This is everything I have in my 'TKB Wizardry.txt'. All of it is from
this list's archives so some of you may recognize your own writing!
Let's call this a 'best of SMOG - TKB edition'.
"Random select selects a random sequencer stage. BUT in order for it
to work, not only do you have to patch a clock pulse into it (from DSG
or DTG or whatever source) but you have to patch it to RESET at the
same time. So run a short patch cord between RESET and RANDOM SELECT,
and plug in your clock signal into the back-end of the banana
(effectively multing everything together) connecting RESET and RANDOM.
One thing worthy of note on the TKB -- its 'reset' input is one of the
few, if not only, inputs on the Serge which can safely gang inputs.
What this allows is for two or more subsequences to be created on the
TKB. Patch, for instance, stage 5 and stage 15 into the reset. Start
the sequencer. It will run from 1 to 4 and reset. Now touch pad 6 and
it will begin cycling from 6 to 14. Touch any pad below 4 and the
cycle begins on that stage and resets at 5. Touch any pad above 6 but
below 14 and it resets to that stage as the start point, and cycles at
15. This is, for me, one of the coolest aspects of the TKB.
Here's a few things to try with the TKB:
1. Clock your TKB in the normal way. Then also route your clock
signal, wherever it's coming from, to a DSG or DTG or NCOM (divide by
N)... Send this divided pulse to HOLD and /or UP/DOWN so the sequence
pauses and reverses in different ways. Use an LFO or random source to
voltage control the pulse division amount going to HOLD etc so
sometimes it's dividing by 2, sometimes by 3, etc.... in other words,
varying the division.
2. Use one layer (say Layer A) to control a VCA modulating the
loudness of your VCO. Play with the loudness sequence in addition to
the pitch sequence.. You can come up with some jazzy combinations here.
3. Use a variable divided pulse as in #1 above to the RESET input,
giving you a voltage-variable number of sequence steps.
4. Clock ONLY the VERTICAL input, set up to 16 different 4-step
patterns (cycling through layers A,B,C,D for a given stage), and
select the different 4-note patterns with the touch pads. Use the ABCD
output to control your oscillator.
5. Use one TKB layer to control the speed of your clock. That way the
step rate won't be steady even & equal, but varied from one step to
the next. This takes some careful tweaking for good results.
6. Use a random pulse to clock your TKB. So you have a predictable
stream of tones but with unpredictable beat.
In "normal" sequencer operation, each of the row (A, B, C, D) jacks
will output the voltage of its respective pot at the current stage.
For each sequence stage (or "step") you get four different voltages,
each of which can be routed to the control destination(s) of your
choice. You might, for example route row A to the pitch of several
oscillators, row B to linear fm amount of an NTO modulated by a PCO,
row C to the wave multiplier cv input, and row D to filter cutoff.
The ABCD output works in conjunction with the Vertical Clock input,
allowing you to run sequences as long as 64 steps, but with only a
single voltage output (in contrast to the four outputs in "normal"
mode). When the Vertical Clock input receives a pulse, the ABCD output
is switched from its current row to the next row. For example, patch
one of the individual stage pulse outputs (1, 12, 16, etc, you pick!)
to the Vertical Clock input. Advance the TKB horizontally using a
regular clock input, and once the sequence reaches the stage you
patched to the Vertical Clock, the ABCD output will switch from the
row A voltage to the row B voltage. The next cycle around, it will
switch to the row C voltage, then row D, and finally back around to
row A again.
Here's one approach...
1. Use one layer of the TKB to control volume. Set all the knobs on
that layer to fully-on.
2. Send that voltage to one of the VCA's in your UAP.
3. Send your audio signal into that VCA, then send the output into
another VCA that your DADSR or DTG is enveloping the audio with.
Now to 'turn off' or silence any sequencer step in your melody just
turn off the layer you used for step 1 above. The advantage here is,
in addtion to creating a 'rest' beat(s) wherever you want, you can
also create accenting... daDA DADAda ... by varying the volume control
stage pot on the TKB.
More efficiently, this could also be done with one VCA if you can mix
DC signals together ... You mention you have an inverter/processor..
can that combine control voltages? I have a Dual Processor & a CV
mixer that do that. Anyway take the TKB volume control layer and your
ADSR and mix them together, biasing the whole thing by -5v, and use
this to control a VCA for your audio. You should only hear notes where
the TKB volume control stage is fully on. This can do the 'accenting'
as above.
It's also fun to use another TKB layer (by layer I mean the A,B,C,D
rows of 16 steps going across) to control pan position. Makes your
sequence dance nicely!
All the above needs only simple TKB patching... clock the clock input
like you always do, and set the sequence length using a stage output &
reset input, or just run 16 step sequences.
Set up a nice 8 step sequence, triggering it off the DTG as you
describe. you will have a steady rhythm with a note at every step.
feed output D from the tkb to the CV input of the DTG and apply a
little CV offset pos or neg. depending on the voltages set on line D,
the DTG will either speed up or down at that step and then move onto
the next one, speed up, etc. so, to set up a simple 4/4 with
note-rest-note-note, set up a 3 (yes) step sequence with a touch of
voltage from step 1 row D (causing DTG to hold) and none on D for 2
and 3. Adjust accordingly.
You can also try this type of thing with the "hold" input on the tkb
and a clock divider.
My favorite technique is to DC mix envelope and a TKB layer (like D),
then negative-bias the result to control VCA so that when layer D is
0, you get no sound, when it's fully CW, you get full volume. This
allows 'accenting' beats as well as rests and full-on notes."