Track Simulation

Discussion in 'General Discussion' started by bravotangosix, Oct 29, 2018.

  1. Filip

    Filip Registered

    Joined:
    Feb 9, 2017
    Messages:
    1,208
    Likes Received:
    937
    ...for those that intent to drive the real track or read real drivers expereinces.
     
  2. bravotangosix

    bravotangosix Registered

    Joined:
    Nov 21, 2015
    Messages:
    265
    Likes Received:
    277
    If your not trying to simulate racing then why use a simulator?
     
    mantasisg likes this.
  3. Comante

    Comante Registered

    Joined:
    Nov 20, 2013
    Messages:
    1,694
    Likes Received:
    1,218
    I think there is a substantial difference between realism in the car compartment and in the track compartment.
    The first is strictly related to laws of phisycs, so, you want to have them right, because physics laws can't be modded or changed in real life.
    For the track, despite the level of accuracy in the look and on the surfaces, you know that at least every player racing on the track is facing the same conditions. You can reference yourself with AI or Humans playing on the same field. Comparisons with reality will never be accurate, and they matter for a very tiny number of people. the final consideration, is that supporting one idea doesn't mean disregarding the other. I don't think it's true that there are people that DON'T WANT scanned tracks, is just that they don't care enough and play whatever track they enjoy, realistic or fantasy.
    I'm of the idea that every track should have the detail depth of Sebring ,at least for the track surface, in all honesty I don't care how that could be accomplished (scanning the real track if present or sampling different surfaces and baking an hypothetical road surface) because I agree that what make me return to RF2 every day is FFB. Once tracks will have such depth of FFB details, how they have been obtained would matter even less for even more people.
     
  4. bravotangosix

    bravotangosix Registered

    Joined:
    Nov 21, 2015
    Messages:
    265
    Likes Received:
    277
    @Comante Is it not true that the FFB comes from the vehicle traversing the surface of the track and then how the cars physics interpret that undulation?
     
  5. Filip

    Filip Registered

    Joined:
    Feb 9, 2017
    Messages:
    1,208
    Likes Received:
    937
    :rolleyes:
    If bump on real track is 62m before apex and bump on modded track is 34m before apex what does it have to do with "simulating racing"?
     
  6. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    Well Sebring is known to be very very bumpy track, not something that is usual. I am also fascinated how many refers to FFB as some bumps machine, road texture scanner, while it is primarily a translator of car physics and there is the main beauty.

    IMO simracers should try to rely less on some pseudo-principles and untrue ideas, because simracing can become cliche. But we can have nice discussions.

    @Filip 28 meters is crazy distance. One second to travel at 100km/h. IMO it is more than possible to place the bump more accurately, assuming that bump is outstanding and there is a reference..
     
    Last edited: Oct 29, 2018
    vittorio and bravotangosix like this.
  7. bravotangosix

    bravotangosix Registered

    Joined:
    Nov 21, 2015
    Messages:
    265
    Likes Received:
    277
    Racing is cars on a track. A racing simulator is an accurate replication of cars on a track. If you only accurately simulate the car then what you have is a car simulator not a racing simulator. RF2 is a racing simulator. :rolleyes:
     
    vittorio likes this.
  8. Emery

    Emery Registered

    Joined:
    Oct 24, 2010
    Messages:
    3,035
    Likes Received:
    1,654
    I like experiencing historic tracks, to increase my appreciation of what was experienced in other times. Even if the track surface can be laserscanned today, it will only be a pale representation of what our heroes faced in the past because the ground shifts and pavement degrades. We *have* to settle for non-laser scanned in these circumstances!

    Full devotion to laser-scanning leads to a static picture that often is no longer true. iRacing's Lime Rock Park, for example, is horridly out of date and doesn't even include the layouts currently used most often, so there's no way people are using it to train for real-life racing. In this case, the real-life racer is better off using rF2's non-laser-scanned Lime Rock!
     
    David Kolody likes this.
  9. Filip

    Filip Registered

    Joined:
    Feb 9, 2017
    Messages:
    1,208
    Likes Received:
    937
    But it doesn't matter be it 3m or 30m why should the bumps be at exact place like on real track if I'm not practicing for a track day and what bump position has to do with racing simulation accuracy? If you mean racing lines and lap times will be different well of course it is not exactly 100% the same track and we are all aware of that but question is do we need/care about that?
    You wont be able to make perfect comparison video of a lap around modded Nordschleife with real lap but it doesn't diminish racing simulation on that track lets call it Nortschleuffe if it is easier for purists to accept.
     
  10. vittorio

    vittorio Registered

    Joined:
    Jan 11, 2012
    Messages:
    1,118
    Likes Received:
    540
    Laser scanned tracks for me are magic, because it is like I get that track most realistic in my living room. Same for realistic modeled and simulated cars.
     
    Adrianstealth and bravotangosix like this.
  11. bravotangosix

    bravotangosix Registered

    Joined:
    Nov 21, 2015
    Messages:
    265
    Likes Received:
    277
    Of course it's out of the question to expect tracks long gone to be accurately reproduced to a laser scanned level but that doesn't mean they shouldn't do what they can to try. And that's what has been done. That's perfectly fine. But should we only replicate the latest model of cars since the others are out of date or old? For me this goes hand in hand with tracks. What Iracing have is a capture of Lime rock at a given time. Same as S397 Sebring. That's priceless. Those tracks are a picture in time. All tracks move and change. Personally I'd love to have a scan of Historic Spa but it's not there as no one did it ( never mind that the tech wasn't available). But would you not prefer to put a classic car on a classic track? I would
     
  12. fsuarez79

    fsuarez79 Registered

    Joined:
    Nov 23, 2016
    Messages:
    544
    Likes Received:
    582
    Personally, it's not that I don't care. If I was able to choose between a laser vs non-laser track, I'd always choose the laser scanned, but it's about realistic expectations.
    We can't expect to have 100% laser scanned tracks considering costs, the fact that we have Historic content and I'd say around 90% of the tracks out there are mods.
    So if the question then becomes non-laser vs no track at all, to me it's a no brainer.
     
    bravotangosix likes this.
  13. Filip

    Filip Registered

    Joined:
    Feb 9, 2017
    Messages:
    1,208
    Likes Received:
    937
    Ok would it be acceptable for you if non-laser scanned have fantasy names: Lumerick Park, Seypung, Bazooka... ?
     
    Mankiewicz likes this.
  14. bravotangosix

    bravotangosix Registered

    Joined:
    Nov 21, 2015
    Messages:
    265
    Likes Received:
    277
    LOL I'd love to race Bazooka, it sounds awesome. The debate for me is about simulating the complete package otherwise it's only half the story.
     
  15. patchedupdemon

    patchedupdemon Registered

    Joined:
    Jan 3, 2017
    Messages:
    1,602
    Likes Received:
    1,151
    What about laser scanned cars, is this not an important aspect of a simulation.

    Thoughts
     
    lagg likes this.
  16. KittX

    KittX Registered

    Joined:
    Jan 11, 2012
    Messages:
    147
    Likes Received:
    62
    mmm not really, maybe only as a references to build upon to.
    You don't have real-time aerodynamics modelling based on actual car shape, you don't have airflow modelling too. Even flight sims don't do that, or maybe very very simplified, because such calculations aren't the tasks current computers capable to do realtime, and have a room for calculating everything else.
    Looks-wise, well I'm not too much for half-million polygons on the outside model, like every panel gap or nut or washer being exact the same size or curvature as in a real car, because most of your time you spend in a cockpit. Yes the models should look sensibly accurate and not visually off from the real counterpart (which imo is the problem with ISI Cobra, it's simply too "square" especially the front intake. But again we don't know which blueprints were used and of which generation/body style). On the other hand I'd love to see more fine details in the cockpits, and not just modelling wise, but some random dirt, dust or cracks, or even fingerprints in the texture. Some current gen sims and simcades have the cockpits IMO too clean and shiny.
     
    Last edited: Oct 29, 2018
    Mankiewicz likes this.
  17. vittorio

    vittorio Registered

    Joined:
    Jan 11, 2012
    Messages:
    1,118
    Likes Received:
    540
    All my friends I talk about racing and simracing (sim = simulation: because of realism) have driven the Nordschleife multiple times.
    They often visit me to drive Nordschleife (ATM laser scanned AC conversion). They come because of realism! If I tell them we call it Nortschleuffe because the simulated track is completely off (every corner is off and every bump is artificial) they would only come to play a bit and drink beer, we could even play something else like Monopoly.

    The magic about a laser scanned Nordschleife is that it is the best digital representation possible with consumer hardware today. For me that is magic. Similar for laser scanned Sebring. Although I will never have the chance to race it in real life, but I know its is the best digital representation possible.
     
    mantasisg, bravotangosix and KittX like this.
  18. bravotangosix

    bravotangosix Registered

    Joined:
    Nov 21, 2015
    Messages:
    265
    Likes Received:
    277
    This.
     
  19. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    To put it short... Why not just try to do your best anytime you do something ? You can make unscanned track better than someone would do same track with pointcloud if you'll really work your a** off. It is better to save time though. For historic tracks you may never get it 1:1, but you can achieve great results if you get actual elevation data, get some footage to understand some specifics like cambers, widths, kerbs, runoffs, macro undulations, significant bumps.... The less data you'' pull off the less you'll be tied with accuracy. As a track creator you have to be the most passionate person about the track for a time period while you work on it, so you'll do best possible job. It is better to work three months and do as accurate as possible to you version, than two months and have soulless track you don't want to drive yourself. It is better to invest a lot to accuracy, because working on all the rest makeup and beauty stuff is far more time demanding, though easier ( unless you had plenty of helping data, such as terrain elevation point cloud).

    I have worked with LiDAR data for goodwood, that data seems to have picked up even macro undulations, or even got correct camber at lavants curve (a slightly negative camber)... I continued learning about track from sim and from real life videos too, I spot bumps in sim and then IRL videos, and then vice versa...

    Some interesting details how to "see" macro undulations/bumps, you actually can see them IRL.

    When it is hot it is shimmering, looks like it is water there:
    [​IMG]

    When it is dark, headlight beams can create shadows, in this case it is very visible because there are rather significant macro undulations (not bumps), and the lavant straight is going uphill:
    [​IMG]

    Vehicle simulation does not go without track simulation, all the little details are the ones which gives most emotions and makes good stuff great stuff.
     
    Last edited: Oct 29, 2018
    vittorio, bravotangosix and KittX like this.
  20. KittX

    KittX Registered

    Joined:
    Jan 11, 2012
    Messages:
    147
    Likes Received:
    62
    Sorry I'm sometimes nerding a lot and tend to get very deep into nuances, so there's food for thought came after reading your post:
    So you took Nords and Sebring as examples, and they're particularily interesting in that topic because those roads aren't only used by race cars. First of all Nords is being used and abused a lot more than almost any other track almost on a daily basis, plus some buses and trucks go there ocassionally. Sebring is still being used as an airport if I'm not mistaken, so those tracks have extra forces applied to surface which tends to more agressive changing of track surfaces/bumps/etc. Do you believe in that case those tracks should ideally be re-scanned annually or like each 3-5 years? Because apart from being all the magic and joy the laser scan adds to the reference experience. And if the track is, say, not laserscanned, but very accurate layout, but the surface/bumps are more, say, "schematic", when you're going to the real track you don't take that part of experience to you on track, you're basically open for this new knowledge. But if the track has some pre-made bumps which you've learned in your rig a lot, like you now have additional habits, like lift off the throttle, or change your line a bit, if that differs from the real track, won't you think you'll subconsciously use those habits on a real track too, which at this point won't be valid anymore?
    Sorry there might be language barrier, I hope you get what I'm trying to say.
     

Share This Page