Hello guys, been loving the sim for a long time (since demo days )

I apologize for raising subjects we've already discussed.

My question for Scawen is; Are there and if there are, what are the plans to improve physics and
setups? (tire,suspension,chassis,differential,aerodynamic)

My question for the community is; does the majority care
about the physics and setups needing refinement and improvement?


I know there are a lot of setup guru's here, so lets start a list
of either something that needs improvement in the physics, setup parameters,
setup options etc.

I'm sorry guys, this is my favorite Sim to drive and its the
only one I use consistently. I'm just afraid that we are asking for correct
"looking" interiors,different track layouts...etc. when the sim needs
attention at its very base, the physics and setups.

4-way adjustable shocks
packers and bump stops (formula cars)
3rd spring (F1 car)
Spring-pre load
Anti-squat/Anti-Dive
Adjustable Bump-Steer (toe difference in droop-ride height-bump)
Adjustable through travel (camber difference in droop-ride height-bump)

springs that don't change ride height

Differentials - more realistic options and physics

tires
-tires that act more realistically in extreme camber
(they seem to have more grip over bumps with more negative camber)
- extremely small difference in braking with a lot more camber
-small difference when the outside
of the tires are only 5 degrees warmer than they were static
-maximum pressure
should never be near the optimum pressure
-the tire doesn't roll over the tread surface
at extreme low pressures the sidewall stretches instead

There are tons more, of different importance that I can't
think of right this moment.

Discussions not arguments please!

Quote from KartRacer :

My question for Scawen is; Are there and if there are, what are the plans to improve physics and
setups? (tire,suspension,chassis,differential,aerodynamic)

Scawen rarely frequents here, or gives away his plans.

However, it is commonly understood that physics are what drives LFS.

The reason why we don't get a slew of physics updates every time there's a patch, is that everytime they change the hotlaps need to be wiped - if this happens too often, it pisses off the hotlappers.

Hopefully that explains it.

Quote from KartRacer :

My question for the community is; does the majority care
about the physics and setups needing refinement and improvement?

Yes.

Quote from KartRacer :

something that needs improvement in the physics, setup parameters,

There's probably no point, they're all more than likely on the todo list. But knock yourself out

I know Scawen rarely frequents these forums, I've been using them for a very long time(RSC as well). This is the best place to "contact" him.

The following is a reply after I sent a message to the "LFS dev's" about setups and physics

"To tell you the truth I can't really answer that question. That is up to
Scawen to decide, and even he will probably not know for sure. Lots of
things in LFS come about as a 'natural' development from what the devs have
been working on at any particular time".

"Of course we want LFS to be as realistic as possible, so the set up options
will be improved and developed as the sim progresses".

"I see that you are active on the forum and this is really the best place to
make comments about the kind of improvements you would like to see".

"Thanks"

I'm not expecting Scawen to give me a ring about this post, I'd be ridiculously
happy if he replied to it.
My point is we've had the same base setup options and suspension physics since the first release of S2. If we can have 1,000s of pages about what cars and tracks to put in this sim, I'm sure there are a few people willing to discuss how those cars do, should and will work on those tracks.

"S2 is a significant improvement over S1; important features have been added. For example dynamic camber (improved simulation of suspension and different suspension types), highly detailed tyre physics and a damage model have been introduced. Also the additional content in S2 should be noted - where S1 mostly contains the 'medium' fast cars and tracks, S2 introduces several 'fast' classes (and 'slow')." -LFS Mainpage

We've had tons of updates and we have gotten differential preload since S2. That didn't change the hotlaps, but thats it, no added options for the setup etc.,

I've watched this sim. develop and a lot of it has been by suggestion of the community. Thats why LFS is what it is. I agree with you in the sense that Scawen probably already has thought of everything I could ever imagine, but even if he has, I still want to discuss it anyway. I'm not going to say "Scawen, probably already thought of that, no need to mention it", when who really knows? Its an interesting subject anyways

What does everybody think about;
What will get changed/added?
Will it all be in a huge patch to avoid the hotlap issue?
What do other people think and/or want?

These physics issues and many more have been discussed many times…

The tire model needs major improvements imo… all others will work effortlessly when tire model gets better.

Some threads found by searching with keywords “tire model load” and all these on the first page…

http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load
http://www.lfsforum.net/showth ... highlight=tire+model+load


Actually… have a look here at that "Suggested improvements log"
http://www.lfsforum.net/showth ... mp;highlight=differential

Don't ask me. I just found the clutch.

4-way adjustable shocks - Only a slight refinement really, and one we can wait a bit longer for.
packers and bump stops (formula cars) - not really that important if the car is set up well.
3rd spring (F1 car) - Did the Sauber have a 'third spring' that year?
Spring-pre load - Why?
Anti-squat/Anti-Dive - This is a designed in thing, not something you generally change at will
Adjustable Bump-Steer (toe difference in droop-ride height-bump) - How about no bump steer in the first place. It's what you want. Why make a setup option to simulate stupid engineers?
Adjustable through travel (camber difference in droop-ride height-bump) - Camber does vary through travel

springs that don't change ride height - springs DO change ride height

Differentials - more realistic options and physics - Weissman lockers might be nice, and cam&pawl, but otherwise open, locked, clutch pack and viscous are okay aren't they?

tires
-tires that act more realistically in extreme camber - tyre physics are an ongoing thing
(they seem to have more grip over bumps with more negative camber)
- extremely small difference in braking with a lot more camber - LFS's longitudinal physics are not perfected yet
-small difference when the outside
of the tires are only 5 degrees warmer than they were static - yes, we agree something is a bit iffy here, but it might also be quicker in real life. Nobody tries tyres at ridiculour camber angles because of the danger and tyre wear issues. Who knows, Scawen might have predicted real life behaviour without even knowing it!!!
-maximum pressure should never be near the optimum pressure - shouldn't it? Ever?
-the tire doesn't roll over the tread surface at extreme low pressures the sidewall stretches instead - I'm hoping that Scawen will code the sidewall to be a fixed length rather than a stretchy thing. Maybe that will cure it.

Quote from KartRacer :

4-way adjustable shocks

Non linear shocks are needed for all cars really, but the damping should only be 4 way adjustable on the race cars.

Quote from KartRacer :

packers and bump stops (formula cars)

The GTR cars already scrape the bottom of their bodies in high speed cornering, so they would benefit from packers too. We already have bump stops though, they were added back in 0.2 or so.

Quote from KartRacer :

3rd spring (F1 car)

High end race cars often have more springs and dampers than at the wheels in general. I've read about springs other than the 3rd spring before. All would be nice to have on the cars that would have them in reality.

Quote from KartRacer :

Spring-pre load

Absolutely! Atm, to make realistic road going sets, you have to use very long springs to make up for all the deflection from the static mass resting on the springs. Pre-load would help there.

Quote from KartRacer :

Anti-squat/Anti-Dive

I believe these are just a function of the steering geometry, which is already modelled. I've tried (adimittedly somewhat blindly) to add anti-dive to double wishbone suspension in LFS but it doesn't seem to do anything. I could well be doing it wrong though, as anti-pitch should just be a fuction of the physics engine (force transfers and such) rather than being a seperately coded component (like an LSD).

Quote from KartRacer :

Adjustable Bump-Steer (toe difference in droop-ride height-bump)

Dynamic toe would diefinitely be welcome (on some cars, the FZ50 in particular). Should allow for more stable race sets and more interesting rally sets.

Quote from KartRacer :

Adjustable through travel (camber difference in droop-ride height-bump)

That was added with the first S2 release?

Quote from KartRacer :

springs that don't change ride height

How does that work then?

Quote from KartRacer :

Differentials - more realistic options and physics

What's missing? IIRC Todd said that, mathematically speaking, a TBD (e.g. Torsen or Quaife) differential is the same as a clutch pack with equal coast and power settings with zero pre-load*. Seeing as torsen differentials are all that we are missing, I don't see what needs adding to differentials in LFS. I would prefer if they were more mechanically based though (we actually set the spring strengths, ramp angles and number of clutches), rather than the setting the effects (locking factors and pre-load).

*I must add I'm not totally convinced though, as TBDs react instantly, whereas clutch packs have lag, and clutch packs would have an understeer effect during steady state cornering, while a TBD would not.

Until we get a rally car (and stages) there is no need for an eletronic, fully configurable, clutch pack diff.

To add to your list I'd like:

• More suspension types (swing arms, proper (semi-)trailing arms, sliding pillar, solid axles, de-dion axles, torsion beam, muliti-link, and others
  
• superchargers
  
• more engine configurations (e.g. W or H) and more than 12 cylinder engines
  
• more than 7 forward gears
  
• knee-points for brakes
  
• Adjustable rear wheel steering
  

I'm not going to cover tyres though as we have enough threads on such topics.

-Agreed for the dampers, 4 way only for the race cars.


-We don't have adjustable bump stops though. I've heard of cars compressing all of the packers and riding on the bump stops and springs, on purpose. Different bump stops would allow you to have more suspension travel but with the risk of damaging susp. because of smaller stops.

-I have seen different stiffness springs that have been put into the F3 car for testing, they are all the same size, but different coil thickness or more coils. The difference in ride-height because of different springs is minimal. They cars have almost no droop and a small amount of preload, so they sit the same with different springs until they go out on the track. Obviously very different springs will make ride-height different. Or possibly they only have so many stiffness's at the same spring height.

-"Anti-dive is the forces of inertia and mechanical resistance that the brakes create thru the front suspension. On most chassis, the lower control arms are level front to back. The uppers usually have about 2 degrees of angle front to back, higher in the front. This is called anti-dive, because thru the braking force on the front suspension, the chassis will resist dive (weight transfer to the front)."

"Anti-dive can be used to help your corner entry. If you feel that your car transfers too much weight to the right front, you can adjust your anti-dive to help this. How do you do this? You need to have that adjustment built into your chassis to be able to adjust this. You would increase the angle that you have on the right front keeping the front side of the control arm higher. Also, you can decrease the angle that the left side upper control arm has, making it level. Sometimes it can even be set in dive which means that the front side of the upper control arm is lower than the back side. With the settings resembling the settings I just described, the chassis will want to transfer the weight to the left front under braking. When you lift off the brake, the car will set down on the right front causing the chassis to turn."

"Anti-dive can be a good adjusting tool to your chassis handling if you understand how it works and you test it a couple of times. Try different settings so you can understand what changes do what and so you will be able to make the correct adjustments to your set-up."


-Anti-squat is more or less the same thing, but in the rear.


Again, the car we are assembling/testing is a fleet of Lola F3 cars. -The rear upper arm has 5 front pickup points directly vertical of each other about a 1/4 inch apart. Some formula cars have it, some don't.

-Concerning the differential, ramp angles and amount of clutches.

-Air pressure, whether the maximum isn't high enough, or the tires don't change correctly according to pressure, something needs to be done. I find it hard to believe that a Formula Car would be putting in as much pressure as they possibly can. You'd burn up the middle of the tire when you leave the pits, the car would be way too stiff as well. Again, I'm guessing, I'm sure some car some where uses the maximum air-pressure...


-About the camber and toe through travel. A lot of cars, especially the older Formula Mazda's I was painstakingly alignment tested on, have the ability to have 0 toe and 0 camber at ride height and then I can change toe and camber at bump and droop, with it still being 0 at ride height. Some people forget to check their camber/toe through the whole travel, but it can cause some nasty toe out in squat, which was horrible for our school cars.

The bump steer is useful, often used in high-powered rear wheel drive race cars, the rear toes in in bump in order to make the car more stable under power. But you are right, most of the time its best and most comfortable for the driver at 0.

Quote from KartRacer :

-We don't have adjustable bump stops though. I've heard of cars compressing all of the packers and riding on the bump stops and springs, on purpose. Different bump stops would allow you to have more suspension travel but with the risk of damaging susp. because of smaller stops.

You never really want to be on the bump stops, because the wheel frequencies go through the roof, and the cars become very skittish. F1 went through a phase early in the ground effect era, but don't really anymore.

Quote from KartRacer :

-I have seen different stiffness springs that have been put into the F3 car for testing, they are all the same size, but different coil thickness or more coils. The difference in ride-height because of different springs is minimal. They cars have almost no droop and a small amount of preload, so they sit the same with different springs until they go out on the track. Obviously very different springs will make ride-height different. Or possibly they only have so many stiffness's at the same spring height.

Any decent team will have a variety of springs with the lengths so that changing them gives the same ride height without adjusting anything else - makes spring changes MUCH easier. Mere mortals like myself have to redo the suspension geometry whenever we change springs (not often).

Quote from KartRacer :

-"Anti-dive is the forces of inertia and mechanical resistance that the brakes create thru the front suspension. On most chassis, the lower control arms are level front to back. The uppers usually have about 2 degrees of angle front to back, higher in the front. This is called anti-dive, because thru the braking force on the front suspension, the chassis will resist dive (weight transfer to the front)."

"Anti-dive can be used to help your corner entry. If you feel that your car transfers too much weight to the right front, you can adjust your anti-dive to help this. How do you do this? You need to have that adjustment built into your chassis to be able to adjust this. You would increase the angle that you have on the right front keeping the front side of the control arm higher. Also, you can decrease the angle that the left side upper control arm has, making it level. Sometimes it can even be set in dive which means that the front side of the upper control arm is lower than the back side. With the settings resembling the settings I just described, the chassis will want to transfer the weight to the left front under braking. When you lift off the brake, the car will set down on the right front causing the chassis to turn."

"Anti-dive can be a good adjusting tool to your chassis handling if you understand how it works and you test it a couple of times. Try different settings so you can understand what changes do what and so you will be able to make the correct adjustments to your set-up."


-Anti-squat is more or less the same thing, but in the rear.


Again, the car we are assembling/testing is a fleet of Lola F3 cars. -The rear upper arm has 5 front pickup points directly vertical of each other about a 1/4 inch apart. Some formula cars have it, some don't.

Anti squat and anti-dive go in fashions. Next year nobody will have adjustments, the year after they come back (or every 5 years, but you get my jist). It's generally accepted that a very small amount of anti-dive is beneficial, but anti-squat tends to reduce power application potential and is rarely used.

Quote from KartRacer :

-Concerning the differential, ramp angles and amount of clutches.

Maybe one day, but hardly critical, and ends up doing roughly what we have now anyway. A refinement for later.

Quote from KartRacer :

-Air pressure, whether the maximum isn't high enough, or the tires don't change correctly according to pressure, something needs to be done. I find it hard to believe that a Formula Car would be putting in as much pressure as they possibly can. You'd burn up the middle of the tire when you leave the pits, the car would be way too stiff as well. Again, I'm guessing, I'm sure some car some where uses the maximum air-pressure...

Correct - the tyre physics are by no means final, and improvements are likely in time. Patience

Quote from KartRacer :

-About the camber and toe through travel. A lot of cars, especially the older Formula Mazda's I was painstakingly alignment tested on, have the ability to have 0 toe and 0 camber at ride height and then I can change toe and camber at bump and droop, with it still being 0 at ride height. Some people forget to check their camber/toe through the whole travel, but it can cause some nasty toe out in squat, which was horrible for our school cars.
The bump steer is useful, often used in high-powered rear wheel drive race cars, the rear toes in in bump in order to make the car more stable under power. But you are right, most of the time its best and most comfortable for the driver at 0.

I cannot see the benefit of bump steer on purpose. And it's quite easy to be rid of it if the car is designed or adjusted accurately.


I'm not saying none of them are useful, just that they are refinements that will [possibly] be added in time. They certainly don't make LFS rubbish, and they certainly aren't the limiting factor in the overall scheme of things.

OT: I'm keeping an eye on the Lola F3 chassis of late. Might be something to consider in a few years when they go down in price for myself

I'll try and sneak you one out the back door, I'm sure we will have a few wrecked cars and enough to build one up from all the peices After all it is a turbo F3 school car....crazy if you ask me!

Yup, I'm not demanding anything. I'm just discussing what I would like to see in the future, like I said, LFS has been my sim of choice since the demo days. I'm just very interested in how good the driving and feedback could be with tire physics and setup options that more directly correlate to its real life counterparts.

Another topic for discussion though; why are we able to down shift so fast under braking? I would think if you were to do that IRL or even another sim you'd end up backwards before you got the chance to turn in! Is it drivetrain physics, tire physics...I dunno, but something seems a miss to me.

Quote from KartRacer :

Another topic for discussion though; why are we able to down shift so fast under braking? I would think if you were to do that IRL or even another sim you'd end up backwards before you got the chance to turn in! Is it drivetrain physics, tire physics...I dunno, but something seems a miss to me.

Pretty much only the FWD junkies are abusing that trick. Since engine damage is mostly non-existent at the moment, engine braking is super effective. In RWD cars the rear comes around to say 'oh hi! ' if you change down too fast.

Still... It's pretty hard to see why the gear change speed isn't forced to a minimum time according to the used car/gearbox/transmission/something but hey I'm not the one developing the sim.

Quote from spankmeyer :

Pretty much only the FWD junkies are abusing that trick. Since engine damage is mostly non-existent at the moment, engine braking is super effective. In RWD cars the rear comes around to say 'oh hi! ' if you change down too fast.

Not really... People just set the brakes even more forwards than they already do, thus no polite rear...

It could be both the absence of damage in the mechanical parts and (what people more knowlegdeable than me agreed) slightly off longitudinal traction/force combining from the tires... I'd blame the former...

The fast downshifting is just a side-effect of not being able to go directly from 6th to 4th to 2nd or etc... because of paddles. You need to go through every gear, unless you have a H-pattern shifter.

And the fast shifting is most pronounced on tracks where you are shifting from 6th to 2nd. In every car, really. With BF1 you need really fast fingers

Also, you need to keep the revs high or the autoclutch jumps in and causes your rear tires to lock up very suddenly. Or I think the auto clutch is the cause... or the clutch diff...?

Quote from Bob Smith :

More suspension types (swing arms, proper (semi-)trailing arms, sliding pillar, solid axles, de-dion axles, torsion beam, muliti-link, and others

Yeah, a proper yank tank would be great in LFS, something like the Ford Mustang with the GT engine but the same suspension and brakes ...

Quote :

superchargers

A supercharged car would be a great idea ... without wanting to sound like some form of pothead, what about something like a Supercharged Jag.

Quote :

[*]more engine configurations (e.g. W or H) and more than 12 cylinder engines

I assume that latter is a request for a BRM P83 then ...

Quote :

more than 7 forward gears

Other than something with two or more gearbox ranges (like a Land Rover or a Truck), I haven't heard of anything with more than 7 Bob...

Healthy discussion. The tyre pressures are an oddity. You almost are led to believe that were the maximum pressure removed for some reason, then even greater pressures would be used on the fastest setups, still without the negative effects you would see with a real tyre. Then there's the whole tyre-wear issue, wherein tyres worn within fractions of puncturing become lightning fast again due to the cooling. It's almost as if tyre-wear doesn't actually exist in any normal sense, you simply have a meter telling you how long till you get a puncture.

I like the differentials now, but my complaint about them is that there are often too many choices. (Although that applies to other areas of setup too). I would rather suitable choices were made for each car, so that only one type of diff is available (particularly for the "road cars"), and adjustment limitations applied appropriately also, unless in equivalent series' the choice and wide adjustability is available.

What is it that makes worn tires worse than fresh tires in real life? Disregard the negative effects from abusing the tire, like graining and blistering.

The rubber undergoes a chemical change due to the heat (sort of self-hardens in a way), and a mechanical work-hardening process as well I believe.

A fresh set of tyres is pretty much quickest 'out of the box', but scrubbing them in will see them last a lot longer at ~95% of the max potential (rather than working great for 5 laps, then being at about 75% potential for the rest of their lives.

Scrubbing in tyres is REALLY important in club motorsport (where we can only afford a set of tyres once or twice a season), and really important in F1 where hundreths of a second are at stake. In other racing it's not that important, so you see more cars going out on brand new tyres (because they can throw them away). I bet in LeMans they carefully scrub tyres too...

Quote from Hyperactive :

And the fast shifting is most pronounced on tracks where you are shifting from 6th to 2nd.

Well, on most setups brakes are set so they won't lock tires on low speeds at 100% brake power, but at high speeds the tires won't be nowhere near lockup treshold and aggressive engine braking is used to cover up for the unused braking potential.

Atleast, that's how I've understood it

Quote from tristancliffe :

The rubber undergoes a chemical change due to the heat (sort of self-hardens in a way), and a mechanical work-hardening process as well I believe.

Found this interesting paragraph at

http://www.avonracing.com/tech/advice.htm

Quote :

It is relatively easy to generate temperature in the driving tyres as they are transmitting power most of the time. The front tyres, however, will need to be given more time and be loaded up progressively before they will be 'In' fully. It should be noted that it is the loading of tyres that introduces the significant heat, not sliding or wheelspinning.
For qualifying, the best results have been shown to have been achieved when the front and rear tyre temperatures are the same when measured at the base of the tread. If pushed too early, the rear will come in before the front causing understeer, and taken to the extreme, cold graining as detailed above.

In LFS it appears that friction is the main factor that increases heat in tyres were in RL it would appear that the mechanical manipulation (i.e. energy converted to tractive power with some waste heat going into the rubber) of the tyre is the main cause of tyre temperature change. Hence why you don't see V8 Supercars doing drifts and wheel spins to warm up there tyres, but rather working them by accelerating then braking firmly as well as swerving from side to side. All which work the rubber which has a by product of producing heat

So perhaps that would mean that the tyre surface can heat and cool quickly from friction but the temperature of the main body of the tyre is more influenced by mechanical grip (i.e. distortion of the tyre)

Is problems like graining and blistering from excessive surface temp not main body of tyre temp?

That would also explain well why in RL lower pressures see more heat in tyres than high pressures, and my go along way to fixing the abuses of using too low a pressure or high a pressure in LFS at the moment which only changes the contact patch size and therefore the amount of friction based heat. So if heat was modeled on tyre deformation as well too low a pressure would result in much higher heating of tyres than now what happens, same may be true in reverse for too high a pressure. Is that one of the primary reasons tyre manufacturers specifiy an optimum pressure range?

The same probably goes for high camber angles, hence why you don't really see extreme camber on the driven wheels in RL as it would be causing unwanted tyre heat generated even in a straight line accross the whole tyre due to deformation were in lfs it just causes uneven heating...

Quote from Glenn67 :

snip

It's like the awesome Garry McCoy (supposedly) not heating the carcass of the tyre so much, and doing it less damage, by spinning and sliding the tyre more and not loading it so heavily. (Supposedly) despite the higher surface temperatures, the tyre finished the race in better condition.
I'm not saying that (supposedly) applies anywhere else but in 500 Gps with that particular tyre in 2000, though, and you're right about the loading. In LFS you don't need to do that to generate the required tyre temperature, you just need to do a few donuts and some lairy understeering, and with the surfaces green the tyre is ready for 100% performance.

More advanced tires is always for the better but personally I'm afraid that scrubbing is something that is too hard for LFSsers and hence won't be implemented

One way of implementing "scrubbing" would be to offer two sub-types per tire compound, the fresh tires and scrubbed tires. I think scrubbing your own tires and "saving and loading" them would be just too complicated, but scrubbing (and better heat modelling) would defenately add to the experience. But... :/

Quote from Hyperactive :

More advanced tires is always for the better but personally I'm afraid that scrubbing is something that is too hard for LFSsers and hence won't be implemented

I don't think it would take too long for people to catch on once implemented and the benifets of a better tyre heat model in respect to setup consequences and driving styles far outway any negatives imo

I am really looking forward to dynamic track conditions for the same reasons

Quote from Bob Smith :

...

To add to your list I'd like:

* knee-points for brakes

...

Knee-points? What is that?

Quote from duke_toaster :

...
I assume that latter is a request for a BRM P83 then ...
...

Yum yum!!! But only if the engine sound will be authentic

Quote from Primoz :

Knee-points? What is that?

Essentially, dynamic brake balance for a given brake input. Probably not all that useful for racing but could potentially allow you to adjust how the car reseponds during trail braking, without affecting straight line braking.

Quote from Electrik Kar :

Don't ask me. I just found the clutch.

Thanks for making me laugh on a particularly dull and mindless Monday