Mungo Enterprises d0, dual channel Delay

Cwejman, Livewire, TipTop Audio, Doepfer etc... Get your euro on!

Moderators: Joe., lisa, luketeaford, Kent

Post Reply
LunaticSound
Common Wiggler
Posts: 105
Joined: Tue Dec 25, 2018 3:50 pm
Location: Hamburg

Re: Mungo Enterprises d0, dual channel Delay

Post by LunaticSound » Sun Jan 26, 2020 6:15 am

Could someone point me to where I can find out more about this? Or explain a bit more, what is going on there? I am fascinated.

leterell
Wiggling with Experience
Posts: 259
Joined: Sun Mar 02, 2014 8:46 am
Location: Europe

Re: Mungo Enterprises d0, dual channel Delay

Post by leterell » Sun Jan 26, 2020 6:23 am

LunaticSound wrote:
Sun Jan 26, 2020 6:15 am
Could someone point me to where I can find out more about this? Or explain a bit more, what is going on there? I am fascinated.
You mean buttons ARE toys's SoundCloud bit?

Here
Buttons ARE toys wrote:
Sat Jan 25, 2020 4:20 pm
timoka wrote:
Sat Jan 25, 2020 6:32 am
why did you remove the freq shifter example @Buttons ARE toys? it sounded good and it's an extremely interesting patch!
I was going to upload something large to soundcloud and needed to clear space, and decided to remove demos first. But then I ended up ditching the thing I was going to upload soooooo



The audio signal is both processed by the D0, and used to clock it, allowing the delay to apply a constant 90° phase shift regardless of frequency.

The 0° and 90° phase signals each go to separate ring modulators. Then the 0° and 90° signals of a quadrature oscillator are used to modulate each respective ring modulator. The upshift is produced by summing the ringmod outputs, while the downshift is produced by subtracting the 90° product from the 0° product.

Downshift panned left, upshift panned right. Half way through, feedback is introduced in varying degrees, sending each ringmod product back into itself and into each other for different flavors.

It was a bit hard at first to get the D0 to produce a 90° shift using a complex audio signal so the way I did it was to send the 0° output of my quadrature LFO into it and then compare the output to the 90° LFO output, adjusting the delay until they matched, and then unplugged that and plugged in the audio signal. Also you need to turn the slew up quite a bit to avoid glitches.

LunaticSound
Common Wiggler
Posts: 105
Joined: Tue Dec 25, 2018 3:50 pm
Location: Hamburg

Re: Mungo Enterprises d0, dual channel Delay

Post by LunaticSound » Sun Jan 26, 2020 6:33 am

leterell wrote:
Sun Jan 26, 2020 6:23 am
LunaticSound wrote:
Sun Jan 26, 2020 6:15 am
Could someone point me to where I can find out more about this? Or explain a bit more, what is going on there? I am fascinated.
You mean buttons ARE toys's SoundCloud bit?

Here
Buttons ARE toys wrote:
Sat Jan 25, 2020 4:20 pm
timoka wrote:
Sat Jan 25, 2020 6:32 am
why did you remove the freq shifter example @Buttons ARE toys? it sounded good and it's an extremely interesting patch!
I was going to upload something large to soundcloud and needed to clear space, and decided to remove demos first. But then I ended up ditching the thing I was going to upload soooooo



The audio signal is both processed by the D0, and used to clock it, allowing the delay to apply a constant 90° phase shift regardless of frequency.

The 0° and 90° phase signals each go to separate ring modulators. Then the 0° and 90° signals of a quadrature oscillator are used to modulate each respective ring modulator. The upshift is produced by summing the ringmod outputs, while the downshift is produced by subtracting the 90° product from the 0° product.

Downshift panned left, upshift panned right. Half way through, feedback is introduced in varying degrees, sending each ringmod product back into itself and into each other for different flavors.

It was a bit hard at first to get the D0 to produce a 90° shift using a complex audio signal so the way I did it was to send the 0° output of my quadrature LFO into it and then compare the output to the 90° LFO output, adjusting the delay until they matched, and then unplugged that and plugged in the audio signal. Also you need to turn the slew up quite a bit to avoid glitches.
That's what I meant, yes, thank you. I had read that post and I would like to understand, why a phase shifted feedbacked ringmod produces a pitch shift, for example. And why 90 degrees? Is it even a pitch shift or is it something like side bands rather?

Basically, I want to get to a point where I could come up with something like that on my own. To understand the maths behind it, I guess.

Thank you!

User avatar
Navs
Super Deluxe Wiggler
Posts: 4140
Joined: Fri Sep 18, 2009 3:49 am
Location: Berlin, Germany

Re: Mungo Enterprises d0, dual channel Delay

Post by Navs » Sun Jan 26, 2020 8:43 am


User avatar
Buttons ARE toys
Wiggling with Experience
Posts: 364
Joined: Wed Sep 14, 2016 9:24 am
Location: Earth

Re: Mungo Enterprises d0, dual channel Delay

Post by Buttons ARE toys » Sun Jan 26, 2020 9:05 am

LunaticSound wrote:
Sat Jan 25, 2020 7:52 pm
Wow. How did you come up with that?

Man, ButtonsAreToys' demo plus Damases post before that make it pretty much unjustifiable to not get the D0. Damn.
Yeah it was desolation jones who pointed the possibility out to me, and then I used Navs' blog as a guide on how to do the actual patch.
LunaticSound wrote:
Sun Jan 26, 2020 6:33 am
That's what I meant, yes, thank you. I had read that post and I would like to understand, why a phase shifted feedbacked ringmod produces a pitch shift, for example. And why 90 degrees? Is it even a pitch shift or is it something like side bands rather?

Basically, I want to get to a point where I could come up with something like that on my own. To understand the maths behind it, I guess.

Thank you!
I can't help you there, I'm far from understanding why it works, myself. I do know it has to do with sidebands rather than pitch shifting though. I don't know much more than that, but this is similar to the way shortwave radio signals are encoded and decoded for transmission and they refer to the different modes of transmission as single sideband and double sideband. That was one of the main reasons I was so interested in this patch, so I could use it to simulate shortwave broadcast sounds and mess with my existing shortwave recordings.
Astronomical Dusk - a web radio show of synthetic delights and collected oddities. Airs monthly at astronomical dusk.

damase
Super Deluxe Wiggler
Posts: 1722
Joined: Sun Apr 19, 2015 11:33 am
Location: Texas

Re: Mungo Enterprises d0, dual channel Delay

Post by damase » Sun Jan 26, 2020 9:12 am

Bravo for that patch! for all the variables and possible failure points it sounds fantastic

desolationjones
Wiggling with Experience
Posts: 420
Joined: Sat Jan 21, 2017 5:55 pm

Re: Mungo Enterprises d0, dual channel Delay

Post by desolationjones » Sun Jan 26, 2020 9:16 am

Buttons ARE toys wrote:
Sun Jan 26, 2020 9:05 am
That was one of the main reasons I was so interested in this patch, so I could use it to simulate shortwave broadcast sounds and mess with my existing shortwave recordings.
Your podcast rocks btw. Did this patch make an appearance in ep. 2?

User avatar
Buttons ARE toys
Wiggling with Experience
Posts: 364
Joined: Wed Sep 14, 2016 9:24 am
Location: Earth

Re: Mungo Enterprises d0, dual channel Delay

Post by Buttons ARE toys » Sun Jan 26, 2020 9:39 am

desolationjones wrote:
Sun Jan 26, 2020 9:16 am
Buttons ARE toys wrote:
Sun Jan 26, 2020 9:05 am
That was one of the main reasons I was so interested in this patch, so I could use it to simulate shortwave broadcast sounds and mess with my existing shortwave recordings.
Your podcast rocks btw. Did this patch make an appearance in ep. 2?
Hey thanks! No, I hadn't put this patch together yet when I made that episode. Most of the frequency shifting in there was just from recording shortwave broadcasts with the radio out of tune. I'm already putting the patch to good use for episode 3 though.
Astronomical Dusk - a web radio show of synthetic delights and collected oddities. Airs monthly at astronomical dusk.

Jaypee
Super Deluxe Wiggler
Posts: 1053
Joined: Sat Nov 22, 2014 11:46 am

Re: Mungo Enterprises d0, dual channel Delay

Post by Jaypee » Wed Feb 05, 2020 6:20 am

Is it the correct and most optimal way to do KS patch and still have a 'normal' delay?
Capture d’écran 2020-02-05 à 12.17.20.png
I know burst noise is not a "must to". Can be anything. Will try to ping another filter and use it as a "burst.

I put the original version if anyone has something to add/edit
Capture d’écran 2020-02-05 à 12.15.01.png
Cheers!
You do not have the required permissions to view the files attached to this post.
"Those aren't your daddy's waveforms " Cynthia Webster

desolationjones
Wiggling with Experience
Posts: 420
Joined: Sat Jan 21, 2017 5:55 pm

Re: Mungo Enterprises d0, dual channel Delay

Post by desolationjones » Wed Feb 05, 2020 7:50 am

Delay B looks good for KS but Delay A is not hooked to anything and I don't know where you're monitoring from. Maybe you meant to use Delay A for the KS and use the built-in mixer to feed the KS to Delay B? In addition to d0's mixing and the KS feedback mixing you will need to mix the KS output with the normal delay output otherwise you won't hear the first signal, just repeats.

Filter pings should be good for less metallic KS sounds since the comb filtering would be less pronounced. Honestly I never patch filter pings as the input for KS since they sound so nice on their own.

Jaypee
Super Deluxe Wiggler
Posts: 1053
Joined: Sat Nov 22, 2014 11:46 am

Re: Mungo Enterprises d0, dual channel Delay

Post by Jaypee » Thu Feb 06, 2020 3:05 am

Capture d’écran 2020-02-06 à 10.40.02.png
Yes, now the KS OUTPUT and normal delay are represented and correctly patched...I think! Will try this version today!
desolationjones wrote:
Wed Feb 05, 2020 7:50 am
Maybe you meant to [...] use the built-in mixer to feed the KS to Delay B
Yes :)
You do not have the required permissions to view the files attached to this post.
"Those aren't your daddy's waveforms " Cynthia Webster

Jaypee
Super Deluxe Wiggler
Posts: 1053
Joined: Sat Nov 22, 2014 11:46 am

Re: Mungo Enterprises d0, dual channel Delay

Post by Jaypee » Thu Feb 06, 2020 7:45 am



Nothing musical, but still amazed by the tone these modules can produce :hail:
"Those aren't your daddy's waveforms " Cynthia Webster

User avatar
jjclark
Professor Wiggler
Posts: 712
Joined: Wed Apr 01, 2009 6:06 pm
Location: Montreal

Re: Mungo Enterprises d0, dual channel Delay

Post by jjclark » Mon Feb 10, 2020 6:05 pm

damase wrote:
Sat Jan 25, 2020 1:45 pm
ive had both and can compare
the rainmaker, sounds good and is all in 1 package, but the comb maxes out at 3.8 ms which is just so much less than the d0 which can go up to .2ms...5k frequency i think.
This isn't true (about the rainmaker comb maxing out at 3.8msec). The rainmaker comb can go down to 0.1 msec. There is an internal added delay of just a bit less than 0.1 msec due to the interpolation filtering, so the minimum comb time is just under 0.2 msec. The maximum frequency I can get it to ring at is about 5.2KHz.
To get this you need to set the #taps to 1, set the feedback to Raw, set the clk src to Encoder and push in the encoder to get it to step by 0.1 msec. You need to adjust the comb feedback so that it is on the edge of ringing. Then ping away! You can easily get kilohertz beeps.

Note that the internal resolution of the comb delay line is 0.01msec (10 usec). If you set the number of taps to something 10 or more, and set the comb size to 0.1 msec then the time between the individual taps will be set to 0.01 msec. The behavior of the comb becomes complicated when you have more than 1 tap, so its not exactly Karplus-Strong anymore, but you do get ringing at multiples of the tap spacing.

(sorry to highjack the Mungo delay thread - now back to the cool D0 patches...)

User avatar
Mungo
Super Deluxe Wiggler
Posts: 1130
Joined: Fri Jan 04, 2013 8:12 pm

Re: Mungo Enterprises d0, dual channel Delay

Post by Mungo » Mon Feb 10, 2020 7:18 pm

jjclark wrote:
Mon Feb 10, 2020 6:05 pm
damase wrote:
Sat Jan 25, 2020 1:45 pm
ive had both and can compare
the rainmaker, sounds good and is all in 1 package, but the comb maxes out at 3.8 ms which is just so much less than the d0 which can go up to .2ms...5k frequency i think.
This isn't true (about the rainmaker comb maxing out at 3.8msec). The rainmaker comb can go down to 0.1 msec. There is an internal added delay of just a bit less than 0.1 msec due to the interpolation filtering, so the minimum comb time is just under 0.2 msec. The maximum frequency I can get it to ring at is about 5.2KHz.
To get this you need to set the #taps to 1, set the feedback to Raw, set the clk src to Encoder and push in the encoder to get it to step by 0.1 msec. You need to adjust the comb feedback so that it is on the edge of ringing. Then ping away! You can easily get kilohertz beeps.

Note that the internal resolution of the comb delay line is 0.01msec (10 usec). If you set the number of taps to something 10 or more, and set the comb size to 0.1 msec then the time between the individual taps will be set to 0.01 msec. The behavior of the comb becomes complicated when you have more than 1 tap, so its not exactly Karplus-Strong anymore, but you do get ringing at multiples of the tap spacing.

(sorry to highjack the Mungo delay thread - now back to the cool D0 patches...)
There are several modules with very short delays possible on internal feedback, that important distinction should be made clear. What is the shortest delay possible when using an external feedback path?

The resolution you mention of 0.01ms is a little confusing, is that the finest resolution of the interpolation or just the calculation for spacing of taps?
Generally comb filter is used to describe the result of a single delay element/tap:
https://en.wikipedia.org/wiki/Comb_filter
Combined with short delays that have very fine resolution then its possible to play tuned filters/notes that track accurately. Even with the constraints of using internal feedback, steps of 0.1ms wouldn't be able to tune the filter as it would jump from roughly 5kHz to 3kHz in one step. Using much higher resolution the d0 tracks accurately all the way to the minimum delay time.

P.S. Love the concept of the rainmaker, I had a similar module prototyped up when you released it!

damase
Super Deluxe Wiggler
Posts: 1722
Joined: Sun Apr 19, 2015 11:33 am
Location: Texas

Re: Mungo Enterprises d0, dual channel Delay

Post by damase » Mon Feb 10, 2020 8:14 pm

jjclark wrote:
Mon Feb 10, 2020 6:05 pm
...The maximum frequency I can get it to ring at is about 5.2KHz....
that sounds like something i need to try, thank you 8-) . i had no idea i could step it higher with the encoder. happy to be corrected, my apologies

User avatar
jjclark
Professor Wiggler
Posts: 712
Joined: Wed Apr 01, 2009 6:06 pm
Location: Montreal

Re: Mungo Enterprises d0, dual channel Delay

Post by jjclark » Mon Feb 10, 2020 11:47 pm

Mungo wrote:
Mon Feb 10, 2020 7:18 pm
There are several modules with very short delays possible on internal feedback, that important distinction should be made clear. What is the shortest delay possible when using an external feedback path?

The resolution you mention of 0.01ms is a little confusing, is that the finest resolution of the interpolation or just the calculation for spacing of taps?
Generally comb filter is used to describe the result of a single delay element/tap:
https://en.wikipedia.org/wiki/Comb_filter
Combined with short delays that have very fine resolution then its possible to play tuned filters/notes that track accurately. Even with the constraints of using internal feedback, steps of 0.1ms wouldn't be able to tune the filter as it would jump from roughly 5kHz to 3kHz in one step. Using much higher resolution the d0 tracks accurately all the way to the minimum delay time.

P.S. Love the concept of the rainmaker, I had a similar module prototyped up when you released it!
Yes, I was just talking about the internal delay paths. For external inputs there is about a 2msec end-to-end input-output latency (mostly due to the A/D and D/A chips), so any external hookup for Karplus-Strong would be limited. But you can do Karplus-Strong using the internal feedback.
The 0.01msec is the time between successive delay line taps, i.e. before interpolation. 0.1 msec is the minimum time that can be set using the encoder. But there is also a 1v/oct input for setting the delay time which uses interpolation which can get you down to 0.01msec. So you can use this for smoothly varying pitched sounds, up to about 3KHz or so.

User avatar
Buttons ARE toys
Wiggling with Experience
Posts: 364
Joined: Wed Sep 14, 2016 9:24 am
Location: Earth

Re: Mungo Enterprises d0, dual channel Delay

Post by Buttons ARE toys » Mon Mar 30, 2020 2:17 pm

So I haven't found that many interesting ways to play with the frequency shifter patch outside of what you would typically do with one. Doing things like FMing the modulation signal or whatnot tend to just result in harsh noise. But, something I've been doing is using my stereo sampler module to record the 0 and 90 degree shifted versions of my carrier audio, which then frees up the D0 to apply a phase shift to any arbitrary signal rather than having to rely on a quadrature oscillator. That has resulted in some very fun stuff!
Astronomical Dusk - a web radio show of synthetic delights and collected oddities. Airs monthly at astronomical dusk.

Post Reply

Return to “Eurorack Modules”