Step 01 / 5
Build the detuned source and amplitude contour #
Set giStage init 1 in the full script. Use 3 band-limited saw voices at -9 dB each, spread by 6 cents between adjacent voices. Reverse the stereo weights across the channels. The amplitude opens in 0.0044 seconds, decays over 0.582 seconds to 0.630957, and releases over 0.186 seconds. Basic Shapes is represented by this explicit waveform; Shimmer, where present, becomes slow independent pitch drift rather than the original unison engine.
| Csound control | Value |
|---|---|
| Sources | 3 × saw, -9 dB |
| Adjacent pitch spacing | 6 cents |
| Amp A/D/S/R | 0.0044 s / 0.582 s / 0.630957 / 0.186 s |
| Instrument gain | -6 dB |
| Velocity multiplier | 0.4 + 0.6 × velocity |
Sound after step 1 — WebM/Opus
Next: Blend the second source and animate the filters →Step 02 / 5
Blend the second source and animate the filters #
Set giStage init 2 in the full script. Add a triangle layer at -13 dB, 0 semitones and 5 cents. Add the quieter octave-down sine sub. The notch center starts from 772 Hz and moves by -0.8 octaves under the second envelope. A positive depth opens the spectral ceiling; a negative depth moves the notch downward. The serial second filter keeps the specified high or low edge under control. Exact wavetable morphing, Warp/Fold and free-running source modulation are not implemented; the code below defines the chosen static spectra and note-local drift.
| Csound control | Value |
|---|---|
| Secondary layer | triangle, -13 dB; 0 st; 5 cents |
| Filter 1 | Notch, 772 Hz base, -0.8 octave depth |
| Filter contour | 0.02 / 1.1 / 0.2 / 0.3 |
| Filter 2 | Lowpass at 6.80 kHz |
| Pulse width | Not used |
Sound after step 2 — WebM/Opus
Next: Round the signal with saturation →Step 03 / 5
Round the signal with saturation #
Set giStage init 3 in the full script. Use +2.4 dB drive with 70% shaped signal and -3 dB output trim. The smooth curve is tanh(input × 3 × ampdb(drive)) / 3; the other portion stays clean. Louder peaks enter its curved region more strongly, adding harmonics and rounding the envelope. This interprets the source saturation curve rather than implementing its proprietary transfer function.
| Csound control | Value |
|---|---|
| Drive / output | 2.4 / -3 dB |
| Shaped mix | 70% |
| Input DC removal | Order 2048 |
Sound after step 3 — WebM/Opus
Next: Add filtered stereo echoes →Step 04 / 5
Add filtered stereo echoes #
Set giStage init 4 in the full script. Use 267 ms left and 267 ms right, 24% feedback and 18% wet. Inject the centered input on the left and cross-feed the filtered echoes so later repeats alternate sides. Filter each echo around 1800 Hz, with a 3.5-octave interval between highpass and lowpass cutoffs. Feedback values are updated once per 16-sample block, adding that small delay to the recirculating path. Held notes leave older colors behind the current tone.
| Csound control | Value |
|---|---|
| Times | 267 / 267 ms |
| Feedback / wet | 24% / 18% |
| Routing | Crossed, left input |
| Filter center / width | 1800 Hz / 3.5 octaves |
| Time LFO | 0 Hz, ±0% |
Sound after step 4 — WebM/Opus
Next: Set the final width and gain →Step 05 / 5
Set the final width and gain #
Set giStage init 5 in the full script. Use 5.27 dB gain after the effects. Scale the side signal by 1, preserving the center. Apply order-2048 DC removal, using a longer window to limit low-note loss. The fixed output-scale control at the end of instrument 90 remains a separate Csound gain setting. Clear the shared note buses only after processing them; effect memories continue carrying their own tails.
| Csound control | Value |
|---|---|
| Final gain | 5.27 dB |
| Width | 1 × side |
| Bass side cutoff | Off |
| DC removal | Order 2048 |
Sound after step 5 — WebM/Opus
Audio & settings ↓