John, how much data is saved with each REW measurement?

[tony22]

@John Mulcahy, hi. I was wondering if all of the "configuration data" associated with a given measurement is saved. I asked because I'm trying to figure out an oddity with a couple of measurements I took earlier this year, and for the life of me it almost looks like I used the wrong CAL file for my UMIK-1 for those measurements (90 deg instead of 0 deg) - at least that's the only way I can figure it without seeing all of the configuration data associated with those Sweep captures. Is that data available somehow?

 

[John Mulcahy]

If you look on the Info panel it will show any cal files used for the measurement. You can see the frequency response of the cal files on the SPL & phase graph. You can change the cal file using the Change cal button or action.

 

[tony22]

Sweeet! Thanks John. You da man!

 

[tony22]

If you look on the Info panel it will show any cal files used for the measurement. You can see the frequency response of the cal files on the SPL & phase graph. You can change the cal file using the Change cal button or action.

Right in front of me the whole time. But now that I've confirmed the cal files are the same, I'm even more at a loss. I'm attaching the MDAT file if you could look at it. Simple two channel measurement of my system taken literally five days apart. Same mic position, same speaker locations, same everything. Nothing changed from one to the other. Is there anything you may be able to discern to explain the differences in the upper frequency portion of the FR? The Filtered IR shows also a rather big difference, along with the RT60 Decay.

 

[John Mulcahy]

If you were really playing both speakers, as the L+R output setting suggests, then even extremely small differences in mic position can have a large effect at high frequencies. You also have an FDW for one measurement and not the other.

 

[tony22]

If you were really playing both speakers, as the L+R output setting suggests, then even extremely small differences in mic position can have a large effect at high frequencies. You also have an FDW for one measurement and not the other.

FDW?

 

[John Mulcahy]

Frequency Dependent Window, that's why the name shows [FDW] in the graph legend.

 

[tony22]

Ah, right. That capture was from a set of measurements I used to create convolution files for Minimstreamer. But that shouldn't affect what I'm seeing with these differences, should it? (at least for FR - yes I guess it will affect what I see in Filtered IR) As to the mic position, I use a laser measuring tool (very precise) to insure that for each measurement to mic height and distance from front wall (where a target is placed) does not vary by even 1/16" inch.

 

[John Mulcahy]

Nonetheless the measurements only differ meaningfully above around 7 kHz, once the windowing is set the same and smoothing is minimised. That is typical of a high frequency cancellation, either between the direct signals from the speakers or due to a reflection from some difference in the environment. It is not a good idea to measure both speakers at once.

 

[tony22]

Got it. Certainly for developing convolutions I do separate L and R channels, and for doing detailed acoustical measurements (I've worked with John Dykstra at GIK in the past). Once in a while though doing a "quick" two channel measurement is... well, quick.