5 replies to this topic

[breadbox]

Having an issue with getting the polywire to uv map straight lines.

I have an example attached of what I'm looking to do.

problem is I start with a very simple polygon object which has only one vert span between its lengths.  this uv maps well because i can use $PDIST in the poly wire UV options to make UV that runs the length of the wire.

However I'd like to resample the polygon object, and then deform it with mountain to get it to make a wobbly line once the polywire is applied.  if I do this then $PDIST samples each sub segment of the line instead of taking in the larger picture of the entire length.  this is as expected however its not what I want.

If I apply the mountian after the polywire then it changes the thickness of the wire in an undesireable way, it gets thicker and thinner in places where i want it to be consistent.

there are a couple ways to approach this problem, one I could maybe create a length attribute before the resample so that each segment knows where it should be organized within the UV map.  or 2 find a way to deform the mesh post wire, so that it doesn't change the thickness of the wire, and I'm not sure how to do either.

any ideas.

Attached Files

•  WireFrame_CaseStudy_V3.hipnc   306.07K   35 downloads

[Macha]

I'd say temporarily convert to nurbs, uv row and column, and then back to poly, then deform, and your wire mesh last.

Edited by Macha, 25 February 2012 - 06:04 PM.

[breadbox]

Macha, on 25 February 2012 - 06:02 PM, said:

I'd say temporarily convert to nurbs, uv row and column, and then back to poly, then deform, and your wire mesh last.

not sure I understand you.

If I deform at last step then what will end up happening is the mesh will get thick and thin like this attached.

there might be a method to deform that preserves the shape of the wire?

I guess Ideally the best case would be to figure out how to UV the polywire with multisegements input, because often I will get meshes that might already be subdivided, and that way the wire stays a consistent width

Attached Thumbnails

• 

[Macha]

Nono, I meant make the polywire last.

Yes, there probably is a way to deform a mesh in such a way that thickness doesn't vary. A rest attribute of sorts, or draw rays perpendicularly to a target surface and displace along them, but I don't think that should be necessary. Or are you saying you have no control of that step; you get a mesh and have to deal with that as it comes?

Edited by Macha, 26 February 2012 - 12:27 AM.

[kgoossens]

breadbox, on 25 February 2012 - 07:03 PM, said:

not sure I understand you.

If I deform at last step then what will end up happening is the mesh will get thick and thin like this attached.

there might be a method to deform that preserves the shape of the wire?

I guess Ideally the best case would be to figure out how to UV the polywire with multisegements input, because often I will get meshes that might already be subdivided, and that way the wire stays a consistent width

Is this what you are looking for?

Attached Files

•  WireFrame_Deformed.hipnc   347.15K   52 downloads

[breadbox]

kgoossens, on 26 February 2012 - 12:41 AM, said:

Is this what you are looking for?

right so this is cool example I was looking into vex noise nodes to do the deformation because the mountain sop creates the thick thin look which is undesired. Vex can preserves the width of the wire a bit better.  however the problem has gotten more complex.

on second look It looks like ill be getting meshes that might be already subdivided in a way that's sometimes more than one division per span.

Macha, on 26 February 2012 - 12:25 AM, said:

Nono, I meant make the polywire last.

Or are you saying you have no control of that step; you get a mesh and have to deal with that as it comes?

so in that case the polywire wont make nice uv's if there are subdivisons along the original span.  so maybe i need to create the uv's another way other than within the polywire?