The online racing simulator
How complicated can a tire model become
(143 posts, started )
Quote from CarlLefrancois :eloquent.

i assume you mean that in the spirit of putting the priority on the online racing experience, getting the model closer to reality is better than adding more diversity.

Exactly. I'm pretty sure Scawen wants that too and his work will be top quality.
Quote from Sobis :Exactly. I'm pretty sure Scawen wants that too and his work will be top quality.

we all want that

i think the question is: can it be done.
Although the thermal camera looks cool () and it is what it looks like, aka factual situation, the truth is that even the large difference in heat displayed on the video has negligible effect on grip performance and as such having the same display information in LFS's F9 instead the current one (i'm guessing for most of us had that idea crossed our minds when we saw the video) would be useless. I'm wondering how many think or know the same? It's highly counter-intuitive.

Fun fact: we already have the same display in LFS. As and addition to current F9 info: press F9, then press ctrl+shift and hold. Now you should see the actual surface temps behaving exactly the same as in the video. But it is not the default view. Why? -I'm guessing the same logic that's behind what i said before.
Quote from Nilex :Although the thermal camera looks cool () and it is what it looks like, aka factual situation, the truth is that even the large difference in heat displayed on the video has negligible effect on grip performance and as such having the same display information in LFS's F9 instead the current one (i'm guessing for most of us had that idea crossed our minds when we saw the video) would be useless. I'm wondering how many think or know the same? It's highly counter-intuitive.

Fun fact: we already have the same display in LFS. As and addition to current F9 info: press F9, then press ctrl+shift and hold. Now you should see the actual surface temps behaving exactly the same as in the video. But it is not the default view. Why? -I'm guessing the same logic that's behind what i said before.

True, but I posted this because it looked really interesting.

Anyway, the ctrl+shift in F9 shows the temperatures from the outside of the tyres and without pressing it the inner, right?
Correct.
#56 - col
Quote from Bluebird B B :That is where quick fixes are needed

Hmm, sounds like an attractive idea, but I doubt it would ever be as simple as you make it sound.

Quick fixes tend to cause more work in the long run. They also tend to introduce bugs and odd behaviour.
Quick fixes are the sort of botch/hack that are used by teams under deadline pressure, which is why many games are seriously flawed when they come to market. That's one of the reasons why LFS exists - Scavier wanted to escape from that approach to game development.
Quote from CarlLefrancois :for a second there i thought that was the HeatCam tm from the new tire model 2 (NTM2) in iRacing 2*. oh wait...

That is their plan. They're gonna introduce "HeatCam" instead of a decent physics model.
Quote from col :Look-up-table(LUT) based systems are notoriously crap at covering the complete range of situations. They're great for a simulation where expected behaviour stays away from the physical limits of the system being modeled, each variable staying within a small range of stable fairly linear values, e.g. simulating a passenger jet in a flight sim where you don't go even close to stalling etc.
A racing sim is the worst possible for LUTs, a driver spends the whole time as close to the physical limits of the system as possible - for this to work well with a thorough LUT system without fudging, you would need astronomical amounts of data. That's why LUT based systems have unnatural feel during physical state transitions. LFS may not always seem 'realistic', but it always feels organic, 'natural' and predictable.
Scawen wants more realism without losing the natural feel. Can't do that with LUTs

LUTs are theoretically unusable as you said because they have spaces in coverage but model needs to be at 20kHz (if thats true, it has alot less than 50ms to calculate because others calculations needs to be done also in the meantime) so I doubt he have much time to do some serious math. Still he can use a lot of memory so he can make pretty decent LUTs for the calculation. Also some interpolation can be used.

Maybe Iam wrong but I think that it uses some conditions, precalculations, LUTS and then calculations and conditions to get the result.

This way it can be really fast but it can be only an aproximation of the bench model he uses.
Quote from L@gger :This way it can be really fast but it can be only an aproximation of the bench model he uses.

It would be good to know how "slow" bench model currently is.
As i understeand, current bench model is best model Scaven could write, and at this stage he tries to simplify it for best cpu performance.

As i wrote earlier, maybe pushing calculations to GPU would be good idea (if possible).
These models could be very slow. They take all possible parameters maybe some that later wont be calculated as their effect wont be noticeable.
It can take minutes to hours to calculate models that reflect maybe few seconds to lfs because the bench model should be physical based and optimalization isnt much priority.

You cant just use a bench model to run real time and want gpu to calculate it. There can be so much time wasting mathematical operations that arent really noticeable in the result.

You can make similar models in some mathematical programs and the c code implementation is more like mathematics and coding play.
Quote from Simpson :It would be good to know how "slow" bench model currently is.

He says in one of the press releases that it takes the bench several seconds to generate one second's data, so a minimum of 400 times slower than required for real-time (3 times slower, 128 times as many wheels to calculate).
#63 - col
Quote from Dygear :http://www.youtube.com/watch?v=PsoGzvWyTjU

Just look how fast the thermals on the tires change. It's incredible.

It's real pretty an' all, but unless we have a key that maps the colour to temperature, it doesn't really tell us anything we didn't already know.
All it means is that the surface rubber of the tyres cools when wind chill of 100-200mph hits it (the surface of the tyre is also rotating at that speed, so I guess that increases the chill factor?). When the driver hits the brakes, the wind chill stops and the surface is quickly reheated by friction due to braking combined with the heat still stored in the rest of the tyre's mass and heat from the brake disks.
It doesn't tell us how many degrees of cooling or reheating there are, or how deep within the tyre surface the heating and cooling happen.

When you think about it, everything else happens very quickly on an F1 car, so it would be very surprising if the tyre thermals didn't.
Quote from col :It's real pretty an' all, but unless we have a key that maps the colour to temperature, it doesn't really tell us anything we didn't already know.

At the race commentator said that black is ~100*C and yellow ~180*C.
Appart from that, there is no graph that will show temperature with colour.
Poor choice of colour scale. It shows nothing most of the time, and then spikes when cornering. So it tells you nothing except that they run a bit of camber - like every racing car in the world.
Here's a longer video, mirrored Monza seems weird. http://www.youtube.com/watch?v=aCph80AO_2s

I think they obviously chose the scale which would be most dramatic. Clearly the surface isn't at background temperature when it's black, tyres don't lose ALL temperature after every straight.
Probably the only thing of some interest is the way you have a cooler band in the middle of the tyre sometimes, I'm assuming it's due to sidewall deformation?
Quote from sinbad :
Probably the only thing of some interest is the way you have a cooler band in the middle of the tyre sometimes, I'm assuming it's due to sidewall deformation?

Can't answer directly, but by LFS setups I've learned that this is caused by under-inflating the tires. I am not saying they have set the car up incorrectly , just noting it.

Also interesting, but expected, are the white flat spots after the locking. We really can't say anything else without a color mapping.
Quote from L@gger :There can be so much time wasting mathematical operations that arent really noticeable in the result.

this is true to a point. Newtonian physics for example give accurate results in almost all cases since these additional "time wasting operations" have almost no effect when the bodies involved are heavy and moving slowly compared to light.

when things are smaller and moving close to the speed of light -- at the limit -- Newtonian physics shows how simple assumptions fail and how important those little time wasting operations really become. (i guess in this case the additional operations are the effects on the bodies by the rest of the universe)

in general, to get to the heart of the behaviour of a real physical system, you don't have to calculate every last motion of every particle, but the smallest effects do still have to be taken into account. we recognise real physics because all these details are there.
Quote from sinbad :cooler band in the middle of the tyre sometimes, I'm assuming it's due to sidewall deformation?

no, it's a characteristic of low tire pressure.

the vertical load on the contact patch comes from a combination of sidewall stiffness and air pressure. the air pressure gives the load mostly on the center of the tire.

in this case air pressure was low so the sidewall stiffness gave more vertical load on the sides of the contact patch than in the middle. when slipping, the contact patch develops more friction force where it has more vertical load on it.
Okay but I was talking about specific model of tire that comes from a bench model. It doesnt have all calculations because of speed optimalization. Not because they are unusable but can be optimalizated or the effect is just too low to be counted in. Bench model may be more universal.

@ F1 thermal camera.
Finally! Maybe F1 teams just dont want to have different scale because other teams could make some assumptions?
who ever believes in the next 2-3 years the update physics tires.
I think the Skaven again took a great vacation for a long time.
Quote from Dmitry[RUS] :who ever believes in the next 2-3 years the update physics tires.
I think the Skaven again took a great vacation for a long time.

k
Quote from Dmitry[RUS] :who ever believes in the next 2-3 years the update physics tires.
I think the Skaven again took a great vacation for a long time.

Something isn't working out.
IMO no updates will coming soon. AND the Rocco will never appear in lFS. I think, it was made only for the GC event, and nothing more.
Quote from ASR Sanchez :IMO no updates will coming soon. AND the Rocco will never appear in lFS. I think, it was made only for the GC event, and nothing more.

I've just seen a round of the Scirocco cup on the telly, complete with magic boost buttons... Meh

Rockingham I'm still looking forward too...

How complicated can a tire model become
(143 posts, started )
FGED GREDG RDFGDR GSFDG