[JLM]

By default I'd think addressing these variables should be the same qualities that make a speaker sound "better". 

We've always been taught to look first for a flat frequency response (on axis, at moderate sound pressure levels).

I ask because over the years I've been reading more about the effects of many other factors.  Phase, constant directivity, room considerations, amp/speaker synergy, and efficiency are just a few examples of possible parameters.

Let's try to answer this with some sort of a scientific basis.

[FullRangeMan]

Sorry I have any scientific basis to show, just my personal taste:
- Midbass range.
- Bass
- Real harmonics to a credible room sound presence
- sweet treble
- Hi SPL capability with few Watts.

[Quiet Earth]

How do you measure an emotional response?

[JLM]

FRM,

- Your reply primarily addresses most of the audible frequency spectrum.
- What does "real harmonics to a credible room sound presence" mean?
- How does one listen to high efficiency?

QE,

- I invite you to describe (qualify if not quantify) whatever sonic attributes evoke the emotional response you may be thinking of.

[*Scotty*]

For myself, I take it as a given that the speaker will have accurate tonal and timbral balance as well as a freedom from bass overhang independent of room effects. With these criteria met to start with, I am primarily concerned with how well the speaker presents the dynamic content inherent in the music as well its ability to create and disappear into a virtual acoustic space that is captured at the time the recording was made.
 Even if it gets timbre and tonality correct , it's still an epic fail if sounds like two boxes sitting at one end the room. It seems to to me that there is very little reason to invest much in the hobby if a convincing recreation of the recording venue. whether the actual live or the virtual reality produced studio, is not produced by the system.
Scotty

[JLM]

I'm very much there with you Scotty, but could you describe how "bass overhang" (pulse response?) can be fully independent of room effects (thinking of speakers pushed too far into corners).

[*Scotty*]

Most domestic listening rooms resonate below the Schroeder Frequency, in other words they behave as though they are a bell with multiple resonant modes related to the rooms dimensions. How long it takes for these resonances to die away determines how far the reproduced bass sound wave that is heard varies from the one contained in the recording.
 A separate issue is how well the driver is damped by its enclosure,if it has one. The driver and the enclosure it is mounted in are a system which has at least one characteristic frequency at which it resonates. How well this frequency is damped and how much energy is stored by the system and how much later it is released time determine whether the speaker brings bring poor bass transient behavior to the party in the listening room thereby compounding any existing problems.
I encountered one of the cleanest sounding systems I have ever heard at the recent CAF show. It consisted servo amped open baffle sub-woofer system consisting of three 12in. woofers per side. Very realistic bass with Very Little room resonance problems and probably zero driver overhang as a result of servo control.
Scotty

[JLM]

Thanks Scotty for the clarification. 

Yes residentially sized rooms will all suffer from resonant modes (echo).  The next question (beyond the scope of this thread) would be what is the best way to address it.

All matter resonates (driver diaphragm/frame, baffle, enclosing walls, enclosed air, air outside, objects in the room, and the room's walls/floor/ceiling. 

[FullRangeMan]

FRM,

- Your reply primarily addresses most of the audible frequency spectrum.
- What does "real harmonics to a credible room sound presence" mean?
- How does one listen to high efficiency?

There are drivers that have a harmonics pattern not suited to the human brain or harmonics that are not what the human brain expect to listen.
There is few drivers of any kind that meet this requiriment.

Hi efficiency? not sure if I understand this question.
Hi effic speakers sound very different  from low eff.

[JLM]

There are drivers that have a harmonics pattern not suited to the human brain or harmonics that are not what the human brain expect to listen.
There is few drivers of any kind that meet this requiriment.

Hi efficiency? not sure if I understand this question.
Hi effic speakers sound very different  from low eff.

 

Maybe we're having a language problem (it's too easy to jump past right into design concepts), but it seems that you're confusing cause with effect.  I'm asking for the effects (what we hear) not how/what produces it. 

[JohnR]

By default I'd think addressing these variables should be the same qualities that make a speaker sound "better". 

We've always been taught to look first for a flat frequency response (on axis, at moderate sound pressure levels).

I ask because over the years I've been reading more about the effects of many other factors.  Phase, constant directivity, room considerations, amp/speaker synergy, and efficiency are just a few examples of possible parameters.

Let's try to answer this with some sort of a scientific basis.

I know that Toole is sometimes quoted without much thought, but AFAIK a flat on-axis response, smoothly varying off-axis response, and low distortion are still pretty worthy goals. As far as phase goes, my own experience is that the best systems (active) have addressed that; others will disagree. I wonder about efficiency, I'm inclined to think that lack of compression is the real underlying goal there. In the end, though, you just have to try things for yourself...

[Freo-1]

Found this write up recently.  Explains what qualities a good sounding set of speakers should have:

" 
 The Design and Development of High Performance Loudspeakers-An Update Review The performance of a loudspeaker can be defined by its linear and non-linear behavior. Linear performance is defined by the impulse response and non-linear performance by harmonic distortion measurements.
The most important elements to consider in a practical design, which are encompassed by the characteristics of linear and non- linear behavior are detailed under the following headings.
1. Magnitude Response
 
The magnitude response of a loudspeaker measured using analogue techniques has been the mainstay of most loudspeaker assessment for decades.
By definition a ”Linear Magnitude” refers to a magnitude response that has a constant level with frequency and only then will it not cause any linear distortion. We all know that this is practically not achievable and that the impulse response of a loudspeaker is largely dominated by the low and high frequency roll-off characteristics and by any resonant peaks in the amplitude response.
It is possible however, to produce loudspeaker systems that maintain a variation in magnitude response within +/- 1.5dB consistently between 100 Hz and 10 kHz and that have an excellent overall balance between bands. We believe that the balance between drive unit frequency bands is critical, particularly between bass and midrange in three way systems, and should always be better than 1dB.  Time is well spent on drive unit development in order to meet this magnitude response limit. It is much more elegant to use properly developed drive units which will then enable the use of simple crossover filters than to use complex equalization before crossover filters in an effort to correct drive unit magnitude response anomalies.
 
2. Phase Response
 
As with the magnitude response, the phase response of a system is usually measured on a single reference axis, midway between the bass/mid and high frequency drive units in a two way system and on the axis of the midrange drive unit for a three way system. A system will be defined as being ”Linear Phase” if the phase response is a straight line, when the frequency response has a linear scale and passes through the origin. The effect is then of a true time delay and will therefore not cause any linear distortion.   In practical loudspeaker systems however, the aim is to design for a minimum phase response free from any abrupt changes that are usually indicative of high Q resonance’s. It is also relevant to include here that the delay between drive units due to acoustic centre misalignment is not audible, we believe, for delays below 2 ms. Therefore, providing the overall delay is within 2 ms. and there are no sharp phase response irregularities, then the system should be free from any subjective phase effects. ATC has incorporated analogue excess phase correction, operating through the crossover regions, in its active loudspeakers since 1982. The result of such active filtering is to give much better control over the filter shapes with greater phase coherence and therefore a more uniform group delay characteristic. The subjective result, when compared with the same loudspeaker system but with a passive crossover, is of a broader and more stable stereo sound field with much more coherent drive unit integration and improved openness and timbre of reproduced sounds.


3. Time Domain Anomalies
 
A high performance loudspeaker should have no high Q or delayed resonance’s and must also minimize multiple arrivals of the same signal caused by reflections and diffraction effects as these add a hard and claustrophobic coloration to the sound, masking ambient detail and confusing the stereo image. Time domain anomalies are without doubt the most intrusive and tiring to the listener of all distortions. Careful drive unit and crossover design can ensure a flat and even magnitude response free from any low Q broad band resonances or response irregularities. High Q and delayed resonances however, which are common in hard undamped diaphragms and poorly designed crossover filters, are not so easily ameliorated. In fact the only successful solution is to design heavily damped flexible diaphragm structures having high internal resistance and great structural integrity even under high drive levels.

At A.T.C. best results have been achieved using quite steep curvilinear and domed diaphragms formed from polyester fabric which is then impregnated with plasticized PVA and/or Opanol viscous damping mediums to control resonant break-up modes which occur at high frequencies.
It is also equally important for the fundamental system resonance to be well damped, that is have a Q between 0.3 and 0.6.
Loudspeakers with an under damped system resonance produce ill defined bass which sounds uncontrolled and excessive and masks midrange detail.
In fact what is really being said here is that for a high performance loudspeaker all resonant systems should be at least critically damped whether they are due to diaphragm break-up or the fundamental system resonance.
 
4. Dispersion and Directivity
 
The relationship between direct and reverberant sound is very important in high performance loudspeakers. It is clear that not only must the on-axis magnitude response be accurate and linear but also that the behavior off-axis must be both broad and even with frequency exhibiting no abrupt dips in amplitude.
In a room when listening to a stereo pair of loudspeakers you first hear the direct sound and then the reverberant field. It is generally agreed and probably true that the reverberant field masks periodic signals, however, it is also apparent that the ear has a precedence effect which means that for impulsive sounds the ear can hear phase dependent effects. Therefore, we believe that any critical judgment of reproduced sound is made principally on the first arrival or direct sound which gives most of the phase related cues and also the low level detail which is quickly lost in the reverberant field.
However, the way we perceive magnitude band balance and the full energy of percussive or impulsive sounds, is dependent upon the power response of the loudspeaker or how evenly it excites the reverberant field with frequency.
Clearly it is impossible to exclude from such a relationship the effect of room acoustics, however, for the purpose of discussing loudspeaker performance we will assume that the listening room has been properly treated and has no serious intrusive problems.
It should also be stated here that the use of D.S.P. to equalize loudspeaker room interface problems is not an acceptable solution to that problem in critical listening environments if it involves modifying the direct sound from the loudspeaker. A dramatic effect of poor midrange dispersion, common in many two way loudspeaker systems, is demonstrated by recording engineers making incorrect magnitude band judgments and applying equalization, usually to the upper midrange, in an attempt to compensate for the apparent lack of energy in that region. Many examples of pop recordings are available which demonstrate this characteristic. That is, a hard strident upper midrange which masks high frequencies, and makes vocals sound recessed while accentuating the bass.
 Non Linear Distortion   
5. Harmonic Distortion

Non linear distortion is the product of non-linearity in a system’s transfer function. There are three principal sources of non-linear distortion in loudspeakers and they are all related to the drive system.
The first relates to the voice coil and magnet gap geometry and the non-uniformity of the distribution of magnetic lines along the length of the magnet gap. A short coil in a long gap renders the best solution regarding geometry, although not the most commonly used, and the distribution of magnetic lines will be improved by the use of an undercut centre pole. Further advantages of this geometry are the improved heat dissipation and therefore reduced operating temperature of the voice coil as well as a reduction in the variation of voice coil induction in relation to its instantaneous position in the magnet gap.
The second principal source of non-linear distortion is generated in the suspension system of the diaphragm assembly and is mainly contributed to by the spider. The spider presents a number of difficult design compromises when longer excursions are required in high power drive units. It must exhibit high axial compliance while also being progressively resistive towards the extension extremes so as to avoid mechanical damage and at the same time be very stable normal to the axis so as to ensure good voice coil centering in small magnet gap clearances. These characteristics together with the added advantage of a reduced tendency to diaphragm wobble on the suspension can be achieved by the use of spiders with a simple triangular roll geometry.

The third source of distortion also relates to the magnetic drive system and is caused by the inherently non-linear magnetic characteristics (hysteresis) of mild steel together with the induction of eddy currents into the mild steel magnetic circuit surrounding the voice coil. ATC has developed a solution to this problem with the help of a new raw material that has the unique properties of high magnetic permeability and saturation together with low electrical conductivity.
When rings of this super linear magnetic material (SLMM) are fitted to the mild steel magnetic circuit, either side of the voice coil: a) the magnetic field in the region of the voice coil will stay essentially the same being maintained by the mild steel front plate and pole in the remaining circuit and b) the close proximity of the SLMM to the voice coil will increase the coils self inductance. This is achieved because the SLMM has high electrical resistance and, therefore, eddy currents induced into the SLMM adjacent to the coil are suppressed causing the self inductance (impedance) to rise. The result is that since the impedance and therefore the voltage across the coil go up when the SLMM rings are fitted, the harmonic components that are induced back into the voice coil stay at the same level because they are dependent only upon the magnetic field. The result is a reduction in 3^rd harmonic distortion by 10-15dB between 100Hz and 3KHz.
 
6. Amplitude Intermodulation Distortion
 
Amplitude intermodulation distortion, however, is much more intrusive than harmonic distortion due to the products not being harmonically related to the original sound.
A recent review of active and passive loudspeakers at A.T.C. confirmed that active loudspeakers, due to each drive unit amplifier operating only over a restricted frequency band, will have much lower amplifier borne amplitude intermodulation distortion than the same loudspeaker operated passively driven over the full audio frequency range. Figures 5 and 6 show a full 20dB difference in amplitude intermodulation distortion in favor of the active system.
 7. Hysteresis Distortion
 
The presence of hysteresis distortion implies that the system transfer characteristic is not always single valued for a given instantaneous input and will vary with both the change of direction and the level of the input and that it will therefore produce distortion that has a different phase to that produced by harmonic distortion.

Hysteresis distortion, as much as it exists in loudspeaker suspension systems and heavily damped soft diaphragm assemblies, does not manifest itself as an intrusive distortion. It is certainly not particularly evident in other measurements, for example, transient response, magnitude response or in harmonic distortion measurements. In fact, if care is taken over the choice of both diaphragm and suspension materials then they will largely have the characteristics of a simple damped spring and exhibit negligible hysteresis.

 
8. Dynamic Range
 
The issue of dynamic range is a complex one and although it is primarily controlled by voice coil operating temperature and magnet total flux it must be considered along with the mechanical integrity and freedom from break-up of the diaphragm and suspension structure. There can be no doubt that system dynamic range significantly effects the clarity of reproduced sound. Even quite simple combinations of instruments, for example a string quartet, will produce a maximum SPL well in excess of 100dB at 2m when starting from just audible pianissimo passages.
A loudspeaker that has significant power compression will tend to sound dull and boomy and the high voice coil temperature and consequent resistance rise will effect the loading of the passive crossover and therefore also modify the magnitude response of the system.
The dynamic range of direct radiating loudspeakers is in fact almost entirely determined by cost. Designers do strive to produce more sensitive small systems through the use of very light diaphragm structures but the scope for maneuver is limited if a correct balance between bass and midrange magnitude response is to be achieved for a given diameter of drive unit. Furthermore, lightdiaphragm structures almost always have low internal damping and therefore a tendency to exhibit high Q resonance’s.

To qualify in all respects as a high performance loudspeaker the requirements of dynamic range will for most designs be the largest compromise. A choice which is made much more difficult as a consequence of the rapid developments in digital electronics during the past decade. Digital recording mediums offer a huge dynamic range with a peak to average of typically 12-16dB which means that even the most modest loudspeaker wearing the tag ”high performance” must be capable of continuous output of at least 94dB at 1m while being driven from an amplifier of 100 watts or more.
 
 
W.J. WOODMAN
ATC (ACOUSTIC TRANSDUCER COMPANY)
LOUDSPEAKER TECHNOLOGY LIMITED  "     

[JLM]

Freo-1,

Boy all that seems so right, but the ATC prices are all so wrong ($4990/pair for the cheapest active ATC, the small 2-way SCM20ASL PRO MK2).   

[jtwrace]

Let's try to answer this with some sort of a scientific basis.

Weird he is but absolutely brilliant. 

[poseidonsvoice]

Freo-1,

Boy all that seems so right, but the ATC prices are all so wrong ($4990/pair for the cheapest active ATC, the small 2-way SCM20ASL PRO MK2).   

JLM,

Price is really immaterial to what you started this discussion/thread on in all honesty. ATC is not really an offender of that in my book judging by how crazy some systems cost at these shows. There is merit in what Freo has posted although I don't agree with it all. ATC, even at their prices have some serious engineering even though they don't have the snazzy looks of the popular designs here on AC and elsewhere. I get concerned when all the loudspeaker manufacturer can talk about for paragraphs is their 'latest cabinet material' and fancy Scandinavian drivers. God dammit I wanna see some f#%^ engineering please. Where are my finely delineated (every 5 degree) barely smoothed (1/10 to 1/24th octave above 300 Hz) horizontal off axis curves? Where is my higher than average sensitivity? Where is my speaker that sounds phenomenal on average electronics and doesn't need a LAMM amplifier to make it sing? Those are the questions I ask. Unfortunately, many of them shy away, switch the subject or refuse altogether.

I'm much more in line with the aforementioned link from jtwrace. It has several pearls that I tightly adhere to.

Both are "white papers" so try to read between the marketing lines. Still, it's far less marketing than the rest of the industry.

Best,
Anand.

[Freo-1]

JLM,

Price is really immaterial to what you started this discussion/thread on in all honesty. ATC is not really an offender of that in my book judging by how crazy some systems cost at these shows. There is merit in what Freo has posted although I don't agree with it all. ATC, even at their prices have some serious engineering even though they don't have the snazzy looks of the popular designs here on AC and elsewhere. I get concerned when all the loudspeaker manufacturer can talk about for paragraphs is their 'latest cabinet material' and fancy Scandinavian drivers. God dammit I wanna see some f#%^ engineering please. Where are my finely delineated (every 5 degree) barely smoothed (1/10 to 1/24th octave above 300 Hz) horizontal off axis curves? Where is my higher than average sensitivity? Where is my speaker that sounds phenomenal on average electronics and doesn't need a LAMM amplifier to make it sing? Those are the questions I ask. Unfortunately, many of them shy away, switch the subject or refuse altogether.

I'm much more in line with the aforementioned link from jtwrace. It has several pearls that I tightly adhere to.

Both are "white papers" so try to read between the marketing lines. Still, it's far less marketing than the rest of the industry.

Best,
Anand.

Well stated, Anand.  I find merit in both white papers.  I especially like what Mr. Geddes has to say about sub woofers.

While I overall lean towards the design principals from the ATC, there certainly is a lot of merit with Me. Geddes approach.  Anand's point about supplying more engineering information vice hype is important with helping to sort out what aspects of speaker reproduction an individual wants in their system. 

[G Georgopoulos]

The most important attribute has got to be, you have to like the speaker (unscientific view, yes!)....

[JLM]

Yes, I quite like what I read from Geddes, Toole, LeJeune, and others.

The most important attribute has got to be, you have to like the speaker (unscientific view, yes!)....

As an audio old fart I'm setting in my ways and can tell at the door if I like that pair of speakers or not.  Such was the case at the April 2015 Axpona audio show in Chicago.  90% of the rooms all sounded similar - artificial/cartoonish: life turned into a poster of only primary colors without hue, tone, or texture.

But even old farts can learn (I hope), thus the purpose of this thread.

[smk]

You can forget the "flat frequency response." Every room I've been in sounds different. All you can do is add acoustics panels out the wazoo & hope for the best.

[werd]

I like lots of air. I like hearing instruments but the glory is in what one doesn't hear. Don't care about off axis response at all. What I like to hear is a room that is lively with reverb and lots of good air around the instruments. All in the sweetspot. Most residential listening rooms are not like that. Built around drywall, carpet, laminate flooring, furniture, shape, etc. It's either dead or the reverb/echo is completely off the mark for hi-fi playback.