Deluge – sequencer, synthesizer and sampler – goes open source
synthtopia.com
synthtopia.com
The Deluge freed me from my desktop, from endless screen time and it allows me to create. It may not be perfect and it may not be perfectly DAWless solution for many because of some missing features like mastering and compression. But for me as a hobby where I just waste time on making crappy electronic music of which 95% will never reach the ears of anyone else (besides my poor wife) it is perfect.
Of course it can also be used by more serious musicians. But for my it enables a unique way of playing with music which I have never experienced anywhere else. And it finally is a DAW which I can (almost) 100% understand.
I've never tried the Deluge myself, but I would really like to.
I use the Deluge to learn music theory, synthesis and a bit of composition and it's so much more fun making music in the garden than to spend another hour sitting in front of the computer.
Synthstrom Audio seems committed to support this device for years to come: they sell spare parts, organize upgrades to newer hardware versions (some more cosmetic, recently though a screen replacement from 7 segment LED to OLED display), offer "refreshment" (new silicon pads, potentiometer and encoders).
Things I hope can be implemented:
* Smooth scrolling: currently Deluge only scrolls in increments of a whole "screen", e.g. 16 steps with a step being anything from a few bars down to something like 256ths of a whole note. This makes it hard to follow and awkward to edit pieces that are not in 4/4 time signature.
* "Accidentals" in scale mode: currently every row displays notes of the same pitch. Deluge allows arbitrary subsets of the chromatic scale (as long as it's at least 7 notes) to form a scale, and in "scale mode" rows for notes not in that set remain hidden. To "escape" the scale one has to switch to "chromatic mode" (a row for every note on the chromatic scale) or to create a new scale by adding notes to rows not in the current scale in chromatic mode. With only 8 rows that means that not even a single octave fits on the screen any more. I hope that something like accidentals can be implemented where the sharpened/flattened notes would simply show up in a different color on the row of the scale note, akin to the use of accidentals on staff notation.
* Support for microtonal scales
* A master compressor
All in all the Deluge is a wonderful piece of gear and I wish Synthstrom Audio well. I really hope going Open Source will benefit their business as much as customers are going to benefit from this.
Things may improve if MIDI 2 ever becomes widespread. But I wouldn't hold my breath.
When sequencing monophonic sequences, one might be able to work around this by adding a MIDI processor into the signal path that prepends certain pitch bend messages to certain note values - but I can't say for certain how well this work, I never tried it.
Here's a link with some ideas (and scripts) to implement microtonal synth voices with today's firmware version of the deluge:
https://forums.synthstrom.com/discussion/comment/16257/#Comm...
Hope that helps!
You're correct, MIDI is very primitive in its support of microtonality, and afaik MIDI 2.0 improves it only a little: the general solution in OG MIDI is to have it send pitch wheel messages along with the note to bend it to the correct pitch, but it's limited in precision, and prevents you from using the pitch wheel for other purposes. You can do it polyphonically, but only by assigning each note in the chord its own MIDI channel, which limits it to 16 simultaneous notes.
MIDI 2.0 doesn't directly support remapping to different EDOs or tunings or anything like that, so microtonal support is still a kind of hack, but it does at least have a dedicated pitch adjustment parameter now in the note on message, with a higher precision, which is separate to the pitch wheel message - so you can use both. Still clunky but it's better.
There's also various standards for OG MIDI like MTS (MIDI Tuning Standard) or MPE (MIDI Polyphonic Expression) but it's very rare for HW synths to support these, and even soft synths that support them are few and far between.
I guess probably the most straightforward way of doing microtonal hardware sequenced synths is by using modular synths, funnily enough - none of these digitally quantised note values there! I just read the Cirklon sequencer, for example, support the MTS standard to store alternate tuning and can then output CV to modulars based off those tunings directly.
Meanwhile last time I was in NZ a few weeks ago I checked the classifieds - partly to see if there was a used deluge at a good price nearby, but unfortnuately I blew more than double that budget on an amazing soprano sax that happened to be nearby instead :/
Which is a long winded way of saying that I'm looking forward to seeing the deluge's code to see how they're running things inside the box.
So the firmware going open source feels natural and I'm excited to see what the community can do. I guess it's a very huge and complex C/C++ codebase, so it will probably take some time to get started...
Things I'd like to be able to add:
- Oscillators from MI Plaits
- MI Rings sound engine
- A monome Grid mode (either via midi, or via USB serial), so that the deluge can be used as a Grid
I've got a couple 256-button monomes (no longer in production, basically -- they only make smaller models now) and waited twelve years for something velocity sensitive with as many buttons to come around. Eventually the Lumatone did, and now the monomes sit in a closet -- but I'd really like something with smaller buttons, because I divide the octave into 58 pieces, which means I can barely reach an octave on the Lumatone (whereas I could easily reach across three octaves on the monome).
(I bring this up because I'm working on a similar project, but not with a regular grid. My keyboard has 113 pressure-sensitive keys. I've been using JLCPCB, who are surprisingly cheap even for large boards and even when buying in quantities of 5 or so.)
It's not clear from their website, but if you order one of these evaluation boards, it comes with a bunch of small swatches of FSR material: https://www.sensitronics.com/products-xactresponse.php
What I do is I glue a sheet of 1/8th inch birch ply to a sheet of felt using liquid hide glue, then use a laser cutter to cut that into keys. (They're irregular polygons, but you could do the same easily with squares.) I have another piece of laser cut plywood (or the "negative space" of the same plywood, but with the felt removed) that serves as a frame to hold the keys in place. I use packing tape on the back of the frame, sticky side up, to grab onto the felt so they don't all fall out when you turn it upside-down. The felt gives the keys a satisfying squishy feel. Wood screws hold it all together.
In other words, from top to bottom a key is comprised of 1/8th inch birch ply, a layer of glue, a layer of felt, a layer of tape sticky side up, a layer of FSR film, then ENIG-coated copper traces on an FR4 PCB with the solder mask excluded. I also have another layer of plywood (1/4") on the bottom for structure.
I made an early attempt at using some kind of silicone rubber, but I wasn't successful. It might be doable and I imagine there are companies that can make that stuff for you but in the end I decided I liked wooden keys better and they were just plain easier.
I don't have leds on the keys, but if you want that, WS2812's are surprisingly cheap and could be placed by a pick-and-place machine by the hundreds -- if you can figure out a way to get the light to shine through all the intervening opaque parts.
No doubt this gives Synthstrom a huge boost to their (already very strong) reputation. But even more so this gives a huge number of developers an amazing foundation on which to start hacking and creating new devices.
(I own a Deluge and am a massive fan of the company)
Recently bought mine so very happy to hear this news!
I'm have only a remote armchair-kind interest in this kind of device, but they always seem like a fun platform to develop (I'm an embedded software developer). The product page doesn't seem to say, but I found this [1] page on Reddit where someone claims it is based on the Renesas RZ processor [2].
That seems to mean it's an Arm core (there is one RISC-V core in the RZ lineup) with some serious onboard memory and DSP muscles.
Having had some experience with Renesas though at $PREVIOUS_WORK, I would no longer expect this to be a fun environment. :)
[1]: https://www.reddit.com/r/DelugeUsers/comments/cf1mco/15_chip...
[2]: https://www.renesas.com/us/en/products/microcontrollers-micr...
"If there comes a time we stop developing the software and feel 'finished', we'll consider it, but we're a long way from that place so it's not worth losing any sleep over thinking about this at the moment. Rohan’s joy is coding. Not gonna take that away from him and let everyone else have all the fun :) Btw, please don't try debate us on this. We're fans of the open source movement and of course understand the benefits it has, this is a decision we've made."
So, does this mean they consider the Deluge "finished"?
So many devices have unavailable firmware which makes me worry for its longevity. I've given up on a couple of things I'd have otherwise purchased because they believe their firmware is a competitive advantage for some reason (skipped out on the Lumatone because of this).
That if some chinese company doesn't decide to clone it... or Behringer
Bought a Deluge, then sold end up not using, then when they launched the live looper bought again, but didn't find time again and sold it, oh dear :X
Nowadays, without having to commute, finally have some time to produce music but sold all the toys and playing guitar and bass :~D
Seriously this is a cool little groovebox, and it is awesome they are going open source.
*Monday, June 5, 7pm EDT* – Codebase made available, documentation released and Patreon launched.
*Wednesday, June 7, 7pm EDT* – Open Group video Q&A session with Rohan. Will be recorded and made available on GitHub following.
*Saturday, June 10, 11am, CEST* – Open Group Zoom session with Rohan. Will be recorded and made available on the GitHub following.
For basic music theory I like Nahre Sol's book "The Elements of Music" which is affordable and basically all you need to jumpstart.
Your next best source are Youtube tutorials where people "reconstruct" the songs you like i.e. you need to begin by mimicking the music you like with the soft synths they use combined with some production/mixing techniques knowledge. Alice Yalcin puts quite a lot of stuff at Youtube. And what's not to love from other channels like Bad Gear, Alex Ball or DrMix! After some months you will become familiar to what the different synths are capable of doing and even recognize the sounds from other people sounds ("that's a Juno!" "that's an Emulator!" "that's a 909!") and you will build your software arsenal around that, maybe even getting some hardware that adapts to your preferred workflow.
Me, I'm a Logic Pro + Korg hardware workstation person but I may or may not spend most of my weekends TV time watching these Youtube channels.
The DAW is software that organizes all your instruments and melodies and so on into a project file and a VST is the instrument itself, more or less. So you work on a song in the DAW using VST instruments.
The Novation Circuit is a good beginner device, not too expensive, and have a big community. Bunch of tutorials for all kinds of music styles on YouTube.
Most devices also help you with scales and whatnot, so you don't really need much music theory understanding, although of course it helps a lot. But that's something you can look deeper into if you actually like making music.