MOTM

> >> here's my request   two knobs per step.  CV and gate
> >> length (time)..   with 
> >> the gate length set to 0 then there wouldn't be a
> >> gate signal for that step 
> >> and with it set to max (depending on clocking speed)
> >> the gate would tie from 
> >> one note to the next..  basic, useful feature.
> 

That's a nice idea, but I'm not sure it should be built in as a 
dedicated set of controls. That would require a dedicated built in 
clock, which would end up being rather basic. Modular's better! It's 
all a matter of priority, panel space, and cost. There are a gigantic 
set of tradeoffs to be made in sequencer design.

If you only have room for 16 pots, you first have to satisfy the 
minimum requirement of a useful sequencer - 8x2 plus 16x1 operation. 
It must be chainable so that two units could be ganged together to 
achieve 32 note sequences. This is harder to do than it sounds 
without using external switching, flip flops, and dedicated VC clock 
modules with start and end pulse signals. You must have the ability 
to change directions either by a DIR input or separate up and down 
clock inputs. You must have at a minimum a RESET, better is the 
ability to preset to any stage. You have to solve the problem of an 
easy way to ensure that your sequence can easily be started on stage 
one (one of the other seq systems - AS or Doepfer? - resets to the 
last stage and touts this as a feature, but I think it's a hokey 
solution). You need a way to control the length of the sequence. A 
HOLD input is nice. An INHIBIT input is nice (kills the output so you 
can wire-OR or mix multiple sequencer outputs together).

Now you get to the "nice to have" stuff. 

Your idea above could likely be achieved from the bare bones scenario 
above by using 8x2 mode, then taking the 2nd row of pots as a gate 
width CV, and simply controlling the PW of your VC clock. If you 
configure the VC clock such that the PW output is allowed to go to 
zero, but the CLOCK output always is guaranteed to output a pulse, 
this works great.

A set of switches to output a common row gate is a popular sequencer 
feature. If you don't need a variable width gate, you don't have to 
sacrifice a row of pots. Note however that this can be awkward to 
include on devices that are 8x2 AND 16x1 - how many switches do you 
put where, and how can you lay it out so that the UI is clear. 
Typically, I only see this feature on sequencers that are dedicated 
16 columns long.

One of my personal favorites is CV stage selection. Instead of a 
clock pulse advancing from stage to stage, the instantaneous voltage 
level determines which stage is selected. Rising sawtooth input = 
forward play, falling sawtooth = backward play, triangle input = 
forward then back play, random input = random stage selection. If all 
this sounds familiar, you may have heard me use the exact same terms 
to describe the waveform playback mechanism of the Wiard MiniWave.

The potential feature set goes on and on and on. Whatever Paul 
decides to put out, I'm afraid lots of people will be disappointed 
because we will happily design in the kitchen sink, without 
considering that the result will be 10U wide and cost $1000. He's 
targetting a somewhat basic sequencer, not MOAS or even DOMOAS.

I've thought about this a LOT, and you can see the reality somewhat 
reflected in the SuperMoe designs. I started off strong, threw in the 
kitchen sink, and ended up with an unmanageable monster. Then I 
designed a small standalone basic unit, and broke the kitchen sink up 
into a modular system. So not to be a wet blanket, but just remember 
when proposing features that you have 5U wide to work with. 16 pots 
and all the jacks and pots for the basic necessary features are there 
already before we add features and cost on top...

(BTW - I'm flattered that this thread was called "SuperMoe feature", 
but that's inaccurate - SuperMoe was designed by me for me, and I 
didn't have to please anybody else!)

Moe

Dave's Hot Rod MOTM Shop
http://www.users.qwest.net/~daveb2