Unstable POP PartId - macOS 15.6

The PartID of the Particle POP is unstable in my macOS instance.

Here is a minimal reproduction of the bug on my system. Literally just a PointGen and Particle.

partidBug.toe (3.8 KB)
You can see a video of this here: https://youtu.be/I4LXISLeOmw

It looks similar to the flickering I was getting in this issue Inconsistent flickering/glitching using Particle POP on macOS M1 - #8 by edeetee .
As with that issue, it only seems to happen once the particles reach a certain number of points and seem to flicker for a single frame.

This causes problems with a larger project I have randomly selecting a subset of points to render with a different instance.

A strange behaviour I was noting was that the flickering also was affecting attribute inputs from before the Particles. Its like the memory of the particles is getting overwritten for a partial frame. Reproduction of this behaviour here:

partidBugExtended.toe (4.0 KB)

I since bought an update license from the previous post and have been having heaps of fun with POPs. Its very polished and I’ve got faith that these little bugs will get smoothed out :slight_smile: Let me know if I can help with the debugging process any more.

macOS 15.6.1 (24G90)
Build 2025.32050

Hi @edeetee,

thanks for pointing this out, I can recreate and have logged it as a bug.

cheers
Markus

Hello @edeetee

It looks like this could be an issue with how the attribute values are displayed in the viewer. This sample shows that the PartId attribute appears to remain constant over the particle’s lifetime. We’ll keep investigating.

partidBugExtended.toe (4.3 KB)

POPbuginstances.toe (8.2 KB)

Hi Guillaume

I have attached an example of the bug extending beyond just affecting the viewer. When I run this project, the instances glitch when they should be constant over the lifetime of the particle.

I selectively render points based on ‘Type’, which is derived from modulo TypeId.

Hello @edeetee

In your example, in addition to the attribute value matching, you have a pattern set on delete2 and delete3 ([*:4]). Because the particles don’t keep a fixed position in the list and their indices change over time, you can’t rely on the index to select particles consistently. When you remove the pattern, it works as expected.

POPbuginstances.Fixed.toe (8.2 KB)

Ah! My bad, I was conflating the two issues. The viewer bug persists but now I can use the PartId/Type to selectively delete as desired. Thanks :slight_smile:

Marking as fixed as this issue is deemed only a viewer bug and you provided a solution for my main issue