The online racing simulator
Physics model - Major issue that needs resolving?
1
(28 posts, started , go to first unread)
Physics model - Major issue that needs resolving?
Hi, I've not come up with a strong enough argument nor found the will to actually right this down because there's every chance the dev team would turn a deaf ear, but I feel in order for the Sim to progress and gather a wider array of drivers and lesser skill gap, the following has to change.

What should be changed, you ask?

Powersliding to gain time. "Mario Karting", drifting, soap bar cornering, however you name it.

B2r Rayman, [FM] GTi Thursday BL1X

This may be a constant annoyance I am sure for those drivers who race in multiple sims - having to adjust the driving style to suit the physics model that encourages an unrealistic handling of the car is definitely not right.

What is powersliding/mario karting, you may ask?
In LFS, slower cars, or cars with lesser downforce require the driver to rotate the car using the rear in order to maintain a higher corner speed and exit. This is only possible with a set-up that is remotely fast. You may see this in tracks with quick, sweeping corners, or in slower cars, particularly the XRG. If you maintained a more realistic and normal driving style by entering the corner neutral, you will often see your time delta fade away compared to the faster men and women. The faster you go, the more exaggerated this maneuver becomes.

Faster cars are not spared either! Seen a BF1 lap of a fast driver around Blackwood, FOX at South City, or a FXR on Westhill yet? Same thing, same problem.



On the contrary, yes, you can indeed drive the cars with a traditional driving style, being slow in, fast out, rear following the front, and still be quite fast on your own accord. But, would you be nearly as fast as the world record holders or faster racers?
Well.... no. It's physically (heh, pun!) impossible in LFS.
To add on to that, in certain tracks, the abuse of the handbrake to lock up the rear axle to promote more frontal rotation just makes the entire immersion and realism of driving quickly in the sim just a chore and not a fun experience for those whom stay true to their driving style and are unable to change it to suit what LFS offers currently. Every car just feels like it wants to be trashed around like a Rotax Go Kart.

For me, after several years of racing here on LFS, I have somewhat grown accustomed to the powersliding style that gains time. However, I find it hard to transition quickly from car to car or track to track and be quick on the spot - even when I consider myself pretty adaptable to different cars. Coming from a good racing knowledge background and watching tons of races across several leagues, it's hard to wrap the brain around the LFS car handling at times.


To end off, this is an example of how cars should behave, and what LFS should aim for in terms of realistic handling.

I'd rather have an slightly understeery, more true to "real-life" base handling physics that set ups could work to improve, rather than a model that encourages the exploitation of lateral grip. Where's the slip limit gone for the rears in LFS?


Slow-ish Toyota GT86s around a fast sweeping Phillip Island

What are your thoughts, and do you think the next major update for the tire physics/handling model should resolve or make changes to this issue?
Agreed. Hopefully that will be fixed along the upcoming tyre physics rework. Good of you to have BL1R's C1 pictured considering that it's where handbrake abuse is the strongest for XRG hotlaps and that slide f*ckery is also pretty high there, though not as much as it is on the hairpin.

Those GT86 race cars really remind me that this game badly needs some proper racing GTi cars with slightly more power, stripped-out interiors, rollcages and semi-slicks. Like the MX5 cup cars. We have street cars, open-wheelers and high power grand touring cars, now we're missing the low power cup cars and LMPs. See you in 2030.
It also kills the tyres so often people complaining about bad tyre physics, tyre heating while they continue to drive slidey like this in race aswell. Just change the camber maybe more pressure, toe etc. And off you go same driving style applied for hotlap and race.
ese Rayman ahí te ha pillao el radar
you really have a lot of reason as to the grip of the car on the track
#6 - Racon
I'm a little rusty on tyre physics, but I think you're saying that the fall-off in grip as the slip angle goes over the threshold is not enough, or doesn't come quickly enough, or both?

I appreciate that there are other dimensions that will complicate the picture, but if we graphed slip angle to grip, you're saying the curve would be wrong at the higher end of slip... but what shape is it and what should it be?

Does that mean we're also getting a little more turning than we should out of riding the edge of an understeer?
It's been known for a while that the game's current approach to tire load sensitivity makes the inside tires grip less in a turn than they're supposed to. It's the main reason that VW everyone gripes about never got added (even with the exact same settings, it didn't handle like the real thing at all). The new tire physics in the works is largely meant to fix that, and I have a feeling this quirk about the current physics is contributing a lot to the sideways driving required for WRs.

Now, I'd need to type a massive wall of text to adequately explain this, but going sideways would basically put a little more weight on the inside rear tire than neutral cornering would. Think of it as the weight transfer becoming less side-to-side and more inside-front-to-outside-rear.
When you consider that the inside tires can easily lose lots of grip with the current physics, every little bit counts, so those aggressive turn-ins would effectively maximize rear grip if done in exactly the right way. You get to quickly rotate the car, quickly stabilize that rotation, and then use more of the inside tire's grip to get yourself around the turn.
Not really the easiest thing to pull off, but it'd give you a clear advantage over neutral cornering if this is exactly what's going on with the tires.

Okay, that ended up being a bit longer than I wanted. But basically, I think this'll go away as soon as the new physics are patched in. It's all to do with some subtle flaws in the current tire physics that took a while to spot, and once those flaws go away, the cars should start driving a bit more normally.
I have a problem in terms of tires since they wear very fast on tracks with many curves and I have the lisors 4 that they recommend (version 0.6E)
Quote from MeteorIDW :It's been known for a while that the game's current approach to tire load sensitivity makes the inside tires grip less in a turn than they're supposed to. It's the main reason that VW everyone gripes about never got added (even with the exact same settings, it didn't handle like the real thing at all). The new tire physics in the works is largely meant to fix that, and I have a feeling this quirk about the current physics is contributing a lot to the sideways driving required for WRs.

Now, I'd need to type a massive wall of text to adequately explain this, but going sideways would basically put a little more weight on the inside rear tire than neutral cornering would. Think of it as the weight transfer becoming less side-to-side and more inside-front-to-outside-rear.
When you consider that the inside tires can easily lose lots of grip with the current physics, every little bit counts, so those aggressive turn-ins would effectively maximize rear grip if done in exactly the right way. You get to quickly rotate the car, quickly stabilize that rotation, and then use more of the inside tire's grip to get yourself around the turn.
Not really the easiest thing to pull off, but it'd give you a clear advantage over neutral cornering if this is exactly what's going on with the tires.

Okay, that ended up being a bit longer than I wanted. But basically, I think this'll go away as soon as the new physics are patched in. It's all to do with some subtle flaws in the current tire physics that took a while to spot, and once those flaws go away, the cars should start driving a bit more normally.

Any source on that, i know they said they had not been happy with the physics but nothing in much detail Smile

I only drive FBM and quite honestly do not see such an issue, obviously quite different than the road cars so expected

@microspecV Showing a still image does not really show what you mean in action, any chance of showing it being done and showing the time advantage given in any possible way Smile thanks dude
Quote from bishtop :Any source on that, i know they said they had not been happy with the physics but nothing in much detail Smile

I only drive FBM and quite honestly do not see such an issue, obviously quite different than the road cars so expected

@microspecV Showing a still image does not really show what you mean in action, any chance of showing it being done and showing the time advantage given in any possible way Smile thanks dude

It doesn't seem like an issue for those very quick drivers who have already grown accustomed to the driving style the physics and tire model require (FBM is a great example).

I'm not in town today but I'll create some kind of visualisation to compare with, but essentially, if we take the FBM as a case study, a downforce equipped car should not really be driven consistently through corners by powersliding and having a neutral steering position. These cars should, in theory and real life, be cornering with the rear following the front in a grip-reliant manner, instead of causing the rear to slip slightly in order to pivot the rear around the front (acting as a pivot to shorten the radius of the corner and converting natural understeer into controlled oversteer, which is unrealistic).

A visual example would be Turn 1 and 2 in Blackwood Historic, where WR setters and very fast drivers essentially drift the car through the corner with little to no steering input, apart from a small turn or scandinavian flick at the start accompanied with rear brake. It's just not... correct. It may be fun and faster, but if LFS is striving for a physics and tire model that best replicates a real-life scenario then it's pretty far from it. The FBM should actually be a little understeery, and the XRG shouldn't even be sliding on any track to gain time, to be fair.

It's pretty evident in lower powered cars where they have a grip defiency and this exploit of the tire model affects far more greatly. Similar to what you quoted METEORIDW stated yesterday in greater detail.
Its evident even in FBM, i dont know how you can say its not.

To better demonstrate MicroSpecV's note, I'll just link this video here. I made it cause that stupid kerb spun me, but case in point mr. Nova wants to make is the first turn in the video.

Driving it like a rally car.

https://www.youtube.com/watch?v=dwrmeiaPgNk&feature=youtu.be
And as for FBM - yesterday we had MRC KY3 race. You take advantage of high rear grip while sliding in turn 1, turn 2, turn 3 and last turn.

Turn 2 being the most evident, step on brake 50 meters before turn, have your setup that way that just as you want to turn in rear tyres lock a bit, lift off brake and watch the car slide perfectly into a turn while you instantly can go on full throttle having the car in superb position just as it stops sliding with rear.

Its just a bit trickery with brakes and differential.
Quote from bishtop :Any source on that, i know they said they had not been happy with the physics but nothing in much detail Smile

Scawen's discussed it a few times on the forums. Here's one post where he specifically goes into it Smile
https://www.lfs.net/forum/post/1863422#post1863422
Quote from Scawen : The problem is the tyre physics. It was impossible to make the Scirocco handle well with the current tyre physics, if we used realistic suspension settings. So I went on a long mission into tyre physics. That is the only reason we have not released the Scirocco yet, and that reason has not yet changed.

EDIT: If you were asking for a source on inside-tire grip being a problem with the Scirocco, that's more something I inferred from what's observable in the game. Other people have mentioned the issue with the inside tires before (particularly how easily you can wheelspin in the UF1), and having much less grip on one side of the car would drastically worsen handling, especially if it's a FWD like the Scirocco.
Quote from MeteorIDW :EDIT: If you were asking for a source on inside-tire grip being a problem with the Scirocco, that's more something I inferred from what's observable in the game. Other people have mentioned the issue with the inside tires before (particularly how easily you can wheelspin in the UF1), and having much less grip on one side of the car would drastically worsen handling, especially if it's a FWD like the Scirocco.

Yes and no. Sure, the tyre physics are to blame, but not exactly for the reasons you mentioned. I wouldn't exactly call it a major issue, but more along the lines of the physics not being punishing enough to exaggerated setups. The physics allow for locked diffs (even the setups using something else are mostly cranked down so much we might as well call them locked) and ridiculous camber angles to create angular momentum, leading to the powerslide-cornering MicroSpecV critized. I am aware that a few people welded their diff in real life, but those were either Autocross or Offroad. I suspect neither tyres nor drivetrain would survive longer races on tarmac.

The reason you don't see it as much with the FBM is because the downforce and the rearward weight bias combined with the quite weak engine probably create too much traction to slide the car properly. The UF1 spins its inner tyre because it has an open diff - so all the torque is transferred to the unweighted tyre. Were it any more powerful it would probably smoke the tyre in record time, alas it bogs down instead...
Quote from MeteorIDW :Scawen's discussed it a few times on the forums. Here's one post where he specifically goes into it Smile
https://www.lfs.net/forum/post/1863422#post1863422

EDIT: If you were asking for a source on inside-tire grip being a problem with the Scirocco, that's more something I inferred from what's observable in the game. Other people have mentioned the issue with the inside tires before (particularly how easily you can wheelspin in the UF1), and having much less grip on one side of the car would drastically worsen handling, especially if it's a FWD like the Scirocco.

thanks for the source of information, as stated by others especially the post below when scawen discussed it, probably trying to make the handling better than it was on a sh1t car. Have you seen easy it is to wheel spin an austin mini and id guess that the issue with the inside of the tyres would be effected by setup or whatever such as toe in, caster ect, not that i have personally noticed one that has altered my race in any way

@bbman you say the FBM being underpowered yet there is enough power to slide or enough things to alter to get it sliding, i think its due to the suspension that makes it keep grip mostly, set the suspension to its highest and it still handles ok.many factors that could cause issues with the inside of the tyre either heating up quicker such as toe in

thanks, bish
Quote from bbman : The UF1 spins its inner tyre because it has an open diff

I'm certainly aware of that, but some people have said they find it exaggerated even for an open diff. Locked diff setups being so effective is also something I chalk up to the inside tires not gripping as much as they should.
But even if there really is an issue with the inside tires, you're right in that I'm probably focusing on it too much. You mentioned ridiculous camber angles being fine, and one of the old news posts has certainly discussed the excessive camber forces you get with the current physics. It's really just a whole bunch of minor issues that only become apparent when people start driving at WR pace.

Quote from bishtop : you say the FBM being underpowered yet there is enough power to slide or enough things to alter to get it sliding

I think he meant to say the engine can't easily overpower all the mechanical and aero grip on the FBM, allowing you to run the diff more open. High-grip cars in real life can certainly do this, so this is something LFS gets right for sure.

At any rate, like I said before, I think all these minor issues are going away as soon as the new physics come in. A few of the cars will probably be much more fun to drive as well.
Quote from MeteorIDW :I'm certainly aware of that, but some people have said they find it exaggerated even for an open diff. Locked diff setups being so effective is also something I chalk up to the inside tires not gripping as much as they should.
But even if there really is an issue with the inside tires, you're right in that I'm probably focusing on it too much. You mentioned ridiculous camber angles being fine, and one of the old news posts has certainly discussed the excessive camber forces you get with the current physics. It's really just a whole bunch of minor issues that only become apparent when people start driving at WR pace.


I think he meant to say the engine can't easily overpower all the mechanical and aero grip on the FBM, allowing you to run the diff more open. High-grip cars in real life can certainly do this, so this is something LFS gets right for sure.

At any rate, like I said before, I think all these minor issues are going away as soon as the new physics come in. A few of the cars will probably be much more fun to drive as well.

ok you are quite right you just have to wait for the update and correct those small mistakes but personally they do not bother me so much but I understand that most players expect more realistic physics
Quote from bishtop :@bbman you say the FBM being underpowered yet there is enough power to slide or enough things to alter to get it sliding, i think its due to the suspension that makes it keep grip mostly, set the suspension to its highest and it still handles ok.many factors that could cause issues with the inside of the tyre either heating up quicker such as toe in

If you just want to slide the car about there certainly are more than enough options, but this is about being faster using excessive yaw angles. I've not driven the FBM competitively for quite some time, but I seem to remember the car being so overgeared anywhere other than SO that any more than a few degrees slip angle lost you a couple of kph on exit even with minimal wings...

Quote from MeteorIDW :I'm certainly aware of that, but some people have said they find it exaggerated even for an open diff. Locked diff setups being so effective is also something I chalk up to the inside tires not gripping as much as they should.
But even if there really is an issue with the inside tires, you're right in that I'm probably focusing on it too much. You mentioned ridiculous camber angles being fine, and one of the old news posts has certainly discussed the excessive camber forces you get with the current physics. It's really just a whole bunch of minor issues that only become apparent when people start driving at WR pace.

Not disagreeing with you just because, I am with you that tyre load sensitivity seems off. The root cause for why you feel the open diff exaggerated in my opinion is just the opposite. Both UF were often the topic of debate, seeing as how easy they were transformed to 2-wheelers just by hard cornering, or with the help of a bump in the middle of a turn... They just love to flip!

In the end, we'll see whether our hunches are right when the new physics are released. While there is definitely things to be improved upon, I feel MicroSpec's post makes it seem worse than it is - and surely not something that should spoil good, close racing! Smile
Quote from bbman : While there is definitely things to be improved upon, I feel MicroSpec's post makes it seem worse than it is - and surely not something that should spoil good, close racing! Smile

Well, for the faster drivers who can adapt to cars and tracks relatively quickly, it isn't much of an issue and hence you may feel it doesn't warrant a huge discussion or resolve. However, for slower or newer LFS members, it may be really puzzling and frustrating wondering where the laptime is slipping from and how drivers are able to seemingly caress a car through a corner with ridiculous speed and a controlled slide.

I know this because I felt this frustration myself back then and constantly see drivers get trumped by the unrealistic car handling.

If it's racing amongst the best drivers then I guess it wouldn't spoil the racing, but let's face it, LFS just isn't as big now and we have a greater mix of drivers with varying skill level than ever before. It's only right that we give everyone a handling model that cannot be exploited and is able to be driven as close to what a real car would handle, or at least, what good handling feels like to the average joe.

You can see this in action by viewing any of the "professional" sim racers on Youtube who review LFS and drive, they are only at best just beating Pro AI and what do you know - all of their driving styles are the point and turn, non sliding style that is present in other sims. They drive the LFS vehicles like how they best know a car should handle and react. And none of them would instintively go "hey, I think sliding around a corner is faster!". Even Jimmy Broadbent in a XRR doesn't bang in a 4 wheel powerslide at BL1 Turn 1. Go figure..

Linking back to the original point, so yes. It does ruin the racing, sad to say. Even for a relatively experienced driver myself, I just am not able to wrap my head around mario karting through corners and have since ignored this driving style to just focus on my own style, and it worked out well, although I am never as fast as the WR or sliding drivers.




P.s It has become so annoying that I have resorted to exploiting another flaw in the handling model to basically replicate realistic handling (rear following front, no powersliding to gain time) through altering the set up drastically, and it's coming quite close to being as fast as the sliding sets, although I have yet to master this new set up trick on all cars but I can assure that it is definitely a flaw in the handling and is probably as dumb as the OP topic on the sliding advantage.

But... there's no point elaborating on this set up style because no dev has hopped into this thread yet to shed some light and I just don't want to waste time attempting to raise concern for a handling model that's quite meh.
But im interested now. What you do differently?
Quote from MicroSpecV :Well, for the faster drivers who can adapt to cars and tracks relatively quickly, it isn't much of an issue and hence you may feel it doesn't warrant a huge discussion or resolve. However, for slower or newer LFS members, it may be really puzzling and frustrating wondering where the laptime is slipping from and how drivers are able to seemingly caress a car through a corner with ridiculous speed and a controlled slide.

I know this because I felt this frustration myself back then and constantly see drivers get trumped by the unrealistic car handling.

If it's racing amongst the best drivers then I guess it wouldn't spoil the racing, but let's face it, LFS just isn't as big now and we have a greater mix of drivers with varying skill level than ever before. It's only right that we give everyone a handling model that cannot be exploited and is able to be driven as close to what a real car would handle, or at least, what good handling feels like to the average joe.

You can see this in action by viewing any of the "professional" sim racers on Youtube who review LFS and drive, they are only at best just beating Pro AI and what do you know - all of their driving styles are the point and turn, non sliding style that is present in other sims. They drive the LFS vehicles like how they best know a car should handle and react. And none of them would instintively go "hey, I think sliding around a corner is faster!". Even Jimmy Broadbent in a XRR doesn't bang in a 4 wheel powerslide at BL1 Turn 1. Go figure..

Linking back to the original point, so yes. It does ruin the racing, sad to say. Even for a relatively experienced driver myself, I just am not able to wrap my head around mario karting through corners and have since ignored this driving style to just focus on my own style, and it worked out well, although I am never as fast as the WR or sliding drivers.




P.s It has become so annoying that I have resorted to exploiting another flaw in the handling model to basically replicate realistic handling (rear following front, no powersliding to gain time) through altering the set up drastically, and it's coming quite close to being as fast as the sliding sets, although I have yet to master this new set up trick on all cars but I can assure that it is definitely a flaw in the handling and is probably as dumb as the OP topic on the sliding advantage.

But... there's no point elaborating on this set up style because no dev has hopped into this thread yet to shed some light and I just don't want to waste time attempting to raise concern for a handling model that's quite meh.

Most new drivers lose time due to factors such as incorrect braking points and bad lines and also to do with a setup that better suits there driving style. The main thing that i have been told numerous times by many of the years and even recently by my own dad is how bad a setup the default ones are, my dad for example just wants to race offline and would like something that he could put his VR headset on and drive at any pace but the default sets suck so bad to give someone a chance to do it proper

I can see that the issues your stating would/could be something are more experienced hotlapper might see as an issue or at least the more experienced

After watching a couple of the south city WR's with the XRR i cannot see what you claim being used which i would of thought would of shown what you claim,so if it was more of an advantage surely it would have been used
Quote from NumberTwo :But im interested now. What you do differently?

All to do with ride height and rear stiffness.

Quote from bishtop :Most new drivers lose time due to factors such as incorrect braking points and bad lines and also to do with a setup that better suits there driving style.

Yes, but the problem itself is the setups that suit their driving styles may not be as fast as the setups getting WR or faster laps (the set ups that encourage rear sliding). For another example, in WE1R XRG, you have to slide the car literally every corner in order to get under sub 2:03. A new driver may choose to set his car up different from said Default setup, in order to make it handle more stable, but, it may not be the fastest way.

That's the point and issue I'm raising. The fastest set up shouldn't be ones that involve any degree of sliding/mario karting/powersliding/lateral slide at all. It's not realistic and it ruins racing.

About the XRR, did I mention it in South City? I believe I only mentioned FOX at South City Tongue
Now hold on, something doesn't add up here or I am missing a big part...

You DO know peak traction occurs at a couple degrees slip angle? We can agree that LfS probably is more lenient when you exceed the optimum and currently doesn't punish you in another way. Still, have you seen the fast guys in any other sim? They're sliding around just as much, because it IS the fastest way. So if you say Broadbent doesn't slide it's because he hasn't spent enough time with the sim, because he definitely does it just the same in iRacing and AC, last I looked. Without the usual death spin for going over "the limit" and LfS' infinite setup options, of course you will spend much more time around peak traction. If you blame LfS because you decided that an understeery setup is the only right thing I fear you're deluding yourself...
Quote from MicroSpecV :All to do with ride height and rear stiffness.



Yes, but the problem itself is the setups that suit their driving styles may not be as fast as the setups getting WR or faster laps (the set ups that encourage rear sliding). For another example, in WE1R XRG, you have to slide the car literally every corner in order to get under sub 2:03. A new driver may choose to set his car up different from said Default setup, in order to make it handle more stable, but, it may not be the fastest way.

That's the point and issue I'm raising. The fastest set up shouldn't be ones that involve any degree of sliding/mario karting/powersliding/lateral slide at all. It's not realistic and it ruins racing.

About the XRR, did I mention it in South City? I believe I only mentioned FOX at South City Tongue

The setups will obviously make a difference or might not be as fast as WR, buts that is reality. You either can have a car that sticks to the road like glue or one that will stick less giving possible faster cornering as it wont take you too wide on exit but more of a chance of losing control

@MicrospecV I used south city and the XRR due to you mentioning that car and south city because it has some tight corners that if it was going to be an advantage it would be used Smile

+ Are you on about the BL1 WR with the XRR ? If so he does a quick flick of the rear to pull the rear around, thats realistic(guess you never watched the BTCC) and not even much of a slide even if one at all
Quote from bbman :snip

eh, might be, I wrote that late at night whilst multi tasking so I may have lost the point there. Nevermind, just ignore that bit, was trying to explain the OP but guess I strayed away from the purpose. Did I Say That?

When I meant by slide I don't mean a little oversteer or exit loss of traction, I mean something more similar to an XRG in Turn 1 at BL1 or Turn 2 + the middle sector in WE1R. That kind of controlled, extended sliding.
1

Physics model - Major issue that needs resolving?
(28 posts, started )
FGED GREDG RDFGDR GSFDG