[Ardour-Users] Plugins for mastering: suggestions ?

Jörn Nettingsmeier nettings at stackingdwarves.net
Tue Sep 8 04:26:52 PDT 2015

On 09/07/2015 05:33 PM, jonetsu at teksavvy.com wrote:
> Hello,
> There are quite a few plugins out there.  Many are Open Source, some
> are commercial, like the OvertoneDSP plugins (between $30 to $120).
> I would like to get into mastering tracks, at least to see what it is
> about and what can be done.  So far I only recorded tracks, each track
> with plugins, sometimes not, and I was happy with this.  Now's the time
> to go one step further which means to read about it, and to try
> things.  Certainly I do not want to spend money now.
> What are your plugin suggestions for bus mastering ?  Not only by type,
> but also by specific name, which ones are you actually using ?

The biggest reason for mastering is to tap the experience and ears of a 
mastering engineer who hasn't been through the grind of mixing and 
offers a fresh perspective.

Conceptually, mastering is a kludge, using elaborate but invasive 
techniques on a two-channel master that, for the most part, could much 
more effectively and subtly be done to the individual tracks of a 
multitrack session.

Mixing is not a black art, but it is comparable to fine mechanics. 
Everything ties together like clockwork. Tweak one thing here, something 
else over there will need attention, too.
If you have a big duffel bag full of wrist watches, some of them will 
certainly have problems.

Why is it that for musicians it feels perfectly natural to seek out 
someone to bang on their bag of watches with a big hammer in the hope 
that in the end, every single one will be working perfectly?
Or to buy special bag-whacking hammers for a premium?

Just open that bag already, look at the problem watches, and fix them.

There is a reason for mastering: it enables a sane workflow and gets the 
job done. At some point, you have to just live with your mixing 
decisions and move on, otherwise the track will never see the pressing 
plant (or the ears of your fans). But don't expect magic.

When mixing, you work on 48k/24bit or more, in a controlled studio 
environment, with reference speakers. What could sound better? If that 
doesn't sound good, work on your mixing skills. The mastering engineer 
will try to translate your mixing _intention_ to a limited medium, be it 
vinyl, 44k1/16 cds, laptop speakers, or FM radio with car noise 
background. She or he will not magically make it sound good. There is no 
magic to mastering, no matter what the plugin vendors are telling us.
There may be magic to good mastering engineers, though. But they don't 
/like/ whacking bags. They only do it because that's what they have to 
work with.

Sorry for evangelising :) But I think the question you should be asking 
yourself is: what does my mix need? It's really hard to find that out. 
The solution is to listen to it with colleagues, to get outside 
opinions, to learn from each other. Once you know what it needs, ask 
yourself: Can the required fix be applied to the master without too much 
damage? Then do it, it's easier that way. If not: fix the individual tracks.

If you want to make stuff loud, mix each track loud.

The magic mastering wand of the decade is the multiband compressor. Why?
If you use a compressor on a master, the loudest noise will duck 
everything else. Your singer will become quieter with every snare drum 
hit. Bad. People used to use elaborate sidechain EQ to fix that (to make 
the compressor follow the voice instead, which is really hard to do). 
That created engineering legends, and legendary engineers. Now with the 
multiband compressor, the loudest noise in each of the four or so bands 
will duck everything else. Less bad, but still kind of stupid, no? Now 
the legends live in the advertising of Waves, TC, and others. If you 
only have the master (and a client with a deadline), the multiband is a 
godsend. If you have time and the stems, compress those for god's sake.
Not to condemn group compression completely: it can have benefits, such 
as to "tie together" a very dynamic drum kit or percussion set, or a 
choir. But in the master? Meh.

If you want to work on the voice, work on the voice _track_.

The magic mastering wand of the last 5 years or so is M/S processing. 
Why? Simple: for pop, the voice is all. As long as the rest doesn't suck 
too bad, all is well. How to get at the voice? It's dead center, 
usually. If you can affect only stuff in the center, the openness of the 
mix (= the difference between the loudspeakers, or what's there to 
actually create a stereo image) won't be affected so much. So EQ that M 
channel until the talent becomes bearable. Fine. Again, ok if the master 
is all you have. Otherwise why bother? There is no magic in M/S 
processing, it's a tool.

If you want to work the bass, work the bass and kick drum tracks. Don't 
slap a "subharmonic synthesizer" on your master that adds organ-like 
undertones to instruments that shouldn't have them, or screws up 
transients in funny ways.

If you want to design transients, work on the transient tracks. Don't 
slap a transient designer on your master that will make the snare sound 
great but screw up the vocal consonants. Or selectively reduce 
transients on _just_ the reverb track for a creamier ambience, don't 
turn everything into a toothless mush.

If you want to add "sparkle", add it to the things that naturally 
sparkle: cymbals, synth pads, even voice consonants (with great care!). 
Don't just patch an exciter into your master bus by default that 
indiscriminately adds distortion to the treble range of everything. (Ok, 
fortunately, that one hasn't been in fashion for a long time. Knock on 

Ok, now to answer your actual question (sort of):

Why I master:

* lack of budget, or
* distrust of cheap mastering service suggested by client (if it's some 
guy with garage band and a huge ego)

Whenever I master myself, I consider it a compromise.

How I master:

* create an ardour session with as many stereo tracks as the final CD 
will have
* arrange them one after the other, carefully tune the silences and/or 
crossfades and roomtone for a good listening experience, leave breathing 
space to dramatical endings, or cut in surprisingly with an upbeat new 
song, or whatever.
* adjust the fader on each track to create a pleasant listening 
experience with an enjoyable, wide dynamic range without forcing you to 
adjust the volume constantly.
* listen for the overall sound of each track and identify how it maybe 
clashes with the rest
* fix that, either by going back in the mix, or with some gentle global 
processing if possible, usually just EQ.

In my master track of that master session, you will find, in the order 
of sound improvement potential:
* an EBU 128 meter to make an informed decision on the relative levels 
of tracks
* an analyser, to view those problems that my speaker system or room 
might not be able to reveal
* a hard limiter set to -0.3dB, to reduce the headroom to sane levels 
(usually it's Fons' zita-dpl1), used such that it springs into action a 
few times per song, at most.

I then create CD track markers in ardour and bounce the whole thing. 
Then, the final master gets wrapped into a DDP and sent to the pressing 
plant. I use Andreas Ruge's tools for that: http://ddp.andreasruge.de/

All best,


Jörn Nettingsmeier
Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487

Meister für Veranstaltungstechnik (Bühne/Studio)
Tonmeister VDT


More information about the Ardour-Users mailing list