Why might a voice or instrument move it's apparent location in the soundstage?

[VientoB]

To do this, you would first need to look into in-the-head localization with headphones and out-of-head localization with loudspeakers.

If this doesn't work for you with headphones, it's probably because your brain doesn't have enough experience to interpret and evaluate/implement the acoustic information. You'll find plenty of reports and scientific papers online about this.

In such a case, I recommend attending concerts in halls with known good acoustics.

Nah I don't think that's the full story.

 

[5-pot-fan]

I'm sorry, but which one exactly is "Scene 10", can you please point out which one it is in the link below to ensure we are listening to the same thing?

Mozart: Le nozze di Figaro, K. 492

Share your videos with friends, family, and the world

www.youtube.com

Section 36 on the YT video. "Voi signor, che giusto siete".
Is there a simple way to send a copy of relevant/interesting segments to you, if you wish?

 

[Scytales]

This discussion is very interesting and down to earth.

As others have pointed out, I think that knowledge about the number and placement of the microphones, directivity pattern of the latter, as well as use or not of mixing between different microphones, is key to determine the cause of some peculiarities in the reproduced sound-stage.

5-pot-fan, if you wish to asses the performance of your speakers with precisely known recordings, I suggest you seek for records from the French labels BNL, Passavant Music or Syrius. The two sound engineers who runs these labels, Bernard Neveu and Philippe Muller, consistently use for almost all their stereo 2 channels works 2 widely spaced (about 3 m) omnidirectional microphones. No mixing, no dynamic compression (turn the volume up!!), no equalization whatsoever.

I myself encountered some mischief with the reproduction of an humble BNL recording of a solo singer and a piano. With the help of one of the above mentioned sound engineer, I quickly understand that adjustment of the toe-in of my two main speakers were required. Once done, the mischief has disappeared like magic.

Just for the sake of illustration, here are two BNL albums I just listened to the past few days :

The liner notes of the CD edition of the latter one contains a photography of the recording session on which the microphones are clearly apparent. That helps to figure out what should happen.

I understand that others sound engineers used to precisely divulged their recording techniques, for instance John M. Eargle (Délos), but I know less about them.

Edit : correction of wrongly identified second album !

 

[Roland68]

Nah I don't think that's the full story.

Of course it also depends on the headphones, a good resolution is beneficial, and also on the music.
Hotel California Eagles Live 1994 is a good recording for this. The applause from around 1:17 is perceived by many people as very 3D. In fact, a system has to be very good to reproduce it as well as with my K812 headphones. But that only applies to me.

I just wanted to give you a hint as to what it is connected with and it is up to you to make something of it. This topic was discussed at length in a German forum and helped many people.

What is of no use to you or others is to doubt that it works for others.

 

[5-pot-fan]

To answer the OP's question, there are two possibilities:

1. The problem is in the recording. When more than one mic is used, there can be phase issues between the mics, especially if the mics are placed close to the vocalist. Then any movement by the vocalist exaggerates movement. I have a few recordings that sound like this, one example is this recording:

View attachment 403123

You will note that on some of the tracks, Bostridge suddenly snaps to the left, sometimes mid-note. I suspect he is moving around whilst singing. Blame the recording engineer for this one.

2. The problem is in your setup - so-called "phantom centre image drift". What it sounds like: some frequencies are centred, but other frequencies drift to the left or right. It is most easily detected by playing a sine wave at different frequencies to both speakers. The sine wave tone should be sharply centred between the speakers. You might find that with some frequencies, the tone might drift to the left or the right. The cause is phase asymmetry between speakers at certain frequencies. Such phase issues may be caused by your loudspeakers being dissimilar (e.g. crossover components or drivers drifting out of tolerance), asymmetric room layout or listening position, improper DSP, etc.

To tell the difference between the two, listen to the recording on headphones. If there is no image drift, you have a problem with your room or speakers.

Hi,
I have run a sinewave test a few times, 1khz to 4khz, and, while there are peaks and troughs (as expected), I cannot discern any L-R movement via speakers or headphones. Your observation may well be correct but I will need to find another method to observe it.

 

[5-pot-fan]

This discussion is very interesting and down to earth.

As others have pointed out, I think that knowledge about the number and placement of the microphones, directivity pattern of the latter, as well as use or not of mixing between different microphones, is key to determine the cause of some peculiarities in the reproduced sound-stage.

5-pot-fan, if you wish to asses the performance of your speakers with precisely known recordings, I suggest you seek for records from the French labels BNL, Passavant Music or Syrius. The two sound engineers who runs these labels, Bernard Neveu and Philippe Muller, consistently use for almost all their stereo 2 channels works 2 widely spaced (about 3 m) omnidirectional microphones. No mixing, no dynamic compression (turn the volume up!!), no equalization whatsoever.

I myself encountered some mischief with the reproduction of an humble BNL recording of a solo singer and a piano. With the help of one of the above mentioned sound engineer, I quickly understand that adjustment of the toe-in of my two main speakers were required. Once done, the mischief has disappeared like magic.

Just for the sake of illustration, here are two BNL albums I just listened to the past few days :

The liner notes of the CD edition of the latter one contains a photography of the recording session on which the microphones are clearly apparent. That helps to figure out what should happen.

I understand that others sound engineers used to precisely divulged their recording techniques, for instance John M. Eargle (Délos), but I know less about them.

Edit : correction of wrongly identified second album !

Thank you for this. I will try and find these (but it may not be easy).

 

[goat76]

Section 36 on the YT video. "Voi signor, che giusto siete".
Is there a simple way to send a copy of relevant/interesting segments to you, if you wish?

Nah, there is no need to send anything as I can easily listen to the YT clip directly on my main sound system. Thanks anyway.

• At 21 to 23 seconds into the track, the male singer is heard from a point fairly close the the left speaker, approximately 2/3 closer to the left speaker than to the phantom center.
• At 25 to 29 seconds, he has now moved a little closer to the phantom center, approximately 1/3 to the left speaker from the phantom center.
• At 30 to 39 seconds, he has now moved to the right side of the stage, approximately 2/3 closer to the right speaker than to the phantom center.

Yes, it's the same sections of the track I isolated in pictures in an earlier post, it has just moved a little forward time-wise as I recorded it from YouTube which created a 1.8-second silence before the song actually started.

I think we hear the recording the same way, if we pinpoint the sound somewhat differently it probably depends on how we physically have positioned our loudspeakers. I listen at a fairly short distance of 2 meters in an equilateral listener triangle which makes it easy to pretty accurately hear where things are panned in a mix.

 

[5-pot-fan]

Nah, there is no need to send anything as I can easily listen to the YT clip directly on my main sound system. Thanks anyway.

• At 21 to 23 seconds into the track, the male singer is heard from a point fairly close the the left speaker, approximately 2/3 closer to the left speaker than to the phantom center.
• At 25 to 29 seconds, he has now moved a little closer to the phantom center, approximately 1/3 to the left speaker from the phantom center.
• At 30 to 39 seconds, he has now moved to the right side of the stage, approximately 2/3 closer to the right speaker than to the phantom center.

Yes, it's the same sections of the track I isolated in pictures in an earlier post, it has just moved a little forward time-wise as I recorded it from YouTube which created a 1.8-second silence before the song actually started.

I think we hear the recording the same way, if we pinpoint the sound somewhat differently it probably depends on how we physically have positioned our loudspeakers. I listen at a fairly short distance of 2 meters in an equilateral listener triangle which makes it easy to pretty accurately hear where things are panned in a mix.

Hi
I have watched this YT video :

and have the following observations:
There are 2 male vocalists here, at least in the YT video.
at 1:33:45 - start of scene - “Volsignor, che giusto siete”
then a short triplet –Count + Countess + Susanna
then Figaro (L of C) starts 21 secs in – phrase ends 'a far', finishing approx 8 secs later (29secs in)
and Count (R of C) starts 30 secs in , ends approx 27 secs in with Marcellina starting a phrase.

I appear not to be able to distinguish these 2 male voices easily by ear, but the different locations in the soundstage are clear, and if the video is of a 'standard' performance then it seems the movements in the soundstage are from different singers.
I am new to listening to opera and would be happy to be corrected/educated further.
However please see my next post as a follow-on to this.

 

[5-pot-fan]

As this recording is of "The Marriage of Figaro," it's quite possible the singers were moving around. Some opera recordings in the early stereo era were staged - London's recordings notably. Considering how important movement is in "Figaro," I could easily imagine this recording was at least partially staged, even if it is slightly outside the time window when the practice was fairly common.

Thank you - from more recent observations I think you may be correct, and that some of the movement is 'real' by the performers. Please see my comments to @goat76.

 

[5-pot-fan]

Have you checked that your hearing is OK and the same on both sides?

Initially I was not too keen on your comment(!) but I am new to opera and do find it difficult to distinguish between voices. Age is definitely also a factor here. Please see my later reply to @goat76 for an example of the issue for me - resolved only by checking on video.

 

[5-pot-fan]

Following all your comments, some further reading, plus some work using REW. I have made a change to my speakers setup, bringing the right speaker forward by 40mm. This has evened out an imbalance in sound at my LP, which was previously countered by an increase of approx 1.5db to the righthand channel. The channel balance is now at zero, and the REW test measurements appear to have a more even SPL.
During further auditions of the piece that started my query, I can no longer detect the sudden jump in vocal position, and although the vocals do move the overall changes are less marked.
I am not surprised that the change in speaker position has made a change overall, but I would not have guessed it would do this - so I am glad to have raised this thread and will try to keep learning.
My thanks to all who have commented so far, and any further points, especially regarding directivity, will be gladly received.

 

[MaxwellsEq]

Initially I was not too keen on your comment(!) but I am new to opera and do find it difficult to distinguish between voices. Age is definitely also a factor here. Please see my later reply to @goat76 for an example of the issue for me - resolved only by checking on video.

It's always worth getting your heating checked if something sounds wrong. I have tinnitus which comes and goes but is worse on the left when it's prominent. This leads to stereo images "moving with frequency", because very high frequencies are slightly masked on the left side by the noise. I've had an MRI scan to ensure the tinnitus is not caused by lesions on the nerves.

So, checking your hearing regularly is worth doing and may show up other issues early! There are online tools to do it.