AW: Powered Subs...on to mastering

["Rainer Thelonius Balthasar Straschill" <rs@moinlabs.de>]

Krispen,

> wonderful on my headphones. There were some really cool and 
> deep things going on in the 44hz range and below, and some 
> others in the 62hz range. It all sounded great through my 
> headphones, but those frequencies were reeking havoc on my 
> consumer stereo systems - car stereo, portable stereo, etc. 
> They were really prominent resonant frequencies that were 
> rattling the hell out of the speakers and causing distortion. 

Simple solution: get a decent car stereo (although my car's Mark Levinson
Amps w/ Infinity speakers and Lexicon processing do some tricky things as
well - that is, the Lexicon processing does it).

> source tracks/wavs, etc, etc. I could spend hours and hour 
> just on one song and still not be 
> satisfied with the results, or waver between two different 
> approaches.   Is 
> there a simpler approach?

The usual approach (as I understand it) is not necessarily simpler but more
effective in the long run:
* get a set of speakers which are good (the typical candidates for active
monitor speakers (e.g. Genelec, Dynaudio) will most probably work).
* learn those speakers. Meaning: listen extensively to stuff for which you
know that it generates problems on other systems, and learn how it sounds 
on
your speakers. Learn how the solution sounds. Basically, learn how a
specific mix "translates" to another, perhaps crappy (e.g. portable stereo,
mobile phone, laptop, AM radio) format.

> I'm wondering what others uses as a consistent approach to 
> mixing/mastering their music.  

One important thing (although not that relevant for your application, more
when using microphones, especially omnis) is to include a highpass filter
*per channel* that cuts everything out which doesn't belong into the music.
If you record a soprano singer, there's no need to record down to 50 (or
even 100) Hz. This can even increase the dynamic bandwidth of your system
(best do it before the converters), as low frequencies contain a lot of
energy.

> threshold do you apply? Do you suck the dynamic range out of 
> your mixes to maximize volume, or are you very conservative 
> and preserve as much of the original dynamic range as 
> possible, sacrificing some volume. 

For me, there's a simple rule which will you give better results
immediately: use a compressor/limiter not to increase loudness, but to make
it sound better. For that, A/B listening becomes rather tricky, because (if
you're listening at lower levels), the ear perceives "louder" as "better".
So, analyze your source material's loudness (RMS will do), then set up your
compressor, set the output gain so the RMS level is the same as before and
compare to the unprocessed signal. If it doesn't sound better, don't
compress.

Now, this may be a problem if you're targeting a market where maximum
loudness is a must - are you targeting that? Basically, this market is
youths with baggy pants in pimped rides. (FM) Radio stations are NOT that
market, simply because they use processors in their signal chain which will
already make the material more punchy and, if necessary, more compatible
with the limits of FM radio. Actually, there has been a discussion within
some US FM broadcaster's organisation to specify maximum RMS levels for
material to be accepted for airplay.
Didn't find that very article, but two others instead:
http://www.austin360.com/music/content/music/stories/xl/2006/09/28cover.html
http://www.orban.com/support/orban/techtopics/Appdx_Radio_Ready_The_Truth_1.
3.pdf

> package.  Ever use Waves MaxxBass? I read some articles that 
> recommended it during the master process, but I did not like 
> the results. It altered too many other frequencies in my mix 
> beyond my original intent.

Can't understand who would suggest that. It works fine on individual tracks
(bd, bass) though, if you're targeting an audience with challenged 
listening
equipment.

> I suppose if I wanted to be a purist, I 
> would only ensure my overall level is at or close to 0db, and 
> not apply any compression whatsoever...because once you do 
> that, you are already altering the original dynamic range of 
> the piece. 

Define "original dynamic range". I believe the thought of an original
dynamic range of the piece for electric/electronic instruments is somewhat
misleading. It's clear for e.g. a concert grand - it's just the sound you
hear when sitting at a distance of about 10m from it in a 50m2+ room. That
would mean that if you close-miked that grand, you had to apply 
compression,
because the mics get hit by a lot more dynamics than the listener's ear.
Then for your guitar: if you adapt your playing and/or sound settings to
what you hear during playing, then the original dynamics is what you hear
while playing (i.e. the signal chain from guitar through processing through
amp through speaker through air), not what you probably are recording
directly after the processing, and again, the dynamic range (and also the
presence of some high or low frequency content) at that point in the signal
chain will be much higher.

> Any thoughts or best practices would be appreciated here on 
> how to be both a sound source preservationist, yet a playback 
> friendly sound engineer at the same time.

Ok, here's another thought: take a recording you like. Rip the CD (or take
the MP3) and load it into your audio processing tool. Look at the frequency
curves and at the RMS level. Make your mixes and masters to be similar.

Personally, I found that everything that is louder than 'round -12dB RMS
becomes unacceptable for me (at least for the stuff I do, referring to
beat-based passages). 

A trick I like, btw, is (as you referred to the Waves stuff) what they
describe as "midrange compression" on their C1 manual. One album where I
found this approach to be used is "Unspeakable" by Bill Frisell.

Best,

        Rainer