Manley VOXBOX - MVB 1996 - 2/2000 and MVBX 6/2000 - 4/2003 - Manual (Page 14)

14
The MID DIP section has a number of typical uses. Most condensor mics have a peak in the frequency
response. It is the normal resonance of the diaphram and is typically in the 5 kHz to 10 kHz range on the large
diaphram mics that we tend to favor for vocals. Sometimes you can often get a sweeter sound by dipping a little
at 5K or 7K and then boosting frequencies above 12K. It may be important to find a good balance between
"presence" or "cut" and "breath" or "air". Sibilance will be focussed on in the De-Esser section. The "Old British
Sound" according to those who created it, was not so much a function of the gear used but often how it was used.
Mids were cut. Pre-mixing tracks and bouncing was a necessity and cutting some mids selectively allowed
"space" for each sound in the bounces. Cutting a bit of 700 to 1K5 reduces the squawk and honk that is all
too easy to get. Another very useful trick is to cut in the 200 to 500 Hz area on some instruments. These lower
mids can sound right when an instrument is "solo-ed" but can "muddy" a mix real fast. Drums, piano and bass
are the usual culprits. Generally, you can get a clearer, punchier sound with fewer mastering problems by trying
to cut in the 200 to 1K5 bands than by boosting highs and lows. Try cutting 300 or 500 on "direct" guitars and
basses to simulate a clean amp sound. Boosting 3 or 4 kHz with the high peak will help too.
The LOW PEAK section is pretty straight forward. The only warnings might be that if you are
monitoring with small nearfield speakers and boosting 20 Hz or even 50 Hz you might be in a dangerous
position. It sounds good now but wait till you hear it in the car. At least check the lows on speakers or headphones
that reproduce deep lows even if most people's systems won't. That killer bottom might explode subwoofers.
Check it and you'll be OK.
The HIGH PEAK is even simpler. We only have two warnings. Analog tape and especially cassettes
can only handle so much highs (especially HF peaks) before a nasty sounding distortion occurs. The other
warning has to do with sibilance and "esses". You should be aware that boosting yet more highs later in the mix
may give you sibilance problems unless initial frequencies have been well choosen.
DE-ESSER. These are used to remove too much highs after somebody put too much on. At least that's
the general consensus. Most de-essers are like a combination limiter and high frequency notch filter. When the
circuit sees "too much" highs it reduces them. The "too much" point is set by the THRESHOLD control. The
offending frequencies are choosen by a rotary switch on the VOX BOX. The "ess" frequencies given are 3 kHz,
6 kHz, 9 kHz and 12 kHz. We also gave you a LIMITER position. The 3 kHz, while low for a de-ess function,
can really help make screamers and sounds that hurt bearable. Our hearing is most sensitive in the 2K to 4K
band. The Limiter is another version of our Electro-Optical Limiter and while quick, it is not a perfect "brick
wall", "zero overshoot", "micro-second fast" limiter - it workspretty damn well, though, and it sounds better
than the "geek" limiters. We don't quite follow why some people insist on wrecking the sound with a
questionable sounding limiter rather than turning the volume down a bit. "More is better" they say. We agree
"less is more".
OK, so back to de-essing basics. Most de-essers are based on VCAs and cheap op-amps which can
do a lot of damage to a well recorded and important vocal track. We build an expensive dedicated passive high
frequency limiter for producers who really care about vocals and shudder to consider a $300 IC based "de-
esser"on the lead vocal. It uses a passive sliding filter based on mastering HF limiters. The VOX BOX uses
newly designed de-esser that not only works great, if used properly, will not mess up the original vocal quality.
It is a passive notch filter combined with another Opto. This allows the fine detail of the original sound to be
left as intact as possible. The de-esser / limiter is after the EQ and before the final line driver. The side chain
is solid state but, as in all Manley gear, the signal path is entirely vacuum tube and never hybrid.
Hints - Don't kill the natural sibilance, just tame it. Natural-like real voice or the produ'er may 'lap
you 'illy. Some vocalists may be "thrown" by hearing the effects of the de-esser while singing. Safest to give
them a straight signal or better yet de-ess in the mix as the last process. Compress (and EQ) before de-essing.
De-essers are not so intelligent that they only react to esses. Loud peaks and high notes may fool them
a bit. If you don't have two VOXBOXes, you might use the De-esser Link output to a spare console channel
to hear the de-ess frequency. The gain will likely have to be turned up. It was not designed to be a sidechain
monitor output. You may only "need" to do this a few times to know what it does.
The best time and way to de-ess is the mix on the final comped (combined) vocal track after all EQ
and compression. It may not be easy to EQ while de-essing. It is far easier to remove harsh sibilance on an
individual vocal track than to have the mastering engineer try to remove vocal esses surrounded by hi-hats,
cymbals, snares, guitars and keys without doing some damage. Boosting overall highs in mastering is easy but
not de-essing. Too many other instruments.