PDA

View Full Version : tp particles leaking through deflector


marcorabellini
12-09-2005, 12:17 AM
This problem plagues me all the time. I'm sending out a stream of tp particles and a few of them "leak" through the deflector. Has anyone figured out a way around this or what's causing it?

I'm guessing it that the particles are moving faster than the subframe rate and don't get "caught" by the deflector.

Increasing the polygons doesn't solve it.

Is there any way to increase the subframe rate?

Thanks,
m

paulselhi
12-09-2005, 05:02 AM
Only TP for max has subframe sampling

I think there is a way to use the DT node for something like this but i really don't undersatnd it

Perhaps some sort of condition based on proximity to the deflector adjusting speed ? a friction node to act as drag ? another deflector at a slightly different postition ?

marcorabellini
12-09-2005, 02:58 PM
I've tried the "deflector layers" technique with no luck. There are sometimes some "magical" particles that refuse to be considered by the deflector. Sometimes it's just one! Madness.

I've thrown some files to Maxon to see if they can figure it out.

m

AkaKico
12-09-2005, 07:22 PM
I've tried the "deflector layers" technique with no luck. There are sometimes some "magical" particles that refuse to be considered by the deflector. Sometimes it's just one! Madness.

I've thrown some files to Maxon to see if they can figure it out.

m

Keep us posted if you get an answer if you can. I've run into the same problem before and am interested in the answer.

paulselhi
12-09-2005, 07:29 PM
you can also set up a particle destuctor for those that get through the deflector, the destructor can have a size such that sub frmae sampling would not be an issue

also maybe a range mapper to set speed to an acceptable level when within a certain distance of the defelector so the particles get "caught" by the deflector

marcorabellini
12-09-2005, 08:56 PM
you can also set up a particle destuctor for those that get through the deflector

Yes, I've tried that and sometimes that doesn't even work. I've tried multiple destructor layers outside of the deflector. Some particles are just very determined to live.

Also, what will happen is you'll get particles popping out of existence which is not necessarily what's needed.

This is definitely something I'd like solved on a more root level. The deflector should deflect properly. Then I won't be spending 14 hours adjusting parameters and still have a particle or two escaping.

I'll keep y'all informed if Maxon comes up with anything.

m

paulselhi
12-10-2005, 09:15 AM
I doubt if this would work but it may be worth trying..how about instead of a deflector using a planar gravity set to repulse the particles using fall off and /or a range mapper to control it's strength over distance

perhaps in this way you could somehow allow particles to just about reach the deflector/gravity and the settings are such that at the point of impact gravity is strong enough to repulse them, maybe some how linking the strength of the effect on each particle to its velocity

On second thoughts this would be a nightmare to set up !!!

Now if you had access to max both it's TP and Pflow systems use sub frame sampling and the TP ( not sure about the pflow) can create geometry from the particles. In this way you could sort of bake a good sim and maybe export it as FBX to C4D

Srek
12-10-2005, 09:54 AM
Hi Mark,
if you use a cube instead of a plane as killer object it will work. The speed of the particles is rather high which will sometimes result in problems when the particle is carried through the detection radius from one frame to another.
Cheers
Björn

rgwarren
12-10-2005, 05:40 PM
Box or plane I still have problems.

I've had problems with particles "sneaking" out especially if I'm animating the deflection box toward the particle direction. There seems to be some boundary condition involved that misses the deflection.

I just add a condition expression and a PDie node to take care of the offending particle.

marcorabellini
12-11-2005, 08:41 PM
...the detection radius...
- Björn

Detection radius!? Any way to access and control that? What's it based on?

Srek
12-11-2005, 08:49 PM
Detection radius!? Any way to access and control that? What's it based on?

The correct name is Offset and you can find it in the PDeflector properties. It "enlarges" the object so a collision can be detected earlier and more easily.
Cheers
Björn

marcorabellini
12-12-2005, 03:39 AM
Wonderful.

I looked this up in the manual but I still have a question or two.

Does this offset the deflect surface as a 2D plane above the surface or as a volume between the actual surface and the offset?

Does a "Both Sides" deflector have two deflection surfaces or one deflection volume? (or neither?)

Thanks,
m

Srek
12-12-2005, 06:19 AM
It defines a volume in. Two sided does the same in both directions afaik.
I found to be the "Offset in traveldirection" the most flexible and usefull.
Cheers
Björn

marcorabellini
12-12-2005, 04:14 PM
Interesting. Sounds like it defines a deflection volume between the surface and what ever the offset is. Could be an answer to my troubles (well, my deflection troubles at least ;)).

I'll experiment when I get a moment.

Thanks for the info.

Now if I could only figure out why all my computers seem to fail at the same time. I swear they are empithetic with each other. (Yes, I believe they are conspiring against me sometimes.)

m

CGTalk Moderation
12-12-2005, 04:14 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.