Topic: Left is Right, Right is Left (omni version)

[MyNoise]

Just realized that at the shows I taped this week with an AT-853 omni mic clipped to each shoulder, I had the L mic on R shoulder and vice-versa.

So I've swapped the channels in Adobe Audition, and now the bass player sounds slightly right of center, which is correct.

But if you weren't there and don't know where the bass player was standing, is this step necessary?  In other words, besides giving an accurate impression of the room layout, is there some sonic waveform reason why the L-R channels of an omni mic recording shouldn't be swapped?  Conversely, does swapping stereo channels after the fact cause harm in some barely-detectable way?

[aaronji]

Time differences create stereo imaging with omnis, so, personally, I think it matters. If they are baffled (by your head, for example), that makes it even more important.

[rocksuitcase]

Conversely, does swapping stereo channels after the fact cause harm in some barely-detectable way?

I do not think this causes any noticeable artifacts

[Gutbucket]

Acoustically it won't matter either way - regardless of whether the recording was made with omnis, spaced or baffled, or directional micrphones, coincident or spaced.. or is a soundboard patch.  this is no different than flipping your head-phones around, or swapping the cables between the two speakers.

Psychoacoustically it can matter though.  It can of course if the listener was present at the performance and retains some memory of how things were arranged, or simply how it sounded.  Or if the listener was not there but there is some expected orientation,  such as the drummer always being off to one side and the organist on the other or whatever.  Orchestra recordings are like this. most often with the violin sections to the left and the cello and bass to the right.

Time differences create stereo imaging with omnis, so, personally, I think it matters. If they are baffled (by your head, for example), that makes it even more important.

Because the same time of arrival delay is retained if one swaps channels, it doesn't matter.  You are correct that time of arrival information is an important aspect imparted by any non-coincident stereo microphone arrangement - with any recording made with acoustically-significant spacing between stereo microphone pair, some time of arrival difference between channels will occur for any sound sources that are not centered, or more specifically, for any wavefront arrival that is not perpendicular to the medial plane. The [edit] potential confusion here is drawing a connection between what are two different issues.

[Gutbucket]

What also matters is if the recording is made up of more than one stereo pair of microphones or sources.  In that case all sources which are not mono need to be correctly assigned to Left or Right before mixing them together, but as long as that is done it won't matter acoustically if the entire resulting mix is swapped Left>Right.

[snip..] this is no different than flipping your head-phones around [..snip]

^I find this is a good check to do when assessing a recording or making a mix, as it effectively resets your sense of stereo perception.  In this way you might find something or some quality of the recording is actually much more unbalanced than you realized and can then decide if it warrants making an adjustment to correct it or not.  Its also a fresh way to hear what the stereo microphone configuration is doing or how the general "stereoness" of the recording is working, and is a quick, easy check to do.

On the psycoacoustic side of things, there have been recordings I've made where I did not realize Left and Right were swapped, yet something just seemed wrong or not working right.. or rather, everything suddenly seemed correct once I swapped channels.  In those cases I was present at the performance and retained acoustic memory of the event on a subconscious level that conflicted with swapped-channel playback, but didn't consciously recognize what the problem was until trying the swap.  Only at that point did I realize the channel assignment had been incorrect. [edit- This happens rarely, but each time it has I've found myself wondering if if would make any difference to a listener who was not present at the event and has no other preconceived notion of what the "correct" orientation actually is]

[aaronji]

The only mistake here is drawing a connection between what are two different issues.

  No mistake. Personally, I prefer the stereo image in my recordings reflects the reality as best as possible. I think that is a no-brainer, really, and probably most tapers would agree. I would've thought the "virtual teleportation time-machine experience" dude would've agreed...

[Gutbucket]

Sure, I have the same preference, and agree that it matters in that sense.

The "mistake" (overly strong word choice on my part) is associating it with the special case of omnis and time-of-arrival differences generated by any spaced pair.  That in itself is irrelevant because a channel swap will affect a coincident pair of directional mics, which does not capture time-of-arrival differences, in the same way. 

Didn't mean to come across as confrontational, only informational!

[aaronji]

You're arguing against something I never said, though. My post said "spaced omnis are stereo and I prefer to preserve that as it happened". Nothing more, nothing less.

[Gutbucket]

Just trying to clarify.  Not arguing against anything you've said, only trying to point out that the two parts of your statement imply a false causal link:

..personally, I think it matters.

Yes, agreed.

Time differences create stereo imaging with omnis, so..

This part is not relevant, as a swap will affect non-omni time-difference-stereo in the same way AND in either case will not be noticed by a listener who has no preconception of what orientation was originally correct.  The implication of some connection in that regard is what I was attempting to clarify. I've edited my original post to change "mistake" to "potential confusion".

[aaronji]

 

[Gutbucket]

My apologies Arron. I suppose I should step away from TS for a while as I seem to be inadvertently causing other TS members whom I respect to get upset, which is certainly not my intent here or in any other thread.

Yes, time differences create stereo imaging with omnis, but that in itself, and the use of head baffling in particular, are not related to the original question of swapping stereo channels in post.  My attempt in clarifying this was simply to dissuade other readers from assuming that it does.

Original question:

..is there some sonic waveform reason why the L-R channels of an omni mic recording shouldn't be swapped?  Conversely, does swapping stereo channels after the fact cause harm in some barely-detectable way?

[MyNoise]

Thanks for the thoughtful input, y'all.
I'm going to swap the L-R channels for the shows I recorded R-L by mistake... easily done!

[rocksuitcase]

I do a lot of post processing listening and "mixing" of two channel sources. I am going to add that the comment about psychoacoustic perception does have some merit here. Both Aaronji and Gutbucket point this out. I said earlier I did not think it should matter to the OP's question. That said,  I have had many instances where I thought something was "not right" with the sound, maybe a balance issue or some instrument seems out of place, so to speak. Swapping the headphones around can reveal the error. And thankfully with DAW software we can keep both original and swapped files to listen later to see if any difference was actually achieved.

The psychoacoustic aspect seems to relate to having been at the event, knowing the bass was to one side or some other visual/directional cue. IMO, the brain is remarkably "accurate" in its' spacial memory definition and one can often tell the channels are swapped on the recording right away. From a purely technical pov, I see no reason NOT TO try the channel swap in post.

[EmRR]

Conversely, does swapping stereo channels after the fact cause harm in some barely-detectable way?

I do not think this causes any noticeable artifacts

Agreed.  The only super esoteric place I’ve seen effects were mixing on a 1970’s Neve Console that had quadraphonic pan pots.  You could get some mangling if you weren’t careful about the way you flipped or panned things. 

[kuba e]

Psychoacoustically it can matter though.  It can of course if the listener was present at the performance and retains some memory of how things were arranged, or simply how it sounded.  Or if the listener was not there but there is some expected orientation,  such as the drummer always being off to one side and the organist on the other or whatever.  Orchestra recordings are like this. most often with the violin sections to the left and the cello and bass to the right.

I read somewhere that the arrangement of the violins in orchestras was different than today. The 1st and 2nd violin sections were split right and left. And composers used this arrangement while composing, for example Beethoven (I know little about classical music. I will be glad if someone can provide more details). And since nowadays the orchestras are in a different arrangement, it is proof that we should not take psychoacoustic as a dogma either. MyNoise, feel free to switch channels as you like.

Aaronji, I'm sorry, I also didn't understand exactly what you wanted to say. You are recording with omni and you care about the accurate preserve of the event? Ok, I accept.
I can't complain, I'm just a passive member here. This could be an inspiring discussion, I can think of other questions. But lets keep it simple. I will also take a step back.