NEWS FLASH!!! Singer-Songwriter/Multi-Instrumentalist Unveils One-Man-Band ROCK Record on the INTIMATE AUDIO label:

1) Mastering Audio, Second Edition: The art and the science (Book), by Bob Katz

Easily my favorite audio book of all-time. I didn't know anything about PARALLEL COMPRESSION, or the HAAS (PRECEDENCE) EFFECT, or why 24 bit mattered (not for the “obvious” reasons, but actually when mixing a 16-bit file in 24-bit) until I read this book. Contains valuable perspectives/realities/consequences on/of the “Loudness War.”

2) Mixing Audio: Concepts, Practices and Tools, by Roey Izhaki

Great book with super specific signal path/setup info to accomplish almost any goal. I learned how to sidechain the bass off of the kick in this book, plus how to do pitch-shifted, super short (panned) delays, and all sort of EQ info. Very helpful if AUX track and SEND/BUS stuff is confusing to you. Comes with DVD (audio files) packed with TONS of stuff. (Link above is for 2nd Edition, which has NO DVD!)

3) Total Recording, by Dave Moulton

Really helpful stuff for Kick drum/bass relationship (I think I also learned how to sidechain the bass off of the kick in this book), acoustics, and using delays. Has visual diagrams which are helpful (for approximating different types of rooms with delays). Comes with CD.

4) The Mixing Engineer's Handbook, Second Edition, by Bobby Owsinski

Contains interviews with numerous mix engineer legends. I didn't know about the importance of mixing at whisper quiet levels (and checking things on a bunch of different systems—and not always sitting in the stereo “sweet spot”... leaving a vaccum cleaner running while checking balances, etc.) until I read this. Tons of other great stuff.

5) Mixing With Your Mind, by Michael Paul Stavrou

This book is great—packed with info that I had never read elsewhere.

The chapter on "uncracking compressors," creating 20sec crossfaded clips of other favorite mixes (to hear differences in overall mixes before your ear has time to "settle in"), the "walking around room with floor tom," chasing the flame perspective, among many others, helped me immensely.

MY PERSONAL RECORDING PROCESS
When recording MANNERISMS MAGNIFIED, I did things in the order below (playing & recording myself), primarily because I had yet to fully realize all the "parts" for each instrument (all songs were composed as just "acoustic guitar" and "vocal")

1) Acoustic Guitar

2) Scratch vox

3) Electric Bass

4) Electric guitar(s)

5) Piano on a couple tunes

6) Instrumental extras (mandolin, accordion) on a couple tunes

7) Drums

8) Real Vox (scratch vox were done waaaay earlier, step #2)

Don't add ANYTHING to tracks you're tring to build without referencing it against a scratch vocal! Meaning: Don't neglect to record a scratch vocal as early on in the process as possible!

DRUM RECORDING
On MANNERISMS MAGNIFIED, I had to record drums "last" (well, again, "real" vox were actually "last") to (#1) ensure I didn't over-do things with any cymbal activity, fills, etc. in inappropriate places, obliterating any vox (they were already in "scratch" form when I laid down the drums), and (#2) to give me the most "lead time" to create cool drum parts & practice them so they were (hopefully) interesting and pretty tight when the time came to actually record them.

I recorded most of the other parts with a click, but had to leave it out (except as a "count in") when I did the drums.... It just felt too weird for me. I jacked up the acoustic guitars (if it was a strummed, percussive bit) & bass in the mix, which helped a bit. I also had to "punch in" a lot of sections. I used those Extreme Isolation headphones to kill off the cranked cans' bleed in the close overhead mics (I used the Recorderman mic configuration, detailed below, plus other mics).

Since I was doing this totally by myself (button pushing included), I think the end result--the parts themselves (not talking about sound), and the way all the instruments/voices interact together--ended up being better than if I did things in any other order. The DRAG though was, since I sort of naturally "lay back" a bit when I play guitar, and I guess, sing, the drums were often a *hair* late; I had to do pretty substantial "shift" edits. Tedious as heck!!!! But worth it in the end, for me. I don't know if I'll ever do that again though, for so many songs.

Below is a YouTube video (not me in the video—just a tutorial that's pretty clear; it helped me a bunch with the mic approach I used). It's called the RECORDERMAN OVERHEAD DRUM TECHNIQUE:

I used that, plus a single mic 3 feet in front of the kick (blended in w/Recorderman setup), and accent mics on kick, snare, and toms, and have been pretty happy with the flexibility. The snare & tom mics are mostly used as different/controlled "ambience" generators (sending them to different 'verb/delay AUX tracks and automating the "wet" track for ambience). The photos below show what kind of soundproofing I used (most important of which, I feel, was the “cage” I constructed over the kit, which reduced the reflections off the ceiling into my overhead mics).

I pan the Recorderman L & R tracks outwards at about 3 and 9 o'clock, audience perspective, and it seems to fit okay, at least for what I'm trying to do. SEE PHOTOS BELOW:

VOCAL RECORDING
Everything below, I use for tracking either myself, vocally, or someone else. Tedious as L, but not tooooooo crazy (gets faster w/practice):

1) I have the artist make a two copies of the lyric sheet for him/her and myself.

2) I may have them note where to breathe on that lyric sheet, if I know in advance there will be edit problems. Or how long certain notes are sustained, for that same reason. (Same with vibrato, if it's going to be doubled later... which I will make from the comp'ing process.)

3) I put a strip of masking tape on the floor as a marker, so the artist will never move from that one spot.

4) In my experience, punch-ins don't work well, in retaining/matching tone quality/resonance, dynamics, vocal character... (NOTE: I didn't say I don't use punch-ins "AT ALL," but "rarely," and that, for me, punch-ins don't work as well as comp'ing.)

5) At minimum, we'll track a verse in its entirety (maybe 5-10 times, who knows), in loop record mode, with a few spare bars at the beginning/end of that section. (Maybe a verse, leading into chorus, depending... Maybe the whole song, but usually not.)

6) I find it takes *at least* a pass or two--or three--till the vocalist can get "into character." Rarely ever is it a "first take is a keeper" situation.

7) As takes are going down, I'll (quietly) take notes on the lyric sheet, marking stuff that's particularly good, and stuff that's not yet nailed. And I'll indicate what "pass" that take was.

8) All these passes sit in a takefile. When done, I will take each one out and put it on an individual track so I can see all the takes stacked above one another. This makes it very easy to see each phrase, and any timing issues between takes, including the length of sustained notes.

9) I play each one (first checking out the ones I thought were good, based on my notes during tracking) and delete everything that sucks (sound quality, intonation, etc.). Or delete individual words, in the midst of an otherwise good phrase. If something is great, but the timing is off, I keep it and shift it, if needed. If a last note of a phrase needs to be longer, for whatever reason (maybe it will be doubled), on occasion (if no takes are good enough) I've time stretched the tail end of that word.

10) I put together whatever is left into the best possible composite vocal (its bits dragged to a separate track as I go), picking/choosing "the next word/phrase" while in loop playback mode, so I hear the "edited" vocal, then the next possible phrase from a surviving take (kicking the other takefiles in/out w/mute buttons). This part is easily the most tedious. After that *maybe* auto-tune a bit if needed, but usually this process eliminates that need.

11) If something needs to be doubled, I audition the leftover bits against the edited vocal, and find the best blending/resonating parts. Or similarly timed vibrato parts. I rarely have someone "sing along" with a take to get a double.

12) Crossfade and move on!

ON COMP'ING VERSUS PUNCHING-IN: All of the above is in the name of insurance, if a singer goes on an improvising burst and wants to try stuff, if a decision is made way later to double a part, etc. And as for doubling... I've had HORRIBLE problems with people who can't sing along with their own vocal to do doubles, or punch-in and nail the exact vocal quality of the phrase that needs a punch in--let alone, sing as well as they do "live" at a show, but with headphones on. You work any way you can with who you get, so the project moves forward, and the artist/producer is happy with the result. If they don't need the help, then obviously all this "comp" stuff doesn't mean anything. But it may help you if you end up with a billion tracks of vox you need to sort through.

Personally, *I* have ALWAYS gotten a better result with assembling a double from alternate takes. Well, at least 90% of the time. That's just me, I guess... In *my* case, even if the singer is VERY good and consistent, the resulting double can be BETTER sounding than if they sang along with an earlier "perfect" take. Ends up sounding naturally balanced, because they sang the exact same thing twice, hearing the exact same thing in their cans both times. The resulting blend is often better. Whatever works--so that it sounds the best (and natural)--is all that matters.

SCREAM ENHANCING/THICKENING TIP: Blending in a "whispered" scream often helps, in addition to things mentioned above: doubles, squashed/tweaked clones, and some level of gain introduced... via a send or applied to any of the above.

MY MIXES: Generally all single-tracked lead vox were sent to a plate reverb, and two (buried) timed delays (different note values, w/LPF). Sometimes I added a manual “doubling effect”—two copies of the track, hard panned, with one transposed one up 9 cents, the other down 9 cents, then offset them by 14 & 19 ms, if I recall, their faders brought up a touch for spread. Double-tracked vocals were generally either sent to a Waves SuperTap delay with a subtle chorusing effect, or two mono, panned reverbs (two similar versions of a TrueVerb preset).