Adding drum pads to the g4…
Length: 92cm (36.22″)
Width: 20cm (7.87″)
Height: 3cm (1.18″)
Here is what the ISHIZENO G4 grid could look like.
This grid is made of 324 keys and 10 knobs. Keys are illuminated by RGB leds and are sensitive to pressure, tilt, and velocity. Knobs use rotary encoders with 24 detents and 24 PPR resolution, as well as SPST momentary switching.
The controller can be used for both playing and sequencing. In playing mode, 8 rows of keys are used for playing, while the top row is used for sequencing. In sequencing mode, 8 rows of keys are used for sequencing, while the top row is used for note selection. The grid supports playing across 4 octaves and 8 channels, as well as 36-step sequencing. When a note is added to a sequence, the 4 potentiometer settings related to the note’s channel can be set from the knobs located on the left side of the grid, while the 6 analog levels of the note’s channel can be set from the knobs located on the right side. In other words, it brings everything together nicely.
With it, a typical setup would consist of:
The design of the i8 is currently receiving some awesome contributions from Meng Qi, who is a musician, synth maker, and teacher living in Beijing. On this thread, he made it clear that we should build a custom controller for our submodular synthesizer. Unfortunately, I had to agree, and I decided to call it the g4.
You can think of the g4 as a mix between the LinnStrument, the Madrona Labs Soundplane, and the monome grid. It would be made of a large block of black walnut and a grid of 36 × 8 keys, or 288 keys in total. This would give it 4 octaves over 8 polyphonic channels, hence the 4 number in the g4 name. As far as the g is concerned, you can think of it as my last name’s initial, or the initial of Gaston Lagaffe’s Gaffophone.
The g4’s keys would be sensitive to pressure, tilt, and velocity, much like the Keith McMillen QuNexus, but they would be much smaller (3/4″ squares). And most importantly, they would be illuminated in order to be used for sequencing.
The g4 controller would have a built-in battery and a single USB port so that it could be plugged to the i8’s faceplate. Combined with an iPad and a small portable rack like this one, it would make for a totally awesome setup.
Current goal: get g4 and i8 up ready for NAMM Summer 2015.
My favorite designer of modular modules is Mutables Instruments, for many different reasons: clean design, advanced features, open source licensing, and so on. With that in mind, I reviewed all their modules and compiled a summary analysis on this spreadsheet. Bottomline: clones of their entire lineup (Elements taken aside) could fit within 3 i8 modules, using 136-HP of horizontal space, versus the 156 that regular Eurorack modules would take. Bottomline: the i8 will not help you save horizontal space on your rack, but it will help you make things a lot more modular, and it will make it possible to migrate existing Eurorack modules to the Arduino shield form factor in order to get dynamic patching.
I did not realize that the i8 is 42-HP wide.
Our knobs will be illuminated by the LED of the rotary encoders. For this reason, we need to find translucent plastic knobs. So far, our best option is the Davies Molding 1226, but it will need to be customized in order to remove the position marker. Also, having something a bit simpler like the knobs used by the monome aleph would be nice. We’ll keep looking for the best possible part, or we’ll commission a custom one…
All 32 knbos of the i8 will use the Bourns PEL12T rotary encoders, with 24 detents and 24 PPR resolution. They also provide SPST momentary switching and RGB led illumination. The SPST switch will allow users to push a knob to select a channel or a potentiometer, while the illumination will be used to provide a minimalistic user interface on the module itself.
All knobs are the same, therefore we cannot offer both potentiometers and rotary encoders, because we do not know in advance what kind of control a given submodule will need. Since rotary encoders can behave like potentiometers with the right user interface (a number pegged to 0 when turning counter-clockwise) but the reverse is not true, we have to use rotary encoders for everything. It’s not perfect, but it should be good enough.
One of the side benefits of our new design is that we have a much simpler faceplate, made of only three types of components: knobs, jacks, and a USB connector. All but the USB connector can be mounted with round holes, which will make the manufacturing of the faceplate a lot easier. And with a bit of luck we should be able to mount all components on a single faceplate PCB.
Now that we have one Arduino per channel, we do not really need dedicated analog to digital converters. Instead, we can use the ones provided by the six analog inputs of the Arduino. They’re on 10 bits instead of 24, but since we’re going to use them for visualization purposes only, they should be good enough (1,024 levels). And we can always add the dedicated converters later on if we feel like it.