Topic: matrix mix question...

[nicegrin]

Hi!

A question about matrix mix downs.  If you record using 2 sets of stereo pairs (A,B) and (C,D) and the pairs are mounted with the same distance (think of it as A and C being adjacent then spaced with 17 cm where mic B and D are placed adjacent to each other).

When you make a mix down in post from the 4 tracks to a 2 track matrix mix there are then 2 possibilities:

1,  Channel A and C mixed down to the same channel (and B and D to the other channel) or

2, Channel A and D mixed down to the same channel (and B and C to the other)

I´d like to know what difference these two possibilites result in and if it´s always preferable to use one before the other.

Any other suggestions for matrix mixing is highly welcomed aswell.

Thanks
N






 

[page]

Hi!

A question about matrix mix downs.  If you record using 2 sets of stereo pairs (A,B) and (C,D) and the pairs are mounted with the same distance (think of it as A and C being adjacent then spaced with 17 cm where mic B and D are placed adjacent to each other).

When you make a mix down in post from the 4 tracks to a 2 track matrix mix there are then 2 possibilities:

1,  Channel A and C mixed down to the same channel (and B and D to the other channel) or

2, Channel A and D mixed down to the same channel (and B and C to the other)

I´d like to know what difference these two possibilites result in and if it´s always preferable to use one before the other.

The real short is; it's preferable, but not required, to keep your Left channels (assuming that's A and C) together and your Right channels together. It will reduce (but not eliminate) image smear and phase cancelation. I find that a little smear/cancelation is ok (even pleasant if used properly), a lot of smear/cancelation is bad, ymmv.

[Gutbucket]

Most of the time doing what Page suggests in keeping the left mics mixed to the left stereo channel and right mics to the right stereo channel is the best idea, but there are situation where mixing feeds from opposite sides together makes sense.

If interested in delving into the technicalities of when and why, you might check out this thread: http://taperssection.com/index.php?topic=110861.0
The relevant portion there to this discussion is the idea behind Twisted-Quad mixing.  The Michael Williams paper linked in my first post there titled The Whys and Wherefores of Acoustic Cross-talk in Multichannel Microphone Array Design, explains what it is and why it may be useful. 

That idea may be extendable to mixing multiple forward facing stereo pairs as well, but the signals in the channels would probably need to be somewhat different from each other to make that practicable.. I'm thinking a situation like omnis plus directional mics, where mixing the omni channels containing mostly diffuse ambient information to opposite sides may help to 'decorelate' their information from that of the directional mics.  In that case, mixing each omni with the directional mic on the opposite side effectively puts more space between the mics mixed together, possibly reducing potential combfiltering of the direct sound that could result from mixing together mics that are close together on the mic bar but not exactly coincident.  The increased decorelation between omnis and directional mics for the ambient sounds may also be helpful in translating a wide, diffuse ambience.

If none of that makes any sense, don't worry about it.  Just try mixing the left mics together and the right mics together in various amounts to see if it sounds better than a single pair alone.