I am never an expert, just wanna be. But relaxation length is distance for tire to roll till it hits 63percents of steady cornering force. So it is a lag, you steer in, but tire will not start to steer immediately, it kinda needs a little bit to react, it is very interesting. It also seems to depend on a lot of things and is not a steady value, IDK what is the reference point that Kunos is using that value for. But if I understand it correctly it is the greater the greater force is, also weirdly seems to drop down with more slip angle.
Flex in Kunos tyre.ini is something like multiplier for slip angle range... so with more flex it should be harder to hit best slip angle, but also should allow more range to fish for it at the time of flex, till you mess up of course. In V5 thread Kunos wrote that Flex is not used anymore, Flex gain used now, but it seems to me that both are working, maybe placebo, but I kept tweaking it...
http://www.assettocorsa.net/forum/index.php?threads/tyre-model-v5-specs.29891/
For sure I don't understand it all too well, how it is implemented in engine so on and on...
After playing and testing with XJ13 I have increased all these "deformation" values higher from what Kunos seems to use. And quite a lot higher. I also made front and rear values more different, so front and rear oscillations would be more different, what I find more believable. While Kunos seems to use very similar values, just slightly different.
I probably have room for tweaks, I did it all entirely on feel, and just with a bit of very not confident logic lol
Now I use something like (in relation between front and rear) less relaxation length and more flex for narrow front tires, and more relaxation length and less flex for wide rear tires. No real base for this. But perhaps logic could be that front is steered, and rear reacts a little after that ? And for flex, perhaps narrower tire flex will cause more significant tire dynamics change, than a flex in bigger tire ?
And it feels somewhat right, feels like little bit of spice is good for really awesome Aphidgod physics, really enhancing it. Now I am trying it every day, to check if this "eureka" thing is stable and was not based on mood, placebo or some kind of bias.
Try something like that in your mod, check how it works, don't be afraid to boost up those values two three or maybe even more times. And tell if you felt any positive outcome of it.
Though I will have to do some more testing with some of your tests to see if I can improve them even more, though my struggle is that the belted tires are difficult to know what values should be more radial like and what should be more bias-ply. Out of curiosity what values of flex are you using?
