mono mixing

at first "mono mixing" seemed to me to be an option to have left and right channels superimposed so that both left and right channels output the same signal vs. time, in the case of a 2.0 sound output of course. I usually call such an option simply "mono" as it was printed in the past on sony walkman.
Actually when setting it on it doesn't at all do this, I couldn't hear any difference.

Looking at the code, what you describe is exactly what it does; for each frame it sums the value of the samples in that frame, divides by the channel count (i.e., the number of samples) and then sets each of the frame's samples to the resultant value.

A listening test here proves that.

Actually when setting it on it doesn't at all do this, I couldn't hear any difference.

I'm assuming that you mean "no difference between mono-mixing off or on", not "no difference between the speaker outputs with mono-mixing".

You didn't say what version of MOC are you using. In older (before 2.6) monomixing was only active when you enabled softmixing as well.

I'm assuming that you mean "no difference between mono-mixing off or on", not "no difference between the speaker outputs with mono-mixing".

yes indeed "no difference between mono-mixing off or on".

You didn't say what version of MOC are you using. In older (before 2.6) monomixing was only active when you enabled softmixing as well.

v. 2.5.1
right, I didn't enable the soft-mixer (w). Now I tried mono-mixing after enabling the soft-mixer and it works. There should be something somewhere in the software interface telling whether the soft-mixer is on or off. At least at the moment the user toggles it there should be an indication.

If you toggle the mixer to the softmixer ('x') then it does.

But that whole area is a little confused in MOC 2.5. It gets a little better in Version 2.6, but it still need a some more attention; behind-the-scenes discussions have been held.

However, it boils down to the best use of the time available for MOC development (not much, at present). This particular monomixer confusion has only arisen twice to my knowledge and is resolved in 2.6. Given that 2.5 is now stable and only accepting bug fixes, backporting the decoupling of softmixer and monomixing from 2.6 would not only be an unexpected change of behaviour in a stable version but a poor use of available time.