John Gillespie Project, Day 2

It looks like John and I will be recording every Monday we can through out the summer.

On this second day of tracking, we recorded some acoustic guitar overdubs, lead vocals for five tracks, and bass guitar for three tracks. We also completed the drum mapping for one complete song.

Above is a miniature schematic of my pre-amp/compressor/eq settings for John's voice and John's bass. Often when recording, many questions arise: When do I compress something? How much do I use? What about those EQ knobs? Honestly, the best thing to do is to do nothing. If you don't understand what compression is, don't apply it. If you don't know what effect turning the 5KHz knob will have, don't turn it.

I have a conservative opinion of compression. Some producers are able to squash the mess out of a signal and make it somehow sound bigger. Because the nature of compression is to make things smaller, I generally don't use it. Faithfully, I use a limiter setting on any vocal, just because the recorded voice is much different from the natural, dynamic character of one's voice.

The difference between compression and limiting is in the ratio. If something is set between a ratio of 1.1:1 and 10:1, it's considered compression. If something is set between 10:1 and up, it's limiting. Of course, the ratio doesn't matter unless you know where to set the threshold. The threshold is the decibel level at which the compressor/limiter starts to apply its ratio. For example, if I were to set the threshold at -10 dB and a signal came through that was -8 dB, if my compression settings were 2:1, that signal would be reduced by 1 dB. Until you understand this, go ahead and leave your compressor out of the signal path.

Also, the attack and release time are important considerations. Some boxes don't have these, but some do. I generally set my attach and release times at "fast" and "fast" for most things. There are advantages to setting slower settings when it comes to the snare drum, kick drum, and bass guitar. This is where you can get a cool "pumping" sound that's in rhythm with the song. It takes practice to set this correctly.

EQ is an important consideration to apply before going to tape or hard disk with a given instrument. One reason for this is that EQ applied after the fact increases the wordlength of a digital signal. Since it's all one's and zero's with Pro Tools, you'll notice a degradation of your track the more plug-in's you apply. These machines are taking giant numbers that aren't necessarily divisible by each other. There are many numeric remainders that end up floating around in the signal. If you're using analog EQ's out of the computer, you don't have to deal with this. But still, once the signal is set, cutting out a frequency raises it's opposite on the other side of the EQ graph. The most important reason is that it's best to get this step right the first time. Again, if you don't know what ranges are represented by a given frequency, I'd leave the EQ out of the signal chain. If you have good instruments and decent microphones, there's probably not much reason to change anything.

For John's voice, I rolled off at around 100 hz on the preamp. I also set a basic limiting for his voice. This helped me get the signal as loud as possible without going over zero. A limiter set to 10:1 at just the right threshold almost insures that. For EQ, I had some issues with sibilence last summer with the Siren Child EP on his voice. With a spectrum analyzer, I was able to determine that the problem frequency for his voice is around 6.6 kHz. So, I rolled off a few dB here. On one sibilant song, a de-esser plug-in was able to catch some offending syllables. I also raised the 32kHz frequency with a hi shelf eq. This is something that the Avalon manual recommended. It helps give the vocal a little more air.

Bass is an instrument that most people agree should be compressed. I used a 4:1 ratio to clamp down on the notes. Bass compression helps to make the low end ring out as long as possible. I adjusted some EQ settings, too. I pulled some of the frequencies where acoustic guitars will live, I also boosted a little around the 1kHz range to boost the presence. Generally, cutting frequencies is preferential to boosting frequencies.

Beyond the technical aspects of this given session, I have to mention that these songs are really incredible. I've been a fan of John Gillespie for more than half of my life (I bought The Bluest Sky when I was 15). I can honestly say that some of his best work to date is represented in these songs. One song "Afternoons" is particularly striking. I woke up with it in my head this morning.

Also, John's singing was very well done. He normally took no more than three takes to get a master take we both liked. At the rate he is able to work, I feel that we'll be an extra day ahead of schedule from what I anticipated.