Sequencer for Subtle Music

Hey there folks,

Hope you are all doing well out there? So for the last 3 years I have gone through various sequencers/controllers looking for the ability to do something specific with music that seems pretty tricky right now. I tried Monome/Cirklon/Elektron/ER-101/Hermod etc.

Basically I make music which is simple but lets say as a reference point (although nothing like as compelling as this piece) I am looking to make something which has similar properties to Gymnopédies by Satie. Right now the only way I can do this is in Ableton as it is the only sequencer (or any DAW) that has the graphical interface so I can see midi data like a score.

At this point I have all but given up on hardware as it is either too limited or has an interface design for X0X sequencing.

I wondered therefor about maybe getting an iPad Pro just to change up things from using a MacBook. I wondered if folks out there have any suggestions of good ways of creating midi recordings on the iPad or any other device that can be played back into a eurorack system. Using a pair of FH-2s for Midi to Cv.

Any thoughts and experiences on this from would be greatly appreciated.

All the best!


I think you’ll have to step away from STEP sequencing to reach to your goals. But I thought Hermod has lots of time resolution for anything “Satie-like”?

There’s an app sequencer Modstep, which is in a way like Ableton Live for midi - are you looking for something like that?


Maybe use a music notation software and just create midi from the score? I’ve had a similar problem and found that for all instances where i don’t want step-sequencing (i.e. extremely simple, repetitive or generated music), I much prefer things like Sibelius over DAWs or any form of hardware sequencing. It’s easier for me to see the relationships and interactions of different parts when they are “properly” notated and not in a keyroll (or even worse: on a xox-interface)


Everyone has their reason for wanting to step away from the daw, I suppose. I respect that. I just find Ableton so powerful for this. Especially for generating ideas. So here’s a quick workflow.

I run an arpeggiator, send it through a few different midi-fx for velocity variation, scale and randomness. Add some Note Echo through Max4live. Resample the generated midi on a new track. Add swing/groove. Edit the midinotes - add something that’s missing (that I hear in my head) remove some other notes to create space. In clip view I can reverse/invert etc.

(I’m really fond of Brian Funk’s (Formerly known as AfroDJmac) free instrument “100year old piano”). Satieesque is not impossible.


which properties do you mean exactly? That would help to get a better idea of what you actually need a sequencer to do.

It’s the phrasing that seems so difficult, the timing and the breathes…


Modstep is a very powerful sequencer and very customisable by design, but essentially it’s still just a step sequencer, albeit one on steroids.

The Winter Modular Eloquencer is a very nice eurorack/hardware option with great randomisation and probability options but also still a step sequencer.

Then on iOS there’s apps like Polyphase, which is more geared towards generative structures. And Rozeta, which again is a sequencer suite of sorts, with lots of customisation options. And Thesys for pitch and modulation focused sequencing.

None of these directly gets you into the territory you describe tho, if I understand it correctly.

1 Like

Each type of sequencer has its ups and downs. I find that the phrasing and spaciousness is easily achievable using a modular system, but then it’s not so easy to get the actual chord progressions.

(case in point: a moving clock, a modulated gate, and dynamic phrasing in one go

I honestly haven’t found anything better than grabbing a MIDI file out of Sibelius, thus getting all the right notes, and then integrating that into the modular where it’s easy to achieve malleability of phrasing, speed, and space between notes.


When I talk of phrasing it is about the ability to add in moments of silence (milliseconds sometimes), often across a whole piece. This seems difficult…

What strikes me though is, aren’t you trying to use tools that are not meant for it at all, to acheive things that are just performance oriented and not programming oriented ? It seems like what you call “subtle music” is just human’s imperfection and/or perfection through humans playing instruments. You’ll never get “that” from a sequencer it’s not what it’s built to acheive, you can probably get very interesting results by trying to get there and failing (to me electronic music is about trying to breath some human flesh into the machine and sort of failing to do it, and human performed music is about trying to reach some sort of perfection and sort of failing to do it). But I mean, Satie’s music is very well written, and can be performed in various ways, it can even be performed in a very mechanical and “unsubtle” way and I’m sure you could still make it work that way, but if what you’re after is the human feel of an interpreter playing it, maybe step sequencers are not what you’re after. Spending a lot of time to add milliseconds of silence in step sequenced based machines seems like a weird thing to do since the beauty of these sidestepped milliseconds is often to be found in the ingenuity and spontaneity of the interpreter.

I don’t know maybe I’m completely mistaken but I have a hard time figuring out what exactly it is you’re after ! Sorry, doesn’t mean to come out as disrespectful.

Also there’s a lot of subtle music made with X0X sequencers it’s probably not the best word. (But putting words on musical ideas is very tough I think)


i don’t think there is a single “sequencer module” that will resolve this for you, but i do think a “sequencer” can do (within a reasonable tolerance) what you are looking to do, it just needs help.

yes, software will give you a better interface and many more levels of control over “subtlety.”

but you also mentioned timing as a component of this subtlety and there are many many options for timing in eurorack separate from pitch/cv sequencing. let’s say you have a single trig/gate source running your satie-ish sequence - attenuated modulation of this sources timing will get you to that “breathing.” combining gate sources will get you there too. skipping steps or shuffling the sequence will get you some
life too.

maybe “post yr rack” and what modules have you been considering!

1 Like

If you’re in the eurorack world, getting in millisecond pauses could be as simple as having a way to briefly pause your clock source.

My favorite sequencing tool at the moment is Orca. It’s a whole other rabbit hole that you might not want to spend the time and effort to go down, but there are myriad ways in it to achieve the sort of thing that you’re looking to do. Then again if you spent all the time and money learning and trying out all those hardware sequencers that you mentioned then a free tool like Orca might not be a big deal to dive into.


I’d second this idea. If you take Satie and play it back as a MIDI file or if it is played on a music box it is probably going to lack all of the emotion and gesture that you feel your own sequenced music is too, and all the swing in the world on a sequencer won’t fix it. Also unless something new has come out I’m unaware of I have the general impression that if you can’t get the Cirklon to do something, no hardware sequencer is gonna do it. The only possible workaround I could think which will mean lots more work entering your sequences would be to double your clock speed and the length of your input, since it would give you a higher resolution where you could enter the shortest rests possible here and there to simulate the effect you are after, and I imagine it would have an effect on how things like ties/slides come through.


I largely agree, but you can also easily reach a middle ground by using a touch-responsive controller to step manually through a predefined sequence of notes. I’ve gotten a whole lot of mileage out of using Pressure Points in this way: wiring the overall pressure output through slew generators to VCA and/or VCF, and the gate outputs directly to sequencer clock input(s). It’s a great way to imbue “expression” into a sequence without exactly playing the notes by hand.


Something you can record in live and then playback would do the trick I guess; maybe the Hermod sequencer?

Or maybe even an Electribe or Elektron box that spits out Midi (not familiar with the Elektron way of working though so I may be a mile off).

I can totally understand wanting to avoid the computer/DAW based approach (there is a lot to be said for the immediacy of hardware).

1 Like

For delicate melodic writing, I really like step sequencers that let you specify a duration per-step, rather than a constant pulse with rest/note triggers.

The intellijel Metropolis is the most prominent eurorack example of this style, but it only does 8-step patterns which is enough for riffs, but not great for melodic work. There’s several Buchla sequencers that can do this - the 248 and the 250 I think, but I’m not a Buchla user.

A newer module you could also look at is the Frap Tools Usta - which seems like a Buchla 250 clone in euro to me, but I haven’t gone super deep into its capabilities or workflow.

That said, it sounds like you might be past working on hardware - my oblique take would be that something like Orca, or even a tracker-style UI for detail-oriented work, might be worth considering if standard notation and piano-roll UI isn’t doing it for you.

What I’m using for this now is my OP-Z - sequences have a max of 16 steps, but you can set duration per-step, and have different step lengths and overall resolution on each track, with lots of ways to add variation and randomness.


Octatrack is pretty wonderful for sequencing. You can do micro-timing, random elements and polyrhythms.


maybe the way to go is ditch the midi grid and work in freely assignable time Intervalls. I‘ve had great fun composing small melodies with looping envelopes and 6 OP FM on the Yamaha SY77. Since there is no way to sync those EGs you‘re forced to do everything by ear, which leads to quite different results than classic grid style sequencing. In eurorack there is the Control Forge if you’re looking for precision and a lot of loopable EGs for more gestural composing. A dc-coupled sampler could also be used as an unconventional Sequencer of sorts.


I’ve had good results using Numerology…it’s deep.


FWIW, long melodic lines is what I have in mind for this sequencer design: Sequencer design for grid.

Some hardware sequencers can do anything theoretically, but their UIs are too modal, requiring a high cognitive load. A piano roll on a computer is lower cognitive load, but I don’t think it suits itself to live experimentation because it produces a lot of invalid in-between states when editing.

1 Like