For future reference, here’s a tridiagonal solver in vex (very much not optimised, but it’s based on some Mathematica code) @jake rice 😮
“` #include
function complex cdiv(complex a, b) { complex conjB = conj(b); complex num = cmult(a, conjB); complex den = cmult(b, conjB); complex ans = cmult(num, CMPLX(1.0/den.real, 0)); return ans; }
function complex[] TDMAsolver(complex a[], b[], c[], d[]) { int nf = len(d); // number of equations complex p[]; complex q[]; for (int i=0;i
Author: admin
Archived post by trzanko
there’s a pretty neat trick to getting noisey edges without the rbdmaterial fracture sop
the stuff they do in rbd material fracture to compute the intersection planes is pretty intense
so there’s this as an alternative
doesn’t pass every test
but with some love it can get pretty nice results
excited for the guiding stuff, rolling my own on a job rn and would be interested to see their approach ( it’s for sure better haha )
i imagine their guiding method point deforms a tet mesh to the anim geo, and then perhaps they extract a rigid anim xform off the derivatives of the rest / anim tet deformation, and activate pieces by the difs in the rest / anim xforms; shear, scale, rot difs, etc
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20195411/07/19/edgeFrac_trzanko_v001.hip
Archived post by mestela
neato, bubble solver for houdini: https://twitter.com/sergenern/status/1180889052859555842
ah geez, its the same guy who wrote the ‘how to model clouds’ blog entry and the ‘how to write your own pathtracer’. what a megadude. sergeneren.com/2019/10/08/bubbleh/
Archived post by toad-bot
A post by @legomyrstan has been selected for the archive! www.fx-td.com/houdiniandchill React to any message with the `:gem:` emoji to vote for a post to be archived.
Archived post by legomyrstan
Archived post by WhileRomeBurns
i’m assuming you don’t wanna use actual sdf volumes but you have closed geometry? you could call intersect_all and if the number of intersections is odd you’re inside
this is the crossing number
geomalgorithms.com/a03-_inclusion.html
if you can precalculate, use winding number sop
Archived post by mestela
little experiment in looping cell noise, usual dance of trying it all in vops, cursing up a storm, rebuilding it in snippets
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20193010/31/19/cellnoise_loop.mp4
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20193010/31/19/cellnoise_vop_loop_v02.hip
Archived post by TOAD$TORM
i don’t know that there’s any easy way, color spaces are a bitch and it’s not as simple as “linear workflow” anymore, with ACES and all that. the VES has the best guide I’m aware of here:
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20191810/31/19/cinematic_color_ves.pdf
Archived post by trzanko
np
this will build points on top of a previous layer of points and do that over n over again
basically an L system
also this could work pretty alright for cloudscapes haha
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20190510/30/19/snowGrow_v002.hip
Archived post by JeffLMnT
“` How to deal with Iterations: Substeps – More expensive, but gives a better solve – especially for higher stiffness values Constraint Iterations – Increase for higher stiffness value to control excessive deformations – great for cheaper solver at high resolutions Smoothing Iterations – smoothing ops, to smooth out errors of excessive deformations, or constraint errors Collision Iterations – Uses the Detangle sop – lower frequency iterations – complicated and time consuming – increase when collisions are fighting against constraints or other collisions However, its probably better to increase substeps and decrease constraint iterations for a more accurate solve – and to reduce load on the constraints – but is more expensive “`