An error I have never seen before

[MikeVW]

Folks: I moved into a new home and am in the process of setting up my system. I received an error that I have never seen before. As systems go at the start, there is nothing obviously wrong with the sound, though the room is pretty lively at the moment. Any thoughts on what this means? The room is about 16 feet wide 12 feet tall the back wall is the kitchen - 40ish feet away. I also don't see a way to check/change the sample rate, but have never had this issue in the past. Thanks for any thoughts. Mike

 

[John Mulcahy]

Can you attach the mdat file?

 

[MikeVW]

Thanks for getting back to me. I should mention that I have subwoofer pairs along the front wall and the dipole speakers are about 6 feet out from the front wall.

 

[John Mulcahy]

That's a very odd looking response. The output device isn't selected in REW, so it isn't clear what you are using on the output path other than that you are driving both output channels. Care to elaborate on the output path? Best measuring each channel individually. It is also best to always select the devices being used and not leave the selections as default, and to check Audio Midi Setup to make sure the sample rates for the devices there match the rate in REW. Since you are using a UMIK-1 everything needs to be set to 48 kHz.

 

[MikeVW]

John, It has worked in the past, but I went in and selected the headphone jack as the output and 48 kHz. I get a similar error. This every odd. I have a system in a smaller room that I set up recently without issue. This room doesn't have excessive reverb, but the delay error makes me wonder if there is a reflection creating an issue. Attached is a new set-up screen showing the correct parameters - I think. Mike

 

[MikeVW]

I just checked my smaller system and now getting the same error. If you don't see an obvious setting problem, I am wondering if the mic has failed.

 

[Holmz]

Hey Mike,
Just record from the mic into someother sound grabbing app, and then play that out of the laptop or whatever… into some earbuds.
If the UMIK has failed, then it should sound dire.

 

[MikeVW]

John, I recorded a test using the mic with GarageBand. It recognized the mic, but didn't ask for a sample rate. I am assuming that it knows the correct rate. In any case, there was echo, dropouts and a garbled recording, so I am thinking it is the mic. I'll get a new mic and go from there. Apologies for troubling you. I was a software developer and did customer support for the product. Crazy the things that can go wrong.

 

[John Mulcahy]

Glad you identified something to address, you would need to be in an Olympic swimming pool for those echoes to be acoustic!

 

[MikeVW]

John, I got the new MIC and had the same issue. I uninstalled REW and then reinstalled it and that seems to have fixed the problem. Should have done that to start. Oh well!

 

[MikeVW]

John, It would seem that the problem is related to the acoustic timing signal. I don't get an error when there is no timing reference, but I do when I turn the acoustic timing reference on. Any thoughts? Thanks, Mike

 

[John Mulcahy]

Without a timing reference there is no clock rate check. You could attach an mdat made without a timing reference to see if anything jumps out.

 

[MikeVW]

John, Attached is a mdat file with both the timing reference on and off. There is a chip before and after the measurements, but the final chirp is a bit weaker than the first.

 

[John Mulcahy]

There is only one measurement in that file, which has the timing reference on.

 

[MikeVW]

John, attached are two mdat files and the associated screen shots for a run of the left speaker with - and without - the acoustic reference timing set. I still have a feeling that something in corrupted, so I went ahead and reinstalled the software a third time. This is a new vision for me. I haven't had any problems in the past. And this is a MAC. I don't have the dmg file from a prior release, but could be worth trying an older version if that is available.

 

[John Mulcahy]

The measurement without the timing reference has the same problem. It is a very bizarre looking impulse response and the delays in it are too large for any kind of domestic space. Saving it to WAV and listening to it confirms the very odd nature of the signal. A spectrogram shows the bands of the signal are arriving at different times and there is a skew in the response.

The strange nature of the response is what is confusing the timing tests. Something you can check is the Captured graph after making a measurement, it shows the actual input signal captured before processing. It should not have gaps/dropouts in it, for example. Perhaps attach a screen capture of that graph. If there is any kind of processing being applied to the mac's output that should be disabled, but I haven't come across that on a Mac before.

Earlier releases are available on the History page.

 

[MikeVW]

John, I found the older versions online and dropped back to 5.19. That works fine. I note only that the older version asks to specify what channel to use as the timing reference. Not sure where to go from here. Mike

 

[MikeVW]

John, Thanks again for the help. The spectrogram was helpful to see the issue. Something clearly got corrupted. I eventually installed the software local and that worked. I then then uninstalled it and reinstalled it as root. It was also time for an operating system update, so I did that as well. Thanks again for your patience. Mike