Variable ScaleZ of Instances

Hello there,

I’m working with particles and trying to change the scaleZ (SZ) of the Instances.
My idea is to change the length of a Tube SOP, based on a Noise CHOP.

When I calculate the life fraction of the particle and add to SX, SY and SZ, the scale works just fine. But when I put the noise channel (that was scaled from -1 to 1, to 1 to 10), the way the Tubes are rendered change, but not the scaling.

How can I make Tube Instances have different lengths?

I’m attaching an image and the .TOE file.

Thanks in advance!

instance_scale.toe (6.06 KB)

There’s something happening here with your rotate to vector operation. If I clear out those pars on the geo then the scaling works correctly.

You can always use an object CHOP to get the bearing info for your instances as a work around and use that in the rx ry rz pars. The technique here would be to use a feedback chop to capture the previous frame’s location data and then use that with an object chop in order to calculate bearing. Here’s an attached toe that does what I think you’re after.

This one might be worth posting in bugs? It seems like the rotate to vector set of pars is behaving a little strangely.
instance_scale_using_feedback.toe (12.9 KB)

Thank you, raganmd

I didn’t know this technique of using Feedback with Object CHOP.

But even when I change the range of the Math CHOP wheres Noise is connected, it seems that every instance get the same maximum scale… is there another technique to get very different length/scale for each instance?

If I use a box SOP instead of a cone I see a pretty good diversity of sizes, so I think the scaling is working correctly. I think the harder bit is that the nature of that cone shape makes it hard to really see a difference in size given that it’s stretched along an axis - making for a longer taper as well as a longer body.

Something to keep in mind is that your noise is longer than your set of particles - so you might also be seeing some repetition in some of those instances. You can manually set the noise length to match your particles, or you can use “Shift to First Interval” on your merge chop to clamp the number of instances.

If you’re after making longer bodies with more consistent front cone head shapes it might be worth doing two different shapes that are instanced? One for the body and one for the head - that way you can still get a variation in body length with a uniform head shape. That’s a little stranger to set up but would work around part of this challenge.

thanks matthew! I will use box for have more variety of shapes.
I often watch your tutorials, they are great!