90% of Reproduced Sound Quality Comes From . . .

tmallin

WBF Technical Expert
May 19, 2010
966
375
1,625
71
Chicagoland
I think that 90% of the total package of sound quality one gets from a two-channel audio system depends on:

1. the listening room (size, shape, construction, etc.);
2. the speakers;
3. the placement of the speakers and listener with respect to each other and room boundaries; and
4. the acoustical treatment of the listening room.

I'm even pretty sure that these factors are listed in the order of their contribution to that 90%.

Differences among all the rest--amp, preamp, sources, cables, and various electrical and vibration isolation tweaks--while audible (we think), are truly swamped in importance by these four. Audiophiles can still hear the effects of these things even when those four are not excellent, but anyone who says that the difference between one amp and another is more significant to the illusion of reality than moving the speakers a few inches is kidding themselves. A non-audiophile will always hear the effects of these four things (or at least the first three), but will often strain to hear the effect of any other change which may be "perfectly obvious" to the audiophile.

Many, if not most, audiophiles do not approach excellence in these four areas. This is not necessarily or even usually out of ignorance. If you lack an understanding family, 1., 3., and 4. cannot be properly dealt with.

Note that the most important factor, the purpose-built room itself, is the most expensive and most difficult to get cooperation with. Proper speakers are also expensive and thus difficult to swing both financially and domestic-acceptance wise. But try to throw the decor issues involved in 3. and 4. into the mix and you either need a VERY understanding family, or you are back to the expense of a dedicated listening room.

Many audiophiles don't even know, or have forgotten, how important these four factors are, having somewhere along the line bought into the Linn "source uber alles" philosophy.

As to the speakers, you hear comments about how "speakers are difficult." Well, yes, they are difficult simply because even today there are night-and-day differences between competing products and you have to decide which to build your sound around. Because they are so "difficult," many audiophiles refuse to spend more than $3,000 (new price) or so on a pair of speakers, knowing that next year or next month they will want a "new sound."

That means that for most audiophiles, upgrading which does not involve speaker replacement takes place at the level of the remaining 10% of possible sound quality, for which vast sums are expended. Look at how many different speakers are sold at less than $3,000 new retail, and how much activity there is in the used market for speakers at that price level and all the rest of the gear. The problem is, every time the system's speakers are changed, what ensues is a valiant, if ultimately doomed, effort to compensate for the, say, 20% contributed by the speakers with electronics, sources, and other stuff which can only control 10%.

These realities have led some audiophiles to gravitate toward speakers which are maximally forgiving of the room around them, so as to minimize the effects of factors 1., 3., and 4. while still being subjectively excellent in terms frequency response, distortion, dynamics, etc. Such speakers should also be designed so that they sound excellent regardless of what is feeding them, how they are placed in a room, where you listen from in the room, how the room is constructed, and the nature of the room treatment, if any. Those factors are still quite audible with such "room ignoring" speakers, but they can not so easily sabotage the inherent excellence of the speaker as they can with other speakers.

Now, the problem is that there is little agreement in the audio community about how to design a speaker to maximally ignore the room and its set up and still sound inherently excellent. The views include, but are not limited to, speakers which are:

a. omnidirectional over their full range (e.g., MBL), since all the reflected sound will tend to have a similar frequency response to the direct sound from the speakers;
b. wide-dispersion (but not omndirectonal) up as high in frequency as possible, usually having narrow anti-diffraction baffles;
c. dipole panels (e.g., planar magnetics, quasi-ribbons, and electrostats) since they put out little sound in the up-and-down and side-to-side directions, thus being less influenced by room modes and reflective room surfaces;
d. increasingly directional in their output starting at a quite low frequency, such as the Gedlee Summa;
e. narrow or at least very controlled in dispersion from top to bottom, such as the Legacy Audio Whisper, the Gradients, and various horn models--all very different approaches to the same narrow/controlled-dispersion goal;
f. constant in directivity up to a 30-degree angle or so up to about 4 kHz, with response rolling off even on axis above that frequency and with increasing directivity above that frequency, e.g., the classic Acoustic Research (AR) speakers such as the AR-3a;
g. constant in directivity up to a 30-degree angle or so up to about 4 kHz, and flat on axis above that frequency but with increasing directivity above that frequency, e.g., the JBL Pro LSR 6332; such speakers necessarily having a fairly wide front baffle.

I know which approaches I favor for those who cannot control room acoustics and set up: e., f., and perhaps g. REG currently believes g. to be the best design approach, but admits it would require room treatment of some sort, or at least a very "softly" furnished room for best results. For those who can't change their room decor from fairly "live"/reflective to "soft," e. and f. might be better choices.

Actually, I favor approaches e., f., g. even in a totally dedicated listening room. In my experience, such speakers are much easier to get to sound truly excellent in even a dedicated audio room without heroic room treatments.
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC

FrantzM

Member Sponsor & WBF Founding Member
Apr 20, 2010
6,455
29
405
Hi

I hadn't read this thread yet.. My personal percentage was a tongue in cheek 87% ... and it can be proven with DBT as the most tin-eared person usually don't fail on DBT when it comes to speakers ...

I agree with Tom 97% :)

Frantz
 

FrantzM

Member Sponsor & WBF Founding Member
Apr 20, 2010
6,455
29
405
Tom

This is one of the first mention of Gedded Summa I have seen in an audiophile Forum... I am extremely intrigued, call it interest in his approach and am even looking to go audition his speakers. What is your take on his speakers, Have you heard them? If this is OT we can goon PM..
Frantz
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
Like the attempted steroid DBT if the effects are that dramatic with speakers no need "to go blind" so to speak. DBT teasing aside. I have said in the past that you are listening to your speakers. Furthermore others have convinced you are listening to your speakers in your room. I am not convinced you need to build your own room. Some older buildings were built very well. the facts remains that some excellent speakers have sounded like crap with a specific supporting cast. We can argue ad infinitum the cause of this. Some speakers present an extremely difficult load. Just for kicks lets' pick on the Apogee Scintilla. Not only could you get bad sound but you could experience some pyrotechnics with. At one point HP of the absolute sound suggested that not only did some amps have their own sound but the coloration mimicked the color of the faceplate.
I hope I am grasping the point here. One speaker that takes this approach is the Gersham Sonogram. They claim there speaker needs no fancy cables, stands or room treatment. They also claim any good electronics will do. One hardly would call $10k of iImms Audio Cd player and integrated amp generic. Many remarked however that in a hotel room that was untreated and the speaker sitting on the floor they got some of the best sound at the show. Speaker price $3k. I Definitely agree get the best speaker you can for the room you like and just as important your favorite type of music.

Of course it is just a beleif until proven in a DBT.:)
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Great post Tom and it includes a topic I thought should have tis own thread which is the different speaker technologies and what they mean in simple terms. So wonderful job there.

This bit then is for completion. Room artifacts, assuming not created by walls and surfaces rattling, are linear in nature. That is, reflections are added or subtracted and as such, do not generate non-linearities. And AMP or DAC though, create non-linear distortions (their response curve is not a flat line) and hence, generate harmonic distortion, resulting in sound getting harsh, bright, etc. Since the room distortions are linear, they do not mask these artifacts, even though the room distortion is far, far higher than these artifacts.

Here is a simple test. Put a DVD of a concert in your player that has both PCM and Dolby Digital soundtracks. Then switch audio tracks between them using the remote. You will surely hear the difference in any room. Yet all the room artifacts are still there. DD at low rates used in DVD creates non-linear distortion, creating artifacts that are not masked by the room.

So while your point is absolutely correct, that room optimization should be the top area of focus for people (and if not, at least use room correction DSP), there common misconception that the other distortions will be covered by the later is not correct most of the time.
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Like the attempted steroid DBT if the effects are that dramatic with speakers no need "to go blind" so to speak.
Actually studies have shown the need for DBT is just as valid there as it is in other spaces. Here is a funny story related to that:

I was in Fry's about 10 years ago when home theater was just getting popular. I was walking around and see this nice theater with a sign saying demonstrations in 5 minutes. So I go in and a couple of other people also show up. I look around and there are these floor standing speakers in the front that were about 4-5 feet tall. And somewhat smaller but still larger speakers all around. He then proceeds to play a chase scene with helicopters and the like, at pretty substantial levels.

The demo finishes and the guy walks up the floor speakers and pulls off the cover, only to reveal the Bose Acoustimass speakers. If you don't remember what they are, this is a picture:



As you can tell, they are tiny little speakers that would fit in your hand. What they were hoping to do was to fool the viewers first into thinking they are hearing these giant speakers, and then say, look how little space they take. I must admit, even though I thought they sounded shrill and strained, even I didn't think they were using these little things to fill that large room.

Unfortunately for them, the trick did not work. There was an older gentleman sitting in the front row and the presenter made the mistake of asking, "didn't that sound great?" The person chimed in, "I couldn't hear what they were saying!" Surprised, the presenter asked, what do you mean? He said, I couldn't understand what the actor was saying! He was right of course. The shrill sound and muffled mid-range did make it hard to understand dialog.

Still, it shows that you can fool people fairly easily with what they are looking at. And there is good industry research to support that as in Folyd Toole and Sean Olive's excellent book on speaker design titled: “Loudspeaker and Headphone Handbook”:

” In practice, the principal difficulty with subjective evaluations is to control what the listeners are responding to. Listeners who, because of room acoustics, hear different sounds from the same loudspeakers cannot possibly agree. Listeners who see the products they are evaluating cannot completely distance themselves from the biasing influences of style, price, size, brand reputation, and so on. Listeners who hear sounds that are unreasonably altered by atypical acoustical surroundings cannot express a view that represents that of a larger population. Music that reflects only a narrow spectrum of what customers play cannot be assured of revealing a balanced perspective of performance. We need to get serious about the conduct of listening tests – so that ‘opinions’ can take on the attributes of ‘facts’, to create a system of subjective evaluation that is a proper companion to technical measurements.

It is well known, in subjective evaluations, that humans are susceptible to influences other than the parameter or device under test. We simply find it very difficult to ignore the evidence of brand, size, price, and so on.

...In loudspeaker evaluations, there is no argument that there are audible differences between products, and clearly audible differences traceable to interactions with programme and rooms. In comparisons of evaluations done in both blind and sighted conditions, it has been observed that listeners substantially altered their ratings of products when they were in view, following biases suggested by visual cues. "

He then goes on for a full chapter, talking about how to create proper double-blind tests and environment for reliable evaluation.

I hope follow ups to DBT itself can be in the other thread :) but the point should remain that they have value even if there are gross differences.
 

tmallin

WBF Technical Expert
May 19, 2010
966
375
1,625
71
Chicagoland
Tom

This is one of the first mention of Gedded Summa I have seen in an audiophile Forum... I am extremely intrigued, call it interest in his approach and am even looking to go audition his speakers. What is your take on his speakers, Have you heard them? If this is OT we can goon PM..
Frantz

I have not heard them. I'm not commenting on their sound quality, just how the apparent design is intended to minimize room interaction. For Earl Geddes' Web site, see http://www.gedlee.com/ . Links to consumer reviews, Dr. Geddes white paper on speaker directivity, and a program which can display the response of Gedlee speakers at different frequencies and angles are found there.
 

tmallin

WBF Technical Expert
May 19, 2010
966
375
1,625
71
Chicagoland
Here is a simple test. Put a DVD of a concert in your player that has both PCM and Dolby Digital soundtracks. Then switch audio tracks between them using the remote. You will surely hear the difference in any room. Yet all the room artifacts are still there. DD at low rates used in DVD creates non-linear distortion, creating artifacts that are not masked by the room.

So while your point is absolutely correct, that room optimization should be the top area of focus for people (and if not, at least use room correction DSP), there common misconception that the other distortions will be covered by the later is not correct most of the time.

I agree that we can hear through the 90% to errors in the remaining 10% and I think I said that. It is just that if we are not kidding ourselves, it is my perception that these differences, while audible to those who care about reproduced sound, are relatively small compared to, for example, the difference in sound caused by moving your speakers a few inches.
 

tmallin

WBF Technical Expert
May 19, 2010
966
375
1,625
71
Chicagoland
Some speakers present an extremely difficult load. Just for kicks lets' pick on the Apogee Scintilla. Not only could you get bad sound but you could experience some pyrotechnics with.

My argument is based on the assumption that the system components are basically compatible with one another. Obviously, if one is trying to drive a speaker with a one-ohm impedance and relatively low sensitivity to high SPLs from a low-power tube amp, there will be problems, audible and perhaps pyrotechnic. The rules of the road need to be obeyed. You probably need a high powered, high current, low output impedance solid-state amp in such a situation.

One can imagine a number of other component mismatches which will cause significant problems for either mechanical or electronic reasons. If you use a high compliance cartridge in a high mass tonearm, the stylus will have trouble maintaining contact with the groove during LP warps or because of footfalls and may even skip out of the groove. If you use a high output impedance tube preamp into a low input impedance solid-state amp through high capacitance interconnects, you may have significant treble roll off and loss of bass impact. I'm assuming that such compatibility issues have already been reasonably addressed in selecting and connecting the components of the audio system.
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
My point on the steroid test is they could not evaluate the placebo effect because the effect of the steroids was so dramatic the control group immediately knew they were given a placebo. Thus there was no need to "go blind." The experiment still dramatically demonstrated the effects of steroids.
I attended a similar Bose demonstration. A box was placed over the right speaker creating a serious imbalance in the stereo image. The device they were trying to sell electronically corrected the balance to a perfect stereo image. Two of the speakers were covered by a a replica of an old speaker. The room was very dark. The only real bias I had was against Bose. Their point was they could easily compensate for a bad room. The point was that they use the covered speakers to help correct the stereo image.
As far as DBT is concerned. I'm ready let's do it. But lets' do it across the board for everybody.
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
Like I said let the testing begin. For everybody!
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
My argument is based on the assumption that the system components are basically compatible with one another. Obviously, if one is trying to drive a speaker with a one-ohm impedance and relatively low sensitivity to high SPLs from a low-power tube amp, there will be problems, audible and perhaps pyrotechnic. The rules of the road need to be obeyed. You probably need a high powered, high current, low output impedance solid-state amp in such a situation.

One can imagine a number of other component mismatches which will cause significant problems for either mechanical or electronic reasons. If you use a high compliance cartridge in a high mass tonearm, the stylus will have trouble maintaining contact with the groove during LP warps or because of footfalls and may even skip out of the groove. If you use a high output impedance tube preamp into a low input impedance solid-state amp through high capacitance interconnects, you may have significant treble roll off and loss of bass impact. I'm assuming that such compatibility issues have already been reasonably addressed in selecting and connecting the components of the audio system.

Those rules of the road are sometimes the subject of furious debate. Take my ML CLS. Rigorous adherence to the rules of the road would have caused me to have discarded it long ago. As I stated in another thread the amp speaker interface is very important. It remains one of life's most interesting aspects. Get one parameter right and the other falls out of alignment.
 

marty

Well-Known Member
Apr 20, 2010
3,025
4,169
2,520
United States
Another excellent post, Tom. But I'm not entirely convinced that room trumps speakers. I think a good deal depends on the speaker design (radiation pattern in particular) and whether the best "you are there" effect is facilitated by more direct sound than reflected sound or a combo of direct/reflected sound from the room. As a former dealer many moons ago, I've set up identical speakers in many rooms and have usually elicited more similarities than differences in sound that I attributed to the fact that the speaker was the same, despite the rooms being different.
 

tmallin

WBF Technical Expert
May 19, 2010
966
375
1,625
71
Chicagoland
Those rules of the road are sometimes the subject of furious debate. Take my ML CLS. Rigorous adherence to the rules of the road would have caused me to have discarded it long ago. As I stated in another thread the amp speaker interface is very important. It remains one of life's most interesting aspects. Get one parameter right and the other falls out of alignment.

In my opinion, the reason speakers sounds so different with many amps is that the comparisons are often done with amps which cannot handle reasonably designed speakers. Well designed amps have high voltage capability, high current capability, low distortion, and a very low output impedance. That last factor is the major reason why some speaker/amp combinations seem so critical. Tube amps--all or most--have output impedances of one ohm or more. Stereophile's amp tests have long shown the frequency response variations caused by a dummy speaker load. For tube amps the variation can be plus or minus several decibels over a very wide range. Such variances are sonically obvious to anyone, not just audiophiles. Well-designed amps have very low output impedance which means that even when mated with speakers whose impedance varies wildly and may drop below 4 ohms at some frequencies, the frequency response of the amp varies by only a small fraction of a decibel. Sure, you may still hear some differences among such amps, but they will be audibly subtle.

Put all these requirements together and this means that most inherently good amps are high-power solid state ones, not low-power tube SETs, or even high-power tube amps. The real sound of your CLS speakers or any other speakers are the sound they have when driven by an amp of high power, high current, low distortion, and low output impedance.
 

tmallin

WBF Technical Expert
May 19, 2010
966
375
1,625
71
Chicagoland
Another excellent post, Tom. But I'm not entirely convinced that room trumps speakers. I think a good deal depends on the speaker design (radiation pattern in particular) and whether the best "you are there" effect is facilitated by more direct sound than reflected sound or a combo of direct/reflected sound from the room. As a former dealer many moons ago, I've set up identical speakers in many rooms and have usually elicited more similarities than differences in sound that I attributed to the fact that the speaker was the same, despite the rooms being different.

I guess you don't have the acoustical pig of a room that I do, Marty. :)

But, seriously, I suppose there is reason for swapping speakers and room in the order of things. Certainly part of good setup in my opinion is taking the listening room's own small-room acoustics maximally out of the listening experience. In my experience, listening room reflections NEVER help enhance the realism of what's on the recording. The reflections occur too early and the reflections will be a uniform overlay on all program material, not just stuff recorded from a distance in a big room. A lot of listeners apparently like the effect of listening room reflections, but for me they are a giveaway that the set up is imposing a huge sonic thumbprint on every recording you play.

The difficulty involved in minimizing the second-venue effect of your listening room's acoustics is why I put rooms above speakers in my pecking order. To adequately minimize the effect of your listening room's acoustics so that you are hearing a reasonable semblance of what was actually recorded , you need: (1) near-field listening, which I define as less than 6 feet from each speaker; (2) fairly directional speakers, especially in the mid and high frequencies; (3) toe in of the speakers to directly aim at your ears; (4) lots of absorbing room treatment on all the room surfaces at the first and second reflection points--this includes especially the floor and ceiling; (5) speakers and listener positioned well away from the room walls--I prefer the 1/3 points if possible with a given room geometry; and (6) adequate stereo separation, being AT LEAST 60 degrees and ideally more. Some will add (7) low frequency electronic equalization to counteract room modes.

Since most hi-fi set ups do not come close to meeting these criteria, the listening room acoustics dominate. As I mentioned, they dominate to the extent that moving the same speakers a few inches within the room can often make a huge difference in the perceived nature of the presentation.
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
Moscode 402au v ML CLS I

In my opinion, the reason speakers sounds so different with many amps is that the comparisons are often done with amps which cannot handle reasonably designed speakers. Well designed amps have high voltage capability, high current capability, low distortion, and a very low output impedance. That last factor is the major reason why some speaker/amp combinations seem so critical. Tube amps--all or most--have output impedances of one ohm or more. Stereophile's amp tests have long shown the frequency response variations caused by a dummy speaker load. For tube amps the variation can be plus or minus several decibels over a very wide range. Such variances are sonically obvious to anyone, not just audiophiles. Well-designed amps have very low output impedance which means that even when mated with speakers whose impedance varies wildly and may drop below 4 ohms at some frequencies, the frequency response of the amp varies by only a small fraction of a decibel. Sure, you may still hear some differences among such amps, but they will be audibly subtle.

Put all these requirements together and this means that most inherently good amps are high-power solid state ones, not low-power tube SETs, or even high-power tube amps. The real sound of your CLS speakers or any other speakers are the sound they have when driven by an amp of high power, high current, low distortion, and low output impedance.


I know forum members are tired of me discussing my system, But lets face it, they are an improbable match on paper. Sterophoile reviewed the 402au and found them to have an output impedance of .22 ohms and and a rise in distortion as the speaker impedance dips below 4ohms. http://www.stereophile.com/solidpoweramps/moscode_402au_power_amplifier/index4.html.
Lets put that toghether with ML CLS. CLSfig1..jpg http://www.stereophile.com/floorloudspeakers/650/index11.html Assigning the CLS an impedance of 4 ohms is being generous. 33 ohms at 1khz, 2.5 ohms at 10khz and 1.9 ohms at 20khz. Sterophile suggests avoiding speakers that dip below 4 ohms.
My listening experience makes them an almost ideal match. I even get excellent bass. Many have called the CLS edgy and and bass shy. One can assume I have no interest in a speaker that sounds that way. Manny ahve suggested that a solid state behemoth is the only way to go. They generally have very low output impedance and the ability to remain stable with complex loads. Such combos have left me cold or tend to be prohibitively expensive. So the question what's going. We can design a near perfect amp based on the parameters we know how to measure. But those devices have usually been found wanting musically. It is why stereo design is part science part art.
Finally I would suggest i would not even consider an SET with he big Wilsons. They must be a good match or why would Steve Bother. I'll be hearing them later this month and evaluate what's going on.
 

tmallin

WBF Technical Expert
May 19, 2010
966
375
1,625
71
Chicagoland
I know forum members are tired of me discussing my system, But lets face it, they are an improbable match on paper. Sterophoile reviewed the 402au and found them to have an output impedance of .22 ohms and and a rise in distortion as the speaker impedance dips below 4ohms. http://www.stereophile.com/solidpoweramps/moscode_402au_power_amplifier/index4.html.
Lets put that toghether with ML CLS. View attachment 276 http://www.stereophile.com/floorloudspeakers/650/index11.html Assigning the CLS an impedance of 4 ohms is being generous. 33 ohms at 1khz, 2.5 ohms at 10khz and 1.9 ohms at 20khz. Sterophile suggests avoiding speakers that dip below 4 ohms.
My listening experience makes them an almost ideal match. I even get excellent bass. Many have called the CLS edgy and and bass shy. One can assume I have no interest in a speaker that sounds that way. Manny ahve suggested that a solid state behemoth is the only way to go. They generally have very low output impedance and the ability to remain stable with complex loads. Such combos have left me cold or tend to be prohibitively expensive. So the question what's going. We can design a near perfect amp based on the parameters we know how to measure. But those devices have usually been found wanting musically. It is why stereo design is part science part art.
Finally I would suggest i would not even consider an SET with he big Wilsons. They must be a good match or why would Steve Bother. I'll be hearing them later this month and evaluate what's going on.

Just so we're clear, Gregadd's amp is a Moscode amp. That amp does not have a tube output stage, but is a hybrid amp with a small-signal tube input stage and MOSFET output stage and no output transformer. Thus, this amp has a decently low output impedance of .22 ohm. And while the Stereophile distortion graphs do show a rise in distortion with 4-ohm loads compared to 8, the distortion remains down around .2% up to more than 200 watts output. While not a behemoth, power-wise, compare these findings to the Lamm ML3's published specs.
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing