PASP DSP Application Notes

Research date: 2026-06-22

Primary reference requested by Shane:

• Julius O. Smith, Physical Audio Signal Processing: https://ccrma.stanford.edu/~jos/pasp/pasp.html

Central knowledge-bank version:

• docs/DSP_Knowledge_Bank.md

Access note:

• The PASP top page was not directly accessible through the browsing tool during this pass, but related CCRMA/PASP chapter pages and Smith's digital waveguide index were available and are consistent with the known PASP body of work.

Useful source links:

• Digital Waveguide Models: https://ccrma.stanford.edu/~jos/pasp/Digital_Waveguide_Models.html
• Frequency-Dependent Damping: https://ccrma.stanford.edu/~jos/pasp/Frequency_Dependent_Damping.html
• Nonlinear Elements: https://ccrma.stanford.edu/~jos/pasp/Nonlinear_Elements.html
• Digital Waveguide Synthesis index: https://ccrma.stanford.edu/~jos/wg.html

Core Insight

The strongest SpaceAge takeaway is not "build academic simulations." It is this:

Use efficient physical illusions: exciters, delay lines, damping filters, resonator banks, and carefully controlled nonlinearities.

That aligns beautifully with SpaceAge. We want instruments that feel alive, respond musically, and run efficiently inside a focused microDAW.

General DSP Principles To Apply

Exciter Into Body

Many physical models can be thought of as:

exciter -> resonant body -> damping/tone shaping -> output

The exciter is the thing that starts sound: a pick, mallet, air burst, membrane strike, stick hit, noise burst, or impulse. The body is what rings: string, bar, tube, membrane, plate, shell, or room.

SpaceAge should expose this in human language:

• Strike
• Pluck
• Pick Position
• Body
• Material
• Size
• Tension
• Damping
• Brightness

Avoid exposing low-level DSP names unless the user is in a clearly advanced view.

Frequency-Dependent Damping

Real instruments rarely decay evenly across the spectrum. High frequencies often decay faster than lows. PASP-style models use damping filters inside feedback loops or resonator paths to mimic that behavior.

Applications:

• Plucked strings should lose brightness as they decay.
• Kicks should keep low-end longer than click.
• Bells should let upper partials decay at different rates.
• Reverb tails should darken naturally instead of sounding like static white mist.

Stable Feedback Comes First

Feedback loops are powerful but dangerous. Delay-line instruments, resonators, shimmer, reverb, comb filters, and nonlinear synth behaviors all need stable gain limits.

Application rule:

• Any nonlinear element inside feedback should be conservative, filtered, and preferably oversampled if it creates bright harmonics.

Modal Resonators

Modal synthesis uses banks of tuned resonators to model objects that ring at several partial frequencies. This is excellent for bells, plates, drums, bars, tubes, cymbals, and strange space-age percussion.

SpaceAge should use modal resonator banks where full physical modeling would be too expensive or unnecessary.

SpaceAge Application Map

Physical Model

Current direction should become a true exciter-plus-body instrument.

Priority additions:

• Fractional-delay pluck voice.
• Loop damping filter.
• Pick/strike position.
• Material macro.
• Body resonator amount.
• Tuned A440 behavior for pitched percussion.

Potential factory sounds:

• Modal Marimba
• Rubber String Bass
• Orbit Kalimba
• Glass Tube
• Dust Bell
• Space Harp

Kick Lab

PASP-informed improvement:

• Use a layered model: click/transient exciter, sine/sub body, knock resonator, shell damping.
• Add one or more tunable resonators for body tone.
• Add frequency-dependent decay so click fades fast but body remains controlled.

Potential controls:

• Beater
• Skin
• Shell
• Knock
• Sub Hold
• Damping
• Body Tune

Snare Lab

PASP-informed improvement:

• Combine a membrane/body resonator bank with filtered noise.
• Let "snare wire" be a noisy exciter/resonator layer, not just static noise.

Potential controls:

• Shell
• Wire
• Snap
• Ring
• Damping
• Body Size

Hat Lab

PASP-informed improvement:

• Use inharmonic resonator clusters plus filtered noise.
• Let Flux Nodes vary strike, damping, brightness, and cluster spread per note.

Potential controls:

• Metal Spread
• Stick
• Choke
• Air
• Damping
• Flux Amount

Redshift

Redshift can borrow physical-model flavor without becoming a physical synth:

• Body Comb can be improved with safer damping and tone compensation.
• Metal Shimmer can become a small resonator cluster rather than a vague brightness control.
• Drift and X-Mod should be alias-aware if pushed into harsher territory.

Halostar

Reverb and waveguides are related. Halostar can improve by borrowing:

• Delay networks with damping.
• Allpass diffusion.
• Stable feedback limits.
• Tone-dependent decay.

Future controls:

• Tail
• Diffusion
• Damping
• Orbit
• Shimmer
• Width

Future Modal Objects Instrument

This may deserve its own internal synth engine:

Modal Objects could be a lightweight resonator-bank instrument for bells, plates, tubes, bowls, ceramic hits, glass hits, struck strings, and hybrid percussion.

This would support the SpaceAge identity strongly: elegant, strange, musical, not generic.

Highest-Value Build Order

1. Upgrade Kick Lab with resonator/body modeling.
2. Add a dedicated Karplus/waveguide pluck mode to Physical Model.
3. Create a small modal resonator bank shared by Physical Model, Drum Labs, and possibly Halostar.
4. Improve Halostar damping/diffusion.
5. Build a SpaceAge factory preset pack around these features.

Guiding Principle

PASP gives us serious DSP foundations, but SpaceAge should translate those foundations into musician-facing controls. We are not selling equations. We are selling instruments that feel alive.