I’ve come across some of my original sketches for GMS on my hard drive and I wanted to write a post to describe the process of designing Genome MIDI sequencer. I think it’s always interesting to see an idea go from a sketch on a piece of paper to a full application months or years later. I first got the idea for Genome after hearing about the Line6 MIDI Mobilizer hardware (pre- CoreMIDI and camera connection kit). Having developed bleep!BOX, I knew that mobile devices are always going to lag behind their PC counterparts in terms of CPU power. This limits the quality and quantity of musical instruments that can be generated. I’m sure we’ll see some really impressive DAW’s on tablets at some point, but at the time it made sense to have something that can take advantage of all the cool gear that musicans own and the awesome capabilities of the iPad.
Personally, I’ve never taken full advantage of the gear I owned because I found hardware sequencers too clumsy to use and if I was going to use a computer, I might as well use plugins for convenience. Having a portable MIDI sequencer with a large touch screen and a nice interface would be a really cool thing and would defintely help me get more use out of my gear. I wanted to make something that people could play live or jam with, so being able to change loops on the fly was a must. Here’s my initial sketch:

The original design had no song mode and I ended up re-working some of the user interactions. For the pattern editor, I wanted to make sure that the core actions such as drawing notes and deleting notes only took one tap. Selecting and modifying notes takes a couple taps.

This is a shot of the sequencer from an early prototype. Some of the art was ‘borrowed’ from bleep!BOX. The core layout hasn’t changed much, but a lot has been added.

Same with the pattern editor. I’m a big proponent of building a prototype before getting too far with a design. It does bias some of your design decisions, but it ensures that what you are building is something worth using and helps you figure out exactly what controls are needed, how things will work, etc. It also helps you figure out early if any of your assumptions are wrong or if things need to be changed. Designers also love having the interface thought out in advance and a good designer should know when to argue for a different layout or different UI controls. The prototype proved that real-time looping of MIDI patterns was at least possible on the early Line6 MIDI interfaces.

Here’s where the final design ended up. I experimented with a couple different look and feels and took a couple weeks to get it just right. The main goal was to make an app that looks professional and matches the level of polish and class of the iPad itself. The amount of controls keeps growing, but there’s room for future additions (and of course things could be re-organized into sub menus where appropriate). I want to keep the core functionality as simple as possible, while still accommodating as much of the more advanced features as I can. I won’t sacrifice ease of use for for advanced features. Mobile apps need to be as simple as possible. But it has been, and will continue to be, an interesting challenge trying to bring the features of a desktop sequencing environment to the more limited touch screen.