Changes in Mastering Approach

Anyone here do mastering with a multi-band compressor instead of a “standard” compressor? What about a maximizer?

I’m thinking of doing that to squash the kick drum (at the fundamental frequency) and possibly the mids (for guitars) so that the upper end of the frequency spectrum gets some space. And then I may add a maximizer (don’t use one now) to fill out the overall mix.

If you use either…

What settings do you use on the MB compressor?

Where do you put the maximizer in the effects chain during mastering? I’m guessing after EQ but before tape saturation, exciter, or limiter.

Hi Larry, good topic; I hope you get some in-depth responses.

I wish I were any sort of expert. I drop Ozone into one of the last slots of the master channel, and take it from there. But tape saturation is ahead of that, and before that, a bus compressor.

Ozone does it in this order: equalization, harmonic distortion, stereo expansion, then comes multiband compression, then maximizing. Maybe I got the harmonic distortion and stereo expansion reversed.

So in their logic, the last two things are multiband compression and then maximizer. The maximizer includes dithering.

I would say that coming to a compromise between bass guitar and kick drum has to be handled in the mix first. There are others here who have suggested a mix-level low-mid frequency cut to reduce the competition in that space among instruments that have a lot of energy there – guitar, keyboards, as well as bass, getting rid of “muddiness”. I am thinking of a post from Jeff Hayat several years back. Maybe it wasn’t even on this site. But Jeff’s stuff is excellent. I do it myself to some extent. And I roll off low frequencies in guitars and keyboards in individual tracks.

As to MB compression, I find it to be really tricky, and I wish I could give suggestions based on my experience, but I would say instead that I need advice. To my ears, it ends up being a range booster (certainly if you use presets for the Cubase plugin), which is not entirely different than adjusting the EQ band in the first place. So I try to ensure that the compression of each band is somewhat consistent, and the gain of each band is also consistent.

I’ll just conclude by saying I’m not sure what I’m doing with it, but I use it.

Of course there are lots of ways to attack the same problem.

I created a mastering template, 24 bit_48khz… I usually go back and forth from final mixes and masters until I hit the spots… In the master add a compressor, limiter maybe some eq but not likely as I prefer it in my final mixes, are usually below zero and my masters are way above zero… Feel the magic of a pro song produced by Cubase :smiley:

I generally use PSP Vintagewarmer2 in multiband (and fat) mode with a hint of compression as the first in the mastering chain, it works for me 90% of the time just to get the balance right across the bands. Lots of people would call me a heathen for doing so but I love what it does to the sound. Ozone I generally only use Multiband compression to fix stuff I missed in the mix (if I haven’t got time to go back) I do use multiband expanders sometimes which can work wonders on a dull mix but generally put a dynamic EQ in front to tame any massive transients. To my ears this adds a lot of bounce to the mix and can really accentuate the rhythmic elements.

The reason why I’m even asking this is because it was pointed out (in a video :laughing: ) that compression during mastering compresses all frequencies when you may want to compress only certain bands and leave the high end alone so that you don’t lose the overall sheen. I currently “keep the sheen” (#winning!) using M/S EQ (high shelf on the sides @ 2,500 Hz or thereabouts), but I’m interested in this approach especially since taming certain bands may allow me to add a Maximizer to fill out the total frequency spectrum.

Here’s my current mastering chain:

FabFilter Pro-C2
FabFilter Pro-Q2
Variety of Sound SlickHDR
Variety of Sound FerricTDS
FabFilter Pro-L

I’m wondering about replacing Pro-C2 with Pro-MB (which I own) and adding a Maximizer before the limiter.

Somewhat related, I saw another video that described using two stage limiting during mastering to achieve something similar (though not the same) to parallel compression. The first stage limited the upper end while the second stage brought up the bottom “peaks” of the signal. This was all done after EQ obviously so the net result was limiting the dynamic content without making it sound too squashed. Both stages were done using a single instance of vladg’s No6 mastering modular limiter.

My ultimate goal is to take my final results to the next level. I’m still chasing Lenny and Kim in terms of the quality of the result :mrgreen: since they are my personal gold standard in terms of how to properly mix / master a project.

Edit: I don’t know how I neglected to include Jet in that list of gold standard people. :confused:

I think your mastering chain is solid, however sometimes it won’t work and you have to creative again…I guess it’s all about being adaptive when needed.

I watched many tutorials in how to master, there are a few similarities in your workflow…for the rest it’s a subjective matter of opinion and choice depending on what you have, what you use to listen and what room you have and then the experience.

I only use multiband compressors when I feel there is a need to compensate things, otherwise I just fix it in the mix.
But don’t take me to serious about this, I master for fun, the few times projects needed to be mastered it is done by others as I am to biased regarding my own mixes.

my standard mastering chain:

slot 1 - room / reverb (just a little and only when doing reggae/ska, but that my main genre)
slot 5 - eq (waves hybride EQ or T-Racks or Maag eq4)
slot 6 - dynamics
slot 7 - limiter
slot 8 - dithering

that’s it.

mroekalea:

I will soon write a tutorial on my way to use reverb to create realistic depth in the mix without diluting the presence of the various tracks, which avoids putting a reverb on the master buss.

Any particular reason why you put dynamics after EQ? My question is due to the fact that dynamics frequently change the EQ, so you want to modify the EQ after you’ve tweaked things using compressors.

I guess the use of “single band” compression vs. multi-band is really dictated by whether you simply need to control transients overall vs. needing to prevent specific types of instruments (e.g. bass-heavy) from overtaking other types of instruments, resulting in a muddy result. Could this be done in the mixing stage? Yes. Could it be done in the EQ insert during mastering? Yes.

I guess my struggle is in wondering if I can improve my result by changing single band compression to multi-band to control things in a band-by-band way. Maybe I’m over thinking this.

Ooh yes I put EQ mostly in the beginning as I feel that compression sometimes exaggerate frequencies which makes me overcompensate.
But that’s me!

That’s as good an answer as any. It’s often said “mastering is an art, not a science” (or something to that effect).

I had some projects where the EQ sat in the front and in the back, fortunately we’re in a creative business and you could make your own rules :slight_smile:

nice to watch:

I saw this a few days ago. Very nice video, but unrealistic since most people don’t have access to the equipment she does.

I don’t use anything like a ‘standard’ approach to mastering. If you’re doing that, you aren’t listening to what the music’s telling you.

Use the plugins you need, with the settings you need, when you need them - John Scrip - Massive Mastering.

If you have a standard approach to anything in music, you’re just regurgitating. Might as well be Justin Bieber.

The new “Lurssen Mastering Console” software seems quite interesting…

I agree that you can’t use the same settings.

But if the project being mastered is your own meaning that there is some expectation of consistency in the results between it and previous projects then I disagree that you cannot expect to use the same plug-ins. 99% of the time, my mastering chain is the same (with differences to the compressor, EQ, and limiter settings), though there have been the exceptions to this.

Interesting read. Reminds me of iZotope Ozone.

Larry, going back to your earlier comment, I think the aim of compressing across all tracks (at mixing stage) is to have all of the tracks, across all bands, smashing together, so that if it’s the guitar that’s too spiky, it will get tamed, but if in another passage it’s the kick that’s too spiky, it will get tamed. It’s been referred to as “glue”. It’s the first place in the flow where the tracks interact through a process. If your tracks are such that one track is consistently too loud, then this compression is possibly somewhat defeated, as it will consistently only operate on say, the kick. Which I guess is why this whole mixing process is called an art!

But later, in “mastering”, after being more or less satisfied with the overall balance all the way through, and after having done equalization at the master level, there might be something to be clarified within each band range, i.e., the kick is still spiking in the lowest range and masking the bass guitar there, or the guitar is competing with a piano or organ, and that’s what I gather is the point of multi-band compression. Regular mix-level compression wouldn’t have fixed that. And I haven’t even mentioned voice, since I haven’t been recording voice.

Yeah, I guess the principle of using multi-band compression makes sense, but your point is just as valid.