Wire parameters not working as expected...

[pete]

hey dude, usually when i am setting up wired constraints i look at the objects pivot in gimble rotation mode so i can see what axis i'm actually wiring up. Sometimes if you check the object out in local mode you'll think 'oh yeah that's the Y axis' but it may actually be another in gimble mode and that is what max actually uses for wiring. Check you pivots out, there may even be a combination of two axis doing the rotaion of what you thought was one..

Sorry it's all a bit confusing but hopefully it's of some help

[SFDD]

Hi Pete:


Thanks for the reply. It does appear as though when I move one object on the one axis, the other object is moving on that axis plus another. It seems odd because I have set up axis constraints, but that doesn't seem to help. What's really odd is that when I move the first object, the second moves all screwy. (I have set the wire to be bidirectional.) But when I move the second object, the first moves properly. This has happened to me in several different files, so it must be something that I'm screwing up. If I start a new file from scratch and make a few dummy objects, it works as expected.

[pete]

Don't you hate that!
Why does mirroring have to be so evil yet so increadibly usefull

Good luck anyway

[Flipped_Normal]

You said you are using something like

obj1.y_rotation | obj2.y_rotation
________________________|___________________
|
|
obj2.y_rotation | -obj1.y_rotation

in the wire dialogs, correct? That would not make sense. If obj1 rotation was 10, obj2 would try to set the rotation to -10. But then if it tried to update obj1 rotation, it would look and see it should be obj2 rotation (no negative), and set it to -10. This would cause some very weird errors. BOTH sides would need have a - for it to work correctly. Make sense? Of course, I may have misinterpreted your first post, in which case ignore me ;)

Also, I can think of two quick reasons this might not work properly. 1 is that 3d graphics programs use quaternions not euler angles internally, so the conversions could be causing the other axes to be screwed up. The other possibility is because with mirroring in max (for some stupid reason, I'm sure it could be done better) there is a negative scale applied along one axis to get the mirror. This could be messing with the calculations too. Unfortunatly, I am not far enough along with my learning of graphics programming, so I do not see any easy solution. But someone else here may have something to chip in.

Of course, I could be wrong in every way. In that case, pretend I wasn't here :shrug:

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.