Scoring Central
stereo vs mono solo orchestral instruments - Printable Version

+- Scoring Central (http://scoringcentral.mattiaswestlund.net)
+-- Forum: Technology (http://scoringcentral.mattiaswestlund.net/forumdisplay.php?fid=5)
+--- Forum: Recording & Sampling (http://scoringcentral.mattiaswestlund.net/forumdisplay.php?fid=9)
+--- Thread: stereo vs mono solo orchestral instruments (/showthread.php?tid=114)



stereo vs mono solo orchestral instruments - Paul Battersby - 09-26-2016

Is there any reason to have stereo samples of a solo orchestral instrument? What I'm finding with solo instrument samples is that the sound moves around from left to right, back and forth randomly in the stereo field. Seems to me, the samples shouldn't do that.

Any thoughts from the resident sampling experts?


RE: stereo vs mono solo orchestral instruments - Mattias Westlund - 09-26-2016

Kind of depends. If, like you say, the samples are clearly moving around across the stereo image, mono is a better option. OTOH I don't know why this is happening -- does it sound like the player was actually not sitting still in front of the mics or is it more a matter of the L/R channel volumes not being balanced for whatever reason (e.g. some formats like soundfonts treat left and right like separate entities and you could technically have 6dB of attenuation on one channel, making it significantly softer than its counterpart)?

As for whether you should have your woodwinds in stereo or mono... I think there was a thread on this very forum on that topic where Samulis and I basically agreed to disagree. Smile

So to end this where it began: it kind of depends. Does mono sound better? Use mono. Does stereo sound better? Use stereo.


RE: stereo vs mono solo orchestral instruments - Paul Battersby - 09-26-2016

(09-26-2016, 08:22 PM)Mattias Westlund Wrote: I think there was a thread on this very forum on that topic where Samulis and I basically agreed to disagree. Smile

That does sound vaguely familiar. I must have seen that at a time before it was relevant to me. I'll go read it after I'm finished typing.

The best way I can describe the issue is this. Picture a VU meter measuring the volume of each channel. I would expect the relative volume between the channels to be constant. Perhaps one channel is a little louder than the other but I would think the ratio should be a constant. If I turn down the volume and just look at the VU meter, it would be indistinguishable from that of an ensemble.

This has been an issue for me as I'm trying to correct the stereo imbalance of various samples. With a solo instrument, I shouldn't hear different notes coming from different places in the stereo field and I was having trouble fixing it because the stereo position of the solo samples kept changing over the duration of the notes. An easy solution, that I'm now thinking is the correct thing to do, is to make all the solo samples mono.


RE: stereo vs mono solo orchestral instruments - Otto Halmén - 09-26-2016

I did some tweaks to my scoring setup about a week ago, among them more exaggerated front-to-back placements of instruments.

Most of my solo samples come from the old Jam Pack libraries, and they're all close-miced in stereo. They're dry, but not anechoic. In my case, when I pull the width control to almost zero, they sound very distant. Once I go full mono, though, something happens that makes them sound "out of place", kind of like they'd turn really small and play back a few inches from your face. It's hard to explain.

Hence, I'm just going to second Mattias on this one: Try them both, and see which fits your samples and scores better. Smile


RE: stereo vs mono solo orchestral instruments - Nayrb - 10-04-2016

Is it possible that the phenomenon is a result of room ambience in the sample? Seems a bit far fetched but I was just musing on this the other day...


RE: stereo vs mono solo orchestral instruments - Paul Battersby - 10-04-2016

(10-04-2016, 01:52 AM)Nayrb Wrote: Is it possible that the phenomenon is a result of room ambience in the sample?

Maybe ... if the walls were moving around as each note was being recorded.  Wink


RE: stereo vs mono solo orchestral instruments - Nayrb - 10-05-2016

(10-04-2016, 02:50 AM)pbattersby Wrote:
(10-04-2016, 01:52 AM)Nayrb Wrote: Is it possible that the phenomenon is a result of room ambience in the sample?

Maybe ... if the walls were moving around as each note was being recorded.  Wink

Good point! Then again, you never know...


RE: stereo vs mono solo orchestral instruments - Samulis - 10-09-2016

(09-26-2016, 09:06 PM)Otto Halmén Wrote: Once I go full mono, though, something happens that makes them sound "out of place", kind of like they'd turn really small and play back a few inches from your face. It's hard to explain.

This is because of phasing between the two channels.

Basically there are two main methods behind capturing a stereo image: coincident arrays and non-coincident array. A coincident array means the sound [theoretically] reaches both mics at the same time (like how a coincidence is something that you would not expect to happen at the same time but does). A non-coincident array, the sound reaches the mics at different times. There are also near-coincident arrays, where the sound reaches the mics at a relatively close time, but it is not identical.

Now, let's think about waveforms for a second. We know sound moves through air as "waves" of denser and less dense air molecules emanating out from a sound source. Well, what happens when a very sparse region and a very dense region collide? They cancel each other out. This is what happens in signal flow as well: when a peak ("hill") collides with a trough ("valley"), they create essentially silence. We can observe this by combining a signal with the inverse of that signal- i.e. if sample #243 has a value of +12, the inverted form of the waveform has a value of -12 there, and -12 plus +12 is 0.

So what does that mean?
Well, if our microphone array is non-coincident, that means... yes, a particular peak can arrive at one mic, at the same time the trough before (or after) it is arriving at the other mic, because one is closer than the other. We call these signals being "out of phase", and combining the signals (i.e. 'summing to mono') will result in certain harmonics and frequencies being cut or boosted- creating exactly the effect you described.

The standard recording array for the "main" mics in orchestral recording is called a Spaced Pair (AB), which is literally what it says on the box: two mics, spaced apart a distance anywhere from 2-6 feet. This is thus a non-coincident array. As such, it is rarely a good idea to simply 'sum' "main mics" in orchestral samples. Note: you may hear about "Decca Tree" or "Tree"- this is essentially a fancy spaced pair with an extra mono mic in the middle and slightly forward to help add more center; as such, it is *slightly not as bad* to sum a decca tree signal- however, you will still get some phasing because the spaced pair portion is not phasing itself away like it would with MS.

So what choice do we have then?
Well, the best option to convert a stereo signal from a non-coincident array into mono is to literally delete one channel and fill it with the other.

Now, there are other microphone techniques that DO frequently reduce down to mono well- Mid-Side or M/S is theoretically phaseless (as the stereo field is GENERATED by phasing and thus perfectly cancels out when summed), XY is very good, and ORTF and NOS are quite good (being near-coincident).