Highest resolution and most clinical , zero forgiveness monitor speakers out there.

[sage691]

The single quarter sized stock speaker in an old Mac G4.

Secret weapon for sure -- you don't need a pair of Barefoots.

Trust me

[audio ergo sum]

Originally Posted by tpad

I don't believe that the KH120 is minimum phase, since it most certainly includes some allpass phase correction networks. Anyway, if you think that there is something unusual about it's low end phase response, then compare it to other speakers (if you can find the information!). I think that you'll find the same general trend in most.

It is well established that you can't hear group delay effects at the lower frequencies due to our inability to resolve detail very well below 100 Hz. Here are some well established numbers:

CCITT Recommendation J21

Table 2/J.21
------------
40 Hz 55 msec
75 Hz 24 msec
14 kHz 8 msec
15 kHz 12 msec


AT&T Program Transmission
(threshold of audibility)
-------------------------
100 Hz 15 msec
upper HF cutoff 8 msec

This is the first time I hear someone citing ITU numbers when it comes to quality standards for high end sound transducers.

These might be well established numbers - found by compromise between what is desirable and what is technically doable - for telecommunication devices and broadcasts transmission lines(ITU). But you are in for a major surprise, if you think those numbers represent or come even close to what human hearing is capable of, objectively, measured scientifically.

Our ear can discriminate time differences in the micro second range. That is prevalent in loudspeaker reproduction, when it comes to the psychoacoustical phenomena of localization and perception of "sound stage".

And nobody at ITU standardization committees cared about transients. In the time domain they care about parameters like speech intelligibility etc.

The whole most important chapter of transparent transient reproduction has been neglected by the scientific community much, yet it is most important for the recognition and high quality transmission of musical sounds. Transients are the red headed step child of the scientific audio community.

Again, the whole theorem of "our ear is relatively insensitive to phase" is based on research with continuous "swung-in" sound signals, disregarding transients.

For the beginning I suggest a simple ABX test. Transient rich full frequency spectrum instruments (Piano, guitar, harpsichord, pizzicato Bass, drums etc.) reproduced with phase shifts as shown in the KH120 graph and without. I predict the result shall be obvious in favor of easily identifying the difference.

[matt thomas]

Originally Posted by Audio X

Most people I know of who work with the L707's also incorporate a sub or two.

For any speaker that needs a sub and doesn't fully cover the bottom octave, I would not consider as being "the best" loud speaker.

You are thinking of only their smaller speakers like the 707. Their signature range specs state:

Frequency response: 37Hz-20kHz +/- 1 dB (20Hz-40kHz +/- 3 dB)

But yes, I do use a sub with my Lipinski 707s , but really only because I like sub frequencies. Othertimes I use the sub with my Proacs, and run my Lipinskis sub-less.

They are amazingly accurate speakers.

Matt

[Audiop]

Originally Posted by audio ergo sum

Our ear can discriminate time differences in the micro second range. That is prevalent in loudspeaker reproduction, when it comes to the psychoacoustical phenomena of localization and perception of "sound stage".

How was these studies performed? Did you do them or do you have a link? Can you clarify what you mean with the first line?

Localization depends on interaural time differences and interaural level differences.

Again, the whole theorem of "our ear is relatively insensitive to phase" is based on research with continuous "swung-in" sound signals, disregarding transients.

Based on test with music including transients.

For the beginning I suggest a simple ABX test. Transient rich full frequency spectrum instruments (Piano, guitar, harpsichord, pizzicato Bass, drums etc.) reproduced with phase shifts as shown in the KH120 graph and without. I predict the result shall be obvious in favor of easily identifying the difference.

Can you be more specific about your proposed test? In particular it would be interesting to find out what to do to exclude other sources of detection of an audible difference such as changes in dispersion and energy response.

Thanks!

/Peter

[tpad]

...The whole most important chapter of transparent transient reproduction has been neglected by the scientific community much, yet it is most important for the recognition and high quality transmission of musical sounds. Transients are the red headed step child of the scientific audio community....

Peter, you're wasting your time. ergo is just publishing more audiophile subjectivist nonsense. If he had any FACTS (meaning credible scientific test results), it might be worth discussing...but he doesn't. The AT&T group delay limits actually were derived by LISTENING to program material.

[Roland]

Transient response is the red-herring of the audiophile community, in so much that when listeners are presented with the higher dynamic range provided by them, the preference is always for lower dynamic range. Hence our love for analogue tape, compression, loud mastering (though this has been, at times taken to silly extremes).

In spite of some of the sillier arguments put forward here, the KH120's do spec very well, are pretty cutting edge technology and subjectively have a lot of fan's, I for one know of a former ATC SCM20 user that made the change because he thought they were a significant improvement. As has been pointed out above, many manufacturers are reluctant to provide much in the way of measurement details, a good example was a speaker some posters here were raving about, I took the time to take a look, little in the way of published data, spec's were not particularly impressive and just a general look at the design would suggest some fundamental problems such as very low cabinet "Q", potential baffle diffraction issues and lack of any kind of time alignment.

Other speakers like the Lipinsk's and Revel's have great reputations and they all spec well, particularly in terms of smooth response plots.

[audio ergo sum]

Originally Posted by tpad

...
Peter, you're wasting your time. ergo is just publishing more audiophile subjectivist nonsense. If he had any FACTS (meaning credible scientific test results), it might be worth discussing...but he doesn't. The AT&T group delay limits actually were derived by LISTENING to program material.

(not so) nice try of an ad hominem attack. What about your FACTS? I can HEAR the KH120's bass being very much out of phase, more so than other speakers. Now what?

Why don't you just listen?
Those group delay limits are in no way acceptable for high quality music reproduction. Those listening tests - I have been there, done it - all have the problem that probandi have not the right idea what to listen for and that they should listen particularly to transient heavy material and concentrate on the faithful reproduction of transients compared to the source.

Anyway, EVERYBODY who is not deaf can hear a group delay of 16 ms between lower and higher frequencies (as ITU J21 allows) regrading transient reproduction. It's equivalent to a double bass in front of you emitting high frequencies only and another instrument 5-6 meters behind the "high frequency bass" for the low frequencies only.

Or 16 ms is equivalent to a mid-high frequency speaker positioned 5-6 meters in front of a low frequency subwoofer. Kind of obvious effect...

Again, our ear's relative indifference to phase is a fact for "swung in" signals, where phase does not relay any important information about the sound. Transients are a different story.
The limitation in phase coding is due to the limit of our neurons to fire a maximum of app. 300 pulses per second. That is an issue for periodic signals, not for the onset of sound as in transients... (as I have stated numerous times)

Phase or "group delay" on the other hand is important for the onset of sound, for the transients. Different story.

The onset of sound is coded into neuronal impulses with a delay of app. 10 micro seconds.
(Purves et al. 2001)
http://www.ncbi.nlm.nih.gov/books/NBK10799/

same here in earlier research (1974), app. 10 microseconds, most sensitive for LOWER frequencies (remember the KH120 graph...)
http://asadl.org/jasa/resource/1/jas...sAuthorized=no

somewhat similar numbers (app. 12 microseconds) for a rough noise burst test here:
http://asadl.org/jasa/resource/1/jas...s5?bypassSSO=1

other research found the just noticeable time/phase difference - dependent on level and frequency - to be around 2 deg phase shift.
http://asadl.org/jasa/resource/1/jas...sAuthorized=no

now let's all have a hearty lough about your and ITU's 16 ms limit, one thousand times above the human threshold of time discrimination in hearing...

Thus onset of sound (read transients) are coded relatively fine and precise, while for periodic "swung-in" sound the phase is only roughly coded, and not at all for frequencies above 2-3 kHz.

That 10 microseconds threshold also defines the maximum resolution, with which our ears/brain can localize phantom sources from binaural hearing btw.

Anybody who has a device like the spl vitalizer et al can check it out. It plays with the phase in the bass frequencies. Or transformers shift phase in degrees, not constant time delays, thus introducing different group delays depending on frequency, an obvious effect.
Transient "designers" do not change the frequency response, yet they change the transient shape and thus the sound. Now what, you believers of the (old) book...

Just listen. (!!!) Play any track with a nice bass drum sound, and you need not more than 10 seconds to hear how faithful transients are reproduced by a particular speaker.

That we have very rough standards for phase transparency is a lot due to the fact that linear phase is (or was at the time) technologically almost impossible to achieve within reasonable budgets.

[audio ergo sum]

Originally Posted by Roland

Transient response is the red-herring of the audiophile community, ...

I disagree. Transient response in terms of phase is the most overlooked parameter when it comes to identifying differences in sound for devices that measure almost identical for overall frequency response etc. That holds true for many microphones, preamps, or loudspeakers on the market.
Often those perceived differences can be traced back to phase response.

Why you are also throwing in dynamic range here, I do not understand.

[Roland]

Originally Posted by audio ergo sum

I disagree. Transient response in terms of phase is the most overlooked parameter when it comes to identifying differences in sound for devices that measure almost identical for overall frequency response etc.
Often those perceived differences can be traced back to phase response.

And this is ONE of the reasons why two microphones with very flat response curves can sound different, obviously alongside capsule resonance, distortion and off axis response.

Ultimately what is being talked about here is semantics. Show me a set of speaker specifications for a commercially available speaker that are significantly better than the Neumann. The fact that many manufacturers don't even show that level of information suggests that they either don't have the technology to test for this or that their design doesn't hold up under this level of measurement.

Poor phase response will lead to speakers not measuring flat.

[Kyle_PL]

Originally Posted by matt thomas

You are thinking of only their smaller speakers like the 707. Their signature range specs state:

Frequency response: 37Hz-20kHz +/- 1 dB (20Hz-40kHz +/- 3 dB)

But yes, I do use a sub with my Lipinski 707s , but really only because I like sub frequencies. Othertimes I use the sub with my Proacs, and run my Lipinskis sub-less.

They are amazingly accurate speakers.

Matt

Lipinski Sound L-707 loudspeaker Measurements | Stereophile.com

... ~48 Hz at -3 dB . Linearity are not even +-3 dB (ugly +4 dB bass bump at 110 Hz).

[audio ergo sum]

Originally Posted by Roland

...

Poor phase response will lead to speakers not measuring flat.

Obviously there is a logical fallacy problem with that statement.

Not any phase shifts translates to a change in frequency response.

There is a whole bunch of outboard equipment, that does exactly that, change phase relation, depending on frequency and also on level, yet not changing the measured (!) frequency response.

Phase directly influences psychoacoustically preceived frequency response though, due to many psychoacoustic effects of masking etc.

I don't know where this logical fallacy mantra of some, that phase and frequency response are directly mathematically related at all times, comes from.

[Roland]

As does the misinformation about group delay. A simple bit of web research makes it quite clear that group delay is a byproduct of all low frequency loudspeaker drivers.

Using a sealed box with a Q of 0.5 yields the lowest group delay possible, however, the response will roll off gently, but early. For this reason most speaker manufacturers tend to aim for a Q of about 0,7 that delivers the flattest response. Group delay for vented boxes is slightly larger, but then many prefer the nature of vented boxes over IB's.

One manufacturer that tends to use a low Q design in a vented system is ATC, however, they sometimes get accused of subjectively being a little bass light.

The general consensus that I have seen has been that anything less than 2 cycles at low frequencies, subjectively doesn't seem to bother listeners, possibly due to reverberation in listening environments, however, lack of smooth response seems to subjectively be more of a problem.

[audio ergo sum]

Originally Posted by Roland

...
The general consensus that I have seen has been that anything less than 2 cycles at low frequencies, subjectively doesn't seem to bother listeners, possibly due to reverberation in listening environments, however, lack of smooth response seems to subjectively be more of a problem.

I think it's up to you guys now to put some hard data regarding this on the table.

How was this measured, what were the parameters of the test(s).
I simply believe, there is an epic misunderstanding here.

2 cycles are very audible, when it comes to transient response. Just listen to a nice full spectrum kick drum recording when testing this in real world.

It's very obvious.

[Roland]

Originally Posted by audio ergo sum

I think it's up to you guys now to put some hard data regarding this on the table.

How was this measured, what were the parameters of the test(s).
I simply believe, there is an epic misunderstanding here.

2 cycles are very audible, when it comes to transient response. Just listen to a nice full spectrum kick drum recording when testing this in real world.

It's very obvious.

It was you who proposed this hypothesis in the first place and to be honest you have failed to back up your argument with any hard data.

I will give you the benefit of the doubt that you are not being a "troll" and offer some further details.

Group delay is likely to be of an order of around 5-8ms at around 40hz, best case scenario, however, this would be a very low Q design and subjectively "light" sounding in the bass. A vented monitor like the KH810 is around 14-15ms at the same frequency. Being that the 40hz frequency will take approx 25ms to complete a cycle.

You have to also factor in the crossover and the effect that would have on any speaker.

A great article can be found here that talks in real terms about the issue.

group_delay

Being that you can't get rid of the group delay, it would only ever be of a magnitude worse if you were to introduce a delay, but what would be the point?

I suspect that almost all commercial designs would be within a few milliseconds of each other in real terms, particularly as I don't know of any speaker built in a sealed box with a low enough Q to make a significant difference.

Of course if you do, point out the data that shows this.

[audio ergo sum]

Originally Posted by Roland

It was you who proposed this hypothesis in the first place and to be honest you have failed to back up your argument with any hard data.

You didn't see the links I posted? I suggest you study those first.

Seriously, I don't have to prove anything. Just listen! It's very easy. What is so difficult about this?
Any child can hear the difference between a phase correct bass drum reproduction and one where the low frequencies are delayed by up to two cycles as you suggest.

[Andy_bt]

Originally Posted by audio ergo sum

Seriously, I don't have to prove anything. Just listen! It's very easy. What is so difficult about this?
Any child can hear the difference between a phase correct bass drum reproduction and one where the low frequencies are delayed by up to two cycles as you suggest.

I think one of Roland's question is what system do you listen that bass drums on?
A.

[audio ergo sum]

Originally Posted by Andy_bt

I think one of Roland's question is what system do you listen that bass drums on?
A.

different Loudspeakers with different group delays. Quads, Geithains, B&Ws, Genelecs, K&H/Neumann... Some have less phase shift and group delay than others. There are huge differences and they are audible, particularly in the frequency range below 1 kHz.

Now [Zwicker et al, Electroacoustics, 1984] mention a group delay of 1-2 ms, above which speech becomes smeared, "t" start to sound like "ui".

Yet we are told that ITU standards of group delay of 16 ms are acceptable... haha.

Many authors in numerous studies found the just noticeable threshold of the ear for time delay to be around 10 µs, which directly relates to transient perception.

Yet we are told that up to two frequency cycles of delay are inaudible. (for 50 Hz that would be 40 ms)

And someone seriously calls ME a troll? Amazing.

[tpad]

ergo, you didn't accurately reference Zwicker and Fastl (Psychoacoustics Facts and Models). What he really said was that for three tone complexes (isolated tone complexes), you could hear phase shift of one tone as a just noticeable difference, but that "for frequency separations greater than 200 Hz, i.e. larger than about one critical band, the just noticeable effective difference increases drastically". Which is basically what I had already stated above, where I noted that unless the phase shift was so radical that it caused an envelope (amplitude) disturbance within a critical band, you couldn't hear it. Personally, I don't make a habit of sitting around listening to pathological tone triplets and neither do most other people I know. When you listen to real program material, these effects (if barely audible at all) diminish to nothingness.

So you think that you can hear a one cycle delay in a KH120 at 50 Hz? If that's the case, then how did you un-delay the low frequencies? Describe how you were able to remove ONLY the low frequency group delay without disturbing ANYTHING ELSE. How do you know that you aren't hearing something else?

Furthermore, I still haven't seen any data published for the uber expensive high-end speakers that were discussed at the start of this thread. Except for JBLPRO, speaker manufacturers generally don't publish phase or delay data. So how do you (and everyone else for that matter) know that you aren't listening to loudspeakers that exhibit worse group delay performance than the KH120 if you don't have any data to compare? So' I'm calling your bluff - lets see some data instead of just blowing out a lot of hot air.

[matt thomas]

Originally Posted by Kyle_PL

Lipinski Sound L-707 loudspeaker Measurements | Stereophile.com

... ~48 Hz at -3 dB . Linearity are not even +-3 dB (ugly +4 dB bass bump at 110 Hz).

The guy I was replying to was probably thinking of this speaker (or some other in their range). But I was pointing out that other speakers in their range go much lower down than this, which is true. Look at the stats on the signature range which I mentioned.

The article you linked to also says:

"The broad, 3dB-high peak in the upper bass is entirely due to the nearfield measurement technique, which assumes a 2pi (hemispherical) environment; the L-707 is actually maximally flat down to 70Hz, with a slow 12dB/octave rolloff below that frequency. "

I maintain that these are amazingly accurate speakers. As does the article you linked to: "Following my experience with the Lipinski Sound speaker in my own room, and listening to a pair playing some of my own recordings in Ray Kimber's studio in summer 2004, I share LG's enthusiasm for the L-707" (note that LG's review raved about them)

Matt

[Audio X]

I wasn't familiar with the signature range from Lipinski, but just checking now it looks like they are segment speakers which accompany the 707's to carry the low frequencies. I think the L707's are really good speakers maybe falling just a bit short of what the title of the thread dictates. 2c

[audio ergo sum]

Originally Posted by tpad

ergo, you didn't accurately reference Zwicker and Fastl (Psychoacoustics Facts and Models). What he really said was that for three tone complexes (isolated tone complexes), you could hear phase shift of one tone as a just noticeable difference, but that "for frequency separations greater than 200 Hz, i.e. larger than about one critical band, the just noticeable effective difference increases drastically".

First of all I quoted from Zwicker and Zollner, (Elektroakustik, chapter 9.2.2) and I quoted correctly.

In the source you mention, Zwicker states how phase differences are noticeable. to give a little more context Zwicker writes (chapter 7.3):
"Three effects are clear: our hearing system is most sensitive to phase differences when the frequency separation is small..."
In the figure 7.14 he states the Just-Noticeable Phase Difference to be around 10-20 deg effective phase change. Measured under laboratory conditions, in real living rooms the values go up.
But we can resume now, knowing that our ear effectively IS sensitive to phase changes even for quasi-stationary signals and tones.

And again, I know I'm talking to the wall here, all these are measurements for "swung-in" signals or bursts without real world transients.

And those transients are where the shit hits the fan. It's obvious and very well audible, when small frequency dependent delays are introduced. From 1ms and more speech becomes slurred, hard consonants sound smeared etc.

Which is basically what I had already stated above, where I noted that unless the phase shift was so radical that it caused an envelope (amplitude) disturbance within a critical band, you couldn't hear it.

No, that is not what you stated, unless you would call a 10 deg phase shift radical. Then what is your 360 deg phase shift for the KH120? Out of this world?
And look at the graph again. The KH120 phase shift is in a narrow band. So according to Zwicker, it should be easily audible. Which it is. Thank you for bringing up Zwicker.

Personally, I don't make a habit of sitting around listening to pathological tone triplets and neither do most other people I know. When you listen to real program material, these effects (if barely audible at all) diminish to nothingness.

They don't. It depends what we listen to. For continuous tone and timbre our ear is not very sensitive to the phase changes below ap. 1 kHz, depending on tone relations (see above), and not at all sensitive above app. 2-3 kHz, nobody questions that. But for transients we look at a different model of perception. For transients the first incident matters, just noticeable time/phase difference there is in the 10 µs range. If the transient is smeared, the full spectrum noise is smeared in time, then our ear will detect just that.

The phenomena there is, that the critical band that gets the first incident with the highest level, will also determine the psychoacousticallly perceived timbre/frequency response. Despite when the signal is measured in a time window bigger than 20 µs it will measure the same for speakers with equal frequency response, independent of phase, critical band getting the first incidence.

Which is why everybody who has ears and a bit of experience will confirm, that there are differences in the perceived timbre of loudspeakers that measure in the laboratory almost equally flat, yet have different degrees and polarities of phase distortion. Speakers with trailing low frequencies will sound thin when reproducing transient rich bass signals, to name only one effect.

How phase distortions are well noticeable at values of 1 ms and bigger is well documented. (see my links above)
And how it plays a role in localization etc. starting at phase changes around 10 µs is also well documented. (see links above)

So you think that you can hear a one cycle delay in a KH120 at 50 Hz? If that's the case, then how did you un-delay the low frequencies? Describe how you were able to remove ONLY the low frequency group delay without disturbing ANYTHING ELSE. How do you know that you aren't hearing something else?

Well, I HEAR the KH120 has a muddy and smeared bass transient response, compared to other speakers who are measured with a less drastic phase distortion in the bottom yet with comparable frequency response.

It is therefore reasonable, to correlate these two phenomena.
Probably the KH120 designer overdid the EQ in the bottom, to make the frequency response look better on paper, on the cost of phase linearity. A Pyrrhus victory from a sonic point of view, but probably good for marketing.

So how do you (and everyone else for that matter) know that you aren't listening to loudspeakers that exhibit worse group delay performance than the KH120 if you don't have any data to compare? So' I'm calling your bluff - lets see some data instead of just blowing out a lot of hot air.

Again, it's not too complicated. Just listen. To low frequency transient rich material (e.g. bass drum, piano etc.)
And compare it to the real world source.

Decades of lousy phase response and grave phase distortions in the loudspeaker designs - due to the constraints of physics and analogue electroacoustics - have made our ear somewhat insensitive to these phenomena. But who ever experiences the revelation to hear a (reasonably) time correct musical sound reproduced over loudspeakers, will immediately understand, what he or she missed all these years.

[matt thomas]

Originally Posted by Audio X

I wasn't familiar with the signature range from Lipinski, but just checking now it looks like they are segment speakers which accompany the 707's to carry the low frequencies. I think the L707's are really good speakers maybe falling just a bit short of what the title of the thread dictates. 2c

Yup, pretty much, except that the segments are added to the 707a, which is the newer version of the 707. You can get the 707a's by themselves, or with one or two of the additional woofer segments. I haven't heard them, mine are the original 707's.

And yes, I agree that the original 707's would fall short of the title of this thread due to the lack of low extension. However, the range they do cover is superb.

matt

[Roland]

Originally Posted by audio ergo sum

different Loudspeakers with different group delays. Quads, Geithains, B&Ws, Genelecs, K&H/Neumann... Some have less phase shift and group delay than others. There are huge differences and they are audible, particularly in the frequency range below 1 kHz.

Now [Zwicker et al, Electroacoustics, 1984] mention a group delay of 1-2 ms, above which speech becomes smeared, "t" start to sound like "ui".

Yet we are told that ITU standards of group delay of 16 ms are acceptable... haha.

Many authors in numerous studies found the just noticeable threshold of the ear for time delay to be around 10 µs, which directly relates to transient perception.

Yet we are told that up to two frequency cycles of delay are inaudible. (for 50 Hz that would be 40 ms)

And someone seriously calls ME a troll? Amazing.

Yes, but you are talking about speech, and group delays of 1-2ms at what frequency? The "t" sound you are talking about is in a much higher frequency range where a group delay would be far more critical due to the much smaller wavelength. The information that I have seen suggests that there study only dealt with frequencies in the mid band, the ones you are discussing of 16ms are at very low frequency where the wavelength is in terms of several metres.

Your discussion of say a BD is not very accurate in that all the speakers you have mentioned will exhibit group delays of about the same as it is a function of the cabinet design, q of the speaker in the cabinet and the crossover. pretty much any ported design is going to be give or take around the 16ms, it can't be less. An IB cabinet can reduce this a little, but not significantly.

As an aside, I along with my recording work, do a fair amount of live sound work, where group delay will likely be a good deal worse due to limitations with speaker placement and there is no problem making a BD "slam" like crazy. Listening to a real BD (and I am assuming we are talking about one from a drum kit) there is much less low end than we would use in a recording situation, the sound of a BD on a record, really only has a passing resemblance to that of a live kit of drums.

That there are speakers that you prefer to others, that can be a very personal matter. In my many years of listening to hundreds of different speakers in different locations the one observation I have noticed is that speakers with good flat response and low distortion (measured) tend to sound good subjectively. I would be interested for you or anyone to name a great sounding speaker that doesn't.

Speakers like the B&W 805, Lipinski 707, Revels, ect, are all pretty much universally respected. I would further add that your assertions about the KH120, seem to place you in a minority of one, I know countless, well respected engineers that beg to differ with you.

[tomdarude]

My pick for the most truthful and revealing speakers would be:

Strauss MF-2 or GrimmAudio LS1 (3-way)


....and for mixing, while they don´t go far deep, I def. trust my United Minorities Ginkos

[tpad]

...Well, I HEAR the KH120 has a muddy and smeared bass transient response, compared to other speakers who are measured with a less drastic phase distortion in the bottom yet with comparable frequency response....

AND, so where is the data of these "speakers who are measured with a less drastic phase distortion"? I've yet to see any measured data other than what I posted. I'm not interested in your group delay platitudes, I want to see measured data of speakers comparable to the KH120, and I'll draw my own conclusions.

[Doc No]

Originally Posted by tpad

...Well, I HEAR the KH120 has a muddy and smeared bass transient response...

I don't hear that at all! Maybe the KH 120 are not expensive enough to HEAR perfect transient response? ;-)

[audio ergo sum]

Originally Posted by Doc No

I don't hear that at all! Maybe the KH 120 are not expensive enough to HEAR perfect transient response? ;-)

Or maybe they are cheap enough to NOT hear it. ;-)

[audio ergo sum]

Originally Posted by tpad

...
AND, so where is the data of these "speakers who are measured with a less drastic phase distortion"? I've yet to see any measured data other than what I posted. I'm not interested in your group delay platitudes, I want to see measured data of speakers comparable to the KH120, and I'll draw my own conclusions.

Besides of your arrogant attitude toward empiric evidence (we call it listening) - and your ignorance to above listed scientific data, which you can put aside as "platitudes" but so are telling us something about you, not about the subject at hand...

I don't have the data ready for presentation. I have seen a few phase graphs, but all have the most diverse scaling.

Someone should do the heroic thing and measure a few mainstream monitors's phase response in freefield conditions and present the results. Btw, measuring phase is not as trivial as measuring frequency response, since several corrections have to be made to get a valid result. One more reason why phase is the step child of electroacoustics, not because it isn't important, but because it is more difficult to do right.

I would predict few speakers ( in the pro-audio high-end league) having such a drastic (360 deg) phase shift over a narrow band as the KH120 below 100 Hz. For f>200 Hz the KH120 seems to be fine.

To my ears, comparable speakers like Geithain 906, Genelec 8030 and others have a smoother phase response below 200 Hz. Possibly by introducing less EQ in counteracting the natural LF rolloff, thus not sacrificing timing for bass energy as much.

Anyway, I would guess that much of the subjective sensitivity toward phase distortion has to do with our musical preferences and what our ears are used to regarding phase distortion. Someone grown up with modern pop(ular) music is used to hearing phase distortion all over. Many of the instruments used in pop music are unkown to us regarding their sound in a pure acoustical sound field without electroacoustical transmission. Phase distortion is imminent to pop music and time correct loudspeakers do little to change the random phase distortions already present in such recordings.

It's a different story for people who grew up with a lot of listening to classical/acoustical music and particularly with listening to instruments naturally "through the air" more than listening to electroacoustic reproductions. Those brains might have stored the acoustic patterns of natural transients and be more sensitive to phase distortions in electroacoustic transmission chains.

[Doc No]

Originally Posted by audio ergo sum

Or maybe they are cheap enough to NOT hear it. ;-)

good parried

[dubmunkey]

What about the same question but with a budget of £2000 ($1600)?

I'm dumping my vxts and am looking at Adam pa22

Is there any better for that price range?