Particle turbulence behaves very differently in 4.1.dev3

[Novark]

Godot version

4.1.dev3.official [a67d37f]

System information

Windows 11 [Vulkan]

Issue description

The work performed in #77154 drastically changed the visual characteristics of particle turbulence. I am no longer able to get the same "turbulent" behaviour that existed before these changes - no matter which turbulence settings I change. Please let me know if there's something I'm missing about the new implementation that would give this result.

Steps to reproduce

To reproduce this is fairly simple:

1. Create a GPUParticles2D Node
2. Emit a significant number of particles (I'm emitting 50,000)
3. Set Initial Velocity [Max] to around 250, and Direction Spread to 180. Enable turbulence
4. Set the turbulence_influence_min property to 0.1 (default) and turbulence_influence_max to 1
5. Observe how the turbulence appears in v4.1.dev3.official with the changes in Reworked Particle Turbulence #77154 versus the previous build without these changes. The new implementation just results in some minor particle "jitter", but not the actual turbulence that it's supposed to have.

Minimal reproduction project

No sample project provided, as the reproduction is quite simple to replicate.

[Calinou]

cc @KdotJPG @RPicster

[KdotJPG]

I was able to get the turbulent look in both, but I had to change the scale for some reason.

I don't know why this difference is showing up. No changes to the scale value curve were merged since 4.0, and the same is the case for how it's effectively used.

[Novark]

I was able to get the turbulent look in both, but I had to change the scale for some reason.

I don't know why this difference is showing up. No changes to the scale value curve were merged since 4.0, and the same is the case for how it's effectively used.

[Image]
[Image]

Edit: Disregard my original comment below, I was looking at the wrong image when I copied your settings. I was able to produce the same results as you by changing the noise scale as well - which I agree is an odd change. The scale characteristics are bizzare - there's basically no effect at either end of the value range - it needs to be right around 3.0 to see any turbulence, and even then, small changes centered around that value seem to result in wildly different visual behaviour.

That's very strange...I just tested it again with your exact settings and see the same result that I got before - no turbulence.

[KdotJPG]

^ found the solution!

[Novark]

^ found the solution!

Glad to hear it! Do you know if this change would also resolve #74541? It may be worth taking a look at while you're in the same headspace having just solved this one.