LYNCH MASTER EQ — private 0.1 Small, deliberate mastering moves with four original EQ architectures and an independently enabled Tube or Transistor color stage. The ±4 dB band ranges, broad Tilt and aligned controls favor finishing a mix. Shift-drag gives fine adjustment. These are musical interpretations, not measured hardware replicas. VOICES Passive uses broad low shelving/high bell and separate low/high attenuation. Inductor uses broad, slightly shifted shelves and musical mid contours. Precision uses constant-Q mids and ordinary shelf/bell ends. Lynch uses parallel mid contributions for a different interaction of overlapping boosts/cuts. All four share your stored frequencies/gains when switching voices. Passive attenuation only affects Passive; its HF Bell switch is redundant. BANDS AND FOCUS Low and High have frequency/gain and shelf/bell choices where applicable. Both mids have frequency/gain/Q. Tilt tilts around 1 kHz using complementary broad shelves, up to ±2.5 dB overall contour. EQ On bypasses bands and Tilt together. Stereo / Mid / Side selects the band-EQ contribution. Mid affects the common center; Side affects the difference between channels. Side band-EQ has no effect on a mono input. Routing changes crossfade smoothly. The stereo high/low cuts, color stage and noise are global and independent of this band focus. COLOR AND CUTS Tube uses asymmetric valve-inspired saturation. Transistor uses Class A and transformer-inspired saturation. Color On is separate from EQ On; Drive and Color Amount control excitation and blend. Output is manual: no forced autogain. The high pass covers 20 Hz Open to 180 Hz; low pass covers 6–20 kHz. Each has its own power switch and a fixed musical Q. This is a finishing EQ, not the full-range creative filter in Lynch Filter or the general-purpose Lynch EQ. RESPONSE AND METERS The response display uses the same linear band/Tilt coefficients and topology as the audio. It shows the selected EQ path's curve, excluding color, cuts, noise, wet mix, channel focus and the small oversampling-FIR passband effect. It is not an analyzer. Input/output RMS, output sample peak and protection show actual audio. 8x oversampling at 44.1/48 kHz, 4x at 88.2/96/192; fixed 96-sample host latency. Init is flat with color/cuts/noise off. Mix 0 and Bypass are exact delayed dry. Sample Protect is optional sample protection, not a true-peak mastering limiter. VINTAGE NOISE Noise On and Noise Level are independent, from −96 to −30 dBFS nominal RMS. Tube/Transistor choose related hiss textures. Noise follows input: signal above −90 dBFS arms it, then it holds for six seconds and fades for two after silence. Cold silent instances produce no hiss. A conservative nine-second tail includes filter/alignment decay. Noise is synthesized original audio, not a sampled unit. PRESETS AND HELP Grouped starts are modest musical suggestions, including selected Mid/Side work. Init, arrows and Save/User manage settings. Every sound control supports host automation. ? enables contextual help; Guide is integrated and closes with Escape. Scale offers six sizes within the current screen. No hover-only parameter edits. Primary architecture context: https://pulsetechniques.com/mastering-eqs/eqm-1a/ https://pulsetechniques.com/mastering-eqs/meqm-5/ https://www.manley.com/s/MSMPX_manual_2018.pdf https://www.w3.org/TR/audio-eq-cookbook/ These inform operating concepts and coefficient design, not a claim to duplicate those manufacturers' circuitry. Private Apple Silicon AU. Actual Logic-session workflow and musical listening remain Robin's final artistic approval.