Sequencer Feature Wish List

MickeyDelp

Hi Everyone! I am designing (yet another) sequencer and am looking for your input. Really, there can never be too many sequencers in the world. I have read other threads wherein the electro-music.community collaboratively designed a circuit and I think it will be a lot of fun.
What features do you consider must-have?
What strange new features have you been looking for, but have not found in an existing sequencer?
I have some ideas of my own of course, and I would love to hear whether you think they are any good.

My thoughts:

Controls:

Modes

Inputs

Outputs

brother303 · Joined: Nov 02, 2010 Posts: 139 Location: ruhr-area/germany

Hi,

here are my suggestions:

- I´d like to see individual gate/trigger-outs per step

- Two or three gate-busses (similar to the Klee) would be cool

- No start-trigger input! I like the solution of Foniks VCPS8 or the Klee. When a clock is present at the clock input,the sequencer starts

- A reset jack to start from step one each time a trigger comes in

- A rotary-switch to dial in the number of steps per sequence (> MFOS 16step)

- Another rotary to select an action,when the step chosen by rotary 1 is reached.(e.g. reset/reverse > MFOS 16step again)

- 1x16/2x8 mode (Klee),but with independent play directions in 2x8 mode,
e.g. steps 1-8 play forward while 9-16 play up+down

- built in quantizer with a number of scales to choose from

- a chord-function (3 or 4 cv-output jacks required,have a look at the Toppobrillo Quantimator!). Each step puts out the root-note and 2 or 3 additional cvs to form a chord. Chord type selectable via rotary-switch

Only some ideas,let`s see what´s possible.

Wink

MickeyDelp · Posted: Tue Apr 05, 2011 5:35 am Post subject:

Good ideas, brother303! Keep 'em coming.

sduck · Posted: Tue Apr 05, 2011 11:11 am Post subject:

16 steps per row, minimum 3 rows if possible.

Basically I want a milton!

MickeyDelp · Posted: Tue Apr 05, 2011 1:01 pm Post subject:

That is a big question: number of steps and number of CV outs per step. The bigger it is, the more expensive it will be - not so much the circuitry, but the pots themselves and the cost of the panel and/or enclosure. My original thought was that if it was chainable, then it could be eight steps, and if you wanted sixteen steps, you could buy/build two of them. Maybe it would be better if there was an engine module that had all the brains, and would control cheaper add-on modules that would give more steps/CVs. What do you think?

brother303 · Joined: Nov 02, 2010 Posts: 139 Location: ruhr-area/germany

Hi,

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

There are some great ideas implemented on the M-185 by Ryk, like its ability to have each step divided into distinct pulses (1-8 pulses per step--or stage, whatever you want to call it), a step/stage mode select (no output on the step, output one pulse, output multiple pulses per the pulses/step switch, hold the gate open the entire step).

These turn an 8 stage sequencer into something far more varied.

Also, portamento on/off per step.

More esoteric/complicated:

Next Step switch, giving each step the ability to direct the sequence to any other step...and if that was voltage controlled....well, Bob's your uncle.

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

Putting this on a separate resply 'cause I don't know what to call it...

I came up with this idea while working on an ARP Axxe Keyboard Memory circuit (which is actually its S&H circuit)...I'm not sure exactly the best way to describe it, but you might call it a CV Programmer function

Imagine a CV input called "Programmer CV In" or something like that.
Next to that you'd have a "Programmer Gate In".

You'd connect a CV and Gate to these from a keyboard or MIDI-CV converter, or a sequencer, or any other CV/Gate combination.

On each step you'd have a Program On/Off switch.

Every time a gate signal is sent to the Programmer Gate In jack, any step that has its Program On/Off switch set to "On" would be fed the CV from the Programmer CV In jack. Each of those steps would have their base voltage changed to the voltage present at the Programmer CV In jack. In this way you'd be able to program multiple steps at one time, live.

Added benefit: since the CV from the programmer input would be replicated at each step, you wouldn't really want or need a quantizer as long as the input voltage is at the pitch you want, no problem. PLUS!! If you feed the programmer input an alternative scale, your sequencer would instantly be playing that alternative scale!

Now, originally I was wondering if an analog S&H would work for this, but you'd need a S&H on each step. And I wasn't thinking of using a CV pot on each step. You might be able to do that, simply summing the 2 voltages, but you'd run the risk of destroying alternative tuning scales.

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

Per step Octave and/or semitone transpose switches...expensive to build, but would make quick sequence alterations fun.

More = More!

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

Actually, expanding on that idea, maybe have brain PCB, a CV/gate PCB, and a trigger PCB. So you could combine normal and trigger sequencing driven from the same brain...or is that kind of a dumb idea?

MickeyDelp · Posted: Wed Apr 06, 2011 3:06 pm Post subject:

That's a lot to digest, emdot_ambient! Thanks for the input. I definitely want the ability to have each step last more than one pulse. Behind-the-panel patching of expansion modules is exactly what I had in mind, much cleaner. The ability to program step values from an external CV is great. But then the pitch knob for that step would not be showing the actual value. Have to think on that one.

Cynosure · Posted: Wed Apr 06, 2011 3:47 pm Post subject:

I have been gathering some ideas for my next project, which will be a simple cmos step sequencer. I am planning on using a rotary switch with at least 12 positions (or 2 in tandem) instead of pots to control the pitch. I don't know how you guys know what you are playing when you use pots. I think it would be difficult to make changes on the fly and keep everything in key - especially if you are trying to add more instruments into the mix. I plan on having each position in the switch tuned to a specific note.

The chord idea is a really good one!

diablojoy · Posted: Wed Apr 06, 2011 4:56 pm Post subject:

brother303 · Joined: Nov 02, 2010 Posts: 139 Location: ruhr-area/germany

Hi,

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

Rykhaard · Joined: Sep 02, 2007 Posts: 1290 Location: Canada

MickeyDelp · Posted: Thu Apr 07, 2011 5:52 am Post subject:

MickeyDelp · Posted: Thu Apr 07, 2011 5:56 am Post subject:

jordroid · Joined: Jan 17, 2010 Posts: 193 Location: ithaca, new york

Regarding the semitone switch idea, if each steps CV is generated by a microcontroller analog input reading a voltage divider you could pretty easily use a pot (cheaper, able to program non half step intervals) or a 12 position switch with appropriate resistors tied to each lug (cooler and easier to use, but essentially always "quantized"), if that makes sense. Tom Wiltshire shows this method in the datasheets for his tap LFO, and i've used it for those and the Boogdish quantizer. It might be nice to have a choice of what to use here, IF it easily works with the uC.

A 12 position half step switch and a 3 position octave switch per step would probably be enough for me, if a common 3 position toggle switch could be used. Two rotary switches per step seems too big and expensive for me.

I like the idea a somewhat "modular" approach to CV generation, one could have, for instance a row with switches for 1v/oct stuff, and a row with pots for variable CV, for filter cutoff or non in tune VCO stuff.

I like the idea of 8 steps per, easily set up for series or parallel operation, it sounds like that is pretty much a given at this point. I'm not sure how "pin hungry" all the features are going to end up being, keeping it down to 8 steps seems like a good number for that reason as well.

I like to run two sequencers together, one off a divided clock, and sum the CV from the two together, and then stick a quantizer set to a major scale at the end, so when the slower sequencer "transposes" the line it stays in key, if that makes sense. This would often be necessary even with half step switches. If that is an easy feature to include it might be cool, if not it's not hard to use a separate quantizer at the end. You of course want to be able to turn quantization off at will if it is included.

I dunno, just trowing things out there, sequencers are one of the more "fun" modules, and i'm always exited about a new one.

sounds like a great project Smile

jordan

MickeyDelp · Posted: Thu Apr 07, 2011 8:38 am Post subject:

jordroid,
Thanks for your comments. I agree that pots would be better than rotary switches from the perspective of cost and flexibility. You would have the option of non-quantitized or quantitized in any number of scales. So, we would have the uC read the pots (super easy) and output to a DAC instead of switching the output between the pots. It basically takes the pot out of the analog signal chain. I'm not sure how the purists will feel about that, but for my money I prefer a nice clean DAC output. I'd be interested to hear opinions on that.

It brings up another question. If the control voltages are being produced by a uC+DAC, then the range is going to be 0-5V. Is that a wide enough range, or should we have the option to multiply that (via an op amp)? For that matter, is 0-nV acceptable, or should we have the option to shift that up/down?

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD

emdot_ambient · Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD