I've been experimenting with this a fair amount, and I think I finally have it down. I have to say, it is making a huge difference in my sound, and for the better. I really like this thing now that I've learned how to use it.

However, the manual sucks. At some point, I'm going to type out step by step instructions, but here are some general tips for now. Note that this is just what I do; other people with different components, rooms, tastes, etc. may find something else works better.

First, here's my setup: I'm taking a digital signal out of the optical Toslink output from my Squeezebox (essentially equivalent to a CD transport), and running it into the Behringer's optical input. I then take the digital Toslink optical out from the Behringer, and run it into the digital optical input of a Benchmark DAC1. I take the balanced XLR analog outs from the DAC1 and run it into my amp (Pathos Logos driving Magnepan 1.6qr speakers). So all the processing is done in the digital domain.

1) I plugged the ECM8000 microphone into the RTA input, and place the microphone right where my head would be in the normal listening spot. I measured the distances from the tip of the mic to each of the speakers to make sure the mic was equidistant. The mic was pointed straight ahead to the spot directly between the speakers.

2) This is crucial: I put the Behringer in dual mono mode, NOT stereo link; this lets you EQ each speaker independently. When I first got the thing, I didn't do this, and was getting some pretty crazy results. To do this, press the Utility button, scroll down to Channel Mode and change it to Dual Mono. Don't forget to press the B button to accept the change.

3) Now hit the I/O button. Make sure on page 1 (called "Select Input") that the input is set on "Main In." I think this is the only way to EQ each speaker independently. I initially thought this had to be set to Pink Noise, but in fact you'll get pink noise automatically when you go into Auto EQ mode.

4) Still in I/O mode, on Page 2 (called "Aux/Dig. Out"), select GEQ-PEQ. This will make sure the EQ'd sound is coming out of the Behringer.

5) On Page 3 ("Select RTA Input"), select RTA/Mic.

6) Now press the GEQ button. Flatten the curve, putting all frequencies on 0db. Make sure you do this for both channels, pressing the A button to switch between them.

7) Make sure you've got the proper input sensitivity and phantom power for the microphone you're using. If you're using the ECM8000, I think the default settings should have it covered.

8) Turn up your amp volume fairly loud, and warn your wife that a jet plane is about to land in the living room. Make sure nobody is walking around, or otherwise making noise.

9) Hit the RTA button, and cycle through so that the Auto EQ label shows up next to the B button. Adjust the Max and Range levels so that you'll get a good window on the frequency read out. You want something narrow enough so that you can easily see differences between the frequency levels, but wide enough so that you can see all the peaks.

10) Start the AutoEQ process. Note that you'll have to do this twice, once for each speaker. The pink noise should only be coming out of one speaker at a time. If for some reason noise is coming out of both, you need to change what you're doing.

I like to have "Room Correction" on, which "tilts" the target frequency spectrum from flat to slightly weighted in favor of the low freqs. Just sounds better to me.

I also didn't do any automatic correction below 100Hz. Instead, I manually adjusted them later, after the AutoEQ had flattened everything else.

I also put the noise level at about -1db, so as to avoid clipping. Keep an eye on the level meter and make sure there's no red flashing. If there is, lower the noise level; if there isn't, raise the noise level until there is, then back off slightly.

Note that during the AutoEQ process you can toggle between pages to compare the RTA readout with the adjustments the Behringer is automatically making to the GEQ. You can also switch between "Fast", "Med" and "Slow" to determine how fast it reacts. I like to start it at Fast, then change it to Med after a minute, then to Slow for a minute or so.

I set the Delta Max and Span to their maximum values, and didn't have any problems. Others may have to narrow these parameters, e.g. if you've got a big problem with your room or setup somehow.

At some point, after a few minutes, it should get to the point where the ongoing adjustments are fairly modest, and you can stop the AutoEQ then. Again, note that you have to do this twice, once for each speaker.

After you've run the AutoEQ, go back to the I/O page and change the input to Pink Noise. Noise should be coming out of BOTH speakers now. Then go back to the RTA, and look at the spectrum. You may notice -- as I did -- that it is no longer flat, now that you have both speakers going. I had a dip right around 12khz. I went back into GEQ, and adjusted the EQ (for each channel -- so you have to switch between Left and Right, making the same changes in each), then toggled back to the RTA, back and forth, until it was fairly flat.

Next, I manually adjusted the GEQ below 100Hz to get something fairly flat. (My speakers roll off below 40Hz, so naturally I didn't try to fix that range.) Note that again, you have to use the GEQ on both channels.

Finally, change the input back into your music input, and do some listening. Season to taste; I generally like a little more bass than you get out of a perfectly flat response.

That's it! Practice and experiment with the thing, and eventually it will have your setup sounding *very* sweet. The improvement is quite dramatic, in my case.

It's definitely a good way to spend a few hundred bucks. Compared to switching out interconnects, tubes, etc, where the differences are subtle at best, this is a REALLY noticeable difference. You can hit a button to bypass it all if you like, and compare it to the unprocessed signal. BIG difference.