The standard way to quantify speed stability of analog transports, done the way AES6-2008, IEC 60386, IEEE Std-193, CCIR 409-2 and DIN 45507 define it. Feed a recording of a 3150 Hz test tone (3 kHz works too - the carrier is auto-detected between 2 and 5 kHz) and the meter performs FM demodulation: it mixes the signal to complex baseband around the carrier, low-passes the result, and reads the instantaneous frequency deviation. The deviation series is then perceptually weighted - the AES6 curve peaks at 4 Hz where the ear is most sensitive to pitch wobble, falling to 0.5 at 0.8 Hz, 0.508 at 20 Hz and about 0.03 at 0.2 Hz (anchors published in AES6; log-log interpolated between them, extrapolated to 200 Hz at the 4-20 Hz slope). Reported numbers: weighted RMS (WRMS, the JIS-style headline figure), weighted and unweighted 2-sigma peak (the AES method: the deviation exceeded only 5 percent of the time, so a single scratch does not fake the reading), and band breakdown per the classic definitions - drift below 0.5 Hz (unweighted), wow 0.5-6 Hz, flutter 6-100 Hz, scrape flutter above 100 Hz. For reference: the best studio machines measure around 0.02 percent weighted, a very good cassette deck around 0.08 percent, and typical consumer decks land at 0.1-0.3 percent. Program-material mode is offered for transfers where no test tone exists: it pitch-tracks sustained tonal content (200-2000 Hz) with autocorrelation and treats the relative pitch deviation as the flutter signal - clearly an estimate, since vibrato and intentional pitch expression count too; a test tone is always the trustworthy measurement. At least 5 seconds of audio is required so that 0.2 Hz modulation is representable; analysis covers the first 3 minutes of longer files.