The “starting point” of the noise module always resets to the same position whenever it is simulated anew, this creates a repetitive motion pattern. I don’t want to use “random between constants” for strength because it is a subemitter for creating a consistent line of trail. Random strength will disperse the trail which I want to avoid for specific occasion. Is there any way to randomize the Noise starting point?
Sounds like you need a new parameter called “Scroll Offset” which supports a Constant and a Random Between 2 Constants option?
Yes. Even though I actually think the offset should always be randomized, does not hurt if you make this option available when we want the same offset for different emitters but with different strengths to create sth exactly like Last Dreamy 15’s Warp Strike.
I was thinking about this some more. I would expect the noise to be different each time if you use “Auto Random Seed” in the Main settings.
But, if you set a seed, then the effect should be exactly the same each time. I can see from your video that Auto Random Seed is checked. So I’m not sure why the effect isn’t changing on each play…
That brings back the memory of the unrandomized emission seed back in 5.4 which did not affect the “randomize” option of Force over Lifetime, so I am not surprised if the Noise module uses a separate seed independent from emission seed.
Yeah, this seems odd. Can you send me the project, so I can investigate properly?
Perhaps, if easiest, just log a bug, mention my name in the report, and post the case number here when you have it.
Then I will grab it.
Thanks
Alright there you go: 873346
I’ve had a look at this.
The issue is present in 5.5.x, but is fixed in the 5.6.x cycle.
Given that it’s not a regression or crash bug, I think the safest option for us is to not attempt to backport the fix… (all bug fixes incur a risk of breaking something else, and, looking at the code, it’s not totally safe to attempt to bring this fix back to 5.5).
Does this sound ok? Is it a critical issue for you?
Not really a critical issue, so it’s okay to close this case if it has been fixed for 5.6.