Potential bug: Comb-filter during sequential measurement

y_brehm

Registered
Thread Starter
Joined
Apr 29, 2024
Messages
1
During a recent series of measurements I noticed a very strange comb filtering problem when using the sequential channel measurement feature. The result of this potential bug can be seen in the FR plot below:

REW-SequentialMeasurementBug.png

Every measurement except for the first one showed a comb filter like FR below ~ 4kHz.

In poorly treated rooms this could easily be mistaken for something caused by reflections and poor room acoustics. I only noticed this because the room was very well treated.
Given the perfection of the comb filtering, it is almost as if the measured signal is mixed with a modified and delayed version of itself. Looking at the impulse response of the broken measurement supports this theory.

The setup consisted of an RME Fireface UFX III connected to the test system via Madi. A Trinnov room optimiser was also used. This is why there is a huge delay in the data.
REW version used: 5.30.9
OS: macOS Monterey 12.6 on Apple M1 Pro

I tested this using the multi-input pro feature as well as the regular single input. This had no effect.

As the sequential measurement feature is potentially a huge time saver, it would be great to get some clarification on this.

Thanks in advance!

Yannik
 

Attachments

  • REW_sequential_measurement_bug_hunt.mdat
    3.3 MB · Views: 2
Top Bottom