You could also just play sine waves thru them and then feed all signals to a set of bandpass filters. Measure the amplitude of the filtered signals and you know which one you have. But perhaps your PWM technique would use less CPU, if that's an issue.
That's an interesting idea! In my situation I'll be using low-cost microcontrollers to generate the signal so PWM would be easier.
you can just have a set of pre-loaded patches and switch between them using [switch~]. I've used that with good results in a production environment.
Huh.. interesting, but wouldn't this be combinatorially prohibitive? I.e. if I have 10 patchable outputs to 10 patchable inputs, then that's (n!) = 3628800 combinations.
Hmmm.. —t3db0t
On Jan 20, 2013, at 11:53 PM, Hans-Christoph Steiner hans@at.or.at wrote:
On 01/20/2013 12:08 PM, Tedb0t wrote:
Hi all,
Does anyone know of existing designs to mirror the state of physical patch cables in a Pd patch? In other words, I'm going to have an installation with a bunch of physical patch cables plugged in between various pods and I'd like them to control a Pd patch.
hey tedbot,
You could also just play sine waves thru them and then feed all signals to a set of bandpass filters. Measure the amplitude of the filtered signals and you know which one you have. But perhaps your PWM technique would use less CPU, if that's an issue.
So far I've been thinking I could generate different PWM signals at each unique cable source and measure them at each receiving socket, then send the graph data to the computer and use dynamic patching to control the patch. However, I'm wary of using dynamic patching in a production environment—any thoughts?
you can just have a set of pre-loaded patches and switch between them using [switch~]. I've used that with good results in a production environment.
.hc
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On 01/21/2013 02:41 PM, Tedb0t wrote:
You could also just play sine waves thru them and then feed all signals to a set of bandpass filters. Measure the amplitude of the filtered signals and you know which one you have. But perhaps your PWM technique would use less CPU, if that's an issue.
That's an interesting idea! In my situation I'll be using low-cost microcontrollers to generate the signal so PWM would be easier.
Indeed, that makes much more sense.
you can just have a set of pre-loaded patches and switch between them using [switch~]. I've used that with good results in a production environment.
Huh.. interesting, but wouldn't this be combinatorially prohibitive? I.e. if I have 10 patchable outputs to 10 patchable inputs, then that's (n!) = 3628800 combinations.
I was dealing with < 10 different options. You can treat the patches as modules that send and receive audio from each other. But sounds like you want to dynamically load stuff if you want 3628800 possible combinations.
.hc
Hmmm.. —t3db0t
On Jan 20, 2013, at 11:53 PM, Hans-Christoph Steiner hans@at.or.at wrote:
On 01/20/2013 12:08 PM, Tedb0t wrote:
Hi all,
Does anyone know of existing designs to mirror the state of physical patch cables in a Pd patch? In other words, I'm going to have an installation with a bunch of physical patch cables plugged in between various pods and I'd like them to control a Pd patch.
hey tedbot,
You could also just play sine waves thru them and then feed all signals to a set of bandpass filters. Measure the amplitude of the filtered signals and you know which one you have. But perhaps your PWM technique would use less CPU, if that's an issue.
So far I've been thinking I could generate different PWM signals at each unique cable source and measure them at each receiving socket, then send the graph data to the computer and use dynamic patching to control the patch. However, I'm wary of using dynamic patching in a production environment—any thoughts?
you can just have a set of pre-loaded patches and switch between them using [switch~]. I've used that with good results in a production environment.
.hc
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
hi! maybe this helps you? not sure if its close to what you want. http://www.openmusiclabs.com/projects/repatcher/ and pd patches: http://wiki.openmusiclabs.com/wiki/Repatcher
d.
El 21/01/13 20:59, Hans-Christoph Steiner escribió:
On 01/21/2013 02:41 PM, Tedb0t wrote:
You could also just play sine waves thru them and then feed all signals to a set of bandpass filters. Measure the amplitude of the filtered signals and you know which one you have. But perhaps your PWM technique would use less CPU, if that's an issue.
That's an interesting idea! In my situation I'll be using low-cost microcontrollers to generate the signal so PWM would be easier.
Indeed, that makes much more sense.
you can just have a set of pre-loaded patches and switch between them using [switch~]. I've used that with good results in a production environment.
Huh.. interesting, but wouldn't this be combinatorially prohibitive? I.e. if I have 10 patchable outputs to 10 patchable inputs, then that's (n!) = 3628800 combinations.
I was dealing with < 10 different options. You can treat the patches as modules that send and receive audio from each other. But sounds like you want to dynamically load stuff if you want 3628800 possible combinations.
.hc
Hmmm.. —t3db0t
On Jan 20, 2013, at 11:53 PM, Hans-Christoph Steiner hans@at.or.at wrote:
On 01/20/2013 12:08 PM, Tedb0t wrote:
Hi all,
Does anyone know of existing designs to mirror the state of physical patch cables in a Pd patch? In other words, I'm going to have an installation with a bunch of physical patch cables plugged in between various pods and I'd like them to control a Pd patch.
hey tedbot,
You could also just play sine waves thru them and then feed all signals to a set of bandpass filters. Measure the amplitude of the filtered signals and you know which one you have. But perhaps your PWM technique would use less CPU, if that's an issue.
So far I've been thinking I could generate different PWM signals at each unique cable source and measure them at each receiving socket, then send the graph data to the computer and use dynamic patching to control the patch. However, I'm wary of using dynamic patching in a production environment—any thoughts?
you can just have a set of pre-loaded patches and switch between them using [switch~]. I've used that with good results in a production environment.
.hc
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 2013-01-21 20:41, Tedb0t wrote:
You could also just play sine waves thru them and then feed all signals to a set of bandpass filters. Measure the amplitude of the filtered signals and you know which one you have. But perhaps your PWM technique would use less CPU, if that's an issue.
That's an interesting idea! In my situation I'll be using low-cost microcontrollers to generate the signal so PWM would be easier.
you can just have a set of pre-loaded patches and switch between them using [switch~]. I've used that with good results in a production environment.
Huh.. interesting, but wouldn't this be combinatorially prohibitive? I.e. if I have 10 patchable outputs to 10 patchable inputs, then that's (n!) = 3628800 combinations.
well, use [connect( and [disconnect( messages to the patcher. (iirc, "disconnect" was introduced to allow dynamic wiring of patches controlled from the reacTable, which is not so different from what you want to do)
fgamdrs IOhannes
-
dani -
Hans-Christoph Steiner -
IOhannes m zmoelnig -
Tedb0t