What do you look for the most in stereo loud speakers?

[Mart68]

I understand your point, but I just don't really think of listening to music in those terms. It's all (or mostly) about the sound of musical instruments and voices to me.

But the sound of those things is heavily influenced by the frequency response of the speakers? For example a speaker with a mid-range lift can make a vocal or solo instrument sound hyper-real and 'live' but will render some recordings as shouty and unlistenable. For which the recording will often be blamed and the speaker praised as 'So revealing that it really shows up bad recordings.'

 

[Sal1950]

You mean what sounds 'right' on playback isn't necessarily what is accurate to the recording?

I can go with that, in that an inaccurate loudspeaker can sometimes sound more 'right' than an accurate one, for tone/timbre.

But I tend to avoid those speakers as they will throw up other problems depending on what you play through them.

Something I've gone round and round with Matt about for supporting the sound of inaccurate speakers/components.
Your basically talking about using an inaccurate speaker, tube amp, etc, to compensate for a poor recording.
That will work on occasion. but then all other (quality) recordings will be poorly represented.
Much better to have a accurate chain, and use tone controls, DSP, etc to do a little tweak when needed.
They're much easier to remove from the path.

 

[Mart68]

Something I've gone round and round with Matt about for supporting the sound of inaccurate speakers/components.
Your basically talking about using an inaccurate speaker, tube amp, etc, to compensate for a poor recording.
That will work on occasion. but then all other (quality) recordings will be poorly represented.
Much better to have a accurate chain, and use tone controls, DSP, etc to do a little tweak when needed.
They're much easier to remove from the path.

No, I'm not. I'm talking about perfectly good recordings being subjectively enhanced by inaccurate speakers. The recording must be simple - e.g acoustic guitar and vocal, nothing else in the mix. The sort of thing often played at shows to impress the rubes.

Lift up the mids, the singer will be right out front and '3 dimensional'. The guitar will have more body and impact. Compared to the accurate speaker system.

Then put some rock music through those speakers and you'll clear the room.

I agree with you that accurate speaker system (at least in the ballpark of accuracy) plus EQ where required, is the way to go.

 

[JohnnyAudio]

Magic that that I've never heard before.

 

[Duke]

Some like the illusion of phantom sources being localizable in the listening room ´just as if the musicians would be there´, detached from their own reverb on the recording or the reverb in the listening room dominating or being colorated.

What causes musicians to sound "detached from their own reverb"?

 

[Sal1950]

No, I'm not. I'm talking about perfectly good recordings being subjectively enhanced by inaccurate speakers. The recording must be simple - e.g acoustic guitar and vocal, nothing else in the mix. The sort of thing often played at shows to impress the rubes.

Sure, that too. Tons of poor measuring (and sounding to us) gear are sold to folks who enjoy an inaccurate presentation.
There's no accounting for bad taste.

 

[Mart68]

What causes musicians to sound "detached from their own reverb"?

Drugs?

 

[respice finem]

or the absence thereof

 

[MattHooper]

I do not really like this term personally, as high end folks seem to associate something completely different with ´holographic´. Some like the illusion of phantom sources being localizable in the listening room ´just as if the musicians would be there´, detached from their own reverb on the recording or the reverb in the listening room dominating or being colorated. I consider this to be against the intension of the recording engineer.

I don’t see it like that.

To me “ holographic” is a term that naturally comes to my mind when I experience certain loudspeakers that seemed to present Sonic
“ images” in a way that is more “3D” than other loudspeakers.

My MBL Omnis created this impression.

It had nothing to do though with the Phantom sources sounding “ detached from their reverb” but rather simply sounding more dimensional within the recorded space.

 

[Arindal]

What causes musicians to sound "detached from their own reverb"?

Mainly the reverb in the listening room showing significant differences to what our brain would expect from the given combination of sound source and reverb of the recording venue. Can be a different tonal balance of the additional reverb, contradictive angles from which the reflections are reaching our head, a reflectogram of early reflections not matching the concert hall or alike.

If you have the feeling that the voices/instruments of an acoustic recording are positioned ´in holographic manner´ in the listening room while the reverb is coming in mainly from the sides, the rear, appearing tonally colorated or anything like this, it is the phenomenon I described as ´detached reverb´.

loudspeakers that seemed to present Sonic
“ images” in a way that is more “3D” than other loudspeakers.

My MBL Omnis created this impression.

This ´3D´ coming from omnis is in most cases a dominating reverb pattern of the listening room overriding the original reverb on the recording. While our brain can make something out of it - a ´holographic picture´ of instruments in the listening room, it usually comes with compromised localization precision/stability, and has little to do with what is on the recording.

 

[Barrelhouse Solly]

What I think of as a "natural" sound. That's my subjective perception of flat frequency response and lack of distortion, I think. I've been surprised more than once to find that speakers that sound "natural" to me measure well. My audiogram proves that my hearing is well below average, so I'm never confident of my subjective impression. Attack is another area I listen for. It's what was listed under "bowing and tonguing" in the instrumental music section of my grade school report card. ('50s public school)

 

[MattHooper]

This ´3D´ coming from omnis is in most cases a dominating reverb pattern of the listening room overriding the original reverb on the recording. While our brain can make something out of it - a ´holographic picture´ of instruments in the listening room, it usually comes with compromised localization precision/stability, and has little to do with what is on the recording.

I disagree, at least in my experience.

As Toole points out, typically the recorded reverb or ambience cues in recordings dominate our perception even within the context of room reflections.

As I mentioned, in my room the Omni maintained good localization - everything was in the same place as with the imaging any other speakers. And the Omni is also preserved the recorded acoustic, so that the acoustic space created around the Omni was chameleon like with each record recording.
However within the recorded space, the Sonic images seemed more dimensional.
To slightly exaggerate to make the point: by comparison regular speakers tended to produce images with an acoustic space in which the individual images seemed a bit two-dimensionally placed in that space.
With the omnis there was more of a “rounded” impression, as if the space around the instruments continued right around and behind the instruments making them more dimensional (though not “ detached” from the acoustic of the recording). (and as I mentioned since I’m fascinated by live versus reproduce the sound I pay a lot of attention, eyes closed, to the impression of real sounds in real acoustic spaces, and for me the Omni did that the best, as well as doing that the best when I did live instrument versus reproduce comparisons through those speakers and others).

Those are my impressions anyway so I’m just saying that they don’t seem to track exactly what you are saying.

 

[Sal1950]

What I think of as a "natural" sound. That's my subjective perception of flat frequency response and lack of distortion, I think. I've been surprised more than once to find that speakers that sound "natural" to me measure well.

Don't be surprised, that was the general result of Harman's listening tests.
If you haven't already pick up a copy of Floyd Toole's book.
There so much solid info in it on the "Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms"

 

[Jim Creek]

For me it’s all about imaging imaging imaging.

 

[kemmler3D]

I still don't really think "imaging" is a great choice for "attribute of a speaker" because it's not a definite, measurable characteristic of one, and it depends a great deal on the room. Beamwidth, DI slope, cardioid, smoothness of dispersion, etc. - all things we can seek out directly that affect imaging, but not imaging itself.

 

[Sal1950]

For me it’s all about imaging imaging imaging.

If a immersive listening experience is what you really love. go multich.
At least to a 7.2.4 level or better.

 

[Arindal]

As Toole points out, typically the recorded reverb or ambience cues in recordings dominate our perception even within the context of room reflections.

I have been disagreeing with Dr. Toole on that matter, at least for listening setups in which the additional reverb in the listening room is dominating by level, altered by tonal balance, angles from which it is coming in and reflectogram. I see no evidence why the reverb cues on the recording should dominate if the (unwanted) listening room reverb is louder, colored, coming in earlier and from different angles than the original reverb in the concert hall.

I would rather assume that the ´holographic´ impression many omnis are producing, is indicative of a total dominance of listening room reverb, as it gives our brain at least a coherent picture of the reverb. What might also contribute to this, is omnis usually producing a tonally balanced indirect soundfield, so again our brain is perceiving this as coherent with the expected tonality of the reverb created by the direct sound.

This all does not apply to well-treated rooms such as studio control rooms or damped testing facilities, though.

in my room the Omni maintained good localization - everything was in the same place as with the imaging any other speakers.

It would be interesting to conduct a listening test comparing phantom center vs. real center, using dedicated mono mixes on an ideal center speaker in terms of localization width (a small coaxial would do), comparing it with the phantom center created by omnis in terms of localization width, stability and proximity. I would expect the omnis to show significantly degraded localization stability and extended or blurred width thereof.

 

[rdenney]

The assumption of ASR-pleasing directivity is the most baffling one for me--that's one assumption that hardly seems justified by the products on the market.

I'm trying to remember who it was that said in a small room, we are listening to a speaker's directivity pattern more than to the speaker itself. Yes, we want flat on-axis performance simply because that's the easiest starting point. But Toole tells us we can equalize the anechoic frequency response. What we can't do, however, is separately equalize the off-axis response differently than the on-axis response, and yet that will be determinative in small rooms. If there's one thing Toole's research has made clear, it's that.

He also says "people like bass"--that is a preference response, not necessarily an accuracy response, even with qualified listeners if they are giving their preferences. I gathered from his description that if a speaker enhanced the bass response somewhat relative to higher frequencies, that would achieve higher preference scores. But it is a move to less fidelity by any measure, and one that might well be exacerbated by room modes (see above paragraph).

To me, a disconnected reverberation tail is how someone might describe a lot of ringing in the bass but very little in the higher bands. I call it boomy bass. I have experienced this in two ways: 1.) I listened to a pair of Martin Logan speakers at a Best Buy a few years ago. Boom-boom-boom-boom plus tinny-tinny-tinny-tinny. The boomy built-in subwoofer in the base unit sound like it was from a different recording than the hyper-clear (read: No reverberation and directivity from a different planet than the woofer). But maybe Best Buy had cranked up the bass to match that preference for more bass.

2.) I just bought a Dayton 12" sub to provide something resembling bass with a pair of Canton GL260 bookshelf speakers that have 6" "woofers". It works pretty well, actually. But with the crossover set low (60-80 Hz) there was a noticeable perception gap between the sub and the speakers. When I turned the crossover up to its highest setting (180? I'm going from memory), it filled in that gap. But when it was loud enough to make 80-100 Hz really sound like "big bass", the rumble and modes in my shop area made the air vibrate below 30 Hz in very unpleasing ways. (The maximum sound length in that space is 60 feet.) When I turned the sub down to the point where I couldn't clearly hear it in the mix, that effect went away. I was driving the high-level sub inputs from the Speaker B outputs of my little Kenwood integrated amp, so I could switch the sub in and out instantly. When I did that, all perception of bass vanished and I realized how much the sub was contributing to the sound. But It was crossed over higher than I would have expected and at a level much lower than most might expect. I don't have much experience with subs and this changed my perception of them.

My conclusion is that flat on-axis response, which above the Schroder frequency can be achieve using EQ, is a requirement for fidelity. But so also is producing relatively flat frequency response curves off-axis, even with the spectral tilt that comes from beaming, and that is not something a lot of speakers achieve successfully. The off-axis performance may have a bigger effect than the on-axis performance in a lot of rooms, if it is sufficiently different from the on-axis sound.

Rick "there's more to fidelity than just flat response on axis" Denney

 

[PristineSound]

5 pages in and over 80 votes later. Spatial performance and low distortion is the most important by a large margin (that's after linear FR and controlled directivity as stated in post #1).

This is inline with my expectations, no surprises here.

The one pleasant surprise is sensitivity, only 1 vote; I would have expected a significant amount of people voting for this because sensitivity comes up often in discussions. But I am glad that people realized that in modern day speaker with the available cheap amplification, sensitivity is of very little concern.

 

[Snarfie]

No, I'm not. I'm talking about perfectly good recordings being subjectively enhanced by inaccurate speakers. The recording must be simple - e.g acoustic guitar and vocal, nothing else in the mix. The sort of thing often played at shows to impress the rubes.

Lift up the mids, the singer will be right out front and '3 dimensional'. The guitar will have more body and impact. Compared to the accurate speaker system.

Then put some rock music through those speakers and you'll clear the room.

I agree with you that accurate speaker system (at least in the ballpark of accuracy) plus EQ where required, is the way to go.

Alternative Take: What Do You Look for the Most in a Song/Track — Not Just the Loudspeakers?

Probably a bit off topic

The original question asks what we value most in stereo loudspeakers — but that got me thinking: shouldn't we also ask what we value most in the music itself?

After all, the way a track is mixed and mastered determines how reverb, imaging, dynamics, and overall tone are "locked in" — regardless of what speakers you play it through. We’re listening to a finished product shaped by the creative and technical decisions of the mix engineer.

So maybe the more revealing question is:

"What do you look for the most in a song or track — instead of just in loudspeakers?"

For example, I've been remixing multitrack stems for personal enjoyment in my DAW. Recently I made a stripped-down version of Nirvana – In Bloom. I removed the wall-of-sound compression (for a better expression ) focused on drums ( Toms kick drum minimale highats) bass, Curts voice, and mixed everything in mono — no panning, no polish, just intimacy and balance. It’s far from the original, but it resonates more with me.

This leads me to wonder: how long before AI can remix or even produce music entirely based on our individual preferences? What would that mean for listening culture — and for how we evaluate playback gear so for instance you have monitors that lack some bass you can individually enhance that in the bass track. Basicly the possibilities are endless.

Curious to hear how others approach this — do you think more about the sound system, or the sound design of the music itself?