“ are rfactor2 physics broken” video

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

  1. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    I been doing some hotlaps in normal driving technique everything seems fine BUT then next time around i start to add Tad bit more steering angle even if i'm already understeering and looking at delta i'm faster......I hate this.
    In my normal driving i can already see front tire are on the edge they are heating up (i can see that on HUD) , i already feel understeer both visually and in FFB.
    BUT still if i abuse the tire to add couple of more steering angle OVER the understeer i get more bite.
    My explanation is tire model need to be checked properly.
    And if the data in tire physics is officially provided and I can still do this... then the core physics has problem instead of car alone.
     
    Nieubermesch likes this.
  2. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    I am not sure of what I am about to say, but I think some ammount of oversteer in a corner can actually be faster than grip, as if the underteer isn't too prounounced, it will allow you to take lines that are faster the apex under grip limits. I will run some laps here. Even if what I said was true doesn't mean that Rfactor 2 way of doing this is correct
     
  3. Comante

    Comante Registered

    Joined:
    Nov 20, 2013
    Messages:
    1,694
    Likes Received:
    1,218
    It must the noted that each kind of tire get the maximum grip at a certain slip angle. For old tires this is around 7 degrees (if I remember correctly) for modern tires it could be half that amount. This mean a certain amount of understeer actually get you more grip, again, without measuring how much the effect is pronounced, it is impossible to determine if it is excessive or not.
     
    Bruno Gil and Nieubermesch like this.
  4. Comante

    Comante Registered

    Joined:
    Nov 20, 2013
    Messages:
    1,694
    Likes Received:
    1,218
    On a second note, steering angle without information about load on tires mean little: one thing is to understeer under full throttle application (in fast turn, or in corner exit) , one thing is to understeer in corne entry, under even slight braking to load the front axle.
     
    Nieubermesch likes this.
  5. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Seems like you have some sort of degree on physics, Comante. Am I wrong? Maybe also some modding?
     
  6. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    Slip angle of tire has nothing to do with perceived understeer. If driver turns the steerign wheel he perceives the response of a car/tire, he does not perceive the precise difference of vehicle direction and tire direction angle. Driver knows that car is understeering when applied further steerign angle doesn't make the car to steer as effectively or any more at all.

    There are two points on slip curve which could be perceived through steering feedback very similarly in terms of tire performance. One point is before the peak angle, other point is after it. Interesting thing about slip curve vs self aligning torque (SAT) of a tire is that due particular effects peak S.A.T happens before peak slip angle. So it makes sense to steer extra bit for extra turn in, when it might seem through a steering that there is an understeer. However, it must not be mistaken with the second point after the peak angle. But I think Slip_Angel does not mistake it, and he does mean that he gains more laptime even when steering way past peak slip angle. Many fast drivers IRL can often be seen sawing steerign wheels while at the turns, they basically are guessing the peak slip angle by jumping in between those two points - when SAT is maximum and when tire performance really begins to drop down.

    This is complex issue, can't judge it easily. There perhaps could be some times when little bit of extra understeer would not only not hurt laptime, but perhaps somehow provide some benefit. Also speaking of delta time, immediate delta time reading is not whole thing, you might have gains during the turn, and then possibly have losses if you happen to exit slower.... I don't know, it can't be easily judged. But I think that Eau_Rouge to Radillon porsche understeer oboard you have shared some time before (in post #263) @Slip_Angel, has illustrated rather well what you are talking about.
     
    Last edited: Jan 15, 2021
    Bruno Gil, avenger82 and Nieubermesch like this.
  7. Comante

    Comante Registered

    Joined:
    Nov 20, 2013
    Messages:
    1,694
    Likes Received:
    1,218
    Not at all, my math suck too much to allow me to study advanced physics, I'm just a curious person that read these forums since day1.
     
    Nieubermesch likes this.
  8. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Ahah, well, it gives you a little less authority, but respect, as you seem very knowledgeable for someone who isn't actually propperly educated on some of this stuff.

    I am studying mechanical engeneering and I have to say my math also isn't helping me, but it's getting better. When I think about the brains needed behind a physics simulation I have to suspend my more critical judgements or at least the way I present them and most of the time I will probably be actually wrong about what I am criticizing.



    Leaving this here since Slip Angel brought an important subject. It's a superficial look, but maybe it can help judging from a different perspective the understeer aspect.
     
  9. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    @mantasisg that video i posted is exactly what i'm talking about.
    I look at delta going negative (fast) after exit if i drive in such manner.
    (although not as much as that guy but enough to feel frustrated on this behaviour)

    in rf2 i noticed many times that i have to overdrive a little to be fast.
    This seems unrealistic IMO.

    Using more steering than what my sense, experience is telling me
    Using throttle to put car in semi sliding state where I don't have to correct it with steering.

    What i think is realistic is driving either exactly on the limits or just under it. IRL if car is understeering it can be forced to a VERY small degree but this region is much bigger in RF2
    Similarly if car is over the limit at rear there is very small region where driver car "drift" it out , again this region is IMO much bigger in RF2.

    One of my observation after driving in ACC and comparing it to RF2 is->>
    I generally have hard time feeling if either i'm at peak or sliding because the sliding state in RF2 is so well controlled it doesn't seem to be overdriving.
    ***Note i'm speaking relative to ACC here not that i can't feel overdriving in RF2.***

    While in ACC on the limit and just over the limit state is very narrow and when you slide (either understeer or oversteer) the feeling is very ON/ OFF switch like, i.e you can very clearly both SEE and FEEL that you are understeering or oversteering.

    Edit: This is the reason why even though setup parameters (ARB, camber, wing etc ) are working intentionally BUT it is the tire that is providing us the exploitability.

    The operating window of tires in all aspects (like slip angle , heat , too much or too little camber, lack of rear wing downforce ETC ) needs to be more narrow. I'm very confident that problem lies in tires physics.
    Not individual cars because this sort of exploits are working in all different cars, different categories of cars.... hence it has to core physics issues.
     
    Last edited: Jan 15, 2021
    Nieubermesch likes this.
  10. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Last time I checked ACC also allows a fair ammount of understeer to be used to drive faster, but I agree the simulation of grip/sliding transitions is more believable, but ACC has a problem at how much slowing down tyre sliding provides, so it isn't quite right there for me, apart from other behaviors of the car that seem a bit damped making the car feel less agile then what I feel looking at real life onboards, while Rfactor 2 nails this better and then also seems to go overboard on that aspect... They should meet on a middle ground, I don't know.
     
    Slip_Angel likes this.
  11. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Well, some of the new cars feel a little bit more unforgiving, as I've been testing some exploitative driving and I get very snappy behavior on some GTE cars, so try more of them.

    I believe tyres can be made to have different characteristics, and probably devs will tend to continue allowing some of the behavior for all the tyres built, sharing some of the characteristics. Modders also copy a lot of times the tyres made by Studio 397 so that might be why they share the same feelings.

    Wished I could actually mod in this game and try to make a tyre idealized to what some of us feel is a more realistic behavior.
     
    Slip_Angel likes this.
  12. Dave^

    Dave^ Registered

    Joined:
    Oct 14, 2019
    Messages:
    263
    Likes Received:
    121
    What are you comparing to in terms of realism? Not having a go, I genuinely want to know your experience of sim vs reality. What is your sim setup etc?

    You say "overdrive a little", but in reality, you have to overdrive slightly to get the most out of the tyres (see above with the discussion of slip angles... etc.) With zero slip, you're underdriving.

    Speaking of slip angles -

    https://www.windingroad.com/articles/blogs/speed-secrets-3-ways-to-drive-faster-part-2/

    https://speedsecrets.com/q-what-is-slip-angle-how-do-i-use-it/

    Also - https://speedsecrets.com/eBooks/ - not just slip angles but many aspects.

    I don't know how "real" RF2 is, I've only ever driven my shitbox MX5 on track, & that was with street tyres, so I have no idea how a GT3 car should feel.

    I'm also using a fixed seat with a Logitech G920 & a 43" TV, so I've no idea what the sim is trying to tell me while I'm doing it.
     
    Bruno Gil and avenger82 like this.
  13. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    @Slip_Angel It is totally the tires that commands the functionality of setup and results of a car working like a on/off switch like modern high performance twitchy tire should work, or like some low performance forgiving street tire.

    Also IMO ACC has been slowly migrating towards having increasingly forgiving tires the whole time since it s first release in EA.
     
    Last edited: Jan 15, 2021
  14. Andrew Hollom

    Andrew Hollom Registered

    Joined:
    Mar 14, 2019
    Messages:
    39
    Likes Received:
    14
    Won't winding extra lock on to get grip catch up with you very quickly during a race?

    I've raced with people who have used this technique for years, and you can spot them by the smoke pouring off their outside front tyre. They tend to qualify reasonably well, then fade badly during the race (anything over 20 minutes is too long for them).
     
  15. Comante

    Comante Registered

    Joined:
    Nov 20, 2013
    Messages:
    1,694
    Likes Received:
    1,218
    Well it could be a good learning experience to build a square car from scratch: standard tires from database, no suspensions, no chassis flex, no aero, just 4 corners with tires. Investigate how it behave, and then, step by step add some functionality as soon as it is learning it: simple suspension geometry at first, chassis flex, ARBs, differential, aero. It is probably a multi year hobby, but at some point you will know almost everything and how it work.
     
    Nieubermesch likes this.
  16. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    Unfortunately my talking is based on GTE cars majorly. (i drove lmp2 as well little bit)
     
  17. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    My comparison is based on what other sims are doing.
    The 2 biggest competition of RF2 in GT cars is ACC and iracing.
    ACC and iracing share similar punishing traits of driving. So it is 2:1 ratio where only RF2 is doing its thing alone while other 2 are quite close.
    My sim setup is extremely simple a T.V 20 ish inch (idk exact size cuz its old T.V not monitor) close to my G27 on table and my driving seat is a chair.
     
    Last edited: Jan 15, 2021
  18. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    The point is that such driving technique would NOT even work for one hotlap. It is touch too excessive.
     
  19. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    I agree. I sincerely hope developers look into tire quickly.
     
    Vladie and Nieubermesch like this.
  20. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    I was testing and I wonder if part of it is down to slowing down when car starts udersteering, wich in turn will actually diminish understeer, giving back grip, where you can then apply more steering lock, feeling like you can understeer more and apply more lock, when in actuality is just understeer as faded somewhat with loss of speed. Is this realistic? Probably to an extent. A game that showed such behavior was PC2 in a lot of cars. I also remember about a member called Spinelli that used to rant about the cars yaw behavior, wich also can contribute to this, as when car is on great steering angle and understeering, if somehow yaw rotation can be "exploited", giving some forgiving nature of the tyres, it can be compensated to actually take advantage of inducing understeer and then keep the speed rotating the car, on a perfect balancing act we see some aliens do.
     

Share This Page