These are the results of the tests I did this afternoon. I ran white noise, 1 kHz & 100 Hz sine wavs into it and monitored with the coax output connected to my PC in WaveLab.
There were really no surprises at all.
The Recording Trim control works in .5 dB steps the one step between the whole numbers is .5 dB, but it does not show that on the display.
The channels track together very well. I assume that's because of digital controls, rather than analog pots. On my box all the inputs matched each other well within +/- 1 dB. Channels 3 & 4 were +/- .1 dB in the MIC (LOW) setting.
There is a 23.5 dB step between Line In & MIC In (LOW)
There is another 23.5 dB step between MIC In (LOW) & MIC In (HIGH)
The "Clip" indicator does show when you are <2 dB from clipping. I hope they can make that user selectable to show a real clip in a new firmware release. The meters on the DR-680 matched up very well with the readings I was getting in WaveLab. I also tested them against the meters on a Rockboxed iHP-120. Which I have learned to trust. They matched up very well.
The optional Limiter kicks in around 4 dB from 0, and acts like a brick wall. It just chops the top off of the WAVs. It wasn't pretty. Needless to say, I won't be using that function.
Knowing that the way you wired the TRS inputs on the MicroTrack II for unbalanced input was so important, I made two sets of cables. One set tied Pins 1 & 3 together at the input, Pin 2 hot. The other cables ran pin 3 to the ground on the unbalanced cable, Pin 2 hot and Pin 1 (Ground) not tied to the unbalanced ground, "floating".
As Mr. Chang had told me it really didn't matter which wiring to use, I confirmed that there was no difference in the noise when using either set of cables. < I would defer to someone that has the proper test equipment on this one though. I used my ears and the meters in WaveLab to make that judgement.
I'm testing batteries and external power now...
If anyone has any other ideas for tests, let me know.