PDA

View Full Version : Instancing: nodal scaling with particle age



jeric_synergy
09-24-2012, 11:29 PM
regardez le zhjeeeeeypeg:

Using Instancing, I'm tryiing to get the instanced items, which are generated by particle, to shrink to nothing as the particles age.

I've tried both the color output of the gradient (white to black) and the alpha of the gradient (opaque to transparent), but neither seem to affect the instance size at all.

What's going wrong here? Besides me?
108068

AlexxelA
09-25-2012, 01:24 AM
what´s your global instance scale settings?
uniform or random? ... try it with uniform or with scale min, max inside node and look if something happen!

Red_Oddity
09-25-2012, 01:58 AM
Make sure that anything you want to set via the node editor is set to Uniform in the Instancer settings tabs (i often find this is what makes things all of a sudden work.)
Also, make sure you check 'Output Size' in the particle emitter, otherwise it will never output any scale information.

jeric_synergy
09-25-2012, 01:31 PM
Make sure that anything you want to set via the node editor is set to Uniform in the Instancer settings tabs (i often find this is what makes things all of a sudden work.)
Also, make sure you check 'Output Size' in the particle emitter, otherwise it will never output any scale information.
Thanks, fellas: UNIFORM in the SCALE TAB did indeed seem to be key to making it work, although I haven't quite got it working yet.

Now, the the instanced objects (mediated by particles) do indeed shrink with age, but they all shrink simultaneously instead of with their associated particle's age. So, somewhere along the line it must be possible to address the specific instance with the specific particle's age, which isn't happening yet.

Thanks for your help! :thumbsup:

Current Scene attached.108080

Red_Oddity
09-25-2012, 04:10 PM
Thanks, fellas: UNIFORM in the SCALE TAB did indeed seem to be key to making it work, although I haven't quite got it working yet.

Now, the the instanced objects (mediated by particles) do indeed shrink with age, but they all shrink simultaneously instead of with their associated particle's age. So, somewhere along the line it must be possible to address the specific instance with the specific particle's age, which isn't happening yet.

Thanks for your help! :thumbsup:

Current Scene attached.108080

That's because the build in ParticleInfo node seems to be bugged (or at least i've never been able to get ANY reliable data from it with particles (which, oddly enough, is the entire function of said node), it seems to just always take the data from the very first particle created on the emitter.)
If you use DP Kit's Particle Info node however, you can use the Relative Age output and it return what you would expect. (Don' forget to set the Emitter and to keep Particle set to Nearest on the Particle Info settings.)

jeric_synergy
09-25-2012, 07:07 PM
That's because the build in ParticleInfo node seems to be bugged (or at least i've never been able to get ANY reliable data from it with particles (which, oddly enough, is the entire function of said node), it seems to just always take the data from the very first particle created on the emitter.)
Well, that's 'not good'. :devil:

It's been fogged, right?


If you use DP Kit's Particle Info node however, you can use the Relative Age output and it return what you would expect. (Don' forget to set the Emitter and to keep Particle set to Nearest on the Particle Info settings.)
I kept wondering "why 'Relative Age'??" and puzzling over other behaviour, as idiots do when they don't read the dox, when I realized 'Relative Age' must be giving a percentage of the particles' set lifetime. In a way, this works better than what I was thinking, as one could adjust the shrinking behaviour totally w/the particles' lifetime parameter.

So, thanks to Red_Oddity, it works! I still want to futz around with it and get it to work with specific particles' "age-in-frames after generation", but at least I understand what's going on here.

And we all stand warned of the non-working state of the native Particle Info node. :mad:

dwburman
09-26-2012, 10:07 PM
Yup. In LW, 'relative particle age' is each individual particle's lifespan and 'particle age' is kind of the group age based on (I guess) the first particle's lifespan.

It's not the best way of labeling it.

<shamelessSelfPromotion>
I demonstrate the differences between the two types (as well as nearly all the other Gradient input parameters found throughout lightwave) in my gradients tutorial set. (See my sig for the link) :)
</shamelessSelfPromotion>

jeric_synergy
09-26-2012, 10:43 PM
Yup. In LW, 'relative particle age' is each individual particle's lifespan and 'particle age' is kind of the group age based on (I guess) the first particle's lifespan.

It's not the best way of labeling it.

<shamelessSelfPromotion>
I demonstrate the differences between the two types (as well as nearly all the other Gradient input parameters found throughout lightwave) in my gradients tutorial set. (See my sig for the link) :)
</shamelessSelfPromotion>
I'm happy to repeat your ssp. :)

Maybe you can enlighten me: it seems the RELATIVE AGE parameter comes out as a PERCENTAGE: how would one acquire the absolute particle age?

Red_Oddity
09-27-2012, 07:53 AM
With DP Kit Particle Instancer the Age attribute returns the age in time i believe (1 / fps).

dwburman
09-28-2012, 10:26 AM
Actually, now that you mention it, particle age might be based on absolute time rather than the 1st particles lifespan. I'll have to take a look at my own tutorial again. It's amazing how one can learn something and then have to relearn it a year later. :)

Relative particle age is a percentage because you can +/- a particle's lifespan, so some particles will live longer than others.

EDIT:
Okay, I rewatched part of the gradients video I made when working out how particle age worked and it looks like particle age is locked to the base particle lifespan, probably starting with the first particle emitted. I uploaded an excerpt from the HyperVoxels tutorial where I show the difference between the two types. Interestingly enough, I think this is the case where HVs uses frame numbers for particle age and Pixie Dust uses seconds for particle age. Another fun inconsistency in our favorite 3D app. :)

http://youtu.be/7IjAdhu6-w8

jeric_synergy
09-28-2012, 12:27 PM
EDIT: Interestingly enough, I think this is the case where HVs uses frame numbers for particle age and Pixie Dust uses seconds for particle age. Another fun inconsistency in our favorite 3D app. :)
Grrrrrrrrr. :devil: From a user viewpoint, LW is like a fixer-upper house that was built by 3 different contractors and sprinkled with maddening inconsistencies, like some light switches are upside down, and sometimes the Hot water is on the right, and some taps turn the the opposite way.....

Each of which becomes a speed bump to creative flow. And their aggregate effect is user frustration.

They are slowly being addressed, but there sure is a lot of them. I don't envy the devs having to go 'round and tidy them up years after the fact.

dwburman
09-28-2012, 07:09 PM
Grrrrrrrrr. :devil: From a user viewpoint, LW is like a fixer-upper house that was built by 3 different contractors and sprinkled with maddening inconsistencies, like some light switches are upside down, and sometimes the Hot water is on the right, and some taps turn the the opposite way.....

Each of which becomes a speed bump to creative flow. And their aggregate effect is user frustration.

They are slowly being addressed, but there sure is a lot of them. I don't envy the devs having to go 'round and tidy them up years after the fact.

That's a good analogy. :D

Oh, I found other inconveniences too. I reported most of them and I think at least one got fixed. There's another difference between HVs and PixieDust (can't remember which input parameter right now) where you had/have to set the key position in PD to something like a 10th or 100th of the setting for HV.

jeric_synergy
09-29-2012, 12:01 AM
That's a good analogy. :D
I should really try to make my living w/words instead of pictures. ::sigh::

(Fun fact: European/UK switches are generally 'upside down' from a US viewpoint. I wonder what the history there is.)

jeric_synergy
10-18-2012, 11:53 PM
Still confused a bit about "age" versus "relative age" (but also got a cold and that's not helping w/the puzzle).

Here's an image of DPKit Particle info node and its corresponding dox. As you can see, the dox are pretty terse, and many outputs are not referred to at all. ( "Temperature"!? ) If there are more current dox, or perhaps an article detailing each output in excruciatingly microscopic and simple detail, I'd love to hear of it.

108614

In this thread we seemed to have determined that "Age" is from the first emission of the first particle, not the age of each particle.

"Relative Age", despite Dana's and my own comments, still seems mysterious as to whether it's a % of the particular particle's, or a particular particle's absolute age, whether in frames or seconds. For a while I thought it was %, so would vary between 0 and 1, but now I'm not so sure. Messing about with InputSpy I've seen a specific particle's age DECREMENT, so I'm back to being confused. (Or my methods are questionable.)
108615

dpont
10-19-2012, 12:32 AM
...Here's an image of DPKit Particle info node and its corresponding dox. As you can see, the dox are pretty terse, and many outputs are not referred to at all. ( "Temperature"!? ) If there are more current dox, or perhaps an article detailing each output in excruciatingly microscopic and simple detail, I'd love to hear of it...

Particle Info is not an emitter, just a data provider
from the LW Particle Buffers filled by a native or a third party emitter
which may differ in their process and doesn't necessary fill all potential buffers,
in the case of a LW Emitter there's no "Temperature".
Particle Info provides only the available buffers
if they exist, for detailed documentation see the emitter itself.

Denis.

jeric_synergy
10-19-2012, 12:05 PM
Well, we both know that plenty of information in the LW dox is sketchy and/or missing.

So, my plea for links to articles/posts/blogs or whatever that would fill this info-gap still stands, just like clarification on "Relative Age".

dpont
10-19-2012, 01:02 PM
A Relative Age is an Age relative to the maximum Age at a current time,
Age divided by the maximum Age gives a percent-Age...

Denis.