Jump to content

Review and Discussion of Toole's Book, Third Ed.


Chris A

Recommended Posts

On 12/11/2020 at 7:52 AM, Chris A said:

The discussion in the above post ^^^ explains why the "reviewers" on other forums simply miss the point with Klipsch loudspeakers, and why so many people prefer them to "correctly designed and implemented" loudspeakers (in their opinion), usually little monkey-coffin boxes on sticks that measure well, sound okay at low SPL, but simply cannot reproduce the music like fully horn-loaded loudspeakers (i.e., including horn-loaded bass bins). 

I have a layman's hypothesis for why this is.

 

The reason we love the Klipsch sound is because it is a "live" sound. It accents certain elements, amplifies detail, and makes a wide as-well-as deep soundstage. And they sound great loud! Horns are a better vehicle to deliver that live sound because they electrically, mechanically, AND acoustically amplify and deliver soundwaves. This makes horns interact with the space/room surrounding them in a particular way.

 

In order to create/maintain this liveliness (this particular "brand" of liveliness was envisioned and developed by PWK and is what Roy continues to develop), the horn driven speaker must have a specific FR curve (I know that it's more complex than that [transfer functions, etc.]).

 

I have built crossovers for my Klipsch speakers that measure very flat. But they lose the Klipsch Sound. I always end up going back the original FR curves with only minor modifications.

 

Anyway, as I learn I will continue to see how and/or if my hypothesis holds up. 

Link to comment
Share on other sites

7 hours ago, MechEngVic said:

I have a layman's hypothesis for why this is.  The reason we love the Klipsch sound is because it is a "live" sound...

This is a true enough.  Why is it that those people that don't like the sound of horn-loaded loudspeakers (horn-loaded loudspeakers that have been designed and implemented well, that is) fail to acknowledge this critical difference in sound quality?

 

This is to me a statement of the divide between horn-haters and horn-lovers that will not cease until that "liveness" is acknowledged and explored by those who say they don't like the horn sound.  (Notice the one-sided nature of this dilemma.)  To illustrate, I'll relate a short story that occurred last night...

 

In the evenings, my wife and I knock off from our daily activities to prepare and eat dinner (around 5-6), feed the dogs/cat, and then retire to the setup: Jubilees/K-402-MEH/AMT-1-Belles/TH-SPUD subs) for video streaming and video discs. 

 

Last night for our dinner hour, I was playing ripped DSD surround music from a USB external drive from my SACD collection (around 100 discs or so) through my Oppo BDP-103.  We played a Pentatone multichannel SACD (the former classical division of Philips), with Mari Kodama on piano performing Beethoven's 16th-18th piano sonatas in 5.1 format, native DSD:

 

images?q=tbn:ANd9GcSrY496Sv4ohZMi-QoiPHc

 

As we finished up in the kitchen and moved to the setup in the same room, I had turned up the volume on this recording to something approaching 80-82 dB to listen to the excellent recording.  My wife sat down and listened to the music for a while while I was tending to the dogs.  She spontaneously mentioned how good it sounded.  I concurred and said that it was the reason why I had turned it up (something that I usually don't do before playing video discs and streaming).  It sounded surprisingly and breathtakingly dynamic/live--just like a real grand piano sounds, with a completely smooth texture, with no defects in timbre/tone or audible distortion (which is uncommon while playing classical piano recordings on hi-fi setups).  No other "extras" were present--little evidence of compression on the recording, no artificially hard edges in timbre, no defects in the (life-size) soundstage, no attenuation of lower registers.  A subconscious feeling of "you are there" was apparent. 

 

I knew that this was a difficult thing to achieve.  I have spent a large amount of time dialing in the room and the loudspeakers, but it is something that I learned the hard way (I hope to help others learn this much more quickly and easily so they too can begin to concentrate on the music instead of hardware).  I now enjoy the music, and collect good recordings instead of having to tinker with hardware and software/firmware.  This is why I spent time on the setup, correcting its transfer function performance (SPL and phase response), as well as tailor the in-room acoustic treatments and loudspeaker placements.

 

This is something that would never be achieved by using direct radiating loudspeakers of any sort. I was walking around behind the listening chairs (off-axis), and she was sitting to the right of the traditional sweet spot. 

 

And this is where the story actually begins: any "Klipsch sound" has no place in this scenario.  What's necessary in this case is neutral sound...something that others that pan horn-loaded loudspeakers try to achieve with their direct-radiating loudspeakers, but without first solving the more important issues of low distortion and directivity control mentioned above, and fail to achieve the experience of "liveness" because of that oversight.  Getting neutrality of sound is easy using fully horn-loaded loudspeakers via tools and hardware/software that's widely available and affordable nowadays. 

 

[If you personally want to accentuate your bass or treble, or compensate for creative mastering EQ used by popular music mastering guys (a one size fits all approach), that's your prerogative.  I instead use demastering EQ, etc. to separately correct recordings that have issues .  The recording we were listening to had none of those issues.]

 

So this means that the typical "Klipsch sound" rise in mid-bass response and upper bass is EQed out, as well as any irregularities in HF response anywhere in the audible spectrum, including a drop-off in SPL above 12-14 kHz, is EQed flat as well as the typically large phase growth through the crossover region(s) are flattened, too.

 

All Klipsch loudspeakers have transfer function issues if not dialed into their in-room placements:
 

  • What is needed is a means to measure these SPL/phase irregularities (REW shareware and a UMIK-1 is sufficient), and a way to flatten both the SPL and phase response (DSP crossovers).  Doing this using the old analog EQ boxes is usually accompanied with a big loss in fidelity and in higher noise levels). 
  • Trying to use only upstream digital EQ from the preamp (like in a PC) leaves you with the issues around the crossover regions and back-EMF disturbances in the sound quality from moving mass effects of lower frequency drivers on higher frequency drivers. 
  • But these things are a snap to do nowadays using DSP crossovers and some form of "room correction" or FIR filtering, and so transparently that you cannot hear any change in fidelity.
     

What's this got to do with Toole's book?  A lot. This is probably the crux of all the issues that arise if you try to use Toole's book (only) to achieve good sound reproduction.  This subject will be raised multiple times in coming exploration of Toole's book, chapter by chapter.

 

Chris

  • Like 1
Link to comment
Share on other sites

PWK once said music is dynamics and no speaker is more dynamic than horns. Klipsch speakers were not reviewed well in the rags of the 70 and 80's because they only looked at lab results and frequency response. My first serious speakers were the AR3a's which were well rated as having a flat frequency response and low distortion. Just to keep it short dull and boring, not dynamic. Fine to use while reading a book. Bought LaScala's in 80's and have never looked back. I have tried full range speakers in an open baffle setup and they do sound good but detail is missing. I can  listen at low volume levels and the music still sound dynamic with my LaScala's. My best friend who recently died from Covid preferred the full range setup so all I can say is different strokes for different folks. I will stay with detail, dynamics and super low distortion.  

  • Like 2
Link to comment
Share on other sites

One of the proponents of "full-range drivers" (actually midbass through treble) in the past is Nelson Pass--notable impresario of Amp Camp amplifiers, First Watt, and Pass Labs, has always been stuck on full rangers with some sort of bass bin or subwoofer to fill in the large missing gap. I think that these guys are also trying to hear the detail and liveness, but without the bottom end, and having large amounts of modulation distortion and impulse response irregularities, they leave a lot to be desired from enthusiasts like myself.  But I do "get" what they are after. 

 

Nowadays, if you will accept loudspeakers the size of a Danley SH-96 (i.e., the size of a K-402-MEH), you can have it all, but you've generally got to use the modern tool called a DSP crossover to get there.  Danley sells the SH-96 in slightly prettier finishes (just like the SH-50):

 

dd9d8cfa_vbattach209354.thumb.jpeg.f76579dfb57c1a0badd2dc2b580bd42c.jpeg 

 

1156003924_SH-50veneered.jpg.d160d51abe656a9cb70d0b70db6a1757.jpg

 

...and the SH-96:

 

697083311_TheHardpopSH-96s.jpg.6b6ccac70125e264c0cf0b6b49692390.jpg

 

 

...but they charge about 2x to 3x what I think they should to do it (i.e., the SH-50s go for ~$3K USD and the SH-96s go for ~$8k USD, each).  That's a lot. Today's Klipsch Heritage prices are themselves too high by more than a factor of a little over two--based on historical prices and using the government's CPI inflation calculator.  So the Danley prices are probably not greater than Klipsch's, but no deals can be found with them either.  Consumers aren't getting break on price in either case.  However, these can be made DIY for half or a third of these prices fairly easily if one can build a box and a dual flare plywood horn...and using the best drivers available (very good woofers and BMS4592ND dual diaphragm compression driver).

 

First generation Jubilees are equally as good subjectively (in my opinion, albeit about 3x the height of the K-402-MEH or SH-96), but including the added wood veneer front panel on the bass bin. 

 

Perhaps the second generation Jubilee promised by Roy (now called the Heritage Jubilee) will be released this spring--but I fear the prices that are going to be asked.  I do recommend moving the K-402 downward and forward a little to remove the hiccup in the directivity at the crossover point...

 

759685053_HaasJubessmall.jpg.2bb46bec188eaa4c61063d5055592d73.jpg

 

...and use a zero phase growth crossover arrangement that I've also detailed on this forum)...

 

TAD TD-4002 Jubilee (Red Trace) vs. Danley SH-50 (Green Trace) Phase.jpg

 

Chris

 

Link to comment
Share on other sites

Nelson Pass is who got me interested in full range drivers. I bought some 6" Fostex speakers mounted them and designed and built a crossover where I could use my LaScala bass bin for the bass. Along with my sub they did sound good. I could not fault them other than loss of detail which was evident to me on music I was use to. The setup is now stored somewhere in my house. 

Link to comment
Share on other sites

Continuing on with Toole's book...

 

One comment that I thought that I'd make.  Toole talks the JBL M2 up in his book, but was made recently aware that the M2 likely requires more DSP crossover filters to flatten its SPL and phase response than 15 PEQs per channel. 

 

JBL_0.jpg

 

That's a lot, and helps to remind everyone that JBL M2s require the use of a DSP crossover. That's a pretty important data point for the mossback audiophiles that think that JBL/Harman can do no "wrong"...in their eyes.  DSP crossovers are the future of home hi-fi, as they have been now for at least 10 years in the studio monitor market, and 20 years in the PA/cinema market.  And with the capabilities and performance of those units like the Xilica XP or XD series, it's basically a no-brainer.

___________________________________________________

Toole talks at length on "spin-o-rama" and various off-axis and room-bounce SPL response curves. These are good, but in context of basically the total absence of phase/group delay response and its correction, I think that the focus on SPL response is a bit over the top (and also considering that it's basically correctable as I discussed in this post) and everything else is a bit underdone, especially distortion at elevated sound pressure levels (SPLs).  Toole says that distortion isn't a factor.  I call BS on that. 

 

3f995-152B-2B1.jpg

 

I really believe that he's trying to suppress the fact that horn-loaded loudspeakers do have large advantages over direct radiating ones (a subject that he's basically deleted from his 1st/2nd editions of his book in his 3rd edition).  If he were to admit this, he'd be putting one of Harman''s biggest rivals (Klipsch) into the driver's seat.

___________________________________________________

Within the narrow construct of Chapter 5 (Characterizing Loudspeakers), Toole's focus on getting decision models that work based on the factors that Sean Olive identified in his patented process have actually worked (which is the source of a bit of a chuckle for me--since decision models for just about everything have been around for a long time--and Harman isn't the first to use them...just ask the "big data" folks about decision models and watch their incredulous reactions to the notion that they're "new"). 

 

The problem is that Toole and Olive didn't include the factors that really do make a difference, like low modulation distortion at elevated SPL (well above 80 dBC) and the subconscious effects of having both good directivity and very low phase growth together--that these two guys forgot to add into their decision criteria (...and I'm being nice when I say "forgot", because I have reason to suspect these two factors were deliberately omitted for reasons of marketing by someone at Harman/JBL...).   Those two factors probably yielded the "wrong answer" for the marketing guys.  It's probably not a mortal sin, because I've seen the same thing happen many times in the corporate world, but it is at least a sin nevertheless.  It's a bigger problem if Toole says they're not factors (and Toole did commit this sin). 

 

Chris

  • Like 1
Link to comment
Share on other sites

Chris, the book sounds very interesting and exciting for those interested in the mechanics of sound reproduction. I just now looked over the entire thread. At one time the subject and book would be something I would jump at to get. I hope some of the younger members on this forum take an interest in the acoustics of sound reproduction from this thread. Great thread for this forum. 

Link to comment
Share on other sites

Chapter 6 (Loudspeaker/Room Systems--An Introduction) is a very brief preparatory chapter that introduces some ideas:

 

1) The low frequency/room mode frequency band, the higher frequency (midrange and treble) frequency band, and the transition between the two: the Schroeder frequency.

 

2) The flawed notion of treating the "first reflection points" in-room (instead of treating just the extreme near-field around the loudspeakers--within the first wavelength or two).

 

3) Toole introduces the precedence effect, and the ideas of spaciousness/envelopment, immersion, timbre changes in small rooms, and speech intelligibility

______________________________________________________

 

Chapter 7 (Above the Transition Frequency--Acoustical Events and Perceptions) is a much longer chapter (58 pages), so we'll take it in sections.

 

Section 7.1 deals with early reflections, and introduces a very important psychoacoustic issue with stereo imaging...the "fundamental flaw of stereo" which shows up as a perceived dip in stereo response about at 2 kHz due to the interaural distance (space between the eardrums).  For those not familiar with this phenomenon, this is an interesting read.  I've experienced this effect directly using K-402 horns and listening to them directly on-axis.  You will hear a dull sound when you suppress the early reflections from K-402s by listening to them directly on-axis.  This is the fundamental flaw of stereo in action, and it has nothing to do with measured SPL response or phase, etc.  You will need a "dummy head" (e.g., KEMAR head, etc.) measurement using two microphones embedded in the dummy head where human ears would be located.  When you measure a stereo signal directly on-axis using the dummy head and using flat frequency response, there will be a dip in SPL response with the two "binaural" microphones hooked together.  This is one of the first steps that audio enthusiasts make when learning about the effects of how we hear.  It's the room's early reflections that fills in that hole in response.  It's an interesting experience the first time you hear it, and it will make a believer out of you.

 

ll3headsfig4.jpg

 

Understanding the implications of the KEMAR head experiment will help you in understanding what the limitations of stereo are, and why we have multichannel music.

 

Section 7.3 deals with wall reflectivities and absorption.  This is where you'll read about diffusers of different types.  I see a lot of hype about diffusers, and I hope that you'll see the subject as I do: diffusers are great, but not terribly useful unless they are of broad band effectiveness, and this is where most diffusers fail--they are typically only effective over somewhat narrow frequency bands, say 500-2000 Hz.  While it is important to note that the reflections in-room do give you a sense of envelopment and immersion, it is equally important to note that using diffusers having limited bandwidth effectiveness can create as many problems as it solves. 

 

Section 7.4.2 dives into the effects of loudspeaker directivity on subjective sound quality assessments--sound quality and spatial quality--of three notable loudspeakers still in production in 1985: the Rega Model 3, KEF 105.2 and Quad ESL63:

 

2+regas+full+front.jpg 

Rega Model 3

 

105-2-high-fidelity-speakers-from-kef-se 

 KEF 105.2

 

quad-3551.jpg

Quad ESL 63

 

This section of chapter 7 (7.4.2) is actually one of the more important sections of the book, in that it first discusses the prime importance of power (polar coverage) response of the loudspeakers, which is actually much more important than on--axis SPL response (incorrectly called "frequency response" without also showing the phase response) since on-axis SPL response is correctable via EQ.

 

One other point that Toole makes here is the importance of listening to loudspeakers in mono rather than stereo in order to hear the differences in sound and spatial quality much more clearly.  Figure 7.14 show the relative subjective performance of the above three loudspeakers in mono and stereo, and this one difference produces 3x to 4x the differential sound quality ratings using mono only.  For those wanting to audition loudspeakers both in their own listening rooms and in unfamiliar listening spaces of commercial sales or shows, it is much more revealing to listen only in mono.

 

Musical genre also has some effect on subjective ratings, too.  Toole's examples include choral, chamber, jazz and popular (assumed to be a mixture of rock and pop).  So this really does support the theory that loudspeakers perform differently using different genres of music.  In general, Toole assigns the subjective performance differences to the different levels of lateral reflections, apparent source width (ASW--a measure of the subjective soundstage size of a single loudspeaker), and the early reflections from the nearfield just around the loudspeakers.  The problem is, he didn't actually test these assumptions, so it's still unclear what the source of the music genre-specific performance of loudspeakers actually are. 

 

It could also be associated with different modulation distortion levels of the loudspeakers, and even their phase/group delay response, since some music genres also thin out the timbre textures (jazz and chamber music) while other genres have more densely packaged textures that would show up the higher levels of modulation distortion.  Since Toole has already screened out distortion (modulation, impulse, phase/group delay, etc.) as sound quality decision variables in his book, it's unclear what the source of the differences actually are.  The KEFs are three-way loudspeakers, and all other things being equal, will present lower levels of modulation distortion (i.e., "the mud factor" or opaqueness) than the two-way Rega, and that shows up in the more complex musical timbres as the KEF wins on these genres. 

 

Also more complex harmonics of violins, violas, cellos, etc. will show up much more clearly if the loudspeaker displays more or less phase growth vs. frequency.  Strings sound "steely" if their phase and group delay response is steeper.  Flat-phase loudspeakers will have a smoother sound with more listener involvement (if the early reflections around the nearfield are controlled).  It looks as if all of these factors are playing some role in the genre-specific subjective differences between loudspeakers (two-way, three-way direct radiating, and electrostatic flat panel bi-pole radiator).  

 

Chris

Link to comment
Share on other sites

Section 7.4.3 [Loudspeaker Directivity and Wall Treatment Together - Choisel (2005)]:

 

This section details the same effect that I found (quite independently) that nearfield absorption around the loudspeakers--within the first 3-5 ms (4 to 6 feet) be used to control early reflections.  Essentially what Choisel is saying is that, by the time you get out to 9.5 ms (10.7 feet) away from the loudspeakers, the effect of suppressing nearfield reflections via absorption, etc. was found to not affect the resulting "image localization" (a.k.a., soundstage image).  All of my room treatments are within this 10.7 feet radius of the loudspeakers, even the two bass traps, one in each corner above the TH subs, but for the reason that they can't be easily located anywhere else in the room.  Bass traps generally affect the sound field below the room's transition frequency (i.e., Schroeder frequency), which is the subject of the next chapter--chapter 8). 

 

One of the largest advantages of of having Jubilees in the room (as well as Khorns, La Scalas, and Belles) is that they hold their directivity down to ~100-200 Hz, below the room's transition frequency.  Toole never talks about this and how it affects the soundstage imaging.  If you use some absorption to absorb early midrange reflections using these loudspeakers in the room corners, you have a capability that other direct-radiating loudspeakers simply do not have, even those with multiple 15" woofers will lose their directivity below 500 Hz--such as the KPT-415-LF bass bin, while the other horn-loaded bass bins of the mentioned Klipsch loudspeakers (and I might add, the other full-range multiple entry horns of Danley) hold their directivity down to at least 200 Hz:

 

 IMG_2164.JPG

The K-402 on top of two different bass bins in the room corners--
and only one of these bass bins can control its directivity below 500 Hz: the Jubilee bass bin on the right.

 

Why is this so important?  Because soundstage imaging usually exacts a very large price in almost all audiophiles' listening rooms.  Everywhere you look, audiophile advice says to move loudspeakers away from the room corners...ostensibly in order to "decouple from the room modes to avoid exciting them". 

 

This is actually bullsh!t :pwk_bs: : moving the loudspeakers away from the room corners robs the loudspeakers' ability to produce clean, effective bass without simultaneously adding huge amounts of modulation and compression distortion when playing at realistic playback levels.  Having loudspeakers that can control their directivity to just below the room's Schroeder frequency is a huge advantage, that, when combined with relatively small amounts of nearfield absorption just around the loudspeakers, yields a soundstage image that completely fills the entire room with clean, transparent bass, and outstanding soundstage imaging.

 

This is probably the most documented subject that PWK addressed: placing loudspeakers in room corners.  The reason why the "audiophile advice" is to move the loudspeakers out into the room (thus robbing the owner of significant real estate in-room) is because the loudspeakers they own do not control their directivity below ~500-1200 Hz (depending on effective woofer diameter), and therefore paint the corner of the room with uncontrolled acoustic energy.  So the advice is probably the worst thing that owners can do in terms of getting clean, uncompressed bass response, to make up for a serious deficiency in the loudspeaker designs themselves.

 

There is a bit more to this: I've found through my own experimentation that the K-402 Jubilee design, in either in the two-way configuration or three-way using a BMS 4592ND dual diaphragm driver, significantly outperforms the Khorn and Belle (and by similarity--La Scala) when it comes to soundstage imaging because of the K-402 horn can control its directivity much more consistently than the three other horn-loaded bass bin Heritage models.  (More on this subject later.)

 

This is a very important section in Toole's book.  Too bad he really didn't expand it to include the most important loudspeaker capabilities associated with Choisel's 2005 article.  He missed a lot here, IMO.

 

Chris

  • Like 1
Link to comment
Share on other sites

Sections 7.4.4 [The Nature of the Sound Field (Klippel 1990)], 7.4.5 [Observations of an Audio Enthusiast (Linkwitz)], and 7.4.6 [Toole 2016] can be lumped into a single entry here:

 

Klippel points out that from subjective assessments, the limitations of loudspeaker's "feeling of space" account for either 50% or 70% of "defects" in the two subjective measures of merit ("perception of naturalness" and "pleasantness", respectively).  This has been a subject of interest for many years, and I believe a lot of people go to even extreme measures (in terms of which loudspeakers they pick) to achieve these capabilities.  In general, dipoles and bipoles have increased subjective depth of field (artificially, I might add), but the height and width of the soundstage is a factor that is typically factored high in the precedence of requirements for loudspeakers. 

 

However, the problems of planar radiators have a distinct limitation on the dynamic range of the loudspeakers and size of the listener's area within the room (i.e., "head in a vise").  

 

Line sources, including Air Motion transformers--AMTs, seem to do much better at dynamics and increasing the size of the listening area within the room, but due to the fact that the AMTs are in fact line arrays, some limitations exist with them with respect to the vertical coverage area at the listener's positions (while the horizontal coverage is quite good at 90 degrees included angle...from 20+ kHz down to perhaps 500 Hz, with outstanding consistency of horizontal area.  The limitation of vertical coverage angles can be addressed using a "shaded line array" (see paragraph #3), but note this includes a fair amount of DSP crossover use, at least three AMT-1s, probably using wings, and three separate amplifier channels to achieve.  While today, this isn't really a big deal to provide and even dial in (the AMT-1s are easy to dialed in terms of PEQs, even with wings), it is seen as somewhat complex by casual audio enthusiasts not possessing some skill with using DSP crossovers and multiple amplifiers, as well as using something like REW to dial everything in. 

 

This could be packaged into an add-on kit that would easily work with Klipsch Heritage loudspeakers and include one Xilica XP-4080 or a miniDSP 4-10 HD to do the DSP duties.  What this does is provide greatly increased vertical coverage, and fine control over the size and performance of the 500-20K Hz band, and when mated with horn-loaded bass bins (Jubilee, Khorn, La Scala, or Belle, or even a Peavey FH-1 or FH-2 bass bins), provides a fully horn loaded and controlled directivity solution for those that don't care for the size or aesthetics of the K-402s on top of horn-loaded bass bins, in like fashion to the CBT-like array:

 

index.jpg

 

While there are downsides to using such a design (probably reducing the good horizontal listening polar angle by a small amount), the upside is a "slim looking" high frequency unit that will more than keep up with the horn-loaded bass bin.

_____________________________________

 

Linkwitz's and Toole's narratives that are laid out in the book are worth reading, but perhaps don't have quite the meat of information that Toole's other chapter 7 sections have.  It's good reading, nevertheless.  However, I do see a failure to recognize the importance of the following loudspeaker/room placement factors:

 

1) lower loudspeaker distortion (especially modulation distortion),

2) using room boundaries for better performing bass production,

3) the lack of recognizing the need for nearfield absorption with loudspeaker in room corners (instead, placing loudspeakers having poor directivity control below 1 kHz out on listen room's floor away from the walls--where the owners and their guests can run trip over them while trying to get around them--not to mention the negative WAF factor for placing them out on the floor+), and

4) the advantages of using horns/waveguides to help in the polar control consistency vs. frequency, especially at lower midrange and bass frequencies. 

 

These are large blind spots I believe, and will eventually be found to be part of the "audiophilia" history of direct radiating loudspeakers, which I believe will continue to be questioned more and more over time by the types of audio enthusiasts that roam these pages...and which are likely to increase significantly percentage-wise as time passes.

 

Chris

  • Like 1
Link to comment
Share on other sites

Section 7.4.7 [Floor Reflections: a Special Case] is interesting in that the author points out that the human hearing system has largely accommodated for floor bounce issues--as opposed to wall and ceiling bounce issues which it has not accommodated so well.  This is an interesting psychoacoustic finding. 

 

Toole seems to excel at the psychoacoustics portions of the problem domain (or this author is trailing farther behind him in this specialty area).  But I think Toole's real expertise lies in the intersection of the psychoacoustics and the engineering of loudspeakers and listening spaces.

________________________________________

 

Section 7.5 [Profession Listening vs. Recreational Listening] is an area that I believe is more wrapped up with recording engineering (recording, mixing, mastering) practices than with the prime audience for Toole's book, that is, consumers.  There is "much ado about very little" here, in my estimation, and many opinions, but little real data.  If the typical consumer were not aware of these issues that the recording/music production staff face, they would not be missing much, in reality.  I think that the recording/music production staff would do better to use less multitrack recording techniques and borrow a lot more from the jazz/classical recording practices.  Recording costs and production times would also significantly be reduced if idea of "all musicians play together and are recorded together" would again take precedence in today's practices.  The music quality would also increase dramatically (assuming the "loudness war" practices are also overcome at the same time).  With the fidelity of home hi-fi systems, now multichannel systems, the past practices of multitrack music production are becoming much much the limiting factor in sound quality.

 

People generally like the effects of pleasant "second room" reflections from their own listening rooms (the first room being the original recording venue for the music).  But there are strict limits on the reverberation times and strength/correlation of the in-room reflections that must be adhered to. There will be a lot more discussion on this point as the chapters go by.

________________________________________

Section 7.5.1 [Hearing Loss is a Major Concern]

 

This seems to be a problem for Toole's and Olive's loudspeaker design decision models, and not really for the consumer marketplace.  Klipsch loudspeakers are fully capable of playing cleanly at much higher listening levels than most people might find comfortable, i.e. the Klipsch engineers have largely done their jobs. 

 

What I think the issue is--is that Toole apparently wants to encapsulate the consumer marketplace with their (Harman's) type of largely direct-radiating loudspeakers--direct radiating woofers/bass bins--and those people that hear differently due to hearing loss or even music training (amateur and professional classically trained musicians, etc.) throw a kink into his "one size fits all" loudspeaker requirements, at least in the requirements hierarchy that he is peddling with Sean Olive.

 

Hearing loss is a concern, but it's not really part of the loudspeaker design and implementation business--that's up to the consumers to protect their own hearing, and the better the loudspeakers at maintaining low distortion and consistent polar coverage (i.e., power response), the more there is a tendency for consumers to turn it up.  I can't say anything here other than this: tools of any sort can be used or abused, and sound reproduction systems are tools...but do not represent community health programs in and of themselves.  The larger issues of hearing loss are important...just not in a book about loudspeaker and room design, however.

________________________________________

Section 7.6 [Perceptual Effects of Room Reflections]

 

This is the section where Toole begins his decomposition of loudspeaker and listening room requirements.  Rooms can be of differing shapes and acoustic internal acoustics properties, but the human hearing system is capable of great accommodation without effort from their hosts.  Different loudspeaker types in differing rooms seem to come out in the same rank-ordered patterns regardless of the rooms themselves being used for auditioning. 

 

However, "me thinks he has an axe to grind".  If truth be told, the difference in performance levels of the different loudspeakers is subjectively not the same in a tiny little room as opposed to a large and well-engineered home hi-fi listening space.  I've heard absolutely spectacular performances in home hifi-sized rooms, and the same loudspeakers sound like excrement in very small/poorly performing (acoustically) listening rooms.  We've all experienced this.  What Toole seems to be doing is trying to tell us that all loudspeakers will improve or degrade by approximately equal factors depending on the size and quality of the listening space.  This is patently false.

_________________________________________

 

Section 7.6.3 [The Effect of Rooms on Speech Intelligibility]

 

This is a subject that I believe is passed off much too quickly by consumers (and by some in the music production business).  My experiences show that as the speech intelligibility of the loudspeakers increase (and there is a direct correlation to low modulation distortion levels and early reflections in loudspeaker performance), the overall subjective fidelity of the system increases, sometimes disproportionately.  Controlling early reflections is a major portion of this in home hifi-sized listening spaces, but so is low distortion (especially at elevated SPL), something that Toole disputes.  While it is good to control early reflections, and perhaps longer-delayed reflection levels in some larger listening rooms, this typically is an early reflection and reverberation time issue.  Toole spends a lot of time on this subject, and it's good for those wanting to go on to understand larger venues than home hi-fi listening rooms, my experience is that much of this subject can wait for those that wish to get involved in professional and religious music performance/production. 

 

Toole also introduces the precedence effect here (a.k.a., "Haas effect").  This is a central issue in how the human hearing system actually works, and is the controlling element in what can be achieved in small rooms.  I recommend highly to pay special attention to this subject.  Toole does a fairly good job of covering this, so I will not spend time on this subject, assuming the readers here will have their own Toole books to read and absorb this subject area.  I recommend thinking about this effect often and keeping this knowledge in the forefront of recall when considering listening room and loudspeaker issues in-room in home-sized listening spaces.

 

Chris

Link to comment
Share on other sites

Chapter 8 [Below the Transition Frequency: Acoustical Events and Perceptions] is Toole's chapter on bass performance and how we hear it.  Curiously, he starts out the chapter talking about room dimensions and room shapes, then looks more closely at how sound waves behave in rooms that are too small (in at least one dimension) to support a full wavelength.  This is a pretty good definition of the Schroeder frequency--where acoustic waves are so long that they no longer "travel" like light waves and don't behave like optical ray-tracing problems (first bounce locations on walls, etc.).  Acoustic waves in this portion of the spectrum behave like the videos that you might have seen of resonances in flat plates that are excited by acoustic transducers, and what you might have guessed is occurring inside loudspeaker bass cabinets. These acoustic waves no longer are seen in time-domain plots as passing waves.  We are now talking about standing waves for most home-sized listening rooms, whose pattern may modulate over time, but doesn't appear to "travel" to different places in the room.

 

In general, I find that the calculation of room modes from simple formulae is a waste of time.  It's much better (and nowadays much easier) to measure these modes near the listening positions or over an area of the room that is going to be considered for use as a listener area.

_________________________________________

 

Section 8.1.1 [Optimizing Room Dimensions: Does an "Ideal" Room Exist?]

Toole discusses the relative sizing of listening rooms to avoid deep nulls and peaks in the standing wave response vs. frequency (notably Cox, D'Antonio & Avis 2004, JAES Vol. 52, No. 6, 2004 June).  While I'm not sure that "best" room dimensions exist, I can say that "worst" relative room dimensions do seem to exist.  Avoidance of listening rooms in general with smaller linear dimensions than 12 feet (~3 m) horizontally and in depth, and square room dimensions is clearly something that many can agree on and that there is strong evidence that these room shapes/sizes are truly undesirable from an home hi-fi listening room perspective. 

 

If the listener is faced with such room dimensions, it is widely  recommended that headphones replace loudspeakers, or that a large internal reflective boundary approximating the respective interior wall size is used to artificially change the internal shape of the listening space to be non-square/non-cubic.

_________________________________________

8.1.2 [Are Non-Rectangular Rooms the Answer?], pg. 222. 

 

One of the comments made on this forum back in August of this year was that irregularly shaped or fan-shaped rooms will avoid standing waves. This is not true.  Toole in fact makes a point to address these kinds of misconceptions in this section.  I welcome comments from those readers who have digested Toole's point in this section and still think that non-parallel walls are the answer.  Toole mentions that only the calculations get more difficult in non-rectangular rooms but the standing waves are still there (figure below extracted from the 1st Ed. text)...

 

394649215_Figure8.7ComputedPressDistr(non-rectangularrooms).thumb.GIF.8e457cca449c670836949e577577a4a6.GIF

 

Chris
 

Link to comment
Share on other sites

Let's briefly survey the typical behaviors of home hi-fi owners (also including home theater [HT] multichannel owners):

  1. The interest is mainly on "buying stuff" instead of working on their setups.
  2. There's a disproportionate focus on electronics and players instead of room, loudspeakers, and recording quality.
  3. Little time or effort is typically spent deciding on listening room size and its acoustics, and many forget to add this to their view of setup performance.
  4. Many are satisfied with all of it looking good, regardless of its real contributions to sound quality. 
  5. The more loudspeakers in the array, the less focus on individual loudspeaker performance.

____________________________________________________________________

So, considering point #3 above (the room), it seems as if there is a disproportionate focus in Toole's chapter 8 on the room acoustics at low frequencies, the most of which is material that owners are unwilling to change.  Even the hint of modifying acoustic diffusion and/or absorption in-room is usually met with disinterest.  So why the entire chapter on bass performance?  Toole, like many academicians, tends to focus on technical subjects, but not take a step back to ask the question, "does it matter".  Much of chapter 8 I believe fails that test because few if any hi-fi enthusiasts are going to do much about it. 

 

So the entirety of section 8.2 seems to be rehashing material (the author's own assessment of that material).  Most of the material of interest will be in chapter 9 [Adjacent Boundary and Loudspeaker Mounting Effects], where the greatest gains (and losses) are usually found in real listening rooms.

 

Perhaps there will be re-visitation of material of section 8.2 later, but for now, I'll add or subtract nothing from what Toole has provided.

 

Chris

Link to comment
Share on other sites

On 12/17/2020 at 7:51 AM, henry4841 said:

PWK once said music is dynamics and no speaker is more dynamic than horns.

Bought LaScala's in 80's and have never looked back. 

I will stay with detail, dynamics and super low distortion.  

Purchased mine in the early 80s

and despite a lot of searching and auditions

Have not found anything I like better.

  • Like 1
Link to comment
Share on other sites

Chris,

Thanks for your great write up

and many other thoughtful comments and suggestions.

My response is to your clear and thoughtful review of the book.

Which sent me searching for 4 books I have, one I am looking under the furniture for.

 

Given that electricity, microphones and speakers have been with us a relatively short time. These are my thoughts:

 

The brain compares to what it knows, does it sound Natural. Things that are not natural can be irritating or jump out. The application of psycho-acoustics to trick the brain for low bit rate is perhaps the low-point of the Recording Industry; surpassing millie vannili and gang rap.

 

Greek and Roman theaters the players had to be heard and intelligible, some of the theaters are quite large, many are still considered acoustic marvels, resembling clam shells or a flared horn cross-section. Not sure how they calculated the design or did the math, but they did. The bottom line was selling tickets, then and now.

 

With electricity and amplification, this was a substantial R&D undertaking and attracted a lot of great minds, almost every engineer jumps at the chance to design audio gear today. The mass application that was a money maker was telephones. They had to reproduce uncle Floyd and aunt Betty well enough to be commercially successful.

 

I believe the mass introduction of telephones pre dated amplified sound in theaters, which also had to sell tickets. I'm guessing that a lot of AT&Ts theater sound development came out of the phone engineering work and engineering groups. The test recording for these groups were either exclusively male voice or 99%. So all engineering decisions and trade offs were based on the experience of listening to male voice. This is still with us today.

 

Pure analog mid 1970s systems forward, and higher sampling rate and

and better decoders and algorithms may have overcome some or most of the built in engineering bias. PCM and ADPCM were male voice biased along with the entire telephone network.

 

Consequently, as we depart the geometry of the amphitheaters, and introduce something less than quality  Movie theater size speaker cabinets and drivers, a series of trade-offs ensues. The math and geometry on the cabinets hasn't changed much since the 1950s. Analog electronics plateaued in the mid 70s and early 80s. Digital is about to do the same.

 

As to Horn loaded vs something else, I believe the horns remain unmatched in speed, accuracy and efficiency.

 

I have not heard a speaker smaller than a Heresy that I would purchase for listening to music. I have listened to a number of direct radiators the same size, but not Lifelike.

 

The magic of the LaScalas, is the ability to faithfully reproduce female voice and piano, the two most demanding that I can think of. And flexible placement.

 

An engineering and or architectural point of view requires substantial knowledge and engineering and real world knowledge base. Some rooms just sound better, we hear this over and over again from musicians. Sting recorded Ten Summoner's Tales in his estate's Kitchen, stoves refrigerators and all. The musicians loved it. Would a model have predicted the musician's joy in this room ?

 

How much engineering, EQ, Phase adjustment, or damping does it take to turn a shoe-box room into an amphitheater or a squeaking mouse speaker into a roaring lion movie theater speaker.

 

As you point out, what problem is the author attempting to address, a design reference book, or a consumer oriented set of guidelines that can be easily applied to our undersized sub optimal spaces.

 

As consumers, I believe we are left with general guidelines like more damping is better, hard surfaces are undesirable, sometimes the long wall is better etc...

 

Again, thank you for your thoughtful and stimulating write up.

Link to comment
Share on other sites

Chapter 9 [Adjacent Boundary and Loudspeaker Mounting Effects]:

 

This chapter actually serves as a preparatory chapter for the following chapters (even at its 17 pages of length). In and of itself it doesn't provide strong guidance for hi-fi consumers in buying, installing and listening to their hi-fi loudspeakers in their homes.  There are clues about what not to do in terms of loudspeaker placement in-room, but no special guidance related to design trades in real consumer listening rooms. 

 

As this chapter is really preparatory, I give extra consideration to the author (just like chapter 8, the immediately preceding chapter), but there are many places where the author could show the positive benefits of boundary gain, etc. that I feel is so sorely lacking at present.  The author actually confounds the reader with unequal alternatives in order to (I believe) subconsciously tip the reader to the notion that boundary gain is somehow "bad". 

 

Paul Klipsch spent his life using boundary gain to his advantage in his loudspeakers, and is completely cut out out of the bibliography in this 3rd edition of Toole's text (which probably is for the best, since Toole was not very supportive of PWK's writings in the 1st/2nd edition).  This should be a warning to anyone that values any Klipsch loudspeaker and what it brings to the marketplace.

_____________________________________________________________

Chapter 10 [The Sound Field in Sound Reproduction Spaces]:

 

This chapter in many ways is like the preceding chapter--introductory and concepts-based, but in this case, I find the examples and point given actually help the reader generate thinking about how to measure and use reverberation time  curves, and what it actually means. 

 

I find that RT curves are very appropriate for hifi-sized listening rooms in that the measurements are typically done by measurement applications (such as REW), but are also augmented with more measures that add information to the basic RT measures (i.e., RT20, RT30).  In particular, there are issues related to the direction of travel of acoustic energy that is not possible to measure easily using omnidirectional measurement microphones (or even cardioid microphones).  This is a major area of weakness in RT measurements, but not one that render the measurement and use of the RT curves useless.  Quite the contrary. 

 

Reverberation time measurements are at least very good first measurements on what's going on in the room in terms of diffuse sound fields, and whether or not the total absorption in the room is matched to the total room size and shape...and whether it can be used for HT-type environments (i.e., surround sound whereby in-room reflections are not as important as in stereo music-only rooms to make up for the center 2 kHz subjective hole in the perceived SPL response, and the lack of side surround channels to provide "LEV"--listener envelopment). 

 

There is much room for improvement over and above RT measurements, especially in the areas of speech intelligibility, but I think that there are recent articles and presentations (in particular David Griesinger's web site) on these subjects that absolutely generate those "aha!" moments in what's occurring in good performance venues and in poorer venues.

 

Diffusion is a subject that Toole introduced earlier, and here he zooms into the technical issues and problems of diffusion.  It's not an easy subject, and I believe this is one area where the older "stereo only" listening spaces diverge in terms of requirements from the surround sound home theater and multichannel music rooms.  Perhaps more can be said on this subject, but in particular, and like I alluded to earlier, getting good broad-band diffusion in the right places in the room while simultaneously putting the right amount of absorption at the right frequency bands (including low frequency bands below 300 Hz) and physical places in the room--is a challenge.  There seems to be a lot of anecdotal information on this subject, and few general statements that can be applied.  I think this is an area where the informed consumer might need to do cut-and-try exploration, while trying to hold the cost down and aesthetic appeal up.  Candidly, it's a subject that I plan to continue to explore as the budget and levels of resulting negative aesthetics allow in my listening room.  There is much to explore here.

 

Toole's discussions of early reflections, I feel, are far too brief.  But perhaps I've had a bit more time applied to the use of full-range directivity loudspeakers with particular focus on first-hand experience in trial-and-error experiments.  While I do not discount the theory presented here, I do contest (at times) some of the conclusions that Toole draws in this narrow technical area.  Perhaps I will link at a later time to other threads that I've posted into to expand on my meaning on this subject, but here's one that will get you started:

 

Chris

 

  • Like 2
Link to comment
Share on other sites

Chapter 11 [Sound in Cinemas] appears to have been coalesced into a chapter from the 1st/2nd editions of the book, where Toole talked about the "X-curve" EQ curves used in large commercial cinemas interchangeably with home hi-fi listening rooms.  This chapter is here I believe to eliminate the confusion of switching back and forth between home-sized listening rooms and commercial venues (i.e., apples-and-oranges comparisons).  Home theaters are almost never the size of commercial cinemas, and the acoustics properties of the two are not coincident (the precedence effect is in full control in home-sized listening rooms, and almost not a factor in commercial cinemas).

 

So, for the purposes of the home hi-fi/home theater enthusiast, this entire chapter can probably be skipped and only referred back to in later chapters if the concepts presented are unknown/unclear to the reader.  I'll leave chapter 11 alone in terms of review, except for this: Toole is basically leading up to the notion that the "X-curve" is always wrong, and other less-severe EQ curves on high frequency roll-off EQ are almost always preferred by the listeners.  I believe I would agree.

___________________________________________________

 

Chapter 12 [ Sound in Home Listening Rooms, Home Theaters and Recording Control Rooms]:

 

This is the chapter that most of the readers of Toole's book will probably gravitate toward.  It's common for book authors to try to combine more than one customer type addressed in their books in order to address a wider audience (to increase sales), but in this case, I believe the home hi-fi/home theater customer segment is many times larger in terms of readership than commercial customer segments--including recording control rooms. 

 

In a way, I think that this chapter should be at the front of the book, but there is so much tutorial work that is needed to get most readers to this point that Toole has placed it more than 2/3's the way through the book (the book encompasses quite a sweep of material breadth, in other words).  However, characteristically of this author, there is still a much greater focus on the research rather than practical application of the material.  So many readers will be struggling to infer the connections between his discussion of certain research papers and authors' writings to bring the material down to the "so what?" question: "what does all this mean for me in my listening room".  I think this is both a strength and weakness in Toole's approach. 

 

I would stick to more "how to" and "how not to" guidelines, and allow the readers to catch up over time, but that's just my realization after many years of teaching and providing both training and education as a side profession to my main engineering job.  Too many times I've run into cases where the reader is unable to generalize and integrate the material into a cohesive and tractable/remembered whole.  It takes time and effort to do this by the readers, and I think Toole's more research-oriented focus, while useful for people like me that value the references and dig deeper into the referenced papers, probably well over 90% of the readers aren't doing this...and will probably never do it.  So focus on practical application of the information is usually the first order of business, in my experience.  Here, Toole's approach looks very fragmented and uneven in focus (even though it may be in focus in terms of his thinking and approach), with too many links to both internal and external references.

 

But to focus on this chapter in particular, note that this is where Toole starts his focus on SPL response (...not even phase response, the other half of the transfer function, that's also quite audible), and does not also simultaneously integrate directivity requirements of loudspeakers and the required in-room acoustics, to give the reader a balanced perspective on typically bought home-theater and home hi-fi (stereo) loudspeakers. He immediately goes to "flat SPL response is king" (something that is easily correctable nowadays by almost all home theater electronics), and then folds loudspeaker directivity requirements underneath the off-axis response curves (sort of strangely and at an unbalanced level of detail). 

 

In other words, it looks very uneven in focus.  And probably most importantly Toole makes the assumption that the loudspeakers that the consumer has to choose from largely exclude those classes of loudspeaker performance that Klipsch owners typically enjoy.  Toole is basically saying "given the current low-to-medium performance and priced 'monkey coffins' (i.e., small direct radiating loudspeakers) in the marketplace, here is what performs the best" but then excludes the loudspeaker performance you get from Klipsch, et al.  Almost any horn-loaded loudspeakers will outperform these other loudspeakers subjectively, especially once they have been dialed into flat SPL and phase response using typical "room correction software" (Audyssey, Roon, JRiver, Accourate, YPAO, ARC, etc, etc.).  This is basically inexcusable, in my engineering experience.

 

Why Toole does this is simultaneously understandable and also obscure.  I believe that he has an axe to grind, i.e., he still hasn't cut himself off from his previous employer to give the reader a better balanced perspective on what is actually available on the home audio marketplace, especially from Harman's competitors.  In this respect, I fault him unwaiveringly, and this fault is probably the main motivation for me to put together this entire thread. 

 

More to come on chapter 12...

 

Chris

Link to comment
Share on other sites

 

Chapter 12 [ Sound in Home Listening Rooms, Home Theaters and Recording Control Rooms], continued:

 

While Toole's recommendations on flat SPL response, combined with flat or descending SPL response off-axis, is good, here is something that Toole doesn't say: that flat off-axis performance need to especially continue into the midrange and upper midbass range (down the the home hi-fi room's Schroeder frequency, where the sound waves no longer "travel", but are more-or-less stationary and form stable peaks and valleys in SPL response depending on where you are listening within the room.  This is extremely important, because is really is hiding from the reader the performance of full-range horn-loaded loudspeakers (and there is more than one manufacturer that provides these type of units).  Harman only produces direct-radiating woofer loudspeakers that all lose their directivity control below 500-1000 Hz. 

 

Additionally, Toole fails to tell the reader that lower distortion (mostly modulation distortion, but also compression, impulse, and lastly/least: harmonic distortion) is more important in the "bones and muscles" of a loudspeaker than is flat SPL response.  Additionally, having the kind of flat phase response that I showed much earlier in this thread  of fully horn loaded loudspeakers, combined with a small amount of nearfield absorption to control early midrange reflections in-room, together yield in-room sound reproduction performance that completely eclipses any direct radiating loudspeaker performance, including those that use diffraction to approximate fully horn-loaded directivity at midrange frequencies (e.g., Dutch & Dutch-type approaches). There is a good paper on the effects of diffraction effects of line arrays (basically the same technique used by Dutch & Dutch loudspeakers) that explains the reasons why line arrays don't sound very good unless you're seated in a particular place in the room, and then still have off-axis artifacts that detract from the "sweet spot" listening--relative to fully horn-loaded loudspeakers (Khorns, La Scalas, Belles, Jubilees, MEHs, etc.).   Toole never points out the downside of sing direct radiating woofers for bass reproduction--and those downsides are large and very audible for anyone with ears.

 

So to gravitate back to chapter 12, once we get past the "flat SPL response", including room reflections, etc. Toole moves on to "room curves".  The problem with Toole's presentation room curves is that--I know he knows--he is not identifying the real factors that precipitate the use of them:  room curves are not really used because listeners preferred the curves", but because the room curves are there to undo two issues, neither of of which are related to "listener preferences".  I've found that listeners want a 1/f cumulative SPL vs. frequency curve (about -5.2 dB per octave or -16.6 dB/decade cumulative power spectral density) in their music reproduction.  What the room curves do is:

 

  1. To partially undo (in one-size-fits-all fashion, regardless of music source and provenance) the mastering EQ that is applied to the tracks that isn't in the original recorded tracks.
     
  2. To partially correct for the loss of directivity control of the loudspeakers in-room, because they lack directivity control below 500-1000 Hz. 

Nowhere was point #1 brought home to me as when I compared the highest quality music tracks that I own to the typical "remastered" music tracks found today. I found that small changes in the overall slope of the 1/f curve produce enormous changes in the sound signature of the tracks, and that the best tracks I own are ones that are indistinguishable from the "as-recorded" mixdown tracks directly from the mixing tables (in SACD, which cannot be edited without first being converted back to PCM in some format).  The -0.4 dB/octave that Toole is describing is minuscule as compared to the mastering EQ that's typically applied, so all of his "room curves" are really in the noise, completely covered up by the "mastering EQ" used by the mastering guys. (I won't call them "mastering engineers", because I don't believe that's what they're actually doing.)

 

And nowhere was point #2 brought home as when I listened to the difference in the K-402 horn vs. the K-510 horn in-room on top of two different Jubilees in Hope in September 2009.  Considerable re-EQing was required in-room after the K-510 was EQed flat in the anechoic chamber directly adjacent--something that the K-402 didn't require.  This means that the loss of directivity below 600 Hz has a massive effect on the sound quality of the loudspeakers--and Toole NEVER talks about this--but assumes that this is just part of the "bad" that comes with the "good".  Nothing could be farther from the truth, and there are loudspeakers in the marketplace--at or below the price points of the highlighted "good" loudspeakers that Harman produces--that control their polars below 600 Hz.  What's the difference in these two horns in-room?  Directivity below ~600 Hz.  That's all.  That experience is something that I will never forget, and I still thank Roy Delgado for that demonstration.  It was probably one of the two or three biggest things he showed us over a Friday-Saturday weekend that really transformed my viewpoint on loudspeaker design.  It was a transformational moment.

 

So the story continues.  We will see even more of this sort of thing in the remaining pages of chapter 12--a pivotal chapter in Toole's book.

 

Chris

Link to comment
Share on other sites

So continuing on with discussion of the -0.4 dB/octave EQ down-slope that Toole professes is "required", one might ask why the mastering guys are putting too much up-sloping EQ on (largely popular) music tracks.  The answer may not be immediately obvious (and I have reason to believe Toole might already know why, but didn't say in his book).  Two obvious characteristics of recording studio monitors is they are almost exclusively direct radiating loudspeakers of the "small" variety, which have issues with compression distortion starting at ~80 dB at one metre that are used for mastering playback and "translation", which tend to mask the transients of the music, so to compensate for the increased dullness and lack of dynamics of the loudspeakers, simply crank in more up-tilt of the EQ curve to make the tracks sound more lively on their nearfield studio monitors (the black lines in the figure I posted on page 2 of this thread). 

 

These same studio monitors also have over 900 degrees of phase growth from 20 kHz down to 200 Hz (and much more below 200 Hz to the extent of their bass output).

 

Toole Loudspeaker Preferences - Phase Response groups 7.0 to 7.9.GIF

 

What does this phase growth do?  It flattens out the peaks in the time-based playback of the recorded music tracks.  The following is from Dave Griesinger's online PowerPoint presentation on "clarity":

 

1961924739_Griesingerspokenwordclaritywithwithoutrandomizedphaseofharmonics.GIF.8cb9d8a7ccbff7d007044f12b5439bd4.GIF

 

Loudspeakers that have a great deal of phase growth from high frequency to low frequency exhibit the same conditions of "shaving off the peaks" from the time-based SPL trace, thus leading to muffled sound--due to loudspeaker or playback electronics phase growth--ALONE.  In response, the mastering guys crank in more HF tilt to the EQ curve to compensate for inherent phase growth characteristic found in direct radiating loudspeakers to make up for their perceived muffled sound (on studio monitors).

 

The result of these two factors is that, when the same music tracks that were mastered on little direct radiating loudspeakers with all of their compression distortion at higher SPL and gobs of phase growth throughout the passbands are played back on loudspeakers that do not exhibit compression distortion (because of their efficiency due to the transformer-like action of the horns) and also do not exhibit the level of phase growth that the studio monitors exhibit--the sound is heard for what the mastering guys actually did to it: they cranked in too much high-end EQ to compensate for both of these studio monitor issues (that fully-horn-loaded loudspeakers largely don't have, i.e., the horn loaded loudspeakers are not distorting, but the studio monitors are). 

 

Why do I demaster my music tracks for almost all of my popular music that has been "mastered" (including re-EQed)?  You know why now. 

 

So how do I know all this?  If you go back to the 1970s vinyl and reel-to-reel tapes when most popular music was being mastered on largely horn-loaded loudspeakers, none of them sound harsh or overheated on the high end. 

 

Think about it...

 

Chris

Link to comment
Share on other sites

Section 12.4 [Dialogue Intelligibility in Home Theaters]:

 

This is actually one of the more important sections that I've encountered.  Since most of the dialog intelligibility in a typical surround sound setup comes from the center loudspeaker, my personal experiences with attempting to get a working center channel between Jubilees is likely a lot more extensive than average.  The biggest surprise that I encountered with the K-402-MEH was its increased speech intelligibility over previous center channel incarnations (Heresy, RC62, Cornwall, Belle, and JuBelle--both in two-way and three-way configurations) between the Jubilee/TADs.

 

I do have measurements of the clarity of most of these loudspeakers (the RC62 lasted but a minute or two--just until I could stop laughing).  What I found is a sensitivity of the C50, C80 and D50 measures that you get from REW to factors other than the loudspeaker itself, so it's difficult to show a measure of the speech intelligibility differences. 

 

Here is one plot show some of those differences using the "D-50" measure ("Definition"...first 50 milliseconds vs. 51 ms+, in percent energy--the right-hand scale, below).  The Right Jubilee suffers only because it is located in the room corner (all others are elevated mid-wall positions), which affects the Clarity calculations since it is picking up the boundary gain of the room's walls and floor more effectively.  There are other differences that are noted between these measures, but note that the K-402-MEH shows a persistent performance advantage vs. the other loudspeaker types:

 

60029170_D-50ClarityComparisons(CornwallJubJuBelleMEH).jpg.e486cc62008af99957f7f05a60365636.jpg

 

"So what?"  Well, a lot.  I suspect that this is one measure that does show significant differences between horn-loaded loudspeakers vs. frequency and direct radiating loudspeakers.  Toole seems to only show the advantages of horn loaded loudspeakers when it's to the benefit of the constituent companies of Harman, so quite uncharacteristically, he provides no modulation transfer function (MTF) measures of different loudspeakers.  In fact, he doesn't even talk about MTF measures at all.

 

Chris

  • Like 1
Link to comment
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...