Hi,

When choosing your AB setup with omni microphones do you have your favourite time difference level you aim for?

How do you combine it with a directionality of high frequencies and the level difference parameter?

Thanks!

can you elaborate? i'm not getting your questions...

Sure!
Using Visualisation of stereo microphone system by Sengpiel:

We get information about time difference depending on how far are the mics set from eachother.

When Orchestra Angle is 90* and mics are 50cm from eachother we get around 1ms time difference between left and right microphone.

http://www.sengpielaudio.com/HejiaE.htm

This is my first question - do you have your favourite time difference you aim for? And does it change depending on the fact how large the group is or what the venue is?

Second thing:

We know that even in omni mics high freqs have directionality, so they behave more like subcardio or cardio mics. Than level differences arises.

On the Sengpielaudio chart you can observe a difference in phantom image between time and level differences what is called Phantom Source Shift "b1 - level difference" and "b2 - time difference"

Then when switching the mics type from omni to subcardio and cardio at the same setup we can see how high frequencies will behave in our setup.

Was wondering if anyone is using this information to clarify the image of the ensemble and use it to place correctly low, mid and high frequencies?

1st: question: no, i (mostly) don't aim at a specific (or just even consider the) time offset/delay much as it varies with the position of the instruments on stage/their angle relative to the a/b system.

2nd question: no again - i base my pattern selection on the width of the ensemble, the distance to the ensemble/recording angle and the wet/dry ratio; however, i do occasionally use either apf's and/or tools to adjust the stereo width depending on frequency.

(that said, i admit that in general i use spaced a/b omnis only very rarely, mostly just as a second main microphone system to give the producer or broadcaster an alternative)

Cool, thanks

For the quote - I like Ozone Imager a lot to order things in this matter, what do you use?
What do you mean by apf's?

I always pan out the AB microphones slightly, say 45° or so, to increase the imaging benefits from the HF directionality.
Especially since my DPA4006TL's have been modified to 60V by Rens Heijnis, this is an audible difference and practical advantage.
Whenever I do not want this improved imaging (e.g. piano solo or harpsichord solo recording) I would rather aim above the instrument, not to favor any particular strings or instrument region.
I do not think I have ever used a larger AB system than 60-65cm (which I use for a 4-mic Faulkner array). I usually start at about 40cm which is my minimum/standard as a main pair and then increase when necessary after listening. I never calculate angles or Singspiel things, it's all between my ears.
For closer capture (say solo voice or instrument spot) I would go for smaller spacing to avoid L-R jumping.
Using the APEs does increase the directionality even more, usually also resulting in more HF pickup but also a more severe L-R separation and thus more hole in the middle even at smaller AB distances. Use with caution if you have to rely on the main pair -- with spot microphones or with a third C mic the gap is easily filled if necessary, though always at a cost.

Oh, and I would never mess with the stereo spacing of the main pair in post. It has to be right on the spot. Phasing/combing issues otherwise -- hard enough to handle with spot microphones, let alone in the main pair!

i'm mixing on a large digital broadcast desk and digital outboard so cannot comment (much) on plugins... (i'm mostly using the drawmer 2476 for frequency-dependent stereo adjustment).

'apf' stands for are 'all-pass filter': a fairly powerful tool and quite commonly used on the way out (to get better coherence between different speaker systems) but what works on the way out also works on the way in... - short: i'm sometimes (mis-)using apf's for frequency-dependent phase alignment of an a/b mic-system, almost exclusively to get a more narrow, mono-compatible low end (if not using a .1/lfe mic in the first place).

True Korneel, but when there are circumstances where you can't put things perfect, or you don't know how than Imager is pretty powerful tool to organize things afterwards, even phase issues when they are always there since AB is a phased array.

And anyway - it is really nice to know how one can be able to correct things after.

I don't follow. AB has no phase/combfilter issues when hard-panned L and R. Nothing from one channel is mixed into the other or summed together in any way, as long as you do not add another microphone to the equation are narrow/widen the image in post.
It is a phased array because of the time/phase differences between L and R, but that does not imply phase issues within one channel of course.

the opposite is true: ANY spaced array yields a partially incoherent soundfield! use an fft if you cannot believe... - now not any two signals which are partially incoherent need to sound unpleasant...

The generally accepted rule of thumb on AB spacing is that an audible hole in the middle begins at about 1 meter.

The great Decca engineer Kenneth Wilkinson began later in his career, to use a simplification of the Decca Tree that was a simple spaced pair as mains, plus the flanks/outriggers. This was soon also adopted by John Dunkerley for many of his sessions. The two mics were spaced about 1 meter, aimed about 90 degrees apart (using Neumann M50s) , and the slight hole in the middle was mitigated by panning the mics in a bit (in your use of the word 'panning', you really meant 'aiming'). The panning was slight enough that any combing that resulted was nearing the threshold of audibility.

Tony Faulkner later simplified this further by bringing the mics just close enough (26-27") that panning-in was no longer required; an effective compromise of very slightly less envelopment and spaciousness of 1 meter spacing, but with absolutely zero (per-channel) combing. The stereo spread of 1 meter panned-in, and 27" panned hard L/R are nearly identical. He uses the same AB spacing in his 4-mic array.

30cm spacing for an omni pair
there is a reason why

Well - that settles that!

End of thread.

seems like it

"The great Decca engineer Kenneth Wilkinson began later in his career, to use a simplification of the Decca Tree that was a simple spaced pair as mains, plus the flanks/outriggers."


How are flankers/outriggers typically used?

My starting point has been omnis spaced about half way between the center and the edge of the ensemble. For the smaller ensembles I usually record, this typically results in flankers spaced about 10 feet apart. And, they are usually on the same line across the front of the ensemble as the main pair. But, usually somewhat higher.

Thank you.

DG

'Standard' Decca technique (on large orch) had the flanks 10-11' either side of the mains (20-22' apart) and a few feet further back; same height as mains. Aimed out just a few degrees (at the middle of their respective sections).

This is, of course using all M50s; with other (non-APE) mics things would probably need to be adjusted.

Thank you.

The reason my flanks are (typically) in line with the main pair (usually ORTF) is that the location I most frequently record in has only a single catwalk, and I have to hang all my mics from the same catwalk. I can adjust the horizontal spacing, of course. And the height. That's it.

DG

Im my personal experience, I seem to like flankers out about 6' from the (tree, main pair, whatever) and aimed ("aimed") out at about 45*, or into the center of the musicians on each side of the stage. 12', tried and abandoned, made the perceived stereo spread too wide for me on speakers. YMMV.

D.

Well, I love an AB array with omnis, and since the time differences between the mics results in different image locations it seems to me it can all be done by ear - listening to the resulting stereo image achieved and adjusting.

And as mentioned, one can achieve a more precise stereo imaging of higher frequencies by angling the mics outward. This too can be done by ear.

I guess you could try to calculate this all out in advance but is seem a bit laborious. But hey, more power to ya! I'd be impressed.

On headphones.
On loudspeakers, of course. Unless you are missing one ear, or have an exceptionally big head.

36cm surely? There is a reason why (depending on mic length...): https://drbadphil.com/omni-mic-pair-in-a-single-blimp



Cheers!

Roland

This, I think, is the primary reason why the standard Decca placement of the flanks was a few feet further back than the mains - the delayed time of arrival reduced the tendency for them to over-widen the image.

Indeed. And this is why we almost never put the flanks further back (nor delay them to the mains), as our mains are coincident double MS. Or a modified tree with ribbons, and the mid a stereo MS ribbon - which means the imaging for the winds & percussion is spot-on.

So, to answers the OP : my favorite time difference is 12 ms + Of course, at those distances you also get significant level differences.