Physic discussion thread

I apologize. I neglected to mention an important part. The cup car still had the "spikey" diff using default, but it was easily fixed by locking it a bit using a standard preload/power/coast ratio, like 80/60/30. I didn't mean to suggest there was no tire fire...lol...(new term to me).

Regarding the tire grip question. Mainly a hypothesis. I imagine a max grip tire (on a max grip track, like TT) could handle more lock (lateral/longitudinal force), but when it unlocks its more severe and snappy. A less grippy tire, might not be able to handle as much lock in the first place which could lead to more gradual opening.

Looking at the image, a fresh set of rubber might have a higher max lateral/longitudinal force, like the red line, but a more abrupt loss of grip at higher slip angles, but a worn tire might have a more gradual force curve, like the green line.

Im not suggesting that the tire model is the culprit either, just that it could be a contributing factor as IMO the tires are quite grippy. (I come to that conclusion because improperly cambered and pressured tires can set WRs on the LBs in the hands of a skilled driver, when common sense suggests that shouldn't be possible)

 

Ah ok. Yeah you can fix it via setup. Terminology and slang is hard, but one tire fire is such a boss term.

Hmm. This gets into the whole open differential physics thing that I don’t fully understand. But I think I kind of understand. The thing for me is that the LSD clutch pack is like an unbreakable force atm. But when it opens it might as well not have any friction.

It has been hinted at before that it might be the tires or their driveline config, because of the way open diffs are also spiky (but don’t have any of that unbreakable force thing).

 

New update, new car, new type of differential.

1.1.1.0 introduced the Ginetta G40 and Geared LSD is back on the menu. I hear it used to be a choice in the options many many versions back. And in PC2 you can choose it many different cars. To be fair to AMS2, it's good that you can't just slap a random differential to any car.

Let's start with the good. This is a really fun car and reacts to trail braking. It's easy to drive and probably a good choice for close racing.

However it looks like there's a major issue with the geared LSD or the driveline. From my testing it used to do a lot in PC2. Telemetry showed increased rear tire speed differences off throttle. It showed in game as expected agile turn-in behavior off throttle (you can tune this to your liking if the differential is configurable). So I had hopes that this would really demonstrate what the cars could feel like with throttle control.

But here's G40 (Cup class) in AMS2 at Goiania Short. it's completely locked 99.5% of the lap. (See the Tyre RPS graph)



Just for sake of comparison, here are the front tire speeds instead. These are not a reference by any means, but just to show you what the graph looks like when the tires have minor speed differences most of the lap (when turning).



Once again the car is certainly drivable and I would say it's easy in the sense that it doesn't do anything unexpected in slow corners. You can drive and race it just fine if you just ignore this detail. But it does slip and drag its rear tires all the time. It lacks the major on/off throttle dynamics you expect. So in my eyes it's missing what would really make it mega fun. And many drivers feel lost without these fundamental behaviors.

I won't comment on why the geared LSD is like this, because I couldn't possibly know. But it is interesting that it's completely unfunctional.

 

Hey, again some nice and fast finding of one of the most affected cars! (actually one, that drives impressively, but still suffers from it big time and the geared diff is actually not unfunctional, i know, this sounds like a contradiction)

First, i want to inform this thread, that this issue is known and also in process of being reviewed by Reiza Studios and also receives internal testing. I'm also involved in testing and can agree with a lot of stuff, you have found!

Some clutch LSD cars are actually way better now, especially GTs, Prototypes and other higher powered cars, but the lower powered ones still suffer and it's an ongoing process, involving many components of the driveline (some visible in the recent changelogs) and it's not the differential model itself but a combination of several things, due to the high complexicity. (Some cars, like Group A and M1 Procar even differentiate too much on power side right now, for example) I'm sometimes indeed also frustrated, it takes so much time for all these small steps, but it's how it is and things keep improving, for sure, which could also be a good cornerstone for possibilities in the future!

Yes, the G40s are showcasing, that also the geared diff is indeed affected by the driveline quirks and back then, we were able to use the F-Retro with the geared diff, that was night&day, which additionally indicates, that the differential model itself is actually working. The G40 showing different results is another indication, that it isn't the differential, but the whole driveline, needing a lot of "Fingerspitzengefühl" to be taken in the right place.

Be aware, that i can confidently say, i'm spending hours to annoy @Renato Simioni with more and more wheelspeed graphs, feedback for the workwise of different cars and can see progress on different fronts already, also discussing and learning, what might be wrong and right. I'm also looking into the threads from time to time still, to keep track of what others say, because more impressions are helpful, of course.
Therefore, keep pushing, every information is helping!!

Especially the most affected cars, where quite some are already identified, share some quirks, that need more time to look into it.

Basically: The more cars are light, underpowered and grippy, the more likely it is, that they will still suffer diff issues.

I hope, i was able to shed a bit of light at this topic, to assure you, that this is taken care of with a very high priority and awareness!!

Have faith!

 

Hi and thanks for the update! It does give me confidence that the driveline will be sorted out eventually and perhaps even be next level in detail. Thank you for handling it with Renato!

I have yet to test the usual clutch LSD suspects. I will try them to see where progress has actually happened.

 

Don't expect the snake oil super healing over night, like basically summarized, though.

Renato and co. are handling it, their work is outstanding! I'm just moaning about things being odd

 

I have to pop in and say that I just finished QCing my Porsche Cup 4.0 builds. The diff ratios make sense now and the car is ___dam wonderful to drive.

 

Had to adjust steering angle to 28 degrees?! Maybe reversed? But fixed Cayman and Cup 4.0 for me. They really dance now!

 

same

 

Ever since the last update (v1.1.1.0) the cars seems to spin way too much, it wasn't easy before, but now it's just ridiculously hard.

The cars spin at every chance they get, I brake? spin, I start driving? spin, I turn? spin...
And if I accidentally go to the grass, there is no chance of getting out of there, constant spinning.
I know it's supposed to be a simulator, but comon.... I doubt that this is the way cars behave in real life.

 

Crimson in the 'alpha' testing is one of the best news for this game.

 

Hi Avo,where have you been all the time?Good to hear you again.Thought ,you gave it up.

 

Naa, just some crazy stuff at work draining all mental and physical resources.

 

Which cars in particular are you having issues with?

 

.
Hi also A.
Just tested Procar and Bt44 for response to:

Both are better all round esp Bt44 Whats obvious is the extra subtlety in handling and car response /feedback, its been a pleasure beating old TT times just for that.

 

May you please share one of the combinations in car/track, that seem very affected to you?

 

Does anyone think the P4s have significantly regressed in terms of driveability in the latest patch? It feels to me like these low power high downforce cars have lost most of the planted feeling that made them such an appealing drive. I'm assuming it's something in the tyre model revisions. Nearly all the other cars have made giant strides lately, unfortunately I don't class the P4s among them.

 

Its my understanding that geared lsd’s don’t “unlock” like a plated lsd, so you shouldn’t technically see a wheel slip difference. They are essentially open, applying torque evenly as you apply power. The more power, the more they lock on both wheels, regardless of the difference in rotation. In the event a wheel slips, torque is simply biased to the gripping wheel.

 

The graph by @oez shows wheelspeed difference in second monitor.

EDIT: (I think, RST uses the wheelslip %age thing, right? Wheelslip graphs are good for analyzing how to optimize effectiveness of the diff, providing it works, of course, but the Wheelspeed graph is more reliable, to see clearly, if any diff is doing the differentiation.)

Of course you also see wheel'speed' differences with a torsen/quaife diff, or you could go with a spool anyway. There is also a coast side and you definetly should see different wheelspeeds off-throttle into corners.

Btw. Just looking at this animation shows me again, how genius engineering can be

The geared diff can show some reaction with very low torque bias settings and extremely aggressive driving, but of course it's not enough or "natural" yet, as we know:
G40 GT5, Velo Citta, sloppy driven lap, hitting kerbs etc.:

 

Ahh, I see. Yes, I was definitely interpreting the graph as a slip graph.