thanks for uploading the cd @jasonw22. Got mine in the mail the other day, very cool to see it all come together.
I realized I had forgotten to follow up on what I did for mastering, so wanted to share a little bit about that.
I started by listening to the record with my notebook in front of me, trying to write down what I thought about the qualities of each track that stuck out to me (and also if any sounds stood out to me in a strange way to look at once I started diving in there). I used these notes along with listening to how the natural transitions would allow the tracks to flow into one another. There were also a couple cases where I asked a few of the participants if they would be interested in doing excerpts of longer form pieces, just to help balance track length across the release a little bit. I also tried to balance the tracks across the two discs (thinking in the back of my mind, since the record is about 90 minutes, someone might want to listen to half at a time to make it a little easier to consume).
Once I had the overall order down, I put things into Ableton. The first disc took quite a bit of time because I was learning a lot about various technical aspects of mastering. Most tracks had a signal chain of Ableton devices: pre-chain utility, eq8s (one stereo, one m/s, one l/r), < 2:1 ratio compressor, (if needed) soft-limiter compressor (high threshold and ratio), limiter (which I later replaced with an oxford limiter I had access to, once I learned about true-peak limiting).
A note on eq’ing: After a while, I decided that my limit of eq’ing was going to be +/- 6dBs in order to not do too much to the tracks. And almost always, I would get the notch sounding right to my ears, then lay off a little bit and increase the q, so it would be lighter touch. Almost always my goal with the eq was to balance (especially with things like the l/r eq, which hardly ever got used), and to help soften any aggressive/painful frequencies.
A note on mono’ing bass: at first I did this using a low cut for mono on my m/s eq, but after a while of a/b’ing, it was changing the nature of the tracks more than I felt comfortable, so I decided against doing it.
For the second disc, I used a trial instance of FabFilter’s Pro-Q 3. A dynamic eq is a fantastically useful thing for mastering in my experience, and this plugin’s approach to allowing an arbitrary number of curves and for each eq curve to control a specific thing (m, s, l, r, etc.) helped me really consolidate the plugins in use. I basically just used a utility, instance of Pro-Q 3, and the oxford limiter as my mastering chain on the second disc.
A note on transitions: On the second disc, I really got the hang of getting solid transitions, I think. I basically started the mastering process on that disc with just doing transitions. I would work to do tons of crazy eq automation for the various segues (I did not apply the +/- 6dB rule here), so that I could match tonality between different tracks, and also pull out/add pressure as the amplitude change with the fade. I set up the automation so that I would have a completely flat eq once the transition was over (and do any of my balancing eq tasks with different bands).
A note on loudness: if you are interested in learning more about this. I highly recommend this SOS article, especially the sections in grey at the bottom that get into the technical details. A thing I didn’t realize before doing this is that you can measure loudness at an instant, over a short time ( similar to an RMS meter) or over the whole track (which is what is used for streaming loudness normalization). I tried out several plugins, but the best (and free) to me was this Hofa 4u meter. Not only can you use it as a plugin on a track, but you can also drag a wav file on it and get the overall tracks LUFS.
One other important thing to note is that mastering engineers do not normalize everything to the exact LUFS value. After doing analysis, I found it is common for albums to have +/- 5dBs of range between various tracks’ lufs levels.
The way I used loudness once I understood for the second disc was as a starting point (I probably wouldn’t do this for a non-compilation-based project, but it worked okay for that). I set everything to -16dB lufs, and then as I listened I moved from that point to get it to feel right. For the nature of these recordings being sort of “experimental” field recording based, some were full spectrum, some were very simple and constrained to a part of the spectrum, so ear was definitely need to get it right.
I could keep going, and I’m happy to if anyone has any specific questions, but I think this covers the main points. If you read this far, thanks for dealing with my rambling hah.