1. #1

    About OPE on curved tracks ...

    So I've got a problem I never came around on evo.
    In one of my tracks called "Supraconductor" I used area effect forces on the bikes tires to simulate an magnetic effect.
    I'm fiddling around with that again and like previous times I can't make it work properly on a curved track.

    I've the position down nice and simple with ope's and position x/y/z feed on them. Because the forces need to point in a fixed direction relative to the bike (always down / in direction of the wheels - even if the wheels point upwards) I can use one single angular feed on the ope to achieve that.

    Thats how i make it work on a straight line thats aligned to one of the global axis.

    I've alwasy tried to make it work on a curved track simply by taking position x/y/z feed and angular x/y/z to ope but as soon as I drive the curve all the directions seem to get mixed up and my area effect begins to wobble around in seemingly random directions. I've tried for hours to change the angular axis, local/global ect.
    But somehow there's something in the logic about how that works I can't wrap my head around.

    So if there's somebody out there who has an idea or a solution I would appreciate it. Thanks.
    Share this post

  2. #2
    Lukeyy19's Avatar Member
    Join Date
    Mar 2014
    Location
    UK
    Posts
    449
    Any way you could take a few pictures of all your logic stuff and the issue or even a video, even if it's just recorded on your phone or something? It's going to be hard to figure out what might be going wrong without seeing it.
    Share this post

  3. #3
    Wouldn't you also just have to use one angular feed for the curved lines instead of 3 different ones(x/y/z)??
    Share this post

  4. #4
    Okay let's try this. I'm not aware how to switch my ingame text to english but I'll guess there won't be a lot of translation neccessary.

    So this is my straight line setup (Winkel means Angular):


    The following two pictures show the forces apply and point in the wanted direction even when jumping:



    Now I try the same setup on a curved track and upon driving the curve the forces point in various directions depending on where i am in the curve and how my bike moves:


    Then I try to add the other angular informations to the ope and hope that it would work:


    But the result isn't satisfing:





    I should add that the positions and angels are not fetched from the wheels but from the attached bike parts - otherwise the angels whould constantly spin as the tires spin.
    I aslo tried to set the angular information xyz in diffrent ways to bank pitch and roll as well as global and local settings but it's always screwed up.
    Share this post

  5. #5
    Lukeyy19's Avatar Member
    Join Date
    Mar 2014
    Location
    UK
    Posts
    449
    If they are both always pointing down then is there really a need to be moving each one independently? Have you tried setting them as a glue group and tying the glue group to the main body of the bike with just one OPE, you say you've tried mixing around the inputs from pitch, roll, yaw etc, which would have been my first thought, but maybe as the track curves, the parts of the bike that you are OPEing to are flexing weirdly, causing their data values to be strange, but the body of the bike may not suffer this issue.

    As another diagnostics test, move your Object Position Data Sources into view of the game camera on a curve and try testing the track with "show editor icons" enabled, this may give you a better idea of which data source is causing issues.
    Share this post

  6. #6
    best way probably is

    make a glue group with the 2 forces and ope those to the bike center
    the hard part is making the glue group and using the right angles(you'll need to retry a few times)

    use snapping
    the most important thing is changing the orientation(you'll need to mess with this till you find it out) of what your glueing before you glue it

    if you are using a glue joint the orientation of the joint matters before you glue and you'll need to rotate this before you glue around one of its axis
    if this orientation isn't right it wont work and it will give the issues you have right now

    also its a bad idea to add a value to global angles because this will cause stuff to break like that too
    Share this post

  7. #7
    The glue group just wobbles around like the area effects themself if i drive through a curve and/or do wheelies/jumps. I tried each orientation for the glue joint. It always beheves strange.
    Share this post

  8. #8
    Aha! I finally found a solution. It's rather simple an not very elegant but I think it will do. What I've done is to use simple strong point to point physics joints. Works perfectly. I'm just too obsessed with ope's I think. Thanks guys, you made me think in the right direction. ^^
    Share this post