Master Thread: “Objectivism versus Subjectivism” debate and is there a middle ground?

[Blumlein 88]

I am skeptical about height information being recorded in the first place and then reproduced.

It seems definitional that height information requires vertically differentiated microphones and, during playback, vertically differentiated speakers. In a two channel chain, there isn't an opportunity for this.

Hence, advent of multi-channel systems and Dolby Atmos, which specifically allocates channels for height information and includes additional speakers to reproduce it.

Height is sensed by comb filtering caused by the outer ear at the ear canal entrance from frequencies in the 6-12 khz range. Greatest sensation of height is around 11 khz. You wouldn't expect it to be recorded. However I think some coincident microphone techniques like say Blumlein and XY may record something that is heard as height. Because you have two diaphragms that are offset vertically by an inch or so. This would cause objects higher or lower than the microphones to have interchannel comb filtering which might evoke some vague height sensations even though they are probably inaccurate vs real height differences.

 

[DimitryZ]

Height is sensed by comb filtering caused by the outer ear at the ear canal entrance from frequencies in the 6-12 khz range. Greatest sensation of height is around 11 khz. You wouldn't expect it to be recorded. However I think some coincident microphone techniques like say Blumlein and XY may record something that is heard as height. Because you have two diaphragms that are offset vertically by an inch or so. This would cause objects higher or lower than the microphones to have interchannel comb filtering which might evoke some vague height sensations even though they are probably inaccurate vs real height differences.

Interesting. The speakers would have to be pretty darn good to reproduce this high frequency phenomenon. Also, wouldn't ear's own filtering during reproduction interfere with or reinforce this?

 

[Ingenieur]

I wish I could hear 12 kHz flat vs. 6 kHz.
I have my hearing tested annually, it is good, equal to the average guy 10-20 years younger, but still limited vs. my system capability.
At least it's not impaired or damaged.

 

[tuga]

It is a reproduction, a copy, a recording of a performance.
Just like a film, art, photography.

I do not know why people do the things they do.

How many people do you know who calibrate their TV sets? Are they destroying the art by not doing it or by adjusting the image settings according to their preference (and not ABX’ing)?

 

[Blumlein 88]

Interesting. The speakers would have to be pretty darn good to reproduce this high frequency phenomenon. Also, wouldn't ear's own filtering during reproduction interfere with or reinforce this?

No, from speakers which are more or less level with your ears, the ears aren't providing any height cues. So comb filtering super imposed upon the signal would make you hear height even though the source is level with you.

 

[Ingenieur]

How many people do you know who calibrate their TV sets? Are they destroying the art by not doing it or by adjusting the image settings according to their preference (and not ABX’ing)?

They are not optimizing it. Then again, I doubt they are seeing things not in the signal, like depth, 'air', etc., based on their preferred settings.

As I said, we have 0 influence over the recorded performance done xx years ago and xxxx miles away.
Unlike the Uncertainty Principle, our listening does not change the measurement.

 

[Blumlein 88]

How many people do you know who calibrate their TV sets? Are they destroying the art by not doing it or by adjusting the image settings according to their preference (and not ABX’ing)?

I've been calibrating mine for more than 20 years. I hate when people have obvious color issues in their TV. Drives me nuts.

 

[Descartes]

Newly acquired audio-visual associations bias perception in binocular rivalry

When distinct stimuli are presented to the two eyes, their mental representations alternate in awareness. Here, such “binocular rivalry” was used to i…

www.sciencedirect.com

Auditory illusion - Wikipedia

en.m.wikipedia.org

 

[Descartes]

. JSmith

This is excellent❗️

 

[DimitryZ]

No, from speakers which are more or less level with your ears, the ears aren't providing any height cues. So comb filtering super imposed upon the signal would make you hear height even though the source is level with you.

I better get my "Depth of Image" LP and check out if there is height there!

 

[Benedium]

If subjectivists were recording artists, might they make changes to the studio speakers instead of the content? Heheh.

 

[Larry B. Larabee]

Comb filtering is inaudible

 

[antennaguru]

I better get my "Depth of Image" LP and check out if there is height there!

I have owned that Opus3 record and CD, as well as several others from their catalog for years and frankly never noticed the height difference in that song noted in the controlled study which recorded the results of 29 listeners. I too will enjoy listening to that track again for those instruments. BTW, the "Depth of Image" record is an assemblage of many tracks from many different records in the Opus3 catalog, and Opus3 and their recording engineer Bo Hansson were famous for using simple twin microphone figure of eight pattern miking as Blumlein 88 highlighted in his post. Opus3 also were very hands-off when it came to the mastering process.

However, I have heard height imaging clearly portrayed on some systems playing two recordings I have of Flamenco music, Pepe Romero and M&K Flamenco Fever (a track from this is also on the Ohm speaker demonstration record). With Flamenco music the Castinets are played clacking over the head(s) of the dancer(s) and their thick heels are pounding on the wood floor as they dance, while the guitar playing and singing is at the normal heights in between the clacking and pounding. I was always amazed at how well some systems portrayed the height imaging differences between these sounds, and also how some just didn't, but frankly I wasn't sure if it wasn't at least partially expectation bias/psychoacoustics as I have seen/heard live Flamenco music performed in a dinner club - and know what is "should" sound like. That some systems didn't portray the image height variations was the part that always puzzled me because if it was expectation bias then I should have been hearing it with all systems playing these very same recordings.

The study and its conclusion seem reasonable to me, especially as I have heard image height variation before.

There is no test equipment I can think of to measure image height variations other than listening.

 

[DimitryZ]

I have owned that Opus3 record and CD, as well as several others from their catalog for years and frankly never noticed the height difference in that song noted in the controlled study which recorded the results of 29 listeners. I too will enjoy listening to that track again for those instruments. BTW, the "Depth of Image" record is an assemblage of many tracks from many different records in the Opus3 catalog, and Opus3 and their recording engineer Bo Hansson were famous for using simple twin microphone figure of eight pattern miking as Blumlein 88 highlighted in his post. Opus3 also were very hands-off when it came to the mastering process.

However, I have heard height imaging clearly portrayed on some systems playing two recordings I have of Flamenco music, Pepe Romero and M&K Flamenco Fever (a track from this is also on the Ohm speaker demonstration record). With Flamenco music the Castinets are played clacking over the head(s) of the dancer(s) and their thick heels are pounding on the wood floor as they dance, while the guitar playing and singing is at the normal heights in between the clacking and pounding. I was always amazed at how well some systems portrayed the height imaging differences between these sounds, and also how some just didn't, but frankly I wasn't sure if it wasn't at least partially expectation bias/psychoacoustics as I have seen/heard live Flamenco music performed in a dinner club - and know what is "should" sound like. That some systems didn't portray the image height variations was the part that always puzzled me because if it was expectation bias then I should have been hearing it with all systems playing these very same recordings.

The study and its conclusion seem reasonable to me, especially as I have heard image height variation before.

There is no test equipment I can think of to measure image height variations other than listening.

I will have to find and listen to those recordings!

The author of the paper has a reputation of finding outlier cases to demonstrate general findings.

 

[antennaguru]

I will have to find and listen to those recordings!

The author of the paper has a reputation of finding outlier cases to demonstrate general findings.

The Pepe Romero "Flamenco" CD is made by LIM. It's on my music server now, with that CD now stored with thousands of others in boxes in my basement so I can't dig it out to get more info - but that one is the very best example I have heard of Flamenco music, and the vertical imaging is very clear on some systems. I've been a part of controlled depth imaging testing between amplifiers, but not height imaging other than uncontrolled listening.

 

[Chrispy]

How many people do you know who calibrate their TV sets? Are they destroying the art by not doing it or by adjusting the image settings according to their preference (and not ABX’ing)?

Then you like gear that can be highly tuned for its various capabilities, perhaps like capable dsp/eq in audio systems? Calibrating a tv for various sources can be a pain, too. Depends how much fussing you want to do. If you like fussing with vinyl, tho....

 

[Blumlein 88]

Comb filtering is inaudible

Care to clarify what you mean by this?

 

[DimitryZ]

The Pepe Romero "Flamenco" CD is made by LIM. It's on my music server now, with that CD now stored with thousands of others in boxes in my basement so I can't dig it out to get more info - but that one is the very best example I have heard of Flamenco music, and the vertical imaging is very clear on some systems. I've been a part of controlled depth imaging testing between amplifiers, but not height imaging other than uncontrolled listening.

I just listened to a sampling of the tracks on "Flamenco" - 1990 release, no exclamation point in the title.

On my system, the height imaging is there, but it's very subtle. On track 8, for example, the effect is mid-hall, with the castenets being about a foot above the guitar, with my listening position being about 4 meters from the plane of the loudspeakers. Guitar is just about listening level.

At the same time, my mid-high planar drivers are about a foot above my ears, while the low frequency dynamic driver is about 2.5 feet below. My speakers are essentially a line source above 1 Khz, so perhaps height imaging is not their strong side. Maybe modern Quads or minimonitors would be better at this.

My favorite modern flamenco artists are Ottmar Liebert, Jason McGuire and Gipsy Kings.

And check out this flamenco-fueled rendition of Hotel California:

 

[Holmz]

This analogy is a bit dumb...

Radical "Objectivists" are afraid of listening as much as alt-"Subjectivists" hate measurements.
Are the deaf leading the blind?

Close:

The problem here, and I’ll quote Alan Shaw again, is that “‘soundstage’ is a mental construct like ’love’”. There can be no assurance that two people associate the ‘soundstage’ property with the same audible (and/or inaudible) stimuli…
...

Why is it then that the Mrs notices when I get the speakers moved around and the sound stage has depth?
I just said, “listen to this”, and did not mention depth… Ir did it need to be double blind?

It seems like you are saying that people hear differently to each other, which seems either unprovable, or supposition, but largely untrue.

 

[antcollinet]

Close:

Why is it then that the Mrs notices when I get the speakers moved around and the sound stage has depth?
I just said, “listen to this”, and did not mention depth… Ir did it need to be double blind?

It seems like you are saying that people hear differently to each other, which seems either unprovable, or supposition, but largely untrue.

Well, our perception of sound clearly differs. For example rock music may be an energising exciting experience for one person but unbearable racket for another. Or classical an uplifting transformational roller-coaster of emotions for one, and dull as dishwater for the next.

Why would that be if the way our brains process sound is all identical.

Obviously things like stereo soundstage imaging are more difficult to demonstrate differences in perception, but it is a reasonable supposition given that some people claim depth and height are audible, where others can listen to the same, and struggle to detect anything.

The point is a lot of what we detect and perceive in audio is processed in the brain, not the ears. This is *always* susceptible to how we feel, the environment, how we are listening, confirmation bias etc etc. And is why perceived differences between components which measure transparent are generally unlikely to be due to technical differences in the way the components perform.