< Back to IRCAM Forum

Order of execution startup

Hi
I have some problems with my patch. The problem is to start the patch easy. I have to load my file with speaker positions to speakerconfig multiple times before I get it to work. Or I load it, send a bang to oper and hope that it will work. I tried various ways to do it but can’t seem to figure out the logic.
Is there a certain order that we should load different things?

Once I get it to work it sounds very good. But it would be nice with a more stable setup. Anyone have an idea on how to fix it?

Spat 29 250113.maxpat (452.7 KB)
29Frisk.txt (3.5 KB)

Hi,

First, have a look at “spat5.tuto-osc-3.maxpat” which provides information about initialisation order.
As mentioned there, it is recommended not to mix multiple initialisation strategies. It seems like that’s what happens in your patch.

You have the speaker coordinates typed in the spat5.oper box, using @initwith. Supposedly, that should be enough for proper init.

However, your patch also contains spat5.speaker.config which is triggered multiple times: from loadbang, and from the [live.menu 29] which is “initial enabled”. So that causes some confusion (and multiple inits with actually different speaker coordinates).
When spat5.speaker.config receives the “/speaker/number 29” message, it first initialises with a default speaker layout containing 29 speakers. This default setup is planar, which triggers error messages from your vbap3d renderer.
I would recommend that you simplify the patch by removing/disconnecting some of these elements.
Feel free to use [spat5.osc.print] to try to see which coordinate messages are actually triggered in the various places.

Meanwhile, I’m also investigating if things can be improved on the spat5.oper / spat5.spat~ side. Maybe with more user friendly debug information…

Hope this helps,
T.

Thank you! I think I added more “loadbang”:s because I had issues before. I think I got it to work now, have to try in dome this weekend