TOPIC:

On the "Save" tab, select "IR Samples" and press "All" to see all the IR's overlayed.

Thanks. I see the colour of the average IR is different. Is there a way of making the ref measurement/IR stand out from the rest?

All I'm trying to accomplish is perform multiple measurements in REW, average them out in Averager while maintaining the time integrity between impulses/measurement in the process, and then exporting the averaged magnitude and phase response as an impulse wav file to either FIR Designer or back to REW.

I see it here too when I load your measurements annotated with 48000 in the filename. That just means that either a) REW has a different peak reference for some of the measurements, compared to others, or b) the soundcard had different in/out latency for different measurements.

I don't know what more I can do to help and I've already spent a lot of time guiding you through this. I'd suggest digging a little deeper into understanding soundcard drivers, how REW works and how to lock the time reference in REW when doing multiple measurements.

Pro measurement tools like SysTune (AFMG) and SMAART (Rational Acoustics) use ASIO drivers and can lock the time reference. The workflow is to take a measurement in these tools, adjust the IR to peak, then lock the IR for subsequent measurements.

Also, soundcard drivers other than ASIO often don't have repeatably in/out latency. ASIO drivers are common in pro audio because (amongst other things) they have repeatable in/out latency.

musan wrote:

All I'm trying to accomplish is perform multiple measurements in REW, average them out in Averager while maintaining the time integrity between impulses/measurement in the process, and then exporting the averaged magnitude and phase response as an impulse wav file to either FIR Designer or back to REW.

I understand. But from the exported measurements you have provided, REW isn't maintaining the time relationship between the impulse responses of the measurements. As I previously mentioned, ensure you are using ASIO drivers (if on windows) and then confirm that you can take and export correct time relative measurements from REW. Once that is working, then Averager can do what you want it to do.

BTW there are some good WAV viewing tools like Sound Forge.

I appreciate your help.

I use the Steinberg UR22 MkII USB audio interface in ASIO format (latest driver) exactly for the reasons you mentioned. The measurements are carried out with a loopback cable connected to one of the outputs back into the card audio input for reference and clock adjustment which REW automatically does. The main reason for considering Averager is because REW does not have the averaging modes available currently.

The measurement delays will vary for the test scenarios that I usually perform, since I move the microphone within per-determined locations during these tests. The latency however has remained fairly consistent from what I have observed. I have had SoundEasy and Clio for over 15 years now but I wanted tools with a faster workflow for this type of work. Smaart is on my radar too.

I guess I will have to resort to manually aligning the IRs which will work, it's just a tedious process and something I wanted to avoid.
Thanks.

musan wrote:

It's samples, as described on the link I provided above.
eclipseaudio.com/loudspeaker-measurements/

Since REW is putting the peak at or near sample 48000, simply put that number at the end of all the WAV file names. E.g. "Measurement 1 - 48000.wav", "Measurement 2 - 48000.wav" etc..

I had already done that and I it's what confused me when the IRs did not line up like in REW. Please see the screenshot below which is for IRs imported with the second method where IRs files are appended with 48000.

We've implemented importing of REW txt IR files, using the metadata provided in the files, and I think the result is what you're after. This import method will be available in the next update to Averager.

Thanks Michael,

That is exactly, what I was looking for. Hope we can get the update soon

Hi Michael,
I noticed that when using the "Time Align all to Reference" feature, some of the impulse responses will shift such that they start ahead of the ref impulse, while some are behind it. What does this feature look at or use when aligning the impulse responses?

The time alignment uses cross-correlation. The alignment results will vary depending on how clean and/or similar the measurements are. The alignment can always be checked and adjusted on the 2nd tab by comparing each measurement with the reference - either using the mag & phase plot or the IR plot.

Thanks for the feedback.

Would consider having an option to turn off the graphing of the averaged response on the Save Tab?

I think it would be helpful to be able to view the correlation between all the magnitude, phase and especially impulse responses, without having the averaged response overlaid on top of everything.

musan wrote:

Thanks for the feedback.

Would consider having an option to turn off the graphing of the averaged response on the Save Tab?

I think it would be helpful to be able to view the correlation between all the magnitude, phase and especially impulse responses, without having the averaged response overlaid on top of everything.

See the latest update.

eclipseaudio wrote:

musan wrote:

Thanks for the feedback.

Would consider having an option to turn off the graphing of the averaged response on the Save Tab?

I think it would be helpful to be able to view the correlation between all the magnitude, phase and especially impulse responses, without having the averaged response overlaid on top of everything.

See the latest update.

Yes I saw it. Thank you

To maintain relative time alignment in IRs exported as WAV from REW use windows, as explained in the help .

Export all impulse responses as WAV

Export the impulse responses for all current measurements in WAV format. The exported file will have a minimum duration of 1 second. The dialog provides options to choose whether to export the measured IR, the IR after any EQ filters have been applied to it or a minimum phase version of the IR. The response can optionally be normalised so that the peak value is unity (0 dBFS). There is also an option to apply the current impulse response window settings to the response before exporting it. The sample format can be chosen as 16, 24 or 32-bit signed PCM or 32-bit Float. Float is recommended if the application using the data can accept this, particularly if the response is not normalised. The peak of the exported response occurs 1 second after the start unless the IR window has been applied, in which case the export starts at the first windowed sample and the peak location depends on the IR window settings. That can be used to preserve relative timing for exported impulse responses that have been captured with a timing reference: define IR windows for the responses that have the same window ref time and the same left window width - use rectangular windows if the data should not be altered by the window. The exported IR for each impulse will then start at the start of the left window, which will be at the same absolute time for each, so their relative timing will be preserved.

I'm not familiar with Averager, but you can generate vector averages directly in REW using the control on the All SPL graph.

JohnPM wrote:

To maintain relative time alignment in IRs exported as WAV from REW use windows, as explained in the help .

Export all impulse responses as WAV

Export the impulse responses for all current measurements in WAV format. The exported file will have a minimum duration of 1 second. The dialog provides options to choose whether to export the measured IR, the IR after any EQ filters have been applied to it or a minimum phase version of the IR. The response can optionally be normalised so that the peak value is unity (0 dBFS). There is also an option to apply the current impulse response window settings to the response before exporting it. The sample format can be chosen as 16, 24 or 32-bit signed PCM or 32-bit Float. Float is recommended if the application using the data can accept this, particularly if the response is not normalised. The peak of the exported response occurs 1 second after the start unless the IR window has been applied, in which case the export starts at the first windowed sample and the peak location depends on the IR window settings. That can be used to preserve relative timing for exported impulse responses that have been captured with a timing reference: define IR windows for the responses that have the same window ref time and the same left window width - use rectangular windows if the data should not be altered by the window. The exported IR for each impulse will then start at the start of the left window, which will be at the same absolute time for each, so their relative timing will be preserved.

I'm not familiar with Averager, but you can generate vector averages directly in REW using the control on the All SPL graph.

I played around with using IR windowing as per the help manual, the alignment was most improved by using "Apply Windows To All" option but it was still not 100 percent. I could only get perfect preservation of alignment when exporting all measurements as text or exporting impulse responses as text files. Please note that I am talking from a perspective of viewing these files from within Averager. It could very well be that I am not using IR windowing feature correctly, am not sure.

I need to read up a bit more on Vector averaging because I find that when I use it in my spatial measurements, there is significant drop of the midrange to high frequency level. Sometimes the level drop is upward of 15dB when compared to power averaging; the level drop to me at least, does not seem to correspond with what I hearing. Averager provides a bit more in terms of averaging options which is what I am currently evaluating and I think the Averager and REW are a match made in heaven for some design scenarios

If you are making measurements at different positions with a timing reference the results will preserve the timing differences resulting from different distances to the source. That is not what is needed for spatial averaging. For spatial averages use the "Average the responses" button in the bottom left of the All SPL graph to produce a power average. If you wish to vector average responses from different positions either use the "Estimate IR delay" feature on each measurement to remove measurement delays first or make the measurements without a timing reference.