Physic discussion thread

Discussion in 'Automobilista 2 - General Discussion' started by Avoletta1977, Jan 3, 2021.

  1. steelreserv

    steelreserv Well-Known Member Staff Member AMS2 Club Member

    Joined:
    May 25, 2020
    Messages:
    978
    Likes Received:
    1,239
    double post
     
  2. oez

    oez Mayor of Long Beach AMS2 Club Member

    Joined:
    Aug 27, 2020
    Messages:
    697
    Likes Received:
    680
    Yes you are correct in that using a word like locked with a geared LSD is probably wrong. But the rear axle is in fact practically locked according to telemetry :D.

    So whatever part or combination in the driveline is causing the lock end result needs to be looked at. I think at this point it’s best if I don’t jump to conclusions by blaming the diff code. But from a setup point of view the diff is still definitely locked when turning and affecting car dynamics.

    Crimson said the rest better, I just wanted to address my use of ”lock”.
     
    Last edited: Feb 2, 2021
  3. CrimsonEminence

    CrimsonEminence Administrator Staff Member AMS2 Club Member

    Joined:
    Nov 4, 2018
    Messages:
    9,790
    Likes Received:
    9,860
    How is it with lower rear ride height, or higher front ride height, or also higher rear wing? Additionally does your rear brake lock up?

    Otherwise, increase diff preload by 20-30 Nm and decrease coast ramp angle by 5-10°, does something change?
     
  4. steelreserv

    steelreserv Well-Known Member Staff Member AMS2 Club Member

    Joined:
    May 25, 2020
    Messages:
    978
    Likes Received:
    1,239
    I don’t think the term locked is wrong. Geared diffs are designed to be locked and progressively lock on power. They transfer torque based on traction, not clutch plates slipping unlocking).

    The distinction I missed is between “wheel slip” and “wheel speed”.
     
  5. oez

    oez Mayor of Long Beach AMS2 Club Member

    Joined:
    Aug 27, 2020
    Messages:
    697
    Likes Received:
    680
    Definitely. Honestly geared is still a big mystery to me. I've seen the detailed 3D animation video on Youtube explaining it, but I just can't wrap my head around it from a setup point of view :D. All I know is that I would get the same dynamics with a spool right now (or in .0 anyway).

    Yep I noticed that there was a misunderstanding. I've dealt with wheel speeds all along as a basic way to know what the actual differential setup is. Tho it's an interesting idea to compare the two wheel graphs. If I'm turning in, but the rear wheel speeds are identical, should it show up on the wheel slip graph instead? Unless I'm misunderstanding what wheel slip means here.
     
  6. steelreserv

    steelreserv Well-Known Member Staff Member AMS2 Club Member

    Joined:
    May 25, 2020
    Messages:
    978
    Likes Received:
    1,239
    My first thought is that the wheel speeds should be different, like the front tire example, but the more I wrap my head around the geared lsd, im not so sure. It's designed to limit wheel speed difference and do so really smoothly.

    From my understanding, based on what I have learned:

    Coast bias is the easy one. Its based on engine braking torque only. The higher the bias, the more stability.

    The power bias is a bit tricky. The diff recognizes the amount of traction loss in a wheel and vectors torque to the wheel with traction in an amount based on the bias. with 1:1 if one wheel loses traction, the other wheel loses power in equal measure. With 5:1 if one wheel loses traction, the other wheel gets 5x the amount of torque based on how much traction the slipping tire still has.
    Setup wise, you want it high enough so you don't lose power, but not too high to create understeer or snap oversteer. If both wheels are losing traction, like a power slide, the diff is constantly sending torque to the wheel with more traction and ping ponging back and forth instantaneously. It should feel smooth, like how the ginettas feel when drifting a little bit.

    Now I'm curious about how a wheel slip graph would look. I will try and set aside some time to gather some RST data because if your graph is correct, and there is no wheel speed difference (and the rears are behaving like they are locked, which is also what my research indicates), this means if the inside tire is spinning faster than the track then a wheel slip graph should pick this up. BUT NOTE: if the either tire is spinning and losing traction, lets say 50% traction, then the outside gripping tire should have basically 50% less power (with a 1:1 power bias ratio). You're going nowhere fast like that and that's why the geared lsd is disfavored in racing. But it will prevent a knucklehead in a supercar from killing himself.
     
  7. CrimsonEminence

    CrimsonEminence Administrator Staff Member AMS2 Club Member

    Joined:
    Nov 4, 2018
    Messages:
    9,790
    Likes Received:
    9,860
    Mate you're overthinking it a lot.

    If you run 1.0 coast torque bias and there is no differentiation at falling wheelspeeds, while driving through/into a corner, the geared diff is not working. Easy as that.

    The wheelslip graph will probably still show slip, but way less or in the expected ratio. (Tires should always slip at least a bit, even with torsen diff approach, it's what tires do ^^')
     
    Last edited: Feb 2, 2021
  8. oez

    oez Mayor of Long Beach AMS2 Club Member

    Joined:
    Aug 27, 2020
    Messages:
    697
    Likes Received:
    680
    Geared LSD is also known for not working (or it practically becomes an open diff) when one side is in the air. Is there a nice loading dock somewhere in AMS2 where we can test this :D? Like the grass-tarmac unlocking test, but for geared LSD.
     
  9. CrimsonEminence

    CrimsonEminence Administrator Staff Member AMS2 Club Member

    Joined:
    Nov 4, 2018
    Messages:
    9,790
    Likes Received:
    9,860
    Velo Citta turn 2. The inside kerb is like a ramp, if you hit it hard enough. Also the chicane sausages are really high there.

    The graph a few posts ago, i uploaded, shows differentiation over kerbs, when hitting them aggressively, btw.
     
  10. azaris

    azaris Well-Known Member AMS2 Club Member

    Joined:
    Aug 30, 2020
    Messages:
    1,177
    Likes Received:
    584
    Like night and day, thank you. Seems this car could use an update in the default setup.
     
    • Informative Informative x 1
  11. oez

    oez Mayor of Long Beach AMS2 Club Member

    Joined:
    Aug 27, 2020
    Messages:
    697
    Likes Received:
    680
    With throttle mapping having seen some updates, I thought the McLaren F1 LT throttle response now feels really immediate. AMS2 hasn't felt so sensitive before and I like it. Or then it's placebo...

    It almost feels like there's more going on than just the throttle pedal being sensitive. It feels sharper. Does the ported AMS1 throttle logic actually change throttle response timing per car as well? So not just mapping throttle position to torque at given RPM, but also how quickly the engine responds. I understand that this would be the ultimate goal especially when dealing with reasonably performant NA engines, turbo cars and really really high performance NA monsters like the McLaren's V12.

    Aside: I drove the latest Fiesta ST for a weekend two years ago. Lets say my pants were too big for that bucket-y seat so I couldn't actually live with one :D. The warm hatch version with "only" 140 horses is quick enough to be fun, but I now know from experience how much the ST's throttle response changes things. Just another fun aspect of a car that is never shown on paper...
     
  12. F1 Hero

    F1 Hero Active Member

    Joined:
    Jun 14, 2020
    Messages:
    176
    Likes Received:
    98
    So, what do you physics experts say now? Where is the game physics at this point?
     
    • Like Like x 1
    • Friendly Friendly x 1
  13. azaris

    azaris Well-Known Member AMS2 Club Member

    Joined:
    Aug 30, 2020
    Messages:
    1,177
    Likes Received:
    584
    The low-speed tyre damping fix feels like it has made a big difference in many of the cars. They just feel more controllable at low speeds, less bouncing, more responsive on turn-in. Or maybe it's just a placebo effect.
     
    • Like Like x 1
    • Informative Informative x 1
  14. Fernando

    Fernando Active Member

    Joined:
    Mar 11, 2016
    Messages:
    204
    Likes Received:
    94
    We share the same placebo effect then :D
     
  15. CrimsonEminence

    CrimsonEminence Administrator Staff Member AMS2 Club Member

    Joined:
    Nov 4, 2018
    Messages:
    9,790
    Likes Received:
    9,860
    Nah, it's not placebo. In some situations, handling was actually even a bit worse, when the first iteration of the tire damper additions arrived, causing some cars to plough etc.

    Now after further tweaking, it seems to actually improve the cars handling.

    The better response at turn-in also has other reasons, but overall, handling has definetly improved, especially at slow speeds, so it didn't just "fix" the low speed bouncing of formula cars.^^
     
    • Like Like x 2
  16. InfernalVortex

    InfernalVortex Active Member

    Joined:
    May 10, 2020
    Messages:
    415
    Likes Received:
    239
    I know this is a post from January, so apologies if this has long been resolved. Here goes...

    They could do this with the geared diff. I think it acts much closer to how a clutched diff should act in this game than the clutched diff does. This is a very unpopular opinion because I think people care more about the label something has being authentic than the feel it has. They will take a faithful 3d model of a 911 that has the physics of a Corvette over a bad model of a Corvette that has an impossibly brilliant, accurate model of 911 physics. I'm probably not even much better about this than the rest of us.

    Street cars typically have geared diffs or non (or barely) adjustable clutch diffs. The clutch diffs with ramps/clutches are race car items. If you have a common enough differential that is used in racing (like, say, a Ford 9") you can buy 1-way, 2-way, or 1.5-way diffs, or just regular adjustable ramp and clutch type diffs and use it in a street car, however. The number of "ways" is a crude way of referencing ramp angles. Clutch diffs are, by far, the most ubiquitous type of diff for racing cars, though, due to their extreme adjustability and flexibility. It is worth getting these correct.
     
    • Informative Informative x 2
  17. CrimsonEminence

    CrimsonEminence Administrator Staff Member AMS2 Club Member

    Joined:
    Nov 4, 2018
    Messages:
    9,790
    Likes Received:
    9,860
    Nope, also proven to have issues, actually. (Have tested the G40, which features a Torsen/Quaife diff, a lot at release, it's a miracle it does something now, after magic by Reiza was done, it was perfectly spool at release)
    So another hint, that it's the whole driveline, that needs care, instead of pinpointing it to the diff. (that works better with some more polished cars, than others)
    Or often def. open differentials.
    Definetly. AMS2 is not far away from being correct anymore, though. Believe it or not, but besides the still occuring "peakyness", the clutch LSD is now not totally broken anymore. ^^' (still has issues, of course)

    And btw. don't tell me, you f***ing F-Vintage ace don't notice anything different now :p
     
    • Funny Funny x 1
  18. Maser V6

    Maser V6 Assume nothing._ Verify everything._Have fun AMS2 Club Member

    Joined:
    Mar 27, 2020
    Messages:
    2,502
    Likes Received:
    1,422
    Agreed here, from flywheel back (retros) and also still engine lift off throttle lag imo. though its seems better on longer faster tracks this issues is still prominent on tracks with tighter corners.

    Promising, though time marches on and related issues still pop up a fresh from newer members it seems.

    If it it was possible to give a post's 2 ratings then I would have, I know you've had track time together!
    Funny and informative.
    @InfernalVortex good luck in Gen 1 tonight. I may look in later
     
    • Like Like x 1
    • Agree Agree x 1
  19. SlowPoke80

    SlowPoke80 Active Member

    Joined:
    Jun 8, 2017
    Messages:
    197
    Likes Received:
    28
    I've noticed that you can't get away with gunning the throttle nearly all the time now, at least with some cars. They still feel a bit too forgiving, though. I was using the V12 earlier. But that's one in particular that should really be more of a handful that it is. All the drivers from that era said the cars were difficult to master, because they only had one off-season to design them without driver aids.
     
  20. Wedsley Dias

    Wedsley Dias Active Member

    Joined:
    Mar 10, 2016
    Messages:
    223
    Likes Received:
    130
    V.1.1.2.5
    Opala Old Stock Race with physics Super Truck (AMS1)?
     
    • Like Like x 1
    • Informative Informative x 1

Share This Page