Speakers: Perceptions, Descriptions, and Tastes

[Shadow_Ferret]

As I've participated on this board and elsewhere, and read reviews, I hear phrases like "good for rock" or "better for jazz." I've heard descriptives like warm, overly bright, boomy, laid back and so on.

And every response, rightly or wrongly, to the inevitable newbie question of which of these speakers is better is, "Listen and decide, we can't decide for you."

It seems that every speaker is a compromise and that's probably just the nature of the beast.

But it also strikes me that a piano only sounds one way. An acoustic guitar only has one sound. Violins have a definate sound. A trumpet is a trumpet is a trumpet.

Forget "bright" or "a good rock speaker" for the moment. Forget perferences or prejudices.

Do any of the speakers we've been talking about recreate the sound of a piano so that you actually believe there's a piano in the room with you?

Because if there is, if that speaker exists (and if I can afford it), THAT is the speaker I want (even if it's a Bose and I have to give up my Audioholics membership).

However, I don't believe such a speaker exists or we'd all have it, right? Why would any of us choose speakers that color or exaggerate certain frequencies if this one perfect speaker did exist?

I mean, if a person knows what a piano sounds like, and most of us do, why wouldn't they get the speaker that accurately recreates the sound of a piano? Or do we carry our own prejudices with us that color our judgement? Do we know what a piano sounds like but when we listen to a speaker if this one has a heavier mid-bass that we find satisfying, do we choose that one even if a piano DOESN'T sound that way? If we like "crisper" high frequencies will we choose say a horn tweeter because it makes the top notes of the piano pop more even if it isn't a true representation?

In other words, do we deliberately (yet subconsciously) NOT choose the most accurate speaker but the ones that are better suited to our own personal tastes in regard to specific sounds, speakers that lean toward boomier bass or over-emphasized mid-ranges or piercing highs?

There'll probably be a few who now yell, "Well, my Blue Squeekers don't color anything! They reproduce things just fine!" and yet for every fan of those Blue Squeekers, there'll be another who says they can't quite warm up to their sound, the highs are too hot and they'll trumpet their Red Boomers while another complains the Red Boomers have an artificial bass boost.

Personal preference. But why is that?

A piano is a piano is a piano, is it not?

 

[WmAx]

A
But it also strikes me that a piano only sounds one way. An acoustic guitar only has one sound. Violins have a definate sound. A trumpet is a trumpet is a trumpet.

Forget "bright" or "a good rock speaker" for the moment. Forget perferences or prejudices.

Do any of the speakers we've been talking about recreate the sound of a piano so that you actually believe there's a piano in the room with you?

In theory, a single speaker would be ideal, regardless of the genre. Heck, it's even demonstrated[with very high statistical probability] in controlled listening tests in blinded conditions that a narrow set of speakers with nearly identical response parameters are chosen in controlled tests by subjects. However, I believe the problem lies in, primarily, the absolute lack of controls/standards in the recording/mixing/mastering processes. Do you realize that there are countless ways the pianos are miced, for example? Each one will produce a different sound/spatial effect. How about that most mics used in the studio are not accurate[having distinct significant upper midrange and treble colorations] in the first place? How about the recording distance/environments? How about effects/e.q. in the mixing stage(s)?

How about the room effects on speaker sound? Just another pebble in the bucket..... How often will you be able to get the SAME speaker to sound the same in a DIFFERENT room?

How can you expect have an accurate representation of the original acoustic source considering the above issues?

-Chris

 

[STRONGBADF1]

The dream

The real problem is that there is no one standardized way to record that piano. (room size, mics, mixes, compression...)

people who record rock have different ideas then people who record jazz...classical...whatever...

once we solve that then we can build the perfect speakers

 

[STRONGBADF1]

In theory, a single speaker would be ideal, regardless of the genre. Heck, it's even demonstrated[with very high statistical probability] in controlled listening tests in blinded conditions that a narrow set of speakers with nearly identical response parameters are chosen in controlled tests by subjects. However, I believe the problem lies in, primarily, the absolute lack of controls/standards in the recording/mixing/mastering processes. Do you realize that there are countless ways the pianos are miced, for example? Each one will produce a different sound/spatial effect. How about that most mics used in the studio are not accurate[having distinct significant upper midrange and treble colorations] in the first place? How about the recording distance/environments? How about effects/e.q. in the mixing stage(s)?

How about the room effects on speaker sound? Just another pebble in the bucket..... How often will you be able to get the SAME speaker to sound the same in a DIFFERENT room?

How can you expect have an accurate representation of the original acoustic source considering the above issues?

-Chris

JINX!!!!!!!!!

I guess we were authoring at the same time.

 

[miklorsmith]

Good question!!

Tough answer. Essentially, "chasing the piano", as it were, is the preeminent goal of any speaker designer, or the maker of any audio equipment. Of course, it's much more complicated, as speakers must be able to accurately track tone and dynamics of many instruments at once. Some people prefer lots of bass and others can live without it if they can get great midrange. Every speaker is compromised from perfect and the choosing of imperfections and character is where the personal elements of selection arise.

A tonally correct middle-C on the piano requires very different resources from a speaker than convincing reproduction of a kick drum. Speakers that can do both very well tend to be large and very expensive. So, most of us choose speakers that compliment our listening preferences.

Why don't we all flock to speakers that can play piano well? The piano is an excellent example of the problems speaker manufacturers face. The left-hand keys begin at around 30 hz and the right side goes up to about 4,000 hz. In a standard, 3-way speaker that means two separate crossover points where drivers are being blended. A piano is obviously "voiced" to reproduce each note with a common character. The overall shape, size, and materials are chosen specifically with this in mind. Look at how big the piano has to be to accomplish its function.

Now look at the loudspeaker. Some signals are reproduced entirely at the woofer, some at the midrange, and others at the tweeter. And, there are many frequencies which must use a combination of drivers. Seamless integration, as with a piano, is very difficult to achieve. And, this is assuming a piano is the only thing the speaker has to sound like, which couldn't be less true.

My personal opinion is that these very crossovers kill much of the dynamics and life of the piano (or voice, or whatever).

This difficulty is one reason why a true "full-range driver" has been pursued as ideal. However, no truly full-ranger has ever been produced. Of the ones that have tried, most have significant frequency response issues and are very light in the bass. Lowther enthusiasts, for instance, embrace incredible transient speed and midrange purity which make conventional speakers sound slovenly and "dead". But, they also live with peaky mid/treble response and limited bass response. Lowthers sound unreal with Patricia Barber but would cry like a baby with Metallica.

There is no perfect speaker. Figure out what you want it to sound like and choose strengths and weaknesses accordingly.

 

[j_garcia]

Speaker design is all about tradeoffs. You are hearing what the speaker's engineer(s) consider to be good sound. Everyone has a personal opinion of the type of sound they like to hear, which means they have a preference for a certain speaker or speaker type.

You are correct, no such speaker exists. EVERY speaker colors the sound to some extent, just some less than others, so it can be tough to find one that you like. I found mine. Patricia Barber sounds excellent on them, and Metallica still sounds like Metallica, but not quite showing off what the speakers can do. Jazz, techno, vocals are very well suited to my speakers, and that is exactly why I chose them.

Don't forget the room. The room has a HUGE effect on how a particular speaker sounds within it.

Contact the manufacturer of my speakers. He has some demo pairs that travel around and they're worth a listen.

 

[Steve1000]

Great question! You are really getting at the crux of this whole business.

But I wouldn't even go as far as you do in terms of finding a constant in chasing the elusive "right" sound. Try playing an acoustic guitar in your bathroom and then in the woods and tell me it sounds the same. Or listen to a trumpet in a jazz club and then in a concert hall. Or listen to a violin in a small room and then in a concert hall. Or listen to McCoy Tyner play the piano and then listen to Keith Jarret play the same piano (if you could be so lucky). And here we're just talking about one instrument at a time. Between the performance setting, the artist, the recording process, the mixing process, the listening room for the speakers, etc., the variables become mind-numbing.

Then, as WmAx has pointed out in another thread, in loudspeakers people apparently tend to prefer a little treble roll-off, though there's not a consenses on the degree of treble roll-off. I have a feeling I prefer less as opposed to more treble roll-off, though I surmise it varies from person to person.

For what is reputedly a pretty good playback device for accurate tonal balance, get yourself a pair of Sony V6 headphones. Much modern recorded music will sound quite bright on it.

And just to put things into perspective a little more, WmAx has mentioned a study where people were asked to criticize loudspeakers behind a curtain. People found much to criticize, but, alas, it was a live performance.

I read mags and get advice from just a few folks whose opinions I respect (such as WmAx) and buy equipment that is objectively supposed to be pretty good. I've become quite skeptical of spending megabucks, in light of how elusive the goal really is. Then I adjust for personal taste with EQ, positioning, etc., and generally I am very satisfied and content in the end. If I am really digging the sound, and it's objectively pretty good, that is good enough for me. I consider my pursuit of the hobby successful.

But it also strikes me that a piano only sounds one way. An acoustic guitar only has one sound. Violins have a definate sound. A trumpet is a trumpet is a trumpet.

Forget "bright" or "a good rock speaker" for the moment. Forget perferences or prejudices.

Do any of the speakers we've been talking about recreate the sound of a piano so that you actually believe there's a piano in the room with you?

...

A piano is a piano is a piano, is it not?

 

[Nick250]

Not to mention that our hearing changes as we age and the frequencies you hear and what I hear mostly likely differ.

 

[pikers]

I think people have resigned themselves to the fact that no recording scenario sounds the same; hell, no two songs sound the same on any album.

Video tends to be a little more objective; audio is not only subjective, but far more personal, because unlike video it appeals directly to emotion.

Listen to music you like on speakers you audition, and decide based on that. Be happy!

 

[The Dukester]

Do the speakers speak

[

A piano is a piano is a piano, is it not?[/QUOTE]

I would say no. Does a baby grand or the old upright in the choir room sound the same as a grand? Of course not. As for the guitars, does a Yamaha FG75 sound the same as an Ovation or Martin D4? No.
The makers of all these insrutments make them to sound the way they like (for a certain price point) and we buy them accordingly. Likewise, to an extent with speakers. Since the single all accurate driver has not been created yet, there has to be some kind of tradeoff, IMO. Music to one persons ears is noise to another, so we tend to buy what we like.
It's the same with most anything. Harleys sound great to some and others hate them, etc.
All the above posts are correct in stating the various reasons speakers don't sound like a live instrument such as micing, room acoustics, etc. Even the one where an insrument sounds different in diff environments/rooms. A totaly accurate speaker would deliver the sound just as the engineer recorded it to sound. Unless you were that engineer or were there when he recorded it, it would be tough to say whether or not it was correct Electric instruments present even a tougher problem in that they sound different live according to the settings on the sound board or amp, etc. So what does a guitar, piano, etc sound like?
We take our speakers like we take our coffee. Purists (like me!) take it black (or accurate). Now, is that Columbian or Central American? Dark roast or espresso? Some prefer to color their coffee, just like some like colored sound.
I think we buy speakers based on the sound we like, the style music we like and what we perceive to be accurate based on the live music we hear. Try your speakers just like you try your coffee. Find the flavor you like and go for it! Don't worry about whether someone thinks they sound mushy, boomy, or tinny. Opinions are like a certain part of our anatomy...everybody has one. I just gave you mine FWIW!

BTW...don't forget to get the really accurate speakers that faithfully reproduce the piano, you have to buy the $2k speaker wires to go with them.

 

[Pyrrho]

As I've participated on this board and elsewhere, and read reviews, I hear phrases like "good for rock" or "better for jazz." I've heard descriptives like warm, overly bright, boomy, laid back and so on.

And every response, rightly or wrongly, to the inevitable newbie question of which of these speakers is better is, "Listen and decide, we can't decide for you."

It seems that every speaker is a compromise and that's probably just the nature of the beast.

But it also strikes me that a piano only sounds one way. An acoustic guitar only has one sound. Violins have a definate sound. A trumpet is a trumpet is a trumpet.

Forget "bright" or "a good rock speaker" for the moment. Forget perferences or prejudices.

Do any of the speakers we've been talking about recreate the sound of a piano so that you actually believe there's a piano in the room with you?

Because if there is, if that speaker exists (and if I can afford it), THAT is the speaker I want (even if it's a Bose and I have to give up my Audioholics membership).

However, I don't believe such a speaker exists or we'd all have it, right? Why would any of us choose speakers that color or exaggerate certain frequencies if this one perfect speaker did exist?

Are you new to audio? Of course people pick things that exaggerate certain frequencies! People often pick what is pleasing to them, not what sounds most like something real. After all, most people don't listen to anything real; they only listen to recordings.

I mean, if a person knows what a piano sounds like, and most of us do, why wouldn't they get the speaker that accurately recreates the sound of a piano?

Most people do not know what a piano sounds like (and different ones, or the same one in different rooms, sound different to each other anyway). Most people do not listen to live music, or if they do, they listen to live music that is amplified and run through speakers. And most recorded music is not a "record" of an event that took place (the original idea behind recording), but a creation in its own right (a great example is the Beatle's Sgt. Pepper's Lonely Hearts Club Band album).

Since rock music is always run through speakers, speakers that resemble the ones used in rock venues have the "right" properties for such music, which explains why Magnepan speakers are often not regarded as best for rock. Accuracy in reproduction of the electrical signal is not what is desired; it must be given that "speaker" sound to sound like what one would hear live.

Additionally, in most cases, absolutely no effort is made to objectively determine whether or not the speaker accurately reproduces the sound. To be fair, it would be difficult for an individual to do satisfactory testing, but most people seem to be completely unaware of this shortcoming in their selection process.

Or do we carry our own prejudices with us that color our judgement? Do we know what a piano sounds like but when we listen to a speaker if this one has a heavier mid-bass that we find satisfying, do we choose that one even if a piano DOESN'T sound that way? If we like "crisper" high frequencies will we choose say a horn tweeter because it makes the top notes of the piano pop more even if it isn't a true representation?

People absolutely do this (except they don't know what a piano sounds like). This is the subjectivist way of selecting: Whatever sounds best to you is best. (Except, of course, when someone else picks something that does not meet with your approval, such as Bose, in which case the person obviously has something wrong with them or they are an idiot, and must be denounced and berated in most uncertain terms on audio web sites.)

In other words, do we deliberately (yet subconsciously) NOT choose the most accurate speaker but the ones that are better suited to our own personal tastes in regard to specific sounds, speakers that lean toward boomier bass or over-emphasized mid-ranges or piercing highs?

There'll probably be a few who now yell, "Well, my Blue Squeekers don't color anything! They reproduce things just fine!" and yet for every fan of those Blue Squeekers, there'll be another who says they can't quite warm up to their sound, the highs are too hot and they'll trumpet their Red Boomers while another complains the Red Boomers have an artificial bass boost.

Personal preference. But why is that?

A piano is a piano is a piano, is it not?

It is not.

 

[Tomorrow]

A piano is a piano is a piano, is it not?

It is not! A Steinway Grand is much different sounding than a Kawai upright. An Ovation acoustic guitar sounds nothing like Gibson acoustic guitar. In fact, I'll go so far as to say no two of the same instruments sound exactly alike, though obviously more similar than between brands. You ever hear anyone play a trumpet that 'sounded' like Louie Armstrong? The musician can easily color the sound of any instrument.

Then you have the issue of the recording studio and the ambience that it features. Each studio will flavor music to some extent.

Perhaps even more relevent than the studio is the recording engineer who will flavor sounds according to his or her prejudices of what sounds good.

While you were taking a good shot at trying to normalize sounds played by speakers, this isn't the path. Unfortunate, but true. Finally, we color a speaker's identity/sound/value with yet a different set of values. It's amazing we all don't have different speakers, lol.

 

[Geno]

I've been going through this very process for almost 40 years, and I've probably got my last ( I'm kidding myself) set of speakers. I auditioned at least a dozen sets at a price point that I felt was a good "band for buck" and , for my personal tastes, looked good in my home...( wanted nice hardwood cabinets this time out; no more black vinyl). There were a lot of very nice choices out there, and I'm sure I could've been just as happy with many other choices, but I had to make a decision. I've got what I consider to be a very satisfying listening experience now, and am happy with my choice.

 

[GregorB]

Back in the mid-1960s Acoustic Research did some demonstrations to prove the accuracy of their speakers. They set up their AR-3a's next to a string quartet playing outdoors, and switched between the live quartet and a recording of them made at the same location. Listeners were asked to identify when the music changed between live and reproduced. The result were relatively conclusive in showing that the speakers did not introduce noticable coloration to the sound and were essentially impossible to detect as a separate sound source from the original.

IMO this is what a truly accurate speaker should do - accurately re-create the sound of the original in the same space. Technically this type of speaker would place a live performance in your listening room, provided that the speakers and their amplification had the dynamic range to keep up with the material.

Which brings up another area of concern: amplification. It's pretty obvious that in most systems the speakers are woefully underpowered. 60-70-100 wpc is simply not enough power to drive any speaker to even semi-realistic levels without compression and clipping. So an accurate reproduction of the recorded waveform is simply impossible no matter how accurate the speaker may be.

All this said, I think a proper combination of speaker, amplifier, and room acoustics can conceivably bring the listener to the recording venue (acoustic recordings only need apply). Bringing the artists to your room is a much more elusive goal.

But I do not believe, and never will, that accuracy in a given space is subjective. Given the time, effort, patience, and resources, almost anyone can audition enough speaker options to find one that provides a convincingly real simulation of reality in their home. But how many do?

Greg

 

[Rip Van Woofer]

The best possible speaker (according to the consensus of audio engineers AFAIK) will have:

• Flat frequency response
• Wide frequency response covering the entire audio band
• Low distortion (below the threshold of hearing at all frequencies)
• Wide, uniform polar response

The rub, of course, is that complete, valid measurements of those are beyond the ability of most laypeople and, for that matter, beyond the ability of most audio reviewers (assuming they even accept the validity of measurements).

The best, and admittedly still imperfect, solution for the lay audiophile (to boil down the essence of the well-informed posts above) is to become as familiar as possible with natural, acoustic, non-amplified sounds and to judge how well they reproduce well-recorded unamplified sounds.

Ah...but "well recorded": how do you recognize that? Well, in my experience even imperfect speakers (the ones that aren't total crap, anyway) will sufficiently reveal the good from bad recordings once you have "calibrated" your ears to natural sound.

As a rule of thumb, if a speaker "grabs" me immediately I am suspicious. Chances are something was "goosed" to make that initial, "Technicolor" sound impression. Ultimately whatever was manipulated ("crisp" or accented highs, "slam" in the bass) will become false and possibly even fatiguing at realistic levels. Real life sound is subtle and does not draw attention to itself. I chose the Technicolor metaphor quite deliberately: real sound is to artificially "good" sound as the overly saturated Technicolor* film image is to what we see through our eyes everyday in the real world.

A bit of personal experience may illustrate this: In my younger days as both a budding audiophile and classical music fan, I used to wonder why the bass and highs of a real orchestra didn't have the same "oomph" and "presence" of my stereo! It gradually dawned on me....

------

*That metaphor, come to think of it, might be a bit dated. I'm thinking especially of 1950's and '60s Technicolor flix. Another comparison I've read is the difference between Ektachrome (more artificial) and Kodachrome (more natural) slide film...but then that only makes sense to photo geeks from the pre-digital era. Sigh...it's hell getting old....

Anyway, check out a DVD of something like the old "Spartacus" with Kirk Douglas to see what I mean by Technicolor...you young punk whippersnappers!

 

[tbewick]

Speakers are transducers, and as such can distort sound in and extremely complex way. There are many ways of measuring speaker distortion, and there is a link that I have recommended before on the Stereophile website that was quite informative:

http://stereophile.com/features/99

I can't really agree with what I feel Chris is implying, that the recording industry should be standardised. I feel that this would stifle innovation.

A lot of this way of looking at things I feel is wrong. You shouldn't concentrate on "does this piano sound right or not?" when listening to the music. Surely the main pleasure should be derived from the quality of the music itself.

I believe a point earlier was made about, I'm guessing, electrostatic speakers having limited bass response. This is a result of their directionality of sound reproduction, in that, unlike ordinary speakers, bass frequencies are much less strongly emitted from the back of the speaker enclosure, thus the speaker does not excite room nodes as strongly as conventional designs. Electrostatic speakers can be very transparent in their reproduction (eg. the Quad electrostatic).

 

[majorloser]

No musical instrument will sound today exactly like it sounded yesterday. It will not sound the same in your house vs. mine. A speaker may be as close to accoustic perfection as possible in a lab but put it in my house and you will probably have crap (especially if I set it up).

No human ear is capable of receiving a sound wave exactly like another. The physical shape of your ear vs. mine influence our ability to perceive sounds. And of course I probably damage my ears from lack of protection for many years. I'm quite sure my Cantons sound like crap to quite a few people. I'll still recomend them as something to compare when a newbie is looking.

As you stated, all I can ask for is that the sounds of that piano I heard in a live performance are replicated as close as possible in my home. And yet my memory will fade from the time I hear it in the concert hall to I get home.

 

[WmAx]

Back in the mid-1960s Acoustic Research did some demonstrations to prove the accuracy of their speakers. They set up their AR-3a's next to a string quartet playing outdoors, and switched between the live quartet and a recording of them made at the same location. Listeners were asked to identify when the music changed between live and reproduced. The result were relatively conclusive in showing that the speakers did not introduce noticable coloration to the sound and were essentially impossible to detect as a separate sound source from the original.

Dunlavy Labs did similar things. However, this is not represenative of accuracy except in the conditions as used [ basicly no environmental acoustics to interact with the speaker or original sound sources, thus only direct arrival sound, which is easy to record and playback accurately as long as the first condition is preserved ] for the experiments.

IMO this is what a truly accurate speaker should do - accurately re-create the sound of the original in the same space. Technically this type of speaker would place a live performance in your listening room, provided that the speakers and their amplification had the dynamic range to keep up with the material.

Room interactions and no standards for recording/mixing/mastering make such accuracy impossible.

Which brings up another area of concern: amplification. It's pretty obvious that in most systems the speakers are woefully underpowered. 60-70-100 wpc is simply not enough power to drive any speaker to even semi-realistic levels without compression and clipping. So an accurate reproduction of the recorded waveform is simply impossible no matter how accurate the speaker may be.

You can't generalize like this; you have to quote specific power in relation to a specific sensitivity, room gain characteristics and listening distance from speaker. Also, live unamplified music is not as loud as many seem to imagine it is, unless they are imagining that they are sitting beside a loud instrument. However, from an audience perspective, SPLs that can not be reproduced by a reasonable quality full size/range hi-fi speaker are not a common occurence.

But I do not believe, and never will, that accuracy in a given space is subjective.

True. However, most people don't know what live unamplified music sounds like; to realize such at a reasonable level requires specific effort such as listening in various venue and of various sound sources[and with eyes closed as to not let visual cues further distort the memory/experience ]. Also, we come full circle to the room acoustics/recording standards issue(s); the best one can do in real situations is find a speaker/room set up that sounds as accurate as possible on non-standardized recordings based on subjective interpretations of what accuracy should sound like compared with a lengthy time delayed memory of a different acoustic event entirely...

-Chris

 

[WmAx]

Speakers are transducers, and as such can distort sound in and extremely complex way. There are many ways of measuring speaker distortion, and there is a link that I have recommended before on the Stereophile website that was quite informative:

http://stereophile.com/features/99

Stereophile is not an authority on what is important concerning loudspeaker reproduction. I will refer people to real authorities, such as the NRC and Harman Kardon, which have produced conclusive scientific evidence correlating specific measured parameters with listener hearing.

I can't really agree with what I feel Chris is implying, that the recording industry should be standardised. I feel that this would stifle innovation.

If you want to ever be able to reproduce the sound exactly as it sounded in the original performance, such standardization is not optional[and realize that loudspeaker and room interactions would have be standardized as well]. However, I don't intend that this should be applied to all recordings. I believe it should be a certification that some audiophile labels would adhere to for certain types of recordings, such as classical and other unamplified type acoustic performances, as an example.

-Chris

 

[bpape]

I think there's a difference between being tonally accurate (reproducing the piano and violin) and parameters that affect things like high pressure levels (rock), imaging (NOT rock), etc.

For instance, a pair of Quad ESL 63's driven by a Berning EA-230 will do a STUNNING job of reproducing the piano, violin, and trumpet, etc. It will image like mad on Jazz, Blues, Orchestral, etc.

Would I choose that combination to use primarily for Rock? No. Simply doesn't play loud enough.

Is there a 'perfect' speaker? No. Are there some that come awfully close? Yup. The kicker has been mentioned though - VERY few can afford them - much less the associated equipment to allow them to perform to their fullest.