Something is odd in these settings (turn compensation and stuff). I am more than 10 years thinking about it.
To check the linearity I look at the arc direction gauge. The visual wheel I still don't get it and lately I removed from screen.
In my opinion the settings behind it and the way it works and effect the wheel drawing should be rethought and or rephrased.
I don't mean to be harsh and I can love without it but something is off and if nobody can easily figure it or explain it then it could be done better.

I wrote something about that in 2021, but you replied there so I'm not sure what issue there is now.
https://www.lfs.net/forum/post/1977600#post1977600

It actually isn't very complicated. Wheel turn compensation (which is recommended to be set to max) does two different things depending on if the game controller wheel turns more or less than the in-game car wheel.

1) Game controller wheel can turn more than the simulated car steering wheel:

Perfect match is possible.
Game controller wheel stops turning (a 'stop force' is applied) when simulated car steering wheel reaches limits.

2) Game controller wheel can not turn as much as the simulated car steering wheel:

Perfect match is not possible.
Non-linear approach is implemented.
Steering wheel turn amount matches in the middle, but simulated car wheel turns more as you increase the steering angle.

NOTE: This relies on the "Wheel turn" setting in Options - Controls matching your controller wheel turn degrees. E.g. if your game controller turns 900 degrees, then "Wheel turn" must be set to 900.

EDIT: If I am missing something, or some bug has crept in, I would obviously like to know about it. But when I have checked in the past, it's all working as it should.

Quote from Scawen :

I wrote something about that in 2021, but you replied there so I'm not sure what...

So, I did some more testing and I'm sure something is wromg with either the code or my wheel's communication with the game. To remove all sources of error, I checked my wheel "axis" for linearity outside of LFS. Everything is working as intended, full range of motion and functioning bump stops if you decrease lock via the Logitech gaming software. I also checked the functionality in other games. As far as I can tell, everything behaves normally (eventhough most games aren't as transparent about their mechanics as LFS, so take this with a pinch of salt).

The problems I mentioned in my previous post still persist and I found another issue. On cars with 900deg lock-to-lock the turn compensation does nothing which makes sense (yet I don't have linear steering). Things get weird if you go below 900deg. My testing-car (prev. post) at 540 lock-to-lock behaves very linearly at 0.00 turn compensation (?). The problem starts if I add compensation at around 30deg lock at my physical wheel (and the virtual one): At 0.01 the virtual wheel suddenly jumps to a higher angle (about 45deg) and stays there all the way to 1.00 compensation therefore decreasing linearity. This effect increases as I add more lock on my physical wheel. At 180deg for example I get around 270deg on the virtual wheel.

Additionally, I can't "activate" the bump stops in LFS even if I decrease my maximum lock at the tires to 10deg. I still get 900deg steering wheel rotation at my physical wheel. The "controller wheel turns" slider appears to do nothing for me.

I tested everything in the 2015 0.6J version to confirm my findings and found the same results.

What I suggest is some data collection from other Logitech G29 players to figure out if this is an issue with my setup or the game. If the issue can be reproduced we should create a seperate thread and figure out what's going on with the code.

Before I start up the other computer (with G29) to check, can I ask you to check with the official GTR cars?

They have 540 degrees of lock. You can see that in "Steering" (section of the Garage/Pits screen) some text:
Car's steering wheel turns 540 degrees (lock to lock)

I guess you see this same info text with your testing car, but please check to confirm that.

I suggest testing with an official car, or a specific mod that we all have access to, so we can compare results.

Quote from Scawen :

Before I start up the other computer (with G29) to check, can I ask you to check...

Took the FZR, default setup and 540deg.

Screenshots taken at 90 and 180 and 360 at my physical wheel, compensation is at 0.00. As I said, 540deg seems to be the "magical number" where everything (below 270) works.

I haven't checked here yet but I've been looking on the internet and apparently thing may be affected by a "Steering Sensitivity" setting in the Logitech software.

https://steamcommunity.com/app/227300/discussions/0/1639787494950318456/

EDIT: Also you may be able to check if the non-linearity comes from the wheel/driver software by looking at the raw output values at the bottom right of the Options - Controls screen. Those are the raw input values that LFS receives from the software.

Quote from Scawen :

I haven't checked here yet but I've been looking on the internet and apparently...

Seems like a good idea, I will check if I'm back at the screen. I remember there being some weirdness with the steering sensitivity slider in Richard Burns Rally

I turned the "sensitivity" slider to 50%, this restored linearity. Now the compensation works aswell, as long as the "wheel turn" is set to the correct value

Quote from Scawen :

I wrote something about that in 2021, but you replied there so I'm not sure what...

I remembered that conversation happened but forgot the explanation in itself. I think the wording is either too obscure to me or my English is not high enough (I use English non translated version), causing me to forget the way it works.
What are the intermediate compensation doing? Any curve or formula of what´s going on?
I also admit to very often using less controller turn than real life, and this is where it gets weird. Maybe were some different solution can be found.
Not a big deal but I don´t think a beginner can easily understand and I also think that a lot of seasonned player forgot about it....
Anyway it would be nice to have more clear user friendly or more synchronous to used wheel, but I can live without.
I hope my previous message did not feel too harsh because that was not the intention but my wording was not the best.

Quote from loopingz :

I remembered that conversation happened but forgot the explanation in itself. I...

I think I figured it out now. First of all, make sure that there is no software at play that changes the raw output data of your steering wheel. In my case the Logitech Gaming Software messed up the signal before it reached LFS. (For some reason "sensitivity" in the LGS means "linearity". Also 50% sensitivity somehow equals 100% linearity. )

Then make sure your physical wheel turn matches the setting in LFS and set the "compensation" slider to 1.00. If the car inside LFS has less "wheel turn" than your physical wheel, LFS will set up "bump stops" when you hit maximum lock. If the car inside LFS has more "wheel turn" than your physical wheel the game will compensate at the far ends of your steering lock. The steering will behave linear closer towards the center and less linear towards the end of your "travel".

You probably want to set up a specific profile just for LFS, since it's the only game that REALLY only uses raw output data from your controller. I found that most other games somehow filter the data before it is displayed on screen, therefore they seem to work better with some adjusted settings in your wheel setup software.

Hi!

So, I have no idea is this an actual bug in LFS itself or is this an issue with one mod in particular.

I've tried to generate AI path for mod CAMMERA JGT (https://www.lfs.net/files/vehmods/79E657) by ivancsx on RO1 track. While it does generate it in single player, the moment I try to add AI in multiplayer after generating it, it tells me that "AI path info must be regenerated" which results them shutting down their engine, similarly how AI behaves on open config layouts. If I then try to return that AI to pitlane or add them again after spectating it, it tells me that it "Could not load AI path info: [knw-file path]"

And, I can make AIs drive that mod in single player with no issues. However, the moment I try to add AI in multiplayer, it deletes that knw file altogether from that [LFS-instance]/data/knw folder. I then have to generate the AI path for that mod in single player again, if I want to have AI drive that mod in that mode.

But here's a thing. During same multiplayer session, I've managed to add AI's using DORITO, IMPREZZIVE and TROENO JGT's and had no problems whatsoever with those mods. And yes, I did think it is a possible bug with this mod in particular, but CAMMERA JGT has worked with AI when using other tracks (at least in BL1, AU3 and AU4) before. So, no idea what has happened there

Quote from tankslacno :

So, I have no idea is this an actual bug in LFS itself or is this an issue with one mod in particular.

Okay, I managed to solve this. The reason why this happened is this:
1) I did previously save a single player grid where AI used Cammera JGT version 10
2) I loaded that grid and managed to generate AI path for that car. No issues at all!
3) I entered multiplayer and tried to add AI with that car. However, what I didn't realize is that the newest version of that mod is version 11. That's why, when I added that AI on multiplayer, it deleted that knw file, because I had generated that AI path for version 10, not 11.

- I solved the issue by just adding a new AI in single player and making sure it used the newest version of that particular mod.

So, this issue happens because LFS Grid files don't take this possibility into consideration: the fact that mod may be updated by author after user has saved grid containing AI using that mod. They are always using that version of the mod which was used when grid was saved.

Great bug finding again! Thanks for working out the reason. The car is probably a fraction wider in the later version, and at some point in the generated path the car goes right up to the driveable limit.

Quote from Scawen :

Great bug finding again! Thanks for working out the reason. The car is probably a fraction wider in the later version, and at some point in the generated path the car goes right up to the driveable limit.

According to the ivancsx (mod author), the changes in version 11 were following:

Quote :

- changed tyre size from 265-285 to 245-305 as suggested by fellow AHPP member (due to stress factor on rear tyres are uncanny)
- retune suspension stuff to accomodate that tyre size

That second part of your message got me wondering: so is it possible for mod authors to do small changes for mods without having to regenerate AI paths? Because I've thought that the simple fact of different versions being used between the mod in multiplayer and the one which generated the AI path alone is enough to cause this issue

Yes, the AI path doesn't store the mod's version number. It's only really about vehicle width, a wider one is likely to invalidate the path.