Particle Creation - Grey scale

[Lyonz]

Hi,

Im looking to set up procedural tools that create particles based on grey scale images. And im wondering if anyone has tried this and know some approaches ? As im currently looking into using cop's to bring the image in and thinking Vop's will be able to activate particles based on greyscale images. Any Advice appreciated

[Macha]

Like this?

emitter.hipnc

[Lyonz]

Like this?

Hi i tried this with my image sequence but it doesn't seems to update as the pattern grows. I have changed the birth rate and constant activation but doesn't make a difference, It stays all in one corner. However this is great for static images kinda like the image blow-away tutorial.

The end result that im after is to create unique firework patterns and thought this would be a good approach. By using B/W images as sprites that split off a single firework particle, to start/end the firework images in the sky. A similar effect to this can be found at the end of Kung Fu Panda 2 as shown on the image below obtained from google images. Hope im going the best way about this problem ?

Otherwise have been looking at tron fireworks to get the desired asthetic this using geometry to create the patterns: http://work.gmunk.com/1316279/TRON-Fireworks

[Macha]

Append [$F] to the file path and see if that works with sequences, and of course make sure emission isn't set to $FF==1

Edited June 5, 2011 by Macha

[Macha]

Like so:

emitfromb.hip

[Lyonz]

Append [$F] to the file path and see if that works with sequences, and of course make sure emission isn't set to $FF==1

Test Render working will update when i have something more towards what im after.

Particle_FX_01.mov

[sanostol]

emitting by primitive attribute alone often shows the emitter resolution, so sometimes I do this, maybe it helps

filteremission.hip

[mawi]

Here is a quick test using attributes and groups to randomize geometry goals.

Tried to comment stuff but it´s a bit fast and not so procedural so please ask.

fireworks_pattern_v001.hip

[Lyonz]

Here is a quick test using attributes and groups to randomize geometry goals.

Tried to comment stuff but it´s a bit fast and not so procedural so please ask.

Hi,

Thanks for the reply this network is really helpful. If i was to freeze the split particle geometry though, after say 10 frames is there a node for this ? Speed limit.. So that i could apply my fire cracker particle effect to each of the split particles, for a time period before fading off ? Hoping to get my animated geometry moving and crackling in the sky for a selected time before fading.

[Lyonz]

Update,

Created a spray effect from a single particle and a split node. This particle then dies and activates a second split node for the 3d geometry to use. How can this split node reference the cube points, so that the particles are stuck to it ?

I have looked at the pattern method, however dont understand the logic behind this ? how the birthpos and goal attribute can offset the geometry from the world space origin ?

Thanks

Firework_Pattern_v002.hipnc

Edited June 13, 2011 by Lyonz

[mawi]

Update,

In the process of linking an animated cube to the pop network.

Have created a spray effect from a single particle and a split node. This particle then dies and activates a second split node for the 3d geometry to use. How can this split node reference the cube points, so that the particles are stuck to it ?.

The setup i did is not going to work with animated goals. The particles are finding their goal position only at birth frame. Also, you will need the same amount or more points on your geometry then particles since they all are trying to match $ID with $PT on geo.

I have looked at the 2d pattern method, however dont understand the logic behind this ? how the birthposition and goal attribute can offset the geometry from the world space origin ?

Since the split particles are looking for a point on geometry and that geometry is in origo, you will have to add to geometry position to center were the particles got birthed. If not, the particles would fly towards origo.

After this im using a speed limit to slow down and freeze the particles, so that the animation can be displayed before fading away. Is there a way to split particles randomly based on particles ? And hide the original particles used with the split but keep splitting more. Hope this makes sence.

Thanks

I´m not really sure I understand what you are trying to do.

Edit.

One idea is to build the basic motion and targeting in 1 popnet and then do sparks and stuff in another.

Sometimes it becomes a bit messy when having alot of groups, splits and forces in on network.

Edited June 13, 2011 by mawi

[Lyonz]

The setup i did is not going to work with animated goals. The particles are finding their goal position only at birth frame. Also, you will need the same amount or more points on your geometry then particles since they all are trying to match $ID with $PT on geo.

Since the split particles are looking for a point on geometry and that geometry is in origo, you will have to add to geometry position to center were the particles got birthed. If not, the particles would fly towards origo.

I´m not really sure I understand what you are trying to do.

Edit.

One idea is to build the basic motion and targeting in 1 popnet and then do sparks and stuff in another.

Sometimes it becomes a bit messy when having alot of groups, splits and forces in on network.

Hi,

What im trying to do is simply have a single particle create a trail with the split node which i have done.

Then once that particle dies use the dead group to create another split, which is fine but this time creating a piece of geometry out of particles (one particle per point).

Im then hoping to slow these particles down with a speed limit node keeping the shape in the sky as the geometry animates.

Then for each particle it needs to create a spark effect randomly before the animation fades out (again another split node for each point on the shape).

And atm im trying to understand how your method works.. Why a birth position attribute is needed ? As without it the shape flys away ?

Also i understand that the attribute random pattern is pulling in the shape but for the goal attribute how does it know where the particles should go. Then i get that the Vop pop is using world P and subtracting to get the combine offset but why is the goal needed.

Kinda lost with this as couldnt find a GOAL local variable..

Thanks In Advance

[Lyonz]

Hopefully this shows what im looking for..

1. How to link the split particles to the points on the geometry as it explodes outwards in the correct position ?

2. Once in the correct position setoff a trigger for the spark animation.

Thanks

Firework_Pattern_v003.hipnc

[mawi]

Here is a "easy to follow" particle-goaling file with and without offsetted goal.

particleGoals_v001.hip

Edited June 14, 2011 by mawi

[Lyonz]

Here is a "easy to follow" particle-goaling file with and without offsetted goal.

Thank you get how this is working now

Wondering though ive tried to replace the scatter with a delete sop set it to keep points only and impluse burst on 8 particles, yet 1 point never seems to get its particle would you know why this is ?

Also i get one vector minus another gets the distance and direction Goal - current position but why is the point velocity attribute used and what controls the speed this happens ?

David

[Lyonz]

Thank you get how this is working now

Wondering though ive tried to replace the scatter with a delete sop set it to keep points only and impluse burst on 8 particles, yet 1 point never seems to get its particle would you know why this is ?

Also i get one vector minus another gets the distance and direction Goal - current position but why is the point velocity attribute used and what controls the speed this happens ?

David

Here is the Hip file for the delete sop problem, where one point doesn't get its particle but instead overlaps on another. If you could tell me why this is ?

Thanks

David

particleGoals_Delete sop.hipnc

Edited June 20, 2011 by Lyonz