Err yep, I’m going to post a new project that you can make at home, every week starting now. My last post about making Piezoelectric crystals at home seemed to be something of a hit getting featured both on hackaday.com and CreateDigitalMusic.com so I thought I’d give you more of what you like while building a sample library for my own compositions.
I thought of covering:
- Contact microphones – construction and creative usage
- Simple electronic music circuits – warm up your cmos chips
- Prepared instruments
- Max/MSP usage – Maybe some generative stuff?
- DIY music controllers
- Studio techniques
If you have any ideas please comment here and I’ll try and cover them. Keep your ideas simple and doable in a 5 min video!
Next time you see the soldering iron logo there will be a tutorial under it.
A little output from a software patch i’m working on to make music out of user input. The input in this case is any number. All the patterns are generated from this number, there is no randomness used in the software, so if you enter the number twice you get the same output. These examples also use one source sample which i was changing throughout. Still lot’s I want to do with this, the melodic parts could do with a lot of work for example…
I’m wondering if I should make this into Forester 2? And if there is any interest I could do a generative music video tutorial of some sort… hummm…
Been super busy here for the last few weeks – I’m making not one but two albums dont you know…. More about them later. For now here is a little video of me testing the currents in Forester 2. Don’t have time to annotate the video I’m afraid, but im basically changing 4 parameters of the current using a midi controller – yes forester 2 will be easily midi controllable. How fast the sound ball thingies move depends on their size and the strength of the current.
Leafcutter John x.
I’ve been building a DIY foot pedal controller over the last couple of weeks and it’s getting close to completion. As you can see above the LEDs are working well. I had the idea that it would be great to be able to get away from the laptop at my live shows so the idea for this controller was born. I’ll show you more as it slowly comes to life.
The m-log is a ‘build your own controller’ kit currently in development by the Owl Project. Using it you can control; Pure Data, MAX/MSP, SuperCollider, Processing amongst other software. The idea is that you have a set of sensors provided in the kit (accelarometer, 2 potentiometers,2 switches, 2 buttons) then you choose how you want to arrange them. You can easily add your own sensors to the kit and it plays happily with most of the sensors from iCube, Infusion systems or phidgets as well as ones from you favorite electronics retailer. I have been involved in a small way in it’s development and I recently gave a talk at an m-log building workshop where unfortunately I had to give back the m-log I had been borrowing from them. The good news was that they gave me one of the new kits they are developing. As I’m currently thinking about re-designing my on-stage set-up this could not have come at a better time.
The standard kit contains:
- Pre-routed wooden body and back – mine is Laburnum
- muio interface board
- Accelerometer (roll & pitch) not shown in picture!
- USB interface chip
- Analogue input chip (4 channels)
- Power LED
- USB cable
- Various Potentiometers, switches, buttons
This standard kit allows you up to 4 analogue controls (which output values between 0-254) in addition to 8 digital inputs (on or off). I wanted more analogue controllers which is acheived by simply adding an extra analogue input chip. You can connect a whole host of different sensors to the m-log including: Light dependant resistors, switches (tilt, reed, button, toggle),passive infra-red detectors, thermistors, potentiometers, bend sensor, pressure sensor…..
The first thing I did on receiving the kit was to have a good look at the wooden body. The owl project have made a good job of selecting very pretty wood. Test fitting the interface board it’s clear that the routing is of very high precision, the USB connector sitting snugly in it’s square hole. Because trees come in all sorts of shapes and sizes the logs in the kits vary quite a bit in size and appearance. When I handled my log-body It felt quite fat and I was not particularly into the feel of the rough bark on my hands so I decided to remove the bark and some of the wood around the sides.
I marked the extents of the routing on both ends of the log to make sure I would not remove too much material. I used small sharp plane to remove wood from the sides until it sat comfortably in my hand.
The next step was to think about what kind of sensors I wanted to put into my log and where to put them. I really like using the joystick controllers in my joy-pad so I decided to include a joystick (which i took out of a cheap USB game controller). I also wanted to have a light sensor (LDR) and a selection of knobs and buttons. I sketched out some ideas and finally came up the following design.
I paid particular attention to the spacing of the knobs and buttons. The controller would not be fun to use if it was a strain to reach the different controls so I took plenty of time to make sure I could reach all easily. The buttons on the side were laid out by holding the body and marking where my fingers fell naturally. I would warn against putting controls too close together because they’ll be difficult to operate (esp knobs). It’s also worth noting that if you cover your entire controller with knobs and buttons you might not be able to hold it without accidently activating some of them (actually this might be quite fun). Finally check that all the components will actually fit into the space provided in the log!!!
I marked out the positions of the holes I wanted in my log and used brad-point drills in an electric hand-drill to make the holes. Warning I would recommend you securely clamp down the body while drilling. The texture of the wood varies which can cause the drill-bit to jam and forcefully rotate the the log or worse, jump out and make a complete mess of your hand. The one inch spiral cut up the inside of my index finger really hurts!
The top of the log is about 5mm thick so you can’t get a potentiometer to attach to it properly. You’ll need to recess the top surface a little to get the nut and washer on. I managed to find an attachment for my Dremel multi-tool which did this perfectly. After the finger incident I didn’t want to take any chances so I clamped the log down and wore safety glasses to do this part.
I made a nice leaver for my joystick out of a piece of Pine dowel. I shaped it with a chisel and sand paper before super-gluing it to the metal shaft of the joystick mechanism. Be careful with super-glue and components, if the glue runs down into the mechanism it might ruin the part.
Owl Simon showed me a neat way of making wooden switches for the m-log. You’ll need tactile switches, a hole-cutter and some extra wood to cut the buttons from.
I used a 16mm hole cutter. I assume the dimension refers to the resultant hole and not the core as mine ended up about 12mm across which ended up being perfect. I cut down the cores into shot lengths suitable for buttons and sanded away any sharp edges. The buttons will be hot glued onto the tactile switches shown below.
For the 4 button section on the front of the log Mounted the tactile switches on perf-board. The brown part of the tactile switch was a bit too long so I clipped them using wire cutters. On reflection It may have been better to drill out the buttons a little more to accept the longer tactile switches, but the buttons I made were not very deep and I didn’t want to risk drilling completely through them.
The perf-board will be hot-glued into position on a little wooden riser inside the log. After this is done the 4 buttons are carefully hot glued to the switches.
I mounted the two side buttons in a very similar way using thin strips of perf-board and hot-glue. I didn’t manage to get a good photo of it but you can just about make them out in the photo below.
Soldering all the components together is fairly straightforward, I had Owl Steve on the end of the phone to help out as complete wiring diagrams are not finished yet. The board is logically laid out and a lot of the connections are repeated so it won’t take long to get into the flow. I used solid core wire for my first connections which was very stupid of me and caused a few problems including a lifted track on the board when I had to re-solder a mistake. Some of the traces are quite tiny so I’d recommend thin stranded wire or stranded ribbon cable (which I took from an old printer cable). The ribbon cable keeps things nice and neat too which is a bonus when you have lot’s of sensors going the the board. My problems were all easily rectified and I soon ended up with a nice rats nest of cables. I don’t have time to go into the specifics of the wiring here but have a look at the muio site for further details.
When wiring up It’s important to keep test fitting the board into the log. This will make sure that your wires are long enough and that everything will fit when the time comes to glue the back on. I also made sure all my connections were correct by periodically plugging the board into the computer and using the muio server software to check the readings coming from the log. It’s a good idea to do this as you go along so you can catch wiring mistakes one at a time. I was wiring quite late at night and I did make a few mistakes but the board didn’t complain and kept on chugging along.
As you can see, there is not a lot of wasted space inside the m-log. All the components live under the board so it’s important to guard against accidental shorts. Before gluing on the back make a final test with the muio server and adjust the on-board trimmers to set the bounds of the analogue input range(s). This enables you to set the high and low boundaries of your accelerometer.
I choose not to use the supplied back because I wanted something really thin and happened to have a nice 2mm thick piece of mahogany veneer lying about. I rough cut it to size, and glued it on with wood glue. While it was drying in the clamps I cut some more wooden cores using the hole cutter to use as knobs. When the back was dry I sanded it into the contour sides and ends. All done!
I’m really pleased with my m-log, It works well and communicates as expected with the computer. The accelerometer gives nice control when rolling and pitching the m-log and I’m looking forward to making some Max/MSP patches to exploit this data. I had a few problems with some of the buttons sticking, but I think i can sort this out without too much trouble.
I’ve started making some abstractions to help in dealing with the m-log data in Max/MSP which will be downloadable as soon as they are ready. I’ll also be making some little m-log software instruments for download soon.
Finally the kit is not available to buy yet, I don’t know when it will be or how much it will cost but if you have ideas or suggestions please leave a comment and I’ll make sure the developers read them.
I’ve been super busy the last few weeks hence the complete lack of posts here. I have a couple of gigs coming up this week (sat/sun) which I’d like to let you know about. I’m also involved in some great projects at the moment which i’d like to share.
1. The first gig – Sat 20th June – is at the Cross Kings Pub, York Way, London N1 0AX. I’ll be doing some new tunes, old ones and some improv bits. Also playing: Liberez, Benjamin Wetherill & the Trumpets of Death, B¿Layachi Cousins Duo. Cost:£6, adv £5
2. The second gig - Sun 21st June - is an early one 6:00 > 7:00pm, Royal Festival Hall ballroom, London. This event is on the closing day of Meltdown 2009 and is before this years curator Ornette Coleman takes to the stage. I’m really looking forward to this, I have managed to get a group of amazing musicians together to play some of my animated graphic score pieces. We’ll be premiering a new one where the audience suggest words and phrases which will make up the score. It’s a free gig too!
3. I just finished a BBC commission for Radio 3, I was asked to work with Shetland poet Jen Hadfield on a new work. I’m not sure exactly when it airs but i’ll let you know when I do. It should be available to stream from the BBC website so you can have a listen even if you miss the original airing.
4. I have spent some very pleasurable days with my friend Tom Haines playing about with hydrophones in a little pink paddling pool. We are working on a new piece which will actually be performed (and listened to) in a public swimming pool using an underwater sound system combined with a regular air speakers.
If you want to come and experience the first and probably only performance of this piece it will take place on July 22nd at Clissold swimming pool 63 Clissold Road,London, N16 9EX. Bring your swimsuit! More Info.
5. Last but not least I have been tinkering with Forester II and have come up with the beginning of a lovely new interface. Here you can see audio files rendered as circles.
I’m still some way off having finished software for you so please be patient.
See some of you this weekend!
Leafcutter John x.
An experiment using Max/MSP playing a circular pattern of sounds. The distance between two nodes determines the time taken to go from one to the other.
A couple of things make this Vid a little confusing to watch. The frame rate can’t keep up with the position a lot of the time – take it from me the pattern remains the same: 1>2>3>4>5>6>7>8>repeat… etc. Also the tempo sometimes changes without the nodes changing position. This is because there is a master time multiplier off screen which i’m using to scale the speed. The relative times of the nodes stay the same. Also it’s hard to see the connections between the nodes on the video, that’s youtube’s video compression. Oh yeah and then there is the yellow marker who just sits there doing nothing, that’s my lazy programming.
Anyway if you like the sounds of machines starting up or spinning down you might like this.
Leafcutter John x.
p.s. and i nearly forgot to mention that if you want to try out some nodal music there are several free applications available to do this. I’ve not played with them much so i’m not sure what they are like to use.
Tiction Is quite similar in appearance to my program though it’s more fully developed. Mac, Win, & Linux.
Nodal is a fairly deep “generative software application for composing music” Mac only.
I seem to be having a very intense Max/MSP week. Last night I was thinking about interesting ways of generating rhythm. In particular how to create structures which balance recognisable loops with random choices. Saw a really interesting page of flash design pieces which gave me the idea of using a nodal structure with one sound mapped to each node.
In the first incarnation of my software (shown in the video above) the nodes fire out midi messages when they are hit by the red marker. The midi messages go to Logic Audio and play specific samples in a soft-sampler. The connections between nodes are created automatically. The rules are very simple. The marker must move on every beat (a regular pulse) it may only move to another node to which it is connected. The connections are directional so if the connection goes A to B the marker cannot move B to A unless another connection is made in that direction. If a node is connected to more than one node a random choice of the available targets is made. As the video progresses I randomly add more connections which increases the number of patterns available.
The Video below shows a later development of the software where the node locations are positioned by hand. Connections between them are also made by hand. The rules remain the same. In this example there are two discrete systems with two markers. As the video progresses the systems are linked. Again the nodes trigger midi messages which this time trigger piano note samples. I built in probability to this version so fore example you can connect A to B three times and A to C once. Which would make A to B three times more likely to occur than A to C. The connections with stronger pink lines have stronger probabilities. Note that some of the nodes are ‘rest’ nodes and produce no sound.
The output from these systems can become a little monotonous after a while due to the lack of volume dynamics and the regular pulse. All the same I’m very pleased with my days work. I’m hoping to be able to build some of these ideas into the new version of Forester…
Hope you enjoy,
Leafcutter John x.
has anyone got an EMF meter they can lend me for a few days?
@nancynancybeep 'Dancing' is Just WOW
- Building Things
- feather study
- field recording
- Graphic Scores
- Live Shows
- Polar Bear
- Site News
- Software News
- The Making of the 5th Leafcutter Album
- Things Thought
- video tutorial
- Weekly DIY Music Thing
- live cam washington dc mall on Vietnam Day 3 Field recording with Richard Eigner and Workshop with Vietnamese Musicians
- Demetria on Vietnam Day 3 Field recording with Richard Eigner and Workshop with Vietnamese Musicians
- Helen on Vietnam Day 3 Field recording with Richard Eigner and Workshop with Vietnamese Musicians
- Allison on Vietnam Day 3 Field recording with Richard Eigner and Workshop with Vietnamese Musicians
- gironoturno.com.br on Chekhov Challenge now on BBC iplayer