Kill nParticles on selected collisions

[Janak Thakker]

Hey guys,

My mate here is trying to figure out how to kill stray nParticles, any ideas?

Basically it's the guys running off after hitting/missing one collision object. Then it would be neat if they could die when colliding with specific objects or something like that? (tried the coliision events thingy but can't get it to do anything but kill 'em on all collisions). Missing something trivial here? :-)

Thanks,
Janak

[Florian Koebisch]

Ok,

maybe this is a quick and dirty solution, but you might be giving it a try. Probably you can place some fields that are prepared like this:

http://www.sigillarium.com/blog/lang/en/297/

or maybe you connect certain fields to the particles after impact.

Best,

Flo

2011/9/13 Janak Thakker <ja...@fatlittleindianboy.com>

[Steve Davy]

I'm not entirely sure if it's the same workflow with nParticles, but with classic particles you can trigger per-particle events by using the collisionGeometryIndex attr. Whether that would work for you depends on what you're trying to do.... it also involves writing a simple expression.

---

Date: Tue, 13 Sep 2011 16:11:27 +0200
Subject: [maya_he3d] Kill nParticles on selected collisions
From: ja...@fatlittleindianboy.com
To: maya...@googlegroups.com

[Cosku Ozdemir]

I also tried this today, looks like nParticles return "0" as
collisionGeometryIndex for all n-passive-colliders.

I haven't done any work with nParticles, so I'd also like to learn
what "the new way" is :-)
Cheers

-Cosku Ozdemir
Visual Effects Supervisor, Robotika Films
Maçka Talim Yeri Sok. 8/1, Maçka, İstanbul / Turkey
t: 2122277230   ///   f: 2122277231
www.robotika.com.tr /// co...@robotika.com.tr

[mgw]

I asked this question about a year ago (maybe a bit longer) and have not
got an answer.

my conclusion was to go back and stick with the old particles.

if there is a way to query the collision-geometry index with nparticles
I also would like to know.

cheers
michael

[Janak Thakker]

Thanks for the input guys.

Skickat från min iPhone

[Janak Thakker]

Damn iPhone...anyways, thanks for the input. I'll see if we can use any of those ideas. I think we might just brute force it tho' as time is running out.

If Autodesk would want to do something really nice for the Maya community I think they should steal Max particle system now that they own everything...haven't used it for ages but 5/6 years ago it already kicked Mayas particles in the groin, especially in terms use friendliness...

Skickat från min iPhone

[Florian Koebisch]

Maya's particle System is very old- school and it was a delight if they just enhanced it in a manner that it has more event- driven capabilities. What I don't get is, according to Autodesk's docs nParticles are inherited from standard particles (in the sense of inhertinace in OOP). So nParticles should be able to access all Methods of Standard Particles plus bringing in their own, otherwise this is strange Software Design.

2011/9/14 Janak Thakker <thakke...@gmail.com>

[mgw]

well, then you should call it strange software design!

with nparticles all the collisions and dynamics are based on the nucleus solver which originally was the framework for ncloth. I guess that's why collision based stuff works different than the old particle system.

[Duncan Brinsmead]

Yes, unfortunately the nucleus collisions are handled quite differently than the old system, so it is more problematic getting collision object ids. (There is a way of getting the object currently but it is very convoluted and involves using callbacks)  The old system has an array on the particle node with input connections from all the colliding objects. The id is the index number in this array. With nucleus the nCloth and nRigid nodes connect to two different arrays on the nucleus node, which makes it trickier to come up with a good id. (perhaps nCloth nodes could have negative ids or offset by some value) There are internal ids on nucleus nodes but these are created at runtime, so they change between sessions and are not usable for this purpose. One thought is to have a mel call one can use to either return the nCloth,  nRigid or nParticle node name from an id and a particle system name.(you need the particle name because there may be multiple nucleus nodes) Alternately we could return the name of the cloth output mesh or nRigid mesh.

 

At any rate this is something that should have been implemented in the first nParticle release but has kept getting pushed back. Any feedback as to what you would like in the way of ids and the interface would be helpful.

 

Duncan

[mgw]

hi duncan,

well I guess what we all would like to see is a modular node based approach to particles and dynamics like we see nowadays in all 3d packages (houdini, 3dsmax (pflow and thinking particles), xsi(ICE), etc.).

I think it's save to say that maya is the only package which doesn't offer this approach which is very strange because maya is a node based system in the first place so why not reveal those nodes in a user friendly way and interface.

just my 2 cents.

michael

[Carlos Rico 3D]

Absolutely!

2011/9/16 mgw <mige...@googlemail.com>

--
Carlos Rico Adega
-------------------------------------
LinkedIn
carlos....@gmail.com