ICE path constraint ?

[Renato Polimeno]

Hello guys,

I did recreate a path constraint following this tutorial without problems.. (it seems like a 'fake cyclic', Doesn´t it ? Get object kine.pos, check closest location on curve, set the result back to its kine global)

However, I´m looking for a solution to avoid the current object translate and use a 0-100 percentage control instead (or 0.0-1.0 normalized value) How to do a scalar slider to travel my object along a curve ? 

Any hints would be highly appreciated.

Thanks in advance.

--

______________________________

 

-> Renato Polimeno

+55 11 7539 3615 

renatopolimeno.com

facebook.com/renato.polimeno
@renatopolimeno

[Alan Fregtman]

Hi Renato,

Here's an example tree:
http://s3.darkvertex.com/hlinked/ice/ICE_example_kinematics_pathOrCurveUConstraint.png

You would have to use the UV To Location node to turn a U value (0.0
to 1.0) to a location then to a position and so on. Divide your
desired percentage by 100 and you get the U value.

Hope that helps.
Cheers,

-- Alan

[Renato Polimeno]

Hey Alan,

that is exactly what I´m looking for. Didn´t know that UV to location node, thanks a bunch ! 

cheers,

[Arvid Björn]

Sorry for digging up an ancient thread! I'm looking for a way to do curve/path constrain using ICE.

I tried your setup but the null is jumping all over the place, especially if I step back and worth between frames. I'm not sure if I'm missing a setting somewhere, do you have a stable path solution using this setup? Any other ways of doing this?

Cheers :)

[Dan Yargici]

Have you accidentally created a simulated ICE tree rather than applying it in a non simulated ICE tree?

DAN

[Arvid Björn]

I've tried both! I assumed it shouldn't be simulated, but I get the same result either way. It seems to build on top of the current null position, so it never starts out from the same position or rotation, it gets weirder every time I play it. I would have thought this tree would just over-write the kine.global and have it fixed like a constraint, but it's sort of half-loose.

[Eric Thivierge]

Do you have the ICE Tree that is calculating the position on the object itself or on an empty point cloud or null that stays at world origin?

--------------------------------------------
Eric Thivierge
http://www.ethivierge.com

[Arvid Björn]

It's directly on the null itself, I assumed it would work because the SetData node was writing to self.kine.global in the example image. It works much better if the tree is on a static null, and writing to the moving null's kine.global! I guess that works for me, is that the only way to make it work?

[Eric Thivierge]

Its the best way for most ICE Kine operations as it ensures the transform space in which all the calculations are done stays constant and doesn't move around. You can do calculations on a stationary null and set a custom attribute that stores the transform then on the null itself have another ICE tree that reads the data from the stationary null ICE tree.

--------------------------------------------
Eric Thivierge
http://www.ethivierge.com

[Arvid Björn]

I'll try that, thanks a lot!