Have a look at EMW's stuff, also...they have a matrix switch network for controlling things such as timing pulse destinations, etc.
Hmmm...well, let's see. As someone who went through the academic music system (until I got to Illinois, which taught me a lot of "interesting lessons" about academic composition...to the point that I quit academia altogether), the idea here to create a "teaching synth" is good...but you have to impose certain limits on the build so that you don't have a lot of confusing gewgaws in there that no one will likely encounter in the real world. As for me, I learned on an ARP 2600, which I STILL say is perhaps the sine qua non of educational synths. Given that I've found that there's little to no significant differences between any of the Rev. 2, 3 or 4 2600s and the Behringer 2600, having one of those around is something I would strongly suggest. You can also mount it in a heavy rack, which discourages it from "going walkies". The Palette 104, on the other hand, is EMINENTLY stealable...it has no Kensington lock port, it's small enough to fit in a backpack or under a coat, and its Meanwell brick is even more stealable; if someone needed a brick for their own rig, there'll be a temptation to snarf that supply. Also, the tile row is still a bit of an "unusual" feature that's still not 100% in common use. So, I banged out something here...
The idea was to create a teaching synth, a system that's specifically designed to show what the basic functions of subtractive synthesis are and what they do...and also, to show what happens when you start interconnecting things and arrive at that "more than the sum of the parts" result.
So, this is set up so that all of the main aspects are amply demonstrated. The top row is all audio, middle is all modulation, and the bottom is all control. Here's what's here...
Top row: This starts with a Cavisynth module, their Bufflide...this contains a 1-3 buffered mult AND a slew generator for portamento. Then there's TWO Plaits, because you want to show not only how these work, but what happens when you put two VCOs together and slightly detune one, or to show how you can tune one VCO to a fundamental and then use the second to add the harmonic content (very Plaits trick, that). After that, a dual ring modulator from Tenderfoot (basically, the same passive ringmod module I have hiding in two of my routing patchbays), then a Veils...which comes before the filters because you can show how to "strum" VCOs via modulation signals. Then Ripples, and a Nonlinearcircuits Dual LPG...because it's very useful to use a VCF for "broad" filtering and then use an LPG to contour that into individual notes, or to use the LPGs with some noise (see below) for percussives. Then the audio processing...a Tiptop Echoz provides delay methods, and then the Beads does its granular thing. Last thing there is an unbuffered mult; normally, I'd leave these out on a small build, but it's worth having it here to show how you use these.
Middle row: Quantum Rainbow 2 is a brilliant way to show what "noise color" is about, plus what you can do with the different noise weightings. The Shifty functions both as a sample-and-hold (in single mode) or an analog shift register so that you can "carry over" CV values based on timing and get quasi-polyphony. Stages is next (very useful pairing with the Shifty, also) then the Batumi + the Poti expander. After that is a Happy Nerding 3xVCA, which gives you three linear VCAs for altering modulation signals, then a Shades for polarization, mixing, adding offsets, etc. And Maths. Yeah, you said "no Make Noise", but...well, Maths is what it IS. Not only is it popular in Eurorack, but I think Tony's redesign of the Serge DUSG is a staple in synthesis in general these days, and therefore, it should be in this instructional build. And after it, you'll find an Intellijel Dual ADSR because, while you can get other envelope behavior out of the Maths and Stages, it'll be important to have proper 4-stage EGs in here for both musical and instructional purposes.
Bottom row: An Intellijel uMIDI gives basic MIDI control over a single voice PLUS clocking and other MIDI CC value outputs. Then, yep, Pam's...not only as a clock, but also as a pattern generator and a few other things. Then the next several modules are for screwing around with what Pam's is doing: an Intellijel Diode-OR allows combining of pulse signals, a Ladik Dual Delay gives you two channels of clock pulse delay, then two Doepfer clock mods give you various clock division schemes or multiplication/ratcheting, depending on which module you're talking about. Then Frequency Central's Reset Simulation is a brilliant Boolean logic implementation that not only shows what Boolean gates can do with timing, but gives graphic examples of what the Boolean states actually look like. And of course, if there's a Boolean module, there's a sequencer...and I went with Xaoc's Moskwa 2 and their Ostankino 2 sequencer control module for a robust but simple sequencing environment. After that is your output modules: another Veils for VCA control over audio levels, then this feeds a Doepfer A-138s to allow stereo panning and manual level control. And at the very end, the Happy Nerding Isolator not only gives you output isolation and a ganged stereo level, but it also contains transformers for the isolation function...which you can "punch" a little to warm up a signal, a very worthwhile little trick that should be taught.
The basic idea here was to use as many "big name" manufacturers as possible...not because I'm down on boutique builders, but because I wanted to make this as "bog-standard" as possible, so that anything learned on this will translate decently to systems built around more esoteric modules. So the only "might be difficult to get" module here is actually that Cavisynth Bufflide...which, if that proves to be the case, can be fixed by using a Ladik C-015 in place of the Bufflide and then you'll move the passive mult at the right end to the left end and switch it out for a 2 hp buffered mult. Otherwise, much of this is either fairly close to "off the shelf" or, such as in the case of Ladik, it's from long-running, experienced smaller makers. Annoyingly, this build wasn't "cheap" as such (it sort of is for Eurorack in general, though), but it should be simple enough to implement if the funding is there, and it should be a worthwhile educational synth for many years to come.
Now...about that case. Yeah, three rows...at 84 hp. Basically, this is designed to go in a Doepfer LC9 case (which is only a whopping $36 more than the Palette 104). And the rationale there is that the LC9 cab is a big ol' chunk of MDF wood...into which you can drill a hole, attach a large eyebolt, and then padlock a chain onto that, assuring that this synth ain't goin' NO WHERE without that key. Knowing how equipment in academic situations can mysteriously "grow legs,", I felt that this was extremely important to implement...and the "gravy" here is that the three-tier cab effectively shows the primary division between audio, modulation, and control sections of a basic 21st century modular rig. Should work if the $$$ can be made to work.
For that size, probably the ES-8 would make the most sense...IF you opted to go with a module as a solution. However, CV Tools and so forth can ALSO use any DC-coupled audio interface to do the same thing. Not only is this way cheaper, it's also just as expandable since you can chain up more I/O via the ADAT Lightpipe interface. I currently use a MOTU 828 mkii for this, works just great. But here's the real point: you can get these older "obsolete" interfaces for dirt-cheap. Right now, the original 828s go for around $75 on Reverb, with the mkii coming in only a bit higher, and either are just fine for this purpose. So if you felt you needed 16 channels of CV/gate/trig/clock send, you can get there this way for $300 LESS than the ES-8...and also wind up with more outputs than the ES-9!
- Productivity: a Modular is by far the least productive thing you want to use if you want to make music. You tend to get entangled while finding a sound or simply by patching. You may end up with something interesting, but it possibly is not what you wanted or needed for the composition you had in mind (if you had one).
-- suomynona
Actually, this only seems to be a problem with people starting off in modular, while they still have more of a sense of exploration...which, frankly, is good. Modular synths allow for lots of that. But I wouldn't call any of this exploratory effort unproductive...rather, it's just the electronic equivalent of figuring out what all of the keys on a sax can allow you to play. But after several years of living with modular synth gear, it's actually QUITE easy to patch up what you need once you've got the instrument(s) sorted out...and exploration is the key to getting there.
About the only "Maths alternative" aside of having a full-on analog computer would be to have a RandomSource Dual Universal Slope Gen (an exact Eurorack replica of the Serge original) plus some various routing and modification mojo...a MISO or Frap 321, a few VCAs, and maybe a mod or two to the RS module itself. It really IS one of the very few "mandatory" Eurorack modules out there!
I think it would be interesting to be able to take sounds from a track in Ableton (for example, i have an iPad with some cool/different synths (Animoog and Sound Prism to name just a couple)...and run them through various modules in my rack and then back into another track in Ableton...would that be accomplished with some sort of 'sampler' module?
-- jb61264
Nope...you'd first set up an external send and return path inside Ableton so that you can get the signal out and back. Then you'd use your DAW interface to get the signal into/out of the analog domain for the synth. Route the "send" through a suitable preamp to raise the signal to synth levels, and attenuate back down at the end so you're at line level to hit the return to the Focusrite. This is, of course, if you use the Focusrite for this; if you had an ES-9, you could send and return the audio using that module alone, plus a suitable digital link to the Focusrite (like an ADAT Lightpipe). This would also result in having dedicated interface I/O for the modular alone, which isn't a bad idea because, well, CV Tools. And yes, you CAN simultaneously move audio and CV signals over the same Lightpipe and Expert Sleepers module.
+1 on expanding the case here. Doing so would make for a more effective build, plus with something such as a Mantis, you can probably wind up saving some money as you're not buying two cabs, two power supplies, two busboards, etc. Little cabs like these DO have uses, just not as the "canvas" for a full system build.
If you're planning to use CV Tools, I'd strongly suggest looking at one of Expert Sleepers' interfaces...the ES-8 or ES-9 would work perfectly, depending on how much I/O you need there. Plus, by going directly to that, you can forego MIDI altogether to the modular, and use the DAW's sync to keep the MIDI devices locked-up with the modular. Definitely more efficient.
Nope, I don't. These were all build reworks for other users that I did on here. When I work on those, I'm also working with my 40+ years of experience in electronic music plus a sizable background of material I've studied (and still do) to stay more or less up to speed in Eurorack.
Now, if you're in Norway, it would probably make more sense to go through Thomann, as they've got a huge selection of suitable cabs for rack frame mounting. Plus, they have cheap ATA cases, which will let you travel with everything prepatched...just pop off the lids, jack in, and cut loose!
One other thing, also...since these Ladiks are so small and cheap, you might even consider getting a small skiff for several of them plus a power supply (or a powered skiff) to use as a dedicated CV source/controller that you can patch at will across the entire system. That would also give you a tad more space in the build, besides being a straight-up ergonomics WIN.
The problem with making your output mixer too small is that it'll be a major PITA to use. You CAN find mixers in that 10 hp and down size that will give you stereo, but as you make the module smaller, you also eliminate the room on its boards for VCAs, autopanners, etc etc while, at the same time, giving you a very fiddly mixer with no CUE function (no room for it, remember?), no VCAs, and so on.
Again, this gets back to this "beauty case" problem. The mixer you want here DOES exist...in fact, there's a few of them...but there's no way you'll be able to jam one of those AND all of the synth's basic modules into a small skiff like the above. Or rather, you COULD...but the result will be very fiddly and annoying to use due to the tight spaces left for your fingers in amongst the knobs and cables. Personally, I would go with a larger cab and larger panels to minimize ergonomic issues than going this small, which will just maximize them.
To that? Well...not really. There are a few min/max discriminators, but pretty much nothing can track CV direction that I know of, with the exception of this Ladik module. However, there is one arithmetic module worth looking at IF you can find another 2 hp so it'll fit, and that's Mystic Circuits' ANA: https://www.modulargrid.net/e/mystic-circuits-ana
It won't read movement, but it does have some useful twists on the usual discriminator circuit in that you can not only get your minimum and maximum CV levels out, but the difference value between them as well as a multiplication function for those. It also has a very interesting variation on the typical sample-and-hold, plus a BIPOLAR squaring function that results in variant square waves. Like I said, not exactly the same thing, but more than capable of trickery that would fill the bill.
Still, it may just as well be a situation where you have to wait for Ladik to crank out some more.
Well, if you want to go with open rack frames, I think there's a housing solution here that you'll dig (provided you're in the USA). I ran across a firm out of Missouri that builds serious, solid portable racks with a very shallow footprint, and I now have several pieces from them. Check this out: https://prospeakerparts.com/collections/rack-cases-carpet-covered-2f/products/procraft-10u-12-deep-equipment-rack-10-space-made-in-the-usa-w-rack-screws This could also hold a power conditioner or a tile row at 10U, but if you only want the space for 9U worth of frames, they've got that too: https://prospeakerparts.com/collections/rack-cases-carpet-covered-2f/products/procraft-9u-12-deep-equipment-rack-9-space-made-in-the-usa-with-rack-screws And there's several others that they do that are fairly Eurorack-friendly, including some snazzy angled racks for studio use.
But the back's open! Ahhh, that's easy too. They carry unmounted rack rails and both solid and vented rack panels, too! You could even go as far as mounting a 7U panel on the backside of that 10U unit and then a Eurorack frame in the bottom 3U for mounting power inputs and supplies, or even a dedicated 1/4" output pair...then just slap some blanks on the rest, and there you go! And that 12" depth is super-easy to tote, but if you go with a smaller cab, you can get into 9" depths: https://prospeakerparts.com/collections/rack-cases-carpet-covered-2f/products/procraft-6u-9-deep-equipment-rack-6-space-made-in-the-usa-with-rack-screws And in fact, there's one sitting to my immediate left right now, housing the remaining patchbays on the right side of the desk. Nice stuff!
Tony Rolando really knocked it out of the park with the Maths design, actually. The Maths is ridiculously capable as a modulation source or, since it descends from the Serge DUSG, you could even use it as an oscillator. It turns up all over the place, almost like a standard device, because it's VERY difficult to overtax it in terms of complexity...despite it being deceptively simple to use. 20 hp of lightning in a bottle!
+1 on the "bigger case". Also, you might reconsider that $1200 budget point...Eurorack ain't cheap. Plus, as this stands right now, you have no way to justify having a panel meter in that 3U skiff; builds of that sort (which many of us here discourage, actually...they're NOT apt to come out like what one sees on YouTube, which is where this idea seems to keep coming from) simply don't have space to allow a "toy module" of that sort. And I call it that because there's not really anything in that 3U build that needs metering as such...it's just there for looks, in other words, and you need function here more than eye candy. sacguy71's right on point here; a Tiptop Mantis will still be eminently portable (hell, Tiptop even sells a gig bag that's sized for it!), not a huge budgetary hit at $335 street, and provide WAAAAY more space. Granted, with that space comes the death of your $1200 max idea, but it opens you up to putting together a far more capable build that can be worked out long-term, instead of dropping the $$$ on it right off the bat.
Well, they're not white, but you can fit two of them into 8 hp: https://www.modulargrid.net/e/ladik-u-081 Also a fader, but with a lot more tricks up its sleeve.
Well, it wasn't that long before "Atom Bomb" when much of that gear was dirt-cheap. It was around that time (1993-ish) that I got my TR-909 that I had for years for a mere $700 from Daddy's Junky Music, and not long before that when I snagged my Jupiter-6 for $750.
Yes, synth shopping used to be fun! You didn't feel like someone was sneaking around behind you, trying to snag one of your kidneys for payment!
I had a long look at this, and came up with a variation on it that seems to do a better job at sound design overall. No percussion modules...sort of...
See, the problem is that when you try and build tons of functions into a small build like this, you wind up shortchanging ALL of those functions as a result. So what I did here is to kick this back to being ONLY the synth. With a few surprises...
Top row: Voicing, obviously. The dual input preamp is at the left end, with a pair of envelope followers below this in the middle row. The single VCO was replaced by Void's Gravitational Waves, which gives you a complex oscillator plus a ring modulator. Then, to process the Gravitational Waves, I added an Intellijel uFold after the 4-in 2hp mixer, which will allow for even more complex timbral behavior out of that dual VCO. Then this can be submixed again via the dual inputs on the Ripples VCF, making it possible to combine both the wavefolded AND clean outputs. After that, an Elements clone from Antumbra...which is the "sort of". Since you can use all sorts of physical models as "activators", and since you have mallet models for this, the Atom can then give you two channels of all sorts of bonks and thumps, PLUS it gets its own stereo VCF (Bastl's wild new dual-peak stereo VCF, the Ikarie). But if you want even MORE bonks and thumps, I put in a Make Noise LxD after the Mordax, so you can process its outputs through those lowpass gates.
Middle row: Mostly modulation. First are the envelope followers for your input channels, which allow you to extract dynamics from inputted signals and then output them as modulation CVs. Links, then a Doepfer noise/sample and hold module as both of those things were missing from the original version. I also put in a dual slew limiter in 4 hp (Ladik) which doubles the job of the Joranalogue while halving the space! Plus, you get selectable slewing for CV up-motion, down-motion, or all motion. Then the Clep, Stages, Maths, and Shades...followed by a Happy Nerding 3xVCA to give you some dedicated VCAs for modulation use. After this, a Quadrax/Qx setup gives you four loopable 2-stage (or 3-stage, depending on the control input and mode settings) envelopes.
Bottom row: Control. The little white thing is a Konstant Labs PWRchekr, which lets you have visual feedback on your DC rails. Then the MIDI interface was replaced with Expert Sleepers FH-1, which can work as an interface for the DAW's MIDI, or you can just as easily connect any class-compliant USB MIDI controller to this. Tempi got added because it has some special interconnectivity with the Rene, allowing you to be able to toss much of the previous clock gen/modifiers as a result. After that, I put in a mono-in, stereo-out digital effect specifically set up to work with the AUX setup on the Toppobrillo Stereomix2. That mixer gives you VCA control over levels, autopanning, VCA over AUX sends, mutes per channel, a CUE bus, headphones, and a full AUX setup with a mono main send and stereo return. I then kept the Befaco OUT as it not only has your ganged stereo output control, but it also makes the switch from CUE to output a tad easier if you patch the Stereomix's CUE out to the CUE in on the Befaco.
So...uh...where are all the VCAs for the audio? In the mixer, that's where. Also, everything was rearranged for ergonomics, especially the move of all of the hands-on controller stuff to the bottom row. And where's the power supplies? They're in the rack itself, as I opted to rework this for a powered cab such as the Doepfer LC9 or, better still, one of Case From Lake's portable 3 x 84 cabs. If you can go with an INTERNAL supply, you do three things...
1) You open up 12 hp of space
2) You can get a beefier supply than the uZeus
3) You avoid having power circuitry anywhere near the audio path, which is something you definitely DON'T want! Not to mention, those little uZeuses can get HOT, and internals can be overspecced to the point where they don't do that. The Doepfer LCs have their hefty, toroid-equipped ones, and Case From Lake is pretty much the Burger King of Eurorack cabs: special orders don't upset 'em, so you can have it YOUR way!
Now, one thing that made this rework kind of...ah...dicey was the size. A really good sound design rig needs quite a bit more of that, which again gets back to Case From Lake and their ability to resize from their basic designs. Instead of 84 hp for a width, you could go to 3 x 104...or even 3 x 120 or 3 x 126! They do that. If this build were allowed to sprawl out more, it would then have the space to be taken up to a serious sound-design level. This build is good...but if more space could make it BETTER, well...
Also keep in mind that the Penrose will quantize ANY incoming voltages, just like any other quantizer. You could just as easily input a fixed offset voltage, controlled by a knob, and you'd be able to "play" the offset, which is useful for setting up drones and being able to then transpose those with the offset knob. But the Penrose's ability to user-define scales means that your knob-twists will ONLY fall into the designated pitched programmed into the quantizer.
This ability to "force tuning" is one thing that makes quantizers...especially scale-definable ones...super-useful for generative work.
Well, we were all sort of dazed and frazzly as this was the "morning" (ie: around noon) after the event, which had run to just before dawn. But I hit breakfast at a Denny's (in Kalamazoo) with him and several other DJs plus one of the promoters. He's a great guy, very low-key, but DEEP into the music. Also, this was a hot minute ago...like, say, 1998-ish. Interesting breakfast conversation that kept drifting from the post-WWII electronic scene to the present day; Surgeon definitely had caught the connection between the two, and the discussion kept getting a tad mind-blowing for some of the DJs (albeit NONE of the Detroit guys...they've known about that for a loooooong time!).
Talk with them for a bit, too...doing "modifications" to their base-model cabs is something they're experienced at. If you need more power, wider rows, etc...they do that.
Yep, you'd BETTER keep that Doepfer cab...because, as we all know, empty Eurorack spaces have the ability to summon modules to fill them! Do not anger the wandering Eurorack modules!
The other thing is that Case From Lake has THE cab you want...go and check out their pages on Reverb (or right now, Facebook; they seem to have blown thru their stock listed on Reverb!), and lo and behold, they've got "flat" cabs that curve slightly to raise the top and middle rows, making for a much better ergonomic result. The other thing I like about them is that they can and do modify their base-level designs. More power? Extra tile rows? Extra width? Yeah, they do that. Plus...and this is the huge one...they're CHEAP! You normally don't see cabs like theirs without prices that go several hundred higher.
He was packing the joints when I did an event with him back in the late 1990s, actually! Surgeon's skills aren't just due to the imposed limitations, though...he has a very singular ability to use the minimal elements and STILL make tracks that sound fully-formed. And that's not an easy trick; when you try to do more with less, it's invariably far more difficult than trying to do more with more.
OK. let's go over these questions one at a time...
1) Noise can be used with Marbles in a number of ways, besides the point that it's useful for a number of other purposes...both modulation AND audio. But as an example, let's take the "jitter" input. By introducing noise at varying levels through this input, you can increase the instability of the Marbles. But there are loads more possible than JUST that. Suffice to say, noise is pretty essential in general in ANY synth, and there's tons of points in the build where a totally random source of that sort would be useful.
2) By using the Penrose quantizer. What you'd do there is to define a set of notes that fit the overall musical scheme you're working with at the time, then feed a modulation source into the "Input" and a clock pulse to the "Trig". The trigger fires the sample and hold circuit in the Penrose to lock in a single voltage value from that modulation curve, then the Penrose "rounds" that value to the nearest assigned scalar step. By using very sparing clock pulses (Skipper at, say, 3/4th of maximum stochastic skips) to add harmonically related blips every time the voicing connected to the Penrose fires.
When you consider what's going on there, it's really very sample-and-hold-like...but the resulting control voltage values get constrained to ONLY the ones that you've defined, not totally random values.
You should be able to order the COUNT module directly from Adventure. Go here: https://www.adventurepedals.com/eurorack/count and then drop them a line inquiring about shipment.
Another great source...especially since I'm betting you're in Italy...would be Case From Lake in Brescia. Won't fix your VAT in the EU, but at least the locality will lower the shipping cost!
+1 on the Mantis (or similar) suggestion. Actually, have a look at these guys: https://reverb.com/marketplace?query=case%20from%20lake&make=case-from-lake Bespoke construction, wood cases, all very amply powered. If you can't find a Mantis, just check out their 2 x 104...which comes in at a whopping $29 more than the Mantis while offering the same room plus a serious wood cab.
Microfreak isn't a bad idea here, also. It's definitely got more of a "west coast" sound going on...plus you get the touchplate keyboard, which I know tends to wig out synthesists that have no modular experience. But really, it's more like the bastard love-child of a Sound Easel and a Microbrute...quite interesting, really. The other nice thing is that Arturia is always coming up with new oscillator modules and other voodoo that can be "blown in" via USB.
But, yeah...$1k isn't a doable goal here unless you're willing to bust out the soldering gear. But if you DO want a way to get proper modular performance while not spending an arm and a leg, then go here: https://www.tangiblewaves.com/ I have a 160-space AE system here, soon to get punched up to 180 spaces when I add the controller/mixer tier. And while that might sound HUGE (and yeah, it IS), the AE stuff is extremely affordable; my system, thus far (which includes the new 20-space row and the modules slated for it) has run me around $6k-ish...and $6k for something that big isn't something you can do in Eurorack, even if you DO kit-build the system yourself.
Or you could sum the four outputs down to either stereo or mono, and feed it through a stereo filter (like the Morgasmatron or the Dual Borg) or any mono VCF you prefer. Even better, if you want to keep the outputs totally separate, you could even go with several different filters for coloring each output differently. Lots of possible options there...
No, I think Jim means "64 hp" is too small. And I concur.
Also, those "very large" mixer modules are NOT simply mixers. Unlike DJ mixers, modular mixers can contain loads of other SUBmodules. Have a look here: https://www.modulargrid.net/e/toppobrillo-stereomix-2-silver-panel
That there is, in my opinion the best balance of control, capacity, and amenities in a Eurorack performance mixer. At 24 hp, it's actually pretty small for what's inside it...which consists of:
4 input VCAs
4 autopanners
4 AUX send VCAs
Mutes/solos
Cue (this is the thing you're looking for, right?) bus
Separate cue mix output
Mono AUX send bus
Stereo AUX return
Headphone preamp (with selectable Cue monitoring)
Ganged stereo output control
Metering
...and of course, all of the manual controls. Basically, you can use this same module to control levels, your FX, stereo placement and more all thru CV. So what it ALSO does, therefore, is to replace all of the modules that you'd need to build one up in...oh, geez, I don't even know how many hp that would take in discrete modules!
Hence the point behind Jim (and I) telling you to ditch the "beauty case". Sure, they LOOK convenient...but when you fill out 64 hp with a scad of small-spaced modules, what you're REALLY going to wind up with is a small box with an eff-ton of tiny, annoying to use controls, slathered in patchcords due to the lack of space, which then complicates things even MORE because they get in the way of the tiny, annoying to use controls.
Sure, small cabs have their uses. Using them to house/power a few "mission specific" modules to augment something else is a great idea. They also seem to be very convenient for YouTube presenters to misrepresent how to build a PROPER modular system, which is one reason why they pop up on there so often. Other than that, hell, they don't even make for a decent doorstop! Seriously, circular-file this idea and look into a suitable cab instead...or if you're still also wondering why, then...
Intellijel Palette 62 (3U + tile row at 62 hp, powered) = $299
Tiptop Mantis (2 x 104 hp, powered) = $335.
So, $36 more gets you a cab that you can build a real, workable, PLAYable, and fully-outfitted system in. I know what choice I'd make...
And "cheap" should NEVER be used in the same sentence as "Eurorack"! If you go with "cheap" there, you're apt to wind up with a boxful of crap! And in fact, that's exactly what's linked above...
Here's a better solution, one which should satisfy the patchcord jones AND let you get way on into the more complex aspects of modular without costing an arm and a leg (maybe just everything below your elbow, though), without being too confusing, and WITH experience on many of the primary modules while all in one device. And we call it the "0-Easel".
First up, go here: https://reverb.com/item/38531145-impossibleshape-0-rack-3-tier-stand-for-makenoise-0-coast-0-ctrl-strega-2020-birch What this is is a 3-tier stand specifically for Make Noise's 0-Ctrl, Strega, and 0-Coast modules. By putting those together in a single rig, you'll actually wind up with a rather potent Eurorack starter setup which can be easily built off of AND which has tons of capabilities.
But here's the thing...and this explains the odd name...what this is is a redux of the same ideas behind the famous Buchla Sound Easel, albeit without Don's separate control/audio scheme and a few sneaky "extras" thrown in thanks to the mad genius that IS Tony Rolando and the addition to this "ecosphere" of the Strega, which opens up avenues that the Buchla doesn't necessarily have. Better still...
Basic Eurorack build (that works): avg. $2500-4000
Buchla Sound Easel: If just the 208c module: $4000. Adding the appropriate case and the annoyingly-discontinued 218 controller: double that (because Buchla + discontinued = NOSEBLEED PRICE)
This thing: Rack = $90, Make Noise devices = $1497.
If I were in this situation, I'd be shooting for that LAST deal! Comparatively, it IS che...ah..."inexpensive" relative to other solutions, AND it has enough sonic potential that you'll still be finding new tricks several years down the road. AND...when you're ready to add more Eurorack stuff, it all connects 1:1. Sounds like a much more viable solution!
Y'know, what we REALLY need here is a multichannel quantizer that can accept/understand Scala tuning tables. Maybe blow 'em in via a USB connection...? I don't think such a critter exists at present so if any module developers are reading this...well, you get the idea.
I did this a bit differently. Instead of just fixing the gap, I opted to whip up a potential "destination" rig. But since the rack space is really constrained (2 x 88 [not 89!] hp), what I wound up doing is to shrink things down across the board so that you can get a lot closer to a true generative rig.
I tried to keep as much as was possible here, but several devices got chucked out in the process. One of the first to go was the buffered mult; since you'd use the VCOs separately(ish) in generative work, and also because it stole 4 hp, it became unnecessary here...and tbh, in a small build like this, mults in general should be avoided as they rob space from function. Use inlines instead.
Top row: Starts with the P/S, of course...then the Disting, followed by the quantizer (I presume this is for quantizing modulation curves, which is a basic generative technique), then all four sources. After these, one of the Veils (I resized both to the new model) for level control before filtering with the Ripples. Then I placed the Xfader after this, as it offers an interesting opportunity to set up autopanning. The Doepfer mixer is next, then there's all of the effects modules, followed by a Happy Nerding OUT. Now, the reason for that specific output module (aside of the transformer isolation) is that it fixes several problems. For one thing, it gives you a headphone preamp. But much more importantly, it has a second input pair...which then allows you to split the mixer's output between the Verb and the Monsoon, and then you can fly one of those in over the other via the second input. Also, the ganged stereo controls here are definitely easier to balance than the separate controls on the Doepfer passives.
Bottom row: A 2hp Nse first, because there weren't any noise sources for the Marbles to play with. Then the Marbles itself got smushed down to a Pachinko...same module, more or less, minus a few hp that gets used elsewhere. And one of those places is the Quad LFO next to it; to really have a decent generative setup, you NEED extra utility LFOs to provide modulation curves that other modules can reprocess. Maths, Veils, and then a little Frap 321, which allows you to attenuate, INVERT, combine, add offsets, etc. It's pretty much a modulation mangler, allows you to generate more complex modulation results. Then a dual ASDR is after this, giving you a couple of 4-stage envelopes for your audio chain, etc. Then the cryptic-looking thing that's next is a dual window comparator...and THIS is definitely the missing link for generative work. Comparators generate gates under specific voltage input conditions...so, say, an LFO is patched into one and the comparator's threshold is set for 3.5V. The comparator will output a gate each time the LFO's output exceeds this, meaning that you can also use your modulation sources as on/offs for signals that are thruputted through VCAs. But this is no normal comparator, but a WINDOW comparator...meaning that it has TWO thresholds, and you can get gates and triggers when the input is below both, between both (the "window") and above both. This way, you can derive several possible signal states from one single modulation source...and of course, since you have TWO here, well... Anyway, speaking of derivation...the next thing is exactly that, a Derivator. This module also fires off gates...but this time, they're dependent on the direction of the CV that's being inputted. So on this, you have gate-outs for "rising", "falling", any sort of movement at all, or "steady". Now, if you're wondering "why all these effin' GATES?"...the next module answers this. Yep, a Boolean logic module...but this one's a lot more capable than the original Ladik one, as ALL of the gates are usable simultaneously. So, with all of the gate outputs based on CV/mod behavior, you can then take "steady" gates and use the Boolean module and combine everything in those, and get a WILD panoply of possible off/on activity. And to make that even more bonkers, the thing after the Boolean module is a dual probabilistic pulse skipper...so you can actually have random(ish) dropouts in your controlling gate signals. Coupled with logic, like here, that's a potent module. And then the clock divider ends things up.
Now, look closer. You'll notice that, in my build, there are a lot of modules that depend on other modules...and not necessarily the SAME modules each time. For instance, you could use the Skipper for the Minibrute's clock output, keeping things in the modular in pace...except for those odd little "hiccups" that also cause their own strange things to happen, or you could just as easily use one channel of it to randomly "suppress" control gates from the Booleans, and so on. THIS is how you do generative...it's not a matter of "patch A to B and turn knob" at all, but a system that contains LOADS of complex possibilities depending on how it's patched, plus in this case, you have to factor in the Minibrute that's being augmented by the build. IMHO, this does that pretty well. But no, it doesn't use everything you have presently AND it uses some things you don't have on hand. Like I said, it's a "possible end-result" build. Hopefully it's of SOME help.
Hmmmm...OK, the first immediate issue here seems to be module size vs. case size. If you're insisting on using "big" modules like the above, you're going to have to go much bigger on the cab itself. As in, say, 2 x 168 hp big.
OR...you can go to a 2 x 104 like the Mantis or even some 126 hp widths like you see from MDLR Case and Case From Lake, but shrink the modules down to a more sensible size.
Case in point: you have a Grids in the build. Now, ignoring the simple fact that Grids is discontinued, it fits into 16 hp...if you can find one. Now, click here: https://www.modulargrid.net/e/calsynth-ugrids-matte-black-aluminum That's ALSO a grids...but a clone of the original, with the same (open source) circuitry, but it fits into HALF the space. It's also available new, whereas the original is not. Another example here would have to be the Behringer 911. OK...it's an ADSR envelope generator, no CV ins, fits in 8 hp. Then you flip around to Doepfer's listings, and there you'll find their A-140-2. 8 hp, check. But one ADSR? No, this has TWO. And you also get CV over duration! Yes, it's more expensive...but as a rule, much of Eurorack follows the ubiquitous "You get what you pay for" rule.
My suggestion would be to take this rack and then see how to replace the single function modules at the larger sizes with multifunction ones at smaller sizes. With the possible exception of the Maths, Pressure Points, and Pam's, everything else you see there has variants of some sort that up the functionality, shrink things down so that more can be added, or both. And again...larger case. Fact is, you should design this in a MUCH bigger cab first...then start shrinking things down while aiming for that Mantis form factor.
Basically, what you need is simply a single module: an input preamp. At least, if all you plan to use the modular for is processing. If you want to control the modular with the M1, then you'll need a MIDI-CV conversion module as well. As for whatever goes in between those, that'll depend on what you're trying to do with the build overall.
However, one thing I'd also like to point out is that if you have all of these synths and drum machines, you really should consider treating THEM as "modules" as well, routing all of their I/Os to a central patchbay. I've used this working method for over 25 years now, and it massively speeds up the interconnection process between devices.
I had a bash at this...but it should be noted that I've "exorcised" the drum functionality. Sort of.
Yeah, it's sorta different. But at the same time, more solid and less confusional. Normally, I use the upper row for "voicing", but given the timing situations inherent in the build, I opted to reverse the upper and lower 3U rows here. Here's why...
Tile row: OK...there's a Noise Tools tile at the left end. Not only does this provide a sample/track & hold and noise, it also gives you a slew limiter AND...most importantly...a single clock control for global tempo, if circumstances dictate. And the reason for being able to use one knob for all clocking is right next to it: a Plum Audio Temps Utile, which can reprocess that single clock signal into several channels of...STUFF. The QuadrATT is next for attenuation, polarization, submixing, etc. Then the Audio In...and a surprise, the Intellijel Stereo Mixer. Now...remember what I said about "sort of" removing the drums? This is part of the "sort of". Instead of leaving the excess of drum modules in the build (seriously...a drum machine is a FAR better and cheaper way to go), I removed them, but put in the MSCL stereo comp/limiter at the right end of the bottom row and the extra stereo input on the "output" (now the Stereo Mixer, which connects directly to the cab's 1/4" jacks) so that you can route a stereo drum machine feed into the modular, screw around with it some, MASH the crap out of the dynamics so it bangs properly, then mix that stereo signal onto the main synth's stereo out.
Top 3U: Konstant Labs PWRchekr for monitoring the DC busses, the Triple Sloths, then four modules for messing around with clock signals. The COUNT is a pulse counter, outputting gates for each "tick" in a sequence from 0-7. Then a Ladik Dual Delay is next, for shifting clock pulses in time. The "meat" here is in the middle, though, where you'll find a dual Boolean logic module, Frequency Central's Logic Bomb...and after that is a Ladik Derivator, which generates gates based on incoming CV behavior. All of those modules, taken together, are capable of serious alterations of your clocking signals. And right after that, of course, is the sequencer; the Eloquencer was just too HUGE, so I went with a similarly-featured but smaller device, 1010 Music's Toolbox...also eight channels, but with touchscreen control, micro SD sequence storage, and BOTH analog and MIDI signal outputs. SSF's wonderful Tool Box multiutility module then gives you some waveshaping, a comparator, an arithmetical CV processor, inverter, diode OR combiner, and a 2-in/1-out electronic switch. Following this are four free-run LFOs, Maths, then an After Later DVCA which is a pair of VCAs based on the Veils design that also offers 2-into-1 mixing. Next to that is a CV/modulation manipulator/mixer, Frap's handy 321. And lastly, envelopes courtesy of a Xaoc Batumi/Poti.
Bottom 3U: Ensemble Oscillator, then this is paired with the Moon Phase stereo VCF. After that are TWO Knits (two, because you want to be able to detune them to thicken up the sound) then a Recovery Motormatic v2, which is a ring modulator into which you can feed both Knits or, if you want something more off-kilter, just one Knit and then you'll use the Motormatic's internal (and somewhat ill-behaved, which can be quite fun) sine oscillator. Another DVCA then controls the dynamics of the Knit outputs before this gets sent to the phkia VCF/VCA combo. Then instead of the Monsoon, I tossed in its logical successor, namely Mutable's Beads. Last section handles mixing and output levels, starting with a very simple stereo processor, an Erica PICO DSP for providing basic reverb, chorus, phasing etc. This becomes VERY useful with the Toppobrillo Stereomix2, which is a 4-in, stereo-out performance mixer. This mixer has CV over level, panning, AND FX send on each input channel, which also have a CUE and MUTE function on each. The CUE is useful for tuning, etc via the Stereomix2's headphone preamp...OR you could use it to feed the sidechain on the MSCL comp/limiter. The main purpose of this is, of course, to "crunch up" the drums via the external stereo in, then feed this to the Stereo Mix tile...but you can also send the drums directly across to the Stereo Mix tile and then use the MSCL to hammer the crap out of some other signal.
So, why no drum modules here? Basically, it has to do with space and cost; let's say that your basic drum module is, oh, an average of 8 hp width, plus you also need its sequencer, which also eats space bigtime. Now, if you want to emulate, say, a Roland TR-909, you'll need ten modules for the voicing, plus a honkin' big sequencer such as Erica's Drum Sequencer...which burns 42 hp in of itself. Soooooo...assuming each drum module averages $200 ($2000) plus the Erica ($600) and a cab for 122 hp (or more) of housing for JUST those things, you could probably buy an ORIGINAL Roland TR-909 (albeit in "beater" condition...for "nice", double that) at today's typical 2nd-hand prices. So, uh...yeah, screw that. It's far better to make the synthesizer more elaborate and capable, because you can get ample drums for LOADS less in a format that steals a minimum of module space by just getting a good drum machine, and you then have things in the synth to mess with the drum machine's feed. This is how it really works...instead of trying to make a modular into a specific device, it makes more sense to use a PROPER modular's capabilities to mess with external signals, because whatever you use internally can ALSO quite often be used on external audio. And also, you can easily crosspatch control signals from various devices, such as clocking everything from ONE device (easy-peasy, given that nearly all synth companies use a +5V gate/trigger for clocking). So the best purpose for modulars is often to use 'em as a "signal nexus"...sort of a massive spaghetti-bowl-interchange of signals coming and going all over. That's what's up here.
First up, Jim's comments are very much spot-on. Let's try and find some fixes...
Sound sources...there's an easy fix here, one that gets you more space while avoiding a functionality trainwreck, and that's removing the drum modules altogether. In the words of Rocky J. Squirrel, "That trick NEVER works!". The reason is that you're really MUCH better off using a dedicated drum machine...plus, if you have one with trigger outputs, you can always use that to fire some modular modules to add THOSE sounds alongside the drum machine's. The crux of the problem comes down to two points: expense and space. For one thing, building something akin to a TR-909 in modular can easily start to veer off into the general price range of an ACTUAL, ORIGINAL 909. And it would be large, more than likely taking up much of an entire Mantis cab in of itself, crowding out other modules.
Clock modifiers...OK, you first need a head-first dive into Ladik's listings. Right now, I'd peg them as the BEST overall source for stuff to mess with timing and/or logic. Plus, they're CHEAP...and offer hard-to-find things like their Derivator, stochastic Clock Skipper, and so on. Anything that you can add to tamper with the even, metronomic flow of time works here...to Jim's list above, I'd also add trig/gate delays, electronic switches, stochastic sequencers (Euclidean, etc), chaotic modules (Turing Machine, etc) that output trig/gates, and on and on.
As for the Tirana sequencers...don't use them for pitch OR modulation. Instead, they're very useful, along with a chromatic quantizer and an adder, for automating transpositions of CVs. Tempo changes, key/mode changes, and the like...that's the very best use for them, sort of like the "shorter" Serge Sequential Programmers.
Lastly, what does "crazy FX" mean to you? Weird time-warpy things? Hideous audio mangling? Something else? The Deflector Shield is definitely "odd", to be sure, but frequency shifting's been part of modular for a while...hell, it was part of the later iterations of the pre-modular Mixtur-Trautonium! Plus, some things become "crazy" when you put them together, but not necessarily on their own. Case in point: working with signal inserts. The Doepfer A-106-1, a variant of the MS-20's Sallen-Key filter pair, has one of these...so you could put a delay in there and have increasingly-filtered delay returns. BUT WAIT...so, you could THEN add an Alright Devices Chronoblob2. It's a delay, sure...but in its "original" mode, you have...yep...YET ANOTHER INSERT, this time in the feedback path, so that every repeat in THIS can be altered as well by yet another module. So...how about some spring reverb? Add an Intellijel Springray II...and OH HELL THERE'S ANOTHER INSERT!!!
...and on and on and on. So it's NOT simply a case of finding the right module, per se...combinations of specific modules that might SEEM to be fairly "vanilla" on first glance can actually turn out to be utterly bonkers when patched the "right" way.
I don't have much of an argument against the current module categories, but there DO seem to be some points where this could be improved.
For example, "Delay" contains EVERYTHING that delays a signal of some type somehow. So if you want an echo-type module, it's here...but then, so are trigger/gate delays, which aren't at all like the first example. You can kindasorta get past this by using the second category blank ("Clock modulator" finds the binary signal delays, "Effect" helps with the audio signal delays, and so on) but if the user who posted the module didn't add any of that to the "Category tags" on the module page, you'll miss it.
The "not-dual/not-quad" issue is another one. In some cases, posters for these sorts of modules that aren't either of those occasionally will put "triple" units into one or another...but it's far more normal for modules that fit that criteria to only be listed in their primary function category. For example, Zlob's VnIcursal VCA: six VCAs behind one panel...but if you're not looking for a "VCA + Mixer" in categories, you either might not notice it or, if looking in the "dual" or "quad" categories, you can miss it altogether!
Another annoyance: buffered mults vs. passives. Pull up "Multiples" with 2 hp chosen as a size limiter...and you get deluged, requiring you to pick through the results all the same. A "buffering" category could be an appropriate fix here, plus it can include devices other than buffered mults.
A few years back, I would've likely said this was all nitpicky. But given the explosion in the overall size of the databases here, some of these "basic" descriptors are running into problems as their categories are being stuffed chock-full of modules! "Oscillator", for instance, contains 1,235 entries in Eurorack alone. Sure, you can limit this further with a second category...but now, even THAT results in a plethora of entries; going to "Oscillator" and "Digital" still turns up 211 entries.
The worst part of this is that any fixes will require not only site recoding, but it'll ALSO require a knowledgeable crew to go through ALL module entries to add/remove flags as needed. And yes, without the manufacturer lock in place; this is another problem altogether, though, as manufacturers use the lock and then more or less abandon the entry pages, which helps no one. Frankly, if I had a dollar for every time I've had to point out that Clouds is a discontinued module, I probably coulda bought Mutable Instruments themselves, lock, stock 'n' barrel! So it's a potential nightmare if approached wrong...but if done right, it'll STILL be a huge undertaking, but it'll result in a fully-revised database.
Well, let's see...I'm presuming this is based on a pair of Intellijel 7U cases joined by the Joiners.
First up, there were a number of modules in this that just didn't fit or make a lot of sense. For example, using the AJH Mini Mod when, fact is, you can get the same general functionality out of adding Intellijel's Noise Tools tile + the Quantum Rainbow that was already in there, and save money in the process. I also chucked out the extra buffered mults (not necessary...you really only need the one, up by the VCOs; the rest can be UNbuffered inlines), the force sensor and the tandem I/O tile.
Also, Maths stayed as well as the PanMix...I'll explain why when I get to those...
Top tile row: MIDI interface added here, using the cab's USB port for the DAW connection. Then the buffered mult and Zeroscope, followed by the QuadrATT. After that, there's one of the two Dual VCAs, primarily for controlling output levels to or from the Panharmonium, then I also put in an FX send/return with an extra pair of jacks.
Top 3U row: The little white sliver is a Konstant Labs PWRcheckr, which gives you some visual indications of your DC rails' health while operating. Then the next thing is Shakmat's SumDif CV adder/subtractor for tinkering with CVs directed toward the oscillators. Then your Cs-L and Odessa, followed by the Quad VCA...and then, the XPan. What this pair does NOW is that it provides a number of CV-controlled stereo ins and outs that feed the stereo input of the Panharmonium, allowing you to generate a stereo mix to that module's stereo input. The XPan also has sufficient inputs to allow a "backflow path" from your mixing section below BACK to the oscillator row.
Second 3U row: A Plankton ENVF got added to provide an envelope follower for one channel of the Stereo In below. After that is the Quantum Rainbow 2, which can feed all sorts of noise signals to various points, but which is mainly for providing a more complex noise source for the Intellijel Noise Tools if needed. Poti/Batumi, Contour1, Qx/Quadrax...and then I took care of the ADSR needs with a Xaoc Zadar + its expander; this is, in fact, what happened to the Instruo ADSR gen, as it made very little sense to have only ONE EG in the same space that could be occupied by a far more capable QUAD EG instead. Following this are your main timbral modifiers: Evolution, Blades, and the dual LPG.
Lower tile row: The Input module is here, then Noise Tools to provide S&H, slew, clocking, etc. After that is Ritual's Pointeuse, which is a bidirectional CVable switch. Next is the other QuadrATT, then an Intellijel Stereo Mix...added to allow you to manually sum the Arbhar, Magneto, as well as other stereo signals back into your main mix, as if you had a stereo return on the PanMix (which it doesn't have, but which this tile DOES take care of). Another Dual VCA then provides either levels for this, or for other modules inputting into the mixing chain. And the Output is last here.
Third 3U row: Modulation fun...Sloths first, then the Morph4 for global-ish modulation processing. But I also put in a Frap 321 for this, which can make for a useful "internal" modulation modifier/mixer for just the Maths, or for anything mod-wise. Right after this is a dual VCA from After Later, based on the Veils VCA design, which again provides CVable level control and/or mixing for your modulation signals. Then Maths...and all of the stuff before, I should note, are there to REALLY complicate what Maths is capable of. That's one thing people overlook about Maths; since it's billed as a "be-all, do-all", most people don't always consider modifiers for it...but when you DO, and when you set things up right, Maths can REALLY shine. So, between the left side of Row 2 and this end of Row 3, plus the extra widgets, NOW this has a serious modulation section, VERY capable of outputting some very complex voltage curves! After this, your Arbhar and Magneto are located above the PanMix and below the output stereo submixer.
Bottom 3U row: There's the second cab's PWRchekr, then Pam's. But after Pam's, there's several modules that optimize what your Plog is capable of turning out. The 2 hp thing is a clock counter which ticks off 1-8 (actually, 1-7, as your thruputted pulse is always the "0"). After this, you've got a 1-in 4-out clock delay, then a probabilistic clock skip for adding semi-randomized clock "dropouts". And all of that provides more complex signals for the Plog to chew on, so that IT can output lots of potential rearrangements of your clocking. I also put the dual comparator and that Ladik Derivator next to the Plog so that they can input their gates to the Plog as well. Both sequencers are next, and they're followed by the Harmonaig and Shifty for quantization and arabesque processing. Then the PanMix returns...and here's why: First of all, this NEEDS a proper stereo mixer with panning and with VCAs at the output stage. Secondly, the XPan just doesn't do this in the "right" way, as it's mainly designed for stereo input/output situations, and some of what you'll be mixing is mono...ergo, panning (with CV control) is pretty useful. And having your final VCAs in the final mixer is a great way to combine functionalities to save space AND up the build's potential.
This isn't TOO much more complex than your initial design...but I put in quite a bit of "sneaky" functionality that, if and when it's needed, can take this to another level altogether. Much of this is going on in the modulation sections in rows 2 and 3; adding the 321 as a "pre-processor" to feed a composite modulation signal in with the Morph 4's processing of other signals just adds mod signal complexity...plus you also have that dual VCA/mixer in that slot. By adding these, the modulation really gets kicked up to another level by just 12 hp of panel space. And there's several points in the build where that sort of sneak happens...so even though this primarily uses most of your present modules as well as the ones indicated in the second speculative build, there's little bits here and there that synergize in that sort of manner. In the end, there's not all that much ADDED...but what was added really stitches the whole build together.
It's definitely NOT modular...but from my experiences in both hardware and the present-day software versions, the Synclavier is the sine qua non for this sort of harmonic series tinkerage. And nowadays, it's no more expensive than a typical multi-VCO module: https://www.arturia.com/synclavier-v/overview
And if you MUST make it part of a modular setup, then you can either cobble up a way for Silent Way/Volta or Ableton CV Tools to talk to it...or you can use VCV Rack and the included Host modules. Definitely loads easier to integrate as software than it EVER was in hardware, from my experiences.
OK...I had a bash at this, starting with my recommendation for the cab...in this case, I used a 3-row 126 hp design by Case From Lake that runs roughly $650, powered. The nice thing about Case From Lake is that they can do bespoke alterations to their base-level cabs, so if you need beefier power, more depth, etc etc...they do that.
Woooooo...was this ever NUTS! Normally, I would say that drums don't belong in a build, but in this situation, the requirements all revolve around timing and sequencing, so I knuckled-down and had at it, deleting a number of modules that either made no sense in the build's context or which were better-filled by other modules.
Top row: This is generally "voices" in my builds, and it's that here as well...with some tweaks. All of the original percussive modules are at the left side of the row, and then I added a pair of Bastl state-variables (mainly for the Chimera and Plonk) and a six-in manual stereo mixer to submix the entire percussion complement down to stereo. I didn't bother with VCAs here, because percussion largely doesn't require them. After that, there's the Spherical Wavetable stereo VCO and an Intellijel Morgasmatron...in this case, also without VCAs as the performance mixer below has those.
Middle row: The percussion sequencer is on the left end, conveniently located below the percussion modules. Then the next several modules are all modulation: a Synthesis Technologies Morphing Dual LFO, Intellijel Quadrax + Qx, a dual VCA based on the Veils VCAs, Tiptop's MISO for modulation signal alterations, then ADDAC's Stochastic Function Generator. Since so much of what's in the build depends on a certain degree of stochastics, I figured that this would be a more appropriate substitute for Maths. Then a Happy Nerding FX Aid is located right by the Toppobrillo Stereomix2, which gives you four input channels with VCA over levels, CV panning and AUX send, a headphone preamp (with a CUE function), manual mutes, and an FX send/return path.
Bottom row: Mainly, this is timing and control: Mutant Brain for MIDI, Pam's, O&c, Sloths...then a Doepfer noise/sample and hold module. Then the real clock manipulation starts with the QCD setup, followed by a dual gate delay and a dual probabilistic gate skipper, followed by the Illyana Boolean module. On the other side of this, there's a Joranalogue Dual Window Comparator for picking gate signals off of modulation. Then the Tool Box also has another comparator, along with a number of other useful submodules that all have potential uses in this build. The Ladik Derivator also picks off gates, but this time they're dependent on CV movement. And for CVs, on the right end is a Time's Arrow, a stochastic pattern generator which I've tandemmed here with the Stochastic Inspiration Generator to cause much more complex melodic/harmonic pattern behavior than just the SIG alone.
This is a pretty complicated build. Things that I dropped included the Hexmix, all of the mults (use inline ones...no sense in wasting functional panel space!), the trig-to-MIDI modules, and the multiplexer...mainly because they didn't appear to have any functions that couldn't be filled by changing other modules in the build, or because they simply didn't appear to make sense. So there was a good amount of "streamlining" going on here, even though this result might still look overly-complex. Some others, such as the PEG, were just too big to be sustainable here, plus they also were able to be swapped for similar/identical functions in other modules that made more sense, space-wise. Plus, some things here violate my usual build pattern (most notably, the performance mixer being on the middle row) because...well, space has demands. But this final result is pretty cohesive, partly from dropping some modules, and partly from speccing different modules that could take up the slack for the removed ones. So, it's better...it's COMPLICATED, but yeah...better.
See Page 59 infra here: https://back.moogmusic.com/sites/default/files/2020-07/Mother_32_Users_Manual.pdf
There's two possibilities here...one of them is that the clock input might be set to the wrong mode, and the pages above show how to correct that. The other could well be that the Euclidean module is only outputting trigger pulses, and the M32 prefers to see gate pulses, which have a longer duration.
One other point: you won't have full CV control over the M32 from Eurorack without the use of a module such as Erica's MSCALE, as the M32 uses -5V to +5V as its scaling, and generally you don't see negative CV values in Eurorack modules.
The point about utilities being the secret sauce of modular is interesting. I guess, like Nick said, its hard to grasp at first because unlike oscillators, samplers, filters, envelopes, LFOs etc, 'utilities' don't really exist in the conventional world of music and are unique to modular (to my limited knowledge).
Sure they do! Let's look at an actual instrument...in this case, bass clarinet, as I know those like the back of my hand.
While you won't have a sampler on a wind instrument, everything else actually IS THERE...it's just not apparent. The "oscillator" then becomes the vibrating reed that energizes the air column, "filter" is all sorts of timbral modifications that can be done with fingering, altering embouchure, etc, "envelopes" are controlled by breathflow and tonguing, and "LFOs" are whatever you do to induce vibrato...embouchure movement, changes in breathflow, physically moving the instrument, etc. Plus the "controller" is all of those keys and pads and stuff. But are any of these things "modular", per se? Again...sure. Modern compositions are chock-full of what you'd call "extended techniques"...and one that comes to mind is using a trombone mouthpiece instead of a single-reed mouthpiece on a tenor sax. Sounds pretty "modular" to me!
I live in Australia which has unbalanced mains power. How much of a factor is noise? I've read that modules tend to put out a pretty hot signal, but yeah I might just follow Jim's advice and try without, then attenuate, then balance. I'm only running it into either an octatrack or external soundcard.
With UNbalanced power (which we also have in the USA), yes, isolation becomes important to deal with all sorts of sonic crud, hum, etc. And one other useful point: if you get an output module that has a balanced output via transformers (like Happy Nerding's Isolate), you can overdrive the transformers a bit and get a nice signal warmup. That's great for basslines and percussives, to get them a little more crunchy and in-yer-face.
Re VCAs: what's the difference between a VCA and an envelope/LFO module? also VCA vs attenuator?
A VCA (stands for Voltage Controlled Amplifier) is a circuit that controls gain. You feed a signal in one side, then send a control signal to it, and the original signal is now under the control of that control signal, which can be all sorts of modulation signals. There's two topologies, also...AC-coupled and DC-coupled...and two types.
AC-coupled VCAs are definitely for audio ONLY. These don't have DC coupling, so they can't pass signals below the subsonic range. As a result, AC-coupled VCAs tend to be logarithmic in response so that their dynamic changes fit with what human hearing considers to be changes in "apparent loudness", which we perceive on the logarithmic dB scale.
Then there's DC-coupled VCAs. These can pass DC, which means they can be used to control the amplitude of modulation signals. And since the changes these make are linear, it's possible to use these VCAs to modify modulation levels to make the modulation follow specific scalings, etc which are VERY necessary for oscillators, VCF cutoffs, etc. But also, there is a hazard in having these as your output VCA(s) as the DC they can pass isn't something you can hear...but your monitors will notice, and enough DC fed to a speaker will cause it to pop as well as also potentially causing amplifier damage.
It's also worth noting that VCAs usually don't provide a gain factor higher than 1, meaning that if you send 2.5V into a wide-open VCA, you should get 2.5V out the other side. Some do, however, and these are often useful at the end of the audio path...either as a preamp for the mixer, or to "push" the level after the mixer.
Envelope generators and LFOs are modulation sources. You can feed their outputs through VCAs to control their levels, but they themselves aren't designed to be the control circuit for the level changes. Instead, a VCA needs a modulation source to be fed to its control input so that its gain factor can be changed to anywhere between factors of 0 to 1, imposing those changes on the VCA's thruputted signal.
Attenuators also control amplitude. But attenuators can control CV voltage levels, so if you need to have a device track at 1/2, you'd use an attenuator in the device's control path, and this would allow you to have half-scaling for that device. This can be very useful for manipulating VCF cutoffs so that they can track in proportion to, but not the same as, the main pitch CV. Plus, there are attenuverters...these have their zero-point at the CENTER of the pot, and their full outputs are at either end...but with one end, the signal is inverted. These are great for inverting modulation signals for a number of things; you can cobble together a crossfader, for example, with two VCAs and an attenuverter to provide the negative control signal for one of them.
-- ffsimpson
Uhm...yeah. Or rather, no. Jim's done a great job at the debug, and I'll have to concur with him that this looks like no fun whatsoever to work with. Plus, it doesn't really have all that much to it that complements the Mini 2S; I would expect to see more clock manipulation, logic, etc...the sort of modules that can make sequencers do utterly crazy things. But this looks aimless, like a lot of stuff tossed in with no particular order or signal flow, so for starters I'd suggest making that aspect more cohesive FIRST...then the operational flaws should be a lot more apparent.
Yeah, the uses for the send/return modules go deeper than just an insert point for stompboxes. For one thing, it's basically an "unfixed" effects processor, as the actual processing done depends on what you choose to drop into the insert, and this can be as simple as a fuzzbox or as complex as an entire pedalboard. And then there's their uses as extra mono-outs or external signal inputs.
One thing I need to try is using something like this to admit a no-input system's signal into a modular for final manipulation. My AE system has a couple of their 4I/O modules (like these, but with four "soft" I/Os instead of the dedicated input and output), so it should be a good candidate for this.
Hi, folks...since I'm about to bang into the rack limits on MG, I'm going to need to delete a BUNCH of the various builds on my account to get some rack slots back for use. So, if I've done a build for you and you've NOT copied any of the builds, your build is at risk of deletion in the next couple of weeks. If you don't know how to copy a build...
1) Click on the rack itself in the Forum. This will bring up a dialog that allows you to "copy" the build to your account.
2) Copy that build! Then after that, use the "Duplicate Rack" function in the "Edit Rack" pulldown to make a copy that you can mess with.
Please note: for those without Unicorn accounts, you CAN copy the build even if it exceeds your free account's limits on hp and rows, but you won't be able to tinker with it further unless you get a Unicorn account...which, frankly, I endorse the HELL out of! $20-ish for a resource as expansive and useful as Modulargrid is kind of a no-brainer, really. And while these were done for different individual users, they're still super-useful as studies on how successful basic builds are put together, so if you see something among these that you like feature-wise, copy away!
OK...well, for starters, take the DFAM out. It's already got power and a housing, so it really doesn't belong in there unless there's some massive, glaring reason for having it there. The other problem is that this layout is pretty chaotic -- it doesn't have any clear signal paths, so what I did with what was there (after chucking the DFAM) was to try and illustrate what that would be like with what's there now:
Now, this makes more sense. The general "flow" now runs left to right, audio on the top row, modulation middle, sequencing and mixing on the bottom. However, there's still problems here, as reordering the rows now reveals some potential issues. For starters, there's no way to do any major manipulation of modulation signals. The audio row could be expanded and made easier with some additional modules. I also got the power supply modules away from the audio modules, as this can sometimes lead to noise problems.
I don't have a list of what you've got but which isn't in the build, but I went ahead and did an example based on how I would approach the build with the aim being to complete the build. This is what I got:
Top row: added a dual slew limiter for different portamento responses, then a buffered mult as there's several possible CV destinations. After that, a Doepfer module provides noise and basic sample-and-hold activity. The oscillators were doubled, plus the STO pair also has one of Happy Nerding's FM Aids for complex and/or thru-zero FM between these. The mini stereo mixer isn't for "stereo", but to give two different audio routings which can also be tandemmed. After it, I put in a wavefolder for more timbral complexity. Then, Rings, Morgasmatron and the Quad VCA.
Middle row: The clocking and randomness takes up the left end of this, with one of the uZeuses, then an Evaton CLX, which provides two clocks with some integrated logic for manipulating the output. Marbles is next for more complex random behavior, then a Noise Engineering Fractio Solum gives you CVable clock division or multiplication. The COUNT then ticks off each clock pulse from 1-8 to split out various points in time from the overall clock. The Tool Box offers a load of utilities related to timing, logic, as well as a comparator for picking gates off from modulation sources. Last in the series there is a set of Boolean logic gates, which can take several different gates and output new/different ones based on the logic operation as well as what incoming gates are present. The other end of the row is modulation: Batumi (added the Poti expander), Maths, the triple linear VCA, then a Frap 321 for combining, inverting, altering modulation signals. Then for envelopes, you've got a Quadrax with the Qx expander, and this lets it also be used as a very complex cascading modulation source.
Bottom row: I set this up primarily as "control". The MIDI interface is there, then the Turing Machine to which I added its CV and gate expanders. Along with that, there's a 2hp TUNE for quantizing the CV expander. After this, the Metropolis, then the Magneto, and finally the 4ms mixer.
This should give you some ideas as to what additional modules will really open this build up, as well as how a cohesive layout that takes signal flow into account. Notice that the middle row's implementation is easily directed up toward the audio row or down to the control row, depending on what's needed at the time. This is how you keep the patching under control, as it helps avoid the "patchcord snarl" that you tend to get when the various modules are scattered all over the rack. It's not a "final result", though...there's still quite a bit that could be done. For example, my instinct would be to go with a much smaller delay line module so that a proper performance mixer could be implemented here, with this then offering CV over levels, panning, and AUX sends depending on which mixer seems right and which fits. However, I stuck with the existing modules with the exception of the Contour, which became superfluous after I put the Quadrax/Qx in.
I think it'd make more sense to go for the extra hp with the Mantises. For one thing, like you noted, you DO get more panel space with those. It would also let you avoid the compromises that happened to the 4-row build; whenever you condense a build, it's a good idea to try and shrink it down while still maintaining the same general functionalities, plus you can't simply put things back based on size alone; the Rene might fit, sure, but it's not intended to take up the same percentage of space if the build gets cut in half.
This is why the miniskiff issues crop up all the time. Synthesizers need FOUR specific elements: generators, modifiers, modulators, and controllers. When there's ample space, you can go with bigger modules. But if you shrink that available space down (especially this much!), you still need the same four elements AND the same proportion of those...and there IS a limit to how small this can go before you either fall into having insufficient panel space or have to make compromises that cripple a build.
