Chainmail Part 2: Deform the Chainmail Using UVSample

Support us on Patreon: / entagma
Part 1 : • Chainmail Part 1: Crea...
In this second part of the chainmail tutorial Manuel shows you how to create a Vellum cloth sim and how to attach the chainmail geo to the simulated cloth using the VEX function UVSample().
Download Scene File: www.entagma.com/downloads/H_C...
00:00 Intro
00:17 Fit the chainmail into the 0 to 1 space
02:03 Create a Vellum cloth simulation
04:30 Put Uns on the cloth
05:26 Create the UV deformation setup
06:04 Write some VEX code inside of a point wrangle
09:07 fixing the Y dimension
10:45 sweeping a circle along the deformed chainmail splines

[@vfxskull6208]

Hey Manu, Once at Aixsponza you gave a masterclass on how the primuv works - how to find a uv position on a primitive - and i've been using this knowledge on each and every project i work. It's always super useful in one way or another Thank you so much! - Alvaro

[@Entagma]

Hi Alvaro. Hope you're fine! Great that this was useful for you

[@jefflhama]

Hi, your comment just made me look after it and it's amazing! thanks

[@petarpavlov]

@@jefflhama Now I am curious to find that video - if you could share the link, that would be wonderful!

[@JoeBissell]

wow I leanred so much!

[@babajaiy8246]

This doesn't 'work' because the rings get deformed too. The ideal I would think is to sample the circles prim center and use it's normal as an @orient to copy to point rings. Will have to try a setup like that. I had one for a 'neck' chain using a bullet solver although it works it's computationally heavy, thought this might be a good alternative. Edit: So I tried sampling the rings prim centers instead, created new points from those positions and set those newpoints normal attributes to that of the sampled normals of the prims position . This works in that it solves the deformed rings so that they are now like a true ring. The last 'problem' to solve is that there are many cross over intersections between parts of the ring. Since I'm only wanting a chain, by reducing the 'hex' columns to only one or 'hex' row to only one; Then there is far less interactions to deal with and with some minor manual adjustments those interactins can be elliminated. Perhaps with such a set up it can be run through a bullet solver and since it's already near it's final position that the calculations might not take to long just using the same object in the vellum sim, lower its' elevation a bit and/or adjust so that it is clear of any ring interaction.

[@user-zc9eh1qn5s]

very helpful, I was using vellum hair sim for those line,but not rigid as this

[@makowiec7200]

Thanks for the tutorial!

[@vipergx]

Thank you, professor.

[@jimjamran]

So so cool!

[@yonlaycabrera4431]

Why not just point deform the grid?

[@foreignlanguagedisco6379]

Houdini newbie here, been trying all day to get this to work. I'm trying to integrate this into another cloth sim tutorial and they use a simple copy to points set up but when I go to deform the geometry I can't find out a reliable method to deform the geometry post-solver (seemingly using the merged geometry by creating an null at the end of the UVdeform node into copy to points just "solves" indefinitely on my machine). The UV deform node set up tries to join all 3 instances of the geometry together (creating web of vertices between the instances) rather than correctly mapping to each instance uniquely. Is there any method to achieve what I'm asking?

[@unveil7762]

Boom 🤯