fraction

i m aim to do create a stereo with representation, represented by a main central point (mesh) and two side small spheres.

This works fine in 2D mode, where you only work with azimuth and distance for the main point. Right now i m simply applying an offset calculation to the main point azimuth to position the side spheres. But this is where i m stuck. When moving the elevation, it obviously changes the offset angle, as the Z position affects the initial angle. So that technics doesnt work for 3d.

I m not sure how to address and resolve this problem in jitter - maybe using jit.anim.nod hierarchy? 

Please have a look on the patch. you will need spat5.converter external.

Need help to represent a specific mesh behavior (using polar system)

need-help-to-represent-a-specific-mesh-behavior-using-polar-system

this should be close to what you need, as far as I understood. 

I'm using jit.gen to address vector math, mainly for clarity, but it is easily replicable by implementing "normalize" and "cross" with standard Max objects.

If you need instead to keep the speakers at the same distance from the center and control the width expressing an angle in degrees, you can try this other method:

Hi Matteo!  many thanks for your help - i tested quickly, but didnt have the time to implement it yet and adapt it to my workflow. I think i ll do this tomorrow. There might be reverse thingy somewhere, as the arc size should increase (distance between secondary sphere) with the distance and not the other way around (which seems the case with your path?)

 similar as the behavior bellow (in a 2D mgraphics/ on the left).

i m wondering what is actually the most clever way to represent such behavior in gl world? maybe such an arc as well instead of two sphere?  what do you think? (i know this goes beyond the little help haha)

oh but this second version seems to work better! 

for some reason, the polar to cartesian that you made in gen has a different output than the spat5.convert that im using. so the result is really uncoherent. I have fixed your first jit.gen that converts to xyz just to compare with the spat5 object . Not sure if its a mistake or a different projection system (?)

now i need to give a look to the second jit.gen... because of course now the result doesnt work in that reference

yes it's been a while. I put this issue on the side as this wasn't on high priority  stuff to implement and i was stuck trying to figure out. But as i got back to it, starting again from what you posted, while i was focusing in the jit.gen to try to understand the issue, it was "simply" that i have a world projection rotated - so object have a different position in worldpos and pos. To use your help (thx again) i probably have to convert my data, in worldpos first make the calculation, then convert back in my projection. Just a matter of packing at the end. It seems to work now.