Jump to content

Particle Birthing


ragupasta

Recommended Posts

Hey guys,

Had a quick search first but nothing popped out at me.

I have a scene in which there is geometry moving at high velocity. These geometry are copy stamped to particles. The same particles are fed into a new popnet to create a dust like effect which will be rendered as points. The main problem I have is with clumping of the birthed particles.

As I understand it the work around is to:

Have the input geometry fed into a timeblendSOP

Oversample the particle system so the network cooks multiple times per frame.

In the animation globals, turn off integer frame value.

Set the step value to the inverse of the oversampling.

I have the oversampling set to 5, so the inverse of that would be 1/5. In decimal thats 0.2, right?

I'm still getting clumping, although not as bad. And I don't want to increase the oversampling much further as things are starting to slow down dramatically.

Is this the proper way to achieve a more uniform/non-clumping birthing from high speed geometry/points? Or is there a better way?

Also whilst I'm here, whats the best way to add a seed value to my particles? Perhaps to the curl noisepop? The particles are birthed from points, so there is no chance of having a birth location seed. The popnet is a very simple one, as the effect is simple. There is the sourcepop, that is fed into a curl noisepop and intern that is fed into a dragpop. I would like to add some sort of seed value, that I can change, so I can then render out multiple layers of 1,000,000 particles and composite them together again. The seed would ensure that the particles are not in the exact same place as the previous render.

I used this method to create a sand test effect a bit back that can be seen here

The difference was in the video I used an isooffsetSop to convert to a volume, then scattered the points inside the volume, and used the seed value in the scattersop to randomize the particles birth location, but kept the overall particle movement the same.
Link to comment
Share on other sites

I usually use seeds in either a force or curl noise POP to randomize multiple pop nets for one effect. As for particle birthing, you can try switching on accurate births, which is described in the help docs as follows:

Recook source geometry before emission of each particle.

This ensures particles are emitted from the correct location when the source geometry is animated. If the shape of the source geometry does not change, you should turn this option off.

Link to comment
Share on other sites

Actually accurate births didn't help at all. Also; it says "If the shape of the source geometry does not change, you should turn this option off." I'm birthing from point's, not geometry. The slowdown I'm getting is due to the timeblendsop, without it the simulation whips along fairly quickly, even when oversampling.

As for seeding, I've run a few test's just using the offset parameters on the curlnoisepop and I should be able to get what I'm after using that method.

Anyways thanks for the input. :)

Link to comment
Share on other sites

Just an idea because I don't exactly know how they clump and what you want to achieve, but how about you use your initially emitted particles only as the birthing positions for another stream of particles that come from a split pop. That way you could perhaps add randomness into the system.

Edit: Do you use constant or impulse birth? (If I remember correctly constant emission will try to emit a certain number of particles. If it misses it's target in a step it will catch up later, thus producing bursts of particles)

Edited by Macha
Link to comment
Share on other sites

Thanks for the help guys.

@macha: Yeah I'm using constant birth. Thanks for the file macha, the clumpyness is similar, although yours is very exaggerated compared to mine.

I've attached a screenshot with point display turned on, and you can see the particles clumping (thankfully not as bad as your file macha :D ).

post-2039-12779798902_thumb.jpg

Link to comment
Share on other sites

I think it may be related to velocity then. If they inherited the source's velocity then they all end up with a similar speed in a similar direction, so the effect is even more visible when it is animated. Even if they are emitted at slightly different time positions you always end up with these clumped pulses.

If you append a drag node and enter an expression like

if($LIFE<0.01,100,0)

does that make a difference?

Edited by Macha
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...