> With my 744T I use a Y-cable. I set the gain on the first mono channel and
> the gain on the second mono channel about -20dB. That way, if a bird sits on
> the end of my mic and overloads the first channel, I hope to have enough
> headroom on the second channel. Said differently, I send the mono signal to
> two mono files at different gain levels. To the best of my knowledge, there
> is no built in circuitry for this, but the Y-cable accomplishes the same
There is no built in circuitry because it is regarded as not needed. Unless
you have a poor digitiser, 16 bits gives you a much larger dynamic range
than you get with natural sounds. 16 bits gives you 15 level bits plus
parity. the lower two or three bits have dither added to minimise
quantisation noise, leaving 12 bits free and quiet for recording. That's a
range of 72 dB which spans all but the loudest natural noises down to a
natural background level.
Recording at 24 bits single track takes up less digital space and it gives
you another 8 bits, 48 dB, on the dynamic range. You can't use this in
practice of course, because the analogue input and output is nowhere near
clean enough to need it.
Do a quick experiment. On playback, bring up the low level track to be the
same as the high level and see if you can hear any difference between
tracks. Swap tracks to randomise the test. I've checked out my cheap digital
recorders and I can get away with recording at 20dB below peak as standard,
which takes care of almost all unexpected peaks except close thunder, which
actually sounds better with a bit of distortion.
Being able to under-record is one of the joys of digital recording. When we
went digital in the BBC Film Unit, the standard peak lineup level was
changed to 8dBs lower than for tape - 14dB headroom instead of 6dB.