The online racing simulator
[OLD] Tyre Physics Progress Report
(4436 posts, closed, started )
Quote from IlGuercio :How can a video be useful?

Just compare the tyre deformation from the M3 video, and the tyre deformation from LFS, you will see that in LFS is too much exagerated.

I had answered your doubt?
Quote from Napalm Candy :Just compare the tyre deformation from the M3 video, and the tyre deformation from LFS, you will see that in LFS is too much exagerated.

I had answered your doubt?

No,it's just a qulitative observation.
Programming needs data,thus quantitative comparations.
You've just shown the consequences,based on real life observation,but didn't provide any basis for a serious calculation that could represent the model of tyre physics.
It's a bit like saying that a car goes slower than another whilst not providing anything to justify why it happens.
Quote from Napalm Candy :Just compare the tyre deformation from the M3 video, and the tyre deformation from LFS, you will see that in LFS is too much exagerated.

I had answered your doubt?

Those tires had rim protectors (thick sidewall that protrudes out past the edge of the rim). None of the tires in LFS have that feature. The thicker sidewall would cause less deformation than one with out the thick sidewall. Also, the person driving might not be pushing to maximum G like we do in LFS, there is no way of knowing the amount of flex in the video is the limit or not. Using real data gathered by tire manufacturers is the only accurate way to go about coding a new deformation model.

The doubt is valid.
Quote from IlGuercio :No,it's just a qulitative observation.

Ok, now search for a Formula1 Video and see the tyre deformation and compare to BF1

If you can't see any difference I don't know what think...
Quote from Napalm Candy :Ok, now search for a Formula1 Video and see the tyre deformation and compare to BF1

If you can't see any difference I don't know what think...

So,are you saying that you can extract a magic formula out of a video and make the tyres work properly?
I doubt so...devs probably knew that LFS tyre model was not so accurate but probably they couldn't have time or sources to investigate about where to fiddle to have proper grip.
It's not like "Hey LFS,i discovered your tyres grip too much,so watch the video and learn to make them work for good".
If it just was that simple...
Im not saying you're wrong,but you have to take into account that different cars,rims,tyres,temperatures and tarmac and god knows what can affect grip in such a big amount that a video can really tell nothing about it.
If you can extract formulas out of those videos,well,have fun and be helpful to devs
Did i explain myself well?
Am i typing japanese?
Quote from Napalm Candy :Ok, now search for a Formula1 Video and see the tyre deformation and compare to BF1

If you can't see any difference I don't know what think...

Don't forget one of the main side effects of the 'current' physics is that it looks very likely that the outside tyres in a turn are able too take more load than they should, which would naturally equate to more tyre flex.

We can be fairly certain that 'tyre load sensitivity' is one of the key areas that Scawen has/is working on in the new physics.

Also as much as we go on about LFS physics tyre design within LFS, they would still be a rather simple and unrefined tyre compared to even performance tyres in real life let alone formula 1 tyres, much like the new ESP in the VW will be a much simpler version of the real thing
Is the CMX viewer updated now?
Quote from legoflamb :I agree that camber adjustments are available on some road cars. Especially ones similar to what is available in LFS. Camber adjustment is part of the alignment. I am sure the range does not allow for +/- 4 degrees of camber, but the option is available.

Here are some cars that have slight camber adjustments from stock that I have come across on the internet:

Subaru Impreza
Nissan 240SX
Mazda RX7
Mitsubishi Starion
Honda Civic

I am glad that the setup limitation will most likely be a server-side option.

I.M.O.:

The road cars in lfs are spec class cars that allow adjustments to the suspension.

For calculating tyre-pressure the camber-angle is also important.
From that they calculate the camber-factor. The reference-pressure has to be multiplied bij the camberfactor. Up to 2 degrees camber-factor stays 1 . and maximum is 4 degrees where camber-factor is about 1.14. so the end pressure has to be multiplied by 1.14. 3 degrees about 1.06 .
cars with speed over 270km/h maximum 3 degrees adviced.
aspect ratio under 30% (235/30 R 18 for instance) maximum 2 degrees so camber factor stays 1.
For in a came mayby you can chanche the formula to a sinus curve or something to calculate the exact effect to the grip and temperature.
The basic formula is made in 1928 for diagonal tires and only the power in it chached from 0.58 to 0.8 for calculating load from pressure.
basic formula is A=(C/D)^125 * B* E. to calculate needed pressure from load and C=(A/(B*E))^0.8*D for calculating load from pressure.
A= actual pressure or needed pressure
B= reference pressure of tire ( not the maximum pressure, if you see "at xxxkPa( cold) "then that is the reference pressure)
C= Actual load or load that can be bared at the actual Pressure( A)
D= Referce maximum load on sidewall in KG LBS or the loadindex
E= camberfactor that can be calculated with the camber-angle

The Load percentage I mentioned in post# 170 , and the bouncing border of 85% could also be worked in the sensors or the game, so you feel more vibration when your Load-percentage is to low, but then also in a vloating schale.
All waiting for this ...
sweet thanks for the report. its funny i was going to open a post and ask for a report on this haha
Quote from IlGuercio :So,are you saying that you can extract a magic formula out of a video and make the tyres work properly?

Really you read this from my post? I'm wizard now! WOW!

Man you don't has sense
@ LFSDevs, thanks for listening to the community and providing us a status report Please continue doing so in the future!

Quote from Matrixi :I think a lot of people are expecting a bit too much out of limited setups. If someone is fast as hell, a limited setup won't stop him being fast.

Yup many say the same thing, you're assuming all of us are bitter envious pricks, and even if there sure are ppl like that, you shouldn't lead or be led into thinking that's the majority.

As of now road cars can't be raced online with realistic sets, unless you're an alien and then maybe you can keep the pace of a midfield guy with a funny set.

We just want to be able to race road cars that feel like the ones we drive every day. Driving some weird contraption which has no relationship whatsoever with reality? there are plenty of racing games just like that out there
are you saying i am fast in real life

no i am not mocking you in any way or trying to prove you wrong, but if you say yes then i will be extremely happy
Quote from logitekg25 :are you saying i am fast in real life


Your bus driver says, he drives every day at the limit, so : Yes you are fast !
Quote from NightShift :you're assuming all of us are bitter envious pricks

Not at all. All I can say is think about iRacing and the beginner Solstice class which is pretty much a limited setup. It's racing just like it is with the advanced Solstice, except it's simply a bit worse to drive than it is with your own tweaked setup. Even though the advanced class is quite limited aswell, there's still enough options to tweak the drivability for your own taste.

People keep blaming unrealistic setups being too fast in LFS.. how about sorting out the physics engine so those unrealistic setups would become awkward (or mechanically impossible) to drive like they would be in real life, instead of overcoming the physics problem with limited setups?

:twocents:
Quote from Matrixi :How about sorting out the physics engine so those unrealistic setups would become awkward to drive like they would be in real life, instead of overcoming the physics problem with limited setups?

That's my preference also and is what we will be edging towards with the new physics arriving here soon.

But I think we will still be some way off from where I hope we will be one day. The most needed enhancement in physics going forward is to get away from the perfect static environmental physics we currently have. Particularly important imo is dynamic rubberised racing line that can also be set like wind atm, variable marbles and dirt off the racing line, track temperature variation in real time along with differences in shaded and sunny positions, changing light conditions both from time of day and moving cloud cover, rain, etc...

Having said all that though I do want to see the option for limited set ups in the road cars for one main reason, it would be a good entry point for new players as they can feel confident that others are running similar set ups to what they have and can learn how to make set up changes at a basic level rather than be thrown into the deep end.
Totally agree with Glenn67
Quote from Ziploc :There are McPherson struts that allow camber adjustment:
http://www.perrinperformance.c ... roducts/139/744_large.jpg

And as far as I know, they are pretty common.

Trailing arms/solid axles are a little more problematic. But multilink trailing arms sometimes allow camber changes as well.

The average car is going to have camber adjustments. And cars in LFS aren't 'repmobiles' anyway. So I would be expecting some camber changes to be allowed.

I own a 2001 VW Jetta (MKIV, A4 chassis) that has mcpherson struts and a trailing arm solid rear axle and the 'factory' adjustments are very limited. The shock inserts into the control arm, and is not attached in the the manor of the above picture.

The best you can do is have a local shop drop the sub frame and position it(usually with the use of a jig) to give the most caster/camber which is around -.8 on both sides or so (specified range is 0 to -1.1 IIRC). i have a friend that calls this a 'performance alignment' and it will usually cost you ~$100 vs $40 for a standard toe only alignment, but that is besides the point.
There are kits that replace the plate where the strut attaches to the strut tower that give 1-1.5* more of negative, so you are looking at maybe 2.5* negative camber.

I think these limited setups will be great instead of using a tracker tool to force a certain set to be used, I just hope that the ranges are realistic. Will there be ranges that can be set server side? or will it be fixed values/ranges decided by the developers?
Quote from Glenn67 :

But I think we will still be some way off from where I hope we will be one day. The most needed enhancement in physics going forward is to get away from the perfect static environmental physics we currently have. Particularly important imo is dynamic rubberised racing line that can also be set like wind atm, variable marbles and dirt off the racing line, track temperature variation in real time along with differences in shaded and sunny positions, changing light conditions both from time of day and moving cloud cover, rain, etc...

I think some of the later stuff is kinda far away TBH, but a dynamic grip level and "clag" simulation would be really cool. As is you can go wide and not really be penalized too much for it. In real life if you get off-line way out in the heavy marbles it feels like you're driving on gravel.
#195 - Dac
A year or so ago I started a thread in the improvements/programming section asking about the posibility of a traction circle graph. However I was told this would be far too difficult to code (if not impossible!) so my question to Scawen is will this new tyre physics model be able to support a circle of forces graph? I believe this would be a extremely good training tool to teach people how and why they aren't using the cars performance to its maximum (or rather why they spun off lol).
Quote from Dac :A year or so ago I started a thread in the improvements/programming section asking about the posibility of a traction circle graph. However I was told this would be far too difficult to code (if not impossible!) so my question to Scawen is will this new tyre physics model be able to support a circle of forces graph? I believe this would be a extremely good training tool to teach people how and why they aren't using the cars performance to its maximum (or rather why they spun off lol).

Someone said that would be hard or impossible to do?

I agree it'd be useful. Possibly more useful would be a g-g diagram. Those aren't too hard to keep an eye on while driving the track. A couple years ago I tried that. It'd be great for replays but it's hard to watch everything while driving hard. The center circle is a g-g diagram. The other circles are the traction circles (true circles in this case but elipses would be easy to do too). The blue bars indicate slip ratio. The green lines in the tire circles show the slip velocity vector at each tire (slip angle indicators, basically).

http://performancesimulations.com/files/ToddSim-GenIV-10.wmv

Take opinions on this sort of thing from people that have never done it with a grain of salt. Generally speaking, folks usually don't know what they're talking about
Quote from Dac :A year or so ago I started a thread in the improvements/programming section asking about the posibility of a traction circle graph. However I was told this would be far too difficult to code (if not impossible!) so my question to Scawen is will this new tyre physics model be able to support a circle of forces graph? I believe this would be a extremely good training tool to teach people how and why they aren't using the cars performance to its maximum (or rather why they spun off lol).

Quote from jtw62074 :Someone said that would be hard or impossible to do?
[...]
Take opinions on this sort of thing from people that have never done it with a grain of salt. Generally speaking, folks usually don't know what they're talking about

You mean something like this:
http://www.lfsforum.net/attachment.php?attachmentid=42594 (coloured slip angle)
http://www.lfsforum.net/attachment.php?attachmentid=42593 (coloured slip ratio)
?

This can be very easily extracted from RAF files (with RAFTyreExtract) and is in this case shown with MultiDim (that plots arbitrary data on a 2D pane). If someone told you this is incredibly complicated to do then they 1) had no idea whatsoever and 2) sound like the usual LFS "it's not implemented because it's too complicated" apologists.
arent you kind of limited to purely pacejka type models with that sort of display though? i suppose with any sort of fem model and presumably with other physical models theres no way of really calculating where the limit of combined traction is at any given moment and vertical load
Android: Yes, that's it.

Shotglass: You're not limited to an empirical model to get friction circles (mine aren't empirical). If you know the friction coefficient and vertical load (max force), you have the radius so can draw the circle. If you have separate longitudinal and lateral friction coefficients you can draw an ellipse just as easily. Depends on how the particular model works, but there's probably a way to work it in almost anywhere. FEM would be more challenging, but nobody's using those in sims.
Great to see progress reports! Thank you!
This thread is closed

[OLD] Tyre Physics Progress Report
(4436 posts, closed, started )
FGED GREDG RDFGDR GSFDG