It worked out well that I’d laid all the foundations for the mastering session on Monday, meaning that when Tuesday’s session got started I was free to focus on listening right from the start (and not fiddle about with more menial tasks like file importing and naming tracks etc). When there are lots of different tasks to perform it’s often a good thing to arrange the work so that it allows for a strong focus and concentration on the most technical (or creative) element. In the case of mastering there’s a lot to listen out for, and a lot to pay attention to. Rather than the microscopic investigation into the interplay between different instruments within the same frequency range during mixing, such as tweaking the tonal and volume balances between a kick drum and bass instrument, or altering the depth and breadth of lead instrument doubles, the view over a mastering session is one that requires a step back for perspective.
I already had in mind some of the changes I needed to make to the entire project (such as reduce the bass and bring up the treble a touch). The way I set about deciding what to work on is by listening to the album on as many different sound systems as I can, that I’m very familiar with (in this case there’s the studio monitors/subs, two sets of studio headphones, a portable sound dock, a sound stick/sub system in the kitchen, the sound bar that we use in the living room, the in-ear-monitors I use with my iPhone, and the car stereo). I also set up play lists of tracks that are either just right (in this case some material from The Knife’s Silent Shout), too loud/bright (such as Zomby’s After Light), too bass light (such as Squarepusher’s Plug Me In) and too bass heavy (such as Janek Gwizdala’s Parallel Worlds)*. When I say ‘too such-and-such’ it’s personal taste. I’m not trying to suggest there’s something objectively wrong with these pieces and the creative process that resulted in their distributed form, but it’s important to hold up examples you work towards, as well as some you can use as a boundary, or extreme (even if the extreme only differs from the target by a tiny amount, it’s still useful).
Even though I bought SoundBlade several years back I find it still far too buggy and glitchy to function as a reliable mastering DAW. On paper, and sonically too, it sounds fantastic, but for the purposes of making sure my day was just that, I decided to work within Logic Pro. I laid out the tracks (two stems from each tune) across a time line. This meant that I didn’t have to mess about with muting/soloing when jumping between tracks, and could also lay out the timing of the album, how one track flowed into another, or how long the silence was between pieces. The decision to use two stems was one I took so that I could brighten the ‘everything but the percussion’ stem more than (and in some cases instead of) the percussion stem.
I used Brainworx’s digital EQ on almost every track, set to M/S mode. One or two needed some gain adjustment so I used Izotope Inc’s Ozone 5 (overkill maybe, but it let me dial in the gain quite precisely). I set up two busses that had the following plug in chain: Ozone 5; Ozone 5; Dynamic Range meter. The first Ozone 5 on the busses applied a light bass cut, light treble boost, light single band parallel compression and a touch of limiting. The second Ozone 5 applied 4 band stereo imaging. Finally, the dynamic range meter allowed me to visually check the RMS values and dynamic ranges of each piece to ensure that album was well balanced, loud in the right places, and quieter in the right places too. The second bus was a duplicate of the first, except without the parallel compression (which was important so that I didn’t reduce the dynamic range of pieces that were naturally hitting the target range). For the most part the pieces were hovering between DR14 and DR12 (with the RMS values sitting between -14dBFS and -12dbFS and peaking at -0.5dBFS). At the loudest sections (of which there are only a few on the entire album) the range dropped to between DR10 and DR8. There’s a lot of electronic music out there that is produced to be loud (what this means is high RMS level, low Dynamic Range). Whilst I’ve experienced the problem of The Smiths following Alt-J (for example) at a cocktail bar in the city, and the drop from somewhere that sounded like DR8 to probably somewhere like DR14 (which did make The Smiths sound a little flat), I’m not interested in squashing my work into a state of competitive loudness. The end result of the loudness war is lower fidelity music. I finished up the mastering session having decided to review the work the next day.
Wednesday started with listening through to the album start to finish through my studio monitors, before listening again through headphones. While I was listening I took the time to work on the timings in a little more detail. To my surprise there were a could of clicks in one of the pieces so I used Izotope Inc’s RX2 to flawlessly remove them. Feeling happy with how the album sounded, and the timing between pieces, I decided to bounce it down (leaving it in 24bit still) as one single audio file, and make the cuts between tracks in Waveburner, apply dither, and export the regions from there. A bus journey into the city to see an improvised music performance gave me just the window I needed for a final check of the album with my IEMs (in ear monitors, a techie term for really good ear buds, Klipsch Image X10i if you are curious). To my disappointment I discovered another click and a transition between to sections in the same piece that wasn’t as smooth as I remember from previous incarnations. It means that the mastering work is technically entering day three, session three, but it’s very important for me to take care of every mistake now, or live with the sloppy consequences for as long as folk (including myself) pay any attention to this work. In that sense my disappointment at finding mistakes wasn’t really disappointment, it was relief. How frustrating to release work and then discover even a single click that might distract the listener from enjoying the piece to the fullest. The change to the transition is one that I doubt others would notice, but I do, and for the sake of integrity I must take care of it.
So there it is. A little about how things went yesterday, and the day before. As you can hopefully tell, being well prepared, and having taken the time to research the work needed in a session is very important. For all the potential pitfalls with working on your own material, if done with enough care and attention (and skill, and expertise, and in a very good studio environment) I’m confident that great results are possible. I can’t wait to share this work with you, and sincerely hope you enjoy it.
* Not that I’m wanting to try and make a big deal about the ‘mistakes’ in these records, but it’s really useful for me to notice things I’d change with other people’s work, especially when it’s work by artists I have a lot of admiration for, who’ve working in top studios with great engineers. It gives me a further boost of confidence to know that the market place isn’t somewhere of absolute perfection, especially in an objective sense.