Searching Ideas for Standard of Car Simulation

Discussion in 'General Discussion' started by mantasisg, Aug 21, 2019.

  1. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    In other thread I speculated, that in five upcoming years we will have some approach for objectively defining a standard for car simulation. Can we do it ? Or in other words how to set a bar when simulator is a simulator ?

    I mean obviously: "FFB feels great" (not objective), you somehow "feeling weight transfer"(not objective), "car is believable"(not objective), "realistic laptime"(too objective, I mean it is obvious minimum), "realistic G's"(same as for laptimes), "ability to beat Hamiltons laptime"(no comment) and so on... Lets keep those away, and perhaps attempt to be a bit more technical.

    Ofcourse majority of us don't have qualification to do this, and this is obviously big task. What I actually expect, is simple ideas with good logic. Well getting a bit more technical would also be nice, perhaps even inevitable if some guy really knows the stuff.
     
    Last edited: Aug 21, 2019
  2. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    To begin with I am sharing this little idea. Choosing a car, or a couple of cars which license would be easily obtainable (would be available in various simulators), as well as data (as much as possible). Simulating exactly same tires. Have tools to be able to evaluate how virtual physics match the real life data. Use some kind of testing pad that would be same in all simulators. Show comparison results in some public platform, and make them easily readable so most guys would be able to understand it correctly.

    The evaluation itself is the most important part. Tests for some particular parameters has to be designed and standartised. As example the most basic ones: base friction test, kinetic friction test. Some which bugs my mind a lot: steering forces during understeer, steering forces during oversteer, effects of scrub radius, complete slip curves (for angle and for force)...

    I think it would be fun game, to think of what some tests could be like.
     
    Last edited: Aug 21, 2019
  3. xerofer

    xerofer Registered

    Joined:
    Apr 25, 2019
    Messages:
    64
    Likes Received:
    25
    The first thing obviously would be to define in what areas current sims are not up to par with (your) simulation standards.
    And what a simulation standard already is.
    As to your 2nd input, don’t you think devs are already doing what you suggested? Only behind close doors, for obvious reasons.

    Go on, your turn.
     
  4. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    Thats my point. There is no true simulation standard, it does not exist. Only your or mine standard exist.

    I'm sure devs does as much realism check as they can and care to. My idea is obviously something that could be independent and public, and to tie all sims together by simulating absolutely same stuff. Absolutely dead same car, with totally same tires, at totally same place, in completely same testing with completely same conditions.
     
  5. xerofer

    xerofer Registered

    Joined:
    Apr 25, 2019
    Messages:
    64
    Likes Received:
    25
    And what if real live data capture doesn’t give the data you are yearning for? Tire parameters for example?

    And above all, it takes money to do what you described. Did you open a patreon page already?

    I don’t think what you described will ever come into fruition.
     
  6. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    Maybe it won't of course it is very loose. I did make this thread title a bit misleading, the goal is good ideas how to do that. Just brain storming good ideas is what I want to do here. And discussing them, improving them. Perhaps coming up with something... In my mind it is interesting, surely practicality is important for it to really happen, but lets not make it a priority at the beginning. I just randomly guessed that some proper ideas could get collected in five years, maybe this thread will just die now, but I wish it would get updated frequently and getting more clear direction every time. Shouldn't be more difficult than to inhabit Mars.

    The realization of most stuff probably wouldn't happen, but even if it could be possible to get at least few things approached lightly it would be nice. Because now there is literally nothing for that, all just personal opinions and rarely some proof which is lost soon in sea of information. Perhaps some stuff could be generalized for all cars with margins considered. It is all very complex, there is no way to look at complete picture without breaking it all into details. Maybe it would be better not to even create whole simulation standard, but many standards for various vehicle handling features.

    For example should a vehicle roll over ? Not an essential simulation feature, but important dynamic anyway, and still a thing that counts as something to simulate.

    Having all real data and this whole standard thing is at this point just a fantasy thing, but it is not like practically impossible, and why not just think about it ?

    A patreon to make these things real ? Thats an idea, but I personally would not take on it haha
     
  7. xerofer

    xerofer Registered

    Joined:
    Apr 25, 2019
    Messages:
    64
    Likes Received:
    25
    So yes, I got what you are trying to do...

    But first of all, what are your complaints on the existing developer studios not doing?
     
  8. xerofer

    xerofer Registered

    Joined:
    Apr 25, 2019
    Messages:
    64
    Likes Received:
    25
    ...and what are your suggestions getting rid of them...
    :D
    Show us!
     
  9. xerofer

    xerofer Registered

    Joined:
    Apr 25, 2019
    Messages:
    64
    Likes Received:
    25
  10. Emery

    Emery Registered

    Joined:
    Oct 24, 2010
    Messages:
    3,035
    Likes Received:
    1,654
    When it comes to the technical accuracy of a simulation, one needs to figure out an acceptable margin of error. For instance, simulation of McPherson struts with double-A arms is reasonable, but substituting double-A arms for live axle is not (yeah, Simbin did that with Race07's Retro Pack and they're not the only company that has committed simulation murder).

    The simulation needs to be transparent enough that such things can be independently confirmed (yet without giving away trade details) rather than hidden behind encryption. Maybe a viewport showing a diagrammatic representation of the suspension in action?

    Standards for simulation, however, totally rely on setting up FFB so each player with the same wheel is getting the same experience would be a huge leap forward. Proper FFB is way too open to subjective interpretation. As it stands now, players can completely screw up a title's FFB just by messing in their wheel's control panel even if the title ships with a proper wheel configuration.
     
    mantasisg likes this.
  11. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    As much as i like diversity in simulator but a focused sim on one car series will be better.(IMO)
    Just compare Gt3 cars from RF2 and ACC the differences are astounding.
    Not a fanboy but ACC just ruined every GT3 cars in every sim ever hence setting a new standard for me.
    Now imagine if kunos got same license for upcoming hypercar series or rally(old cars since new cars are under trash arcade license) !
    The quality,focus of the product will be amazing to watch.
     
    Dookie, Mibrandt, xerofer and 2 others like this.
  12. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    I don't want to get into this global simracing wishlist. I am also possibly not right about everything. I am often bugged by how forcefedback responds to oversteer, sometimes to me it is weirdly strong too soon. I am often bugged by tires feeling too precise, sometimes feels like I am too easily pointing car where I want. Sometimes feels like cars don't bounce of kerbs strongly enough. Sometimes it feels like cars aren't prone to roll over. Sometimes it feels like rain physics are either too hard or too easy, sometimes at the same time for different dynamcs. Sometimes it feels like grip is not sorted correctly sliding vs base vs speed of change from base to sliding.... a lot various stuff, I do not name specific sims now.

    @Emery Yes, exactly minimum FFB sliders should be one of the basic standard features for real simulation. Now we sometimes have all sort of expectations for feel, and are able to butcher a lot of stuff and then just live with placebo that something was improved.

    @Slip_Angel I agree to me ACC gt3 are also better, feels more subtle maybe, mostly for slipping behaviour. But I think it could possibly be better in rF2.
     
  13. xerofer

    xerofer Registered

    Joined:
    Apr 25, 2019
    Messages:
    64
    Likes Received:
    25
    Now imagine Kunos doing THE pre war series simulation...
    Very much needed, indeed.
     
  14. xerofer

    xerofer Registered

    Joined:
    Apr 25, 2019
    Messages:
    64
    Likes Received:
    25
    Seems like your into the prewar Grand Prix, too.
    :)
     
  15. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    I do too that makes it 2 people notice same things in same sim.
    Don't know about rain though as i don't play rain.
     
  16. xerofer

    xerofer Registered

    Joined:
    Apr 25, 2019
    Messages:
    64
    Likes Received:
    25
    Get ACC to experience rain, really.
     
  17. Rui Santos

    Rui Santos Registered

    Joined:
    Jan 8, 2012
    Messages:
    1,081
    Likes Received:
    1,206
    Let's end this chit chat of the newbies and leave S397 guys work in peace...
     
    David Kolody likes this.
  18. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    Yep so far even rain physics are also the best for me.
    Generally ACC is vehicle dynamics triumph for me it has spoiled me.
    Although RF2 F1 cars are THE BEST now period.
     
  19. Slip_Angel

    Slip_Angel Registered

    Joined:
    Jun 15, 2019
    Messages:
    588
    Likes Received:
    297
    Changing topics, You once said that you are getting weird tyre temp like outside or midside hotter than inside tyre patch ? correct.
    I saw some thermal onboards and if your ouside tyres both front and rear have more temperature while cornering at outside and mid part of tyres than inside then that is indeed how it is IRL.(atleast that's what i observed)
     
    Rui Santos likes this.
  20. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    I renamed the thread, because the name was way too cringe worthy.

    @xerofer Yes I am fscinated by those cars, even have W125 mod in progress. But I am not sure what this has to do do with our topic :D I learned quite a bit from that car though.

    @Rui Santos I will not argue about chit chat of the newbies part..., but how this breaks peace of S397 ? Some of us are just interested in some general simulation stuff, and the topic is clear - ways how to evaluate good simulation. rF2 forums is the only place that I know where could be at least some people at least mildly interested in simulation. If some people hates to think bout it, it is their problem. But that of course makes life less complicated :D Suggest me better place if you think this should not belong here.
     
    Last edited: Aug 21, 2019
    Rui Santos likes this.

Share This Page