#62

5+ midi ports makes me think Cirklon but also the MidiBox sequencer. I’ve always been interested in the MidiBox but it’s a tall investment to not really know what I was getting into.

#63

Not answering your question, but the thing I love about the pyramid that seems missing elsewhere is that it always records unquantized. The swing and quantization are effects which get applied in realtime on playback (meaning you can even dial them in and out as a performance effect). One of my biggest annoyances elsewhere is recording something with quantize inadvertantly enabled (especially to the incorrect beat division) and all the midi notes being irrecoverably placed on the grid.

2 Likes
#64

i would definitely like a system of written notation that’s more friendly with abstraction and experimentation than the current standard

1 Like
#65

Having owned a ton of sequencers including cirklon (twice), mpc (x 4), elektrons (everything prior to the digi series), pyramid, maq16/3, Ableton push and modular Rene, I’ve reached the conclusion that my favourite sequencer is the one(s) I design for myself. Monome and Kyma is now basically my holy grail. I’m still tempted by off the shelf solutions of course but every single one fails me in some important department.
The decoupled nature of Grid/Arc has revolutionised my thinking, quite spectacular…

2 Likes
#66

You could get this kind of behavior from the Mimetic Digitalis with inputs to the CV X and CV Y inputs, FWIW. 16 total “storage locations” each holding 4 voltages

1 Like
#67

It’s been fascinating coming to the Vector from the Komplex. I think it’s clear what the Vector was originally conceived as: eight mono Numerology MIDI step sequencers, each backed with two (relatively stripped down) subsequencers, and some built-in MIDI-to-CV. Unsurprisingly, this is what Vector is best at. While there are some minor inconsistencies and loose ends, and it’s not as open-ended as Numerology or Komplex, it’s still really good.

As additional features are grafted onto Vector around that core sequencer model, things get—unsurprisingly—hazier. Fundamentally, the hardware doesn’t anticipate things that are being asked of it. Copy/paste is an obvious example: two buttons missing from the hardware mean the feature is layered in with button combos and not everywhere it probably ought to be. The “Dashboard” is shoehorned in (and it doesn’t feel well thought-through to me). The drum sequencer and chord modes also look like they might be a bit jammed in there, although I don’t use either of them so I haven’t studied them at all. Anyway, if Five12 keeps piling on features that the hardware didn’t sufficiently anticipate, the inelegances will compound.

Meanwhile, the Launchpad integration is transformative. Launchpad doesn’t add copy/paste buttons, but it significantly alleviates pressure on the existing hardware for navigation and pattern cueing, and it adds some grid-based sequencing features. Once you’re using the Launchpad with Vector, the Launchpad sort of becomes the top level interface, and the Vector itself becomes where you go to tweak things. I understand that Launchpad integration has also been a big deal for the NerdSeq, although I’m not up on the details.

2 Likes
#68

(Your musings on the Vector were a big influence when I wrote my thing about the Vector)

I agree with this fully.

As for the NerdSeq – the Launchpad support is still very much a WIP, and I got the idea that the implementation has more to do with the selection of patterns rather than note input, although I think it can be used for triggers if not CV values. Hoping that the feature gets fleshed out further.

One point with the NerdSeq is that it does have some features that would need further development, and have been like that for a while. But I think that’s almost a given with their one-man structure. There’s a lot of expanders that sound really interesting (HDMI output, USB keyboard/launchpad expander, I2C expander). Launchpad support is another. The DIN MIDI expander also has the very cool Sega controller input port, but that is also something that would need a liiiiitle more work – mapping more functions to more buttons and probably letting users map the key combos.

Maybe one way of putting it is that the NerdSeq is not, uh, very surgical in its concept – it’s more of a candy shop. But it’s a fun space, and the candies are good – incoming is a four-op FM implementation for the sample player part, envelopes and wavefolders, just to name a few. :exploding_head:

1 Like
#69

I admire the NerdSeq, but in countless ways, it is not the right tool for me.

I really wish Five12 would focus on making Vector as good as possible at what it was designed to be rather than turning it into a “candy shop”. For example, I’m skeptical it is ever going to be a particularly good drum sequencer. There will an endless stream of bell and whistle requests that will never really be fulfilled and so—in this culture—Vector will come to be defined as much by what it isn’t than by what it is; a risk of the work-in-progress model. People are throwing all sorts of blue sky (random) feature requests, because why not?

2 Likes
#70

Yeah, I’m not enthused about the chord or drum functionalities, seems like painting through a keyhole.

On the other hand, things that could expand to make it more plentiful to program without much added clutter:

*More mod destinations for the modulation sequencers and inputs

*More available MIDI CC’s

But in the end, I would love it if the focus was on making it as logical and quick to navigate as possible and impossible to crash.

#71

You have a talent for metaphor.

In addition to expanding and improving mod destinations, my main preoccupation would be better modular integration: getting CV in and out.

Other than a general navigational consistency request, my only feature request to-date has been the ability to extract per stage gates (adaptation of the dedicated stage gate outputs you see on Serge and other classic sequencers). I thought it would be clever to base this on the CC1, since that data is already in the model. Thus far, crickets.

Circling that back around to this thread topic, any dream sequencer I would want would have to be highly patchable.

1 Like
#72

I thought I’d try reviving this topic with a video of a sequencer I’ve been working on called STREAMER. Here’s a demo video I made with a laptop-based prototype. It’s indexed so you can skip to the bits you’re interested in.

One person I spoke to described it as a bit like Ansible but using a Push controller instead of a grid and that’s not a bad way of looking at it although there’s more to it than that.

We’re working on porting it to a chip at the moment and that’s going well. The plan is to make a Eurorack module but I’d be interested to konw whether people might be interested in a desktop version?

In terms of relevance to this conversation, here are a few points:

Ergonomic: STREAMER is a step-sequencer that can slip smoothly into parallel universes. The Push 2 offers control without the need to memorise multi-finger button combinations.
Fast: STREAMER has a single layer ‘associative menu’, so there’s no menu diving. Press Setup and activate a control to access all the relevant parameters.
Easy to use: STREAMER’s hint function means you rarely if ever need to consult the manual. The module explains itself.
Permutation: STREAMER employs a unique algorithm to mutate your patterns with the turn of a single knob. The algorithm is highly dependent on the content of your pattern.
Polyrhythms: Each track in a pattern can be a different length. STREAMER is structured with polyrhythms in mind, so they don’t feel like something that was bolted on as an afterthought.
Capture & Rotate: The last few bars of sequencer output can be captured to a new editable pattern with a single button press. Rotate allows you to shift the pattern with respect to ‘the one’. Tracks can be rotated relative to each other.
Expandable: Add output expanders to increase the number of outputs up to a max of 32.
Arrangements: STREAMER employs a unique pattern arrangement method that doesn’t force you into a linear arrangement with a beginning, middle and end. Of course, linear arrangements are also possible!
CV recorder: Record synchronised cv changes to a track using the Push 2’s touch-strip.

Unfortunately, it doesn’t record dreams yet though :slight_smile:

I’d be interested to hear what you think if you have the time to check out the video?

4 Likes
#73

something like the newest iteration of elektron sequencer with step jump and per track condition mixed with thee free form organic tools of modular like sequential switch (boss bow 2 style), and clock divider/multiplier (4ms ping stlye), and lfo controlled movements (rene style). but, since my favorite way to sequence is playing a launchpad pro (in scale mode) and recording live into that sequencer and editing afterward, i would like the ableton (9?) “capture midi” feature that’s constantly listening to the midi input and can recall if you’re playing around without hitting record first

i really like tenori-on style grid based step recording too, so if you could have some kind of maybe norns based step sequencer that has a wider octave range than kria or a scale mode but also utilizes weird phrasing within that sequencing structure like “rebounds”-esque movement in certain sections of the grid or some sort of generative morph feature maybe for experimentation.

#74

Orange knobs and a preset manager…

2 Likes
#75

@Ted your harvestman setup looks incredible. i wanna see it in action

1 Like
#76

@encephalitislethargi It’s growing rapidly. Will post some video soon!

1 Like
#77

To the original topic: I’ve been a sequencer maniac for about 25 years and I feel like I’ve owned them all and tried every different modality at this point, from a vintage Atari running Notator to an MC-4, analog sequencers, eighties and nineties digital sequencers, etc. My main “dream features” can be boiled down to:

  • Not beholden to MIDI metaphor/not starting out as MIDI and translating to something else to go external. This has always felt like handcuffs with its “player piano” mentality, decoupled CC’s/program changes, etc. from notes, 1-16 channels and “drums are always on 10” etc.
  • directly-generated voltages with wide range, not making MIDI and then just turning that into CV after the fact. extremely tight timing.
  • able to be “jammed on” and interfered with on playback dynamically - muting, triggering, launching, pitch shifting, swing/shuffle, suddenly doing things across multiple tracks at will, etc. etc. etc.
  • ability to work on/across multiple tracks at the same time in parallel
  • extremely wide range of options for track lengths in parallel - a one-bar loop should be able to repeat next to a 16-bar pattern, next to a linear track that plays the entire length of the song, next to a fill that only comes in once every 16th bar, etc.
  • ability to do MANY things on a per-note or step basis
  • ability to quickly create entire songs by assembling what you create in a way that’s integrated into workflow rather than in an entirely separate “song mode” that has its own weird limitations or works completely differently or unintuitively
  • as touchable as possible - I want to grab the notes and tweak by hand, dial things in by touch, lay down triggers by tapping keys, “feel” the parameters by twisting them exactly to where I want them to be

Anyway, I have definitely found that there is no one true sequencer that satisfies all of these! I’ve gotten very close to my ideal workflow though with the Cirklon and the Social Entropy Engine used together. The Cirklon really remains the king for all of this and ties an entire room of both MIDI and CV gear together in a tight, integrated, fun, intuitive way. Even working simply it’s very fast for me to get parts going, start collecting those parts as scenes, and then start playing those scenes back in order so that I have an entire song going. I run the Engine into the Cirklon and often write on it, then record the output into the Cirklon for deeper arrangement and development. Where the Engine really beats the Cirklon is fun/speed/jammability of writing parts - the great arpeggiator, ability to parameter lock pitch/gate time/veloctiy (which can be switched to another CC destination like filter) holding a note and twisting knobs to taste, tieing notes together by holding one and tapping the other, note retrigger/ratcheting with a single dedicated button tap and a knob twist for repeat rate, and the ability to jam on and audition all of this without writing to the pattern or only punching in when you choose.

The Cirklon definitely has downsides - the limitation of four modulation rows for a pattern (although with 64 simultaneous channels you can always just make a second channel for the same instrument to unlock four more rows) is frustrating, the song mode is an extremely simplistic pattern playlist compared to the rest of the machine’s depth, and all of the cool features require setup time each time you make a new channel. Editing polyphonic stuff is a nightmare - faster to just re-record or even use a DAW and just recording the output into the Cirklon, which is what I do for stuff I can’t really play by hand. It’s still by far the best sequencer and workflow I’ve ever used. I also run a NDLR and OPZ into the Cirklon inputs to write on and capture the output into the Cirklon.

I recently got a Polyend Tracker and it’s so marvelous that I’m obsessed. While still a very young instrument with a lot of refinement ahead of it, I find it very inspiring to write on between the amazing fill function (I saw someone upthread detail this and it’s basically worth the price of the machine) and the ability to actually view and work across multiple tracks at the same time. I find that I write completely different drum and bass interactions when I’m working on them side by side rather than one at a time.

4 Likes
#78

I’d love an Atari running Cubase as a module, sort of like a giant Modcan Touch Sequencer, but with a MIDI out alongside a whole bunch of CV/gates