Jump to content
The Klipsch Audio Community
Arkytype

Stereophile review of Klipschorn

Recommended Posts

'design by committee' is the worst idea for a true designer and only leads to compromise...

  • Thanks 1

Share this post


Link to post
Share on other sites
10 hours ago, PrestonTom said:

John Atkinson knows what he is doing...

It's my premise that the gentleman that you identified knows what he is doing, but that his motives are less than stellar.  I don't like what he is doing--it's dishonest.  It is my belief that the reason why this thread is here and is growing is that discussing the review really does matter. Talking in depth about the review in a systemic manner rather than just taking poorly aimed pot shots I think is important. And attempting to summarily dismiss this discussion is something that I find to be uncalled for and unfair.  I believe that this type of magazine review does unnecessary harm to Klipsch's products--particularly the Khorn.  I think that's a pretty important subject, actually.

 

10 hours ago, PrestonTom said:

However, why he chose to do it this way will remain a mystery. 

I think that it's not such a mystery. (You probably do, too.)  The reason why I took the time to write what I did--and it took a while to cobble together what I wrote--is that I believe what the magazine did does a great disservice to Klipsch and others that might feel that the Khorn has been almost a singular example of rebutting the easy opinions expressed by this magazine. PWK spent a significant effort rebutting these sort of "drive by" reviews, and also leading by example in his loudspeaker designs and educating his customers and his own staff. So it does matter because there are a lot of people out there (as evidenced by some of the responses here) that don't have enough information to separate the truth from assumptions, and that they actually take the words of a magazine that hasn't done it (and probably can't do it) over the words of those that have (Klipsch).

 

The point of talking about precedence of requirements in loudspeaker design is I believe important--so much so that I intend to extend the discussion to Jubilees and perhaps other closely aligned loudspeaker designs.

 

10 hours ago, PrestonTom said:

Guess what,  my Jubilees have peaks and dips (just like what Roy measures)...

 

25789578_RightJubileeSPLandphaseresponse.thumb.jpg.7a38e46a1170181c398dcf84877045be.jpg

 

Mine don't, but then again I've spent time getting to know them better via testing, updating settings, changing in-room configurations, and experimenting with different room acoustic treatments in order to significantly improve their acoustic performance. This work over the past 5 years is the basis of the above discussion on precedence of loudspeaker requirements.

 

7 hours ago, Arkytype said:

The person who is perhaps the most responsible for quantifying listener preferences in a loudspeaker is Floyd Toole. He has conducted hundreds (if not thousands) of blind listening tests at Canada's National Research Council with participants ranging from professional musicians to volunteers.

I'm very familiar with this work by Sean Olive--and championed by Toole. There are some real issues there that ultimately benefit Harman and its current product lines.  I'll wait on those discussions. Just note that there are reasons to not believe the requirements precedence that Toole apparently pushes, and that it is quite easy to rebut.

 

Chris

Share this post


Link to post
Share on other sites
On 8/21/2019 at 12:10 AM, Khornukopia said:

I love my Klipschorns. 

Most people that own them do--as well as quite a few that don't yet own them but wish that they did--along with a listening room that does them justice.  I think that the recent magazine article writers forgot that.  Richard Heyser didn't, however.  Since Heyser's Khorn review in 1986, things have changed, most notably the ability of virtually anyone to measure their own loudspeakers in-room using even more powerful measurement means--and that the cost of these measurement instruments is now free (except for the cost of the personal computer that most people already have amortized for other purposes than measuring loudspeakers, and the cost of a "good enough" calibrated microphone that costs ~$100).  This is probably 1% or less of the de-inflated cost of the equipment that Heyser used in 1986.

 

On 8/20/2019 at 9:44 PM, Schu said:

'design by committee is the worst idea for a true designer and only leads to compromise...

The Khorn wasn't designed by committee: it was designed by PWK (recently updated by Roy D, et al.).   All loudspeaker designs represent compromise.  It's how those compromises are determined that makes all the difference. If you use the requirements sets and their relative importance promulgated by Atkinson, et al., you tend to get the same results...that all sound about the same. 

 

If instead you recognize that there are people walking around that say that they much prefer the sound of Khorns to those many thousands of "monkey coffin" designs praised by Atkinson, et al., then you have the opportunity to learn something new by examining why that is.

 

Chris

 

  • Like 1

Share this post


Link to post
Share on other sites

Chris, I am not sure we disagree.

 

That is a very impressive graph and you have done a great job especially around the "trouble spots" at 100, 200, and 300 Hz. Given my own experience, I suspect not all of that has been cleaned up with DSP and you cleverly used some acoustic (not electrical) strategies. You are to be congratulated. The factory measures come nowhere close to that.

 

Getting back to the original point however, Klipschorns and Jubilees will have peaks and dips (with factory delivered cabinets and measured in the "correctly loaded" space). That does not seem to diminish anyone's enjoyment of them. I do agree that folks who do not have the ability to listen to the Klipsch offerings ahead of time are bound to rely on the various "measures" floating around. They should be measured appropriately prior to publication. 

 

At the end of the day, John Atkinson and Art Dudley, have little impact on my enjoyment.

 

Good luck,

-Tom

Share this post


Link to post
Share on other sites
17 hours ago, Chris A said:

....the following list of revealed capabilities [are] based on the Khorn's performance in each of the measured areas [that are identified]:

 

  1. Full-range directivity (particularly below 800 Hz)
  2. Modulation Distortion
  3. Compression Distortion
  4. Efficiency/Sensitivity
  5. Cumulative spectral decay (especially below 800 Hz)
  6. Room dimensions/loudspeaker placement
  7. Near-field room absorption around loudspeakers

The following list of loudspeaker capabilities are significantly depreciated with regard to those preferring the Khorn's sound reproduction:

  1. Frequency response flatness (particularly below 200 Hz)
  2. Impulse response
  3. Input Electrical Impedance

___________________________________________________________________________

Continuing with this line of inquiry, it seems to me that there are people that are listening to something quite different than flat phase/SPL direct radiating loudspeakers for a reason.  The iconic example of this phenomenon is the Khorn and the people that love them, in my view.  (There are other offshoot listener groups, such as those that prefer "full range drivers", planar dipoles, quasi-omnidirectional, and even more design types.) These groups of listeners eschew the typical monkey coffin designs that have been discussed, above. 

 

So what is unique about the Khorn lovers?  The first set of factors (#s 1-7 above).  Even to the point of dropping the requirements of the following three requirements (FR flatness, impulse response, electrical impedance), these people are quite happy with what they've got.  Especially when they achieve a good listening environment and have found a way to equalize the on-axis frequency response somewhat using amplifiers, preamps, and the like. 

 

Some Khorn owners go further: they replace the midrange horns/drivers and even the tweeters, crossovers (including Zobel networks to tame the input impedance swings in output SPL when using high output impedance amplifiers), and false corners/enclosed backs, etc.  What's most important to note, however, is that for the people that don't change their Khorns to do these "enhancements" (changes), the existing performance of the Khorn is enough for them.  This means that the loudspeaker requirements set and their relative importance are not monolithic from all listeners, as Olive/Toole would have you believe.  There are significant groups of listeners that have been systematically screened out of the group that Harman performed in its "listener qualification" trials that resulted in their final reported requirements precedence(s).

 

"So what?"

 

There are obviously large variations in what people prefer and listen for.  In the case of the Khorn, we have a body of technical papers and articles that explain these differences (low modulation distortion, higher efficiency, full-range directivity, etc.).  Atkinson, et al. have also tried to screen these type of listeners out--and to glom onto the "Harman bandwagon" in that regard.  The interesting thing is (anecdotally)--this group of Khorn listeners also tends to include professional musicians (formally trained, etc.). 

 

Now isn't it interesting why that might be the case?

 

Chris

  • Like 1

Share this post


Link to post
Share on other sites
25 minutes ago, Chris A said:

Continuing with this line of inquiry, it seems to me that there are people that are listening to something quite different than flat phase/SPL direct radiating loudspeakers for a reason. 

 

In the near field, you can move a foot in any direction, or turn your head to the side, and significantly change the spectral content of whatever you happen to be listening to. That doesn't only apply to loudspeakers; try it next time you're in a face-to-face conversation. The side-effect of this is that we have learned to be rather tolerant of non-flat frequency response, as long as it's not egregiously bad.

 

Quote

The iconic example of this phenomenon is the Khorn and the people that love them, in my view.

 

I don't happen to own Khorns at the moment, but they are among my favorite loudspeakers. I acknowledge their shortcomings, but they reproduce dynamics better than any other loudspeaker that I have ever heard. That alone makes them special.

 

Quote

There are significant groups of listeners that have been systematically screened out of the group that Harman performed in its "listener qualification" trials that resulted in their final reported requirements precedence(s).

 

That is an interesting, and significant, observation. It means that the entire study is tainted by confirmation bias. Basically, they have assumed what they are trying to prove.

 

Quote

The interesting thing is (anecdotally)--this group of Khorn listeners also tends to include professional musicians (formally trained, etc.).

 

Many years ago I brought my uncle, a professional musician, to a local high-end shop to listen to Klipschorns. His immediate comment was, "It's like I'm right there in the orchestra with them."

  • Like 1

Share this post


Link to post
Share on other sites

This issue of "screening" is always a source of problems in decision support models (and I've done a lot of those models over my career).  Screening of participants and other data sources is also a sure-fire way to spot confirmation bias tendencies.  One of the techniques that is used by QFD professionals (Quality Function Deployment) is a "customer segmentation matrix".  The following notional (i.e., non-calibrated) matrix is for home theater audio customers:

 

1413139509_ExampleCustomerSegmentDemandMatrix.thumb.JPG.b45ddad81b3fe16391cb9cc4bc764155.JPG

 

The left-hand column has the identified customer segment types.  The customer segments are usually identified via affinity analysis of interview data.  The needs of the nine (9) identified customer segments in the notional example differ according to the desired/required product attributes in this case.  Kano product analysis tags (B, O, A) and Christensen Product Hiring model attributes (GB, ETB, ETU, LP) are also included in the "Hows" across the top of this matrix.

 

All of the above is provided as a cursory example only--that the analysis done by Olive, et al., is much too much focused on "getting the corporate answer" than "getting the truth".  If Olive had instead hired himself a QFD professional before starting his big decision model, he would have gotten much different results (and he may have, and was later directed by corporate upper management to change his results in order to advertise their work for corporate benefit). 

 

I've seen this issue a million times (it seems ) as a practicing operations analyst for many years, and as a QFD practitioner for even more years.

 

Chris

  • Like 1

Share this post


Link to post
Share on other sites

So let's see how the set of capabilities (in approximate rank order) do when we look at the performance of the  "Klipschorn II"--the Klipsch Jubilee relative to that of the original Khorn:

 

Capability

Khorn

Jubilee

1

Full-range directivity (particularly below 800 Hz)

The first production home hi-fi loudspeaker with full-range directivity, 70+ yrs. of production

Significantly improved bass bin polar directivity at higher frequencies due to reduced splay angle & distance between the two horn mouths);
The K-402 horn/K-691 has much better vertical and horizontal directivity;

crossover (~450-500 Hz) polar coverage angles vertically and horizontally without significant coverage angle mismatch or lobing

2

Modulation Distortion

The Khorn’s horn-loaded bass continues to impress experienced musicians, et al.

The Jubilee bass bin has uniformly lower harmonic distortion (HD) than Khorn. Harmonic distortion is an indicator of amplitude modulation distortion (AMD), which is the most significant form of modulation distortion at low frequencies;
2" throat compression driver on K-402 horn significantly increases the surface area of the diaphragm

3

Compression Distortion

Both the Khorn and Jubilee have inaudible levels of compression distortion at home hi-fi SPL which is much less than direct radiating loudspeakers.

4

Efficiency/Sensitivity

Both the Khorn and Jubilee have very high efficiency/ sensitivity which permit very large dynamic range operation even using very low power output amplifiers.

5

Cumulative spectral decay (especially below 800 Hz)

Both the Khorn and Jubilee have outstanding bass and HF decay characteristics due to their fully horn-loaded drivers.

6

Room dimensions/loudspeaker placement

The Khorn requires a listening room with a relatively wide lateral span of certain minimum dimension to maintain a minimum listening distance from the front face of the loudspeakers of 3-4 metres.

The Jubilee, by virtue of its time aligned drivers, does not require a large listening room or a wide spacing.  Its minimum listening distance is ~1 m

7

Near-field room absorption around loudspeakers

Both the Khorn and Jubilee are corner horns requiring room corner placement and near-field sound absorption or smooth room boundaries with no acoustically reflective furnishings out to ~2 m from each loudspeaker to suppress early midrange reflections.

1

Frequency response flatness (particularly below 200 Hz)

The Khorn’s FR flatness is nominally ±5 dB from 31 Hz—17 kHz

Jubilee frequency response flatness is a function of the dialing-in process, with ±1.5 dB from 30 Hz to 17+ kHz typical values that are easily achievable.

2

Impulse/step response

The Khorn’s IR/step  response is ragged

The Jubilee’s impulse response can be made to be near perfect via use of DSP PEQs and low or zero phase shift crossover filters. 

3

Input Electrical Impedance

Khorn impedance is ragged

Jubilee input impedance is a function of the inherent driver/horn impedance for the bass bin and HF channels (bi-amped & direct coupled)

 

  • Like 1

Share this post


Link to post
Share on other sites

So the point of the above table is to show that the latter incarnation of the Khorn (the Jubilee) addresses the perceived deficiencies of the original Khorn. 

 

It should be noted that the "Khorn I" (in its various minor change incarnations accommodating driver changes over time due to supplier changes) has a very large installed base of devoted owners that haven't yet seen the need to upgrade their Khorns.

 

Many new owners of the AK6 Khorn are very satisfied with their performance (certainly enough to buy them new--and keep them for life).  The AK6 design is an incremental improvement of the original Khorn design with enclosed bass bin back that decouples the need to seal against a room corner at 45 degrees, having an improved tweeter, and incorporates overall adjustments to the passive crossover smoothness.

 

Chris

  • Like 1

Share this post


Link to post
Share on other sites

Thanks again to ChrisA. It nice to see why I like my Jubes more than the KHorns.

Share this post


Link to post
Share on other sites

That is an interesting analysis Chris. Thanks

 

Allow me to quibble on one point however. I choose this point since many folks put an inordinate amount of attention and scrunity on it (even for differences that are sometimes trivial). That is the amplitude response (I assume on-axis and in an corner loaded environment). 

 

You stated: 

"Jubilee frequency response flatness is a function of the dialing-in process, with ±1.5 dB from 30 Hz to 17+ kHz typical values that are easily achievable."

 

Let me nit pick a bit. I really don't see this as "easily achievable" for most users. I believe it would take some perseverance. They would also need to set aside preconceptions about the analog vs digital debate. Some might have the perseverance, but many would be hard-pressed to give up on their notions of "digital" (look how many folks get caught up in minutia of wire, capacitors, MDF vs plywood, etc). I am afraid that the big picture of DSP improvements might get side-lined for far too many. 

 

I hope that you are correct and that I am wrong. Someday, there may be a better acceptance of bi-aming and DSP. Until then ..... folks are going get worried about frequency response graphs that are "supposed to look a certain way". The graphs for Klipschorns and Jubilees will not satisfy them (no matter how good the systems sound, and how much better they could eventually sound).

 

Good Luck,

-Tom

 

Share this post


Link to post
Share on other sites

It seems to me that there are no issues with the DSP bi-amped Jubilee (...and you should know--since you've owned a DIY pair for over 12 years). 

 

To introduce that hackneyed memeplex (i.e., anti-digital bias) yet again is to me like a reviewer looking for something to review once finding nothing significant to criticize in a loudspeaker (like the Jubilee)--so they go in search of a memeplex to introduce into the argument in order to muddy the waters to provide a reason for continuing to promote his/her chosen profession.

 

:wink:

 

Chris

Share this post


Link to post
Share on other sites

I am not sure who the reviewer is that you are referring to. 

 

I have found the Jubilees to be rather simple to setup and adjusting the crossover (when the driver is upgraded) is fairly straight forward when you have time alignment and CD coverage. I would not under estimate the "anti-digiatal bias" and I would not think that I am trying to "muddy the waters".

Share this post


Link to post
Share on other sites
18 minutes ago, PrestonTom said:

I would not think that I am trying to "muddy the waters".

Now is as good a time as any to change to a bit more productive topic of conversation...at least it seems to me...

 

39 minutes ago, PrestonTom said:

...I really don't see this as "easily achievable" for most users. I believe it would take some perseverance.

Anyone with an email account, a calibrated microphone (such as a USB microphone) and a personal computer--and an afternoon to take a few upsweep measurements in-room--can achieve the levels that I stated (if their listening room reverberation time and early decay time pass muster--which is required to achieve an SPL flatness level stated).  I don't think that requires "perseverance".  In fact, the ~3-4 dozen people that I've helped to dial in their systems to date (remotely via emailings of REW measurement files and returned .xdat files) would likely think that it is no more demanding than using some "room correction software" that's ubiquitous in hi-fi usage today to correct for at least the below Schroeder frequency SPL irregularities in-room.

 

Chris

Share this post


Link to post
Share on other sites

If you were able to get all those systems, measuring comparably to what you showed earlier, then you are to be congratulated. My experience with seeing folks use room correction software, has not always been positive. 

Share this post


Link to post
Share on other sites

How's life going in general, Tom?  Any new recordings of note?

 

I've acquired a few classical Blu-Ray performances that are spectacular.  I wish you could hear them on the 5.1 here since I broke to code on how to do Danley-style crossovers with zero phase shift. 

 

Chris

Share this post


Link to post
Share on other sites

I'm getting closer to going digital, but my funds are limited. I am slowly getting some hardware inventory ready so I can get working on it. The simple act of moving my tweeters convinced me.

 

Bruce

Share this post


Link to post
Share on other sites
3 hours ago, Chris A said:

How's life going in general, Tom?  Any new recordings of note?

 

I've acquired a few classical Blu-Ray performances that are spectacular.  I wish you could hear them on the 5.1 here since I broke to code on how to do Danley-style crossovers with zero phase shift. 

 

Chris

Perhaps you have me confused with someone else?

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...