Yes, this is the point where I need to make more intelligent control. With a steering wheel it would be a piece of cake
I have kind of the same issue. Maybe simply allow counter steering to go as far as player wants ? I was thinking of something more controled in case of counter-stering / drifting. Like preventing the player from spinning voluntary while cornering.
Yes, this is my problam, that neither do I understand it Probably it's a bad idea that helps in some situation which should be solved by proper simulation of other (missing) parts (eg: suspension/diffs)
I don't really understand your arcade tweak. But if the car don't want to go straight after a turn, you can try to remove the front/rear weight distribution or tweak a little side weight repartition. I have this issue sometimes because of strange weight balance.
No, I meant less response force is generated by more load at the same SA/SR. Meanwhile I learnt, that it was correct, and it has a name "load sensitivity"
I don't directly use the pacejka formula, instead i plot a curve with values. The curve i currently use is the one KJM gave above. So back to your question. Yes you have less response force in the end of the curve, despite maybe higher load.
And that is what you lose if you use normalized curve.
I also calculated with instant weight transfer which shouldn't happen instantly. If I had proper suspension I'd have more transient transfer, and I'm not sure that all "dynamic" weight should be moved because of suspension and anti-roll-bars.
Weight transfer is really the thing that seems to have the more influence here. I just integrated front/rear & left/right at the same time, but when i do long turn, at some point the front outside wheel get the highest weight load (while inner rear has the lowest), and then my car spins immediatly.
I don't think you should add static lateral friction, that should come from the tyre model. At least I would like to avoid all possible magic numbers here and there
In fact, all my control loss situations and so do make a lot of sense with regards to physics, the problem is not preveting them but making them fun/controlables. The real problem i almost always have is : when i lose control, instead of starting a controled/credible drift situation, car almost always spins immeditatly, which seems really unrealistic. I feel like i should had some kind of "static lateral" friction which would always be there or something like that i may have missed. I use pacejka forces and nothing else, maybe should i ?
A haven't tought of speed sensitive downforce that every car generates, yet. And the differential. Now I have rigid rear axle, and I suppose you have that too
I've seen in other sims using wings also for street cars to "emulate" downforce generated by the chasis shape. I don't think that has too much significance, though. More likely some addition to a more realistic car, and not to a spin or not to spin situation that we have.