A silly (ignorant) question concerning stereo speaker response - advance apology!

[goat76]

Slightly right of center here, with speakers as well as with headphones.

After listening to the song, I was surprised that most people here say the finger snaps are coming half the way to the left, 10% inside the left speaker, or even outside the left speaker. With my speakers, I hear the finger snaps coming from about the same middle spot as the singer with a small tendency to the left of the middle of the speakers.

I wanted to make sure I heard it correctly so I checked it out in iZotope RX.
The largest difference in the level of the finger-snapping I could find in the song was the very first one but it was only 0.6 dB louder in the left channel than the right channel, and there is no time delay or phase difference between the channels that could mess with the perceived position of the finger snaps either.

The level differences between the channels for the first five finger snaps in the song:

1. 0.6 dB to the left
2. 0.4 dB to the left
3. 0.0 dB
4. 0.2 dB to the left
5. 0.2 dB to the right

Tidal has 3 different releases of this track but I found them all sounding the same. The " An Innocent Man" I analyzed was from the album "The Essential BILLY JOEL". If anyone of you hear the finger snaps far to the left in that track, I think something fuzzy is going on in your speaker systems.

 

[Esprit]

Or in the acoustics of the room…

 

[mj30250]

The snaps are very slightly to the left of the vocal on my desktop system. It seems to change in height somewhat, though.

 

[Timmy]

1. 0.6 dB to the left
2. 0.4 dB to the left
3. 0.0 dB
4. 0.2 dB to the left
5. 0.2 dB to the right

All »my« five samples measure a plus for the left – between 0.4 to 0.5 dB. Source is a purchased Flac file from 2006

 

[JeremyFife]

Fun, and easy to check ... streaming from Amazon (SD) ... clear on the left side, left of the vocal and pretty much inline with the speaker. Not sure about relative depth, perhaps perceived as 'behind' the vocal but not significantly. Office system, so nearfield.

Unfortunately, I'd forgotten that I really don't like Billy Joel ... ah well

 

[Joffy1780]

Why bother with songs?
Using the test below one can determine all directions (and there's more tests on the site for the rest of the scene and all the clues)

Virtual Acoustic Space | Auditory Neuroscience

auditoryneuroscience.com

View attachment 388589

Wow this is very cool, thank you

 

[goat76]

All »my« five samples measure a plus for the left – between 0.4 to 0.5 dB. Source is a purchased Flac file from 2006

In the DAW I'm using (Reaper), a small difference of just 0.5 dB higher level in the left channel will pan the sound object only 5% to the left. A 6 dB higher level in the left channel is required for the sound object to be panned halfway (50%) to the left, and a 12 dB higher level is needed to pan the object 75% to the left.

Unless some of the people in this thread are listening to another mix of the song than the one I analyzed, there is something fishy going on with their speaker setups if they hear the position of those finger-snaps close to the position of the left speaker.

 

[TLEDDY]

Why bother with songs?
Using the test below one can determine all directions (and there's more tests on the site for the rest of the scene and all the clues)

Virtual Acoustic Space | Auditory Neuroscience

auditoryneuroscience.com

View attachment 388589

EXCELLENT!!!

The test worked perfectly!

Thank you!

Tillman

 

[TLEDDY]

Here's an interesting set of files that give a more complete picture of your systems imaging/soundstage capabilities.

Online LEDR™ Sound Test | Listening Environment Diagnostic Recording Test

Audio Test Tones : check your stereo system for correct imaging

www.audiocheck.net

Really works dream!

 

[Bugal1998]

1. Does this subjective test make sense?

2. If you try the test on any/all of your own equipment-
A. Do you hear the finger snap
B. If so, where do you perceive it in the sound field?

3. Do you consider the last few minutes reading this as time lost that you can never get back?

1. Maybe... or maybe not. Unless the spectral content is changing meaningfully between snaps, why would the location change from snap to snap on ANY system? And if the relative level between left and right is inconsistent, how do we know if they're supposed to stay in the same spot. What about timing delays? If the snaps were recorded with a close stereo mic then small movements could result in meangingful positional shifts without changing relative L/R levels by much. I agree with @Sokel on the questionable utility of using songs for diagnostic purposes.

2a. Yes, I hear it loud and clear.
2b. Basically dead center or slightly to the left. It doesn't stay locked exactly perfectly in the same place the entire song, but nearly so. It's also drenched in reverb that fills the room, including somewhat above and behind me.

3. Not at all! I've been thinking about starting an imaging thread for ages simply to see what relative differences emerge between listeners and systems. Thanks for posting an interesting question!

Edit: I saw @goat76 's analysis after posting, and it seems there are no timing differences between channels.

 

[Bugal1998]

After listening to the song, I was surprised that most people here say the finger snaps are coming half the way to the left, 10% inside the left speaker, or even outside the left speaker.

This issue could easily manifest simply by one's posture when listening. Even when bringing experienced listeners into the room I seldom observe them getting centered between the speakers, often leaning comfortably left or right in the chair (not judging, just observing!).

If I shift a little left or right in my seat the location shifts exponentially, or certainly more than 1-to-1. For two channel listening the speakers are toe'd out as far as possible without causing the high frequency response below 15khz to drop off. This gives a spacious and "holographic" immersion by putting acoustic energy on the side-walls, but also accelerates the precedence effect by causing rapid volume level increases (when shifting toward) and decreases (when shifting away) between the speakers when moving left or right. By the time I reach the left or right arm rest the snaps and vocal have shifted almost entirely to the left or right speaker.

A little left or right in the chair and the reported snap position could change meangingfully.

 

[goat76]

This issue could easily manifest simply by one's posture when listening. Even when bringing experienced listeners into the room I seldom observe them getting centered between the speakers, often leaning comfortably left or right in the chair (not judging, just observing!).

I often make sure I sit right in the middle of the speakers. My speakers have curved sides so it's easy to see when I'm sitting at the exact sweet spot.

And yes, your observation might be right, some of the people who hear those finger snaps coming far off to the left side are maybe not sitting at the exact sweet spot. But why are most people leaning to the left side and not the right? Hmm...

If I shift a little left or right in my seat the location shifts exponentially, or certainly more than 1-to-1. For two channel listening the speakers are toe'd out as far as possible without causing the high frequency response below 15khz to drop off. This gives a spacious and "holographic" immersion by putting acoustic energy on the side-walls, but also accelerates the precedence effect by causing rapid volume level increases (when shifting toward) and decreases (when shifting away) between the speakers when moving left or right. By the time I reach the left or right arm rest the snaps and vocal have shifted almost entirely to the left or right speaker.

A little left or right in the chair and the reported snap position could change meangingfully.

I prefer the focused sound I get with the speakers toed in towards me at the main listening position. I have them set up so that the axial point is about 20 centimeters behind my head, I find that to be most optimal with my speakers for getting a good balance of high-frequency energy and a focused-sounding phantom center.

I'm not much into the idea of trying to enlarge the sweet spot (which can be done with extreme toe-in crossing in front of the listening spot) as I almost always listen to music by myself, and optimal stereo reproduction is pretty much a one-man sport and can only be fully optimized for a single listener anyway.

 

[goat76]

1. Maybe... or maybe not. Unless the spectral content is changing meaningfully between snaps, why would the location change from snap to snap on ANY system? And if the relative level between left and right is inconsistent, how do we know if they're supposed to stay in the same spot. What about timing delays? If the snaps were recorded with a close stereo mic then small movements could result in meangingful positional shifts without changing relative L/R levels by much. I agree with @Sokel on the questionable utility of using songs for diagnostic purposes.

2a. Yes, I hear it loud and clear.
2b. Basically dead center or slightly to the left. It doesn't stay locked exactly perfectly in the same place the entire song, but nearly so. It's also drenched in reverb that fills the room, including somewhat above and behind me.

3. Not at all! I've been thinking about starting an imaging thread for ages simply to see what relative differences emerge between listeners and systems. Thanks for posting an interesting question!

Edit: I saw @goat76 's analysis after posting, and it seems there are no timing differences between channels.

The reverb for the finger snaps is artificial, and it is probably a stereo reverb which could explain the very small variations in the stereo balance. The finger snaps are probably samples with random velocity.

 

[Bugal1998]

But why are most people leaning to the left side and not the right? Hmm...

Now that's an ASR research question if I've ever heard one!

I'm probably too obsessive about being centered, and like you I have various visual cues that can be used to get centered.

I prefer the focused sound I get with the speakers toed in towards me at the main listening position. I have them set up so that the axial point is about 20 centimeters behind my head, I find that to be most optimal with my speakers for getting a good balance of high-frequency energy and a focused-sounding phantom center.

Toed in I can get a dual mono pink noise phantom image as tight as a large grapefruit (give or take), or fairly diffuse and 6-8' wide when pointed straight ahead. I'd say the current setup yields a beachball sized phantom image that exists slightly behind the plane of the speakers (12.5' listening distance).

The acoustic lens on the HF driver in my sysyem narrows output beginning ~10khz, and I find the tonality and spaciousness improves if the speaker's on-axis acoustic energy is allowed to pass outside the listening position and the excess HF fall-off on the MLP-axis is then restored via EQ. This approach roughly achieves a neutral total power response for the speaker.

I'm not much into the idea of trying to enlarge the sweet spot (which can be done with extreme toe-in crossing in front of the listening spot) as I almost always listen to music by myself, and optimal stereo reproduction is pretty much a one-man sport and can only be fully optimized for a single listener anyway.

100% agreed on all counts! Either the timing between the L/R speakers is aligned or it's not: dead center is always the sweet spot.

 

[sigbergaudio]

Pretty much dead center here as well, on a setup that is generally quite extreme on any such effects.

Test for instance "Woman" by Simple Minds, the effects in the intro pans 90 degrees in either direction on my setup (so way past the speakers).

 

[dualazmak]

If you would be interested in the tracks of "SONY SUPER AUDIO CHECK CD 48DG3" (ref. here, the booklet PDF attached herewith) including very useful test tones/reference-music-tracks, please simply let me know by PM communication.

 

[dualazmak]

Attached also, you would please find spectrum and other data in PDF of all the tracks of "SONY SUPER AUDIO CHECK CD 48DG3" analyzed by Adobe Audiotion 3.0.1 and MusicScope 2.1.0.