Cineversity

hello!
I’m playing around with vector fields and I noticed if I quantize a random field it only seems to quantize the field strength, not the angle, maybe I’m not setting it up correctly.. Is there another method to quantize vector directions? I feel like I missed out on some training on vector fields, is there some good learning examples on how to work with vector/direction fields?

Hi Nick,

You have the Sine and the Direction icon active on each, I assume. I believe you checked with the step size of one and got more than the two expected directions.
Let us stick for a moment with the step size of one. This means out of a large number of angles for each axis, we expect to get two (let name them a and b).

We have three-axis for each object. Each axis receives a random value. Now we split this in either value a or b. Which means, we get for axis R.H perhaps a, for R.B again a and for R.P this time b.
Which would result in aaa, aab, abb, bbb, baa, bba, bab, and aba. (two to the power of three = eight). So we have with the Quantize setting of just one, already eight different directions.

Any number greater than one will result in an even greater number of directions of the Clones, even Quantized. A level of two steps would result in 3^3=27, or three steps 4^3=64. I think 64 is already more difficult to differentiate from a non-quantized field.

If I like to quantize clone directions, the Fields are not needed for that, this can be done with Effectors directly. Fields work like an area (hence the name) and on direction fades into the neighbor. Yes, one could adjust this, but it might be an effort to keep the borders in sync while not having a clone sitting exactly on such a border (if that is avoidable or even produce at all).

Let me know what your target is, and I might get a better idea about the options.

Cheers

I was trying to set up something similar to this https://twitter.com/RuslanFozars/status/1289256391987740672?s=20
My plan was to try to use it with moSpline which allows use of field force and deforms the spline based on a vector field, but maybe thats not how it was done, very interesting!

Thanks for this information, Nick.

Please have a look a the file …_MGmf_02.c4d first. I think that is very close to where you want to go. It is below in the Example files.

That will be all if the result is all you are after. I got that you explore this part of the application at the moment:

To provide a better idea about all of this, I have included some files to explore (…_MGff_01.c4d to 31.c4d). It will give some answers that are better explored by messing with the files. However, some info about:

The MoSpline is in the Field, and it can adapt to a certain degree to the Field.
All the little lines that show the direction of the Field are not the Field itself. The Field would be more a gradient, a flow of all these lines.
These little lines are not islands that can be quantized in a certain amount of steps.

Let us pretend for a moment the lines by itself would be the direction and half the space to the next neighbor line, the direction of the line would not be parallel in that individual space.
From those lines to the neighbor would be an interpolation. Again, the lines are graphical results of the Field, they are not the Field itself.

Here are some examples:
https://www.amazon.com/clouddrive/share/0DmmC0LT0VCjd3PAF3sDMwC1hpmgr7zThEEyPNrp3vm

All the best

P.S.: Preview clip
https://www.amazon.com/clouddrive/share/ybEYhQp7UQF4Cb2E9mxi6dMSbcDGlzh5zJTlZDApOL6

… and I could resist to play a little bit with it:
https://youtu.be/dkmM0Ep2LYo
(This is my YouTube site, and name confirmed)
File is here
https://www.amazon.com/clouddrive/share/261Ycs01AMYmUElOHoSvddCgWLlGRbOsxAeTgkhyHpP

Image: Field Masks