GT Pro Tune - The GT7 Tuning Calculator - Tune every car in Gran Turismo with this app!

[Highlandor]

Not sure what your point is. All of this is what I have built from the ground up, the suspension is based on actual equations, and everything is calculated to start at a balanced and stable state, you can then tweak to improve/ruin it from there! 😂

Can I interest you in trying it out?

That last post was my attempt to explain what was wrong with our calculator, how I was planning to fix it and why I chose that 'fix', and how a much more complex problem arose (I couldn't 'fix').

Now that A1 is around, I wondered whether AI could've/would've solve the problem I couldn't or maybe you had(?) already.

I'd love to try it (much appreciated for the offer), but I don't even have a PS5, rig/seat long gone, old wheel was pretty much on it last legs and subsequently packed away. I'm not here to create another calculator or anything like that. It's about the problem that was too hard for me to figure out. Also, I dunno, maybe there might be someone on here who wondered how it was done with our calculator back then. Probably not, but anyways..

I'll try and summarise a bit more clearly, maybe:

• Our calculator gave a 'middle' setting that 'balanced' the stock custom suspension each car came with
• That stock setting was right in the middle i.e. shown as 100% on our calculator (which had a range of 0-200% - 0% being softest and 200% being hardest) so we could adjust either way by the same amount
• It wasn't long before realising this starting point wasn't very good and it was time consuming to test and find out what 'direction' i.e. softer or harder / front or rear or both, to go in
• For me, it became apparent there was one variable (drivetrain), above all else (such as tyres, power etc), that would make the % values for the front and rear, go in a certain way, the majority of the time
• So for each drive train I created custom % values fr and rr I would input into the calculator as soon as I loaded up the calculator with each cars specific weight distribution and spring rates
• This saved alot time and gave a much better 'base' (with the bonus that it was still balanced) to fine tune from.
• I then applied the same idea to things that weren't on our calculator, such as brakes, LSD and 'toe'. Again, having specific set values (for each drivetrain) to create a 'base' before even driving the car.
• I realised that this could be taken further and fine tune the 'drivetrain' base values to incorporate car types i.e. GT3, DTM & GT500
• Even with quite different % fr to rr (on some cars) the calculator meant there was a natural balance in the setup.
• I felt this was an important reason as to why people were giving feedback that these setups could be carried over to other tyre compounds, regardless of whether those tyres gave more or less grip.
• I looked upon it (tuning) as there were 2 main groups of variables to try and cater for, one was the car, the other was the driver.
• For the 'driver', it started with steering sensitivity, but over time, more and more variables were being highlighted. I can't remember them all, but off the top of my head it was things like throttle control, heel 'n toe, even what view was being used (3rd person compared to bumper cam), ABS on/off, was the setup needed for a race or a time trial, even tracks had different grip levels and what worked on one track was terrible on another) etc.

It felt it was just impossible to come up with a solution to deal with all these 'new' variables. With hindsight I was looking at this in the wrong way i.e. it was a tuning calculator, these things are important to tuning so I have to find an answer. I wasn't looking at this from the user/driver perspective, which was a big mistake.

Now that AI is here, I was curious as to whether that would / could find that solution to the 'driver variables' issue.

I'll try not to de-rail the thread much more. I think we did 'ok', considering how many cars were in the game and our calculator was a fairly simple equation. There seemed to be alot of 'interest' in the 'secrets' of our calculator back then, even though everything had been posted on here openly! So maybe someone might want to know how we were doing it, or it might help someone else in some way (?), or not.

But hey ho, there you go, c'est la vie with t'interweb innit (admittedly, something I didn't realize was going to that bad back then and I struggled to deal with), but enough of that.

All the best and fair play to you for what you created.....much appreciated for the offer, if ever my 'numbers come up' I'll give you a shout, but for the foreseeable future I can't see an opportunity to 'get back into it' unfortunately


H

 

[hookedonturbo]

That last post was my attempt to explain what was wrong with our calculator, how I was planning to fix it and why I chose that 'fix', and how a much more complex problem arose (I couldn't 'fix').

Now that A1 is around, I wondered whether AI could've/would've solve the problem I couldn't or maybe you had(?) already.

I'd love to try it (much appreciated for the offer), but I don't even have a PS5, rig/seat long gone, old wheel was pretty much on it last legs and subsequently packed away. I'm not here to create another calculator or anything like that. It's about the problem that was too hard for me to figure out. Also, I dunno, maybe there might be someone on here who wondered how it was done with our calculator back then. Probably not, but anyways..

I'll try and summarise a bit more clearly, maybe:

• Our calculator gave a 'middle' setting that 'balanced' the stock custom suspension each car came with
• That stock setting was right in the middle i.e. shown as 100% on our calculator (which had a range of 0-200% - 0% being softest and 200% being hardest) so we could adjust either way by the same amount
• It wasn't long before realising this starting point wasn't very good and it was time consuming to test and find out what 'direction' i.e. softer or harder / front or rear or both, to go in
• For me, it became apparent there was one variable (drivetrain), above all else (such as tyres, power etc), that would make the % values for the front and rear, go in a certain way, the majority of the time
• So for each drive train I created custom % values fr and rr I would input into the calculator as soon as I loaded up the calculator with each cars specific weight distribution and spring rates
• This saved alot time and gave a much better 'base' (with the bonus that it was still balanced) to fine tune from.
• I then applied the same idea to things that weren't on our calculator, such as brakes, LSD and 'toe'. Again, having specific set values (for each drivetrain) to create a 'base' before even driving the car.
• I realised that this could be taken further and fine tune the 'drivetrain' base values to incorporate car types i.e. GT3, DTM & GT500
• Even with quite different % fr to rr (on some cars) the calculator meant there was a natural balance in the setup.
• I felt this was an important reason as to why people were giving feedback that these setups could be carried over to other tyre compounds, regardless of whether those tyres gave more or less grip.
• I looked upon it (tuning) as there were 2 main groups of variables to try and cater for, one was the car, the other was the driver.
• For the 'driver', it started with steering sensitivity, but over time, more and more variables were being highlighted. I can't remember them all, but off the top of my head it was things like throttle control, heel 'n toe, even what view was being used (3rd person compared to bumper cam), ABS on/off, was the setup needed for a race or a time trial, even tracks had different grip levels and what worked on one track was terrible on another) etc.

It felt it was just impossible to come up with a solution to deal with all these 'new' variables. With hindsight I was looking at this in the wrong way i.e. it was a tuning calculator, these things are important to tuning so I have to find an answer. I wasn't looking at this from the user/driver perspective, which was a big mistake.

Now that AI is here, I was curious as to whether that would / could find that solution to the 'driver variables' issue.

I'll try not to de-rail the thread much more. I think we did 'ok', considering how many cars were in the game and our calculator was a fairly simple equation. There seemed to be alot of 'interest' in the 'secrets' of our calculator back then, even though everything had been posted on here openly! So maybe someone might want to know how we were doing it, or it might help someone else in some way (?), or not.

But hey ho, there you go, c'est la vie with t'interweb innit (admittedly, something I didn't realize was going to that bad back then and I struggled to deal with), but enough of that.

All the best and fair play to you for what you created.....much appreciated for the offer, if ever my 'numbers come up' I'll give you a shout, but for the foreseeable future I can't see an opportunity to 'get back into it' unfortunately


H

Hi,

Just thought I would follow up, I thought initially you were just coming on here to show off or try and “derail my thread” as you mentioned. But then I thought about it and that wouldn’t make sense. Apologies for jumping to conclusions.

I guess maybe you were looking for a common ground convo, I know making this calculator was a very lonely process, lots of hours in game evaluating and testing and re evaluating! Then even more hours in front of of an excel spreadsheet taping in formula after formula making equations from car suspension tuning books and blogs make sense!

For your comfort, you were definitely on the right track (no pun intended)! The base (stable) setup, is foremost the right place to start then be able to soften or harden to suit the car and drivers preference (both variables that cannot be calculated) I also incorporated a drivetrain denominator that uses a guesstimate weight for the differentials and driveshafts based on the fact if it has those upfront/rearward or both axles.

Also built into the calculations is a multiplier for ROAD/GR4/RACE/VGT or Formula. This give an initial shift of hardness to suit the probable physics model differences of the different suspension types.

There are many other factors that I have built into this that I’m not going to divulge here, but I can safely say that this is the most complex and accurate tuning app ever built for the game. Although I’m sure someone will put me back in my place after saying that! Haha!

If you do ever get back into the game, let me know I’ll hook you up dude!

 

[hookedonturbo]

AI promt suggestions

 

[hookedonturbo]

AI promt suggestions

For ease here is a picture version.

 

[hookedonturbo]

If there is anyone who would like to provide feedback on the AI Race Engineer responses. I would be more than happy to provide assistance and adjustments where needed.

 

[OzinSg]

I haven’t tried the AI feature, but I’m really enjoying the enhancements. It’s a lot less tedious to use now, and it’s great to see less of my annoying typos. Thanks for updating the new cars


Just noticed a couple of details: the FTO GP Version R 97 is FF, but the diff shows it as FWD. The Land Cruiser 74 is the reverse i.e comes up as FF. I noticed it has trouble calculating the tire size for cars with lots of gears.


Also, any chance you could add the Power Pack vehicles?


Great work

 

[hookedonturbo]

What are the power pack vehicles!?

 

[hookedonturbo]

I havnt even bought it yet lol

 

[OzinSg]

They are prizes

• Ford Mustang 2015 American Racer
• Nissan Skyline GT-R GP-Tuned (KPGC10)
• Porsche 911 Turbo Rally (930)
• Toyota AE86 Levin D-Tuned
• Toyota Supra GT Road Car (JZA80)
• Volkswagen Beetle 1966 Desert Racer

 

[hookedonturbo]

I haven’t tried the AI feature, but I’m really enjoying the enhancements. It’s a lot less tedious to use now, and it’s great to see less of my annoying typos. Thanks for updating the new cars


Just noticed a couple of details: the FTO GP Version R 97 is FF, but the diff shows it as FWD. The Land Cruiser 74 is the reverse i.e comes up as FF. I noticed it has trouble calculating the tire size for cars with lots of gears.


Also, any chance you could add the Power Pack vehicles?


Great work

View attachment 1497757

 

[hookedonturbo]

- Bugs fixed.

- Power pack vehicles are being added in the next couple of days. Bear with us!

 

[Highlandor]

Hi,

Just thought I would follow up, I thought initially you were just coming on here to show off or try and “derail my thread” as you mentioned. But then I thought about it and that wouldn’t make sense. Apologies for jumping to conclusions.

I guess maybe you were looking for a common ground convo, I know making this calculator was a very lonely process, lots of hours in game evaluating and testing and re evaluating! Then even more hours in front of of an excel spreadsheet taping in formula after formula making equations from car suspension tuning books and blogs make sense!

For your comfort, you were definitely on the right track (no pun intended)! The base (stable) setup, is foremost the right place to start then be able to soften or harden to suit the car and drivers preference (both variables that cannot be calculated) I also incorporated a drivetrain denominator that uses a guesstimate weight for the differentials and driveshafts based on the fact if it has those upfront/rearward or both axles.

Also built into the calculations is a multiplier for ROAD/GR4/RACE/VGT or Formula. This give an initial shift of hardness to suit the probable physics model differences of the different suspension types.

There are many other factors that I have built into this that I’m not going to divulge here, but I can safely say that this is the most complex and accurate tuning app ever built for the game. Although I’m sure someone will put me back in my place after saying that! Haha!

If you do ever get back into the game, let me know I’ll hook you up dude!

To be fair, you don't have to apologise, you were right. To cut a long story short, too many beers, hefty natural dose of 'fight or flight' and getting triggered, ended up with the inevitable. But, to be fair, I'd been popping back under the same conditions for a while, but always managed to walk away, well until last week. Oh well, managed to get one decent post in. So will try posting when sober from now on

I did write a full on proposal/presentation for an app based on our calculator, but that along with pretty much everything else is now long gone.

I do remember a bit about the 'solution' I was thinking about (from where last post left off i.e. looking at dealing with driver variables the wrong way).

I'll see about typing it up and posting it if you want. In a way it's exactly the same as what you posted (separating problem areas i.e. braking, mid corner etc), but the ideas for how the app and user 'fixes this' are a little bit different, so maybe this might help give some ideas.

One of the negatives of using drivetrains as the starting variable is that sometimes/usually there are cars that are odd ones out (i.e. that don't conform to the handling norm for that drivetrain or type of car). At first it seemed like a problem, but then you get to realize the amount of cars that are like this are a very small minority, so dialling in an individual custom setting especially for these wasn't too much hassle. Although, sometimes they can be a right pain in the butt i.e. Z4 GT3, on GT6 I think.

Sometimes it might not be the 'handling', it might be a setting that sets them apart from the others of a similar kind, like the DTM Merc that had way more aero than the other FR DTM cars. I think we used to run that on hards and let everything else on mediums just to stop it being a 'cheat car'. As good as the drivetrain variable was to start with, there always had to be a way to deal with exceptions to the norm, but I'm guessing you got that covered.

There's a few other things that I never 'went public' with regarding the calc, obviously, got no idea about the app or your tuning history, plus haven't got GT7, so don't know if some if this would help or is stuff you already do / or know?

Sometimes I did something that I called 'reverse tuning' (others probably have different names for it or probably do it as 'normal' nowadays?). I don't think many people did it back in t'day, but happy to be corrected, but I'll try and explain it:

Car understeers, normal fix would be to change something at the front, but instead I'd change something at the rear
Car oversteers, normal fix would be to change something at the rear, but instead I'd change something at the front

i.e. you 'tune' the opposite end of the problem rather than 'tune' the end of the car which has the problem. There's more to this, but don't want to write an essay if you already know. This was the 'secret' to tyre wear (or lack of it ) on some of my setups. But hey ho, might as well go for it now the cat is out the bag, happy Xmas(!), let me know.

There was another 'reverse' trick on certain cars that was from the calc. Normally the dampers should be roughly the same approximate (%) strength as the springs and the reverse of roll bars (which gives that core stability). But on some cars that I used high default % values (i.e very stiff) at the front and low at the rear, you can switch the dampers around and put the low rear ones on the front and vica versa. Not sure if I've explained that clearly(?) or you know about this little 'trick' already. It only ever worked for dampers though and was very handy at helping out on cars that understeered like an ocean liner going into a corner.

I'll shut up now, and go see what I can dig up that I might have left and write up my proposed 'solution' to the driver variables as best as I can remember it if you want.

Keep rockin'

H

 

[OzinSg]

- Bugs fixed.

- Power pack vehicles are being added in the next couple of days. Bear with us!

Awesome - thank you

 

[PatCD]

Getting back into GT7, and found this post. Went to try out the app, I want to get the Advanced version without the AI, if I just hit the 'Buy Now' button, it doesn't do anything. If for some reason, I check the box for the AI, and then hit Buy Now, it goes to the purchase page, but only has the monthly AI listed, not the Advanced version of the main app.

Any help would be appreciated.

 

[hookedonturbo]

Getting back into GT7, and found this post. Went to try out the app, I want to get the Advanced version without the AI, if I just hit the 'Buy Now' button, it doesn't do anything. If for some reason, I check the box for the AI, and then hit Buy Now, it goes to the purchase page, but only has the monthly AI listed, not the Advanced version of the main app.

Any help would be appreciated.

Hi Pat,

Sorry, bit of a technical glitch happening at the moment! Bear with us while we fix the database.

 

[hookedonturbo]

Getting back into GT7, and found this post. Went to try out the app, I want to get the Advanced version without the AI, if I just hit the 'Buy Now' button, it doesn't do anything. If for some reason, I check the box for the AI, and then hit Buy Now, it goes to the purchase page, but only has the monthly AI listed, not the Advanced version of the main app.

Any help would be appreciated.

Hi Pat,

Please login with the email and password that you signed up with. Then click the upgrade link to purchase.

That should hopefully solve it for your case.

 

[Highlandor]

This is 'part 1' of how I was trying to find a solution to driver variable problem. I split this into 2, part 1 was about getting user to communicate feedback accurately, part 2 was taking that feedback and applying it to the setup. I used flow charts (I can't remember if there were 2 or 3 charts to be honest), which were very helpful to 'see' if any areas of the solution were being duplicated or it was becoming maladaptive and was going to end up chasing my tail. Unfortunately, this all got chucked a few years ago when I packed up my wheel etc.

All I got left are a couple of images used on my old sight (that might be applicable?) and a couple of testing videos I was going to 'use' to explain / show how I was testing. Attached are the images, these are just an example of communicating using visual cues rather than words, and these were for 'track limits' in our lobbies. If anyone knows how to post a vid I'll upload the testing vids, that's even if anyone is interested

Obviously, this was all 10 years or so ago, so can't remember everything exactly as it was, but hopefully this shows how looking at things from user perspective can be a bit different to creator perspective, maybe (?). I've tried to do this in a way which could (maybe) make it be easier to extract/copy onto a flowchart if need be. Also, any questions are effectively me 'thinking out loud' to myself i.e. it was basically a brainstorm with myself, which was interesting

I can't remember all the driver variables, but most of them were on the previous post, pretty sure if need be, any ones missed could be found. So, here's the 'brainstorm':

• Problem 1 - what happens when user asks for more than one driver variable - or when someone chooses a large group of different driver variables for the calc/app to incorporate onto base setup?
• Problem 2 - the sheer amount of custom values required to deal with all of this (I have no idea, but now it must be running into thousands), would that even be physically possible? Something like steering sensitivity would have multiple options just on it's own. What if two (or more) of the variables start to 'contradict themselves' in terms of what the calc/app is proposing for the 'answer'.

(NOTE: Part of this is applicable to PART 2 where there needs to be a 'decision' by the calc/app as to whether to keep on amending the original base setup, or to start afresh with a new base setup that is better suited to the users needs and therefore would need less amendments, which would reduce the chances of 'contradiction' occuring).


Possible solution:
The one 'thing' who knows the needs/requirements of driver/user better than anyone else is always going to be 'themselves'. So how about getting 'them' to do the fine tuning 'themselves'? But, that leads to a couple of potential new problems...

• Problem 3 - if user is using a tuning app/calc, chances are they probably don't 'know' about tuning generally (or maybe specific things like terminology i.e. understeer, springs, LSD). So, if there is a terminology issue, how can you get the user to 'fine tune' a setup, the calc/app has given them, to their needs?
• Problem 4 - also, if user has paid for the calc/app, they might not be too chuffed when calc/app asks them to 'do the fine tuning' themselves. I mean, this is what/why they're paying for the calc/app to do(?)

Possible solution:
Maybe if there was a way whereby the calc/app makes the user 'think' that it's the calc/app that's 'doing the fine tuning', when in fact it's the user who is doing it.

• Problem 5 - But, this would need a way for the user to accurately tell the calc/app what the problems are, which potentially takes us back to issues highlighted in problem 3(?)

Possible solution:
Before starting to address what is 'wrong' with the 'setup', maybe it would help if we could establish what is the best way for the user to accurately communicate the problem(s). Not everyone has the same knowledge of the terminology involved

• Problem 6 - this feels like we're going back to what the original problem was (too many variables)

Possible solution:
What about having an option on the calc/app whereby the user can choose how to 'communicate' the problem i.e. a bit like how a person 'personalises' what they see on screen when playing GT i.e. changing the view, what things to display. So, on the calc/app there will be some kind of an 'option'(?) for the user to select between, main possibilites/examples:

Using terminology (words)
Or
Using 'visual' cues i.e. images or video

• Problem 7 - language, what if user doesn't speak english(?)

Possible solution:
Maybe this reinforces the notion of using of visual cues, rather than words, (thus killing 2 birds with one stone?)

END PT 1

I looked upon these issues as driver variables are going to impact a setup much the same way as the settings within the setup i.e. understeer/oversteer, braking, mid corner, grip/acceleration. So the calc/app can 'react' in the same way whether the 'problem' was with the base setup or driver variable. So that is what part 2 was about.

----------------------------------------------------------------------------------------------------------------------------------------------------

Not sure how soon to post part 2 as this took a while, had to do this from scratch, from memory. Hopefully there might be something here that helps in some way, shape or form, or helps to give 'ideas' for the future(?)

H

 

[hookedonturbo]

This is 'part 1' of how I was trying to find a solution to driver variable problem. I split this into 2, part 1 was about getting user to communicate feedback accurately, part 2 was taking that feedback and applying it to the setup. I used flow charts (I can't remember if there were 2 or 3 charts to be honest), which were very helpful to 'see' if any areas of the solution were being duplicated or it was becoming maladaptive and was going to end up chasing my tail. Unfortunately, this all got chucked a few years ago when I packed up my wheel etc.

All I got left are a couple of images used on my old sight (that might be applicable?) and a couple of testing videos I was going to 'use' to explain / show how I was testing. Attached are the images, these are just an example of communicating using visual cues rather than words, and these were for 'track limits' in our lobbies. If anyone knows how to post a vid I'll upload the testing vids, that's even if anyone is interested

Obviously, this was all 10 years or so ago, so can't remember everything exactly as it was, but hopefully this shows how looking at things from user perspective can be a bit different to creator perspective, maybe (?). I've tried to do this in a way which could (maybe) make it be easier to extract/copy onto a flowchart if need be. Also, any questions are effectively me 'thinking out loud' to myself i.e. it was basically a brainstorm with myself, which was interesting

I can't remember all the driver variables, but most of them were on the previous post, pretty sure if need be, any ones missed could be found. So, here's the 'brainstorm':

• Problem 1 - what happens when user asks for more than one driver variable - or when someone chooses a large group of different driver variables for the calc/app to incorporate onto base setup?
• Problem 2 - the sheer amount of custom values required to deal with all of this (I have no idea, but now it must be running into thousands), would that even be physically possible? Something like steering sensitivity would have multiple options just on it's own. What if two (or more) of the variables start to 'contradict themselves' in terms of what the calc/app is proposing for the 'answer'.

(NOTE: Part of this is applicable to PART 2 where there needs to be a 'decision' by the calc/app as to whether to keep on amending the original base setup, or to start afresh with a new base setup that is better suited to the users needs and therefore would need less amendments, which would reduce the chances of 'contradiction' occuring).


Possible solution:
The one 'thing' who knows the needs/requirements of driver/user better than anyone else is always going to be 'themselves'. So how about getting 'them' to do the fine tuning 'themselves'? But, that leads to a couple of potential new problems...

• Problem 3 - if user is using a tuning app/calc, chances are they probably don't 'know' about tuning generally (or maybe specific things like terminology i.e. understeer, springs, LSD). So, if there is a terminology issue, how can you get the user to 'fine tune' a setup, the calc/app has given them, to their needs?
• Problem 4 - also, if user has paid for the calc/app, they might not be too chuffed when calc/app asks them to 'do the fine tuning' themselves. I mean, this is what/why they're paying for the calc/app to do(?)

Possible solution:
Maybe if there was a way whereby the calc/app makes the user 'think' that it's the calc/app that's 'doing the fine tuning', when in fact it's the user who is doing it.

• Problem 5 - But, this would need a way for the user to accurately tell the calc/app what the problems are, which potentially takes us back to issues highlighted in problem 3(?)

Possible solution:
Before starting to address what is 'wrong' with the 'setup', maybe it would help if we could establish what is the best way for the user to accurately communicate the problem(s). Not everyone has the same knowledge of the terminology involved

• Problem 6 - this feels like we're going back to what the original problem was (too many variables)

Possible solution:
What about having an option on the calc/app whereby the user can choose how to 'communicate' the problem i.e. a bit like how a person 'personalises' what they see on screen when playing GT i.e. changing the view, what things to display. So, on the calc/app there will be some kind of an 'option'(?) for the user to select between, main possibilites/examples:

Using terminology (words)
Or
Using 'visual' cues i.e. images or video

• Problem 7 - language, what if user doesn't speak english(?)

Possible solution:
Maybe this reinforces the notion of using of visual cues, rather than words, (thus killing 2 birds with one stone?)

END PT 1

I looked upon these issues as driver variables are going to impact a setup much the same way as the settings within the setup i.e. understeer/oversteer, braking, mid corner, grip/acceleration. So the calc/app can 'react' in the same way whether the 'problem' was with the base setup or driver variable. So that is what part 2 was about.

----------------------------------------------------------------------------------------------------------------------------------------------------

Not sure how soon to post part 2 as this took a while, had to do this from scratch, from memory. Hopefully there might be something here that helps in some way, shape or form, or helps to give 'ideas' for the future(?)

Okay. So,

You have highlighted many problems, rightfully so. However, there is one larger issue to hand here. If the user does not know how to communicate to the app of what the handling issue they are experiencing is, then none of this will work.

This is why GT pro tune produces very stable initial base set up, that the user can drive out of the box. To tweak the set up to their own personal driving style, I have brought in the AI race engineer, to try and translate the users requirements into small tweak adjustments that will help.

The AI prompt that I use as a very strict set of instructions, all the user needs to do he’s just give it a short explanation of what they are feeling, i.e. The car feels unstable in corners, the car jumps over bumps and curbs, it’s hard to accelerate out of slow corners without wheel spinning.

The user can also provide a preference solution if they have some more knowledge, i.e. Suggest settings to make ride more compliant over bumps and curbs by adjusting the dampers.

If you want to upload a video, you have to post it to YouTube or another video host site and then paste the link.

If I’m honest, H, I think you’re overthinking the whole driver variables issue. In order to achieve what I think it is you were trying to achieve you essentially need to train the user in every area of race craft, vehicle engineering, suspension geometry, and all associated terminology and language.

But look, I enjoyed reading your post, please feel free to carry on brainstorming!

 

[Highlandor]

Okay. So,

You have highlighted many problems, rightfully so. However, there is one larger issue to hand here. If the user does not know how to communicate to the app of what the handling issue they are experiencing is, then none of this will work.

This is why GT pro tune produces very stable initial base set up, that the user can drive out of the box. To tweak the set up to their own personal driving style, I have brought in the AI race engineer, to try and translate the users requirements into small tweak adjustments that will help.

The AI prompt that I use as a very strict set of instructions, all the user needs to do he’s just give it a short explanation of what they are feeling, i.e. The car feels unstable in corners, the car jumps over bumps and curbs, it’s hard to accelerate out of slow corners without wheel spinning.

The user can also provide a preference solution if they have some more knowledge, i.e. Suggest settings to make ride more compliant over bumps and curbs by adjusting the dampers.

If you want to upload a video, you have to post it to YouTube or another video host site and then paste the link.

If I’m honest, H, I think you’re overthinking the whole driver variables issue. In order to achieve what I think it is you were trying to achieve you essentially need to train the user in every area of race craft, vehicle engineering, suspension geometry, and all associated terminology and language.

But look, I enjoyed reading your post, please feel free to carry on brainstorming!

Like I say, I don't know about your testing or history, I can only go by what I was experiencing. Maybe if you could explain how this is all tested on GT it could help? Where, how, who do you test with etc?

I had alot of different guys that would just come to the lobby and race, chill or test with. Most of my time on the game was hosting, only a small % was tuning and of that, very little of my own time, so I was heavily influenced by the people in the lobby.

Some didn't even speak very good english, one german guy (DERALTE) spoke hardly any English, but he raced super clean and general not a problem, although explaining manually adjusting the PP system to him wasn't easy. We managed to find out he 'drove' for the DTM, which got us all excited, until one of the other german guys helped translate and it turned out he was one of the lorry drivers supporting the races i.e. delivering portable toilets. But we still called him the DTM driver, which he found kinda funny.

A swedish guy (AHLBY) couldn't speak english, so he never used a mic, and he just typed how he heard words. One of the funniest moments we ever had was when he said wait (don't start the race) as he was going to get a bear from the fridge!

At least one guy drove no ABS, everyone used different views, had different wheel rotation, some were still on a pad, y'know that was the norm for me, trying to deal with all that.

I always tested online, always had tyre wear on, all driving aids off apart from ABS at 1, always had steering sensitivity at 0, both on wheel and game, in fact, I've never played any version of GT with anything other than stock rotation.

I always thought things like extra grip (better tyres) and ballast was hindering 'learning' about tuning, so I never used them, always tried to keep to stock tyres. I only ever used ballast for online races as in the same way real life races do i.e. success ballast, whoever was leading the series/championship got the most ballast.

This was normal to me and was part of the process of ending up doing what I did and knowing what I know. Although GT5 Prologue was also a very big influence, 3 white tuned EVOs running Racing Hards at the front and SPORTS Hards at the rear (although Mikey and B1 chose mediums - the clio crew never knew what hit them at Suzuka ), and well, y'know, the rest as they say is history.

It's up to you, if you don't want to say, then no worries, keep rockin

H

 

[hookedonturbo]

Update complete.

New cars added. Also deployed some code to fix some bugs and for the AI to pickup the track name.

Enjoy!

 

[Terronium-12]

@chrisbaba When you signed up here, you agreed to the AUP which states all messages must be posted in English. If it isn't your native language, that's perfectly fine. We don't ask for perfect English, just for there to be a uniform language across the board for everyone to understand and easily interact with.

If you have to use a translator, please do so. We don't want to discourage you from posting (it's the last possible thing we'd want) but do keep this in mind moving forward. 👍

 

[hookedonturbo]

@chrisbaba When you signed up here, you agreed to the AUP which states all messages must be posted in English. If it isn't your native language, that's perfectly fine. We don't ask for perfect English, just for there to be a uniform language across the board for everyone to understand and easily interact with.

If you have to use a translator, please do so. We don't want to discourage you from posting (it's the last possible thing we'd want) but do keep this in mind moving forward. 👍

Thanks, really well deserved and needed.

@chrisbaba - please feel free to message me directly rather than on the open forum or even email me at support@gtprotune.com

 

[DJ_Smuckers]

Hello! I'm a complete noob to tuning and usually just try and find a car tuning someone posted but there aren't always a shared tune for a car I want to use or they don't have a setup that fits my (poor) driving style, so this calculator is great tool. I'm having some problems with the calculations getting values for settings that my car does not go to. I know this is 100% a user error and I'm sure I'm putting wrong information into the calculator. I did watch the instructional video but still not having any luck. Any tips would be appreciated. Thank you!

 

[hookedonturbo]

Hello! I'm a complete noob to tuning and usually just try and find a car tuning someone posted but there aren't always a shared tune for a car I want to use or they don't have a setup that fits my (poor) driving style, so this calculator is great tool. I'm having some problems with the calculations getting values for settings that my car does not go to. I know this is 100% a user error and I'm sure I'm putting wrong information into the calculator. I did watch the instructional video but still not having any luck. Any tips would be appreciated. Thank you!

Hi,

Happy to help, can you send me some screenshots of the car setup screen and the calculator screenshots as well, I can then see if what the issue is.

 

[hookedonturbo]

Any feedback on the latest GT ProTune update and the AI chat feature?

It would be good to hear how anyone has found tweaking the setup with this?

 

[savior972]

Hello everyone,



I purchased the app quite a while ago and would now finally like to get to grips with it.

I watched the videos on YouTube and tried to create two setups. I got an error message with both of them, as shown in the attachment.

I'm now a little stuck and would like to know what I did wrong.

The attachment shows the entries I made.

The error occurred in both attempts to create the setup.

Once upon a time, there was the BMW M3 '03 and the BMW M3 '97.

I would be very grateful for any help, as I think this tool is really cool.


Translated with DeepL.com (free version)

 

[hookedonturbo]

Hello everyone,



I purchased the app quite a while ago and would now finally like to get to grips with it.

I watched the videos on YouTube and tried to create two setups. I got an error message with both of them, as shown in the attachment.

I'm now a little stuck and would like to know what I did wrong.

The attachment shows the entries I made.

The error occurred in both attempts to create the setup.

Once upon a time, there was the BMW M3 '03 and the BMW M3 '97.

I would be very grateful for any help, as I think this tool is really cool.


Translated with DeepL.com (free version)

 

[savior972]

Okay,

then I'll have to check again tonight.

Thanks for the tip.

One more question:

Do you get the error message when you send it in the app as shown in the screenshot?

 

[hookedonturbo]

Hello everyone,



I purchased the app quite a while ago and would now finally like to get to grips with it.

I watched the videos on YouTube and tried to create two setups. I got an error message with both of them, as shown in the attachment.

I'm now a little stuck and would like to know what I did wrong.

The attachment shows the entries I made.

The error occurred in both attempts to create the setup.

Once upon a time, there was the BMW M3 '03 and the BMW M3 '97.

I would be very grateful for any help, as I think this tool is really cool.


Translated with DeepL.com (free version)

This is also incorrect. How did you calculate the tyre size?

 

[savior972]

I calculated the tire size as shown in the YouTube video.

What kind of value should be there, with many decimal places?

It seems like I've done everything wrong

It's time to really deal with it.