21M.361 | Spring 2008 | Undergraduate

Composing with Computers I (Electronic Music Composition)

Assignments

WEEK # WEEKLY ASSIGNMENTS
1 None
Module 1: musique concrète, form and content
2 Assignment 1.1: Recording Sounds, Basic Studio Technique
3 Assignment 1.2: Technical Composition, Abstracting Scores
4 Assignment 1.3: Real Composition Based on Score
Module 2: feedback and continuity
5 Assignment 2.1: Feedback Improvisation
6 Assignment 2.2: Feedback Composition and Continuity
Module 3: noise and layers
7 Assignment 3.1: Collecting Specific Types of Sounds
8 Assignment 3.2: Layers, Scenes, Nice Sounds, Linear and Dense Noise
9

Assignment 3.3: Noise or Drone Composition

Special assignment: before lab 4.1, download Max/MSP and begin to learn it.

Module 4: synthesis, algorithm, interaction (Max/MSP)
10 Assignment 4.1: Basic Max/MSP
11 Assignment 4.2: More Basic Max/MSP
12 Assignment 4.3: More Max/MSP
13 Assignment 4.4: Final Max/MSP Project
Module 5: sampling, remixing, polishing
14

Assignment 5.1: You Choose!

Special assignment: prepare for concert

15 Concert

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

1. Using any recording equipment available to you (MD/DAT/cassette/video/HD/memory recorder, cellphone, computer, dictation thingy, etc.), record a large number (20ish?) of isolated sounds. These should be intentionally recorded, i.e., not just plucked from a longer recording session; they will in all likelihood be from a single source (broadly defined), and somehow can be heard as the intended sound. To clarify: these sounds should be short—0.5s to 20s. There are two varieties: sounds you cause to happen, and record; and sounds you just happen to record or find and record, without you causing the sound. Both are acceptable. By single source I don’t mean all you sounds should come from the same source; rather, each individual sound has a single source, e.g. a glass being tapped, as opposed to a glass being tapped while someone slurps some soup. Also record a long sound, 1–2 minutes. This should not have intentional sounds in it, i.e., don’t plan for sounds to happen. Just record a stretch of sound you think will be interesting. Some variety in the sound should make Assignment 1.3 a little easier. In all recordings please avoid bathrooms; when we forget to mention this, we tend to get several bathroom recordings. What happens in the bathroom should stay in the bathroom.

2. Transfer your sounds to a Pro Tools session via the auxiliary cable (and not by importing) onto a single (pair of) track(s). Edit out everything except your 8 favorite short sounds, and your long sound. Bounce to a single file (your 8 short sounds followed by your long sound, each separated by small silences long enough for you to be able to announce in class what the sound is) and copy (not move) the file into the ass-1.1 folder, named (with your name) ‘peter-1.1.aif.’ To copy a file on a Mac®, option-drag it; or highlight it and type command-c, and where you want to put it, command-v. Unless otherwise specified, the only acceptable soundfile format is aif.

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

1. Using Pro Tools, cut your first assignment into its 9 components (8 short sounds and 1 long sound) and make them separate files, nicely edited. Using only your short sounds, i.e., not your long sound, compose a piece using all the Pro Tools techniques you have learnt. Nothing more, i.e., no effects or inserts, no extra sounds. This composition should be purely technical, to show that you have mastered basic Pro Tools editing, and that you have grasped some basics in electronic composition. Don’t worry about expression or general musicality. This is a purely technical exercise. Aim for 2 to 3 minutes. Copy your bounced composition into the ass-1.2 folder, etc.

2. Make a descriptive score of your long sound. Be creative, or boring. Your score should somehow represent your long sound, not just to you, but hopefully also to someone else. If, for example, you plot loudness, you needn’t label the graph as representing loudness. The score will be used for your next composition, and the description of a parameter you have abstracted from your long sound might be used to prescribe another parameter in your next composition. Do not try to account for everything in the long sound.

Examples of Student Work from Prior Terms

These files are presented courtesy of the students and used with permission.

Chaitra Chandrasekhar (MP3 - 1.3MB)
Daniel Hochbaum (MP3 - 1.8MB)
Graham Woolley (MP3 - 2.2MB)
Greg Schroll (MP3 - 2MB)
Kelsey Byers (MP3 - 1.7MB)
Mats Ahlgren (MP3 - 1.4MB)
Matt Gordon (MP3 - 1.9MB)
PJ Steiner (MP3 - 1.3MB)
Rae Zucker (MP3 - 2.2MB)
Roland Tung (MP3 - 2MB)
Sheldon Hewlett (MP3 - 1.8MB)
Tilly Whitney (MP3)
Zach Rich (MP3)

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

Using your score from Assignment 1.2, compose a piece using only your short sounds. All your short sounds must be used. Use all the ’tape’ techniques in Pro Tools we have learned so far. Forget about your long sound; you are working from the score. You are not trying to recreate the long sound. This is a real composition, one that you should feel comfortable living with and saying is yours. It should be expressive or somehow else musical, not just a technical exercise. It should be as long as you think it should be; 30s to whatever your score suggests.

Examples of Student Work from Prior Terms

These files are presented courtesy of the students and used with permission.

Alex Vazquez (MP3 - 1.1MB)
Andrew McPherson (MP3 - 2.8MB)
Andrew Harlan (MP3 - 1.9MB)
Austin Taranta (MP3 - 1.5MB)
Christian Camargo (MP3 - 2.5MB)
Ed Platt (MP3 - 2.1MB)
Eduardo Coutinho (MP3 - 2MB)
Graham Woolley (MP3 - 5MB)
Hubert Roberts (MP3)
Katie Olsen (MP3 - 1.7MB)
Kristen Burrall (MP3 - 2.2MB)
Mats Ahlgren (MP3 - 4.2MB)
Matt Gordon (MP3 - 1.8MB)
Michael Miller (MP3 - 7.3MB)
Michelle Fogerson (MP3 - 1.8MB)
Rae Zucker (MP3 - 3.3MB)
Sari Canelake (MP3 - 2MB)
Sheldon Hewlett (MP3 - 1.4MB)
Tilly Whitney (MP3 - 3.9MB)

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

1. Make a 20–30 minute improvisation exclusively consisting of feedback sounds created by various means within the studio. You can use any sound source there other than voice and pre-recorded media (CDs, pre-existing soundfiles, etc.). Your main sources of sound should be speaker-mic feedback loop with intervening filterbank, parametric EQ, delay, and mixing board faders/EQ. You might be tempted to rely mainly on delay and percussive sounds; but please try to stay focused, at least initially, on speaker-mic feedback. You can “play” the microphone as well as all the pieces of equipment and the mixer’s faders, with an ear toward creating continuous sounds. I’ll stipulate a rule: you have to use all outboard equipment, viz. the MXR delay, SND filterbank, Aphex EQ, and Aphex Aural Exciter. If you like, you can see if intervening things (like Neuhaus’ percussion instruments) elicit new and desired sounds. As far as working methods go, feedback is never entirely predictable or repeatable, so try to record everything—there’s no guarantee you’ll be able to reproduce a sound. Be prepared to record long improvisations—and do think of yourselves as performing. And be prepared for the improvisations to be awful, boring, and painful at times. Listen for the occasional good sound, and when things are going right, you might get a long stretch of wonderful sounds. This recording will be the source material for part 2 of this assignment, and for both parts of Assignment 2.2.

2. Make a 4 minute montage of the best parts of your improvisation. In other words, find the parts you like, and string them together with crossfades. This is not a composition; that happens next week. This is merely to show off what sounds (and changes) you think are kind of neat. Leave the result as an appropriately-named aif in the appropriate folder.

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

1. Using your 20–30 minute feedback improvisation, play around with all you learnt in Lab 2.2: use internal feedback loops with delay and/or reverb. This should still be live performance, since you are using real-time inserts.

2. With your original material and now this new material, compose a 3–7 minute piece using the Pro Tools techniques you know (except you may not use any audiosuite effects—not gain, not pitch shift—and the only inserts you can use are delay and reverb in a feedback loop). Aim for a piece that—whatever its means of production—sounds as if it is continuous and evolving. Think more about cutting and less about pasting. This should help retain the original temporality. If you do re-use material in various sections of the piece, don’t make this apparent. The final product should sound as if it is something that is chronologically viable as an improvisation or a prepared performance. Similarly, layer sounds as you like (here of course you have to paste), but remember that the goal here is a rich homogeneity, not the juxtaposition of diverse sound events. In general, crossfade will be your friend. Leave a final aif, etc.

Examples of Student Work from Prior Terms

These files are presented courtesy of the students and used with permission.

Catherine McCurry (MP3 - 4MB)
Ed Platt (MP3 - 4.1MB)
Jim Wagner (MP3 - 6.7MB)
Kelsey Byers (MP3 - 9.2MB)
Melike Yersiz (MP3 - 3.3MB)
PJ Steiner (MP3 - 4.3MB)
Rae Zucker (MP3 - 7.3MB)
Sari Canelake (MP3 - 4.3MB)
Tilly Whitney (MP3 - 4.4MB)

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

Collect the following categories of sounds. Use the best possible recording equipment you can lay your hands on.

1. Two 30 minute “long” continuous recordings. These should be uniform in the medium scale: not like a hair drier which is uniform in too small a scale, and not necessarily like half an episode of CSI, which doesn’t necessarily vary from beginning to end; but more like a film, or traffic, or walking down Mass Ave, or walking through MIT (not just the infinite but various buildings), or the airport where lots of things happen, or a decent TV episode where things progress or at least change, or traveling on the T. etc. Ultimately these sounds might be chopped into smaller pieces and layered, and used as something to convolve with smaller samples (not this week, though).

2. Ten 5 second “scenes.” These are not like your short sounds from Assignment 1.1, which were of specific things. These scenes are more like capturing 5 seconds of “somewhere,” like a random 5 seconds in Starbucks (or, better still, Dado or Pamplona), or in a lecture. Later these might be chopped into different sized pieces, and some parts crossfaded together to produce a short (20-30 second) changing sound, and used as something to convolve with smaller samples (not this week).

3. Ten distinctly “noisy” sounds, as short or long as you like. These will be used to make both horizontal (linear), temporally changing noise that is not dense, and vertical dense noise (not this week). You can steal some of these from recordings, etc. (Within the confines of MIT this is permitted as “fair use.”)

4. Five potential distinctly “beautiful” sounds. This is of course subjective, but nice chords will work, etc. These will later be the short samples for the convolutions described above. If you can’t find enough beautiful sounds this week, they can be constructed next week from simpler sources, like piling together transposed versions of single notes. Again, you can steal some of these from recordings, etc.

So, just do these recordings, edit them, remove unwanted clicks and abrupt beginnings and endings, and leave them in the appropriate folders in the ass-3.1 folder, which is a little more complicated than usual.

5. Read what is required of Class 3.2, and if you don’t like drawing or generally making a mess with pastels, bring along some colored or black-and-white pictures copied or downloaded from somewhere, or created by someone else or by yourself. Wear sensible clothes!

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

Again, there are four parts. Warning: this is a long assignment. Each should take no more than an hour. You might not get your desired result right away, so experiment a little (just like with Assignments 2.1 and 2.2). These are not compositions. Rather, we are developing techniques that might be useful in making a noise/drone composition, such as for Assignment 3.3.

1. Using your long sounds, layered as much as you like, your beautiful sounds, which you can also layer and process, and convolution, make a 3–5 minute nice sound, hopefully with subtle changes. You can use change pitch, change gain, and mix, as well as convolution, and you can use Pro Tools for layering. Don’t forget you can only have one of Pro Tools and Peak open at a time. Label like peter-3.2.1, etc.

2. String together, either in Peak or Pro Tools (using crossfade), parts of your “scenes,” to produce a 20–30 second narrative. This might not sound that great or interesting, but there is another step: convolve this narrative with your beautiful sounds (again which you can work on) to come up with an interesting and nice sounding chunk, not just a drone. Label, etc.

3. Make 30-60 seconds of horizontal (linear), temporally changing noise that is not dense. Make it grating, screeching, nasty. This is not so easy, so don’t worry if it doesn’t work out well. But learning to do it might help with the next assignment. Label, etc.

4. Make 30–60 seconds of vertical dense noise. Noise doesn’t have to sound horrible, nor does it have to resemble white (or other colored) noise. (I’m using the term “noise” advisedly.) Label, etc.

Examples of Student Work from Prior Terms

These files are presented courtesy of the students and used with permission.

STUDENTS 3.2.1 3.2.2 3.2.3 3.2.4
Arushi Dugar (MP3 - 1.8MB) (MP3)

Catherine McCurry (MP3 - 1.9MB) (MP3 - 1.3MB) (MP3 - 1.2MB)

Cristian Derr (MP3 - 2.4MB) (MP3) (MP3)

Eduardo Coutinho (MP3 - 2.5MB) (MP3)

Joey Wong (MP3 - 1.8MB)

(MP3) (MP3)
Mats Ahlgren (MP3 - 1.9MB)

(MP3 - 1.1MB) (MP3)
Roland Tung (MP3 - 2.6MB)

(MP3) (MP3)
Sari Canelake (MP3 - 3.6MB)

Greg Schroll

(MP3) (MP3) (MP3)
Michelle Fogerson

(MP3) (MP3) (MP3)
Peter Coles

(MP3)

(MP3)
Rae Zucker

(MP3)

Adam Paxson

(MP3)

Alex Vazquez

(MP3)

Tim Dudley

(MP3)

Daniel Hochbaum

(MP3)
Joshua Park

(MP3)
Kyle Backman

(MP3)

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

Using your drawings as inspiration (and not so much as scores) compose a 5-25 minute noise or drone piece—5 minutes is certainly acceptable, but if you have a high threshold of boredom, you could try for something longer; it might not require much more work. Among others, use the techniques we looked at for Assignment 3.2; try sculpting noise using filters and convolution of parts of noise if you like. You can superimpose normal sounds, like from your long sounds (Assignment 3.1), on whatever you come up with, if you think it sounds good. You can also superimpose instrumental or vocal sounds, but be careful not to make this sound like an unnecessary extra, like a finishing touch that ruins the effect. There will be a subtle twist to this assignment, which I’ll mention in lab.

Examples of Student Work from Prior Terms

These files are presented courtesy of the students and used with permission.

Alex Vazquez (MP3 - 11.1MB)
Arushi Dugar (MP3 - 2MB)
Catherine McCurry (MP3 - 5.2MB)
Cristian Derr (MP3 - 3.3MB)
Denis Odhiambo (MP3 - 11.4MB)
Eduardo Coutinho (MP3 - 11.6MB)
Joey Wong (MP3 - 4.8MB)
Mats Ahlgren (MP3 - 9.4MB)
Michelle Fogerson (MP3 - 1.8MB)
Peter Coles (MP3 - 5.1MB)
Rae Zucker (MP3 - 9.7MB)
Thomas Carr (MP3 - 2.9MB)
Tim Dudley (MP3 - 9.9MB)

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

1. Using the list of Max/MSP objects given in the lab notes, build a patch that makes some sounds, music even. Try to use as many of those objects as possible. For the ones you can’t, at least play around with them.

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

1. Build on your Max patch from Assignment 4.1, using the new objects from Lab 4.2, listed in the lab notes.

2. Start thinking about a final project for this module.

3. Choose up to 10 minutes of music, anything, email me mp3s of it/them with a brief write-up, ready for Listening 4.3. I’ll post them on the class Web site as extra listening. Please eventually listen; it is respectful to others in the class.

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

1. Write a definite proposal, and show initial work, for your final Max project. See Assignment 4.4 for details. I’ll look at them in class and suggest revisions. Your initial work should be some serious Max programming, and your proposal should be specific, and feasible.

2. Look at Assignment 5.1, and write a clear proposal of what you intend to do. I will approve it, or make suggestions of how to improve it; once the proposal has been approved/revised, it can’t be veered from. Here is Assignment 5.1.

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

Make a Max patch that makes music, and involves some form of interaction (input parameters, live performance, etc.)

Assignments: 1.1 | 1.2 | 1.3 | 2.1 | 2.2 | 3.1 | 3.2 | 3.3 | 4.1 | 4.2 | 4.3 | 4.4 | 5.1

The choices for this assignment are as follows. Choose only one. Last week you had your proposal approved/revised. Execute (as humanely as possible) your assignment and leave it in the appropriate folder, etc. If it is a Max patch, make your own folder in the Ass folder, and include all required files.

(i) Make a piece based on your own composed material, however you like.

(ii) Choose a “single sound”—interpret that as you will—and make a piece, however you like.

(iii) Polish a previous composition. Written evidence must be provided to show that sufficient work has been done. And of course I’ll compare the polished version with the original.

(iv) Sample from or remix a piece/pieces of existing music so that it is unrecognizable from the original in a significant way. To do this, please supply all original material as mp3s in a folder, named with your name and the name of the original. Be careful with this option: it is fun, and popular, and prone to really bad remixes. Generally avoid simply layering a few songs, or doing something overly hip-hop-ish, unless you explain the intricacies I do not already know about-I’m still learning.

For all these options, bear in mind that this is a course in experimental music. And as with all assignments, listen to the work of previous students.

Examples of Student Work from Prior Terms

These files are presented courtesy of the students and used with permission.

Alex Vazquez (MP3 - 2.1MB)
Arshan Gailus (MP3 - 2.8MB)
Charles Amick (MP3 - 2MB)
George Marzloff (MP3 - 2.6MB)
Graham Woolley (MP3 - 4.8MB)
Heather Brundage (MP3 - 1.5MB)
Hubert Roberts (MP3 - 3.9MB)
Kelsey Byers (MP3 - 3MB)
Kristen Burrall (MP3 - 2.1MB)
Mats Ahlgren (MP3 - 4MB)
Michael Miller (MP3 - 5.9MB)
Rae Zucker (MP3 - 6.1MB)
Thomas Carr (MP3 - 2.1MB)

Course Info

Instructor
As Taught In
Spring 2008
Learning Resource Types
Activity Assignments with Examples
Music