can someone explain how Maya's Lattice deformer is industry standard?

[adrian_wyer]

simple task, get a cube with lots of subdivisions in x

add a lattice deformer

try and replicate this shape (softimage, 5 seconds)

instead you get this; 

anyone?

the falloff in maya is just plain wrong... 

[brentmc]

Hi,

Did you reset the lattice settings before creating it.

I am guessing maybe local mode is disabled since it works for me with the default settings?

[adrian_wyer]

here's an example scene.... i can't get it to deform correctly even with a tool reset....

[brentmc]

Hi Adrian,

This is the attribute I was referring to and enabling it on the ffd node makes the mesh deform with the lattice.

[adrian_wyer]

with local on; 

with local turned off;

hopefully this is user error, i'd love it to work!

[brentmc]

Local Mode off seems to be the culprit. The default is on. (so this must have been toggled off in the options before creating the lattice)

From the Maya lattice docs :

---

Use Local Mode
Specifies whether each lattice point can influence only the deformable object’s points that are nearby (local), or can influence all the deformable object’s points. Check on or off (default is on). If on, you can specify Local Divisions.

[adrian_wyer]

ok, so with local now turned on, and the local divisions set to 3 in S (so there should be falloff between lattice control points) i get this;

is my maya broken? using 2023 btw

[damaggio]

User mistake then. 👍

[adrian_wyer]

🤣not really... i would expect a smooth falloff in the deformation between lattice control points (see earlier softimage screen shot), what i get is linear fall off with 'local' turned off, or some weird ass bendy hard edged deformation that is clearly not the intention of the tool

there are plenty of subdivisions in the deformed geometry to accommodate a smooth curve deformation... the tool just isn't providing this

[jmreinhart · Accepted solution]

@damaggio you can get close to what @adrian_wyer is looking for by using the local settings and adding some divisions to the lattice itself, but there are definitely some artifacts at the borders of cells where the deformation is not smooth. The only way to make it smooth is to bump up the local influence and when you do that the mesh gets too flat.

You can work around this by having a low-res lattice drive a high-res lattice, an then drive the mesh with the high res lattice. That way even with a high local influence value you still follows the zig-zags. In this particular case though a wire deformer or nonLinear deformer would be simpler.

So I agree with @adrian_wyer there does seem to be an issue with non-continuous deformation between cells.

[adrian_wyer]

thanks @jmreinhart thought i was missing something.... in the end i went with a wire deformer, which is much more suitable.

the problem with upping the number of divisions in the lattice to get smoother deforms, is that it kind of negates the point of a lattice... fewer points providing larger/high level deformations...  

thanks for testing tho... good to get a smarter person than me doing a sanity check! 😁

[damaggio]

That's great to know, I'll keep a record of your tests. 👍