Theo,

I've never used a C 34, but I'm a little surprised that its figure-8
setting would be so much more sensitive than its cardioid setting. You
may have some other kind of problem there; I'd suggest that you look
for an explanation of the discrepancy.

Anyway, when matrixing from an M/S recording to left/right stereo, you
should definitely have some way of adjusting the levels of the "M" and
"S" signals that you're feeding into the matrix. First of all these "M"
and "S" signal levels must suit the matrix--neither overloading its
circuitry nor being so low in level that the result is noisy.

But yes, as long as the signal voltages aren't too high or too low for
your matrix circuitry, you absolutely can adjust the two signals
relative to each other. That's one of the main advantage of M/S
recording: the amount of "S" relative to the amount of "M" is a direct
control over the apparent stereo width of the recording, and you don't
have to decide this setting while you record (where you may not have
good monitor speakers available)--you can set it afterwards, when you
matrix to L/R stereo.

One way to do this is to start by feeding only a small amount of "S"
signal into the matrix, then gradually bringing it up in level. If you
do this, you can feel the image widen from mono to a reasonable stereo
spread--and then beyond, if you push the "S" level too far (things will
start to get "phasey," particularly when sound sources at either edge
of the sound stage are prominent).

Usually I find that there isn't very much doubt (on any given playback
system, at least) about where the M vs. S levels should be set for any
given recording, since the amount of hall reverberation generally
increases as the stereo width increases, and there is a certain blend
of spaciousness and clarity that feels persuasive, while anything else
seems too far in one direction or the other.

But please don't feel that you have to maintain the relative
sensitivities of the two halves of your microphone; that's just a
physical parameter of your microphone, and in this case, I would even
consider it a little suspicious as I said before. Much more often the
"S" microphone is a little less sensitive than the "M" microphone, and
in addition, picks up less acoustic energy because it receives so
little direct sound.

In any event the capability of varying "M" vs. "S" gain at the matrix
inputs is one of the keys to making good use of any M/S recording.
--best regards