pinknoise output level unstable

[suktan]

Pinknoise output level is not stable. It can be seen by using loopback mode though originally discovered this issue in no mode (no loopback) with output connected to an amplifier driving a loudspeaker.

Data logger for RMS_A(EU) is used to illustrate the problem.

Is it a bug?

[VirtinsTech]

The pink noise generated in the Signal Generator of Multi-Instrument spans 0Hz to half of the sampling rate. The pink noise (a.k.a 1/f noise) contains big-amplitude low-frequency components. If the integration time of the level measurement is not long enough, the measured level would not be very stable. Unlike the case of a "periodic" pink noise, the characteristics of a pink noise are only statistically stable in a long run. Measurements using pink noise or white noise require long-time averaging in order to achieve statistically stable results.