What is RMS in audio world?

LUFS as defined in ITU-R BS.1770. Spotify’s figures from Spotify for Artists: Loudness normalization, checked October 2026.

What does RMS stand for?

RMS means root mean square. The name is the recipe, read backwards: square every value, take the mean of the squares, then take the square root of that mean.

Why not a plain average? A sound wave swings above and below zero, so the positive and negative halves cancel out and a simple mean comes out close to zero. Squaring makes every value positive, and the square root brings the result back to the original units. Both halves of the wave carry energy, and RMS counts both.

What is RMS value?

The RMS value of a signal is the steady level that would deliver the same power. That is why the term turns up in electrics as often as in audio. Mains voltage is quoted as RMS: 230 V in Europe is a sine wave that actually peaks at about 325 V, and 120 V in the US peaks at about 170 V.

For a pure sine wave, RMS is the peak multiplied by 0.707, or about 3 dB below it. Other waveforms give other ratios:

Music sits much further below its peaks than a sine wave, because drum hits and consonants are short spikes on top of a lower body of sound.

What is RMS in audio?

In a DAW, an RMS meter shows the average level of the signal over a short window, often around 300 ms, and keeps updating as the song plays. Both peak and RMS readings are in dBFS, decibels relative to full scale, where 0 dB is the highest level a digital file can hold.

One thing to check before you compare numbers: some meters follow the AES17 convention and add 3 dB, so a full-scale sine reads 0 dB RMS instead of -3 dB. Two meters can show different RMS values for the same file because of this setting, or because they use different window lengths.

Short peaks of a few milliseconds barely register as loud to the ear. What we hear as loudness is the sustained part of the sound, and that is what RMS measures. A track with high peaks and a thin body can read hot on a peak meter and still sound quieter than a denser track with lower peaks.

What is RMS in speakers?

On a speaker or an amplifier, RMS watts is the power it can handle or deliver continuously over a long period. Engineers will point out that “RMS power” is loose wording: it is average power calculated from RMS voltage and current. Everyone uses the term anyway, and on a spec sheet it is the number worth reading.

The math is the same as in audio. An amplifier rated 100 W RMS into 8 ohms puts out about 28 V RMS (power = voltage squared divided by resistance).

Peak or “max” power figures are higher and describe only short bursts, so they make a speaker look bigger than it is. Compare RMS to RMS. And remember that RMS watts don’t tell you how loud a speaker plays. Sensitivity, given in dB SPL at 1 W and 1 m, decides that. This is a different use of RMS from the meter on your master, so don’t mix up the two.

RMS vs peak level and crest factor

The distance between a signal’s peak and its RMS level is the crest factor. An unprocessed drum recording has peaks far above its RMS level. A heavily limited master has the two close together.

Crest factor is a good single number for how dynamic a mix is. High crest factor means sharp transients and room to breathe. Low crest factor means a dense, flattened signal. You can raise RMS by lowering crest factor with compression and limiting, but every dB you take off the peaks is punch you can’t get back.

RMS vs LUFS: which should you use?

RMS treats every frequency the same. Our ears don’t. LUFS (loudness units relative to full scale, from ITU-R BS.1770) runs the signal through a “K-weighting” filter first: deep sub-bass counts for less, and the range above about 2 kHz counts for a few dB more. LUFS meters also ignore near-silence when they measure a whole song, so a quiet intro doesn’t drag the number down.

The result is that two mixes with the same RMS can differ by several LU. A bass-heavy mix reads lower in LUFS than its RMS suggests, and a bright one reads higher.

LUFS meters give three readings: momentary (400 ms), short-term (3 seconds) and integrated (the whole song). Streaming services turn tracks up or down by measured loudness, and Spotify documents its measurement as ITU-R BS.1770, the LUFS standard. So for anything to do with release loudness, read LUFS. RMS is still handy while mixing, as a quick check on how dense a bus or instrument is.

What is a good RMS level?

There isn’t a correct number, and there isn’t a correct LUFS number either. The target is a master that sounds right for the song. Spotify plays music at -14 LUFS, but that is its playback level, not a goal for your master: a louder file is simply turned down. Most pop masters we deliver sit around -9 LUFS integrated, and released music ranges roughly from -5 to -20 LUFS depending on the genre and the arrangement.

The RMS reading for the same master depends on your meter and on how much low end the mix has, which is another reason to judge loudness in LUFS and by ear. We cover the details in how loud should my master be. If you’d rather leave that call to an engineer, our online mastering sets the level by genre and by references you send.

How do I check RMS levels?

Many DAWs can switch their track meters to RMS or include a metering plug-in that shows it. Reaper’s meters have an RMS mode, and Cubase’s SuperVision has level and loudness modules. Third-party options include iZotope Insight 2 and the free TT Dynamic Range Meter, which shows peak, RMS and a dynamic range score side by side.

Meters disagree a little because of window length and the AES17 setting. Pick one, learn how it reads, and compare against reference tracks on the same meter.

Want an engineer to set the loudness for your genre? Our mastering is $79 per song, and mixing and mastering starts at $133 per song. Both take 2–3 days per song with 6 revision rounds included.