Stereo vs mono - what is the ideal frequency response?

[Thomas_A]

I am making an separate thread regarding the dilemma of stereo reproduction and the current research saying that a speaker should ideally show a flat response. I don't know if this has been discussed in length before, although the issue has been touched upon in several threads. According to the research (Harman), listening has been done in the most sensitive mode, meaning a single speaker in mono, both using signals and music material. As far as I know, the research has come to the conclusion that the on-axis response should be dead flat.

Now to the dilemma. If ideal speaker should be flat as judged from mono speaker playing in front of you, you will hear different timbral response if you place the same speaker in a stereo triangle. This is evident from previous research by Shirley, Toole and others. The conclusion from Toole that "stereo is flawed" (which is true) and one should go for multichannel instead. True multichannel recordings are however few, and you are still stuck with tons of ordinary stereo recordings. The real question is however avoided: Given that a flat response is preferred for a mono speaker - stereo speaker cannot show the same flat response if the tonality of the centre phantom image should be the same. So what is the ideal frequency response for speakers playing in stereo?

 

[StevenEleven]

First I don’t think the Harman research shows the speaker should be dead flat in a room in a home. A little extra bass seems to be preferred for starters, and the highs maybe rolled off just a little. The smoother the better within those parameters perhaps. That‘s the broad range average of preference, I think, but an individual may have variations on the preference, plus or minus.

When I calibrate my system I calibrate in mono (center channel and subwoofers), stereo, and multichannel. The same room modes are there for all configurations and that‘s the major thing to take care of for me. Otherwise, interesting question in theory, but I am not sure a speaker or speakers in a real home room is ever going to be close enough to “dead flat“ for it to be a practical question.

I usually listen to stereo upmixed to multichannel, FWIW.

When I’m done I add the salt and pepper and spicey sauce—I tweak the broad-range bass and treble to suit to my personal taste.

 

[thewas]

Given that a flat response is preferred for a mono speaker

I think what is or should be preferred is as similar conditions as possible to the setup used at the mixing and mastering of the track. So assuming that nowadays more and more studios use relatively anechoically linear on axis loudspeakers with smooth directivity the answer is imho obvious. Trying to find per trial and error an "optimal" FR listening to some prerecorded music is just continuing Audio's Circle of Confusion.

 

[Thomas_A]

I think what is or should be preferred is as similar conditions as possible to the setup used at the mixing and mastering of the track. So assuming that nowadays more and more studios use relatively anechoically linear on axis loudspeakers with smooth directivity the answer is imho obvious. Trying to find per trial and error an "optimal" FR listening to some prerecorded music is just continuing Audio's Circle of Confusion.

So how would ju judge the "mono speaker" listening session by Harman? Are they wrong?

 

[Thomas_A]

First I don’t think the Harman research shows the speaker should be dead flat in a room in a home. A little extra bass seems to be preferred for starters, and the highs maybe rolled off just a little. The smoother the better within those parameters perhaps. That‘s the broad range average of preference, I think, but an individual may have variations on the preference, plus or minus.

When I calibrate my system I calibrate in mono (center channel and subwoofers), stereo, and multichannel. The same room modes are there for all configurations and that‘s the major thing to take care of for me. Otherwise, interesting question in theory, but I am not sure a speaker or speakers in a real room is ever going to be close enough to “dead flat“ for it to be a practical question.

I usually listen to stereo upmixed to multichannel, FWIW.

When I’m done I add the salt and pepper and spicey sauce—I tweak the broad-range bass and treble to suit to my personal taste.

I think it is best to leave the room effects (room gain and frequency-dependent absorption) out of the equation, it becomes more complicated. The issue that I really want an opinion on is the stereo system error dilemma, vs previous research regarding the ideal speaker response for a mono source.

 

[thewas]

So how would ju judge the "mono speaker" listening session by Harman? Are they wrong?

Those are a compromise made to compare loudspeakers and actually their radiation / sound power behaviour, not to find an optimal anechoical on-axis response where Toole and previous experiements underlines it should be flat, see here more about it
https://www.audiosciencereview.com/...ut-room-curve-targets-room-eq-and-more.10950/
First sentence:
Almost 50 years of double-blind listening tests have shown persuasively that listeners like loudspeakers with flat, smooth, anechoic on-axis and listening-window frequency responses.

 

[StevenEleven]

Those are a compromise made to compare loudspeakers and actually their radiation / sound power behaviour, not to find an optimal anechoical on-axis response where Toole and previous experiements underlines it should be flat, see here more about it
https://www.audiosciencereview.com/...ut-room-curve-targets-room-eq-and-more.10950/
First sentence:
Almost 50 years of double-blind listening tests have shown persuasively that listeners like loudspeakers with flat, smooth, anechoic on-axis and listening-window frequency responses.

In any event, that’s just about the only way out of the circle of confusion.

 

[flipflop]

The issue that I really want an opinion on is the stereo system error dilemma, vs previous research regarding the ideal speaker response for a mono source.

Is it comb filtering you're worried about? Early reflections partially solve the problem. Chapter 7.1.1 in Toole's book deals with this topic.

 

[thewas]

In any event, that’s just about the only way out of the circle of confusion.

Exactly!

 

[Thomas_A]

Those are a compromise made to compare loudspeakers and actually their radiation / sound power behaviour, not to find an optimal anechoical on-axis response where Toole and previous experiements underlines it should be flat, see here more about it
https://www.audiosciencereview.com/...ut-room-curve-targets-room-eq-and-more.10950/
First sentence:
Almost 50 years of double-blind listening tests have shown persuasively that listeners like loudspeakers with flat, smooth, anechoic on-axis and listening-window frequency responses.

But again, if this is in the stereo mix - the listeners should have preferred a curve according to Shirley et al. All other results should have been noted as audible resonances.

 

[Thomas_A]

Is it comb filtering you're worried about? Early reflections partially solve the problem. Chapter 7.1.1 in Toole's book deals with this topic.

Yes, but it still gives variations of +/- 2 dB 1-10 kHz in a normally reflected room for the phantom centre. This should have been audible when testing with a mono speaker.

 

[Vini darko]

One way to partially solve this. Use three speakers with right and left normally. And run the third speaker mono in the center off the same amp using some 4ohm resistors going to both positives. I've done it before and it gives interesting results. Not recommended with fancy high priced amps though.

 

[Thomas_A]

One way to partially solve this. Use three speakers with right and left normally. And run the third speaker mono in the center off the same amp using some 4ohm resistors going to both positives. I've done it before and it gives interesting results. Not recommended with fancy high priced amps though.

Matrix decoding can solve this, if there is an optional correction curve for the phantom centre/center channel, and if the correction has been made during recording. I know of no such function in commercial products though.

 

[Vini darko]

I had to go look up what matrix decoding is. Sounds a lot like dobly Pro logic. That was able to be embedded in stereo I think. Hardly ever sounded good in my limited experience.

 

[Thomas_A]

I had to go look up what matrix decoding is. Sounds a lot like dobly Pro logic. That was able to be embedded in stereo I think. Hardly ever sounded good in my limited experience.

There has been other ways to extract center (and surround) channels from stereo, e.g. Hafler stereo.

 

[Kal Rubinson]

There has been other ways to extract center (and surround) channels from stereo, e.g. Hafler stereo.

Well, sorta.

 

[March Audio]

I am making an separate thread regarding the dilemma of stereo reproduction and the current research saying that a speaker should ideally show a flat response. I don't know if this has been discussed in length before, although the issue has been touched upon in several threads. According to the research (Harman), listening has been done in the most sensitive mode, meaning a single speaker in mono, both using signals and music material. As far as I know, the research has come to the conclusion that the on-axis response should be dead flat.

Now to the dilemma. If ideal speaker should be flat as judged from mono speaker playing in front of you, you will hear different timbral response if you place the same speaker in a stereo triangle. This is evident from previous research by Shirley, Toole and others. The conclusion from Toole that "stereo is flawed" (which is true) and one should go for multichannel instead. True multichannel recordings are however few, and you are still stuck with tons of ordinary stereo recordings. The real question is however avoided: Given that a flat response is preferred for a mono speaker - stereo speaker cannot show the same flat response if the tonality of the centre phantom image should be the same. So what is the ideal frequency response for speakers playing in stereo?

That's not quite what the research showed. It showed that a flat anechoic response (not in room measured response) with smooth off axis response was preferred. The same preference (speakers rated in the same order) was found whenever stereo tests were performed. Stereo does not affect this preference.

 

[Thomas_A]

That's not quite what the research showed. It showed that a flat anechoic response (not in room measured response) with smooth off axis response was preferred. The same preference (speakers rated in the same order) was found whenever stereo tests were performed. Stereo does not affect this preference.

Even though they do sound different, according to Toole.

 

[Thomas_A]

A mini "preference test" for those who want. Must be listened in stereo speakers (not headphones). Which one is preferred, if any of them?

https://www.dropbox.com/s/h14d9owhlonz6do/cc.wav?dl=0

https://www.dropbox.com/s/20qy8xsl91ymas8/ci.wav?dl=0

 

[March Audio]

Even though they do sound different, according to Toole.

Yes.

This lecture is very interesting, well worth a watch.

He specifically comments on stereo v mono testing.