Question about the tyre model

Discussion in 'General Discussion' started by Rich Goodwin, May 2, 2014.

  1. Rich Goodwin

    Rich Goodwin Registered

    Joined:
    May 3, 2012
    Messages:
    1,219
    Likes Received:
    9
    OK, so first off, I bought rFactor2 not very long after it was first announced. I have dropped in and out from time to time, mostly on new builds just to see how everything is progressing. For a number of reasons, I have never stayed around all that long. I am primarily an iRacer and this is the source of my question or rather, mini novel here.

    Over there, the “New Tyre Model” in its many versions, is touted as the most correct model there is when talking mathematical correctness and how rubber works and things. OK, I get that, I am on board with that, I support that. It sounds logical to approach it in that way even if hugely complicated.

    Recently I joined a league on rF2 that uses the brilliant URD GT cars and I decided to really put some time in to the sim. I very, very, very quickly realised I couldn't just hot lap. I needed to asses my style, and have an amount of sympathy for the tyres. This was after destroying my front right with just a few lock ups.

    I have fully fallen in love with all aspects of the sim. It is so dynamic and in my opinion superior to anything else out there. Just the online over at iRacing is better, I don’t think anyone can argue that.

    I digress. My question is about the Tyre model here. I'm not expecting any insider info or anything like that and I would prefer to stay away from being fanboys.

    If iRacing say they are doing it the “correct” way and if people are to believed, the only ones to be doing it that way, how does ISI do it?
    Many people would simply say that the “numbers are fudged” here or something just as dismissive. Is there any documentation or anything on this? I am just really interested how things are done with different sims. I would love just to have a good dialog with knowledgeable members of the community here.

    Either way, It’s a brilliant sim. Keep up the good work.
     
  2. 88mphTim

    88mphTim racesimcentral.net

    Joined:
    Sep 23, 2010
    Messages:
    10,840
    Likes Received:
    314
    I would wager both iRacing and ISI target the same set of things, and believe they're doing it the correct way. Which makes the differences extremely interesting.
     
  3. Rich Goodwin

    Rich Goodwin Registered

    Joined:
    May 3, 2012
    Messages:
    1,219
    Likes Received:
    9
    This is my thinking exactly. I don't mind admitting I love both sims, just the difference is......well, huge, frankly.

    If only all the great minds of sim racing could work on one, perfect sim!
     
  4. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,072
    the Iracing tire model is very good, however when (& if ) they say (or anyone claims for that matter) that the Iracing tire model is correct & a good match to the given simcar ,
    when it's changed / altered / "updated" & it then makes the sim car feels completely different they must ask the question
    how much it actually matched the sim car before or how much it matches each and every change

    a sim car is someone's interpretation of the real car


    re.rf2 -I think I might complain about there being to much grip, just in an attempt to offset the few that seem to suggest there is not enough.

    p.s I think the grip levels are just fine as they are in rf2
     
    Last edited by a moderator: May 2, 2014
  5. PLAYLIFE

    PLAYLIFE Registered

    Joined:
    Dec 26, 2010
    Messages:
    743
    Likes Received:
    126
    Modelling a complex structure, such as a tyre, there would be many approaches one could take to simulate reality and all of which would include a number of assumptions. Using, what we call in the industry 'engineering judgement', you can make certain assumptions which are considered to be appropriate and valid due to the logical thought process. There may be many different assumptions all of which could be considered to be valid; perhaps there is a degree of 'correctness' associated with them, but given the lack of evidence (the fact that an assumption needs to be made indicates there is still uncertainty), some may be more valid than others but all could be considered to be correct.

    In my line of work, we've modelled things mathematically accurate given the known data but the results just didn't seem or feel 'right'. There are known unknowns which we can accommodate to a certain degree, but then there are unknown unknowns which we cannot account for.

    Further, as Adrianstealth said above, interpretation can pose a whole new threat to the theory behind the model.

    So mini-novel response aside, there would be many ways to model a tyre and all may be correct but they all most likely won't be the same. Which is more right? We'll probably never know.
     
  6. stonec

    stonec Registered

    Joined:
    Jun 19, 2012
    Messages:
    3,399
    Likes Received:
    1,489
    Speaking of the tire model, ISI could make a bit more noise about it in my view, as it's arguably the best in any consumer simulator at moment, but sadly so little talked about outside the mod community. I know there is the tire tool, but it's not market sexy as such. It would be cool to have for example some sort of animation video showing how the tire reacts to different extreme conditions.

    What ISI has achieved with the tire model is most impressive. I remember a slight change of feel in cars around a year ago at the time the Marussia was released. The handling on limit with certain cars was a bit too much on/off before that in my view. But after last summer it has felt rock solid on almost every newly relesead car, from karts to street cars and touring cars. The way iRacing has hyped their tire model you would think it's the next messiah, but it has taken them five iterations and five years to reach a somewhat believable result, each iteration with a completely different feel.
     
  7. 88mphTim

    88mphTim racesimcentral.net

    Joined:
    Sep 23, 2010
    Messages:
    10,840
    Likes Received:
    314
    Dallara use an engine most of the sim community have heard of (not rF Pro). I think it is more an issue of choice than it not being possible to simulate aspects effectively. :)


    Sent using Tapatalk
     
  8. Barf Factor

    Barf Factor Registered

    Joined:
    Nov 2, 2012
    Messages:
    403
    Likes Received:
    18
    I agree with this completely. rF2's physics are its strong point. Unfortunately a lot of people don't seem to appreciate it at first and get caught up on flaws in other areas.
     
  9. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
    Ive been spending some more time with iracing lately due to the pure online aspect of it, not the physics, and there is no way the physics overall are as good as rFactor 2's. Having said that I have only been spending time with the free cars and the couple that I have purchased. Mainly the Skip Barber, Mazda Miata, Spec Ford, and the Star Mazda.

    The FFB differences between rFactor 2 and iRacing are just as large if not larger than the physics differences. Iracing's FFB is terrible to say the least. You are driving almost purely by audio and visual cues. The FFB just gives you bumps and jolts from bumps/curbs and that's pretty much it. No feeling under brakes, no front tyre feeling, you aren't feeling anything from any part of the car, it is so dead in terms of any feel. It's like playing with FFB turned off except for bump/curb effects, and then enabling centre spring in your wheels control panel.

    The online system is great though.

    One last thing. You really seem to need to do some really weird/unnatural things with the car in order to get the real top times in iRacing.

    I still have fun with it and enjoy it though because it never really feels arcadey/semi-simmish, it's just not as good as rFactor 2 that's all.
     
  10. hariseldon

    hariseldon Registered

    Joined:
    Sep 16, 2013
    Messages:
    112
    Likes Received:
    0
    I'm in the same boat as Richard, and I dragged him into the league in question. I'm absolutely loving the dynamic racing, the challenge it presents. Tonight we're doing an hour at Imola and we will have rain half way through the race. Doing that on iRacing is impossible. The drying line, the rubber lay down, timing when to switch tyres, it's brilliant. I've turned into an rf2 fanboy. The physics in this sim are quite simply better than anything elsewhere.

    Sent from my HTC One_M8 using Tapatalk
     
  11. AndyG

    AndyG Registered

    Joined:
    May 3, 2014
    Messages:
    34
    Likes Received:
    5
    With the latest builds I can at long last get into rF2 as my machine can run it decently. I race over at Race2Play. Raced 11 Sims over time. My love is ARCA Sim Racing. All I would say is coming from ARCA (Stock Car racing including oval and road courses too), immediately I feel at one with rF2. Because ARCA has one of the best tyre models out there, full stop. I feel that rF2 is right there with it (and probably as I get deeper into the Sim it's going to actually be ahead). iRacing - there is no tyre wear at all, so I honestly don't get what's so great about that tyre model. I can drive the McLaren MP4-12 for ever on one set of tyres doing pretty much the same times. There has to be drop off. In iR there is none. Please correct me if I'm wrong, but that's what appears to happen when I drive it.

    Any way, back on track, keep up the good work rF2 team:).
     
  12. Marek Lesniak

    Marek Lesniak Car Team Staff Member

    Joined:
    Oct 6, 2010
    Messages:
    1,585
    Likes Received:
    101
    One thing is tyre model itself (the core code simulating tyres, more or less correctly) and other thing - parameters we then provide, to recreate a particular tyre.
    A tyre model might not simulate some physical phenomenons or simulate it in a simple way (due to whatever limitations were the reason). So, on that level some assumptions might be made.

    Parameters of a given tyre are not simple to obtain, especially when several things affect final result we see plotted as cornering forces etc. So, here some more assumptions are likely to be made.

    All of that, gives us a more or less correct simulation of a tyre. Assumptions could be valid. Tyre data also could be more or less correct but when we add that all together, the final result can be noticably different compared to other attempts.

    On a side note - people often say "oh, I love/hate that tyre model because..." without realizing, it might be more about tyre parameters (beeing slightly off) rather than tyre model itself ;)
     
  13. Ivan Baldo

    Ivan Baldo Registered

    Joined:
    Oct 7, 2010
    Messages:
    57
    Likes Received:
    3
    Software setup plays a big role too.
    I seen people fiddle too much with force feedback settings in the wheels control panels as to completely destroy the feel in a given simulator but not on another.
    But also the other extreme, leaving everything at defaults and saying "it stutters" and "I don't feel connected with the car", then you see that they are using much more VRAM than what the videocard has.
    Is funny when you go to someones house and touch a couple of settings here and there and then completely transform their opinion of a given simulator.
    We are lucky, we have more than one good simulator, one cannot say that Assetto Corsa, iRacing or rFactor 2 is bad, all of them are very good, if you feel that one of them is wrong, then look into the settings, try to setup things differently and try again, they are all good and you should feel at home in all of them.
    And another cool thing too: they are changing all the time, sometimes there is a clear improvement, but sometimes there are some step backs while other things improve and then they fix the step back and everything is better, for example iRacing certainly is having some issues with tyre degradation but that started after they improved other aspects of their tyre simulation that broke it, they should fix it in the near future.
    From my experience with friends and their configs, I have a suggestion for ISI: setup everything by default for 1GB of video RAM for the official content with 16 cars, take the most demanding ISI track with the most demanding ISI car and adjust everything to not exceed 896MB of VRAM, and enable the auto quality option aiming for 30FPS; also, disable all driving helps by default, aim for the hardcore simracer otherwise the sim will feel odd with them and give you bad publicity :-/.
    I cannot go to everyone's house's or access remotely to fix their setups :-(.
    BTW, instead of the CTRL+F bar that appears at the top left just show the amount of VRAM used as a number and put it in red when it exceeds the real amount of VRAM - 128Mib, don't add the amount of GART available, you don't want to continuosly read from that memory because in some systems it causes big nasty stutters so its just better to pretend it doesn't exist because really, in some systems it doesn't work as advertised; you can use it from time to time but not very often; maybe even show a message once for the user when the VRAM memory is exhausted and suggest reducing texture quality or object complexity.
    Doing those things shouldn't be too difficult but help a lot for a great experience with rF2 and avoid some bad undeserved publicity.
    Thanks!
     
  14. Andregee

    Andregee Registered

    Joined:
    Jun 23, 2012
    Messages:
    936
    Likes Received:
    394
    about Iracing

    100% agree. I gave up trying to be fast with the Ruf because i never noticed any understeer. The only feeling that i got, was that the car has a very unprecise steering but i was wrong, it was massive understeer, without giving any feedback about it.
     
  15. Marc Collins

    Marc Collins Registered

    Joined:
    Jan 11, 2012
    Messages:
    3,159
    Likes Received:
    162
    Agree about iRacing. Cars change completely from one build to the next. Example: the Radical was horrid from Day 1 and now is one of my favourite sim cars of all time on the latest tire model. But it took until v5 to get it. Some cars on v5 still feel bad. But overall, the v5 tire model cars are excellent, but there will be a v6 I am sure! And the FFB needs some major work in iRacing. rF2 is better for sure in that regard and better or at least equal in physics.
     
  16. Joe

    Joe Registered

    Joined:
    Nov 14, 2013
    Messages:
    372
    Likes Received:
    21
    the FFB better on rF2 is not just better eng physics but also attributed by new tyre model (thermodynamic tyre model).
     
  17. Doc_

    Doc_ Registered

    Joined:
    Jan 14, 2012
    Messages:
    85
    Likes Received:
    0
    hmmmmm......
    I came into the forum looking to find out when ISI were going to fix the tyre issues only to find this is the final tyre model. Most disappointed.
    No matter how much oversteer you dial into a car you just scrub out the fronts in double quick time. You cant balance the car for even tyre wear.
    When checking the inside/mid/outside tyre temps in motec, the mid temps are way too high. You can not get enough pressure out of the tyre to get that temp down. You end up running every car with the minimum pressure all round. Also, the temp gap between inside and outside temps is huge.
    I do think the grip levels are fine, just the physics of operating temps & wear are way off.
    I have now run in 4 championship and every race is the same, people just driving around trying desperately to look after the fronts.
     
  18. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,483
    Likes Received:
    6,656
    You should read the knownissues file.
     
  19. MarcG

    MarcG Registered

    Joined:
    Oct 4, 2010
    Messages:
    6,854
    Likes Received:
    2,235
    where have ISI said that in this thread? FYI the Tire Model is ongoing, like the rest of the game!
     
  20. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
    Incorrect :)

    Just a few examples to prove this...

    A. The readme file in your game install directory mentions that the tyre model is not finished.

    B. When people were discussing the new Indycar, Tim made a statement about core tyre model updates that would be noticeable on ovals.

    C. Tim mention that there is an update coming relatively soon to the tyre model that will improve the GT cars across the board (I think he said this just after B660 came out, but I can't remember for sure).

    Just like weather, damage, the drivetrain model, graphics, sound (hopefully), the overall physics engine etc., (pretty much the entire game) the tyre model, as good as it overall is to drive now ;) , is definitely not finalized and will continue to be even further improved upon.
     
    Last edited by a moderator: Jun 1, 2014

Share This Page