 All right, we're going to talk here about the Nibble Tronic. I have here this brave guy C-trap called since today. He's an electronic engineer and he's really into the fascinating thing of using media controllers and playing with it. He makes musical instruments and he's going to show you how to get there. Unfortunately, the musician is ill today. So I don't know, maybe we have to call someone from the audience to help him. I hope to play some music afterwards. But this guy will show it all. You all right? Yep, thank you. Let's light a fuse then. Okay, first, hello. I'm a bit nervous, but happy to be here. I want to ask you to ask you questions as they arise so you can keep track of what is going on. And yeah, then this is what this will be about. I will tell you what I did with this device. Then I will tell you what others did before me and then we want to discuss if you should do the same. As it is traditional, I start with the finished product. What we have here is the current state of the nebutronic. I should get rid of the paper. We see the mouthpiece at the left. Then there's a pressure sensor. It is connected with a split PVC hose. There are four micro switches right next to the microcontroller board you see there. And then we have a joystick and two more sliders. On the underside, there is another slider and a midi jack. I completely lost track of shit. This is the previous version and I will use it to illustrate the working principle. As you can see, it had quite a lot more buttons and it could only play around 11 nodes. One for each button and another one for no button pressed. The working principle is that you have this mouthpiece, this split hose, which allows you to blow through so it figures like a traditional instrument. You put your fingers on the button and play it like a flute. The pressure sensor either triggers a note, if none has been played before, or controls the volume of the synthesizer. So you can have a volume envelope for every note, like a wind instrument does. The buttons define which note is played or trigger another note when you change your finger placement. 11 nodes aren't that many, so I decided to use fewer buttons to play more notes. And the five of the left hand then were used to encode five bits. So using these five buttons, you can then count a binary up to 31. This worked. You have great tone range. You can play four full octaves with that. If you omit the same tones. But it is a bit hard to memorize 32 distinct finger placement. Nobody wants to do that. And so I came up with a more ergonomic control scheme. Where's the next button here? Screw that. Using only five buttons, you can ditch one more because that one was for the thumb and was kind of hard to play. It became a thumb rest. I then added two buttons from the right hand. And that way was able to encode 60 nodes and with the right hand up to four octaves, which was actually it became playable. You only had to remember 12 finger placements. And you started to play semi-tone. So you had the full tone range of four tapes, which is quite something. Remembering finger settings is better than 32, but not yet ideal. And having 16 numbers available for just 12 notes that I needed, then the final refinement I decided to assign the full notes to the even numbers and the semi-tones to the uneven ones. That way, you only have seven finger placements left. And that actually was manageable and has been played last year on this conference. Not by me. I never learned to do this. This is the resulting control scheme. And this is implemented in the current version. And you see you play with actually only three fingers and have the pinky left as modifier for semi-tones. This works. It still has a lot of unused buttons and has the slight problem that it is pain in the ass to manufacture. All the brass rods and threadings and putting it on, it takes ages. It was kind of unergonomic. There are tons of cables. So time for our next iteration. And this is that, that what came from that. On the other side, we have the four buttons for the notes. And a joystick for pitch bend and modulation as somehow they showed up on synthesizers and never went away. So I thought I'd go with that. And two sliders for whatever. You can assign some continuous controller to that and adjust your sound and whatever you do with the MIDI controller. On the lower side, we have now a slider. The slider next to the pressure sensor is used to encode the octaves. And the whole range is mapped to just four. If you want more octaves, you can program them in. You can actually solve this in software, which is nice. But it becomes hard to find your octaves that way after some time. And we have a battery as a power supply, so fewer cables now. I'm missing the next button. Problem is, nothing of what I did is actually really new, as I found out. And there's tons of prior art out there, especially in the common section of the God of Phone blog, you find lots of people asking questions for their own builds. So there are really, really a lot of EWIs out there. As I said, Instructables. This is the wireless God of Phone. Some build I found. This one is interesting. It is a device by Nia Stina, who started to design these things in the 60s and built the first ones in the 70s, and then actually designed the one big commercially available product from Archi. This is a Nia Stina design. So there's actually one person doing all this thing out there. There are a lot of builds. These things are available. And I have never witnessed one being used in some kind of performance. So they exist, and nobody uses them, which was kind of a mystery to me. So there are the reasons why these things exist. There are a lot of practical reasons. There are artistic reasons. You can control not just synthesizers. You can control your light show. And this might make sense in some settings. But the really big advantage of EWIs is that, by traditional instruments, have an interface that is mostly defined by the underlying physics. With an electronic device, you can do what you want. You have no such constraints, and you can build an instrument for somebody who has maybe a non-standard number of fingers or hands. Another idea would be to ditch the breath control completely, replace it with a foot pedal, and you can build a wind instrument that somebody with a weakened lung or some other problems can actually play. So there is the main reason to design these things, to build instruments that otherwise, for people that otherwise couldn't, can play an instrument. As an example, I have the sketch of an EWI for a Tyrannosaurus Rex with a very long neck, reinforced steel mouthpiece, reinforced claw pads, and another band control input that allows the T-Rex with its poor fine motor skills to actually play this thing. Another good reason to build those is to understand what's there. What we have here is the device that is built by Akai. And in the circle, there's a set of metal rollers. And they are opposite of the left-hand finger pads. They react to touch. They can be moved, but they don't detect movement. And I've seen this, and I didn't understand why somebody would build such an interface. And then I built this. We have a question. Oh, thank you. So if you're really interested, I own this device, and I know how to play it. And if you know how to play a saxophone, it's completely obvious why you want to build it that way. I understood it later, because this is the only wind instrument that I actually own and I can't play it, which is a shame. I'm planning to learn it since two years or so. Excuse me. Are there saxophone players here in the room? Maybe they can help you with that. I actually figured it out, because this setup with the octave selection opposite of the left-hand fingers doesn't work too well. You're pushing on the slider, and at the same time pushing against your fingers, so you can play notes that require you to release all your fingers. And you're just pushing the instrument out of your mouth. And that may be a chromatic element, but it doesn't really work. This could be meditated in several ways, with a neck strap like many traditional instruments. But in the end, I found out that these rollers at the archive are actually used for a tape selection, so no real new idea there. And because they react to touch, you don't really have to push against your fingers. You can move your left-hand fingers completely independent from your thumb. And because these rollers, they give you really good feedback where your thumb is much better than the mechanical slider. So yeah, by building my own device, I kind of imitated accidentally the big one by archive and got to understand why they did what they did, how they came to these conclusions, and I got a really nice understanding about electronic wind instruments. So this is another good reason why you might want to build such a device. Yeah, come on. If you thought you would see something revolutionary here, they exist since the 60s, they make sense. If you have a T-Rex in your life who wants to play wind instrument, now you can build one for him. And you probably should if you want to understand these devices. They pose some interesting software problems. You don't want to spam your synthesizer, but you want to have continuous breath control the whole time and you have to balance that out. The sensor is not quite straightforward and there are a lot of interesting problems to solve there. If you don't want to start from scratch, here are the links and the God of Phone page is great. There are several builds, several different control schemes and there is a lot of theory in there, lots of source code and lots of solutions. At Patchman, there's the historical reference. There are the devices of Niostina and some other exotic builds. Hopkins Electronic Aerophone is actually wireless, which might be interesting to build if you enter that. And of course we have the Nibbletronic, which you have seen here and several pictures. So thank you very much. I have the, I'm under the impression that I've been quite fast this time. I'm kind of as relieved as you and are there any questions? Fantastic. Thank you very much, Seatrap. I think we all appreciate it that you come over to speak about an Nibble instrument. Actually, I can help you as well to point you to some saxophone players who play this acai kind of devices. Are there people here among the audience who play this instrument? Is there a possibility maybe to show as well the instruments, because you have it here in your box? I actually brought it with me. And the full bill of materials is on the page and there are some bought, which you can't order less than six. I don't want to find an orchestra, so yeah. Not yet, but there are enough people here to make that kind of band, I presume. Is this functional? Can we test it without having damages? Not right now. The firmware is kind of bugged currently. I came here to finish the development and to work on the next. iteration because as I said, this solution for octave selection is less than ideal. Probably there are some people who can help you with that. Are there? I thought that maybe some people around here might know something about building hardware. I'm not sure, but... I don't see any hands yet. No one? Okay. We found one. Second. Very nice. Okay, there are questions I presume. Here, from the internet? Yes. The question is why did you choose a non-canonical setup for the finger buttons? A non-what? Excuse me, can you repeat that please? Why have you chosen a non-canonical setup for the finger buttons? I hated playing the recorder at school and I wanted to go as far away as possible from that. I love that. We know that feeling, I presume. Is there another question somewhere here? One second, please take the microphone, sir. Playing the buttons. What exactly does the breath do? What does the pressure sensor do? Is it for volume? It controls the channel volume, I think it's CC7 or something like that, of the whole synthesizer. So you can define your volume envelope for each note. Okay, thank you. Right, and someone else? Question? There are four mics here, someone on the internet. Yes, there, please, sir. Take the mic. So if you want to see somebody who's really good at it, playing the acaiivi, you go to YouTube and search for Original Rays by Michael Brecker. And then you see it's for 15 minutes. Original Rays played by Michael Brecker and his band. Okay, thank you. No one else here with a suggestion. Is there a band that can be now assembled here in the back? No. Thank you very much. Once again, thank you C-Trap for this fantastic presentation.