2010 musings: a brief history of RW

2010 musings: a brief history of RW19.01.2010 10:13:34

This is the first post in a collection of three, reflecting on the last two years and the story behind Ruin & Wesen (from Wesen's perspective), identifying the goals and thinking behind the company, and working out a plan for 2010. The last two years have been very intense years, and these posts are a reflection about what has happened, how it happened, and why it happened. They are both very personal, and on another hand very public, because it touches on issues that I think a lot of creative people out there can relate to. They also try to explain why Ruin & Wesen has been mostly "Wesen" until now, as the company was created to produce and ship the MiniCommand and Monojoystick, and how we are planning to change that in 2010.

First Minicommand prototype

The journey starts in January 2008 with me (Wesen) starting to work on my hardware liveset based around the MachineDrum. I discovered DE9 Closer to the Edit by Richie Hawtin in late 2007, and was so fascinated by its structure and its sound that I tried to recreate it, with the restriction of playing only one hardware machine. In February 2008, I built a very simple controller to control the Delay effect of the machinedrum: it was 4 potentiometers sending out simple MIDI sysex messages. That simple controller changed the way I interacted with the machine, allowing me to tweak the delay live, build soundscapes with it, buildups, pads.

I posted about the device on elektron-users, and people started showing interest. The first prototype being really clumsy, it was time to develop a new version, with a display and a bunch of new features (controlling MIDI parameters, and a special function called the macro knob). Around that time, the idea came up to start building a few devices for other people, and see how far that would go to make a living out of it. I had met Ruin over the internet, and we were spending long hours everyday on webcam building electronics, exchanging circuits and ideas. Ruin got me into building electronic music circuits, and I'm very thankful to him for that. Around May/June 2008, we decided to put up a website, and build a mixture of analogue and digital synthesizers. This hasn't happened yet (although we have quite a few prototypes), but we have been working on it in the background, and are going to focus on this in the coming year.

Second Minicommand prototype

I had a strong background in programming, and bits of experience with microcontroller programming and MIDI devices. However, I had to delve deep down into electronics, basically knowing nothing at all about electronical engineering or manufacturing. I shipped the first prototypes (frankenstein-like monsters I feel kind of ashamed of now) to a few very early adopters I want to thank for their early and ongoing support (Thomas, Ryan, Marianna and Lee, thank you so much). In late summer 2008, I started to realize that the demand for the devices was slightly bigger than I imagined, and started looking into a more "industrial" way of manufacturing them, which led to getting a big loan, and buying equipment, amongst which a CNC mill.

Milling the casings

Around this time, in autumn 2008, the current website was set up, and I started working very intensely on building the devices, switching to a bigger CPU, later on adding more memory and the SD-card storage, and writing more and more elaborate firmwares (including the firmware for the first version of the monojoystick). It took me the better of 3 months to come up with a good casing and a final product, which is when I started taking preorders in order to finance the first batch.

During that time, I wasn't entirely sure about the development model I would use for Ruin & Wesen, and whether to go opensource. I decided first to settle on a robust but extensible firmware for the device, but keep the sourcecode for myself. At that time, none of the more advanced features for the Elektron hardware were present, the most advanced feature of the firmware being the capability to transform MIDI notes into pitched samples for the ROM-machines. I think it never was actually clear to go that far into the possibilities offered by the devices, and that that is something that slowly took shape only later.

Some ideas were slowly emerging however, and I decided to give a workshop about the MIDI programming behind the MiniCommand at the 25c3 (the big german hacker congress) in Berlin in December. To make it easier for the workshop participants (something that didn't quite pan out :), I modified the Arduino environment to work with my code, and support the way of uploading the firmwares over MIDI I was using. That was a quite deep revelation, and over the course of a few days, I wrote more and better firmwares than I did in the 6 months before. As a result, I decided to shift the focus of the MiniCommand. Instead of being a pretty specific MIDI controller with a few special features, it would become this opensource platform allowing users to quickly create customized and powerful firmwares. This shift probably wasn't very clear to people, as it was not really well advertised, and this is one of the issues we are going to work on in 2010.

The response to the first batch was overwhelming, and I got a new lab to start building the series. Looking back, this was a very hard time, because I basically didn't have any experience producing a series of devices. It went fairly well, but took much longer than I would have expected, as I ran into a number of different problems along the way (and with shipping!). The first batch took 6 months to build, and here again I'm infinitely grateful to our customers for their patience and ongoing support. The second batch took much less time to build, although they were still a few rough edges (especially with sourcing components). Adding to the stress, I helped write a book about Arduino programming, getting 240 pages out in Spring of 2009. The relief was great when the first batch actually shipped. The customer's response was very positive, and I again want to thank all of you for your support, kind words and patience.

Synevo VST

Ruin at this time was designing electronics, putting together a few greatly appreciated VST instruments on the website (more to come), and designing a lot of modules and synthesizer sketches for future products. Our blog is a collection of deep articles about the technology that fascinates us: analogue modules for Ruin, and programming issues and manufacturing steps for me. In Spring 2009, I also held a number of workshops and talks about the programming side of the Minicommand, the MidiDuino framework (which is a really horrible and confusing name, I'm afraid). This led to the numerous programming tutorials I put up as video and in the manual. An idea taking shape was the idea to gather an opensource DIY kind of community around the device, having users write firmwares that others could use and put up on our website. Due to the ongoing development of the framework itself, this was not really feasible until now, although our forums are a step into this direction. Now that the framework has kind of settled into a stable form, one of the main goals of 2010 is to make the community happen by providing better documentation, tutorials, and ready to go releases. More about this in the next two posts.

Patch manager software

In August 2009, I moved to Weimar into a new lab, building the second batch of MiniCommands (which got delayed because of a problem sourcing the rotary encoders). In the following months, we published the Patch Manager software written by my friend Christian, that would allow customers to change the firmwares on their devices very easily. The whole summer and autumn was almost exclusively spent on rewriting and extending the MidiDuino framework, which groups the functionality accessible through the firmwares. Basically, the framework is the code that makes the communication with the hardware and the controlled synthesizer possible, and a firmware is a small program using functionality of the framework to provide different controller instruments. I am pretty proud of the framework, its elegance and the amount of work I put in it overall to make it robust. Some of these advanced features are visible in the latest firmwares like MDPatternEuclid. However, it may have grown a bit too quickly and scope, and I also have been advertising some of the planned features a bit too much, only realizing later that it would take me much longer than planned to implement them and keep the code quality at the level I expect it to be.

Second batch of Minicommands

After two years of working almost non-stop (with its fair share of all-nighters and 18h days), December was a fairly slow-moving month. Although I shipped a few MiniCommands during that time, and finished the first prototypes of the MonoJoystick, development and support were put on spare flame, due to overall burnout. The intense work up until then, and the realization that it was not going to be possible to go overboard-workaholic like this for much longer, led me to reconsider a lot of things in early January, and searching for a way to provide the support and care our customers are looking for, while also being innovative. One of the issues is probably the lack of transparency about what our products mean and how they are going to be developed. This blog post is the first part of a series that is addressing these issues, our goals and dream as a company, and also a pretty concrete plan of what Ruin & Wesen is going to work on in 2010.