_Christopher_
The geometry wrangler is just that; not a dop network with a geometry wrangler ?

It has to be inside of DOP network, since it's a DOP node, there is no Geometry Wrangle at SOP level

And yes it's different a bit as it can bind fields and operate on DOP data etc,... but its still similar so while you can also do all the things in SOP Solver and Attribute Wrangle, Geometry Wrangle DOP can handle a lot without such roundtrip

tamte

_Christopher_
Thanks, although the group; bend_stress is not having any points assigned to them, is it because the values in the attribute bend_stress are extremely low to the decimal place so to speak ?

yes, check Geometry Spreadsheet to see what values you are getting, they may be tiny so you need to work within those ranges
that's why stretch_distance, stretch_ratio and bend_angle may be more intuitive attributes to work with

How about fitting the values; or because the values are so low, fitting isn't ideal ?

bend_ratio is better then bend_stress still the values maybe an issue 😉

_Christopher_
How about fitting the values; or because the values are so low, fitting isn't ideal ?

you can remap the values however you want, I don't exactly know what stress measures and what are the expected ranges

you can definitely just pick random max value and divide the whole stress value by it to get more reasonable values

however I'd avoid dynamically fitting to min-max range of the values on the geo as those may be different every substep so it may not be ideal for judging the real stress as it would be relative to min/max stress within the current substep which is useless for any serious computation (except maybe if you want to visualize relative stresses within the substep, which the visualizer can automatically do)

tamte

_Christopher_
How about fitting the values; or because the values are so low, fitting isn't ideal ?

you can remap the values however you want, I don't exactly know what stress measures and what are the expected ranges

you can definitely just pick random max value and divide the whole stress value by it to get more reasonable values

however I'd avoid dynamically fitting to min-max range of the values on the geo as those may be different every substep so it may not be ideal for judging the real stress as it would be relative to min/max stress within the current substep which is useless for any serious computation (except maybe if you want to visualize relative stresses within the substep, which the visualizer can automatically do)

Suppose I wanted to append some procedural noise to a stress attribute although without just having the procedural noise appear on the object on any given frame if it were to be applied using the SOP Solver for example what would be ideal in this situation, sorta like it growing with the stress attribute ? When the stress attribute gets to a specific value, the noise grows but other areas of the noise are masked.

Also since procedural noise applies its self to the whole object, is it possible to mask out areas of the procedural noise ?

You can keep track of max stress value across the sim so far
Either as point wrangle right after vellum post process in the solver or geometry wrangle DOP after SOP Solver it the other Geometry Wrangle DOP (if you decide to use that one to compute stress)

Then remap and use it as a mask for your noise which can also be done post sim for example
These are all pretty standard attribute operations though so I'm not really sure if you are running into some issues or just asking before even trying or researching

tamte
You can keep track of max stress value across the sim so far
Either as point wrangle right after vellum post process in the solver or geometry wrangle DOP after SOP Solver it the other Geometry Wrangle DOP (if you decide to use that one to compute stress)

f@max_stretch_stress = max(f@max_stretch_stress, f@stretch_stress);

Then remap and use it as a mask for your noise which can also be done post sim for example
These are all pretty standard attribute operations though so I'm not really sure if you are running into some issues or just asking before even trying or researching

In a previous message if I didn't want to use a SOP Solver I could use a POP Wrangler or a geometry wrangler in DOPS. Now it seems as if one would have to use a POP Wrangler or geometry wrangler in DOPS after a SOP solver 😉 Unless that still is the case, it's just how you happen to re-explain ?

_Christopher_

tamte
You can keep track of max stress value across the sim so far
Either as point wrangle right after vellum post process in the solver or geometry wrangle DOP after SOP Solver it the other Geometry Wrangle DOP (if you decide to use that one to compute stress)

f@max_stretch_stress = max(f@max_stretch_stress, f@stretch_stress);

Then remap and use it as a mask for your noise which can also be done post sim for example
These are all pretty standard attribute operations though so I'm not really sure if you are running into some issues or just asking before even trying or researching

In a previous message if I didn't want to use a SOP Solver I could use a POP Wrangler or a geometry wrangler in DOPS. Now it seems as if one would have to use a POP Wrangler or geometry wrangler in DOPS after a SOP solver 😉 Unless that still is the case, it's just how you happen to re-explain ?

not really, doesn't matter where you compute the attributes SOP solver, POP Wrangle, geometry wrangle
just easily put another POP or Geometry wrangle after and do the max, or do it in the same one or inside of the same SOP solver
there is infinite amount of options on how you decide to structure your graph, but data flow is very straightforward
how these nodes operate on data and how to chain them one after another are the very basics and shouldn't cause that much confusion

Even though I'm using fit01 to refit an attribute to 0-1, I don't see any of the values changing ?

Here is some values of the refit/remapped attributes;


The left values are the refit/remapped values using fit01 the right is the actual attribute. The values are the same ?