Where the KCM_Selection is just setting the Selection channel to be 1.0. If I don’t have this, none of the particles are selected, and thus don’t get the Color modfication.
Maybe if there was some more specific way to mark a particle for deletion (which is a boolean, really), we wouldn’t have to keep setting that additional “select all” KCM.
Don’t forget that you can use the soft selection to delete a particle once its crossed a randomized per-particle selection threshold, which makes it more than a simple boolean. Whether anyone actually needs this is a different story.
I think the most prudent solution to your scenario is to (optionally?) have the Krakatoa Delete modifier reset the Selection channel to 1 of particles that weren’t deleted. I’ve also toyed with the idea of making a modifier that combines the selection and delete step into one, where the output node is forced to be a bool/float nameless channel that determines if that particle is deleted. This requires a bunch of refactoring of the KCM editor code so I’m leaning away from it at the moment.
I made a post earlier but I had hurriedly read the thread I wasn’t sure if what I had asked about and this are the same sort of thing:
I was curious if it was possible to set the value to be used as a bool, for instance 20 for true 21 for false, instead of just being locked to 1 and 0. Is something like this possible? It seems to makes some sort of sense to me that you could create multiple selection channels this way. Does this make any kind of sense to anyone else?
I do like the reset to 1 if still alive idea, too.
BTW what’s this I read something about triangles and an island known for it fine quality of coffee?
I was thinking that a bool would be ideal, since you would either be marked for deletion or not, and that can be represented for one bit per point, taking up a lot less memory. But now I see that there is no bool type for PRTs. Oh well. So yeah, Darcy’s idea of resetting would work best for me. And sorry, what you are saying doesn’t make sense to me. Do you have a example use case?
seems just something odd I can’t figure out more of a selection issue and not delete (but could be delete), @ frame 30 I lose selection, I would expect the both would either remain or lose selection not just one. seems due to ordering in the stack, as swapping the selection kcms opposes the effect. I think this is due to only having a 0 or greater than 0 for selection. Is it? test_Shot004_v002_001.rar (20.2 KB)
I can’t seem to figure out what this scene is supposed to show. I find the selection is consistent across all frames. Can you make a movie that indicates the problem?
Also, I’m not sure exactly what you were trying to do with the Ray Intersect node. Your use of it is creating a ray at the particle’s position, and using the particle’s position relative to the origin as the ray’s direction. Was that on purpose?
Well this being a failed experiment from an earlier project and since I am not exactly sure how to derive the volume of an object via KCM I figured (with my upside down logic) the if I cast a ray to the same point as its origin it would just simply be a point and the SurfData uses the z-axis so anything from one of the points along the z-axis, I figured would be considered in the volume (of course with no limits in the z- or z+) Doesn’t really matter though they are selected by what they are supposed to be selected by.
So, how do you calculate the volume? (for another topic I suppose ) is it as simple as creating a vector and adding the objects pos. track?
As in the scene though at frame 30 particles AFAIK shouldn’t be shifted like they are, I would think that they would act as if you turned one or the other off, it should be a reflected effect. Instead something strange happens, sure the particles are still selected but they seem to get confused at which KCM they are supposed to be in.
