Potential for increasing the number of sequences?

I don’t see this talked about a lot here, but I’ve been experimenting more with Sequence mode, and I’ve learned very quickly that 32 mute-states/sequences is NOT enough. Unlike the people begging for an increase in the 7000 event limit (something that could likely only be fixed via hardware upgrade), I don’t see why adding more Sequences would take up much more memory, since a Sequence is simply a list of tracks and their mute-states. I would be so thrilled if the new update increases this limit, because it’s been my main creative barrier when using the Pyramid - I run out of sequences pretty quickly and then I’m stuck either making really long tracks to compensate, or just having to jam live in Track mode and completely ignore the Sequence mode, which I feel is a real waste of it’s potential.

Do others here feel the same? Thanks :slight_smile:

right now, i use a max of like 4 or 5 sequences. but i also do not mind jumping between projects when jamming live

How do you make any sort of full track using just 4 or 5 sequences? Especially for tracks that build slowly, 32 is really low. Especially since they just added the Pattern system to tracks, you have so many potential combinations of mute states and only 32 slots to utilize… Even 64 would be a major improvement, though I don’t see why they couldn’t do even more.

Haha, I dunno, using 32 sequences seems totally nuts to me. I really can’t imagine having that many mute states.

Also, I’m usually either programming something with a pop structure so there are only a few different sections or I just have a couple of different loops that I then mute/unmute different tracks on if I’m doing something techno-y. Also, a fair number of the sounds I’m using come from things that aren’t controlled by Pyramid, like tape players or a guitar or a delay pedal.

I guess seq mode also feels pretty inflexible to me in some ways, but I’m not sure why and I often get kind of lost when I’m trying to remember which sequence had which mute states and which patterns on which tracks where.

v :slight_smile: v

I suppose it depends on the kind of music you’re making, but if you aren’t using external instruments, it’s hard to make your music feel really varied or alive with the limited number of mute states. Think of how many different combinations of tracks+patterns (32 patterns per track and 64 tracks!) there are and you can imagine how the mute states could quickly be used up if you aren’t making something pop-structured, or are trying to create a “build” (where you slowly add tracks one or two at a time).

Yeah, I think it’s definitely dependent on the kind of music and also how you’re approaching making music. It sounds like you are putting a lot of thought into what you are making and spending a lot of time working on details and composition, whereas I am using the pyramid more like a loop/beat machine. Basically, I’ll just jam around on it for a while and maybe make a couple of beats I like and then that’s mostly it until I open them up later to jam them again. They’re usually just 8-bar grooves and might have one or two variations that are other sequences and then the rest of it is playing with mute states and blending it with the external devices.

I got really used to composing in FLStudio, so I think it’s hard for me to try and bring that workflow to the pyramid because the pyramid feels much less immediate than the DAW I was using for years (at least for composition).

What kind of music are you making? Do you have anything I can listen to? I’m curious.

Here’s an example of a beat I sent to some friends for them to play over: https://drive.google.com/file/d/0B20FOV4ssNLCd1JNQ1U1RDdoaHc/view?usp=sharing

That is one “sequence” and I’m playing with the mute states and FX. The rain sounds are an old thunderstorm sounds tape.

Hi Stelpa
yes 64 sequences would be nice…
but in the meantime if you make longer patterns for each sequence (we can have 384 bars with the Pyramid so it’s a sequencer not really thought for short patterns like any other sequencers on the market) would it help?

@Stelpa maybe get in contact with @o_0, he’s using his Pyramid as the nerve-centre of his rig with it being the only sequencer he uses and has made several top-notch full tracks with it. He has a channel on YouTube that’s he showcases his work as well as some great tutorials.

Oh, nice avatar btw.

I am a bit of a perfectionist, so I’m afraid I don’t have anything to share at the moment, @ebbletarts. I delete most of what I make, though that is a habit I am trying to break. However, to answer your question, I like to program complex rhythms and patterns (e.g. drum and bass), and that very quickly takes up tracks+pattern slots and warrant the use of additional mute-states.

I enjoy your beat, by the way, thank you for sharing :slight_smile:

@TheLoudest I can always make longer sequences, as I mentioned, but it becomes frustrating to edit them once they go beyond a certain length, plus it removes some of the creative freedom - I enjoy the idea of having a “fill track” for some drums, for example, that I could swap to using sequences instead of having to manually program in the fill which would cause it to happen at the same time in the bar on each loop. It gives more choice to the user if there are more mute-states to work with, and again, I doubt they take up much memory at all since they’re just a list of tracks/patterns and their muted state.

@megamarkd I understand that it’s possible, but every musician works with their tools differently, and I have had trouble with this particular constraint and would really appreciate Squarp addressing it if they are able to, since I think it is possible on the hardware without much modification.

I do like their work and have learned a lot from their videos before, but since the Pyramid is so open to possibilities and different uses, it’s nice for the user to have choices when it comes to how they wish to utilize sequences - I feel the user should control their tool, rather than the tool constraining the user.

And thanks re: the icon, I’m a huge Waldorf fangirl and thought it’d be appropriate :stuck_out_tongue:


@Stelpa I understand completely your sentiment. I’ve been spoilt by 32trk sequencers for nigh on 20yrs now, although my QY700 can run linear sequences along side loops. I did experience the “32trks is not enough!” (as well as event limits) with that. I currently use five sequencers and still am looking at more due to my M.O. of matching sequencers to particular duties (I have a sequencers=band members analogy).
For the type of music Oliver makes, he really does showcase the power of the Pyramid well, which is why I pointed at him. His work really does inspire I find, although we do work in rather different ways.

Also I still have my Waldorf The Synthesiser Company sticker I got with my XTk with the backing paper still on.

Hello! someone @ me… it’s Olivier btw, not Oliver, but you can call me o0.

I have yet to justify having tracks that use more than 16 sequences, but that’s because I really use sequences as cue points into the track rather than only control what happens to a track by switching sequences.

The core of my technique is that I perform the song live, by switching sequences, changing mute states and patterns sometimes modifying patterns on the fly to keep things interesting. So I never use song mode. This allows me to do the following:

The main motivation to use a new sequence is to change the active patterns of tracks, or to drastically change the mute states of multiple track at once.

I often create a sparse version ( most track for that section muted ) and a full version as a sequence, knowing that I’ll be able to control the build-up and takedown in a more fine-grained manner by entering the track mode ( with record off ) and manually add or remove tracks. This way I have 2 sequences by section, instead of 4 or 10 or whatever.

Also I modify tracks by manually bringing an instrument’s volume up or down while the sequence is playing, for a similar, but more gradual effect.

I also always set the sequence switch mode to be close to or a multiple of the number of bars of my main chord progression. This means I can cue up the next sequence, and got back to track mode and keep modifying mutes, of manipulate the sound at the synth level, or using the Pyramid’s encoders.

This means I get a lot more mileage out of fewer sequences. Also it means that it’s easier to remember what each sequence represents. Most of the time I only use the first page of sequences.

Top row is sparse versions, with different sections left to right ( so 1-8 )
Bottom row is the full version of the same sections, also left to right.

My YT channel: https://www.youtube.com/c/OlivierOzoux

Latest Track

I think the entire track is encoded in only 8 sequences. but each instruments/track has between 2 and 4 patterns each, and I probably end up using around 30 tracks ( one page of drums/percussions, one page of synths )



Hi Olivier. I appreciated reading about how you utilize the Perform function of the Sequence mode, and how you lay out your sequences, it’s quite creative and I think I might try something along those lines. However, I believe it does point to a pretty major flaw in the current Sequence mode that you, probably one of the most well known and experienced Pyramid users, do not even bother with the Song or Loop functions of the Sequence mode. These are pretty major features of the device, and I believe that part of the reason they are so rarely used by anybody is because there are not enough mute-states to make a complete song without manually editing tracks in Track mode while the sequences play, which is not how everyone wishes to use the Pyramid.

I like your jam/real-time focused approach to playing with the Pyramid, but it does not really address the core issue, which is that not every user desires to use the same approach - some of us want to program in a song by hand and use the Song function to sequence mute-states and create a finished, hand-programmed song without real-time input, and with the current limit of only 32 mute-states, it’s simply not enough for anything but very simple song structures with few variations and a few builds, since each one takes up an entire mute-state.

I’m a little disappointed that nobody else seems bothered by the small number of mute-states available - I understand everyone uses the Pyramid different, but I’d like to know I’m not the only one bothered by this limitation, especially since it would be so easy to resolve - if they had 64 mute-states, like with Tracks, it would allow many more possibilities without negatively effecting the users who don’t wish to use them all - mute-states only take up memory if they are being used, after all, and even then they have such a tiny memory footprint that I don’t see why Squarp put such a small limit there in the first place.

There might as well be infinite mute-states (or at least more than 32!), since they take up such a little amount of memory on the device, and even then only when they are in use - it would satisfy the users who care about such things and want to fully utilize the Song and Loop functions, and not at all hurt the users who only need a few mute-states.

I agree completely. Let’s see what 3.0 brings ( @jean @tom @benoit how’s the beta coming along? :sunglasses: ) but I agree that the Song mode is completely under-featured. Actually come to think of it there are 3 areas that never really use in the Pyramid ( well 4, but that last one is my fault more than theirs ).

1. Song Mode. always starting from the top is a deal breaker, even if you could have enough sequences and mute states to cover all of your needs without resorting to dirty tricks like live muting or a PiraMidi loopback to force the pyramid to change its own loop state. In fact I would love to have each sequence be a mini song mode, so I could say: “seq1: play pattern 1 4 times, then pattern 2 once, repeat” or “play nothing for 4 bars, then play one of pattern 3-5 on the 4th bar”

2. Euclidean Patterns. Kind of a cute toy right now, but I’m not going to waste an entire track on a 5 step pentagon distributed over 12 steps that can only play 1 note or chord. Also we need to be able to modulate the euclidean patterns because right now I’d rather punch those 5 steps in step mode instead.

3. Chord Mode. Also a pretty weak area. chord mode ( and ratchet/repeat mode ) should not be just ways to input notes, but should be real things that can be preserved and edited on the timeline. when I input a chord I want to see something that shows up as C#min7 and I also want to be able to at the very least control the primary inversions.

The other thing I never really use is Live and Loop mode, but I think it’s just because I’m soo much faster and accurate in step mode that I rarely bother, or when I do I always edit what I record rather than waste a precious MidiFX slot on quantizing. But that’s a personal choice, rather than a deficiency of PyraOS


yeah totally agree with that
and we need a very simple but effective way to do that
I think there is no need of a fancy UI here
maybe a simple “list”’ screen could do the job

(and same thing for a “projects chain” ;-))

I think the Euclid should be an FX
so we could have the melody we want, the chord progression we want (mix minor and major chords) with the length and the variations that we want
then “apply” an euclid fx on it
(plus, we could make automations with the euclid parameters…)

it would work a bit like a “gate” or a chorder/“superarpeggiator” I don’t know how to call it :slight_smile:

yes and yes! :slight_smile:
a REAL chord mode
with chord names for the 16 steps
(rather than a duplicate of the PianoRoll submode screen)

actually, IT IS a deficiency…
because the live recordings are currently pretty messy with the interpretation of the offset in a 0-99% range…
but I think this would be fixed soon… : )
I can’t wait for this and a good looper as suggested in this forum because I want to use the LIVE mode more often

also you forget :

4. a drum mode (with a drum grid where we can choose the notes, more adapted to the drum parts than a pianoroll screen)

5. basic MIDI sequencing functions
like “insert a bar”, “double the pattern length” (duplicate the content) “divise/double note values of the whole pattern”(midi time stretching)…

and a better file management (classify projects by name would be nice)


I agree : )


Thank you very much for that!
My apologies for miss pronouncing your name.

I just got a Pyramid and I was extremely hyped with the SEQ mode until I realised that there‘s only 32 slots… I‘m 100% with you, this is incredibly limiting to the potential of the device, especially when you want to work fluently with several layers and sections. Any news on this issue? Is this being addressed?

I’d be very surprised if the number of available sequences was increased at this point. The available memory is limited as it is, and increasing the number of sequences would basically require increasing number of patterns by equal amount due to the autopattern feature. And who knows what internal dependencies and limitations there are that we’re just not aware of.

Ah, I didn‘t know about the autopattern feature, that makes sense. Then maybe an option to increase the number of sequences when autopattern is off would be possible? A sequence only stores mute states and selected patterns, right? This shouldn‘t use up much memory.

Until this happens, it should theoretically be possible to work around the limitation by storing only selected pattern data in a sequence and using an external controller (i.e. Launchpad) to additionally send mute states — right? Anybody tried this?

Such options don’t generally make for a sensible user interface, and tend to only cause further problems on the way. One would indeed think that sequences don’t take that much memory, which is why it seems there must be something else blocking it. Whether its the autopattern or something else, only Squrp knows.

Bottom line, I think it’s better to just accept the existing limit as a fact of life rather than pine for something that’s almost certainly not going to happen.

1 Like

I would also add that it encourages us to find ways to modulate within a sequence rather than creating sequence after sequence for variation.

Makes me think of a few Octatrack videos I’ve watched of people playing entire songs with one Pattern.

1 Like