The online racing simulator
[OLD] Tyre Physics Progress Report
Hello LFS Racers,

As we said in our news item on 21st August, we have been working on a new tyre physics model. It has a good mathematical basis for the forces a tyre produces under varying circumstances and does feel good to drive, so we are very pleased with the new model. The new physics and plans for the VW Scirocco have inspired some other developments that are now important to get the best experience from the updates.

Most of the remaining development is in the following sections :

- Tyre heating and cooling
- AI driver improvements
- Limited setup system
- Suspension updates
- VW stability control

We'll now talk a little about these things, and some of our plans. Please be aware that these are just plans as we see them now. Plans can change, and they often do, so it is very likely that some things you read below will be different from the final result.

Tyre heating and cooling

As the tyre model is completely different, both in the contact patch (frictional effects) and also the way the tyre reacts to loads (deflection depending on air pressure and sidewall stiffness) the old tyre heating system no longer worked properly. So we have been trying new things, with some success, but at this time there are still noticeable issues that need to be sorted out. We found it impossible to simply use the old heating model with some adjusted figures as we had hoped. Some more real updates are needed to get the heating and cooling of tyres right.

AI driver improvements

Just recently we discovered that the AI were driving around too slowly because they were underestimating the available grip. But because available grip depends on tyre load in a non-linear way, it turns out that some of the programming for the AI drivers needs a little reorganising. Many community members like to drive with AI in single player mode, so it is important for them to drive at a good speed.

Limited setup system

We mentioned before that the VW Scirocco would be our first car with a limited setup. But as time went by, that plan changed a little. We realised that limited setups would be a popular option for online hosts. It is a feature that has often been requested. It would allow people to race in road cars that feel like road cars (not stiffened and lowered) and still be competitive. So we decided the limited setups should be an option that applies to many of the LFS cars. For example, in limited setup mode, the standard road cars would only allow tyre pressure and toe adjustments. The LX6 / RAC / FZ5 class (LRF) would also allow compression and rebound damping to be adjusted along with ride height and adjustable anti roll bars (in some cases). The idea is to allow the same adjustments you could make on the real car without changing any parts.

Suspension updates

In order to implement the limited setup system, we needed some good standard setups, the kind that these cars would have in real life. In comparison with most LFS setups currently in use, the new default road car setups are a lot softer. Our testers noticed that with the new setups some cars can be a bit tricky at times, for example when adjusting the accelerator while cornering or when braking into corners. In real life, the suspension geometry is designed to produce an anti-dive effect under braking and anti-squat under acceleration. This has until now been absent from the LFS suspension systems, but now looks like an important development to support the realistic setups.

VW stability control

The special traction control and stability control systems for the VW Scirocco still need to be completed. We don't intend to make this as complex and refined as the real VW systems. We do want to recreate some of the beneficial effects of those systems that can apply individual wheel brakes to help the car go where the driver wants it to, when things begin to get out of shape.

We hope this progress report gives some insight into what has been going on. As you can see there are still important things to be completed. It's hard to know how long the tyre heat and suspension updates will take, so we are still not able to estimate a release date.

A reminder :
- The new physics will be a free update for all LFS users
- The VW Scirocco will be available to S1 and S2 license holders
Nice report. Thanks.

Looking forward to it all!
#3 - amp88
Sounds very interesting, thanks for the progress report.
very nice that devs team listens to community and makes such reeports
Thank you Scawen

Now we have a report to know more about all the things
Good list !
Great improvements.
#8 - tolas
I guess we will have to wait long time for all this to be made, but I'm patient. It's going to be a great release.
sweet!
#10 - Byku
Yeah, it sounds like months at least ^^, well.. at least i've got something to live for .
nice stuff! always good to hear updates
-
(aroX123) DELETED by aroX123
Thank you very much for the progress report and for giving the knowledge to users about how things goes on, what you did and future plans. Good move!
A progress report!!!! :static:

Thanks very very much for the update Scawen
Nice to hear some news

But 1 little question:
Will there be a new smoke model coming with the new tire-physics ?
Limited setups is minnt
Nice!
Quote from silline15 :
Will there be a new smoke model coming with the new tire-physics ?

As Scawen said in the test patch thread, there will be more places which generate the smoke coming from the tyre. That's how I understood it.
Quote from silline15 :
Will there be a new smoke model coming with the new tire-physics ?

i doubt it, because thats a graphics "problem". LFS would need to be completly rewritten in DX10/11, im guessing.

ATM LFS`smoke is better than iRacings (lol)
Nice stuff coming there. Seems like another long delay of the Scirocco and company.
Quote from JazzOn :i doubt it, because thats a graphics "problem". LFS would need to be completly rewritten in DX10/11, im guessing.

I'm not convinced that the only thing between the existing 'smoke model' and a better 'smoke model' is the version of DirectX being used. I think the way it's generated from the physics is more important - where it is generated (Scawen mentioned it will no longer just be the point of contact with the road, implying the whole tyre can smoke as it would during a burnout), and the conditions that cause it to be generated. Also, the conditions for it stopping as well - lockups are great if they cause smoke, but the puffs/residual smoke on the following straight are clearly not right.

So no, I think the model will come down to a physics change rather than a graphics engine change.
Quote from Flame CZE :As Scawen said in the test patch thread, there will be more places which generate the smoke coming from the tyre. That's how I understood it.

I guess this means that there comes just more smoke, in reality it can stay much longer & can be much more intensive (+ i think tyres smoke earlier in real life but im not sure)
Quoted in the test patch forum (now is closed)

Quote from Scawen :
You are pointing out an issue with tyre heating / cooling that exists in the old physics and is not yet cured in the new physics. I found it was impossible to change a few numbers and fix that - I need to improve the tyre surface and rubber heating model.

I have an opinion, not based on science but in logic. So what I think it is the moment of sliding this friction produces a lot of heat in the surface of the rubber, so for heat up a rubber that is thermal insulation that heat need to be very very high, and produce the smoke because this surface is burning. Also for the same reason (rubber is thermal insulation) this surface cools quickly and the smoke stops.

I don't know how to simulate it but it is something like an infinite number of layers of tire, the heat passing from one to another while isolated from outside, with a determinated coefficient of wear, heating and heat transmitter

You get the idea? I mean very bad English
Nice to hear, thank you Scawen
The new smoke looks GREAT! And the suspension update is awesome!
This thread is closed

[OLD] Tyre Physics Progress Report
(4434 posts, closed, started )
FGED GREDG RDFGDR GSFDG