export impulse response: 192k flavour has a much lower gain

[giacomus]

Hello!Exporting impulse response to a wav file, and selecting the 48k and 192k sample rates, the two output filter files have very different gains:

48k version:

> sox FLX+2dB-trimmed-48k.wav -n stats
DC offset 0.000008
Min level -0.738194
Max level 0.337055
Pk lev dB -2.64
RMS lev dB -51.17
RMS Pk dB -40.70
RMS Tr dB -189.96
Crest factor 267.25
Flat factor 0.00
Pk count 2
Bit-depth 32/32
Num samples 131k
Length s 2.731
Scale max 1.000000
Window s 0.050

192k version:

> sox FLX+2dB-trimmed-192k.wav -n stats
DC offset 0.000002
Min level -0.184549
Max level 0.127396
Pk lev dB -14.68
RMS lev dB -63.28
RMS Pk dB -52.80
RMS Tr dB -204.69
Crest factor 269.10
Flat factor 0.00
Pk count 2
Bit-depth 30/32
Num samples 524k
Length s 2.731
Scale max 1.000000
Window s 0.050

There is a difference of more than 10dB

Is this expected? Is it an effect of the resample algorithm?

With regards
Giacomo

 

[John Mulcahy]

It should be 12 dB lower (factor of 1/4) to correspond to the increased sample rate and preserve energy. If you import each into REW you should see they have the same frequency response levels.