ThomasPara

Cant see where your input geo are merged into the sim and witch attributes you transfer, easier if you support us a file.

//set a wrangle to run over primitives int primpts[] = primpoints(0,@primnum); removepoint(0,primpts[0]); removepoint(0,primpts[-1]);

In a particle sim you can get the attrib @life

if you post your file (or example), then its easier to help you.

your setting @myage back to zero every frame, just remove that line.

As long as your defining your gaps by a distance it wouldnt be a problem to just reapply the same operator (in this case with my solution you would have to create a more solid way to figure out the "val" variable, atm i guess it will be wonky with uneven point distribution). If you want to use a percent approach, the easiest way is to store the original length of the first primitive and pass it on. I would make it so that you pass on two arrays to the operator with position and width for all your doors (so this would be another asset you apply first). And then your doorway asset needs to be able to make several new prims based on the input. Anyway, this is by far a final tool, its just a start.

Wanted to try a pure vex solution. curvegap_01.hiplc

It does both, the new deformer bends and twists. So when you search for a twist deformer it gives you a deformer that can twist, the same with bend.

you can also put in a third argument in the foreach, the index you are running over. //run over primitives, use this on a polyline and you will get the order of the points int primpts[] = primpoints(0,@primnum); foreach(int index;int i;primpts){ setpointattrib(0,"index",i,index,"set"); }

Heres a quick example of how it can be done. fishnet_01.hiplc

Your fitting only clamps your values, instead of remapping them. Your random have a static seed, giving the same result to every point. It would be alot easier to use a zero-centered noise, adding that to your normal. And then normalize your normal instead of fitting it. In vex you can use the "sample_direction_cone" function.

Heres a version thats faster than the for-each, its some more nodes to make that happen. The wrangles "minimize_search", "find_unique", "set_search", and "set_color" has a combined time of around 1 sec, while the foreach is around 13 sec. This runs over 5 mill points example_fix_02.hip

just tested theese solutions, and none of them are faster than the for-each, think this needs a new iteration

//set the wrangle to run over group1 int count = findattribvalcount(0,"point","id",i@id); //find how many points that have the same id for(int i = 0;i<count;i++){ //run over the points and set the color or any attribute of your likeing int found = findattribval(0,"point","id",i@id,i); vector color = set(1,0,0); setpointattrib(0,"Cd",found,color,"set"); } example_fix.hip

do you have an example file?