DRC/Inguz and Microphone Calibration

[muski]

I recently had my Behringer ECM8000 microphone calibrated by Cross-Spectrum Labs with some interesting results (cost $50, http://www.cross-spectrum.com/measur...ike_meas.html). A very different calibration file was the result. See the attached graph -- green is the stock ecm8000.txt file that comes with DRC, red is the calibration file for my ECM8000 with a Tascam US-122L preamp/USB interface. The calibrator's experience with about a dozen ECM8000s is that though they show pretty good response (esp for the price), but they are all over the place.

Haven't done a ton of listening with this, but immediately had to switch from 'normal' to 'soft' -- it was just too bright and there was some occasional weird digital distortion (maybe it was clipping). Further experimentation with different target curves now required. [Edit: disregard -- I think I recorded the sweeps incorrectly on my first try...]

Curious if others have had their mics calibrated and if the results were as surprising.

[pfarrell]

The calibrator's experience with about a dozen ECM8000s is that though they show pretty good response (esp for the price), but they are all over the place.

I have not, but I'm not surprised about his comment, Behringer are on the low end of pro-audio stuff. It would be interesting to try some more serious measurement mic, say an Earthworks.

[muski]

That would be an interesting test. Makes you wonder how close a $50 Behringer mic + $50 calibration fee gets you to a $600 Earthworks M30 calibrated mic...

Another slippery slope (or $lippery $lope)... maybe one day.

muski

[pfarrell]

That would be an interesting test. Makes you wonder how close a $50 Behringer mic + $50 calibration fee gets you to a $600 Earthworks M30 calibrated mic...

Another slippery slope (or $lippery $lope)... maybe one day.

You bet.

My guess (only a guess) is that for occasional usage, the Behringer would be fine. I'd expect that the Earthworks would work over time, like a Fluke meter. But even Fluke meters are supposed to be calibrated periodically.

[krzys]

Muski did you get a *.txt file from the calibrator? Is that usaual with calibrators ?
Chris

[vdorta]

Haven't done a ton of listening with this, but immediately had to switch from 'normal' to 'soft' -- it was just too bright and there was some occasional weird digital distortion (maybe it was clipping)

Wouldn't this suggest the calibration is wrong? In my system, with the stock ecm8000.txt file, the improvements brought by the Inguz "normal" correction over the standard flac (on both music and Stereophile CD1 test tones) are what I would expect: flattening of a 80Hz-120Hz bump and a 60Hz trough, no difference in the rest of the frequency range which is very flat to begin with.

[muski]

Muski did you get a *.txt file from the calibrator? Is that usaual with calibrators ?
Chris

Yes, it was included with the service. I think most calibrators do that. For instance, on the Earthworks audio site it says:

"Calibration files are available for every Earthworks measurement microphone at an extra charge. These files can be used to import the microphone’s amplitude frequency response directly into your measurement system. They are in text format, with frequency in Hz in the first column, and response values in dB in the second."

[muski]

Wouldn't this suggest the calibration is wrong?

No, I think it was 'operator error' on my part -- I don't think that I generated my first set of filters correctly. I have since regenerated everything, and am not hearing the clipping I noticed the first time with the normal filter. The sounds is a little brighter, but that may simply be that I was so accustomed to the old ones. I should do some more side-by-side listening to the calibrated and uncalibrated filters.

Turns out that when I plot the 'normal' and 'soft' filters there is almost no difference between the two in terms of freq response correction. My understanding is that the difference between 'soft' and 'normal' has more to do with the degree of phase correction. I also just read that the 'No Correction' filter just does freq filtering, with no phase correction, so I will try listening to that, too.

I guess it's also possible that you might have a better instance of an ECM 8000 than mine. As I said before, I would be interested to try the same exercise with a calibrated Earthworks mic.

BTW, are you using the 'normal.drc' from the Inguz site or the one included with DRC?

Muski

P.S. Just curious -- you say 'what I would expect'. Is that based on published measurements of your equipment or something else?

[vdorta]

Muski, "what I would expect" means that I am comparing (by ear) the original uncorrected sound of my system (flac files), that I know by experience to have a lumpy bass due to my particular room, but that is also pretty flat on midrange and highs, to the Inguz-corrected sound (flac/InguzDSP files). I use both test tones and music I know well.

You have to compare:

1) the uncorrected sound (as above) to the standard ecm8000.txt-corrected sound;

2) the uncorrected sound (as above) to the calibrated sound.

If sounds in 1) are similar on midrange and highs and the brightness only appears in 2), then the problem has to be the calibration, because unless your room is extremely dead or there is a problem with the tweeters, the Watts are pretty flat above the upper bass.

I am using the normal.drc from the Inguz site.

[muski]

Thanks. My room is very bright (polished conrete floors and glass and plaster), so your observation about brightness makes a lot of sense. Will give it a try this week.

Thanks,
Muski