult levels, from "no, you pull on the handle to open the door" to "OMG HAX".
-More than two leaderboards at a time. Leaderboards for EVERYTHING should be a standard.
-A variety of online seasonals, instead of Catch the Rabbit, chapter 37.
-Matchmaking and quick events (any extra processing is server side, and the code already exists from GT5P).
-Event creator: choose your track, opponents, etc. Save and share with your friends. They have the code already, because they use it to create events.
-Championship creator: As above, with several races linked together. Save and share with your friends. They have the code already, because they use it to create championships.
-More information on cars! Interviews with drivers, designers, critics, reviewers, race videos, car show videos. Do away with the dreary typo ridden text descriptions, show us what's special about the cars we're driving! Let people find out about the cars in-game, rather than on Wikipedia.

How much would leaderboards on everything add to the game? An event/championship creator? A 2 minute video on each car (say on each model for stuff like the Miata/Skyline/S2000 and friends), or even on each manufacturer?

These things aren't for everyone, but there are people that would spend hours and hours on this stuff, just like there's people that spend hours and hours hotlapping cars now.

A large library of cars and tracks can be used for so many things. It's a shame to see them wasted in GT6.
 

Squanto

GT1-2-3-4-5-6 Enthusiast
Premium
641
Netherlands
Dordrecht
Squanto666
That was my point though. There's heaps you can do to GT5 without using a single extra processing cycle.

-A career mode that isn't dreary.
-Simple invisible power and speed boosts to the AI so that they can keep up with the player. Same AI algorithm, just tweak the AI car parameters. Player selectable difficult levels, from "no, you pull on the handle to open the door" to "OMG HAX".
-More than two leaderboards at a time. Leaderboards for EVERYTHING should be a standard.
-A variety of online seasonals, instead of Catch the Rabbit, chapter 37.
-Matchmaking and quick events (any extra processing is server side, and the code already exists from GT5P).
-Event creator: choose your track, opponents, etc. Save and share with your friends. They have the code already, because they use it to create events.
-Championship creator: As above, with several races linked together. Save and share with your friends. They have the code already, because they use it to create championships.
-More information on cars! Interviews with drivers, designers, critics, reviewers, race videos, car show videos. Do away with the dreary typo ridden text descriptions, show us what's special about the cars we're driving! Let people find out about the cars in-game, rather than on Wikipedia.

How much would leaderboards on everything add to the game? An event/championship creator? A 2 minute video on each car (say on each model for stuff like the Miata/Skyline/S2000 and friends), or even on each manufacturer?

These things aren't for everyone, but there are people that would spend hours and hours on this stuff, just like there's people that spend hours and hours hotlapping cars now.

A large library of cars and tracks can be used for so many things. It's a shame to see them wasted in GT6.
If this was in GT6 I would play it every day 👍
 

BWX

2,005
The fact that they have not fixed major game breaking (IMO) bugs such as pitstop tire bug.. and have not released or given release date of much hyped features such as community features and course creator, and that we were outright lied to about so called DLC, and that design decision problems such as short career mode, "let you win" ai, no endurance races, and the like are not even mentioned at all by PD, it sends a very large message to any fan of the game or "customer" of PD.

It is a huge slap in our collective faces. It is a big fat middle finger waving 6 inches in front of our faces while they laugh. It is a big pile of **** that they *** *** *** ** **** (censored) down your throat.

That is all I need to know about Kaz and PD for now. That will tell you all you need to know about any "timelines" you are interested in. Sometimes past behavior can predict future behavior. I think in this case we can conclude that Kaz and PD will do what they want, when they want, and how they want, and they do not care what the fans or customers want or say. Also, they might tell you a few big fat lies along the way if they feel it is in their best interest to do so before a big release. If you want communication, they are silent. If you want action, they stall. If you want bug fixes, they give you "features" 99% will never use. If you want features listed on the box, they give you a rear wing simulator instead, or more racing suites, or gloves, or paint, or some other POS you don't want, never asked for, and don't care about, while urgent bug fixes and advertized features are never mentioned. They are basically an apathetic company with an uninterested and heavily distracted leader.

Go play something else, go do something else, because GT is dead.
 
BWX
The fact that they have not fixed major game breaking (IMO) bugs such as pitstop tire bug.. and have not released or given release date of much hyped features such as community features and course creator, and that we were outright lied to about so called DLC, and that design decision problems such as short career mode, "let you win" ai, no endurance races, and the like are not even mentioned at all by PD, it sends a very large message to any fan of the game or "customer" of PD.

It is a huge slap in our collective faces. It is a big fat middle finger waving 6 inches in front of our faces while they laugh. It is a big pile of **** that they *** *** *** ** **** (censored) down your throat.

That is all I need to know about Kaz and PD for now. That will tell you all you need to know about any "timelines" you are interested in. Sometimes past behavior can predict future behavior. I think in this case we can conclude that Kaz and PD will do what they want, when they want, and how they want, and they do not care what the fans or customers want or say. Also, they might tell you a few big fat lies along the way if they feel it is in their best interest to do so before a big release. If you want communication, they are silent. If you want action, they stall. If you want bug fixes, they give you "features" 99% will never use. If you want features listed on the box, they give you a rear wing simulator instead, or more racing suites, or gloves, or paint, or some other POS you don't want, never asked for, and don't care about, while urgent bug fixes and advertized features are never mentioned. They are basically an apathetic company with an uninterested and heavily distracted leader.

Go play something else, go do something else, because GT is dead.

I don't always bring updates.. but when I do I make sure no one cares about them. Stay thirsty my friends.

Sometimes a simple "Yes, we are aware of that issue and are working on it." would be satisfying.
 
That was my point though. There's heaps you can do to GT5 without using a single extra processing cycle.

-A career mode that isn't dreary.
-Simple invisible power and speed boosts to the AI so that they can keep up with the player. Same AI algorithm, just tweak the AI car parameters. Player selectable difficult levels, from "no, you pull on the handle to open the door" to "OMG HAX".
-More than two leaderboards at a time. Leaderboards for EVERYTHING should be a standard.
-A variety of online seasonals, instead of Catch the Rabbit, chapter 37.
-Matchmaking and quick events (any extra processing is server side, and the code already exists from GT5P).
-Event creator: choose your track, opponents, etc. Save and share with your friends. They have the code already, because they use it to create events.
-Championship creator: As above, with several races linked together. Save and share with your friends. They have the code already, because they use it to create championships.
-More information on cars! Interviews with drivers, designers, critics, reviewers, race videos, car show videos. Do away with the dreary typo ridden text descriptions, show us what's special about the cars we're driving! Let people find out about the cars in-game, rather than on Wikipedia.

How much would leaderboards on everything add to the game? An event/championship creator? A 2 minute video on each car (say on each model for stuff like the Miata/Skyline/S2000 and friends), or even on each manufacturer?

These things aren't for everyone, but there are people that would spend hours and hours on this stuff, just like there's people that spend hours and hours hotlapping cars now.

A large library of cars and tracks can be used for so many things. It's a shame to see them wasted in GT6.

If those features where in GT6 then yes, that would be awesome. I don't know about an AI algoritm with speed boosts and such. If the AI was doing that I would be thinking "where did he pull that speed from.. that's not possible (without NOS)".

AI must be one of those things that they didn't work on much for GT6. To me it's almost the same as in GT5 but the ai cars deviate from the racing line more often now. Even if that is some progress it's far from good. AI cars still doesn't try to overtake each other that much and almost all of them have the same race strategy (brake early). I don't know why they bother to have 16 ai cars on track when almost all behave the same way. Would rather have 6-10 ai cars that have different styles (less ai cars more resources, better) - something like in B SPEC on GT5. It would also be amazing if they had some fantasy track that is not that detailed to free up lots of resources and use those for proper 16 ai car race.
 
12,261
Australia
Adelaide
GTP_Imari
If those features where in GT6 then yes, that would be awesome. I don't know about an AI algoritm with speed boosts and such. If the AI was doing that I would be thinking "where did he pull that speed from.. that's not possible (without NOS)".

I explained that poorly. Let me try again.

I find the AI in GT5 (and GT6 for that matter) to be pretty well behaved in terms of choosing their line and such. I think the current AI system is more than good enough in that respect. The two areas where I find that it falls down are speed on the straights, and speed through the corners. (Apart from the rubber band effect, which I'm ignoring because the first thing to do is throw that away completely. So imagine we're starting with GT5 AI.)

The AI can go fast in a fast car, it just doesn't do a very good job of driving at the limits of the car like a human can. So while I human might be able to get 95% out of a 500PP car, for 475 effective PP, the AI might only be able to get 80% out of the car, for 400 effective PP.

The way people approach this in GT6 is to cripple themselves by driving a 400PP car, but that's not that optimal for a bunch of reasons.

PD has access to, or should have access to, a ton of data on cornering speeds and straightaway speeds for every track in every car, from real players. I suggest that they collate this data, and group them into three to five difficulty levels that would cover say 95% of all players. Aliens and windowlickers, sorry but you're excluded. ;)

How to adjust the AI for each difficulty level? Say we're doing 500PP at Nurburgring. The player is assumed to be in a 500PP car or near it. The AI is given a 500PP car on equivalent tyres. Run the AI around the track, and see how it deviates from the data you have on players in that difficulty range.

If the AI is too slow through the corners, add grippier tyres.
If the AI is too slow on the straights, add power. Not NOS or boosts or anything, just pure horsepower. Twiddle the ballast if needed to control the balance between acceleration and top end speed.

Repeat until the AI roughly matches the player data pace in pure lap times, speed through corners and speed on the straights. While yes, the AI is "cheating" by having more power and grip, the AI's inability to use it to the same extent as the player should even it out and it should be pretty difficult to tell. It'd be hard to know until you actually did it though.

While this would be a nightmare to do for 1200 cars one by one, I strongly suspect that once you'd done a couple of dozen a pattern would emerge, and you could do the rest very quickly.

All of this would be hidden from the player, the AI car would display as 500PP on correct tyres. All we're doing here is generating a fudge factor to deal with the AI's inability to drive to the limit.

And that's how you make the AI in Gran Turismo competitive without rewriting the AI code at all, with no more tools than what the players currently have available (although you probably want finer control on tyres and power than the player has available).

It likely wouldn't be perfect, and I'd expect at least a few edge cases where the AI turns out to be astoundingly dominant or pathetically easy. But on the whole it should provide an adaptable AI that can be quite competitive, without damaging the good racecraft that the AI has now, and without writing another line of code.