I’ve been thinking about this for a long time. I saw that a similar question was asked here about standalone sequencers, but I don’t want to limit the scope like that. I would like to know what is the best sequencer one can use today?
For me being able to use more than a single cv and gate per voice is a must. I want to sequence MPE like data (crazy right, but I think this is a totally normal idea), and to do this I need more than 1 cv out for a voice. 4 CV and 1 gate would be ideal, but would like to be able to adjust that as desired.
I want to be able to sequence entire songs on it. Individual timings on each voice/track.
Being able to play something into the sequencer and have it sequenced back would also be great.
Does anything exist like that today? The Cirklon perhaps (with its year long waiting list)?
First off, I think the idea of finding “the best” sequencer as formulated in this topic is misguided. The best one can hope for is finding the best sequencer for ones use case. Different sequencers present different approaches to sequencing, not simply different quality-levels of the same approach.
That out of the way: What does “sequence MPE data” mean? Record a MPE performance and play it back? Or simply sequence a larger number of parallel data streams? If it is the latter, I think any of the elektron-boxes that are “allowed” to sequence midi can do that. The octatrack for example can send 4 notes and 10 cc per channel/track if I remember correctly. Converting that to CC is a solvable problem. You also mentioned the Cirklon, which can also do similar things.
But I think the greater problem lies in the UI-domain: How do you edit all those sequences, especially if the intended effect is synchronicity in the sense that several parameters change at the same time, but in different ways. With any hardware box I know, that requires a whole lot of context switching.
I think if your goal is working with that many streams of automation data, a computer with a large screen is the best option, especially from a UX point of view. If you need CV output, expert sleepers provides excellent solutions for that.
I like the term best just because while I have a specific goal in mind for it, I don’t want that to be the only goal it is capable of reaching. I don’t want the sequencer to be the best one trick pony.
Sequencing MPE data means the first in what you said. I want to play something on the Roli or Haken and have the sequencer capture that.
What I’ve thought about doing is using the SSP to record the CV coming from the FH-2 as I play something on the Roli or Continuum, and then use the sample player to “sequence” it for later playback. That would work well for just being able to record the play data, but being able to manipulate after is then confining it to the manipulations akin to sample tweaking.
Typing all this down, I think this is far beyond what sequencers are doing today.
I could make make something for the SSP that would allow me to do some of this more intuitivly, but it would be a massive project, and I’m not sure it’d be worth my time.
So i’m curious what you would consider the best sequencer. That is, which one can do the most, in your opinion?
If you feel that the elektron style sequencing is the best because of the control scheme then I’d like to hear that.
Just what is the best one to you, and if possible, through your own experience, which one would be best for what I’m trying to do here? While also understanding I don’t want it to just be able to do this.
Personally, I use multiple sequencers, each for different things. In the modular context, I currently use a Teletype, a ER-101/102, a Five12 Vector and a RandomSource Seq8XL. Each is useful for different applications. I use the ER-101/102 not so much for 4 independent pitch/velocity/gate sequences, but more as a 8-track modulation sequencer with arbitrary sequence length and the possibility for intermodulations. But that’s very different from the idea of recording a performance.
I also have a Linnstrument and do work with that in the modular context from time to time (although I find it cumbersome and prefer to just build instruments in max or supercollider for it). When I do, I record and edit the performance in reaper and send the output out via an ES-3. 8 CV tracks is usually enough for me, but I also have an ESX-8CV for 8 additional outputs if I need them (and more could be added). I think that’s the best approach for what you want to do. I think hardware will only complicate things and as you said, the best thing one could probably do would be to start a computer-in-a-box project that hides the complexity of the computer and just exposes an interface. I don’t know that this would be worth the effort.
editing of MPE is still pretty new, even in DAWs - I think it’ll be a while (if ever) before we see it in hardware.
(ok, the linnstrument, kind of has one, but its not really what we are talking about here)
also, its going to be a small market in eurorack, given polyphony is expensive in modular… again, that’s perhaps slowly changing (e.g. last years doepfer modules)
of course, you can ‘fudge’ it with quite a few things now, but its hardly going to be a nice experience editing MPE data with any sequencer, that doesn’t have the concept of ‘note expression’ … it gets very disjointed quickly.
I think for now your best bet is to create a small midi recorder with playback, thats not particular hard to write - you could then export the midi, to edit in an MPE capable daw (like bitwig, cubase)
This is where programmable eurorack modules come into their own… I wanted the microSSP, but for now Im using a Bela Salt for this, and it works pretty well
btw: you should check out our new independent discourse forum (polyexpression) , which is for all expressive instruments, we only launched last week - but starting to get a good crowd of musicians who are into this area - and this is the kind of thing we are discussing.
EDIT:
btw, Id say the topic title is a bit misleading… MPE is pretty niche.
I say to imply that the best sequencer would need MPE support is a bit of a ‘stretch’… and thats coming from a hardcore MPE fan
Thanks you two! I’ve been lusting after the 101/2 combo for a while. When I finally ordered them, that was the beginning of the 102 parts shortage, and so I cancelled, as I didn’t want to experience the 101 without the 102. Really think I prefer the way the 101/2 works.
As for using a daw to sequence the modular, that is something I’m setup for. Since the FH-2 can act as a host or client, I can send midi data to it from a DAW. So I have that front covered.
I’ve been tooling away on a Android based midi sequencer that I can use with the FH-2 as well. But sequencers are complicated things and doing it right takes a significant time sink. Not sure when that will be done.
Thanks @TheTechnobear for telling me about Poly Expression! I just signed up! I think MPE is a revolution that makes synths every bit the instrument that an acoustic one is. On top of that, master a MPE controller, and you can play multiple modeled instruments. One synth to rule them all type stuff. (I’m exaggerating, but this stuff is what gets me excited the most regarding synth technology.)
I can see your point, but look at sequencers today. They are all constrained by the same constraints for the past 4+ decades. Sure they have cool new control tricks and nice features but we are still mainly just sequencing 1 cv and 1 gate. Some however are trying to accommodate the ways of expressive play, like the 1010 toolbox, and the 101/2 and I think this deserve special highlighting.
I, for example, would not call a sequencer, no matter how capable, the best if it can’t deviate in anyway from the 1 cv/gate paradigm at the very least.
If you want to moderate your subjective opinion on the title, by all means go for it. This isn’t my playground. I just play here.
I think you’re wrong, and have explained my position in this thread. However, I’ve amended the title to make it a little more obvious (in case it wasn’t) that I meant best as it pertains to you.
Sorry for slight OT, but is that viable by now? The last time I tried to build something like that (quite a long time ago to be fair - android 5?!) the scheduling of the operating system made it basically impossible to have any proper notion of timing at all. My understanding was that this hasn’t changed significantly, but I haven’t been paying a lot of attention since I got my iPad.
I haven’t done extensive testing, but have been able to use a few wakelocks and output a small midi sequence. I didn’t check if it was in time or anything, but it sounded good to me. My thinking was, why would there be MIDI APIs if they were worthless to use. I kind of got mired in the GUI of the sequencer after that.
I figure at the very least that work will beneficial if the android version doesn’t work, as I can then roll it over to a PC based VST instead.
I really like the idea of being able to create these massive sequences on my phone while at work or during travel, and then just plug them into the FH-2 to make them happen.
With something like the FH-2 doesn’t any midi sequencer already do this? I can also think of lots of sequencers, Cirklon included, that have multiple CV outputs you can use however you like…
It also depends how you imagine the configuration of a sequencer. For example, the Eloquencer module has 8 cv/gate pairs, but that doesn’t have to be 8 voices. You could use half of them as modulation or expression, or even more. There’s also the Squarp Hermod that has lots of outputs freely assignable to whatever you like…
The Serge TKB has four CV channels on each step/touchpad, allowing for lots of expression.
IMO, one device that would do everything you want wouldn’t be what I would call the “best” since it would be incredibly complex. The closest is probably Cirklon (if you need your cv straight from the sequencer rather than midi>cv), and it’s so complex that it really turned me off. For me the best sequencer is one that is performable and intuitive for real time use. So to echo others’ comments, I don’t think there can be any objective “best” since everyone’s use cases are so different. The right tool for the right job etc…
… MPE is another story, as @TheTechnobear already mentioned.
I’m not afraid of complexity. Complexity is the natural by-product of depth/features.
“Simplicity is the ultimate sophistication” which I think means that making something complex easy without loosing too much is truly difficult. I’d say nearly impossible.
I didn’t know the Elequencer could freely use its outs. Still because they are not all cv capable jacks, you could make 2 voices using 8 cv and 2 jacks. Leaving you with 6 gates useful for nothing.
The Hermod sounds great, but I remember being turned off with the number of outs it had as well. Just 16 outs is just not enough.
The Nerdseq was similar, even though it really appealed to me for various reasons.
I’d love to try a Cirklon, but with a year wait list just to get one, I’m not really taking it serious either. In that time frame, I’ll either finish my own solution or a new sequencer will come around that ticks more of the boxes and is available.
I get that is the way most people feel, but I really don’t understand why. Especially since people are already doing the same things with envelopes and modulation sources. I really don’t understand why the synth community hasn’t embraced MPE with completely open and eager arms.
I use a Rene 1 and a Mother 32, and find the former generally more flexible with performance and editing on the fly and the latter better for modulation and randomness. I had a XAOC Moskwa at one point, but found that so frustrating and limited that I sold it after a few weeks.
Ya seems like you would want a proper computer running max or numerology or something. And learning how to build that would surely give you skills and tech to apply it in many forms.
Personally not a big fan of programming complex step sequencers manually. I like simpler ones with lots of control options interacting together. That said the sequencer on Gotharmans little deformer 3 looks pretty insanely featured with a workflow and unique funtionality incorporating his unique take on randomization that i think i would enjoy taking advantage of. Only 4 CV/gate outs tho and obv no MPE yet. But its got alot alot.
For me the all important question is how quickly can i manipulate and play with a sequence once its programmed in. OpZ seems gr8 and simple fun in that regard.
The thread title asks what we think is the best sequencer. To me that means immediacy and performability, both things mean that mean the best for me sacrifices some “features” for ergonomics… that isn’t true for everyone’s needs.
I think Roli do a reasonable job at explaining this (unsurprising given that they’ve been one of the groups helping push for this, though there have been many others involved): https://roli.com/mpe
Based on your criteria, it sounds like the best sequencer for you would be a laptop and 2 FH-2s + expanders? Hard to see any limits there besides ergonomics, but you can use midi controllers.
Definitely some friction between what you think is best and what others think is best. Maybe ‘what do you consider to be your favorite sequencer?’ is a better question?
I don’t have a favorite, but I love the ideas explored with the combination of Teletype, Ansible, Arc, and Grid.
Sure, could be. Seems a little arbitrary and more a question of semantics, (you know a you say tomatoe and I say tomata type situation) but I get it.
I agree with you (@Jonny) about a computer. I’m already set up for that. Just checking if there isn’t anything else I could get that would work and allow me to be sans a PC.
