If you read the earlier posts in this thread you might have noticed that we finished the main recording session for One Of Us back in 2008 April. So why is it being released now in Nov 2009 you might ask? Funny you should ask 🙂
The astute reader might have observed a subtle hint of burnout when we were done with the recording. Or you might have smelled the burnout several blocks away. At the end of any recording project the last thing we want to do is listen to it. I want to play my guitar, listen to some other music on my play list, do some other stuff, and Ramona would rather play,teach or do her web design. Anything but listen to our shit! Sooo, we basically came back home, I backed up the recording, then I put the disk on the shelf and we went on with our lives. I got really busy doing SONAR 8, and then in late 2008 we took a vacation to India. Then it was back to SONAR 8.3 season, and NAMM and …. And before we knew it literally was April 2009. We had a polite message from Ingrid asking – “hey guys what happened to that recording”. Huh, which recording? Uhhh, oh that one!!! Anyway one fine day in April I woke up and said nuff is enough we need to finish this.
So I cranked up the disk. Thankfully it started, phew that would have been a problem! I opened up ProTools LE, got some advice on how to navigate the sessions (again the astute reader might have observed that I am not a PT fan by now), and imported all all the multi track wave files to SONAR. Ahhh, comfort zone 🙂 All the tracks were recorded in one giant session as is customary with many recording engineers. I split them up into multiple project files in SONAR so that I could mix them independently as I prefer. The original sessions had 8 tracks for drums, 2 for trumpet, 2 for bass, 2 for piano and 2 for guitar, for a total of 16 source tracks, all recorded at 24/96. Pretty simple to start with.
I set up basic gainstaging and pan – I like to mix in offset mode in SONAR so all my gains and pans are set as offsets. I then load up the first project in SONAR and press play. Whee – this is literally the first time I am hearing this music in close to a year. It sounds fresh and much nicer than it did a year ago. I am excited! The trumpet and piano and bass and drums sound awesome. But wait a minute – what is up with that guitar sound? I must have messed up the EQ accidently somewhere. I look – nope, everything looks ok in SONAR. It sounds a lot thinner than I remember it sounding in the recording session. If anything I normally err on the side of thickness not thinness with my sound. Hmmm.. Took me awhile before I finally saw what was going on. I solo the miced guitar track, thats the one sounding thin. Finally, I put two and two together – its the darn downfiring woofer on my Acoustic Image amp. The mic didn’t pick up any of that. Man was I pissed off – the miced amp track is unusable. Why the engineer didn’t spot that at record time is a mystery. A major disappointment on my first day of mixing. Anyway, I have the direct stereo out from my VG-99 recorded as a backup (thank God for that) so I figure I can re-amp my guitar in my studio and fix it where necessary. Note to self: When tracking at someone else’s studio always, always check the printed signal to make sure its what you want. Don’t rely on what you are hearing in the sound booth. I kick myself for not being thorough about that.
So begins day 1, the first in a series of many others over the rest of the summer of 2009. I spent most of my weekends and nights working on additional tracking, editing and mixing of this project. Ramona had several new parts that we needed to record. We intentionally didn’t track the extras in the studio to save time. I track all the additional piano and guitar parts in my studio and reamp my guitar tracks where the sound is not what I wanted. Guitars sound vastly nicer than the studio original recording, since I used my other (non downfiring) speaker cab!
While working on stuff in my home studio, for the longest time I was plagued by some crazy RF interferance I was picking up when using the balanced outputs from my VG-99 to my MOTU. It was the strangest thing – every day after a certain time 8:30pm to be exact, I would get these radio stations mixed in with the recorded signal! So I’d have to stop tracking after that. I tried everything I could think of – new cables, relocating gear, new power conditioning power strips, etc, but nothing would get rid of that. After several days of this, I finally found the root cause of it. I had the sub out’s from VG at +4dB going to my MOTU balanced inputs, and I had the ground lift switch engaged on the VG. That in intself was fine – the root issue was that the MOTU 828 has a software switch in its control panel to set the inputs to -10dB or +4DB. I had missed that that switch was set to -10dB. I flipped the software switch in the panel and voila the RF vanished! Its critical to do this if you are using balanced outs since otherwise you potentially get all sorts of noice induced in the circuit. Anyway that was a big finding for me!
Later on in the process once I have all the 10 songs edited, I do basic mixes then recruit the help of my friend and colleague, Dan Abreu from Cakewalk to track the vocals on this project. All the vocals were recorded in his project studio Pennyco Productions, where we spent many nights doing the initial mixes. Dan’s input was super helpful during this phase. He has great ears and a lot of experience with mixing drums, so it was great getting his expertise in the initial mixing phase. We’re especially thankful to him, since he was so pressed for time preparing for the birth of his baby daughter Penelope. Once we have the core mixes done at his place, we transferred to my studio for all the tweaky time consuming nit-picky stuff.
Work and other engagements made this take longer than I expected, but by Sept I have mixes that are pretty close to what we think are finished. Or so we thought 🙂 Just a few more tweaks and we would be ready for mastering. The devil is in the details however. I don’t have a treated room, so we test the mixes in several locations and on multiple speaker combinations. One of the goals I had for this project is to ensure it sounds good on multiple delivery systems, audiophile equipment, consumer stereo’s, car stereos, desktop computer systems, and the ubiquitous iPod’s with ear-buds. I also solicited feedback from some experienced Cakewalk users to get more ears on the mixes. This is super important since when listening to mixes a hundred times you can adapt to a sound and miss the obvious. I get lots of great feedback on the mixes which help me fine tune them. Special thanks go to Cakewalk user (and fine guitarist) Eric Hansen as well as Tom Jacobs for their valuable input. During this process I go back several times and make adjustments to the mixes to account for something or the other. Acoustic bass, as I learned, can be one of the hardest things to get just right in a mix since the subtlest of changes can result in you over or under compensating other tracks. I would run into cases where the bass sounded great on ear buds but woofy on a stereo or vice versa. The frequencies that make up the ideal elements in the bass sound for a jazz mix can be subtle. I also compared what we had with several reference mixes that we considered to be ideal – caught some good problems during that process as well. I tend to like the way acoustic bass is recorded on some European labels like ECM records with more of the string sound mixed in and less of a thump. After much experimentation, we were relieved when our mixes sounding good across multiple systems. We were finally ready for mastering.
One Of Us: Mastering →[next]