Notes · · Industrial / signal processing

GSRF vs EMA for mid-band oscillation

Restates locked Evidence numbers only. Not “always beats EMA.”

Engineers often ask: should I replace EMA with GSRF to reduce oscillation? The honest answer is band-dependent. GSRF Practical is a soft thermostat in log-space— a spring to a declared normal—not a linear low-pass.

What locked packs show

When EMA can still win on osc31

On the official frequency sweep, GSRF loses osc31 at very fast periods (~P=20) and very slow periods (~P=400–600) versus best EMA. If your “oscillation” is actually slow process drift you want tracked, EMA is often the better tool.

Why the soft thermostat helps mid-band ring

Pure-sine tests show GSRF AC gain plateaus near ~0.37 while EMA gain approaches 1.0 at long periods. Mid-band energy is clamped toward normal instead of fully followed. That is the intended identity.

Practical takeaway

  1. Use GSRF when the costly problem is mid-band ring / chatter around a known normal.
  2. Keep EMA (or k_return≈0) when the job is minimum lag tracking.
  3. Validate on your sampling rate: “period 80” is eighty samples—often ~80 minutes on 1‑min industrial traces.

Sources: Evidence · characterization v4 · probes v5 FULL PSD · excellence hunt v2 · phase6.2. Primary page: /evidence#frequency

Questions people actually ask

Does GSRF beat EMA on mid-band oscillation?

On locked phase6.2-style packs, ~+35% osc31 and ~−70% sweet-band spectral power vs best EMA are the public anchors — not a claim for every band or plant.

What period is the sweet spot?

Roughly periods 50–120 samples; peak osc advantage near P=80. Very fast (~P=20) and very slow (~P=400–600) can favor EMA on osc31.

Is spectral power the same as osc31?

No. osc31 is time-domain residual std after rolling mean 31; sweet-band spectral is frequency-domain power in the mid band. Both are reported.

Request Audit Pack Try snippet All notes