Eurorack: discussion, advice, learning, questions

I’m gonna go ahead and say that Shared System is a perfect modular synth experience. Every time I lust for some gizmo, I just think about how I can do it with my Make Noise stuff. If you don’t mind mismatched panels, you can build one for a lot cheaper than what the BnG costs new.

However! People here have great ideas, and great advice, and it is immensely fun to devise your own synth. Part of the fun is ignoring conventional advice and building the synth that YOU want to build. I’m a big advocate of “too many sound sources” and “not enough VCAs”.

Take for instance some of the setups used by one of the artists that got me into modular, ann annie.

If you posted a picture of that rack in MW or Reddit or whatever, people would immediately point out that they’re are no less than four sound sources and not very many VCAs, not much modulation, etc. Basically this is the rack that all euro-veterans advise against.

Yet he’s out there making incredible music, playing big shows, and just kicking ass all around.

Build the instrument you want to build, and when you get tired of buying and selling modules to figure out what your perfect instrument is, sell everything and get a Shared System! That’s what I did, and my only regret is using a credit card to buy modules. :woman_facepalming:


Awesome advice (really, all around), thanks. I’ll ignore it then ! :laughing:

Ouch for the credit cards, I remember all too well (albeit at a smaller scale) how much of a drag taking a loan for some of my current gear / software had on my ressources. Plus I practically didn’t use one of the things I bought (Maschine Jam). Infuriating…

Generally there seems to be a bias against redundancy when people give advice on modular.
Some of it makes a lot of sense, if you don’t use 90% of your synth then you don’t need it.
But I think people sometimes expect everyone to use their synth for a patch that uses everything at once regularly. Personally I have probably the ability to do 4 simultaneous voices if I really want to but most of my patches are a single voice because it suits me. Each module still provides timbral variety I couldn’t get otherwise for when I need it.

Consider this type of setup that was relatively popular 10 years ago:
All modules necessary for a standard subtractive synth voice, only with 5 timbrally different filters.
Now, this isn’t for myself but this type of setup has a clear purpose and it’s a valid reason to use modular as a format. The user may only use one filter at a time, making most of the filters redundant in that patch (and redundant when you look at how often they’re used compared to other modules) but they’re clearly not pointless. You can make probably any kind of monosynth in the last 40 years of popular music with that.

A bit of an aside but I hope I’m not completely rambling…


I got Plaits yesterday and I really am loving this oscillator. So many great sounds on it, but I am still learning it.

I really love the envelope you can access with the second button but what I am not sure of is the fit with my MMF filter. The LPF sounds anemic at best on this. I need to figure out if there is a way to modulate the filter in Plaits, and if so I may pull this filter out and free up some space.I have a clep Diaz that is really cool and I’m going to figure out a bunch of cool ways to modulate different parameters with it today.

Another good example of this is a recent Elinch video where he goes through his process for a track:

1 Like

(Not sure if this is the right place for this to go, please feel free to move it if it isn’t!)

I’ve toyed around with the high-level concept of building a “performance partner” (what we like to call “digital twin” in tech these days) in the form of a eurorack “control rack” for a couple of months, and with the covid-19 stuff going on I figure it’s as good of a time as any. I envision this to be a cross of algorithmic sequencing/composition with inspiration from snippets of theory like call-and-response, point-counterpoint, organum, the like.

At the most basic level, I’d like to be able to play or sequence a very simple melody (say, a 1 bar melodic sequence) into the rack, and have it output CV/gates for accompanying voices. The intent is for the output to be pseudo-random and musical. I expect this is going to be my long-running project that I’ll be working on as a focus area.

Has anyone attempted something like this before?

Some really basic pieces I thought I could start with (that I’ll start assembling in a rack tonight):

  • Running the sequence into o_C’s CopierMachine with a one-bar delay
  • S&H a channel for drones
  • Multing that drone into another voice, triggered with a burst generator with an AND logic with the drone for rhythmic bass
  • Resampling the trig and CV with Marbles to create a permutation of the original melody
  • Using Maths as a trigger delay and subdivider for repetitions of the original melody at different rhythms

Eventually I plan to pick up a Teletype and exploring whether I can implement some of Tom Johnson’s Self-Similar Melodies on it. I’m also possibly looking into something like the Harmonaig or ADDAC 207, maybe an ACL Sinfonion if I feel crazy enough, if I want to include polyphony.

Some stuff I’ve already gone through in the last 24 hours or so for ideas and inspiration:


Love the concept. Teletype would certainly enable you to explore these ideas further. I would start there and then see what you feel you are still missing. Also, the TXi expansion module could be useful if you want to affect changes in the algorithmic script via human input.

This is really similar to stuff I’m trying to do! I’m definitely not there yet, so I’m thankful for the links to those ideas & inspiration sources.

One question I have is: “Resampling the trig and CV with Marbles to create a permutation of the original melody” – my understanding of Marbles is that you can process external CV, but you can’t use it as the source for its randomness algorithm. It’s basically just a toggle between “use internal random voltage generator” and “use externally supplied voltages”. Am I missing a feature here, where Marbles can take in your supplied pitches, then use them as a seed for its random generator and then slowly permute them over time (via Deja Vu)? Or am I misreading what you’re saying, and you’ll be using its bias/spread/steps controls to just transpose the melody or something?

Yes I believe it can. In fact I will be messing with that today. But the Manual states that you can play notes into Marbles (it recommends playing 50 notes or so, which may just be a 5 note lick repeated 10x, or whatever you want) and it will learn what notes you are favoring and then make it’s own variations of it.

I’m very interested in figuring this out, because thats pretty amazing if it works as I expect.


Oh you’re going to program it in as a scale! I forgot you could do that, and was focused on the “Sampling external CVs with the X generator” section of the manual. But yeah programming it as a scale will make sure it only uses notes from your melody, and it will select from those notes proportionally to how often they appear in your input melody (as you sweep the STEPS input/knob). I don’t think it will necessarily follow the shape of your melody (like, if you play it Hot Cross Buns, it won’t play variations on Hot Cross Buns, it’ll just use the notes of Hot Cross Buns to make a brand new melody), and I was hopeful for a way to have it also follow melody shapes, but it will certainly follow the allowed notes. Awesome!

Thank you all for your words of support!

@21echoes “Using externally supplied voltages” is exactly what I meant by “resampling the CV” - not changing the seed for its random generator. It won’t be the same melody, but if my melody has only 2 to 4 notes, then a locked sequence of that should sound relatively similar. More testing required :slight_smile: Modules are racked, will have to find the time to play around with it now.

Has anyone tried using analog shift registers (for example, CopierMaschine on o_C) for shifting trigs? Trying to think of ways to delay my trigs the same way I’m delaying my CV…

I have patch that I’m currently working on that uses CopierMaschine on o_C in combination with a Pamela’s New Workout to send clocked random voltages at different ratios to a Quadrax. This envelopes are then being sent out to a 4MS spectral multiband resonator to trigger the notes in the current selected notes within a scale. You can get a nice shifting yet repeating pattern this way.

Yes exactly. I just did it now. It’s surreal because it really keeps the focus on the notes I played while generating totally new ideas. I’m sitting here just listening to it riff in the exact scale I wanted while focusing on the chord tones I wanted emphasized most.

This module is really amazing.

1 Like

Are there modules that can take a pitch CV and gate and delay them by a certain number of clock trigs? Shift registers are where my mind is at, but modules I’ve seen do one or the other, but not both in one module (ie 2 inputs). I could hack something together with stock o_C running CopierMaschine for pitch CV, and another o_C with Hemispheres running ShiftGate, but coordinating the modulation for both of these such that I can change the delay length is something of a small nightmare. I was hoping for something with the with a common delay value shared between two inputs and two outputs, without having to reach for a Teletype…

Sounds like an analog shift register like, but I don’t think anything less than a digital solution like Teletype will handle delay by an arbitrary number of clock steps.

The delays in Disting are DC coupled and work for CV, so if there’s a clocked stereo delay algorithm it could do that. Except I don’t know for sure that it perfectly preserves the CV without scaling it in some way.

You can shift gates with a shift register since gates are just high and low CV values. What will happen though is that you lose gate length since it’s going through a sample and hold (so the gate will be held until the next clock pulse). So if you need it to absolutely output short triggers, that will be a problem. You may also experience some retriggering issues if you have two gates in sequence also.

People have mentioned a few delay modules, which I think would work. Issues with delays might have to do with getting a good delay time that actually syncs with your original clock.
I’m also not entirely sure that the delay needs to be DC coupled either. If you send a short click of some sort, many modules (although I’m sure not all) will accept that as a trigger. I often use this when I want to convert gates to triggers by simply patching the gate to a super short decay envelope and using that as my trigger.

1 Like

@GoneCaving @oot yeah, like I said in my post:

Shift registers are where my mind is at, but modules I’ve seen do one or the other, but not both in one module (ie 2 inputs).

Checking on the ASR for o_C (hemisphere) there’s not enough memory to handle two ring buffers, so still stuck with only one CV input. A Synthesis Technology E102 is pretty close to what I want out of it, except that I’d need two, would have 3 pairs of unused shift outputs, and at that point costs more than a single Teletype. :grimacing:

Hi everyone,
I have a new iPad coming next week and started wondering what would be the best way to connect it to the Modular as a sequencer.
I haven’t really followed the development in regards to midi2cv modules, and maybe there are other ways I don’t know of?

So, what is the best way to connect the iPad to a Eurorack Modular in 2020?

…especially with the iPad having USB C.

Thanks everyone and stay safe!

Apple Camera Connection Kit (best if you get the one that also allows you tot charge at the same time) to MIDI interface module (ES-8 etc) on the euro. Have fun!

1 Like