"Trackspacer"-esque Ducking EQ (Stock)

[alliemakesmusic]

Hey all,

Summary:
I got the idea to use BV512 to make a stock combi that performs a similar function to what "track spacer" by waves or the Carve EQ RE does. If you aren't familiar, essentially it takes in a sidechain audio and performs a multiband ducking of different frequency bins across the whole frequency spectrum. This makes it so the main signal source ducks all frequencies being used by sidechain and prevents mud or other kinds of frequency conflict. The best part is if the sidechain audio is in a rest or goes silent the signal source returns to flat EQ. Less automating guitars fighting with pianos

How to use the combinator:
Connect a mono sidechain to the modulator source for the first BV512 and your audio as the combinator's input.

Parameters:

• Scale Cut - Scales the amount of ducking that will be applied. This should always be dialed on a track by track basis to "whatever sounds right" similar to how you would dial in any cut or boost.
• Db Offset - where "0" is on the EQ. 89 is 0db but you can move it up or down. This will have the side effect of making cuts more or less drastic as a side effect as well as any coloration associated with the band count.
• Dry / Wet - Self explanatory hopefully
• Attack - How quickly a band gets ducked
• Release - How quickly a band stops ducking when the frequency content disappears from the sidechain
• Band Count - 4, 8, 16, 32, FFT; as stated in the reason manual, there's only 16 CV outs on the back so if you select 32 or FFT each output controls more than one band. I usually just dial this to whatever sounds the best.
• Hold - applies the hold button to the inner modulator BV512. This will lock in the frequency band ducks for that moment in time. Useful to prevent pumping when maxing out release isn't enough.
• Volume - Master fader

P.S. A brief aside, I have downloaded a crap ton of combinators from this community and I'm very appreciative so I hope you get some use out of this. I honestly think the combinator update is Reason's most underrated feature - it's what made me decide to upgrade to 12 hopefully Reason Studios keeps adding to it and making it even more featureful.

[A.sch3]

Good idea !
I need to try it, but, does it colors the affected sound ? the BV512 i remember to be not very transparent in EQ mode

Thanks for sharing !

[ljekio]

my concern has always been that the frequency of the control signal is 64 times less than the frequency of the audio. Is this enough for the inaudible operation of the device?

[jam-s]

I think it would be a good thing to add the instructions on how to setup/operate your combinator to the combinator itself (either via a custom backdrop in its lower part so it can be hidden by reducing the HE count of the combinator) or preferably via a TMA-4 or TMA-9 included in the combi itself.

[alliemakesmusic]

Good idea !
I need to try it, but, does it colors the affected sound ? the BV512 i remember to be not very transparent in EQ mode

Thanks for sharing !

It does color the sound. The thing I notice the most is the bottom end tends to disappear in higher band counts which may or may not be an issue depending on source material. If you use 4 to 8 bands its fairly transparent to me. In FET mode 0db is 100% transparent but big DB changes can introduce artifacts according to the manual. It worked well enough for what I needed it for (stuffing inane amounts of ear candy against the mixbus) but I might try to make one that's more transparent.

my concern has always been that the frequency of the control signal is 64 times less than the frequency of the audio. Is this enough for the inaudible operation of the device?

I'm not sure I understand. Is the question if the CV changes fast enough for the effect to work? Most harmonic content is not changing fast enough to interfere. If the concern is aliasing, the frequency bins are discrete and fixed that have analysis of amplitude happening at normal DSP frequencies during processing and its only the amplitude of the signal that becomes quantized.

I think it would be a good thing to add the instructions on how to setup/operate your combinator to the combinator itself (either via a custom backdrop in its lower part so it can be hidden by reducing the HE count of the combinator) or preferably via a TMA-4 or TMA-9 included in the combi itself.

Good idea. Maybe in V2 I wanted to do a better graphic for the backdrop anyway

[alliemakesmusic]

V1.1 where I don't upload one with cables missing

[Loque]

You can also build one with some free and quite old Red Rock devices:
https://reasonstudios.com/shop/rack-ext ... equaliser/
https://reasonstudios.com/shop/rack-ext ... -analyzer/

And i think with this, you can also smooth out the gain reduction:
https://reasonstudios.com/shop/rack-ext ... ic-driver/

[crimsonwarlock]

[selig]

Good idea !
I need to try it, but, does it colors the affected sound ? the BV512 i remember to be not very transparent in EQ mode

Thanks for sharing !

Definitely colors the sound:

[selig]

my concern has always been that the frequency of the control signal is 64 times less than the frequency of the audio. Is this enough for the inaudible operation of the device?

Whenever you use CV for ducking you run up against the exact same limitation, so if it's not a problem there it won't be one here. The lack of low end and the low resolution of that range are a bigger limitation IMO, if only because that's the region you need the most control. It's the same issue with some other (paid) solutions where the low response isn't at all 'fine' enough (or smooth enough) to actually deal with the problems you have in that frequency region, let alone do it transparently IMO.

[selig]

Thought this looks familiar…
Here's the version I did (wow, 10 years ago!).
https://www.dropbox.com/s/9k3y9bmeo6afs ... b.zip?dl=0
One difference is I used a single Thor for the CV generation, then simply used a bunch of CV Splitters to get enough signals. Better to generate once and copy than to generate each one separately!
I also had a HP and LP filter, but basically the same. I can't remember the discussion since that would have been two years before ReasonTalk was created. Should be interesting if you want to compare notes...