Flaws of the tyre model simulation

Discussion in 'Gran Turismo Sport' started by Lord Protector, Feb 3, 2018.

  1. kevlar_hybrid

    kevlar_hybrid

    Messages:
    116
    I agree in race driving, but this is the problem, this is not race driving. If you can, watch Kie25 this week at Brands. Just watch his pedal inputs. It's gas -> heavy brake -> massive long coast period into the corner where tyre scrub is more efficient than brakes and gives you turn in -> then full gas at apex.

    So yeah, not quite how Senna describes things, which is the point. The physics don't reward proper driving, the fast guys are playing this as they would play any game, finding how the game allows speed and doing that. This has a low relation to how actual driving or physics are though.
     
  2. Lazybike

    Lazybike

    Messages:
    475
    And in my opinion the game is better for it...it's a level playing field, it just comes down to acquiring the right skill set and applying it.
     
    kembro likes this.
  3. kevlar_hybrid

    kevlar_hybrid

    Messages:
    116
    I'm with you, as a gamer. But then they should really drop "The Real Driving Simulator" stuff for everyones benefit. It's like a nicely done Ridge Racer.
     
    BreakerOhio likes this.
  4. Krypton78

    Krypton78

    Messages:
    604
    Location:
    Germany
    Shouldn't the thread title be either

    ... of the tyre model

    or

    ... of the tyre simulation

    rather then
    ... of the tyre model simulation?

    :)
     
    kevlar_hybrid likes this.
  5. Haitauer

    Haitauer

    Messages:
    1,614
    Location:
    Finland
    This

    Ive been wondering since the beginning why they use such bad setups in BOP fixet setups. Especially the differential. Only thing i can think of, is that its easier to balance the cars with inferior setups than make the slow cars faster. Also drivetrain differences are easier to handle.

    Throw any praiano tune to a cars (preferably one of his non-alien tunes:lol:) and it feels like driving a different physics model.
     
    Last edited: Sep 26, 2018
    RCKakashi14 and Mirror_man like this.
  6. BreakerOhio

    BreakerOhio

    Messages:
    1,731
    Bingo! That last paragraph explains it perfectly.

    I have to adapt to this game style of driving. Watch senna, best motoring, drift king drive street cars, none of them drive the way gts requires for time attack. I know my real world driving techniques reward me in speed and time and this depiction of it is a games method to make a fair and simplified esport game to work for a larger audience to operate.

    I am noticing they are slowly introducing it back into the physics with updates, i cross it fingers they will do it right i hope! , they have to do justice for drivers type of cars, nsx and s2000.

    Updated: I just played the latest 1.28 update, im very happy! The rwd cars are feeling better to drive. Thanks PD for this update. The LSD and tires work much better now. I had faith but also understood the reason this was delayed.
     
    Last edited: Sep 27, 2018
  7. Lord Protector

    Lord Protector

    Messages:
    2,785
    Location:
    Slovakia


    1 and a half year after the game’s release... ford focus st (with stock road tyres) failing to climb a mild uphill...

    :banghead:
     
    AgentBlackDog likes this.
  8. ARTAsean

    ARTAsean (Banned)

    Messages:
    1,308
    You know that’s a 60 degree bank right? A FF car probably couldn’t climb a grade 1/2 that steep.
     
    Pedinhuh, Stencedaddy, VBR and 3 others like this.
  9. Hoopzah

    Hoopzah

    Messages:
    206
    Location:
    United States
    No non 4WD car could do that in 1st gear.. except my SAAB (joke). She's been up a 42°grade in Maryland, but it wasn't from a stop. Thats that's just silly
     
  10. Lord Protector

    Lord Protector

    Messages:
    2,785
    Location:
    Slovakia
    Okay... what you don’t see is... It goes downwards ffs :rolleyes: you can see I use 1/5th of the power and tyres go spinning like crazy and the car moves downwards.... and when using TCS it can climb it NO PROBLEM. Issue isn’t the angle, it’s the tyre model.
     
    The Cake likes this.
  11. Vegard

    Vegard

    Messages:
    2,879
    Location:
    Norway
    It looks more like a lack of friction due to the steep incline combined with GT Sport's clumsy auto clutch modulation.
    The TC negates the auto clutch problem and gets the car moving but ultimately surrenders to Newton's laws of mechanics.

    Remember, when starting on a steep incline you need to overcome both the inertia of the car and gravity.
    And in an FF you'll have very little traction due to the fact that most of the weight is transferred to the rear.
     
    ARTAsean and VBR like this.
  12. Lord Protector

    Lord Protector

    Messages:
    2,785
    Location:
    Slovakia
    have you ever driven a car? seriously... you would know that with 1/5 of the gas pedal the car would probably stall. it would definitely not spin the tires and even if it did, it would certainly not go downwards...

    it is like this because there the tires do not want to regain grip once they lose it... and that is why the car goes up, when using TCS, because it prevents the tires from spinning all the time.
     
    VBR likes this.
  13. Vegard

    Vegard

    Messages:
    2,879
    Location:
    Norway
    Yes, I've driven quite a few during the last 25 years. I'm also fairly well educated in classical mechanics and trigonometry.
    I'm making an educated guess here, a quick approximation. I haven't done any exact calculations so I might be wrong.

    The question of whether the car would stall, break traction or simply drive up the incline depends on COF, normal force, wheel torque, mass of the vehicle, angle of the incline and probably a few others that I'm forgetting.

    Have you done any calculations?
     
    Last edited: Apr 15, 2019
    ARTAsean, silverman_GT and VBR like this.
  14. Lord Protector

    Lord Protector

    Messages:
    2,785
    Location:
    Slovakia
    No calculations, just experience from motorsport...
     
  15. Vegard

    Vegard

    Messages:
    2,879
    Location:
    Norway
    Extreme hillclimbing I presume?

    May I ask what your hypothesis on the tyre model is?
     
    ARTAsean likes this.
  16. Lord Protector

    Lord Protector

    Messages:
    2,785
    Location:
    Slovakia
    i has nothing to do with extreme hillclimbing... SSRX's highly banked corner was used to demonstrate the problem with the tyre model... i demonstrated it in the OP months ago, on a straight line with no elevation... i already explained what is the problem, what else do you want?
     
  17. Groundfish

    Groundfish

    Messages:
    1,010
    Nice job man!
    I always get annoyed by that facet of the tire model when I am trying to climb the wall at SSRX in my Focus st.
    The tire model also had a few probs when I was running Fisherman’s Ranch in the Formula car.
    Coming out of tight turns it’s very difficult to get traction. Doesn’t PD know how sticky those tires are irl.
    It’s important to spend hours and hours doing things like trying to climb walls in a focus st. Time well spent!
     
  18. Vegard

    Vegard

    Messages:
    2,879
    Location:
    Norway
    I want to know why you're so sure that it's only due to the tyre model.

    A front wheel drive car sitting on a 45-60 degree incline will have very little weight on the front tyres.
    When you try to launch a car sitting on a 45-60 degree incline you're going to get even less weight on the front tyres.
    With so little weight you get very little grip.

    With the clumsy clutch modulation, turbo torque, a short(ish) 1st gear, and with very little grip available to overcome the inertia and gravitational pull you're going to struggle to launch it without slipping. I agree that the dynamic friction is less than perfectly modeled, but I don't think your test results can be attributed to that alone.
     
  19. Hoopzah

    Hoopzah

    Messages:
    206
    Location:
    United States
    Try it in 2nd or even 3rd without TCS.. works fine for me. Just as real life, you can't expect to climb a steep grade like that in 1st gear with crap tires in a car that only pulls the front wheels from a complete stop.
     
    Last edited: Apr 15, 2019
    ARTAsean likes this.
  20. Knukel

    Knukel

    Messages:
    449
    Have you tried this test in other games like AC (Monza oval) or PCars2? Temped to test this in AC later today...
     
    super_gt and Lord Protector like this.
  21. Lord Protector

    Lord Protector

    Messages:
    2,785
    Location:
    Slovakia
    I haven’t. I’m tempted to do it in my real car... and I think I will, it’s the only way to prove my point right now, as some folks over here think this behavior is normal :lol:

    Oh and @Groundfish , I talked about the issue for about an hour with PD directly, on two occasions actually :) they are well aware of it, unfortunately the issue seems to be more complex.
     
  22. Groundfish

    Groundfish

    Messages:
    1,010
    @Lord Protector
    Hmm you’d think wall climbing in the Focus st would be a high priority.
    It’s an important part of the race.
    ;)
     
  23. Vegard

    Vegard

    Messages:
    2,879
    Location:
    Norway
    I think you're underestimating how steep that banking is.
     
  24. Lord Protector

    Lord Protector

    Messages:
    2,785
    Location:
    Slovakia
    And I think you’re overestimating how much traction is loss during wheelspin on a track-grade asphalt.
     
  25. zzz_pt

    zzz_pt Premium

    Messages:
    7,697
    Location:
    Portugal
    It's not problem with the banking or the focus. Low speed grip (pretty obvious in tight corners like T2 at Maggiore or Dunlop at Suzuka) has been an issue on GT Sport since day one. I haven't played in over a year and this week I thought about doing some dailies so I came here. I was hoping PD had improved the situation tbh, with the FIA live events and stuff.

    I guess GTS is just another game and this (low speed grip) is part of it by now.
     
    Lord Protector, ASH32 and Scaff like this.
  26. superwally

    superwally

    Messages:
    1,242
    Location:
    United States
    Then why beat a dead horse? Finding the worst-case-scenario to prove a flaw that you know PD is unlikely to correct makes very little sense. You and several other frequent posters on this site are fairly well off the rails with PDOCD. What's PDOCD? It's when you just can't turn on the game and enjoy what it has to offer, without focusing compulsively and obsessively on what PD gets wrong. Most of us know the game is flawed, but somehow we manage to enjoy it. Life goes on.
     
    _ApexPredator and iCyCo like this.
  27. Groundfish

    Groundfish

    Messages:
    1,010
    PD actually improved the games grip and physics quite substantially over the course of several updates within the last few months.
    I recommend it highly. It’s much better than at release with the low speed balance issues on hairpins mr gr3. That was fixed a long time ago. At release t1 Nurb gp was unprovable in certain cars The physics are quite good now minus pinball collisions.
    I only play on wheel tho.
    Pd struck a great balance physics wise imo.
     
    Knukel likes this.
  28. jjaisli

    jjaisli Premium

    Messages:
    2,577
    Location:
    United States
    The relationship between the tire contact patch and the road, and all of the factors in between (friction, temperature, compound, weight, pressure, delamination, flex, lateral load, tire size, width, profile, etc) is highly complex. And probably far beyond the capabilities of the PS4 (at least as far as making those calculations on the fly, and doing so accurately, as the relationship between all of those is constantly changing and each affects the other). That said, as somebody who is NOT Fernando Alonso, nor a trained engineer, but who HAS driven over 7,000-km on various real world racing circuits, tire modeling is probably the single weakest link in every sim or simcade I've ever played. And it's a particularly weak point of GT in my opinion.

    Tests like these are not very scientific but they highlight some very basic flaws in the physics engine.
     
    Last edited: Apr 15, 2019
  29. CallawaySS

    CallawaySS

    Messages:
    88
    The accelerations seem to be more realistic now. Or maybe the guys in this video know how to launch cars properly? Either way, I think it's close to reality
     
    xBSxChadSpeed likes this.
  30. Scaff

    Scaff Staff Emeritus

    Messages:
    23,175
    Location:
    United Kingdom
    I'd disagree. I just done a load of standing start testing in GTS, PC2 and AC, and in this regard GTS has gone backwards over past builds (oddly as the rest of the physics engine has got better.

    Once again you need a car with a massive amount of torque to get any step out off the line from a standing start.
     
    zzz_pt likes this.