Parallel Compression / Scriptnode, Anyone Done One?

[ustk]

@iamlamprey if you connect the gain reduction to a gain, the only thing you'll get is a cancellation of the compression… automatic makeup is just a gain compensation set in relation with the threshold and ratio I guess

[Natan]

@ustk The Snippet Above Doesn't Sounds Good At All.
And The Dynamic Module's Make Up Goes out of control :(

[Lindon]

@Natan why do you even need to do it in scriptnode?

[Natan]

@Lindon Lindon, The Dynamic Module's Makeup Is Too Hot, I Just Tried To Test If The Scriptnode One Can Perform Better.

[ustk]

@Natan AFAIK they are wrapped around the same code

[Natan]

@ustk Thank you for the information, But The Issue Think is more About makeup Gain.
It's not very Even and way too much.
Tried a Hacked Dynamics.cpp By @dustbro
Didn't Go well here, The signal almost gets Inaudible.

[ustk]

@Natan https://forum.hise.audio/topic/3908/parallel-compression-scriptnode-anyone-done-one/7

[Natan]

@ustk

[Lindon]

@Natan said in Parallel Compression / Scriptnode, Anyone Done One?:

@ustk Thank you for the information, But The Issue Think is more About makeup Gain.
It's not very Even and way too much.
Tried a Hacked Dynamics.cpp By @dustbro
Didn't Go well here, The signal almost gets Inaudible.

perhaps add a fixed value compressor (that moderates the make-up, and is only engaged if make-up is turned on...) after your "real" compressor?

[Natan]

@Lindon
I Checked a few Compressors with make up gain, all perform very Smooth, Like ableton one.
@iamlamprey did a Gain matching, it was close But not so musically.

Not sure how to achieve This, Need to Get some rest and Continue Tommorow

[Dan Korneff]

@Natan said in Parallel Compression / Scriptnode, Anyone Done One?:

Not sure how to achieve This, Need to Get some rest and Continue Tommorow

Auto makeup gain is a calculation based on Threshold and Ratio (sometimes attack and release are added to the equations).
Figure out what your nominal input is and plot a graph depicting gain reduction. That will tell you exactly what your makeup gain seeds to be for a particular circuit.

@Natan said in Parallel Compression / Scriptnode, Anyone Done One?:

it was close But not so musically

This is the difficult part to achieve. Without some standard of nominal input level, your calculation may end up adding large amounts of makeup gain before any gain reduction kicks in.
For instance, a low input level would force your to turn down your threshold to start achieving gain reduction, but a lower threshold also increases gain. And a high input level may start adding large amounts of gain reduction with barely any makeup gain based on your formula.