Jump to content


Photo

C4D Thinking Particles - Collision Issue


  • Please log in to reply
8 replies to this topic

#1 Hamilton

Hamilton

    MoGraph Superstar

  • Members
  • 109 posts
  • Gender:Male
  • Location:Victoria BC, Canada
  • Interests:SKATE, BEER, MOTION, PHOTOGRAPHY -- now repeat 24/7

Posted 07 January 2013 - 05:57 PM

Hey,
I seem to be having a problem with a pDeflector that isn't recognizing a particle collision. I'm using a data channel to store Integer info and attach the particle to a cloner object with dynamics. I thought I heard somewhere that when you hand particle information over to mograph that the xpresso for TP doesn't calculate the same. Not sure and am stumped. 

Would someone please take a quick look and see if you spot something?

Thanks,
/Ryan

 

http://www.ryanhamil...e_to_Cloner.zip


Edited by Hamilton, 07 January 2013 - 05:58 PM.


#2 ChrisC

ChrisC

    MoGraph Demi-god

  • Members
  • 538 posts
  • Gender:Male
  • Location:Brighton, UK

Posted 08 January 2013 - 04:03 PM

Hey Ryan - trouble is I think if that you're controlling the particle on every frame, by sticking it onto the clone, then it's not having a chance to collide with anything - it's one side of the surface on one frame, and the other side on the next frame, but has no 'knowledge' that it collided with anything; it's just doing what it's told.

 

Instead of starting with particles to check for collisions, have a look at the Dynamics Collision node - should do what you're after.


Edited by ChrisC, 08 January 2013 - 04:04 PM.


#3 Hamilton

Hamilton

    MoGraph Superstar

  • Members
  • 109 posts
  • Gender:Male
  • Location:Victoria BC, Canada
  • Interests:SKATE, BEER, MOTION, PHOTOGRAPHY -- now repeat 24/7

Posted 08 January 2013 - 04:28 PM

Thanks for that Chris,

I have been trying the collision node also while waiting for replies but it detects the collision of the disk geometry at multiple vertex points at once and gets confused as to where to place the new pStorm. Also I would just like to have one pstorm emit at the center of the object. My file is in a bit of a mess right now from trying different scenarios. I'll clean up the xpresso and repost with the collision node wired up.

/Thanks



#4 Srek

Srek

    Maxon

  • Members
  • 361 posts
  • Gender:Male
  • Location:Friedrichsdorf, Germany
  • Interests:Arduino, RepRap, Science Fiction

Posted 09 January 2013 - 08:00 AM

Good catch Chris, yes that would make the collision node ineffective, basically all the particle nodes that concern particle movement realy only work if you leave the moving to the particle system. Changing velocities is the correct way to handle things, changing/setting the psotion isn't.

Instead of testing for a real collision you could check the distance between the particle and the colliders geometry. I have created a preset (X-Pool>System Presets>Advanced>GetNearestPoint) that can give you the vertex nearest to a given position. Check if the resulting difference vector between that vertex and your particle position has a positive Y component and at the same time check if the clones velocity vector also has a positive Y component (the Dynamic Body State node can give you this).

By mixing three different systems (MoGraph, Dynamics, TP) you made this a rather difficult job, you might be easier off using only MoGraph and Dynamics, or only TP.



#5 Hamilton

Hamilton

    MoGraph Superstar

  • Members
  • 109 posts
  • Gender:Male
  • Location:Victoria BC, Canada
  • Interests:SKATE, BEER, MOTION, PHOTOGRAPHY -- now repeat 24/7

Posted 09 January 2013 - 03:12 PM

Cheers Srek. I think I understand what you & Chris described. Because the particle is updating every frame to see where the cloner object is, and then placing it self there, it doesn't actually have any movement of its own to warrant a pDeflector collision. ( Was that correct? )  

Also I couldn't get Dynamic Collision node to do anything so won't bother reposting that setup.

I will try your nearest point advice though. It sounds really useful and I didn't know about it.

 

Finally Srek you are right in saying that choosing this path was a difficult option. I wanted to make a setup similar to a couple parts in the Nano Bounce spot, but with beer bottle caps.

At 20sec and again at 26sec it looks like a good reason to try it as a TP setup. I figured drop an object with Dynamics - attach a particle to it - on collision fire off some new particles -  let those new particles have dynamics and when they bounce make them trigger and new bunch again. I pictured it being rather simple but was quickly mistaken.

Thanks for your help.


Edited by Hamilton, 09 January 2013 - 03:13 PM.


#6 the_Monkey

the_Monkey

    simian

  • Members
  • 2,011 posts
  • Gender:Male
  • Location:Brooklyn, NY

Posted 11 January 2013 - 12:12 AM

Maybe something like this?

 

-m



#7 douwe

douwe

    MoGraph Megastar

  • Members
  • 260 posts
  • Gender:Male
  • Location:Brussels

Posted 11 January 2013 - 01:45 AM

Maybe something like this?

 

-m

 

now, look at who's still relevant as fuck


www.c4dlounge.eu : dutch c4d community (Belgium / Netherlands)
//////////////////////////////////// douwe on vimeo

#8 Hamilton

Hamilton

    MoGraph Superstar

  • Members
  • 109 posts
  • Gender:Male
  • Location:Victoria BC, Canada
  • Interests:SKATE, BEER, MOTION, PHOTOGRAPHY -- now repeat 24/7

Posted 11 January 2013 - 03:28 AM

Very cool Monkey. Thats a nifty xpresso setup.

Cheers.



#9 Srek

Srek

    Maxon

  • Members
  • 361 posts
  • Gender:Male
  • Location:Friedrichsdorf, Germany
  • Interests:Arduino, RepRap, Science Fiction

Posted 11 January 2013 - 09:09 AM

Sorry, i can't watch the video due to stupid german copyright issues.

You are right about there beeing no real movement.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users