“ are rfactor2 physics broken” video

Discussion in 'General Discussion' started by GTClub_wajdi, Dec 29, 2020.

  1. Yzangard

    Yzangard Registered

    Joined:
    Dec 6, 2018
    Messages:
    351
    Likes Received:
    173
    I counter low rear wing by preventing front roll as well, nothing special here but there might be an exploit somewhere, never tried to abuse the physics so can't tell.

    Do you use telemetry / motec ? Asking because if you do, we have proofs and we can check stuff (despite ACC is only providing a very limited set of channels). I can try it as well but not on SPA because this track have an issue with grip, what track did you use ?

    PS : track surface is a very important factor as well, some ACC tracks aren't realistic in that sense...last made tracks are good but some of the first batch aren't, like SPA for instance, the corner at the end of the straight following the Raidillon is an obvious example, it is very easy to lose rear here, artificially easy.

    Another point : don't forget that due to rFactor 2's incredible ffb, you feel perfectly the car so you might be smoother than in ACC, when you lower rear wing you have to be VERY smooth with break release or you may unsettle the car and lose rear grip, it happens to me a lot in ACC, a lot less in rFactor 2, I thought it was grip but as I said above, grip is pretty much the same in iRacing, ACC and rFactor 2 for what I've tested and checked with telemetry.
     
  2. MarcG

    MarcG Registered

    Joined:
    Oct 4, 2010
    Messages:
    6,854
    Likes Received:
    2,234
    Quick, bounce the AI threads :D
     
    SoloWingX and Dave^ like this.
  3. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    Thank You !
    I will just post couple of bullet points, these issues (or no issues) need to be looked into. (some of which many people agree on and some of which they disagree on).

    *Tire pressure physics

    *Tire is sustaining excessive slip angles in both understeer and oversteer without big consequences as one would expect. basically car is very much working even if there is good amount of understeer and/or oversteer present in driving.

    *Aero sensitivity to YAW, cars are very controllable in high speed slides and sometimes can be recovered from high degrees of rotation.

    *Aero sensitivity to pitch and squat for all aero dependent car. ( This is personal observation but from other sims the similar class cars are very pitch sensitive under braking. In ACC blogs it has been stated many times that car needs bumpstops to stop pitch. otherwise excessive pitch can even stall underbody aero if not in control. Similarly i think that car are not shifting aero back during squat this allows for small exploit that people run much softer springs and dampers at rear without getting SIGNIFICANT power on understeer, it is there but not as significant IMO)

    * Some top guys can be seen to exploit the tires ability to grip even when adding excessive amounts of steering lock. example->> (take a look at 0:24 and 1:19). I have also seen PRO guy named Risto Kappet using such techniques

    *People are running leman setup without big grip loss . especially broken with pairing that with detached ARB (especially rear one)

    *The Lack of aero grip at high speed can be much heavily countered by mechanical understeer setup. It is much more exaggerated in RF2 as opposed to IRL, As far as i understand it cannot be used IRL to such a degree.
    Basically low wing, high mechanical understeer setup is one of a exploit.

    *personally I don't think ARB is having as big effect on overall setup (this is personal if you guys have time take a look, keep it low priority)

    *While camber effect is present but generally all of the fast guys are running minimum camber because penalty is not that big or maybe cambers are not working as intended (Also check the TOE effects on overall performance as this also being used at very low values).

    *people are also running very low amount of rake without significant loss of grip PLUS it also seems that higher ride height are not taking away downforce as much as one would expect

    *Engine inertia issues meaning cars are cutting off engine even if you let go of clutch smoothly at MAX throttle/RPM . basically no burnout ( This cannot be done on any part of circuit just to be clear unless it is raining or very low grip conditions ). IMO all cars are affected by these.

    Overall these exploits can be used on all the DLC cars (probably on all cars).

    there are some more issues and exploits as well these are currently in my head please take them into notice as well, Thank you for reading.
     
    Last edited: Jan 7, 2021
    Nicolay.G, Ermz, Vladie and 2 others like this.
  4. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    ^ As in that video. Yeah, there is slight reseblence to how you'd expect to drive 1950s tire :D I would be interested to see car going up to raidillon still in compression area with such massive understeer, I am afraid IRL such action would massively destabilise the car, probably any car, from any era. It would be interesting to check onboards for any similar examples, it would be interesting to see understeer in there IRL, not even speaking about this amount of it and for pretty much whole curve.

    However, at least in 1:19 there was snap oversteer, although it it so long for front to bite in to cause shift to oversteer and it looked to happen slowly.

    Edit: in very quick search didn't find proper examples from RL, but this video is a bit similar, yet not the same stuff. But here I see understeer happpening there, and how it progress and where it lead to in that case, obviously RL physics sucks, ACC and rF2 has way better physics than RL:

     
    Vladie likes this.
  5. Paul Jeffrey

    Paul Jeffrey

    Joined:
    May 21, 2016
    Messages:
    660
    Likes Received:
    4,336
    Thanks @Slip_Angel that's a useful summary - much appreciated.
     
  6. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Hoping this issues don't take an enormous ammount of effort and time to solve. Even if some of us don't abuse the physics, I do sometimes get caught driving way too stupidly and not getting the penalty I expected.
     
  7. GTClub_wajdi

    GTClub_wajdi Registered

    Joined:
    Feb 28, 2012
    Messages:
    3,238
    Likes Received:
    572
    I just wanted to discuss about rfactor2 physics. You know I love rfactor2 but there are a lot of improvements that should be done to the physics engine.
     
  8. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    Thank you, for listening to community.
    Keep up the good work.
     
  9. Bruno Gil

    Bruno Gil Registered

    Joined:
    Oct 12, 2020
    Messages:
    128
    Likes Received:
    79
    I think you did not understand my point, so I'll explain in a more detailed manner why that test is flawed.
    First of all, you are taking the corner full throttle, from the sound of the engine. "Detaching" (whatever actual roll stiffness that is in this case) the arbs isn't gonna give you any more power so you're Always gonna get to the end of the corner at the same speed.
    So, at max, what you can see in your test is a shift in the grip balance. I believe we can see this: car is rolling less, leading to a more effective diffuser, which shifts the grip rearwards, making you go wide... Here, that's your explanation.
    Also, your attitude is quite arrogant when you get presented with counter arguments, so I'm quite sceptical about having you be our spokesperson when it comes to this. I do appreciate the effort you're going through to see the game improve, but you cant just keep saying "games broke" when you cannot actually explain how.
    Remember, setup changes are relative. Maybe we just don't have that much of an adjustment window as in acc for example, so going an extreme isn't gonna necessarily make the downsides be bigger than the upsides.

    Lastly, about the "Pro" (Esports) drivers thing...remember guys, most pro "anything" don't want competition, and some will actually go as far as spreading misinformation to keep the competition on the back foot... Just a hint for life

    Edit: on that last Porsche video by Henri: first, the porsche in this game got a lot less steering sensitivity than most other gt cars.. However, there is obviously too much lock on some corners. I did a cup with those cars recently and I can tell you this though: while it might be good for Hotlapping, you do this in a race and within 3 laps your tires are shot (at the very least, you need to back out - substantially - for a couple of laps. This resulted in me being able to keep with the pace of other drivers who drove like that for the first laps, while I was driving smoothly, and eventually win some races on tires, as I could sustain that pace until the end of the race and they couldn't. While I'm not sure in RL the tires could take it for a whole lap, let alone three, I don't see this pushing as a huge problem because of that.
    TO BE CLEAR: I do agree with some of the bullet points you wrote in your last post. It's just, others, in my experience, are not an issue and can be understood if you think of other variables when a change doesn't produce the result you expect.
     
    Last edited: Jan 7, 2021
    Yzangard and Slip_Angel like this.
  10. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    Your explanation is good , on why i ran wide with stiffer ARB. I thought of that as well.
    But next thing i tested is i ran same experiment with low wing just to see if my underbody is working better but in results i got more oversteer in with ARB.Not just one corner i tested this on whole lap.
     
  11. Stefan_L_01

    Stefan_L_01 Registered

    Joined:
    Nov 18, 2012
    Messages:
    595
    Likes Received:
    386
    I dont know what car you found, ride heights with 135mm, no common content...
    Here is one quick example if you want:
    It´s a simple acceleration in 6th gear from 0 to full rev. On the left is cursor value, on the right of charts min, max and (not important) average.
    First chart shows suspension pos, sow our suspension is compressed by 10mm
    Next shows individual for tyre deflection (note: at v= 0 deflection is positive, so more positive means a reduction of effective height) and ride height, both directly from telemetry.

    SO if ride height includes the influence of tyre deflection, this would mean it is a combination of tyre deflection and suspension deflection. As ride height gets lower with more downforce, and tyre deflection heigher, one can sum them up to get the difference for suspension influenced ride height only (the absolute value do not interest). This is done in last chart.
    The difference in suspension based ride height change is less than 5mm.

    Can you explain me how a suspension changes pos by 10mm and the suspension based ride height at wheel changes only by 5mm? Every common suspension I know, McPherson or double leaf, will travel LESS than the wheel as it is more inside to the joint, and quite often even angled (I exclude formula and proto types with leverages and force transmissions of suspension)
    BR
     

    Attached Files:

  12. Yzangard

    Yzangard Registered

    Joined:
    Dec 6, 2018
    Messages:
    351
    Likes Received:
    173
    Fair points and I agree that @Slip_Angel was a bit aggressive, even peremptory, but I guess that's because he tried several times to make his point and he didn't necessarily deal only with people who accepted the discussion.

    Having said that, I also think we should try to set up some kind of communication with the people at Studio 397 to expose different points and establish a testing protocol to express those points with data, not just opinions or feelings. In other words, have concrete and facts, not inferences and preferences.

    Not speaking English correctly (I am French after all, worse I am Parisian), I can't do it and I don't have the legitimacy to do so anyway but I can see with the French speaking part of the pilots on rFactor 2 to go back up their subjections.
     
    Comante and Bruno Gil like this.
  13. Bruno Gil

    Bruno Gil Registered

    Joined:
    Oct 12, 2020
    Messages:
    128
    Likes Received:
    79
    I'm actually curious now. This could be due to rear height being higher because of less df compression and thus not allowing the diffuser to work properly. Or you might be onto something. Can you test lowering the car and see if you get different results?

    EDIT: I just checked your setup and, in my limited experience with rf2s gts, 2cm rake is way too much, and the car overall is quite high, so I do believe the "too high for the diffuser to work" theory
     
    Last edited: Jan 7, 2021
  14. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    The point of adding ARB so that i could run higher rake, as my understanding was that stiffer body means stable platform.
    So even if i tested with low ride height i will be compromising cornering speed and/or run little wide on fast turns.
    Not to mention with detaching ARB i also expected that tire to roll very bad on outside edge meaning less grip but again not happened.

    I have no telemetry to back this up.
    Best corner in my testing is final corner and T8 at old zandvroot GP.
     
  15. Bruno Gil

    Bruno Gil Registered

    Joined:
    Oct 12, 2020
    Messages:
    128
    Likes Received:
    79
    In cars where your only aero adjustment is the rear wing, you shouldnt(in theory) use high rake with low rear wing, as both those things push the aero balance forward.
    I don't understand what you mean by the low ride height and cornering speed, but, in general, a lower ride height will give you a lower centre of gravity, and hence faster cornering speeds. Especially on fast corners. Also keeps the underbody aero working.
    So, as you see, you are expecting certain results when you are not considering all variables. At least you now show you're trying to understand that better. So, don't just assume that because "x" caused "y" , "z" would cause "a" without being willing to test "z" . And when results do not match your expectations, do try to understand why. In my experience with rf2, in the vast majority of the times, I will come to an explanation that makes sense AND leads me to better understand race car dynamics. Welcome to setup building...
     
    Comante likes this.
  16. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,346
    Likes Received:
    6,572
    The mod I used is in development, it's an old tintop with very little aero on it. Hence the high ride height. All I was trying to do was isolate suspension movement from tyre deflection, you can see the tyre deflection has fluctuations not present in the suspension position and those fluctuations are present in the ride height.

    Can I ask which mod you're using for that graph? I'd like to look into that more to help come up with answers.
     
  17. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    By high rake i mean 11 mm on porsche RSR. Also what i meant by lowering height is->
    I was talking about lowering rear ride height would cause understeer hence cornering speed.
    True that lowering ride height is better CG but remember these cars need rake to generate good downforce and if i do not want to lift front splitter up in process of lowering rear CG then some rake is necessary. ( the highest rake i used in GT car in RF2 is 20mm on corvette as that car seems to need it PLUS i was running softer springs)
    I might not know much but i have taken quite few things into consideration while making these setup, you are just assuming that i did not because i have not typeed all the considerations,setup philosophy here.
     
  18. Stefan_L_01

    Stefan_L_01 Registered

    Joined:
    Nov 18, 2012
    Messages:
    595
    Likes Received:
    386
    That was a cup Porsche. The 2 following charts are AM GT3 and 488 GT3, the later one I hit coupling at the end to avoid any strong torque moments on the tires (which arent that big in 6th at high speed anyway)
     

    Attached Files:

  19. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,346
    Likes Received:
    6,572
    @Stefan_L_01 the problem with DLC is you can't check things like undertray point positions. And then we're back to the "is it a game problem, or is it a car problem?" conundrum.
     
  20. Bruno Gil

    Bruno Gil Registered

    Joined:
    Oct 12, 2020
    Messages:
    128
    Likes Received:
    79
    Lowering rear ride height would cause RELATIVE understeer to what you have now; doesn't mean it won't be better. If you don't want to include rake changes, just lower front and rear by equal amounts, should do the trick.

    I have no idea what you mean by raising front splitter and lowering rear cg... Theres no rear cg, there's one abstract CG that moves up, down, front and back, depending on general height and rake. By just lowering the rake, you will be moving this CG both down and back.

    You can take QUITE a few things in consideration, but NEVER everything (nor can anyone else). So, the correct process, imo, is: come up with an hypotesis; test it; if it doesn't verify try to understand why; adjust and retest. Isn't that how scientific studies are done? You seem to be trying to skip a few steps, and proceed to be surprised when results don't match expectations. And I'm not assuming anything, I'm just constructing arguments based on the info you've provided me with.
     

Share This Page