This is the question we get more than any other, and the honest answer is “it depends on the room and on the iPhone.” The slightly longer answer is below.

tl;dr — On a calibrated iPhone, T20 and T30 are typically within ±5% of a Class-2 instrument in rooms with reverberation times between 0.3 and 3 seconds and ambient noise below ~35 dB SPL. For laboratory certification work you still want Class-1 gear; for studios, classrooms, offices, and small venues, AcousticCheck is good enough that we use it on our own consulting projects.

What we measure, and how

AcousticCheck implements reverberation time via the Schroeder backward integration method, which is the same algorithm used by laboratory-grade instruments. The math is identical; the difference is the microphone and the analog front end.

The internal iPhone microphone is, surprisingly, quite good — but its calibration varies between models. We maintain calibration tables for every iPhone from the 11 Pro forward, derived from controlled measurements against a Brüel & Kjær Type 4189 reference. Each calibration is part of the app bundle; the app picks the right curve based on your device.

When the numbers will be off

  • Very large rooms (T60 > 3 s) — the noise floor of the internal mic limits the dynamic range of the decay. You need an external mic or you risk a truncated decay.
  • Very small / dead rooms (T60 < 0.2 s) — the decay is too fast for the time resolution. You’ll see scatter between bands that isn’t real.
  • Loud HVAC or street noise — anywhere the ambient is above ~35 dB(A), the SNR of the decay drops below where Schroeder works cleanly.
  • Below ~80 Hz — the internal mic’s low-frequency response falls off. Third-octave bands at 63 Hz and lower carry larger uncertainty.

How to check whether you got a good measurement

Every measurement shows the R² of the decay fit per band. R² > 0.98 means the decay was clean. R² between 0.90 and 0.98 is still usable but worth a sanity check. Below 0.90 the band is flagged and we don’t report a T20/T30 for it — better honest gap than confident lie.

Better honest gap than confident lie. The R² check is there for a reason.

The Class-1 question

If you are submitting acoustic measurements as part of building certification — building code compliance, concert-hall acceptance, room-acoustic certification — the relevant standards (DIN 18041, ISO 3382-2) require a Class-1 instrument. AcousticCheck is not Class-1 and we don’t claim it is.

For the rest of the work — design iteration, before/after of a treatment, troubleshooting a room, documenting a studio refit — Class-1 accuracy is not the bottleneck. Repeatability and immediacy are. AcousticCheck wins on both.