so a rolloff is not like basically taking a frequency and lowering it, or taking a range of frequencies and lowering it?
What's the difference between running a 95hz rolloff, and taking the same recording and running an EQ over it lowering the 95hz the same volume as the roll-off would?
+t for being a fountain of information of things I don't understand at all...
Not sure were you are coming from in the first sentences above. In this context I view the terms "basscut", "bass rolloff" & "high pass" as the same thing. Batteryboxes typically implement the filter using a single capacitor on the signal output port. The capacitor prevents bass energy from escaping the batterybox while at the same time it passes higher frequencies. The transition from block to pass is gradual as in 6 dB per octave. So the highpass affects an entire band of frequencies: the band from 0 Hz and up to the filter knee (where the signal by def is only 3 dB down).
Activating the filter does not lighten the job (much) for the mics sitting at the front of the battery box.
Now, say the PA stack can't deliver truly deep bass. So the sound guys boost the 70 to 90 Hz range by 12 dB to give the impression of a solid low end. You're at the rear of the hall with omni mics and get the full boundary effect (ref PZM mics) from floor and sidewall. So your 4061 sees the 70 to 90 Hz band 24 dB (12 from boost + 12 from boundaries) stronger than the "natural" flat level (where the midrange and highs live).
You run without filter. You glance at your levels at think "wow - hot levels tonight". You turn the record dial down to accomodate the 24 dB bass hump without "over" lights coming on. What you might not keep in mind is that the mids and highs peak at -24 dB.
So you come home. You EQ out the excess 24 dB of bass and are left with a properly balanced recording with peaks at - 24 dB FS. I.e you are left with a recording equal to if you ran the batterybox with it's bass filter turned on AND the record levels such that peaks were at -24 dB FS. If this was a 16 bit system you would be making a 12 (11?) bit recording. You will probably "normalize" this to get the peaks at 0 dB FS but is't still and forever after, a 12 bits recording.
This was an example. Possibly exaggerated to show what the differences are and some of their consequences.
Questions to ponder are
1) would it be equally bad if it was the highs that were boosted 24 dB?
2) what did they do wrt to boost (EQ) in the analog world when recording onto LPs & cassettes.
Jon