sono

I am using jit.gl.multiple to modulate position, scale and rotations of an x amount of gridshapes.

Also, I am applying position transformations on the jit.gl.multiple itself, resulting in offsets for all the gridshapes at once.

What I now want to retrieve, is the position data of the individual gridshapes AFTER the transformations inside jit.gl.multiple.

I am using jit.gl.multiple to modulate position, scale and rotations of an x amount of gridshapes.
Also, I am applying position transformations on the jit.gl.multiple itself, resulting in offsets for all the gridshapes at once.
What I now want to retrieve, is the position data of the individual gridshapes AFTER the transformations inside jit.gl.multiple.

jit-gl-multiple-get-individual-gridshape-properties

if you have to do it manually, jit.gen might help, but the operation may be not so complicated (i'm not sure ?... maybe depends if you use a lot of rotation on the jit.gl.multiple in fact)

the screentoworld message to jit.gl.render miiight also help ? if you are reduced to manually finding the position of each gridshape...

if you have to do it manually, jit.gen might help, but the operation may be not so complicated (i'm not sure ?... maybe depends if you use a lot of rotation on the jit.gl.multiple in fact)
the screentoworld message to jit.gl.render miiight also help ? if you are reduced to manually finding the position of each gridshape...


here's a patch demonstrating vichug's suggestion:

This is a nifty way to do this indeed. However, I am also applying rotations, but I imagine it should also be possible to implement this in the jit.gen with rotation matrix transformations? I know the math, and could fiddle with it, but I imagine this will have been done before. Will do some research on that.

And where do the instances of the multiple gridshapes exist anyway? I mean, somewhere the transformed positions should be available right, as Jitter knows where and how to draw them. This gen trick is copying calculations that are done already. If it works it works, but it seems like double work to me.

see my reply here for how to utilize qrot for rotations: 

https://cycling74.com/forums/jit-gl-node-like-nested-3d-transformations-on-matrix-cpu-side

see my reply here for how to utilize qrot for rotations: https://cycling74.com/forums/jit-gl-node-like-nested-3d-transformations-on-matrix-cpu-side


Then, my next question also involves jit.gl.multiple, but then combined with jit.gl.text3d.

I want to have multiple text3d's in the space, but with varied texts. Now I have it working with multiple instants but the text is set at the associated jit.gl.text3d object, making the texts the same for every instant.

I found out that text3d takes asccii from a jit.matrix, but I don't know if and how it is possible to create a jit.gl.multiple glparam for this as it is not a text3d attribute in this way (the prefix 'text' doesn't work).

jit.gl.text is no compatible with jit.gl.multiple in this way.

you will have to use something like poly~ or javascript for this kind of thing.

jit.gl.multiple: get individual gridshape properties