Which RMS meters to trust: Ozone/Span

[Wilkesin]

When doing "mastering/pre-mastering" or whatever you want to call it I've always used Voxengo span behind my limiter and used the RMS readout to judge the overall loudness. I usually run things to roughly -13 dB RMS which I dont consider very heavy limiting at all compared to modern discs.

I've been getting deeper into Ozone lately though and decided to see if maybe I didnt need Span so I switched the Ozone meters over to their RMS readout and shockingly they indicated that my track was sitting closer to -7 RMS.

Can anyone explain the differences? Which one is right?

[Kihoalu]

.

Part of the difference may be the way that 0db full scale is defined. If 0db RMS is for a full scale sine wave the power (at 0db) will be much lower than if 0db RMS is defined for a full scale Square Wave. (A full scale square wave being the maximum power attainable).

RMS of a sinewave = peak * 1/SQRT(2), RMS of DC (or square wave - which is just alternating DC) = peak so the difference is 0.707 or a power difference of 3db. Of course this does not explain your entire measurement dis-agreement but may be part of it.

The other thing about RMS loudness measurements over a long interval of time (the length of a song) is that they are not the total integrated power (energy) of the song but are usually based on obtaining maximum or minimum measurements over some short interval of time (such as 200-300 milliseconds) in order to approximate the way a human would interpret the loudness. This measurement interval can make a substantial difference in the final numbers. These days I use 300mS interval and Full Scale Sine wave as my measurement references and I usually have about 12-13 db of Peak to RMS ratio. You can check you SFW manuals/settings to see if these intervals and reference match.

.

[Staccato]

Originally Posted by Kihoalu

.

Part of the difference may be the way that 0db full scale is defined. If 0db RMS is for a full scale sine wave the power (at 0db) will be much lower than if 0db RMS is defined for a full scale Square Wave. (A full scale square wave being the maximum power attainable).

RMS of a sinewave = peak * 1/SQRT(2), RMS of DC (or square wave - which is just alternating DC) = peak so the difference is 0.707 or a power difference of 3db. Of course this does not explain your entire measurement dis-agreement but may be part of it.

The other thing about RMS loudness measurements over a long interval of time (the length of a song) is that they are not the total integrated power (energy) of the song but are usually based on obtaining maximum or minimum measurements over some short interval of time (such as 200-300 milliseconds) in order to approximate the way a human would interpret the loudness. This measurement interval can make a substantial difference in the final numbers. These days I use 300mS interval and Full Scale Sine wave as my measurement references and I usually have about 12-13 db of Peak to RMS ratio. You can check you SFW manuals/settings to see if these intervals and reference match.

.

Any desired results may be obtained by adjusting the "RMS Window Size", this is the reason for standards.