View Full Version : particle orientation & lock\bond

01 January 2011, 05:27 PM
my intent is: particles that follow locked surface over its tangents.
seems to be that locked surface not work like I imagine...:cry:
arrows works fine only on the floor.

if someone have a bit of free time, I attach my scene...thanks.

01 January 2011, 08:56 PM
what max version? won't open in 2010. :sad: (don't have 2011 where I am ATM)

01 January 2011, 10:22 AM
sorry johnny,
I always forget that you are the most available particles paladin but uses the 2010 version :)
2010 version uploaded...try again please.

another mistake for me is why arrows don't proceede over the first edge if I don't active a turbolence force.

01 January 2011, 02:52 AM
Hmm, well, that is about as good as I can get it without spending an exorbitant amount of time on it :shrug:

I think softening that edge coming into the vent (?, I am guessing that is what it is) may help, I had to move it a little, there is just too much deviation to get the particles to align with the speed, well maybe one of the other box#3 boys (or girls :) ) can figure that one out. Combining quats and matrices is something that I am not that familiar with.

In your original scene (it was exploding on me, for some reason it didn't really like the shape instance in the global, it usually doesn't mind) You can turn off the forces in the second event and set the Speed by Surface to Continuous. The Rotation speed space follow fails simply because it is doing what it is supposed too, which is point in the direction of speed BUT the particle is being adjusted by lock and bond and speed by surface.

So unless you can figure how to align to the normal (which is what I did in the example) AND figure out how to track a point -> point at an interval of time to get the heading, then combine the two sets of values, I think it is as close as you are going to get. Be cool to see somebody figure that one out, I would like to learn how it is done :)

01 January 2011, 10:45 AM
mmm...first of all, thanks.

your file is an interesting solution but your data operator demoralize me...I'm not a box#3 user (one of my thousands limitations :(...).
open it I see something similar to the kennedy family tree :).
however, thanks for your search for the solution starts from here now.
if solve in a clean way, I post results...

01 January 2011, 04:42 PM
it's me again,
johnny: I tryied to study your file and other similar solution that use a simplifyed version data operator tree (the normals of object I choose become the Z axis of my instanced shapes...right?).
understanding the process is yet a step ahead but I noticed a strange effect on the upper plane when arrows rotate andt proceed like a car in reverse gear!
this is also what happens in your file.

01 January 2011, 08:24 PM
Yep I noticed that too and I haven't figured out a way to compensate for that yet. Rotations are difficult, for me anyway. If I come up with something I will be sure to let you know. There is also the problem, when a particle passes to a new set of normals that it can have a minor twist to it before it aligns with the new normal.

01 January 2011, 10:42 AM
There is also the problem, when a particle passes to a new set of normals that it can have a minor twist to it before it aligns with the new normal.

lol...a like TRON style...perfect for this period.

my first thought: to avoid this, every instanced shape deeds to be oriented by 2 point (those that define the direction of the motion vector) at a distance equal to the length of the mesh.
a little rig.

or a drag rotational force
or a "stop gradually" operator

ok, let's start from monday.

CGTalk Moderation
01 January 2011, 10:42 AM
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.