Upward (Parallel) Compression

[valsolim]

Hello, Vojtech,

according to Bob Katz's terminology, your MDynamics plugin is capable of doing Downward Compression, Upward Expansion, and Downward Expansion (through the Gate feature). The only thing which cannot be done directly within MDynamics is Upward Compression - bringing the quieter passages up in volume. Since this technique is being very commonly used especially for processing drum tracks - and by the way I know you are a drummer - I wonder whether you plan to incorporate Upward Compression into your plugins? The most common implementation of this idea is parallel compression which, of course, can be done with MDynamics placed into a parallel bus. However, I suppose it would be very helpful if something like that can be done directly within a plugin. The workflow could be faster and the project could be less cluttered.

What do you think about it?

Best regards
--
Miloslav

[MeldaProduction]

Hi Miloslav,

I'm not really sure what you mean, because as far as I know, but features are already there. First upward compression increases level when below threshold, right? Then all you need to do is upgrade to 5.00 , and then set a processor into an expander mode (the arrow in its title) and set ratio below 1:1. You can also draw the shape if it doesn't fit your needs.

And parallel compression is nothing else than a dry/wet .

Cheers!

[manducator]

I haven't used Mdynamics but for upward compression (which is parallel compression) you just need a dry/wet button to mix the uncompressed with the compressed signal, and I see that Mdynamics has a dry/wet knob, so do you know fur sure it doesn't work?

[valsolim]

I'm not sure whether my technique is "correct" or not, but I'm used to control the parallel compression with send levels, instead of adjusting the level of the output bus. But the wet/dry mix fader affects the output of MDynamics, not the processor input.

I didn't realize that I can draw my own curve, so I will definitely try that, thanks for the suggestion! But Upward Compression is not Expansion, it's more like "negative gate": actually, it would be a gate with its curve mirrored with respect to the diagonal. (But, of course, it couldn't be a straight line since complete silence cannot be effectively rised up in volume.)

So, I will try to draw some negative gate and will see whether it works. Regarding the parallel compression, for the full possibilities of this technique I would need some processor input gain control (as an alternative to the send fader).

--
Miloslav

[manducator]

I have the book of Bob Katz right next to me and on page 133 he explains:

"New Zealand radio engineer Richard Hulse discussed with me his practice of parallel compression, which is a means of creating an upward compressor".

A few senetences further, Katz mentions this:

"The principle is quite simple: Take a source, and mix the output of a compressor with it."

So you can do upward compression in 2 ways; you make a send track, put the comp in there and mix the compressed sound with the original uncompressed sound.

Or

You can use the comp as an insert, which blends the dry signal with the wet (comressed signal).

I use both techniques a lot.

So if you use a send/return track, you can do parallel compression with any compressor.

EDIT: what you call "neagtive gate" is upward expansion, according to me. It's just a mix up of words.

Downward compression= compression as we know it; bringing peaks down.

Upward compression; mix the compressed sound 'underneath' the uncompressed sound. That way you get the compressed sound but without loosing the peaks.

Upward expansion: making peaks louder, the opposite of downward compression. Sometimes engineers use this technique to try to eliminate bad downward compression.

Downward expansion: = gate, it brings low level passages further down.

[MeldaProduction]

I'm not sure whether my technique is "correct" or not, but I'm used to control the parallel compression with send levels, instead of adjusting the level of the output bus. But the wet/dry mix fader affects the output of MDynamics, not the processor input.

This doesn't seem right to me. As far as I know the concept of parallel compression works by mixing dry source with a compressed one - thus sending ALL of it to the send with compressor and then controlling the amount of output from the send. This way you control how much of the signal is compressed and how much is not.
I think with you case you get pretty similar results though - you are taking all of the output from the send, right? And you are controlling how much you actually send to the send (this sounds funny ). So at the end you are controlling the amount of compression anyway, you just make it a little more complicated, because you have to adjust the levels afterwards, and with low levels there is no compression at all, so I think it is quite harder to use.

At the end I think with MDynamic's dry/wet (which does what I believe is traditional parallel compression) you get similar results in much easier way.

I didn't realize that I can draw my own curve, so I will definitely try that, thanks for the suggestion! But Upward Compression is not Expansion, it's more like "negative gate": actually, it would be a gate with its curve mirrored with respect to the diagonal. (But, of course, it couldn't be a straight line since complete silence cannot be effectively rised up in volume.)

Don't think about the terms too much, they sound complex but it's all pretty simple. Take MDynamics, set ratio to 2:1, it is an upwards compressor. Change the ratio to 1:2 (very low value!!), now it is an upwards expander. Enable the expander mode and it will do the same thing, but to the levels below threshold - with 2:1 it is a downwards expander, with 1:2 it is a downwards compressor.

See? There's basically no difference between compressor and expanders, just the ratio is different. The expander mode is more like "downwards" mode, but most people don't like the ratios below 1:1, because they are quite "dangerous" (you can very quickly get very very high levels).

[valsolim]

Okay, I wasn't aware of the expander mode, so this should be basically what I was after, I will try that. Thank you both very much!
--
Miloslav

[valsolim]

I've just tried the MDynamics processor in the expander mode as you suggested and for the ratios below 1:1 it really works as an upward compressor. But since the minimal ratio is 1:3 the effect of upward compression is quite moderate and insufficient for most applications. I understand that the ratio of 1:3 is definitely enough for the default "compressor mode" since in this setting the effect of upward expansion is really huge. However, I have the feeling that this is not enough for upward compression.

What do you think? Could be the range of ratios for the expander mode extended below 1:3? Ideally all the way down to zero?

--
Miloslav

[MeldaProduction]

I'm afraid it won't be possible because of the backwards compatibility Miloslav. Also I don't really think this is a good idea - think about what you are actually doing. Say you place threshold at -20dB and ratio 5:1. Not feed in a drum track which goes from -100 dB (between the hits). Then you just made -20-(80/5)=-20-18=-38dB from the -100dB ! This means you amplified the noise or DC or whatever there is, a lot!

[valsolim]

Okay, after some playing around with the settings, I was able to achieve what I wanted by means of the default "compressor mode" with adjusted Range parameter. Actually, this was what confussed me before because especially in the expander mode there is a huge leap in the amount of compression when you change the Range parameter from the initial 96 dB to a bit lower value.

So, thank you for the enlightening discussion. Your plugins really offer endless possibilities!

--
Miloslav