I agree with you regarding the issues related to response, as the system
(especially Type I) will magnify any response problems greatly (much
more than Dolby A) due to the companding employed. This was especially
an issue if the tape was recorded so hot the the high end became
saturated, leading to all kinds of bizarre sounds.
In regards to level, though, I'm not even sure how you would calibrate
it, as there is no electrical reference point (such as the Dolby cal
tone) to work to. There is only a setting for overall system I/O gain
(record level match), the VCA trim pot, and a bypass mode. (There other
internal adjustments as well on most of the systems, but little
information about what how they should be adjusted. "Proprietary
Information" I guess ;-) ).
As long as the system isn't driven so hard that you exceed the range of
the level detector, then it *should* track properly at any level. I have
in fact seen systems (especially those built into recorders like the
Tascam), where the I/O level was not set correctly, which will indeed
cause a mismatch as the detector reaches it's unity-gain point.
I have also run across non-linear behaviour of the 2:1 compander, where
the slope of the compander and/or expander don't match, which is
obviously a level-dependent phenomona. To further complicate the issue,
they may track fine with a steady-state signal, but not with a complex
waveform. This is usually due to a problem in the detector circuit.
However, if the units are working properly, and the companders and
detectors are properly calibrated, then the level shouldn't be an
issue.(I've run into similar issues with Dolby NR systems as well,
Bob Olhsson wrote:
>>Actually, the DBX process is not level sensitive, as it employs a direct
>>2:1 linear companding system. This was the big advantage of DBX over
>>Dolby, which is level (and response) sensitive (in all it's flavors).
>That's what they claimed. In practice it was very much level sensitive and far more response sensitive than Dolby A.