This is an excellent comment.
Think about an acoustic guitar. You have six strings (generally) and they
have been designed to sound as similar as possible. However, they are
different thicknesses, they are in different physical positions over the
resonating body and sounding hole, they make a distinctly different tone if
they are "open" as opposed to "stopped"... so even though you conceive of a
guitar as an instrument with simple 6-voice polyphony while you're playing
one, the physics of the situation create a very complex, rich sound.
In fact, when doing orchestral arranging, one thing you want to do is make
sure that the clarity of individual voice movement in a polyphonic mix can
be heard. Instruments like the guitar do this automatically, and the ear can
pick up that distinction. It's subliminal, but important.
Guitar is just one example. Think about the woodwinds section in an
orchestra. You have a row of people all playing flutes. But they are seated
in different locations in the stereo field, and one guy's flute is 15 years
old while the woman next to him has a brand new one that she polished that
morning, and then there's that guy on the end with a gold flute, the one no
one else talks to very much. And each of these people has a slightly
different aesthetic, lung size, etc., which affects their playing even
though they've spent their careers trying to perfectly control their sound.
All of this creates a result much richer than holding down notes on a
polyphonic sampler, or even just detuning VCOs.
So I second JH's comments: start off with the "idea" of making the sounds
the same, but don't be too anal about it, as you may find that having one
just a pinch brighter and another with just a little bit more decay on the
envelope does miracles. This is the kind of thing you will not get on a
pre-packaged polyphonic digital box, certainly not on a sampler. You used to
get a little of it with old poly analogs because the tolerances in the
circuitry made each voice just a wee bit individual.
Also remember that you can cater your patch design to the particular phrase
being played. If you're playing block chords with four voices for this pass,
you really don't need four VCAs, etc. It's like planning CGI animation -
only render what will be seen.
-----Original Message-----
From:
jhaible@... [mailto:
jhaible@...]
Sent:Tuesday, 03 July, 2001 7:03 AM
To:
fuzztone@...Cc:
motm@yahoogroups.comSubject:Re: [motm] Polyphonic MOTM
> has to have at least one vco, vcf, env and vca ∗per
> voice∗ .
[...]
> just be sure to dial your voices up the same way - like
> setting up an oberheim 2/4 voice.
Polyphonic playing on a big modular can be a very special experience. At
first one might think that many identical modules are needed, like four
M-440 filters for a 4-voice modular, or 8 ENV generators ...
But in practice, you can get good (and often more interesting) results with
a mixed bag of modules. Using 4-pole filter for one voice and 2-pole filter
for the next, or using a VCLAG to fill in for missing ADSR modules, is not
just a surrogate for N equal voices, but it can give polyphonic playing a
new quality.
For a start, it's good to ∗attempt∗ to set up a similar sound for each
voice. This will not be possible with different filters, but you can set
them as close as possible (like, using a slightly lower cutoff point on the
2-pole VCF to make up for the less steep slope). Of course playing with N
totally different voices has is benefits as well, but playing with
∗slightly∗ different voices has a charm of its own. I don't know why, but
starting with different filters and trying to get close is more satisfactory
than using identical filters and setting slightly different parameters.
(;->)
JH.