"translate y into x as my pmpd masses are fixed on that axis"

What I meant to say was translate 'z' into 'y' (as y is fixed).

But is this correct, can anyone confirm I'm on the right path?

The numbers that pmpd is spitting out on the z plane travel between -4 and 4 and I'm presuming that they should then be mappable as in effect the y plane for quad panning?

It works in theory if continuing the grid theory of 4 speakers on, in effect x and y, (in my head at least), though I'm still struggling to get this going in any meaningful way.

On 24 May 2013 22:56, Julian Brooks <jbeezez@gmail.com> wrote:

Hi IOhannes,

I think I was attempting to wrap my head around starting from scratch with this but actually I should see it through and carry on with Lorenzo's external, try to figure out where I'm going wrong - I have made use of it a few years ago and it worked well.

I think part of it is that Lorenzo's example patch has used [grid] (which actually makes lots of sense) but I don't have it on my install. So attempting to translate it into vanilla is a ssttretchh.

Gonna bite the bullet and build a few libs and externals.

I think as well that I was approaching this wrong. Instead of thinking in 3D I can just translate y into x as my pmpd masses are fixed on that axis (don't know why I didn't think of it earlier tbh, but there you go).

Regards,

Julian

On 24 May 2013 18:50, IOhannes zmölnig <zmoelnig@iem.at> wrote:

On 05/24/2013 04:43 PM, Julian Brooks wrote:

Hi all,

So yeah - that (the subject).

I would like to map some 3D moving sources (from Pmpd) in quad.

I had been hoping to make use of Lorenzo's [panner_1x4~], from here
https://gitorious.org/pd-panner-4ch
but I just can't wrap my head around the maths involved.

Anyone have any experience with this, or alternate strategies - the leanest
possible please as it's on the RPi.

Speakers are in a square/diamond and very close together - 25cm or so.

It doesn't have to be very accurate and I would just like to have the
impression that the five pmpd generated sound sources are moving around but
my attempts so far sound cack.

Also presuming that as we're not doing ambisonic type malarkey that the
y-axis is pretty much redundant but, well, dunno?

Had a dig around here and hurleur but nothing obvious jumped out at me.

Tips, tricks, pointers or tutorials most gladly accepted

i'm not sure what your problem is, but probaby zexy's [cart2sph] will help mapping the cartesian XYZ coordinates to spherical distance/direction coordinates. for panning you will usually use the direction coordinate(s) only.

fgmasr
IOhannes