Last Monday I spend another day at the Orgelpark, this time working with Anne (La Berge) and Trevor Grahl on more specific details of the parts of 'Shackle Affair: Organ'. I had some trouble getting started, for various reasons, one of them being too little sleep the night before. As the piece involved a lot of different aspects, both technically and musically, I find I really need to make specific plans for these development sessions, otherwise I get swamped in details and totally lose overview. So another learning moment. Next week I'll have another session and will make sure I'm a bit more organised.

Back to this session: I started out connecting it all: MIDI to/from the Sauer organ, audio from my computer to the speakers, audio from the dynamic mics in the Sauer to the speakers, to the Rat distortion pedal and to my audio interface for live sampling, audio from the little organ ('kistorgel') to the speakers, to the ZVEX distortion pedal and again to my computer for live sampling. My naive idea was to do this through the digital Yamaha mixer that was there. I've seen these monsters before, but never really worked with one. I'm sure they are very convenient once you know how to work with them, but it's not something you just setup as a first-time user. So that and too little sleep and there goes an hour (or two) without any progress. After getting a coffee with Anne I decided to leave it and just connect the organs one by one directly to the audio interface. Figuring out the connections at a later stage.

Another frustration was the Rat distortion pedal. I didn't seem to be able to have a decent volume, until I found out that these things just always feedback when you're sending them an signal. My conclusion: just open up the volume and let them scream, switching them off when you're not making any sound (any guitarplayer could have told me that, but hey, I found out the hard way).

I also decided not to amplify the Molzer organ. The Sauer and little organs are fine with the two mics on each, but I couldn't get a satisfying sound from the Molzer, so decided to use it only once in the piece, and then just acoustic. Limitation saves the day once again.

A MIDI issue with the Sauer organ was solved after discussing it with Trevor, who knows the MIDI console very well. I won't bore you with tech details, but just sending a MIDI reset message just before engaging register assignments works really well. Also, the MIDI reset message works great as panic key, so I can (in theory!) make the Sauer organ shut up from behind my MIDI controllers.

I was glad to also have some time finally with Anne and Trevor to discuss and try out musical ideas for the parts. Very inspiring to have someone interpret your instructions in a quite different way as you anticipated! That's why I love working with musicians and improvisors.

One last note: I have been struggling with the structure of the composition. The basic idea is to have about 10 parts in the piece that are presented to the players through the Shackle System: a visual cueing system where players have the power to cancel new part proposals by the system and also can request a new part at their conveniece. So parts of semi-random length will be proposed in semi-random order (semi meaning: between a predefined min/max value). Players can cancel those proposals, and in the middle of a certain part ask the system for a new proposal. There is no guarantee that in a set of about 30 minutes all parts will be played. This is fine for a Shackle set, because we play quite often, so it is actually nice that some parts don't cross the stage everytime, creating a unique performance at every Shackle concert. But for this piece, that due to its very specific instrumentation nature will probably not be played many times, it would be preferable that every part is played at least once.

The same can be applied to the order of the parts: in a Shackle set it is much better to have parts in a different order everytime, so a Shackle set has the same basic ingredients, but is always different - not only because parts can be interpreted differently everytime, but also because how we play a part depends for a great deal on what came before. But for this piece I do have a certain order in mind, and I would like to have an audience hear that.

So after talking to Anne about it, I decided to make it much less flexible: the order of the parts will be fixed, and the players will not have any influence on the structure. Except for myself: I will have the opportunity to cancel a new part proposal if I feel the current part is not done yet, and I will be able to request a new proposal if I think we're ready to move on. The parts will still last a random lenght (between a fixed min/max value) so the lenght of the piece will not be set in advance. But my ability to advance or extend the piece with cancelling or calling up proposals will make sure it will be within reasonable boundaries (probably between 20-30 min).

Over and out now. On my way to mail a couple of Shackle Sticks to the WORM shop.

Last week I spend a day with Piet van Tienen in the flesh (after working remotely for over a year or so) to fine-tune the two songs we've been working on. The process: Piet would send me a raw version of the songs, played by himself on guitar and voice recorded on a webcam. I would then arrange them, adding drums, bass lines, mellotron-like instruments. Also thinking of my own role as producer/player, where to add my joystick-abstract-crackle and where to process guitar/voice. Now in the studio together it was the time for a reality-check.

The arrangements sounded good when played with the live guitar and voice. Levels need to be balanced for sure, but overall fine. Live sampling the voice and guitar also worked well, but with moderation. I really have to adjust, not playing to much, letting the story tell itself. Same for more abstract joystick-action: hold back, moderate. But especially this last aspect I really have to rethink. Not sure if I should not leave it out completely.

I thought of posting some excerpts, but it's still to fragile. So patience please ;)

In the past weeks I've been collecting various parts of the composition The Sound of the Machine. Originally a piece for Disklavier, tape, flute & electronics and laptop-instrument, I wanted to document the piece plus make a multichannel version that could stand on its own as a tape piece.

The documentation version is done: with the generous help of Anne La Berge I first recorded her contribution to the work. This was a challenge for her, as the Disklavier tracks were on tape (and not of great quality, I didn't re-record those yet) and I wasn't playing along in the improvisation sections. So she had little to play against, but still did a great job. In my studio I then recorded my part, 'live sampling' the flute.

As the tape part was of course already there, the last job was to record the Disklavier. I went to Muziekhuis Utrecht, where the instrument currently resides, installed my two Neumann KM184 microphones and two Schaller contact microphones, and made various recordings. This was a challenge, as I don't have a lot of experience recording piano. I tried different angles for the mics, the majority of them having quite some distance to the strings. Later in the mix I found that the few recordings I made with the mics much closer were definitely the best. Luckily I did record all the sections that way (just once, but that turned out to be enough). Another learning experience.

As for the contact microphones: when performing the piece, I would specifically use a Mackie mixer for its great pre-amps to boost the signal of the contact mics, plus use its superior equalizers to balance hum and signal. I totally forgot about that until Jan the technician at Muziekhuis Utrecht handed me a Soundcraft mixer. Now I'm not very particular about my equipment, but I did hear a difference in signal quality. Oh well. It was way better than the previous recordings anyways.

The last stage was mixing. A very enjoyable job, just me, the tracks and a nice pair of speakers. One thing I learned there was to bring the levels down down down! Especially with the piano and the quite harsh flute+sampling I needed that to make the mix more transparent. I also discovered some nice free plugins from Voxengo, of which I used the Tube Amp simulator as a mastering tool. Much better than the build-in maximizer in Nuendo!

When recording the Disklavier I luckily also thought of video-taping the instrument. I still find the piece more rewarding if you actually see the magic of automatically moving keys and the instrument's struggle to play the massive amount of notes I send it. (Of course I also should have video-taped Anne and myself - well, next time).

For this composition I will use the Shackle System as a way to structure the piece. This system is basically a visual cueing system, proposing parts for the performers to play, giving them the option to cancel proposals and request new ones. In the way we use it in Shackle, the proposals and lenght of the parts is chosen in a weighted random fashion by the system. This makes it very flexible and surprising, also for us as players. For this piece I will probably make it a bit less surprising, possibly by setting a specific order but still letting the players decide when to go on to the next one. Or limiting the number of 'cancels' for each player, so that they will have to think hard if they really want to cancel the next proposal. To be determined.

I've actually composed 10 (or 11) parts already, and in the process of finetuning them. I will keep the rules for each part a secret for now, but below is a list of the names, plus explanations. All searched for in connection with 'shackle' - limiting one's possibilities. And then soon you get into darker territory...

• Fetter (a chain or manacle used to restrain a prisoner, typically placed around the ankles)
  
• Trammel (poetic/literary: a restriction or impediment to someone's freedom of action)
  
• Bilboes (an iron bar with sliding shackles formerly used for confining a prisoner's ankles)
  
• Crackdown (severe measures to restrict or discourage undesirable or illegal people or behavior)
  
• Tether (the horse had been tethered to a post)
  
• Yoke (Dutch: juk / a wooden crosspiece that is fastened over the necks of two animals and attached to the plow or cart that they are to)
  
• Halter (a rope or strap with a noose or headstall placed around the head of a horse or other animal, used forleading or tethering it / archaic: a rope with a noose for hanging a person)
  
• Bridle (a horse's bridle)
  
• Manacle (handcuffs)
  

Last Monday I spend a full day at Orgelpark. As they are closed on Mondays, I was on my own. I mainly worked on amplification of the organs for live sampling and processing. My conclusion after this day:

• For the Sauer organ I will use the two dynamic microphones that are already installed in the organ, to live sample and process the sound. I will also put a Rat distortion pedal on the organ. This sounds great on the organ. After this work is done I should try a drone piece for the organ, in Sunn O))) style.
  
• For the Elbert Kistorgel I will use two Neumann microphones stuck into the opened top of the organ. The bleed from the speakers into the mics will be acceptable. I will definitely live sample and process the sound from the organ, and also extend it with a ZVEX distortion pedal.
  
• I'm not really sure yet how to amplify the Molzer organ. I did test putting a wireless Beyer Dynamic mic in there, but placement is very important as I'd like to get a balanced level from all pipes. So next time I'll put two of those microphones in there on stands, and see if that works.
  

I also tested MIDI control to and from the Sauer console a bit. Triggering notes and changing registers with the joystick. Very cool, but very dangerous: a high risk of stuck MIDI notes. Although I trust my SuperCollider programming, I'll have to test this thoroughly, and then some more. What we don't want is hanging notes during a concert. MIDI coming from the Sauer console will trigger flute samples in my SuperCollider setup. Also quite risky, but I have more control there so I can always break in if things go haywire.

A short documentation video: