Jump to content

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


Chris A

Recommended Posts

Excellent and to the point.

 

Last time I stopped by a Magnolia Center, I had them play one track with Poe using their totl Marantz

and switching between speakers, most expensive to least expensive (monkey coffins)

 

Her voice was unintelligible on most, difficult on a few and clear on the Klipsch R series book shelf

 

Sales mgr was upset and stopped the test. Put on something was mostly rock instrumentals turned up.

 

My understanding is that Magnolia isn't allowing people to bring their own music as of 1-2 years ago.

Link to comment
Share on other sites

So one of the biggest differences in individual loudspeaker performance (via in-room measurements) isn't covered in Toole's book...as far as I can tell.  The interesting thing is that no one has apparently caught this.  Would this be an important performance measure to include in all "loudspeaker reviews" online and in print?  I think so--most emphatically.  So what's a major difference between brands in individual loudspeaker performance?  Clearly, this seems to be one of them, except virtually no one is paying attention. 

 

I would also guess that planar dipoles will also perform well on this measure (or maybe not if the backwave interferes at greater than 50 ms delays)--but have other issues related to "head in a vise" and very limited dynamic range/high compression distortion. 

 

I believe we are, starting on pg. 351 of Toole's third edition, just beginning to see real differences in interpretation of loudspeaker performance.  It doesn't stop here, though. 

 

However, I plan to continue the systematic walk-through of Toole's chapters.  We'll continue to see what turns up.

 

Chris

  • Like 1
Link to comment
Share on other sites

12.5 [Recording Control Rooms]:

 

Normally, in those instances where the book's subject diverges strictly from home hi-fi performance, I would recommend going around those book sections, only referring back to them if curious or for clarification of later points on related subjects.

 

However in this instance, I believe the subject is inseparable from the typical recording quality of albums which are most often played on most consumer's setups. Understanding what happens in the recording, mixing and mastering studio monitors will give the reader a much better understanding of why their recordings might not sound very good. 

_______________________________________________________________

 

A bit off topic here:

 

Of particular note, before turning to the performance of the studio monitors themselves (section 12.5 serves as the only section discussing this subject), I wanted to visit the notion of "translation".  On several occasions here on this forum, I've been less than, well let's say, forgiving, of those that happen by that have done mastering as a paying job.  Why?  I find that several notions wrapped up with their "standard operating procedures" (SOPs) to be offensive and destructive of the quality of the music that I can buy.  Most assuredly, they see it differently. Usually the discussion gets down to a mea culpa:  "if I don't do it this way, I won't have a job mastering...".  That's okay with me.  If all the mastering guys had to quit and do something else, perhaps there is a chance the practices that have become SOP might revert back to quality instead of "individual track sell-ability on earbuds".

 

You might feel differently--but over time and trying to fix over 17,000 music tracks that have basically been "translated" into oblivion, my patience has worn quite thin.

 

Why can't I buy the mixdown tracks from the musical artists, and avoid having to deal with "translated" music tracks?  There is no reason nowadays.  Digital downloads cost nothing to make available, and in fact cost less than "released tracks", because the mastering guys don't need to tack on their fees to the musicians' released tracks. 

 

This one-size-fits-all mentality is precisely what shortens my patience.  I don't care for the mastering culture that has arisen over past decades, and I didn't pay what I've invested in my setup to have it degraded down to $500 (or less) loudspeakers that REALLY don't sound very good, e.g., Auratone 5C, Yamaha NS-10M Studio, etc.  Why would someone do such a thing?  The answer to that question is one that defines the modern "RIAA" industry for popular music, which is also bleeding over into the MPAA industry (i.e., the highly destructive "Loudness War" compression/clipping practices, etc.).  I don't like it and it really doesn't need to be this way. 

 

Think about this: how long would it take loudspeaker manufacturers to change their loudspeaker designs to sound good with higher fidelity music tracks?  Not very long if they plan to stay in business.  This is the crux of the issue: the prolongation of bad automotive loudspeaker performance from the 1960s in order to "meet the marketplace demand" (a euphemism for bad sound reproduction). 

_______________________________________________________________

 

Back on topic...historically the most used nearfield studio monitors whose performance is shown in this section are frankly terrible by today's Klippel and Audio Engineering Society standards, in most ways but one:  the most used studio monitors in the market typically have reasonable step response. While that's not saying a lot, its a strong clue to what's occurring across an entire industry. 

 

More to come.

 

Chris

  • Like 1
Link to comment
Share on other sites

The reasons young people are flocking to vinyl is:

 

No quantization error if analog source

 

No loudness wars, won't fit on vinyl

 

No psycho acoustics added to trick the ear for low bit rate and $2 ear buds.

 

Guttenberg interviewed a friend who masters music, asked which mastering is the best, answer, the version for vinyl.

  • Thanks 1
Link to comment
Share on other sites

I do understand why so many people are flocking to vinyl releases. It's easy to hear what happens if the mastering guys can't do what they can do to digital tracks.

 

It's apparently the A&R executives and some of the artists themselves that don't really care about sound quality as much as how much money they think they're going to make, based mostly on flawed thinking about loudness war envy.

 

Do you know what quantization noise sounds like?  Background hiss.  I never hear that in the digital format, unless its from old analog masters.  This is probably because only the last 6-8% of the dynamic range (in dB) of most popular music tracks is typically being used in CDs nowadays (i.e., 6-8 dB of crest factor remaining in the music tracks). So the issue with quantization error I believe is limited to hearing only the decays in music, never the bottom-of-the-scale hiss.  If the mastering guys stopped pushing the tracks to be as loud as possible, the quantization error on decays would be inaudible, but these guys actually pride themselves on how loud they can make it, and if you ask them not to do that, they refuse saying"I can't risk my reputation on making it that quiet".  It's a broken culture.

 

But the problem with vinyl is...when you start to listen in thinned out passages (like bebop jazz combos, etc.) vinyl really does fall apart relative to Red Book CD format (modulation and harmonic distortion, wow, flutter, and surface noise, etc.).  It's very easy to measure what's happening (I've measured it myself)...even if you try to use $10K USD turntables and $5K cartridges--it still falls apart because of the format.  I don't believe that a lot of people want to understand that.  Nostalgia rules supreme.

 

But I do understand that for those having "mid-fi" tastes, vinyl is nowadays the only way to get their popular music with a little snap still in the tracks, i.e., the dynamics can't be completely crushed in vinyl like it is for digital.  The mastering guys and the A&R execs are about as trustworthy as cable company executives...they can't be trusted to leave the fidelity in the music anymore.

 

Chris

  • Like 1
Link to comment
Share on other sites

13. [A Rational Approach to Designing, Measuring and Calibrating Sound Reproduction Systems]:

 

This chapter again rehashes Toole's views of sound in small rooms above and below the transition frequency (Schroeder frequency), and room curves.  He briefly discussions automotive audio, headphones, and commercial cinema.  There is little here that could not have been integrated into the earlier chapters on these subjects, and would have reduced the need by readers to rely on the books very brief index (which itself leaves something to be desired) to find all of the material that Toole references in the book.  While the bibliography is much more extensive, with a book this size, is seems as though putting the references at the back of each chapter instead of one reference at the end of the book would facilitate the reader much more strongly.

 

14. [Measurement Methods]

 

The first few sentences of this chapter are interesting.  I believe the following is "fair use":

 

Quote

1) The preceding chapters have provided persuasive evidence that, if good sound is the objective, frequency response is the dominant factor.

2) Non-linear distortion is audible if sufficiently loud, but in the world of conventional, "full-sized" loudspeakers, it has not been a frequent problem at normal playback sound levels.

3) Professional use can be more demanding.

4) Impulse response and time-domain ringing of resonances are already accounted for over most of the frequency range because loudspeaker transducers are minimum-phase devices--a flat, smooth frequency response indicates an absence of such misbehavior.

5) Rise time is a function of the high-frequency limit--for example, the "speed" of a subwoofer is limited by the 80 Hz low-pass filter. 

6) Phase shifts, polarity concerns and group delays are below the thresholds of detectability in normal listening.

Wow.  If this was on page one instead of page 379, perhaps it would have avoided all the disagreements found in the preceding pages of this thread...because I believe that listeners on this forum would have objected to sentences 1, 2, 4, 5 and 6.  Let's take them one at a time:

 

Quote

1) The preceding chapters have provided persuasive evidence that, if good sound is the objective, frequency response is the dominant factor.

 

Smooth and flat SPL response is important, but PWK clearly said it wasn't the most important--in fact it almost didn't make his list of important factors.  And it is the only factor (along with phase response) that can be corrected using EQ tools that exist in every multichannel preamp/processor that is on the market.  It just isn't worthy of such attention in today's marketplace.

 

Quote

2) Non-linear distortion is audible if sufficiently loud, but in the world of conventional, "full-sized" loudspeakers, it has not been a frequent problem at normal playback sound levels.

 

This is just false.  No one I know that pays for hi-fi equipment wants to be told that distortion isn't audible in today's loudspeakers.  "Normal playback sound levels" is apparently anything below 80 dB/one metre (OSHA standards that must be used for employed persons in work environments--such as professional listeners that JBL employs).  Need I say more?

 

Quote

4) Impulse response and time-domain ringing of resonances are already accounted for over most of the frequency range because loudspeaker transducers are minimum-phase devices--a flat, smooth frequency response indicates an absence of such misbehavior.

 

Loudspeakers (the whole unit of multiple drivers) are not minimum-phase devices.  Only the "transducers" (drivers) can be minimum phase, and many times they aren't--due to internal acoustics issues and all-pass delays/phase shifts in crossover filters.  All-pass delays/phase shifts are always present in loudspeakers, and especially when you place them in home hi-fi rooms.  There is a great deal of difference, particularly at woofer frequencies, of transient response of different loudspeakers.  It has been my experience that fully horn-loaded loudspeakers, once dialed-in with EQ for SPL and phase response flatness (not just "linear phase" but flat phase) and low group delay disturbances, are audibly much "faster" than any direct radiating loudspeakers that can reach 100 dB at 1m without audible compression, modulation, and harmonic distortion. 

 

Quote

5) Rise time is a function of the high-frequency limit--for example, the "speed" of a subwoofer is limited by the 80 Hz low-pass filter.

Ever hear the difference between a fully horn-loaded bass bin of good design and a direct radiator?  The distortion in the direct radiator loudspeakers is audible even at 80 dB levels.  And the horn-loaded bass bins will always sound much faster than the direct radiators.  How are these facts accounted for in sentence #5, above?

 

Quote

6) Phase shifts, polarity concerns and group delays are below the thresholds of detectability in normal listening.

 

This is just false in my experience...at the levels of phase shifts that Toole shows in his earlier chapters for typical direct radiating loudspeakers (550-900 degrees of phase shift from 20 kHz --> 200 Hz). I can clearly hear the differences in flat phase and low group delay disturbances at or below "normal listening levels" (i.e., 80 dB/1m).

 

So what Toole is saying is that--only "normal" people play their loudspeakers below 80 dB at 1m.  I don't know anyone that would agree with that statement.  Why would someone spend thousands of dollars on home hi-fi loudspeakers that could not play above 80 dBC, even in the peaks?  I know that they will not be satisfied for long, in fact they will become quite dissatisfied that their loudspeakers sound lousy above 80 dB, and will look to replace them with loudspeakers having much more capability to replay music at realistic (live) listening levels. 

 

And to say that phase shifts are not detectable at "normal" listening levels--indicates to me that Toole never tests the difference between loudspeakers that can control their directivity below 1 kHz, and those that can't (his preferred loudspeakers). It takes two factors to be under control to hear the effects of phase shifts: 1) less than 90 degrees of phase shift within the loudspeaker itself, and 2) controlled directivity/absence of nearfield reflections to get the direct arrivals from the loudspeaker to the listener's ears without substantially being disturbed by nearfield reflections, and thus rendered inaudible due to the competing reflections.

 

Additionally, Toole spends so much time on SPL response flatness, that there is no air left in the room when the much more important factors of low distortion and full-range controlled directivity are discussed.  This is the difference between Klipsch loudspeakers and those branded as "superior" by Toole (and are owned by Harman...).

 

Chris

  • Like 1
Link to comment
Share on other sites

The balance of chapter 14 is, in my view, severely skewed by the notion that "frequency response is king--we really don't need anything else..." that Toole professes.  Again, it's probably a good thing that Toole waited until very late in his third edition to make such statements (and so plainly). Nevertheless, I will try to find more useful information in this chapter.  I will skip the sections that deal with in-room frequency response as a function of off-axis angle and room reflections, simply because they are over-emphasized already. If the loudspeaker has good directivity control vs. frequency and over the range 100 Hz to 20 kHz, and reasonable accommodations for control of early reflections within the first 5-6 ms acoustically around the loudspeakers, these discussions by Toole are basically moot.

 

I find that the best method to control early reflections and room modes from seriously obscuring the minimum-phase transfer function and loudspeaker directivity effects--is to simply take measurements of the loudspeakers at 1 m in front of the front baffle.  For measurements below the room's Schroeder (or transition) frequency, which is usually from 100-200 Hz in reasonable sized listening rooms, it is also necessary to move the measurement microphone back to the listening position(s) and to attempt to position the loudspeakers in-room and to EQ them to eliminate the response peaks--but not the high-Q drop outs in response.  This is a fundamental principle of in-room EQ that should never be violated, else the effects of room resonances and cancellations are actually exacerbated by any attempts to boost nulls below the transition frequency.  If your listening room has a Schroeder frequency close to 100 Hz (i.e., at least one dimension of the room is fairly long, and there is sufficient absorption in the room to control the reverberation time--RT20 and RT30--generally to below 0.5 s), you will be able to EQ your loudspeakers using 1m measurements down to that frequency and even below.  The check at the listening position is only to make sure that there isn't any response peaks that are significant at the listening positions (only) so that attenuating EQ is used to control those downward. 

 

Spatial averaging of measurements: this subject is probably necessary in very, very small listening rooms where the bass response below the transition frequency can vary greatly.  In larger rooms, this variability in bass with small changes in listening position begins to be less prominent, but will always be a concern in home-sized listening rooms. I find that taking secondary, and perhaps tertiary measurements around the listening positions can lead to better tradeoffs in EQ, but note that these are tradeoffs--not optimizations.  Everywhere outside of these listening positions will be strongly affected by room modes below the transition frequency, and by larger amounts of variation as the volume of the listening room decreases in size.  Time spent in measuring can basically be wasted with such ergodic measurement procedures, and I find lead to minimal improvement in the results.  If there is enough absorption in the room, the fewer measurements approach will work quite well and save significant time.

 

I find that weighting of the results in-room (A, B, C, etc. weighting scales) is one of Toole's concerns, but not one of mine.  I go for flat response in-room, and if that doesn't sound like sufficient bass, then steps need to be taken to further flatten the phase response of the loudspeakers below ~100 Hz. Then you will hear that flat bass response clearly.

 

14.2.2. [Multichannel calibration]: 

 

If the "automatic" room correction software would work well, I would recommend using it for system calibration.  As such, I haven't found a room correction software package that actually works without creating problems.  So I recommend 1m measurements of all the surround loudspeakers (5, 7, 9 channel) to see the combined effects of all EQ compensation packages, else what you will end up with will decrease the value of the sound quality of your system by as much as 75% (or more) of its potential sound quality.  That should be motivation enough to learn how to do it and to actually do it in-room.  After this is done (I recommend DSP crossovers on all channels in the surround array--at least for the 5 main surround channels if multichannel, as especially for the left and right channels if multichannel or stereo setups).  If you can run the room correction software and then turn off the EQ correction only, thus leaving the distance and channel gains intact, I believe you will get the best performance out of your system, assuming that you've EQed each loudspeaker to flat SPL and phase response before running the room correction software.

 

14.2.3 [The Effect of Propagation Distance--A Side Channel Challenge]:

 

One of the things that I've witnessed with surround loudspeakers (left surround, right surround, etc.) is that if their coverage pattern is much wider than the front loudspeakers, you'll never get the surrounds to sound right in multichannel music mode--their timbre will be off.  I think this is self-explanatory.  The machinations that Toole goes through in this section should be tempered by that observation.  I recommend the same sort of loudspeakers in the surround channels as the fronts--if you can accommodate them, unless you host several rows of viewers in your home theater on a regular basis.

 

14.3 [Measurement Microphones]:

 

The reality of today's marketplace I believe skews the measurement microphones to something that is both useful and lower cost.  I find the use of USB microphones with the digitization electronics and computer bus built-in to their bases alleviate huge problems that exist using microphones requiring separate mixers/phantom power sources that I experienced in the early days of taking in-room measurements. In addition to certain loudspeaker models that I have endorsed in this thread, I also endorse the miniDSP UMIK-1 microphone most highly.  I feel it eliminates the source of potentially dozens of procedural issues that existed before its entry into the marketplace.

 

Chris

Link to comment
Share on other sites

15. [Multichannel Audio]:

 

The history of multichannel sound is somewhat informative of how we get things like "LFE" and bass management that are ubiquitous in today's home theater marketplace, but I would guess that most readers are searching for "how to do it" in terms of home theater sound, not so much "how it was done in the past" and in commercial theaters. 

 

So if we zoom to section 15.7 [How Many Loudspeakers and Where?], I think we're into probably the most informative section of the book. Section 15.7.1 [Optimizing the Delivery of "Envelopment"], is a key section that I believe cannot be overemphasized in its importance.  I would encourage the readers here to not only consider positive benefits of having multichannel sound over stereo (if sound quality is more important than nostalgia or apparently excusing notions of "the simplicity of stereo"), but also to consider the placement of the center and surrounds for performance, as detailed in this section.  In particular, I would also emphasize the importance of getting all the surrounding channels (i.e., 5 channels, 7 channels, etc.) the same height off the floor, especially the front three loudspeakers--within two inches (5 cm) of a level line through the center of the midrange-treble horns or drivers. 

 

Section 15. 8 [Surround System Layouts] also includes extremely important guidance on the layout of surround sound systems and acoustic treatments in-room, which I recommend highly.

 

The balance of chapter 15 is interesting, but in terms of usefulness to the typical surround sound or even stereo system owners, probably at a much lower level of importance.

 

Chris

Link to comment
Share on other sites

The balance of the book...

 

Chapters 16 [Loudspeakers and Power Amplifiers], 17 [Hearing Loss and Hearing Conservation], and 18 [Fifty Years of Progress in Loudspeaker Design] are informative, but generally at a lower level of importance to the "how to" home hi-fi reader.  Chapter 18 in particular I think sheds some light on Toole's preferences for flat SPL response, but I think it is informative for those readers that are into nostalgia in older studio monitor designs that today can command exorbitant prices.  A look at chapter 18 might change your mind if you're thinking of investing in yesterday's loudspeaker designs that do not have the same low distortion and effective directivity control of PWK's loudspeaker models (in particular).

 

I plan to summarize the book's contributions to the art and science of home hi-fi in subsequent posts (and answer any questions that arise).  I will also take a moment here to comment on one yet-to-be-addressed glaring omission: the design of loudspeakers themselves.  Toole rightly talks about the "system level" requirements of loudspeakers, rightly so.  But then he completely omits anything below the surface in terms of loudspeaker design.  For those enthusiasts that are good with buying loudspeakers and plunking them in a predetermined room that is convenient in terms of expediency for a listening space, I would hazard a guess that these people could be completely satisfied with Toole's focus at the loudspeaker "configuration item" level, not to peek below the exterior surface of them. 

 

However, in my experience, it is the understanding of what lies below the surface of the loudspeaker itself that is the most informative and useful to the loudspeaker buyer.  Perhaps some of those observations, both from personal experience and discussions with Roy, and from other sources (such as Olson and Beranek) would be equally useful to prospective loudspeaker and home hi-fi system owners as the book itself has been.  I think this is also in-bounds for this thread.

 

In general, Toole has made a grand attempt to consolidate his first/second edition book into fewer chapters and more focused subject areas, but I do believe that more consolidation would be most useful to the typical reader, and a greatly expanded index, along with end-of-chapter references.  I've found that many of the papers that Toole references can be found on-line, and oftentimes provide a totally different view than Toole portrays in his book of those sources (which is particularly true of the sources on the subject of loudspeaker phase response flatness--in which the authors of some of those referenced articles are perhaps totally opposed in terms of their viewpoints than the author of this book).  

 

More to come.

 

Chris

Link to comment
Share on other sites

Below the surface of the loudspeaker exterior (the stuff that Toole didn't talk about)...

 

In general, Toole doesn't cross the divide of talking about specific loudspeaker technologies (i.e., multiway cone and dome-type direct radiators, horn-loaded compression drivers and cones, planar radiators--including electrostatic and electrodynamic transducers, full-range drivers, crossover filters, closed box bass bins, vented, transmission line, passive radiators, etc., etc.).  We could examine all the technologies of books like the Loudspeaker Design Cookbook (Dickason), and others. In particular, I believe it to be a very broad subject area that could easily eclipse the ~500 pages of Toole's text.

 

1) Do I believe these discussions are warranted in direct context of Toole's book?

Absolutely. 

 

2) Do I believe that some of the material presented in Toole's book is in excess of what the typical home hi-fi consumer/owner needs? 

Yes--I even pointed out those chapters that I believe to be effectively extraneous in comparison to what I'm about to talk about.

 

3) Should all the available loudspeaker technologies be discussed?

No--I believe it would end up being a text in of of itself, not a forum thread subject.

 

4) Which technologies should be addressed here?

I believe that horn loading of drivers is the most important, so using the appropriate sections from Olson, Beranek, and Bjorn Kolbrek's High Quality Horn Loudspeaker Systems (albeit the latter sections of this massive tome--starting on page 413-ff).  Any other relevant technologies can be added as needed. 

 

5) Why horn-loaded only, and why Kolbrek's text?

Because that is what Klipsch is expert in doing, and is, I believe, the reason why they have done so well in the marketplace of late: that clean, powerful and crisp sound of horn-loaded loudspeakers derived from high quality designs.  Kolbrek's text is also a massive undertaking, complementing and considerably extending the subject of horn-loaded loudspeakers first introduced in classroom text format over 70 years ago in Leo Beranek's original classroom text at MIT in the early 1950s, followed by RCA's Harry Olson in his text(s).  These older texts are still used today, and represent the original "bible" for all types of loudspeaker technologies then in existence.  Kolbrek apparently has filled a badly needed update and extension of the two major loudspeaker texts from the 1950s in the area of horn-loading.  I believe Kolbrek has done a very good job in that respect.

 

Knowing how deep this subject could become, I think it wise to break this thread at this point and create a new one addressing the information inside the subject of high performance horn loudspeaker design and manufacture.  That thread can be found here:

 

 

Chris

  • Like 1
Link to comment
Share on other sites

  • 1 year later...
On 9/8/2017 at 5:58 PM, Chris A said:

51lDSq121-L._SY445_QL70_.jpg

 

Floyd Toole has become a figure not entirely unlike PWK: he represents a knowledge base in sound reproduction and identification of relevant measurements that correspond to listener preferences. This stands in contrast to what I consider to be a general wasteland and land of confusion and folklore memes filled with pseudo-science and belief systems not backed up by fact.  His newly updated and very recently released 3rd edition of his notable book: Sound Reproduction--The Acoustics and Psychoacoustics of Loudspeakers and Rooms has been largely rewritten and somewhat refocused.  I think that this new book deserves a review and some discussion.

 

Chris
How quickly time flies, recently I seem to have read only the second edition, so I have already read the third. I wonder if there will be a fourth, everyone would be happy. I am currently reading a canticle for leibowitz using https://freebooksummary.com/category/a-canticle-for-leibowitz for this. I also really liked her. I honestly don't know what book will be next. I want some kind of story based on real events.

I read it recently and I liked it.

Edited by Evelon
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...