Hi, I've built a classic stereo delay using [delwrite~] and [vd~], and L/R crossed feedback. This works!
Now, I want to insert an effect into the feedback path. (many possibilities here: add some bitcrush, so each repeat gradually becomes more distorted.... or a [hilbert~] type pitch shift, which might create something like Shepard tones, etc).
*This works when everything is within a single patch or abstraction. *
However, when I try to allow the effect to be external to the abstraction (like a SEND/RETURN insert on a mixer), when I try to complete the circuit, I get "DSP loop detected".
The two circuits are essentially IDENTICAL!
This seems like a bug to me. Any ideas?
I've attached pix of both the high-level schematic, along with a demo patch. (I can upload the actual patch if that would be helpful).
Thanks! BH
-- William Huston: WilliamAHuston@gmail.com Binghamton NY
*Public Service Mapping / Videography / Research / Education / Safety Advocacy* Blog http://WilliamAHuston.blogspot.com -- Facebook http://facebook.com/billhuston -- Twitter http://twitter.com/WilliamAHuston-- Youtube https://www.youtube.com/channel/UCGijK1amWOLglT3YeTyEBNQ?sub_congfirmation=1
- -- Podcast Blog https://billhustonpodcast.blogspot.com/*
*Document collections*: VirtualPipelines http://TinyURL.com/VirtualPipelines -- BHDCSDimockArchive http://bit.ly/BHDCSDimockArchive *Please support my work! -- *TinyURL.com/DonateToBillHuston
On 25.02.20 09:50, William Huston wrote:
I've attached pix of both the high-level schematic, along with a demo patch. (I can upload the actual patch if that would be helpful).
please don't post screenshots of Pd-patches. post Pd-patches instead, they are much smaller (being text only) and are actually of *use* to those who might want to give your problem a glance.
fgamsdr IOhannes
I couldn't find it in the picture, but I didn't look carefully and I agree it'd be better if we had the patch to open. Nonetheless, I strongly doubt there's a bug. The "DSP loop detected" error is pretty straight forward and tells you have made a feedback signal connection, where an output is connected to an input earlier in the chain, and this is bad.
For reference: https://github.com/porres/Live-Electronics-Tutorial/blob/v1.0-beta-17/Exampl...
Anyway, this is happening somewhere in the patch, maybe if you send us it, someone will help you find out how it's happening.
Em ter., 25 de fev. de 2020 às 06:15, IOhannes m zmoelnig zmoelnig@iem.at escreveu:
On 25.02.20 09:50, William Huston wrote:
I've attached pix of both the high-level schematic, along with a demo patch. (I can upload the actual patch if that would be helpful).
please don't post screenshots of Pd-patches. post Pd-patches instead, they are much smaller (being text only) and are actually of *use* to those who might want to give your problem a glance.
fgamsdr IOhannes
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
Hi William,
I second IOhannes and Alex that it's better to attach actual patches, but from looking at the picture it seems like you attempt to do your "external" feedback path via signal connections and this obviously doesn't work because it creates a DSP loop. The workaround is to use a pair of [s~]/[r~] for one side of the aux path, which "breaks" the DSP graph. Note that in your "inner" feedback path you're already doing this, that's why you don't get a DSP loop there.
Note that [s~]/[r~] is basically a delay line of a single block, unless you force the order with signal connections, see "3.audio.examples/G05.execution.order"
Christof
On 25.02.2020 09:50, William Huston wrote:
Hi, I've built a classic stereo delay using [delwrite~] and [vd~], and L/R crossed feedback. This works!
Now, I want to insert an effect into the feedback path. (many possibilities here: add some bitcrush, so each repeat gradually becomes more distorted.... or a [hilbert~] type pitch shift, which might create something like Shepard tones, etc).
*This works when everything is within a single patch or abstraction. *
However, when I try to allow the effect to be external to the abstraction (like a SEND/RETURN insert on a mixer), when I try to complete the circuit, I get "DSP loop detected".
The two circuits are essentially IDENTICAL!
This seems like a bug to me. Any ideas?
I've attached pix of both the high-level schematic, along with a demo patch. (I can upload the actual patch if that would be helpful).
Thanks!
BH
-- William Huston: WilliamAHuston@gmail.com mailto:WilliamAHuston@gmail.com Binghamton NY
*Public Service Mapping / Videography / Research / Education / Safety Advocacy* Blog http://WilliamAHuston.blogspot.com -- Facebook http://facebook.com/billhuston -- Twitter http://twitter.com/WilliamAHuston-- Youtube https://www.youtube.com/channel/UCGijK1amWOLglT3YeTyEBNQ?sub_congfirmation=1*-- Podcast Blog https://billhustonpodcast.blogspot.com/
*Document collections*: VirtualPipelines http://TinyURL.com/VirtualPipelines -- BHDCSDimockArchive http://bit.ly/BHDCSDimockArchive *Please support my work! -- *TinyURL.com/DonateToBillHuston http://TinyURL.com/DonateToBillHuston
**
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
-
Alexandre Torres Porres -
Christof Ressi -
IOhannes m zmoelnig -
William Huston