Posts Tagged ‘form’

Data management

Sep 21st, 10 / 0 Comments

I’m working on a piece now for the free improvisation group Impulse. This is a different compositional approach for me on two levels: 1, I’m not writing notes–it’s a free improvisational environment, and 2, it’s my first major foray into electroacoustic music (Pure Data). It’s a little overwhelming, so I’m working on a data management system. Actually, I prefer to call it a script. It’s not a script in the traditional sense, in which it would dictate every action throughout the piece. Rather, it’s a series of states that the players can activate. So the form of the piece actually comes from the improvisation happening on stage. Here’s the first part of my script, data management. It’s incomplete, and it’s rather lacking in detail–but that’s the nature of works-in-progress…

Time is governed by a master clock counting seconds. This clock sends out the message “time”.  Various patches take note of the time when they begin and at when a preset or gaussian number of seconds have passed, the patch shuts down. The main program also times the sections. The first section will be put in ready-to-end mode after a gaussian time has passed. Once the section is ready to end, other performance factors (described below) will trigger the movement from one section to the next.

Audio files are divided into 4 categories: pads, loops, rhythms, and gestures. Pads are lengthy (~1 min) of ambient sound and are triggered by group performance data. Loops are files designed to be looped until certain performance events occur and are triggered by group performance data. Rhythms are also longer files, but designed to establish a meter or a groove and are triggered by group performance data. Gestures are anything from a single note to around 15″ melody. Gestures are divided into groups by instrument (violin, sax, accordion, guitar) and are triggered by individual performance data. Pads, loops and rhythms belong to a super category of global accompaniments that are triggered by group performance data. As such, only one of these files can be played at a time. Gestures, however, can be played simultaneously with global accompaniments. Only one gesture at a time may be played per performer. In theory, a total of 4 gestures and one global accompaniment might be heard at any given time. However, audio files will probably not be played very frequently throughout the performance. They will generally be played during times of sparseness.

There are 3 to 5 sections (states) (this can be preset before a performance depending on the amount of time available). They order they come in and their exact durations are determined by performance events. Once a section has been performed it becomes unavailable for selection by the program.

Sections:

The section names come from the way I hear an imagined performance. However, these names should in no way dictate anything regarding performance style to the performers. Also, again, these come in no particular order (except for the initial state, which is fairly passive), but are triggered by a combination of time and performance events.

How to parse kinds of musical data:

Pitches, harmony — computer will keep track of the 3 and 5 most frequently played pitches over the past 5, 20 and 60 seconds (list of raw data)

Dynamics — 3 dynamic levels (raw data): p is less than 0.3, m is 0.3-0.6 and f is more than .06

Tempo — number of new notes over the past 5, 20 and 60 seconds (raw data)

Phrase duration — four groups–shorter than 2″, 2″-5″, 5″-10″, longer than 10″– (raw data)

Register — Average MIDI note value, converted to frequency, compared with the instrument’s range (in frequencies), and scaled to 0 (lowest note of instrument’s range) to 1 (highest reasonable note). Output data is 0-1. Parsed in 3 groups–hi (0.7-1), medium (0.4-0.7) and low (0-0.4).

Accents — Number of accents over time (raw data)

Silence — Seconds of silence over time (raw data)

Individual player data messages (to look at for computer’s behavior within states):

$1-3p-(5, 20, 60) — list of the three most frequent pitches in the past 5, 20, 60 sec.

$1-5p-(5, 20, 60) — list of the five most frequent pitches in the past 5, 20, 60 sec.

$1-dyn-(5, 20, 60) — average dynamic level over time

$1-tempo-(5, 20, 60) — average tempo (based on density of new notes) over time

$1-phrase-(5, 20, 60) — average phrase duration over time

$1-reg-(5, 20, 60) — average MIDI notes over time (scaled to instrument’s range, so all players will send out 0-1)

$1-accent-(5, 20, 60) — measure of intensity as the number of accent over time

$1-silence-(5, 20, 60) — total number of seconds of silence over period

Global Messages (to look at for state changes):

g-3p-(5, 20, 60) — list of the three most frequent pitches (global) in the past 5, 20, 60 sec.

g-5p-(5, 20, 60) — list of the five most frequent pitches (global) in the past 5, 20, 60 sec.

g-dyn-(5, 20, 60) — average dynamic level of the group

g-tempo-(5, 20, 60) — average tempo (based on density of new notes over time)

g-phrase-(5, 20, 60) — average phrase duration

g-reg-(5, 20, 60) — average register (scaled to instrument’s range, so all players will send out 0-1)

g-accent-(5, 20, 60) — measure of intensity as the average number of accent over time

g-silence-(5, 20, 60) — total number of seconds of silence over period (multiplied by number of players)

Composing

Sketching

Sep 20th, 09 / 0 Comments

Got down to making some firm decisions on a new work I’m composing for wind ensemble.  To avoid rewriting a summary, here’s an email I just sent the conductor of the ensemble planning to premiere it next spring.

The form and generative/governing concepts are fairly clear to me now.  It is all inspired by the Blue Ridge Mountains and the folk music traditions of the southern Appalachians. . . .  I have transcribed a recording of “Wayfaring Stranger” made in Beech Mountain by Horton Barker, as exactly as possible.  The oddities of that recording (asymmetrical meters, accidental/incidental microtonality) will generate some of the music.  I also plan on using that transcription later in the piece.

I also want to incorporate some abstractions on Appalachian instrumental music, especially variations on traditional banjo and fiddle conventions.  I want to emphasize, however, that I do not want to compose a “bluegrass” piece for band, just abstracting and reworking some hallmarks of that music–like fiddle portamentos and clawhammer banjo rhythmic gestures.

These ideas are perhaps tangential, however, to the main thrust of the piece.  I am fascinated with the timbral possibilities of the wind ensemble and plan to exploit timbre and texture, sometimes to the obscuring of melody and rhythm.  I am planning to include a fair amount of improvisation in order to achieve the kind of intricate textures and independent playing that I want (3 to 5 notes in a box, or melodic fragments played independently of the tempo, etc.).  The harmonic interest will come from a conflict between E min. pentatonic and the octatonic scale C, Db, Eb, etc.  At the end, I plan on using timbre and wide separation in pitch space to help these two harmonic worlds merge naturally.

To help bring together all these inspirations floating around in my head, I got a sketch pad and went to work.  I sketched a few shapes intuitively, with no thought of any concrete ideas.  The top portion of the sketch, which I decided to make my guide for percussion timbres, textural density and volume, is clearly mountainous.  The right side, especially, reminds me of the Blue Ridge Mountains with ribbons of fog below and stars above.

I’ve sketched out pieces before, but this is more graphic than I usually do.  There are, of course, some words and, on the left, some notes on staves, but it’s mostly visual.  There is also a lot of empty space.  I kind of know what’s going to come there, and didn’t feel the need to fill up all the available space.  I found this very useful to me (I’m a visual person) to organize my ideas.  It makes perfect sense to me–the visual representing the aural.  I realize it doesn’t for many people, but I’ve always found it very easy to compose a sculpture or a painting…  makes me wonder why I don’t do it more often.

Of course, the visual doesn’t mean the same aural for everyone (and vice versa).  I’ve often thought it would be a good exercise to give several composers the same sketch and have them write a piece from it.. just to see all the different outcomes.

sketch

After sketching, I notated a more detailed version of the first minute.  It’s convenient that the first minute is so sparse.  Of course there will be a lot of details to work out for the final version, but it will come faster than the more dense textures that come later–and fast is good right now (I want to do a string quartet as soon as this piece is finished).  As the initial sparse clusters/tectonically-slow melody give way to a churning rhythmic section that begins the buildup to the climax, there needs to be a bridge or a transitional section.  For now, I have several 3-5 note motives that can be rhythmicized in many different ways.  These will be building blocks that I could use as improvisatory cells or work out and notate exactly.  They will begin in isolation, then combined into duets and trios, etc.  Either way, it’s too much for today.  I thought about going on to the rhythmic section that will come next, but I want it to evolve organically from the transition, so I need to do that first.  I might start working on the chorale, though, later this evening.

Composing, The Wanderer