-
Notifications
You must be signed in to change notification settings - Fork 1
Exercises
Full index: rebraining.org/practice/music/kaiku
One pitch class at a time. One voice. Whole notes, whole rests. The rest is the test — listen for the reverb tail, the release, the 1.008× beat. Each register presents differently. Do not skip contrabass.
| # | Title | Range | MIDI | What it tests |
|---|---|---|---|---|
| E01 | Soprano — F# Phrygian, sustained | F#4–F#5 | E01-solo | Timbre and beat in the soprano register |
| E02 | High Tenor — F# Phrygian, sustained | F#3–F#4 (upper) | E02-solo | Timbre in high tenor |
| E03 | Low Tenor — F# Phrygian, sustained | F#3–F#4 (lower) | E03-solo | Timbre in low tenor |
| E04 | Contrabass — F# Phrygian, sustained | F#1–F#2 | E04-solo | Timbre and beat in the bass register. Beat period ~2.7s at F#1 |
| E05 | The b2 — G♮, all registers | G2–G5 | E05-solo | The Phrygian blade. G♮ isolated across four octaves. If this fails, everything downstream is wrong |
| E06 | Piano — full range, reference | B1–B5 | E06-solo | Full piano compass as reference. Checks range consistency across all 17 scale pitches |
Soprano against piano reference. Whole notes throughout. Two things are tested simultaneously: FM formant interaction between voices (Stacks A and B can stack unpredictably at some intervals), and the REAPER ensemble render path.
| # | Title | Motion | MIDI | What it tests |
|---|---|---|---|---|
| E07 | Similar motion | Both voices move in the same direction | E07-ensemble | Formant stacking in parallel ascent/descent |
| E08 | Contrary motion | Voices move in opposite directions | E08-ensemble | Formant separation when voices cross |
| E09 | Parallel thirds | Fixed interval of a third throughout | E09-ensemble | Stack A+B interaction at a third — characteristic FM beating territory |
| E10 | Parallel fourths | Fixed interval of a fourth | E10-ensemble | Stack A+B at a fourth |
| E11 | Parallel fifths | Fixed interval of a fifth | E11-ensemble | Stack A+B at a fifth — often the cleanest parallel interval for FM |
All parts. The question is blend, mask, and alias — not timbre. Does the nasal partial (Stack B) disappear in full texture or dominate? Does the wheel carrier create beating artifacts between parts at close intervals?
| # | Title | Parts | MIDI | What it tests |
|---|---|---|---|---|
| E12 | Full texture — sustained | All 5 parts, whole notes | E12-ensemble | Blend, mask, level consistency across the full ensemble |
| E13 | Full texture — mixed motion | All 5 parts, varied motion | E13-solo, E13-ensemble | Alias and artifact under real part-writing conditions |
Attack, sustain, release, tail. Velocity response. Voice stealing. The Hiljaisuus LFO completes one full cycle at ♩=40.
| # | Title | MIDI | What it tests |
|---|---|---|---|
| E14 | Staccato — pp through ff | E14-solo | Click artifacts, release behavior, velocity response at five dynamic levels |
| E15 | Crescendo — sustained | E15-solo | Sustain stability; whether FM character shifts with amplitude |
| E16 | LFO cycle — ♩=40, slow tempo | E16-solo | The Hiljaisuus LFO completing one visible cycle. Should appear as slow amplitude periodicity in the spectrogram |
| E17 | Re-trigger — same pitch, rapid | E17-solo | 16-voice allocator behavior: phase cancellation, amplitude doubling, pop on re-attack |