Daan

I am working on a big project right now and am in the last (optimization) phase. As I was putting some parts of my code in [poly~] objects I realized [poly~] will stop its DSP processing if you connect any of its audio outs (even with lots of DSP stuff in between) to one of its audio ins. No direct feedback here, so I don't understand. Doing the same thing with ordinary patchers/bpatchers seem to work, as long as you don't create direct feedback loops, and also don't connect ins and outs directly (a DSP object like [+~] needs to be in between). With [poly~] you need to use [send~] and [receive~] pairs but this introduces a vector size delay, something that will ruin the mechanics of my patch.

To have something to base discussions on, I quickly constructed a collection of patches (please see the screenshot in the attachment).

A friend of mine told me that -yes- with the [poly~] object (see situation 5 in the screenshot) I can't do what I want to do but he said with an ordinary patcher or a bpatcher (situation 4) it shouldn't work either.

1). Is there anything wrong with a situation like number 4 or 8 in my screenshot? Basically routing audio back and forth into and from a patcher/bpatcher. It doesn't cause delays/noticeable issues, the patch I am working on seems to work perfectly fine. Isn't it just behaving like it would when it wouldn't be encapsulated into a patcher/bpatcher?

2). Why does MAX prevents you from directly connecting an audio out to an audio in of a bpatcher/patcher?

3). So what is with the [poly~] object and not being able to feed audio back in, like I can do with ordinary patchers/bpatchers.

DifferentSituations.PNG

I am working on a big project right now and am in the last (optimization) phase. As I was putting some parts of my code in [poly~] objects I realized  [poly~] will stop its DSP processing if you connect any of its audio outs (even with lots of DSP stuff in between) to one of its audio ins. No direct feedback here, so I don't understand.  Doing the same thing with ordinary patchers/bpatchers seem to work, as long as you don't create direct feedback loops, and also don't connect ins and outs directly (a DSP object like [+~] needs to be in between). With [poly~] you need to use [send~] and [receive~] pairs but this introduces a vector size delay, something that will ruin the mechanics of my patch.

A friend of mine told me that -yes- with the [poly~]  object (see situation 5 in the screenshot) I can't do what I want to do but he said with an ordinary patcher or a bpatcher (situation 4) it shouldn't work either. 

1). Is there anything wrong with a situation like number 4 or 8 in my screenshot? Basically routing audio back and forth into and from a patcher/bpatcher. It doesn't cause delays/noticeable issues, the patch I am working on seems to work perfectly fine. Isn't it just behaving like it would when it wouldn't be encapsulated into a patcher/bpatcher?
2). Why does MAX prevents you from directly connecting an audio out to an audio in of a bpatcher/patcher?
3). So what is with the [poly~] object and not being able to feed audio back in, like I can do with ordinary patchers/bpatchers.

Hi people!
I am working on a big project right now and am in the last (optimization) phase. As I was putting some parts of my code in [poly~

Confusion related to routing audio back into patchers/bpatchers/poly~