Problem with blast effect

DizkoDizko Member Posts: 498
edited November -1 in Working with GS (Mac)
Hello!

So I used more or less the exact rule set from TSB's blast tutorial with some minor value tweaks to create a blast effect. 90% of the time it works perfectly. However I noticed a problem and I am at a loss on to why it's occurring and how to fix.

Essentially the problem is, if the actor to be blasted is moving quickly on the X axis, the blast effect goes haywire. He'll go straight up in the air instead of based on his angle to the blast actor. However in the Y direction, everything seems to work correctly.

This an example of it working the way it should, just to get an idea of what I'm doing.

EDIT: The levels are also twice the normal height (landscape) so... 640 high (not sure if that makes a difference or not.)



Any help would be thanked with many praises to your awesomeness.

Comments

  • DizkoDizko Member Posts: 498
    Here is an illustration of what I mean.

    image
  • tenrdrmertenrdrmer Member, Sous Chef, Senior Sous-Chef Posts: 9,934
    It looks like your actor is getting to a position directly over the blast before all of the rules and calulations run for the blast. You may have to come up with an offset to use for the x value in the expression based on the actors current speed.
  • DizkoDizko Member Posts: 498
    It almost seems like the vector to angle part of the rule set is getting bugged out somehow. I'll have to post a vid later to demonstrate but it appears like it starts to go the proper way for a split second before bugging out and going straight up in the air. Once you do the initial blast and it bugs out, it works perfectly again.

    I don't think it's problem you described tenrdrmer because it happens even when the actor being blasted never gets to the center or above the blast actor. It appears to happen only when the actor being blasted is moving from left to right and when the actor being blasted is to the left side of the blast actor. It seems to work fine when the actor being blasted gets to the other side of the blast actor.
  • mangaroomangaroo Member Posts: 419
    Dizko try changing the x velocity to 0 first before (0.05 even) the rest of the rules, that may neutralize the variations.

    If you don't mind, how did you record your screen with the indicator for touch? or is it a mouse cursor you sly dog...
  • DizkoDizko Member Posts: 498
    mangaroo said:
    Dizko try changing the x velocity to 0 first before (0.05 even) the rest of the rules, that may neutralize the variations.

    If you don't mind, how did you record your screen with the indicator for touch? or is it a mouse cursor you sly dog...

    I tried to do that, I don't believe it has to do with the velocity of the object. Something is up with the angle to velocity, not really sure what because it seems to be intermittent. It might be some other behavior that's interfering...

    To answer your question, that's just ScreenFlow, it has an option to change your cursor after you record the video.
  • mangaroomangaroo Member Posts: 419
    thanks for the tip, read it from DreamLabs little while ago too, good idea

    Perhaps try changing the angle to 0 too (i was picturing my game in the head where the actor stays the same angle). If your actor rotates before the blast, try resetting angle to 0 at the moment of blast to rule it out with one attribute change. hopefully the quick change wont be seen
  • DizkoDizko Member Posts: 498
    So here's a vid, as you can see the first time I fire off the blast it does something odd, but the 2nd time it works as expected.

    Click the video link it won't play embedded, it's unlisted... don't need people seeing my bugs :)

  • DizkoDizko Member Posts: 498
    Fixed it, there was a leftover interpolate behavior in the otherwise of the rule set for the x push. I had changed the way I did the push a while ago, forgot that there was anything even in the otherwise section. Stupid me :(
  • SkyMapleSkyMaple Member Posts: 817
    Hey, awesome sun! :P
Sign In or Register to comment.