View Full Version : Parent/Child Emitter Problem

07 July 2003, 07:13 PM
The basic setup is simple: There are two emitters (parent and child) and one collision plane. The child emitter and the collision plane are on the same level (let's say they represent a water surface), the parent emitter is above them (let's say it's a rain cloud).

The particles from the parent emitter (cloud) fall down on the collision plane (water) and trigger the child emitter by a collision event. The child particles are generated every time a parent particle hits the collision plane.

Everything works fine until frame 189: Suddenly all the particles from the parent emitter generate child particles without any collision event. After that no more child particles are generated.

Here's the scene file:
parentchild.lws (

Please could someone have a look at the file and tell me what I did wrong?

Thanks in advance! And sorry for my bad English.


07 July 2003, 07:13 AM

Come on guys...

Where are all the ParticleFX gurus? I can't believe that nobody is able to fix this...


07 July 2003, 08:22 AM
Id try playing with upping the particle limit and maybe reducing the birthrate a little.

07 July 2003, 08:29 AM
You should report that to Newtek.

I played around with it a little bit, and got it to work with smaller numbers of particles, but it doesn't seem to like the larger numbers.

In the meantime, set your birthrate to 2, and clone your hierarchy of parent/child emitters in place a couple more times. This will make more than one set of parent/child emitters pouring from the exact same spot. It easily runs through your 400 frames of animation without a hiccough and looks about the same.

For some reason, it will handle multiple emitters with lower birth rates, rather than one emitter with a high birth rate.

But, please do report it to Newtek. It shouldn't be doing that.

07 July 2003, 08:50 AM
Originally posted by webfox
But, please do report it to Newtek. It shouldn't be doing that.

I'm a LW newbie and I was convinced that the mistake was on my side. So you really think this is a bug? :hmm:

Well, I will report it to the tech support.

Thanks for your help!


07 July 2003, 06:52 PM
It did the same thing for me.

07 July 2003, 07:07 PM
For 4 frames each particle below the surface gets cloned once.

Isn't there a way killing each clone once they past the surface and triggered a collision?
Or do you still need those?

Because once they'd be gone, I guess your problem would be too.
As a workaround that is.

07 July 2003, 08:24 AM
Originally posted by BladeFX
Isn't there a way killing each clone once they past the surface and triggered a collision? Or do you still need those?

Killing the particles after the collision event could probably solve this. But you can set a collision object's mode either to "event" or "erase". I guess I need another collision plane just below the first one to kill the particles that have passed the first surface.

I can live with the workaround webfox suggested. The animation is just an exercise for me to understand the concept behind parenting emitters. I did it after reading the ParticleFX chapter in the manual just to prove for myself that I got the idea of it. On the LightWave CD are a lot of examples with multiple emitters and collision objects but I didn't find one with parented emitters.


07 July 2003, 08:36 AM
Larry: I finished your free spline patching tutorial a few days ago. It got me hooked on this modeling method! Before that modeling with splines had the flavor of occultism for me. Unfortunately I can't afford your CD training material right now. (BTW: What about the shipping costs to Germany? In some cases the costs for shipping from the US to Germany are higher than the price of the actual product.)


07 July 2003, 08:48 AM
Thanks allaboutkeys!
Shipping cost to germany have not been any worse then anywhere else so far. Usually its making sure that the contents are marked educational.
Splines in LW are pretty flexible and easy to use. Unfortunately most ppl seem to ignore them on not realy get them. Im almost done with the first CD in my organic modeling series which covers splines in great detail.
Have you had a chance to check out the other free material I have? Its parts of my other courses and CDs. Theres over an hour of stuff to check out HERE (

07 July 2003, 05:58 AM
The child emitter birth thing works...but its so tempermental it can take hours of fiddling with sliders to get it right. It should be a bit more forgiving...

What is the magic formula? No one really knows. there are things you can do, and everyone will tell you the same thing, but it doesn't really work well, as do many things in LW.


07 July 2003, 06:34 AM
Usually I go for a method thats dead simple. I ususally dont go for a method that relys on a simulation but rather something where I have more direct control. :)

07 July 2003, 04:15 AM
so have you played with the LW scene - fxspancolleventx.lws?
it uses erase, but may work for your case...

07 July 2003, 08:10 AM
Originally posted by HowardM
so have you played with the LW scene - fxspancolleventx.lws?

Yes, it's basically the same effect like in my scene. But as I said I wanted to explore the logic behind a parented emitter pair.

fxspancolleventx.lws has emitter 2 parented to the collision plane (not to emitter 1). It took me hours to understand the subtle difference between the nozzle types "parent-collision" and "parent-emitter".


07 July 2003, 04:55 PM
I thought it might be good to share with you what the development team found out about my scene file:

The method that you used was not expected for this type of effect. Child_Emitter settings set to be generated by event collision doesn't occur "immediately" because the particles don't support spawning using this method.

To do this type of effect, you can use parent emitter but the particles do not popout under the event plane, so, it is not right on tope of it. Refer to the attached scene.

I have been told that this isn't a bug - just trying to do something in a method that isn't the most accurate.

They sent me a fixed scene with the following settings:

1. As before there are the two emitters where emitter 2 is parented to emitter 1.

2. The mode of the collision plane is set to "erase".

3. The child emitter's nozzle type is set to "parent-emitter(end)".


07 July 2003, 06:02 PM
hmmm, kinda lame eh?
I mean, your method DID work until a certain # of particles hit its limit, then it freaks out...sounds more like a bug to me....
Using an erase workaround is a cop out. Why wont Event work correctly, is a question NT needs to think about...
I started to tweak the demo scenes that come with LW and realized that most of their techniques use an erase instead of an event....really makes me angry actually...I dont think there are ANY sample scenes demonstrating collision event without erase....lame.

07 July 2003, 06:35 PM
HowardM: Well, my method was logical to me - at least after reading the ParticleFX chapter in the manual. Now I must say the erase technique is more straightforward. But yes, you are right, both methods should do the job in the same manner.

After using LW for about two weeks now I'm still more than enthusiastic about this software. (I have to admit that I don't know any other 3d program except for Bryce and Poser.)


CGTalk Moderation
01 January 2006, 02:00 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.