[Sidefx-houdini-list] POP Collision question

Alex Lim list at poh-yee.com
Mon Aug 16 15:36:00 EDT 2010


Hello,

Depending on the type of scenario you're dealing with and what the 
particle is supposed to do after it collided, you may be able to get 
away by figure out a way to detect when a collision happened via series 
of SOP solution.

On the current project I am working on where everything, including the 
camera, is moving at 600-1000mph or more, using POP collision would 
simply cause more pain and trouble than by figure out an approach to 
perform simple collision detection through a post process in SOP. 
Luckily in my case, the particles mainly need to travel in a fairly 
linear fashion with no multiple bounces. Some of the key things that I 
have to think about were the position, direction, and velocity of the 
particle at previous frame vs. the current frame. From these 
information, one should be able to derive whether or not a particle is 
considered to be collided against something and precisely at what frame 
the said particle has collided.


Do keep in mind, however, these are all dependent on your situations and 
needs. There will be many situations (ie. multiple bounces) where doing 
your collision detection is simply cleaner, easier, and faster through 
POP/DOP approach.

But regardless, it was simply fun to figure out the old school ways of 
doing the effect. ;)


Hope these helped some tho.


-Alex



On 8/16/2010 11:13 AM, Alex Czetwertynski wrote:
> My particles are going pretty fast and neither oversampling or cranking up the tolerance fully solve the problem.  They do get rid of some particles, but not all.  I'll try to look into other solutions
>
> Thanks!
> Alex
>
> On Aug 14, 2010, at 12:21 AM, Sebastian H. Schmidt wrote:
>
>> Hi Alex,
>>
>> are your particles flying very fast ?
>> If so you could increase the oversampling on the pop-network but that slows
>> down the whole simulation.
>> Another thing what you could play with on the collision pop would be the
>> collision tolerance parameter.
>> That parameter defines 'Distance from geometry at which particle is
>> considered to have collided'
>>
>>
>> Seb
>>
>> On Sat, Aug 14, 2010 at 3:28 AM, Alex Czetwertynski<
>> alex at franktheplumber.com>  wrote:
>>
>>> I have a set of particles emitted from a sphere going towards another
>>> sphere where they stick and slide along the surface.  The problem I'm having
>>> is a series of rogue particles flying away from the sphere even though the
>>> majority are properly sticking on it.
>>>
>>> I tried increasing the radial force in my Point SOP at Geo level, but that
>>> doesn't seem to help...
>>>
>>> Is there anything I can do in the Collision POP to avoid this?
>>>
>>> Thanks
>>> Alex
>>> _______________________________________________
>>> Sidefx-houdini-list mailing list
>>> Sidefx-houdini-list at sidefx.com
>>> https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
>>>
>> _______________________________________________
>> Sidefx-houdini-list mailing list
>> Sidefx-houdini-list at sidefx.com
>> https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
>
> _______________________________________________
> Sidefx-houdini-list mailing list
> Sidefx-houdini-list at sidefx.com
> https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list




More information about the Sidefx-houdini-list mailing list