1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Discussion in 'Tracks' started by luchian, Aug 19, 2017.

W.I.P. Motorpark, Romania 0.1 alpha

Very early version of Motorpark circuit situated in Adancata, Romania

  1. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    Very good videos guys, will be useful for driving comparison. Now let's work on that layout :).
    PS: nice skills in the Alfa, for an (I assume) occasional track driver.
     
  2. dragos simion

    dragos simion New Member

    Joined:
    Dec 11, 2017
    Likes Received:
    15
    no he is a profesional kz driver
     
    luchian likes this.
  3. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    Maybe this doesn't look like much, but it's an important piece of the puzzle. After a few conversions, I've managed to import the DWG into Blender, and the scale seems spot on. The main loop (orage below) reports 3287 m. The additional loop adds another 870 m, and taking out the small part between them, total track is around 4024 m. Which to me sounds pretty close of the real thing (with some neglectable measurement error - I'll do a more precise one when finishing the layout). Happy user here :).

    DXF_in_Blender.jpg
     
    Gabriel44 likes this.
  4. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    So, I (we) have a small dilemma. I have compared the DWG plant with the layout made by RTB from google maps. And the differences are rather small. This raises a couple of observations/questions:

    1/ I am impressed with the accuracy only based on google data
    2/ It seems the error is not constant across the track (like let's say a 7% deviation or something). It's more subtle, on some of the turns. And makes me wonder which one is actually correct ? The plan ? Or the google data, because maybe doing works introduced some deviation from plan ?
    2A/ is there any civil engineer reading this, that could confirm what is the expected deviation from plan, on works this size ?
    3/ from the people that tried the real track and the current AC version of Motorpark, can you guys tell if some turns seem off ? (like too tight, or not too tight enough) (not speaking about elevations, those will be easy to correct).
    4/ @modelers around: should we decide to go with the plan, which would you say is the simpler option ? To redo everything from scratch or to try and adapt the road to conform the plan ? :D

    Oh, here is an HD screen (click for real size), to have an idea. Green = layout from plan. Black (dense mesh) = road from RTB.
    plan_compare.png
     
  5. MLT

    MLT Member

    Joined:
    Nov 2, 2017
    Likes Received:
    88
    Most of the time first impressions are often translated to the models, which captures the essence, but not the real details. I usually prefer to redo everything from scratch, once I've got acquainted with the model ( the curves, slope, etc. ).

    So I usually first start by dragging, sizing and doing all the do nots. When I think I've nailed it ( shape, look ), I restart from scratch keeping in mind what not to do in comparison to the previous versions.
    This usually ends up with a nice clean model (track).

    For example I redid the Fiorano track layouts many many times, as in each iteration I would gather more information that would lead me to have to rethink and remodel certain parts, and hence at the end I would make the final version from scratch to keep everything as clean as possible.

    EDIT: well probably the simplest thing would to keep on working on what you have, but redoing it will definitely make the track better. It does become a pain in the ass, get yourself a buzz if it becomes unbearable hehe.
     
    Last edited: Dec 16, 2017
    luchian likes this.
  6. Willy Wale

    Willy Wale Member

    Joined:
    Jan 12, 2017
    Likes Received:
    74
    Location:
    UK
    For me, I would go with the plan. When I overlayed the Google data with the Orsnance Survey master map data for my home town the google was almost a track (road) width out on the eastern side if I lined up the western side with the north-south scale roughly correct.
     
    luchian likes this.
  7. Mr Whippy

    Mr Whippy Active Member

    Joined:
    Jan 18, 2016
    Likes Received:
    126
    Apart from the T1 deviation, the rest of the deviation from plan is probably from Google not using orthophotos.
    So where the track runs up/down hill in z, the track in the aerial photos moves in x/y.

    I'd go with the plans, and projected straight down onto the z terrain.
     
    Willy Wale and luchian like this.
  8. dragos simion

    dragos simion New Member

    Joined:
    Dec 11, 2017
    Likes Received:
    15
    So if you ask me, the plan is more acurate. For example in the google maps version turn 1 felt too forgiving and shoud be more thight like it is in the plan, also the first hairpin felt the same.So go for the plans, in real life the track is exactly like in them.
     
    luchian likes this.
  9. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    The plan it is then :)

    Sent from my phone using Tapatalk
     
    Willy Wale likes this.
  10. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    Works advancing nicely. The main loop is aligned, the additional loop needs some work still.
    But loving to work on this thing and very curious to hear some feedback on how the new layout feels compared to real.
    Once we'll nail that, I'll also add a separate physics layer to give it some character.
    No screenshot this time, but it will come in the following days.
     
    MLT likes this.
  11. dragos simion

    dragos simion New Member

    Joined:
    Dec 11, 2017
    Likes Received:
    15
    Can't wait !
     
  12. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    I am rebuilding everything from scratch, to get a clean looking mesh, aligned with the DWG plan. In the mean time, I made a quick (dirty) test, aligning the mesh of the RTB version to the plan. I wanted to see how the turns will look in-game. Since it's made, you can try it out also. HERE.

    Disclaimer: it's not at all representative, since it's just pulling/pushing meshes to fit the plan :).
     
    dragos simion likes this.
  13. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    Just a sample photo: the new road (visual) follows the DWG plan at millimeter.
    I need to play with the heights now. Once the road is set, should go faster.

    adancata.png
     
    dragos simion and Gabriel44 like this.
  14. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    Little bit more progress. Manually (#slooow #boooring) creating height vectors based on the measurements on the plan. It'll get there :). But enough for the day.

    2017-12-24_002605.png
     
    dragos simion likes this.
  15. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    Diferences are rather small, so not even sure we'll feel this when driving, but has to be done. At least for OCD if not something else :D. And with that, wish you a Merry Christmas. I'll go get ready for party :whistle:

    2017-12-24_170057.png
     
  16. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    After all the eating and drinking, what better way to spend some time than math :D

    cote_H.png
     
  17. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    Some progress report:
    • road following the DWG plan on XY-axis : DONE
    • road (center line) following DWG plan on Z-axis: DONE
    • road camber : in progress, 20%-ish DONE.
    So, had measurements for internal and external points. I got to calculate an angle based on these and road width (11.2 m, except for the main straight). Now I am introducing these, 1 by 1, into the model. Another long/boring job, but at least it will be the last one for the road. After this step, we should have an accurate representation of the asphalt, according to the measurements I could find on the DWG. Hopefully, they will correspond to the reality on the circuit :lol:.

    It's somewhat ironic:
    We can agree it's a flat track (when you look at it from a plane :p). But since there were small variations, I had to calculate and introduce them into the model. In the end, the amount of work involved, makes no difference if it would've been a very hilly track, or a flat one. The only difference is in the numbers one is injecting, not in the number of steps or the number of hours needed.
    2017-12-28_183048.png
     
    Willy Wale, MLT and Pixelchaser like this.
  18. Pixelchaser

    Pixelchaser Well-Known Member ACM Club Driver

    Joined:
    Jan 4, 2015
    Likes Received:
    848
    good job, nothings actually flat, unless its in Switzerland. :lol:
     
    luchian likes this.
  19. luchian

    luchian Administrator ACM Racing Team ACM Club Driver

    Joined:
    Jun 3, 2014
    Likes Received:
    638
    Any of the guys watching this happens to have some photos of the entry from new loop to main loop ? (see below). On the plan both left and right are very rounded. In the videos looks more like on my proposal below. Some photo/video would be really helpful.

    Maan, that drone would be useful :D

    2017-12-29_003849.png
     
  20. Gabriel44

    Gabriel44 New Member

    Joined:
    Dec 2, 2017
    Likes Received:
    6