LiDAR circuit models

Discussion in 'General Discussion' started by toodaft, Sep 28, 2011.

Tags:
  1. fanlebowski

    fanlebowski Registered

    Joined:
    Oct 19, 2010
    Messages:
    132
    Likes Received:
    1
    Totally agree. when you make a car with motec (or other) data in the hand, you are immediately stopped in comparison when you want check fast bump/rebound, bumpstops, rideheight and so many others things.

    today, with standard quality track we know in rfactor we can simulate the general car behavior and 50% (or less) of the setup. We have 50% to find in the tracks quality, but also in the physics engine, but this is not the good thread for this part ;)

    this is a huge limit in simracing nowadays. But have we some choices ? if we want some accurates tracks, someone will have to pay :)
     
  2. Alex Sawczuk

    Alex Sawczuk Administrator Staff Member

    Joined:
    Oct 4, 2010
    Messages:
    851
    Likes Received:
    971
    From the pictures I've seen it looks woefully inaccurate.
     
  3. DeDios

    DeDios Registered

    Joined:
    Oct 5, 2010
    Messages:
    298
    Likes Received:
    1
    With all respect about track builders but..well,using laserscanning data, is difficult to see similar discussions about accuracy :D Obviously, you can do some errors, but is really difficult to do BIG errors.
    Would be nice to invent a new laserscanning cheap technology.
     
  4. CdnRacer

    CdnRacer Banned

    Joined:
    Feb 4, 2011
    Messages:
    1,894
    Likes Received:
    31


    Wrong. You obviously can't read.
     
  5. DeDios

    DeDios Registered

    Joined:
    Oct 5, 2010
    Messages:
    298
    Likes Received:
    1
    where am i wrong? explain! :D
    Has F1 2011 Laserscanned tracks? if yes, didn't know.
     
  6. Noel Hibbard

    Noel Hibbard Registered

    Joined:
    Oct 5, 2010
    Messages:
    2,744
    Likes Received:
    40
    Very good point. To add to that, IRL you can learn a track drastically faster than in a sim. There is just no replacement for real life visuals (depth perception and peripheral vision) and gforces. Having a track in a sim that is reasonably close is just as useful for memorizing a layout as one that is very accurate (LS). Trust me, you aren't going to jump from a sim to a real car and drive the track 100% your first lap out simply because you can drive the LS version at 100%. You are still going to drive at 80% for a few laps to get used to the real track.
     
  7. DeDios

    DeDios Registered

    Joined:
    Oct 5, 2010
    Messages:
    298
    Likes Received:
    1
    Yes, i agree. Real life is and..will be (obviously) a different world compared to the simulations. When i drive karts, main difference is about G forces, there are really important for undestand car behaviour.
    Anyway, in a simulation we need to push at maximum level accuracy..about tracks and mods.That's the only way to have most real sensations :)
     
  8. shum94

    shum94 Registered

    Joined:
    Oct 5, 2010
    Messages:
    375
    Likes Received:
    0
    Force Feedback...
     
  9. K Szczech

    K Szczech Registered

    Joined:
    Oct 5, 2010
    Messages:
    1,720
    Likes Received:
    45
    What about it? And what's that have to do with g-forces?
     
  10. shum94

    shum94 Registered

    Joined:
    Oct 5, 2010
    Messages:
    375
    Likes Received:
    0
    ... sensations and forgetting to mention ffb ^^

    As for the thread :

    What matters is fidelity of the track, laserscanned or not. Virtua_LM Mid Ohio is in my opinion great.

    I'm for quality material & textures as well. It add to the immersion and i've seen way too often track release in rFactor looking like the track isnt finished (polished) or worse in beta state with problems with pitboxes or pitlights missing.

    Nowaydays with go pro camera and youtube, it's simplier to check out if the track modeled is + 95 % accurate.
     
  11. DeDios

    DeDios Registered

    Joined:
    Oct 5, 2010
    Messages:
    298
    Likes Received:
    1
    I'm not agree; you can't check track accuracy just using youtube and go pro videos. Only thing is to have reference data, CAD files and other stuff but are not simple at all to have. An another useful thing is to have some feedback from drivers-pilots who did several laps on that track, specially regarding bumps, banking ecc..
    I agree about textures and materials; you can do a marvellous job with 3D but..if you have bad textures your work will not reach the top quality.
     
  12. shum94

    shum94 Registered

    Joined:
    Oct 5, 2010
    Messages:
    375
    Likes Received:
    0
    Check out not help out.
    Once you did the track, youtube is a nice way of checking out if the flow of your track match the real one. Not everybody knows someone who drove the track modeled.

    For exemple Spa GTR2 had the last left before last chicane wrong.

    And let's not forget that real track are creations, and except Turkey GP, every new F1 tracks are not interesting (i hate flat tracks).

    So maybe someday a good track created in a sim will be so good that some people will be interested to recreate it in real.

    PS : I disagree, not i'm disagree ;)
     
  13. DeDios

    DeDios Registered

    Joined:
    Oct 5, 2010
    Messages:
    298
    Likes Received:
    1
    lol sorry for my "inglisc"! :D
    youtube videos are imho too much influenced by cam specs :)
    Yes, you can see if track is "right" but..YT videos are not too much useful for check accuracy. Anyway is just my idea, i'm not a track builder, would be nice to read some "expert" opinion.
     
  14. Revvin

    Revvin Registered

    Joined:
    Oct 5, 2010
    Messages:
    40
    Likes Received:
    0
    From an onboard camera the perspectives and therefore your interpretation of the radius of a corner etc would differ greatly depending on where and how high the camera is mounted, information you wouldn't get from a YouTube video. Fortunately we know from rFactor 1 that adding laser scanned tracks is possible so perhaps if there is a cheaper method found in future we can have them.
     
  15. Old Hat

    Old Hat Registered

    Joined:
    Feb 22, 2011
    Messages:
    163
    Likes Received:
    8
    Well, even the handful of so-called 'aliens', who're supposed to be faster after a few laps than anyone else, struggle in iRacing against fast drivers who've put in lots of laps during the week (an iRacing series spends a week at one track). It's not as though the tracks are new to anyone either. And even aliens admit to infeasibly large amounts of practice in the past compared to a RL driver to get fast in the first place. Sims aren't RL.

    The Virtua_LM Mid Ohio looks the same as iRacing on Youtube but isn't to drive on.

    Does depend on the track (and car) somewhat though. Also, I'm trying to factor in the effect of iRacing's slippery physics here. The V8 Supercar around Mid Ohio is bonkers- touch the throttle at the wrong time and it squirts you off. Seen it happen to the fastest, who's a pro in RL in that car. You won't learn that combo and compete with the fastest after a few laps no matter who you are.

    Regarding wheel FFB: probably the fastest guy in LFS (with over a hundred WRs) apparently didn't have FFB. As I said, I tried it off, and didn't seem to make any difference to speed - once I'd finished fish-tailing out of the pits that is. LOL
     
  16. Hofstee

    Hofstee Registered

    Joined:
    Oct 15, 2011
    Messages:
    13
    Likes Received:
    0
    Even so, GoPro's camera, while being good for general flow, isn't the greatest thing to check for accurate modeling. GoPro being a wide angle camera, has a bit of a distortion effect.
     
  17. 88mphTim

    88mphTim racesimcentral.net

    Joined:
    Sep 23, 2010
    Messages:
    10,840
    Likes Received:
    314
    I wonder if Shum is saying that it's what you can tell from GoPro cameras which is what you'd need to tell?

    If you are a sim racer, never going to the real track, the importance could be that it seems to be accurate.
    If you are a sim racer, testing for the real track, the importance could be that it flows the way the real track does.

    What I mean by that is, if you can go to a track and be even 50% sure of the flow of the track, 90% sure of which direction the next turn goes and even 20% sure of what gear you need to be in, you have a huge advantage over someone who hasn't even run a simulator.
     
  18. 88mphTim

    88mphTim racesimcentral.net

    Joined:
    Sep 23, 2010
    Messages:
    10,840
    Likes Received:
    314
    Plus, every second spent at a real track is different. Wind, clouds (track temp), etc all change the track from lap to lap, so does rubber laid down. Even though rF2 will have those things, even if you had a laser scanned track, the chances of you ever encountering the exact same conditions in-sim and in real life are... probably difficult to work out. There must be an infinite number of combinations of conditions.

    Then you have to consider that every car is different, even if they come off the same production line. The chances of getting the exact same feel in-sim and in real life also is pretty tough.

    No no matter what, you are still going to have to adapt from software to real life. No matter what. So does that little bit of accuracy actually help you or hurt you? Too much confidence in your in-head track map could be a dangerous thing...
     
  19. DeDios

    DeDios Registered

    Joined:
    Oct 5, 2010
    Messages:
    298
    Likes Received:
    1
    The software can be useful for to learn a new track (these are words from several pro drivers) but..surely is not a reality substitute. Can help you a lot but..you can't learn all secrets using just one monitor (i'm talking about our "classic" sim rigs..). Anyway the final objective is to have the most accurate software, the most accurate tracks and cars to trying to simulate reality. This is the simulation, imho.
     
  20. 88mphTim

    88mphTim racesimcentral.net

    Joined:
    Sep 23, 2010
    Messages:
    10,840
    Likes Received:
    314
    Oh, absolutely, that is simulation. But accuracy is relative in this case, as funny as it sounds. Lots of real racing drivers credit tracks that I certainly know aren't accurate to that level.

    Also, surfaces settle, sink, change, get resurfaced. How accurate is the track you laser scanned after those things? Trees grow, buildings get demolished, built, signs get changed, moved, etc. So how accurate is the track side after you laser scanned it?

    That accuracy, is relative.
     

Share This Page