View Full Version : Orient Constraint - Local rotation data

06 June 2010, 11:59 AM
Hi guys. As you know constrains in Maya read world coordinates, which is good at most cases. But what I want is now the opposite: object coordinates. Say that we have two cubes, one is constraint to another. If I group the master object itself, and rotate the group node, the target object is rotated too. Because final matrix of the master object is changed. What I want is that target object should be affected only and only If the master object's local coordinate is changed. I have examined the DG connections and found out that the Parent matrix attr is the cause of this, so I broke that link, and it works just like what I wanted it to.

Am I doing right though?

06 June 2010, 02:16 PM
What you are actaully wanting to do then is not constrain at all

but drive the rotations. By linking the attributes.

In the sample file you are looking for the behaivor of Cube set E and F

A and B = Orient Constriant
C and D = Parent Constraint
E and F = Linked Attributes via the HyperShader

06 June 2010, 03:45 PM
You're so right Darksuit. My bad. But what If an Aim constraint? I am asking this because sometimes it causes double transformation and I think it can be solved this way. What you think?

06 June 2010, 04:12 PM
really it all matters on what you are doing this for and what the end result is going to be.

Best thing you can do is quick fast exeriments. The sample file I attached took a mere 30 seconds to set up. most of that was placement.

Sometimes it helps to just jot down on paper to get the idea out.

Write down what the end goal is that you want to acheive.
answer the following questions:
1. What are you doing this for? (self improvement, Film, Games, etc..)

2. What restrictions do you have, are you limited by the types of constraints you can use, is there a method you have to use for engineering reasons, etc...

3. What is the simplest way to achieve the result? Even if its faking it (not actually doing the process but at least getting the visual to look right). Ie can you key both. Doesn't matter if it takes more time, sometimes the process will teach you more about what you are actually doing, or help someone else understand better what it is that you want.

4. How many different ways can you think of to acomplish the task?

Also if you experiment small, keep it simple. Simple allows for rapid interation and quick protyping, changes and evaluations. This is more or less based on the scientific method.

When ever I answer a question here. =)

06 June 2010, 06:32 PM
Thanks again David for clarifying it. I was making a robot rig. Sometimes I get into too technical things and fail to figure out the solution which in turn to be the easiest technically. I totally agree with you.

CGTalk Moderation
06 June 2010, 06:32 PM
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.