View Full Version : Geo constr. (on nonplanar surface) + Point constr. = problems

09 September 2010, 03:20 PM
Hey people! It's been a while. Anyway, this question mostly in the title. I want to sample information from a NURBs surface with a locator (geo constr.). The locator itself will also track the position of an object (point constr.). When you use a NURBs plane, the locator will follow the object within the constraint of the NURBs surface as expected.

As soon as I try to manipulate my NURBs plane into a non-planar surface however, the locator sticks to the edge of the NURBs surface! Simply put, the locator can't track the position of the object along the inside of the NURBs surface, only along the edge.

A quick visual of the problem:

Thank you for your replies in advance ^_^.

09 September 2010, 04:18 PM
I assume the locator is trying to get as close to the cube as possible (as it is point constrained?) but it has to stick to the nurb surface (geom constrained?).
If this is correct, well, the locator is in the "correct" place, because the edge (in your example images) is the closest point to the cube...


09 September 2010, 07:51 PM
:O *lightbulb* That's correct! I misunderstood the function of the point constraint!

To try and alleviate the problem, I introduced more objects into the system (another torus) and had the locator also point constrained to that torus2 which is parent constrained to the original torus1.

Only thing is, it only alleviates the problem. There's two problems that come up when I use this method. The first is that the locator, although it does have a wider range of motion, tends to "skip" towards the lowest part of the dip and another time when the edge first touches the dip.

The other problem is mainly in terms of functionality that I was hoping to get from this system. My hopes was that I could move the torus closer or further away from the NURBs surface and it wouldn't affect the locator's position (because technically moving it closer or further away would be moving along an axis that the surface can't read).

To fix this, I could theoretically just add multiple instances of objects or locators and surround the NURBs surface that's a bit inefficient, even using MEL code to do, especially since I plan to use this concept of tracking probably 10+ in all of my scenes!

CGTalk Moderation
09 September 2010, 07:51 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.