XSCACE Studio

macOS, Windows, and as an in-room web app served directly from every Xylem amplifier — no separate install needed to reach the local dashboard.

Yes, XSCACE Studio is free for macOS and Windows.

Every XSCACE Xylem DSP amplifier. It also has a standalone mode that can measure and report on third-party systems via AirPlay.

No. It connects directly to the Xylem amplifier over your local network — no cloud, no round trips.

Eleven professional measurement instruments built into the app — X-Sense EQ, SPL Meter, Delay, Levels, Polarity, Frequency Response, Sub Align, Transfer Function, RT60, RTA and Mic Cal — SPL, delay/levels/polarity, frequency response, THD, sub alignment, transfer function, RT60, RTA, and branded PDF reporting.

Yes. The toolkit supports UMIK-style calibration file import (frequency and sensitivity) for fixed-output USB measurement microphones.

Audio calibration software measures how a speaker system actually performs in a room — SPL, frequency response, time alignment and reverberation — then uses that measured data to correct the DSP. XSCACE Studio does this with an 11-instrument toolkit built directly into the amplifier control app, so no separate measurement software is needed.

Yes. The Calibration Toolkit includes an impulse-triggered RT60 measurement (ISO 3382-aligned) with energy-time curve and spectral decay waterfall views, alongside a 31-band RTA with NC background-noise rating.

REW is free, general-purpose measurement software that works with any amplifier but writes nothing back to the DSP for you — you measure in REW, then manually re-enter corrections elsewhere. XSCACE Studio's toolkit measures AND writes the correction straight to the Xylem amplifier's DSP in the same app, with a closed-loop re-measurement to confirm the fix actually worked. It's a narrower tool — XSCACE amplifiers only — but it removes the export/re-enter step entirely.

Validated against REW, the reference tool in this field. Measuring the same system, a wired reference gave an inter-channel difference of +0.551 ms and XSCACE Studio measured +0.56 ms — agreement to 0.01 ms. Separately, the Xylem DSP's own delay block is accurate to 3 µs on a commanded 5.000 ms.

Accuracy is proven either way; precision over wireless is limited by the receiver, and the app is candid about it. A wireless receiver re-buffers between plays and applies its own timing corrections mid-stream, so every run measures its own worst time warp and warns you when it exceeds 0.5 ms, recommending a wired input for sub-millisecond work. It tells you when to distrust it rather than presenting a confident number it cannot stand behind.

Because a connection that reports success is not proof of sound. Compatibility testing found that licensed wireless receivers will accept a stream, report a healthy session, and silently discard the audio unless the sender meets the full specification. The ♪ test button beside the output picker exists so that no measurement is ever taken against silence.

A wired rig aligns speakers using a loopback, where one clock drives both playback and capture. Over a network there is no loopback and the receiver's clock is not yours. XSCACE Studio rides a train of quiet pilot bursts on a fixed carrier channel throughout the whole measurement — including underneath the sweep — and fits a line across them to reconstruct the receiver's real time map. Sound travels roughly 34 cm per millisecond, so a 1 ms error would put a speaker a third of a metre from where it actually is.

It refuses, rather than reporting a number it cannot defend. RT60 follows ISO 3382-1 practice — Lundeby truncation, a minimum 12 dB fit span and an r² ≥ 0.90 quality gate — and when a measurement does not meet them the failure message tells you which way to move: raise the level, or quieten the room.

Yes. Any measurement can export the raw microphone capture as a 16-bit WAV, so a disputed result can be dissected offline against any reference implementation you like. Several corrections during the app's development came from exactly that.

One number: threshold in dBFS. That is the figure printed on the driver's datasheet, so it is the one an installer actually has to hand. The 34-point gain transfer curve is generated on the device from it, with peak and RMS detectors running in parallel and whichever asks for more reduction winning. 33 factory presets across the range ship with thresholds derived from each speaker model's real power handling, and an amplifier lists the presets matching its own channel count.

It synthesises harmonics of fundamentals the cabinet cannot physically reproduce, so the ear infers a fundamental that is not being played. It is not a bass boost — nothing below the driver's range is being amplified, because there is nothing down there to amplify. That is why the control reads "extend down to X Hz": it sets how deep the effect reaches rather than naming a filter.