Having trouble ducking Reverb in Reason

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
Post Reply
calebbrennan
Posts: 315
Joined: 16 Aug 2016

13 Jul 2022

Ducking Reverb in Reason? The best video I could find on youtube by DJ Quazar is a great effort but he goes too fast and doesn't spend enough time showing the patching on the flipped rack so I can duplicate

Any other instructional vids to recommend?


User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

13 Jul 2022

There are different ways to duck reverb (and delay). The traditional way is to duck the audio return level of the reverb. To do this use any compressor with a side chain input AFTER the reverb, and send the ‘control’ signal (the signal you want to CAUSE the ducking) to the side chain input of that reverb.
BUT, you can also duck the send to the reverb, which allows the decay time to ‘ring out’ after you duck the send.
AND, you can do something I feel is even more useful in many cases, which is to duck the DECAY time! So you get shorter reverbs when there is a signal (like a vocal) present, and longer reverbs to fill the gaps between lines. This allows you to have a huge reverb when it can be effective, but not clutter the mix when lyrics need to be heard/understood. To do this you need to go down the CV rabbit hole in Reason…
To duck the decay time (or feedback time on delays for a similar effect), you need a Combinator and a compressor. Using a vocal as an example, split the vocal and feed a compressor such as the MClass Compressor (set to low threshold/high ratio, and relatively fast attack/release times). Take the CV out from the compressor into a Combinator CV input, and assign to the reverb Decay knob. Typically you will invert the CV signal so that high signals send low values and vice-versa. This is so the louder vocals will have a shorter decay time, and when the vocal stops the decay goes to the longest desired setting.

I’m happy to go further into any of these depending on which one you’re interested in exploring. :)
Selig Audio, LLC

User avatar
deigm
Posts: 242
Joined: 10 Oct 2018
Location: Australia

14 Jul 2022

I can't believe i never thought of ducking the decay before! Makes me wonder what else i could be ducking :)
Thank you Selig. Stuff like this is why I come to this place.

User avatar
Loque
Moderator
Posts: 11170
Joined: 28 Dec 2015

14 Jul 2022

deigm wrote:
14 Jul 2022
I can't believe i never thought of ducking the decay before! Makes me wonder what else i could be ducking :)
Thank you Selig. Stuff like this is why I come to this place.
From time to time i duck the main frequency of the source signal in the reverb, to not wash out the mixed audio too much. Maybe i should have increased the send or the other frequencies too, to have more presence of the reverb while the main sound comes in 🤔
Reason12, Win10

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

14 Jul 2022

deigm wrote:
14 Jul 2022
I can't believe i never thought of ducking the decay before! Makes me wonder what else i could be ducking :)
Thank you Selig. Stuff like this is why I come to this place.
If anyone is curious about the concept I actually created a version of this for the Factory Sound Bank years ago called:
“EFX Dynamic Hall”.

There is TONS of stuff you can control dynamically, and it’s easy to do in Reason.
Selig Audio, LLC

User avatar
arnigretar
Posts: 453
Joined: 15 May 2020
Location: Iceland
Contact:

14 Jul 2022

Brilliant, diving into "EFX Dynamic Hall" stuff. Thanks Selig!
https://futuregrapher.bandcamp.com/

Reason 12, Ableton Live 10 Suite, Roland Cloud, Arturia V9, Korg Legacy 3, Soundtoys 5, Waves Mercury, Sonic Charge Bundle, N.I.: Massive, Reaktor 6, FM8. + a lot of Hardware. Windows 7/10.

User avatar
napynap
Posts: 123
Joined: 08 Sep 2017
Location: Palmdale, CA
Contact:

18 Jul 2022

selig wrote:
13 Jul 2022
...
To duck the decay time (or feedback time on delays for a similar effect), you need a Combinator and a compressor. Using a vocal as an example, split the vocal and feed a compressor such as the MClass Compressor (set to low threshold/high ratio, and relatively fast attack/release times). Take the CV out from the compressor into a Combinator CV input, and assign to the reverb Decay knob. Typically you will invert the CV signal so that high signals send low values and vice-versa. This is so the louder vocals will have a shorter decay time, and when the vocal stops the decay goes to the longest desired setting....
Brilliant! Thank you Selig!
visit http://www.napynap.com to learn more about me. Thank you.

GluhovProd
Posts: 19
Joined: 06 Dec 2019

29 Jul 2022

selig wrote:
14 Jul 2022
If anyone is curious about the concept I actually created a version of this for the Factory Sound Bank years ago called:
“EFX Dynamic Hall”.

There is TONS of stuff you can control dynamically, and it’s easy to do in Reason.
Thank you VERY MUCH, Selig! It's very interesting and helpful for me.

But after explore you patch i don't figure out few things:
- why one of m-comp have SC from self-input signal? It's for "classic" ducking one of verb unit?
- what's logic dry/wet control through Thor?
- spider CV splitter have many connections. I don't fully understand the logic behind it. Do I understand correctly that you can get away with one wire connection, but it will not work as smoothly?
- Pulveriser. Why it's here?

I was able to make my patch similar to yours, but it does not work as "smoothly" as yours. I would like to understand more!

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

02 Aug 2022

GluhovProd wrote:
29 Jul 2022
selig wrote:
14 Jul 2022
If anyone is curious about the concept I actually created a version of this for the Factory Sound Bank years ago called:
“EFX Dynamic Hall”.

There is TONS of stuff you can control dynamically, and it’s easy to do in Reason.
Thank you VERY MUCH, Selig! It's very interesting and helpful for me.

But after explore you patch i don't figure out few things:
- why one of m-comp have SC from self-input signal? It's for "classic" ducking one of verb unit?
- what's logic dry/wet control through Thor?
- spider CV splitter have many connections. I don't fully understand the logic behind it. Do I understand correctly that you can get away with one wire connection, but it will not work as smoothly?
- Pulveriser. Why it's here?

I was able to make my patch similar to yours, but it does not work as "smoothly" as yours. I would like to understand more!
Looking back over the device, yea it’s a bit confusing for sure! ;)
The input feeds four destinations: the main dynamic reverb, the input to the envelope follower (compressor), the side-chain input of a ducking compressor, and the dry input for the crossfader (dry/wet). The main dynamic verb then goes into the ducking compressor for additional ‘ducking’ effect, then into a Pulveriser for “sustain/tone” shaping/randomizing, and finally into another RV7k to add a tiny bit of “hall” to the overall effect before heading to the crossfader for dry/wet control.

Oddities: the Tremor button appears to do nothing - DOH! The Post Verb Pulveriser same, not sure what it was intended to do but obviously didn’t make the cut (and thus SHOULD have been removed) - also DOH!

As for your specific questions:
1-Yes, the side chain input is for a ‘classic’ level ducking option.
2-Dry/Wet through Thor to get a 3 dB crossfade curve - using Line Mixers gives a 6 dB curve which doesn’t work well for a dry/wet control IMO.
3-The splitter was for additional options which I’ve mentioned above.
4-Pulveriser was also for additional options, neither of which are particularly necessary for the basic effect (but I’m sure I had a good reason for adding it at the time!?!).

The basic effect could be built with a Sweeper, RV-7000, and something to invert the CV (Thor, Selig Gain, other CV utility). Assigning the reverb Decay to the first Rotary on the Combinator, then using the CV input for that rotary on the back gives you the basic patch. You may also want to be able to control the depth of the envelope effect from the Combinator (Selig Gain could do this too) but otherwise it’s down to fine tuning the settings at this point.

The primary ‘issue’ with any modulation path is predicting how far something will ‘modulate’. Meaning, we typically set modulation ‘depth’ in percent on most synths, and it’s no different here. What would be more desirable IMO is the ability to set the start/end POSITIONS, or in the case of dynamic reverb the decay time for soft vs loud sounds. Meaning, we would want to be more sure of exactly how far that Decay knob moves in response to CV input. The ideal would then be one knob/fader for the ‘soft’ dynamics (or ‘off’), and a second knob/fader for the loud, but it’s not something built into Reason so it has to be custom built.

This would allow you to define modulation by positions of the controls rather than by an arbitrary ‘depth’ knob in percent. Either option uses two knobs, so there isn’t any knob penality for this approach. In fact this approach is similar to synths that use modulation ‘rings’ around the knobs (Massive, Pigments) with the exception that those synths still have a ‘depth’ parameter and my solution uses instead two independent decay time sliders - one for when there is no signal present, one for when the signal is 100% present. This allows you to choose your modulation target (max modulation point) rather than your modulation depth if that makes sense…
This is easier to show than to explain, basically it means the modulation depth is alwyas defined by the distance BETWEEN the two controls rather than a percent above/below the starting point. So it’s more of a literal approach which plays nicely to the type of modulation used here, where the two knobs define the range. This also allows positive or negative moduliation simply based on the position of the two controls - for example, if the louder control is higher than the softer, the modulation is positive. If OTOH, the softer control is higher than the louder, you modulate in the opposite direction.

Instead of going deeper here, I’m going to use this concept as my main subject for this months live stream (this Thursday Aug 4), and will post the Combinator using this technique as well as the math (yes math) behind wiring this up. I initially wasn’t even sure it would be possible to do this particular technique in Reason, but turns out it is fairly simple and would even work on any version of Reason from v4 onwards (due to my tyipical extreme use of Thor as a CV processing device). I’ll post back here when I get the main thread written/posted in the next day or so, hope this helps! :)
Selig Audio, LLC

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

04 Aug 2022

I'll be covering this subject, the simple and the more complex approaches, today in my live stream:
viewtopic.php?f=5&t=7528931
Selig Audio, LLC

manoob
Posts: 36
Joined: 17 Aug 2015

04 Aug 2022

try sidechaining. easiest way and you can sculp it...

User avatar
tomusurp
Posts: 248
Joined: 30 Jan 2022
Location: USA
Contact:

11 Sep 2022

selig wrote:
13 Jul 2022
There are different ways to duck reverb (and delay). The traditional way is to duck the audio return level of the reverb. To do this use any compressor with a side chain input AFTER the reverb, and send the ‘control’ signal (the signal you want to CAUSE the ducking) to the side chain input of that reverb.
BUT, you can also duck the send to the reverb, which allows the decay time to ‘ring out’ after you duck the send.
AND, you can do something I feel is even more useful in many cases, which is to duck the DECAY time! So you get shorter reverbs when there is a signal (like a vocal) present, and longer reverbs to fill the gaps between lines. This allows you to have a huge reverb when it can be effective, but not clutter the mix when lyrics need to be heard/understood. To do this you need to go down the CV rabbit hole in Reason…
To duck the decay time (or feedback time on delays for a similar effect), you need a Combinator and a compressor. Using a vocal as an example, split the vocal and feed a compressor such as the MClass Compressor (set to low threshold/high ratio, and relatively fast attack/release times). Take the CV out from the compressor into a Combinator CV input, and assign to the reverb Decay knob. Typically you will invert the CV signal so that high signals send low values and vice-versa. This is so the louder vocals will have a shorter decay time, and when the vocal stops the decay goes to the longest desired setting.

I’m happy to go further into any of these depending on which one you’re interested in exploring. :)
any chance you can show backside pictures of these different routing methods? although I should probably watch the livestream
"The hottest in the matrix"
My music:
https://www.youtube.com/c/TomUsurp
https://www.tomusurp.com


:reason: :re: :refill: :rt:

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

11 Sep 2022

tomusurp wrote:
11 Sep 2022
selig wrote:
13 Jul 2022
There are different ways to duck reverb (and delay). The traditional way is to duck the audio return level of the reverb. To do this use any compressor with a side chain input AFTER the reverb, and send the ‘control’ signal (the signal you want to CAUSE the ducking) to the side chain input of that reverb.
BUT, you can also duck the send to the reverb, which allows the decay time to ‘ring out’ after you duck the send.
AND, you can do something I feel is even more useful in many cases, which is to duck the DECAY time! So you get shorter reverbs when there is a signal (like a vocal) present, and longer reverbs to fill the gaps between lines. This allows you to have a huge reverb when it can be effective, but not clutter the mix when lyrics need to be heard/understood. To do this you need to go down the CV rabbit hole in Reason…
To duck the decay time (or feedback time on delays for a similar effect), you need a Combinator and a compressor. Using a vocal as an example, split the vocal and feed a compressor such as the MClass Compressor (set to low threshold/high ratio, and relatively fast attack/release times). Take the CV out from the compressor into a Combinator CV input, and assign to the reverb Decay knob. Typically you will invert the CV signal so that high signals send low values and vice-versa. This is so the louder vocals will have a shorter decay time, and when the vocal stops the decay goes to the longest desired setting.

I’m happy to go further into any of these depending on which one you’re interested in exploring. :)
any chance you can show backside pictures of these different routing methods? although I should probably watch the livestream
Happy to share, but the way I approached Decay Ducking was beyond what was needed for the basic effect.
For the basic effect just do this.

Image

The key ingredients:
The dry input signal is used to create the envelope CV.
The CV is inverted so you reduce Decay time the louder the input signal.
The CV trim control affects how far the Decay time is modulated with the CV envelope follower signal (technically modulation depth, basically a 'scale' control).

From there, to add complexity/control:
Any number of devices can alternatively used to create the CV signal.
Any number can invert and scale the CV.
You can control the Decay parameter as above OR you can use a Combinator and control the Knob itself
You can also choose to do the opposite, which is similar to gated reverb which is to start with a short Decay time and use the input signal level to make the Decay longer with louder sounds. The result is long Decay times when a signal is present, and short when the source stops.

Things to be aware of:
When working with an envelope follower you don't necessarily get the full range out of the device like you do with an LFO or Envelope Generator. Meaning, an Envelope Generator sends 0-127 out for it's CV signal every time (unless velocity is involved). But the Envelope "Follower" will follow the input signal and thus may only send a fraction of the 127 if the level is too low. SO the first part of using any dynamically controlled device is getting the levels fine tuned by playing your source signal. With Sweeper there is a Gain In control and a visual aid to getting the max CV level when the loudest part is played. From there, you can then set the modulation depth with a bit more confidence as to what to expect to see. Basically with 100% modulation, you want to be able to take the Decay time from it's max to it's minimum, so in essence you are 'calibrating' the system when you set initial levels in the envelope follower so that 100% max loudness on the inputs produces 100% max modulation when depth is set to 100%. Dang, I know there is an easier way to explain this, it's late and I fear I'm not writing this as clear as it could be but hopefully this gives you something to get started down the path of dynamic control of any device/parameter! :)
Selig Audio, LLC

User avatar
tomusurp
Posts: 248
Joined: 30 Jan 2022
Location: USA
Contact:

12 Sep 2022

selig wrote:
11 Sep 2022
tomusurp wrote:
11 Sep 2022


any chance you can show backside pictures of these different routing methods? although I should probably watch the livestream
Happy to share, but the way I approached Decay Ducking was beyond what was needed for the basic effect.
For the basic effect just do this.

Image

The key ingredients:
The dry input signal is used to create the envelope CV.
The CV is inverted so you reduce Decay time the louder the input signal.
The CV trim control affects how far the Decay time is modulated with the CV envelope follower signal (technically modulation depth, basically a 'scale' control).

From there, to add complexity/control:
Any number of devices can alternatively used to create the CV signal.
Any number can invert and scale the CV.
You can control the Decay parameter as above OR you can use a Combinator and control the Knob itself
You can also choose to do the opposite, which is similar to gated reverb which is to start with a short Decay time and use the input signal level to make the Decay longer with louder sounds. The result is long Decay times when a signal is present, and short when the source stops.

Things to be aware of:
When working with an envelope follower you don't necessarily get the full range out of the device like you do with an LFO or Envelope Generator. Meaning, an Envelope Generator sends 0-127 out for it's CV signal every time (unless velocity is involved). But the Envelope "Follower" will follow the input signal and thus may only send a fraction of the 127 if the level is too low. SO the first part of using any dynamically controlled device is getting the levels fine tuned by playing your source signal. With Sweeper there is a Gain In control and a visual aid to getting the max CV level when the loudest part is played. From there, you can then set the modulation depth with a bit more confidence as to what to expect to see. Basically with 100% modulation, you want to be able to take the Decay time from it's max to it's minimum, so in essence you are 'calibrating' the system when you set initial levels in the envelope follower so that 100% max loudness on the inputs produces 100% max modulation when depth is set to 100%. Dang, I know there is an easier way to explain this, it's late and I fear I'm not writing this as clear as it could be but hopefully this gives you something to get started down the path of dynamic control of any device/parameter! :)
I very scarcely understand this. This is quite a bit to take in and you have a lot more technical experience than me. But I appreciate you sharing this knowledge. I believe all these ducking methods will take my productions beyond the realm of Earth. But I will have to learn step by step this CV process.

I will take a look at your livestream, hopefully it will break this stuff down visually for me and in a way I can understand it’s function.
"The hottest in the matrix"
My music:
https://www.youtube.com/c/TomUsurp
https://www.tomusurp.com


:reason: :re: :refill: :rt:

splitpen
Posts: 141
Joined: 22 Mar 2017

13 Sep 2022

Bit late to the party,
Just a trick to give try
What also works in a lot of times with shaping the reverb (at least for myself). When placed a reverb on a send channel, placing a transient shaper before it is sended to the reverb and soften the attack will sound way more natural in the blend. Certainly when having source sounds with a bit snappy sounds on front. The clicks will normaly be extra pronounced by the diffused delays in the reverb. To get big spaces you can place a compressor with short attack after the reverb to blew it all up. OTT vst will do the trick in most times as well.
Latest track: https://www.youtube.com/watch?v=EjzYHe3D8f8
Reasonstudios 11 Roland Quadcapture | AMD Ryzen 5900x|Elektron Analog Heat Mk2|Diy 3 way SB Acoustics/Scanspeak studiomonitors | AKG K702 | Key­sta­tion 61 MK3 |BCR2000 |X-Touch Ext | MP Midi

Post Reply
  • Information
  • Who is online

    Users browsing this forum: Kypresso and 17 guests