generating new value for each copy per point($PT) ?

[Guest ravin]

In copy SOP i am declaring a stamp variable with value "rand($PT)-0.5" . When I copy the sphere to the copy SOP, each copy gets the same value which is expected. I want each copy to extract a new value from this expression. Can anyone suggest a way to do that. Basically I want to extract a new value for each copy/$PT and arrange 'n' copies around each point.

Hope i made it clear.

thanks,

[Macha]

I'm not sure I understand. That should work.

Is this what you mean?:

stamp.hipnc

[Guest ravin]

I'm not sure I understand. That should work.

Is this what you mean?:

Hey Macha,

I am stuck in copying points to particle problem. In the setup you gave, copy spheres to particles. So if u want to copy 'n' spheres to each particle, you want set the stamp function in such a way that each time each copy of sphere queries the stamp value, it gets a new value. So what I am wanting to do is, generate a random vector for each point to displace the sphere around that point and do this for each copy to that point.

In your setup things are simple coz you know there is one copy for each point, so you get a new value but imagine if you want 10 copies for each point, so now you will have those 10 copies getting the same random value and hence even if I displace them in translate, then will overlap coz they are receiving the same value from that particular point.

Hope I didn't confuse you,

Also I am attaching your file with little modification where I have added a popnet. Now the popnet should release 2 particles and I want to copy say 3 spheres to each particle. So I should see 2*3 spheres moving away from centre. The challenge is to arrange 3 copies around each particle in a random order such that they appear to be another set of particles.

thanks,

stamp.hipnc

[Macha]

Ah, I see. Why not use two copy stamps then, the first one with a sphere perhaps, and then use that as our source for the second copy stamp.

stamp.hipnc

Edited August 8, 2010 by Macha

[Guest ravin]

Ah, I see. Why not use two copy stamps then, the first one with a sphere perhaps, and then use that as our source for the second copy stamp.

Hey ya ...that's a really smart way of doing it. To break the uniform look I can always use some other nonuniform geometry or add some random noise or something. This thing done, my next challenge is to render dust/cloud out of these particles. I know I had posted the topic earlier and had received lot of replies but I think they were a little advanced topics for me. I have never done particle rendering as volumes so how can I achieve this with a simple setup? I have two sets of particles, one which I want to render as dust and another as clouds with that puffy look. Sprites would be one way but I want shadows too and I want to do it the volumetric way. Just a simple file with the setup and explanation would help me a lot.

thanks for the help, i really appreciate it.

[Nerox]

Hey ya ...that's a really smart way of doing it. To break the uniform look I can always use some other nonuniform geometry or add some random noise or something. This thing done, my next challenge is to render dust/cloud out of these particles. I know I had posted the topic earlier and had received lot of replies but I think they were a little advanced topics for me. I have never done particle rendering as volumes so how can I achieve this with a simple setup? I have two sets of particles, one which I want to render as dust and another as clouds with that puffy look. Sprites would be one way but I want shadows too and I want to do it the volumetric way. Just a simple file with the setup and explanation would help me a lot.

thanks for the help, i really appreciate it.

Hi Ravin,

The Mountain SOP is a nice one to randomize your geometry a bit. Especially if you use the stamp() function to change the frequency per object for instance.

[Guest ravin]

Hi Ravin,

The Mountain SOP is a nice one to randomize your geometry a bit. Especially if you use the stamp() function to change the frequency per object for instance.

I will surely take a look at that.

thanks