Speaker Testing: why mono is better

[ahofer]

Maybe I'm weird, but the reason for mono testing seems intuitively clear to me. Assuming the speakers are identical, there isn't any new information *about the speaker* present when you test two at once. What you do get is much clearer L-R cues in the *recording* and more room effects.

I know some speakers have offset tweeters in R/L pattern, and I suppose it's *possible* to design speakers with differing inside/outside radiation patterns. I doubt the latter is desirable. In the absence of asymmetrical pairs, using just one seems like a good way to control for speaker characteristics as opposed to other effects.

 

[Inner Space]

Assuming the speakers are identical ...

Big assumption, often ...

... there isn't any new information *about the speaker* present when you test two at once.

Wouldn't a tiny defect now be more obvious, being 3dB louder?

Asking for a friend ...

 

[krabapple]

From chapter 7:

In monophonic tests, (10, including Toole and possibly Olive) listeners reported large differences in both sound quality and spatial quality.
However, in stereo listening most of the differences disappeared in these data that average ratings for all programs.
The two highly rated loudspeakers kept their high sound quality ratings, but the loudspeaker with low spatial ratings in mono became competitive in stereo.
This was a puzzle, because it had been assumed that it was stereo that would reveal the relative merits in terms of imaging and space.

That is from a chapter about Subjective Evaluations in Real World Situations. Specifically, a subsection on The Effect of Loudspeaker Directivity , the topic of Toole's seminal 1985 paper

The rest of the subsection explores the solving of this puzzle, by closely analysing the data in 1985. I suggest you review it carefully. I quote two relevant passages below, whose implications you should ponder:

In stereo listening, all of the variables in this test were influential, but the nature of the recordings themselves proved to be the overriding factor. The essential ingredients of "imaging " are in the recordings.
.
.
Finally, evaluate loudspeakers in monophonic comparisons (to find out what you really have) and demonstrate them in stereo or multichannel (to impress everybody). Choose the recordings carefully -- they are a significant factor. Subsequent stereo vs mono tests in the intervening 30+ years have not changed these conclusions.

 

[krabapple]

I don't get it. there is no space when listening in mono

Not this again, please.

 

[krabapple]

Big assumption, often ...

If speakers of the same make and model are routinely non identical to a degree that matters, then isn't it hopeless for audiophiles to audition in mono OR stereo? Every pair of speaker you buy would be a crapshoot.

However , if it's a simple output level difference...that's what channel level calibration is for. It's standard on every AVR for years now.

 

[tuga]

In stereo listening, all of the variables in this test were influential, but the nature of the recordings themselves proved to be the overriding factor. The essential ingredients of "imaging " are in the recordings.

I wonder if this means that "spatial quality" is no longer relevant when listening to two speakers because you have...well, stereophony.

 

[tuga]

If speakers of the same make and model are routinely non identical to a degree that matters, then isn't it hopeless for audiophiles to audition in mono OR stereo? Every pair of speaker you buy would be a crapshoot.

However , if it's a simple output level difference...that's what channel level calibration is for. It's standard on every AVR for years now.

Audiophiles with AVRs? Where do you live?

 

[krabapple]

Audiophiles with AVRs? Where do you live?

The United States. You might have heard of it.

 

[krabapple]

I wonder if this means that "spatial quality" is no longer relevant when listening to two speakers because you have...well, stereophony.

You might instead wonder how relevant speaker quality is to 'spatial quality'.... when listening in stereo.

 

[JSmith]

Too bad Toole couldn't get it done.

Are you seriously questioning the research done by @Floyd Toole and Harman?

I highly recommend you watch this in detail;

If you have a more technical reason for this stance rather than a one liner, please share it.


JSmith

 

[kongwee]

I wonder if this means that "spatial quality" is no longer relevant when listening to two speakers because you have...well, stereophony.

Dolby Atmos music mix, usually higher DR rate than stereo mix, even down to two channel. Many people here are begging the market to do it so.

 

[dasdoing]

Not this again, please.

I don't want to spend much time with this subject, so I wont read the whole topic.
I assume that they were efectivily testing dispersion (and therefore reflection) caracteristics. it's obvious that a more disperse speaker will sound wider in mono. now to make the asumption that this translates to stereo is wrong (and is acatualy proven by them). to conclude that image therefore should be better avaliated in mono is hard to understand. Now I am not saying that speakers must be compared in stereo to avaliate their pure performance, but if you want to avaliate stereo imaging you obviously have to listening in stereo. again, I can't even understand why someone would disagree.

EDIT: I think they just use the wrong term. they are avaliating dispersion, not imaging. yes, the dispersion is better identified in mono. now you don't need a subjective test for dispersion. why would you? it is easily meassured

 

[Vuki]

Never solved mystery - additional data lost in time...

 

[SIY]

Maybe I'm weird, but the reason for mono testing seems intuitively clear to me. Assuming the speakers are identical, there isn't any new information *about the speaker* present when you test two at once. What you do get is much clearer L-R cues in the *recording* and more room effects.

I know some speakers have offset tweeters in R/L pattern, and I suppose it's *possible* to design speakers with differing inside/outside radiation patterns. I doubt the latter is desirable. In the absence of asymmetrical pairs, using just one seems like a good way to control for speaker characteristics as opposed to other effects.

Ken Kantor strongly disagrees. The NHT speakers, and before that, his AR designs had very deliberate asymmetry.

 

[krabapple]

Are you seriously questioning the research done by @Floyd Toole and Harman?

I thought he was referring to this:

Sure hope in my lifetime cea2034 ratings will be published by manufacturers

That's something Toole has advocated

 

[krabapple]

I don't want to spend much time with this subject, so I wont read the whole topic.
I assume that they were efectivily testing dispersion (and therefore reflection) caracteristics. it's obvious that a more disperse speaker will sound wider in mono. now to make the asumption that this translates to stereo is wrong (and is acatualy proven by them). to conclude that image therefore should be better avaliated in mono is hard to understand. Now I am not saying that speakers must be compared in stereo to avaliate their pure performance, but if you want to avaliate stereo imaging you obviously have to listening in stereo. again, I can't even understand why someone would disagree.

EDIT: I think they just use the wrong term. they are avaliating dispersion, not imaging. yes, the dispersion is better identified in mono. now you don't need a subjective test for dispersion. why would you? it is easily meassured

Suggest you read the chapter to get the context for the 1985 study.

 

[krabapple]

Never solved mystery - additional data lost in time...
View attachment 194872

Except, Toole spends pages dissecting factors that likely produced this result. You don't seem to be following along.

At the risk of summarizing him incorrectly, Toole has noted that as channel number increases, the detection of some audio system deficits becomes more difficult -- or to put it plainly, we notice them less. In this case, loudspeaker ratings become more comparable in stereo versus the starker differences seen in mono. In particular the rating of 'spatial quality' starts to be driven more by the recorded content (and environment) than the loudspeaker.

 

[Vuki]

Except, Toole spends pages dissecting factors that likely produced this result. You don't seem to be following along.
...

No, he doesn't.
Actually, here is his comment:
https://www.audiosciencereview.com/...n-speaker-preference-score.31454/post-1122595

 

[JSmith]

I thought he was referring to this:

Sure hope in my lifetime cea2034 ratings will be published by manufacturers

That's something Toole has advocated

You are quite correct, I must have been drunk... I should have gone home. 0_o


JSmith

 

[tuga]

You might instead wonder how relevant speaker quality is to 'spatial quality'.... when listening in stereo.

Speaker quality as in transduction of the signal with minimal distortion?
It is paramount.