Physic discussion thread

Personally I would start from bigger and more powerful cars... I believe cars like Caterhams will be the last to be resolved as they will require a level of "rightness" much superior to other cars to achieve natural results...

 

Yeah so far I found the following very off...
Group A - tyres feel way too soft/grippy and the car just wants to spin everywhere.
Mclaren road car - totally gone lol
Stock car 2020- feels better but still too soft, tyres/chassis
Super V8 - too soft, front tyres are too grippy/soft feeling.
Opala 86 - same as the group A
GT4 Cayman - clutch slipping and seems too lose all front end feel mid corner.
But everything else probably feels better so far, all though very different.

 

Considering how the modifications to the tires and drivetrain were conceived and implemented in the past week I would not be surprised if some refinement is needed and quite some tweking of the baseline setups is even more needed... Fiddling with tire stiffness, diffs and driveline all at the same time will for sure command deep setup changes compared to the old hardware which definitely did not have the time to happen

 

Nobody promised global LSD fixes with this update, but I'm just going to confirm that the F309 clutch LSD is still off in that it takes a lot to get it to open. The upside is that it doesn't have wild one tire fire. It's fun.

Clutches: 2
Preload: 0Nm
Power ramp: 89deg
Coast ramp: 89deg



The wait for a major LSD improvement continues, but we should focus on the other changes . For example if the GT1s no longer scare you in low speed corners. I'll go try that.

 

Can we do a little community experiment with the Merc Group A?

Latest version is a huge improvement in my opinion in every way--except it still has the annoying diff issue. I simply bumped the number of clutch packs up to the max of 10 and it mostly resolved the problem. But could you guys who are the real experts suggest ramp angles that clutch packs that should minimize the problem or make the car as driveable as possible?

I'd like to see if several people agree with an A:B back-to-back test. And it might help us pinpoint whether any progress has been made in this release, or, if we're just working around the problem still waiting for Reiza to fix it. I am hoping there was some progress...

 

Agreed, but saying things like McLaren road car totally gone is not helpful or useful. It isn't gone. Suffers more than most from the diff locking/unlocking issue that is the point of this thread, and any road car should automatically be using 100% brake pressure as that is never curtailed or adjusted in a street-legal vehicle. If the diff would stop playing games in slow corners and doing its unexpected business, the car is quite close to excellent in my opinion.

Group A, as I just mentioned above, also suffers from the diff issue. But the new tires and other changes are fantastic--it feels more plausible than ever to me.

Haven't tried the others listed above, but based on beta, the vast majority of cars are going quickly in the right direction.

 

Hmm. I can only say that it lights up the inside tire during exit even with 10 clutches. I don't know how much of an effect 10 vs 4 clutches should have. Either way it's difficult to throttle steer as is.

(don't worry about that massive downward spike near the end... I had a moment)

 

Didn't mean to imply that the 10 clutches was any kind of solution. It merely changes the handling and feel of the car dramatically. I was wondering what combination of ramp angles and clutches SHOULD we try for this car? ...In the interim....

 

High pre-load and small ramps angles is probably the answer if the internal tire is going in smoke?
Maybe as drivetrain issues are addressed in the future we may expect some reversal of the recent "per clutch" reductions we got in the past months? Let's see what the next move from Reiza will be. I'm hopeful everything will fall into its place soon

 

High preload force would mitigate it on its own since ramps only add more force. The group A has low preload with low added power lock from it’s high power ramp value. So indeed little force on clutch packs and probably little locking per clutch.

True, this is a good point.

 

@Synaks I'm replying here in the physics thread. Have you had a chance to test the 1.1.0.4 update yet? It helped the GT1 McLaren F1 be less crazy in slow corners for example.

Oh sorry, I misunderstood that. I need to AB test it again.

You can't change the ramp angles and preload in that car. But if you could I would probably go for a bit higher coast ramp and a bit lower power ramp. Then 60Nm preload with 4 clutches. It clearly needs more preload to mitigate the one tire fire out of corners and as far as I understood the ramps work with a very slight delay. Hence more preload to prevent the inside tire from getting a head start before power ramp increases lock.

So my suggestion for a car like this (relatively light, high-ish center of gravity, semi-driftable slicks, oversteery driving style):
Preload: 60Nm
Clutches: 4
Power ramp: 60deg
Coast ramp: 45deg

Obviously I haven't tried this, cause you can't change them. The Group A should be fine with little differential function at speed, because it's eager to oversteer. In fact then you really want the stabilising rear lock.

This also comes with the caveat that I don't think Group A's current driveline and/or tire setup - or whatever the culprit is - makes the differential work when coasting except at very low speeds. So changing its coast ramp is a bit pointless.

As long as it works in low speed hairpins. Even the Oulton Park U-corner is ok with an almost locked diff (even if its function was more gradual I mean), because it's not that tight in the end. It bothers me more that it's so difficult to throttle oversteer with its current differential setup.

(PS. daily reminder to help confusion with clutch LSD ramps: lower ramp degree = more locking )

 

Sorry for the triple post .

Just to see where we are right now, I tried the Stock Car 2019 at Goiania Short. And surprise surprise I looked at some tire speed graphs .

It actually keeps one tire fire in check pretty well with its default setup. And before I say anything else, this car is really fun! I don't have any major gripes with it right now. I'm just using it as a test bed.

However I wanted to see how open the differential gets when coasting, because that's my personal biggest setup concern with all of the cars I've tested thus far. On the graphs this can be seen as the rear tire speed lines separating. If they separate too much, too early and too easily, this can be seen as rear instability when driving the car and lifting.

I like to use rear instability as a tool in my driving. It's one of the car behavior aspects an LSD minimizes if needed. More locking when coasting = more stable rear. But some like me still like a nervous unstable car that might be a bit crazy at first, but turns in easily.

Another thing I've mentioned before is how too much rear stability minimizes or fully eliminates liftoff oversteer. I would also like to have a lot of that in my setup.

Some amount of both is expected in many cars.

---

Default setup didn't open at all (or rather rear tire speeds were identical) with the default setup. So I loosened it up.

Preload: 50Nm (some amount of locking all the time, minimum value)
Clutches: 2
Power ramp: 40deg (high additional locking on throttle)
Coast ramp: 89deg (no additional locking off throttle)



As you can see the differential does open a little bit at slow speed around the apex of T1. It allows some rear tire speed difference and the car turns more easily. To be fair this is a decent minimum requirement for very tight tracks and hairpins. But it just makes you wonder if it should open easier and earlier than this. Notice how it doesn't open in the chicane which is also pretty tight and could really use the extra turn in that a loose differential gives you.

Sadly, this low preload and number of clutches comes with a lot of one tire fire (rear tire speed difference during acceleration).

---

To have something to compare this to, I tested the Stock Car 17 in AMS1.

I used its default setup.

Power: 22%
Coast: 40%
Preload: 21

I don't know how to compare these percentage based values, but preload goes up to 40. So if I had to guess this is closer to the AMS2 default setup which has a fair amount of preload with higher coast locking (lower coast ramp value) than power locking (higher power ramp value). And neither locking value is super high.



The differential opens (rear tire speeds become different) much earlier, just as I start to turn in. And then there's perfect lock (identical rear tire speeds) without one tire fire during acceleration.

---

Which is better or correct or more realistic? I don't know, I'll let someone else be the judge of that . The Stock Car 19 feels fun and good to drive so it isn't a big issue with this car. But even so you have minimal say in how loose the rear is off throttle. I would like to set it up to be a lot looser than that. And this is a pattern that repeats in many cars. Some of them I would like to be A LOT looser.

Remember that there's quite a lot of preload and coast locking that you could decrease in AMS1 to make it looser. Where as the AMS2 graph is already at its loosest, most open configuration.

And then finally I need to point out that I had to set preload and clutches to minimum to achieve this minimum amount of differential slip (open) in the slowest corner. But this gives me a lot of one tire fire during acceleration. I could add a bit more power lock, but it seems unreasonably high as is. Maybe this is due to something else like @Avoletta1977 (and others?) have been suggesting since open differentials also get too much of it? Either way it's unbalanced from a setup point of view: minimally open coasting, very open on power.

 

It is exactly what all the convos on the diff were on: currently in AMS2 the diff opens too little (if at all) too late and too suddenly. In some cars it is more evident then others depending on their weight, downforce, torque, COG, tires etc. and in some other less. But overall it is not yet there.
Therefore right now you may experience additional understeering at turn in, sudden waves of oversteer when around max lock for a corner (it was the case of some situations for the GT1s earlier) and possibly either a lot of drifting or a lot of understeer on your way out of the corner depending on the car and corner.

 

I couldn’t have said it better myself.

This does get repetitive . It is not my intent to do these tests forever just to complain about an easy target. As you said by now we can see what’s going on with the rear and what the effects are. Instead I just wanted to make sure myself that it’s not cured yet.

Rear tire speed graphs can wait until the next update on the matter. Unless someone finds a car with some interesting behavior.

 

In my opinion, fixing diffs should be at the very top of Reiza's to-do list, do we know if this is the case, are Reiza actively working on a solution to these issues?

 

In my quick testing without telemetry, the Cup car had tons of one tire fire with its default preload. And really locked coasting. It seems to mask this issue with slippy tires that make the car feel loose. This is pragmatic, but in the future a compliant differential would help increase contrast between off throttle oversteer and on throttle understeer that the Cup car allegedly has tons of in real life. Again I won't say any "final word" on this car's dynamics other than if that was the end goal then you would probably need a well functioning LSD.

And to be clear in general (not just as a reply to your comment), us annoying diff people are merely suggesting that a major clutch LSD improvement could unlock AMS2 car dynamics in a big way going forward. There are still other important variables to get right like the tires and the rest of the driveline. Both of which Reiza did work on for the latest update (1.1.0.4) .

Could you elaborate how grippy tires result to this behavior? What if the tires had less grip? Wouldn't we have even more one tire fire? Or if you mean not unlocking immediately off throttle, shouldn't grippy tires force a rear tire speed difference even more when turning (and the outside tire needs to travel a longer distance then the inside tire)? I can see why very sloppy tires couldn't overcome the LSDs friction from preload and coast ramp.

This is doable. But is it the same logic that worn slippery tires would help diff function?

 

Yeah i agree the 2019 StockCar is perfect give or take, so i just wonder why the StockCar 2020 is so bad in comparison.
I know it's slightly different but surely if you have a car so good, it's strange some are so off.
I'm sure Reiza will sort out this Madness engine...

 

This new generation of Stock Cars have been redesigned to bring the cars closer to its roots – although still based on the same chassis of the JL09 that fielded the series in the last decade, these have been reprofiled to become a little more Stock, a little less prototype, featuring several elements of their production counterparts and a lot less aero components relative to the cars of 2019.

I would say it's probably more than just slightly different. It's difficult to find exact information in English.

Heavier, less aerodynamic... it's going to be a softer experience, but of course I'm not sure what you mean by soft.

When you say soft, do you mean it has too much body roll or that it's unresponsive?

It used to be about 3 seconds slower around Goiania compared to the 2019 version and just feels like a tank in comparison. I haven't tried it in a couple versions though. Anyway back then it felt looser, slower to respond to everything and lacked high speed grip which is within my expectations.