Exporting correct IR delays

[twelti76]

Hi John, I know I asked you about this before. I want to export measurements so they have the correct delays, as measured using a loopthrough setup. You had suggested using the "IR Windows" and simply setting the Ref TIme to 0, and "Apply windows to all". This works within REW. I now see the reference time at zero in every case (not at the highest initial peak of each measurement as before). I can read off the correct delays, which match also the note that REW adds in the measurement notes. Example - in REW after apply all, with ref at 0:

I export this measurement as .wav file and import it again, now the reference time is set back to the first peak it seems.

When I exported, I applied the IR window thusly:

What can I do to export with the correct delay? Thanks as always for your help.

 

[John Mulcahy]

To preserve relative delays use this Analysis preference option when importing:

You can then assign t=0 at your preferred position for all the imports by using the set t=0 at cursor option on the Impulse overlay graph. Alternatively you can choose the export option to place t=0 at a sample index, set the index to 256 and use this import option:

 

[twelti76]

OK, I'm trying to put this all in order in my head. One thing, is there a way to get the sample number at the reference time? If I could get that, I could export using it and import using it? Second, if I just export, after setting reference time to 0 in all measurements like I already did, and say I import to Matlab, if i start at sample 1 and go 125 ms in to the IR, I should be at "0 time" reference position right? I tried this in Matlab but it is hard to tell since the IR delay I measured was very small to start with and some pre-ringing in the IR too.

 

[twelti76]

UPDATE: after careful review, it does look like the correct delay is there in Matlab. I did not know, or had forgotten about the import setting in "Analysis" section, so thanks!