[sdiy] Mikado: A DIY Analog Sequencing System

[Jim Johnson]

>My overall quick impression is: the ratio of 'sequencer' to
>'logic' modules is way to logic heavy. It's great to have all
>those logic modules, but why do they need to eat up so much panel
>space? I'd prefer the majority of panel space devoted to the
>parameters that'll be 'played'.

Could be.. but I've always felt that logic is way under-represented in the
analog synth world, and I really want to experiment with that. So there's a
lot of logic (and dividers). These will be playable to some extent with the
manual switches. To the extent that the take up panel space.. if there are
jacks, they have to take space.

> I think it would be much nicer
>ergonomically if the vertical space between the knobs was the
>same as the horizontal space between them.

I used the Serge spacing for this.. wanted to stay consistent. 

>How do you skip a step in a sequence? Would you feed the trigger
>output from the stage to skip into the clock in? How would you
>skip multiple stages?

Can't do it. I wonder if I will miss this... I know some people freak over
this capability, but I've never seen it to be that big a deal. 

>How about a rotary switches for sequence length?
That's why God gave us patch cords :'). Just patch any Pulse jack back to
the Reset input. Anyone building this could easily use a rotary switch to
do this, if they want. (Jim P. also has an End input in the circuit that
would be more convenient than the Reset input for this.)

Actually I think the current arrangement is better than a rotary switch,
because one of the undedicated Toggle switches could be used to switch
sequence lengths on the fly, between two values, without the need to
quickly spin a rotary switch.

>No random mode? Gotta have a random mode!

I might miss this.. Seq-303 gives it to me, and I use it less than I
thought I would. 

>It'd be really nice to have the quantizer have a second 'scale'
>for rythmic values (i.e., 1, 1.33, 1.5, 1.67, 2, etc.).

Clearly you're thinking of the VC clock for this. I don't think of that as
central to the design at all.. non-ostinato rhythms can be generated using
the logic and dividers, or the pattern generator. The VC is there if it's
necessary. Actually, though.. the quantizer tables will be burned in to
ROM. This would not be impossible to do with a different front panel
topology, I think.

>Once again, you've clearly put a great deal of thought into making
>a nice electronic instrument. I'm too much of a chicken shit to even try!

Thanks.. I really hope it's as useful and cool as I imagine it will be!

Jim

*********** REPLY SEPARATOR  ***********

On 3/30/02 at 5:41 PM P.J. Sonnichsen wrote:

>Jim,
>
>First off, you need to be lauded just for tackling such a
>prickly subject. At first glance, analog sequencers seem
>like they'd be so simple to design. Until you start doing it...
>
>My overall quick impression is: the ratio of 'sequencer' to
>'logic' modules is way to logic heavy. It's great to have all
>those logic modules, but why do they need to eat up so much panel
>space? I'd prefer the majority of panel space devoted to the
>parameters that'll be 'played'. I think it would be much nicer
>ergonomically if the vertical space between the knobs was the
>same as the horizontal space between them.
>
>How do you skip a step in a sequence? Would you feed the trigger
>output from the stage to skip into the clock in? How would you
>skip multiple stages?
>
>How about a rotary switches for sequence length?
>
>I'd prefer 2 pattern generator modules.
>
>I'd urge you to replce all the single trigger busses with
>triple ones. For example, in the sequncer module, you could patch
>it up to be a "skip, on, or reset". In the pattern generator,
>you could easily set up a "bass, snare, rest" pattern. This
>triple buss was one of the best things about the Arp Sequencer.
>
>No random mode? Gotta have a random mode!
>
>How about the ability to have more than one stage on at time? The
>Moog 960 could be coaxed into this state and it makes a darned nice
>"constrained chaos" module.
>
>It'd be really nice to have the quantizer have a second 'scale'
>for rythmic values (i.e., 1, 1.33, 1.5, 1.67, 2, etc.). It'd
>make programming non-osstinato patterns a cinch. I know it
>would be a bit tricky design-wise, but Patchell is the "Rainman"
>of DIY synth design. I'm sure he could whip out a design (without
>trimpots) on his coffee break... 
>
>Once again, you've clearly put a great deal of thought into making
>a nice electronic instrument. I'm too much of a chicken shit to even try!
>
>- P.J.
>(giorgio moroder watch out)


Jim Johnson 
Metaphoric Software
-------------------
Makers of Techno Toys
Software for Electronic Music
http://www.technotoys.com
info at technotoys.com