Beyond the Ariel

Re: Goto ....

angeloitacare said:
i remember read somewhere, Goto in the beginning, when developing his drivers, where playing with a piano, recording it, and than play it right away next to the piano, tru his hornspeakers and compare the result. They say they were coming very close to original sound.

Actually a solo piano is a particularly easy instument to duplicate.

A good friend of mine is a concert pianist (Dickran Atamian). I once recorded him playing an electronic piano (Yamaha) directly from the instrument - no acoustics. We then played this back over my loudspeakers in my listening room. Even the pianist was amazed at how real the image was when one closed their eyes. The vision of a piano was right there. There were some small issues with the quality of the sampling algorithm and the "channels" (some keys on one channel some on the other, I had to cross-fad the source file to get a good center image), but on the whole the experience was amazing.
 
Re: observable pontification

auplater said:

Having been involved in electrochemical and aerospace engineering for 35+ years, all I can say with bemusement is the level of "ingineering" exhibited in many (especially loudspeaker) audio components is laughable at best.

John L.

Of what use is engineering in a loudspeaker (or any audio product for that matter) that is sold not based on anything quantitative but on marketing hype and magazine reviews?
 
Re: Re: Goto ....

gedlee said:


Actually a solo piano is a particularly easy instument to duplicate.

A good friend of mine is a concert pianist (Dickran Atamian). I once recorded him playing an electronic piano (Yamaha) directly from the instrument - no acoustics. We then played this back over my loudspeakers in my listening room. Even the pianist was amazed at how real the image was when one closed their eyes. The vision of a piano was right there. There were some small issues with the quality of the sampling algorithm and the "channels" (some keys on one channel some on the other, I had to cross-fad the source file to get a good center image), but on the whole the experience was amazing.

That brings to mind a pet peeve I have... piano image, etc. I have a few sources that are nearly dead-on wrt perceived image, as though the image size is correct between the speakers, at the correct height, with the scale moving from left to right (as though the pianist is sitting back to the listener). Not very many get this correct, nor do many symphonic recordings place the orchestral instruments correctly, for that matter. The effect is amazing on the few that do, but are the recording technicians that inept?

Often times the brass is off-center and up front, the strings are all over the place, or the entire reproduction has a hollow distant sound, as if it was recorded from the lobby. So it's not just the implementation and quality of the reproduction system at fault.

Sorry for the diversion... but I wonder if recording values can descend much further...

John L.
 
Often times the brass is off-center and up front, the strings are all over the place, or the entire reproduction has a hollow distant sound, as if it was recorded from the lobby. So it's not just the implementation and quality of the reproduction system at fault.

Yes, I listen to KING FM (classical music) from Seattle over the cable as I go around the apartment because they play stuff I'd never hear otherwise. The qualitative variety of recorded sound is enormous: ranging from breathtakingly gorgeous to repellent with the flaws you mention and others as well. I am amazed at how many recordings of conventional works have left and right channels reversed or just unbalanced.

I find the piano image you like to be disorienting: at a live performance the pianist sits side on to the audience with lid opened towards them and the long and short string sounds blend.
 
FrankWW said:


Yes, I listen to KING FM (classical music) from Seattle over the cable as I go around the apartment because they play stuff I'd never hear otherwise. The qualitative variety of recorded sound is enormous: ranging from breathtakingly gorgeous to repellent with the flaws you mention and others as well. I am amazed at how many recordings of conventional works have left and right channels reversed or just unbalanced.

I find the piano image you like to be disorienting: at a live performance the pianist sits side on to the audience with lid opened towards them and the long and short string sounds blend.

I don't necessarily "like" the left - right presentation of a solo pianist; just know that the intent was to portray the artistic effort this way.

I agree, the open front grand blended piano is appropriate for onstage concerts.

John L.
 
There is one thing to be said for the subjective

gedlee said:
I am often criticized because I downplay the subjective, but I do this for precisely the reasons that you state. The audio media and its jargon allow people to accept poor reproduction as sound quality because "it sounds good to me" - a position that cannot be argued with. But remove the subjective and its not so easy to support this erroneous point of view - in fact its pretty much impossible. It's the bottom line "I know what I like" that allows for the situation that you describe.

I try for what I think is "accurate" sound reproduction from my system. This must done by ear. Unfortunately I can't make side-by-side comparisons with live performances. Most often this is a smooth FR with at times a gentle downward slope at the top end depending on the tweeter used. However, I'm not always trying for perfect reproduction of the recorded material because much recorded material is not engineered well enough. This means that for a large number of recordings the reproduction is not as pleasing as it is for the well engineered ones. Quite often it seems to be the bass region (possibly more room influence than recording) and the mid to upper treble that cause my frustration.

It becomes something of a tradeoff. I either design for the best recordings to sound close to "accurate" to my ears or for the non-optimal to sound closer to "accurate". It's unfortunate that as my reproduction system has improved over the years, the deficiencies in recordings has become more apparent.

To me the question is what compromises, though small, must I make for the bulk of my recordings that I listen to most frequently? More and more it's towards those recordings that are well engineered and I just accept the frustration when listening to the poorer recordings.

Then there are those that may want or need some form or "equalization" due to hearing issues. I don't fall into that category at the moment, but I think that I may see the day when I'll prefer a possible up tilt in response. This won't be "accurate" to what I hear live, but I see no reason not to adjust a bit for "enhancement" at that time so that I "sounds more accurate" to my ears then. I just hope that this is many years in the future.

All one need do is listen to the old J. Gordon Holt monologue with the various microphones to understand just how much variation can exist between recordings.

Dave
 
Re: There is one thing to be said for the subjective

dlr said:
I try for what I think is "accurate" sound reproduction from my system. This must done by ear.

I couldn't disagree more. "Must [be]" should be changed to "shouldn't be"

However, I'm not always trying for perfect reproduction of the recorded material because much recorded material is not engineered well enough.

So you are going to take it upon yourself to "correct it" in your own system? And this makes sense to you?

This means that for a large number of recordings the reproduction is not as pleasing as it is for the well engineered ones.

It becomes something of a tradeoff. I either design for the best recordings to sound close to "accurate" to my ears or for the non-optimal to sound closer to "accurate". It's unfortunate that as my reproduction system has improved over the years, the deficiencies in recordings has become more apparent.

To me the question is what compromises, though small, must I make for the bulk of my recordings that I listen to most frequently? More and more it's towards those recordings that are well engineered and I just accept the frustration when listening to the poorer recordings.

Any other approach is senseless (except the "to my ears" part, which is wrong). You cannot correct a bad recording, but you can heighten the enjoyment of a good one. I would never compromise the playback of a great recording to accomodate a poorly engineered one. The facts are that not all recordings are that good - thats life.

Then there are those that may want or need some form or "equalization" due to hearing issues. I don't fall into that category at the moment, but I think that I may see the day when I'll prefer a possible up tilt in response. This won't be "accurate" to what I hear live, but I see no reason not to adjust a bit for "enhancement" at that time so that I "sounds more accurate" to my ears then. I just hope that this is many years in the future. [/B]

I do have a hearing loss (heck, I 56 and I played in R&R bands for years). One gets used to their loss and any attempt to change the playback makes the music sound unnatural. Its a fact that people with hearing loses don't realize it - because they have adapted to it. To try and "reverse" the situation is simply folly.

It is well know that people abhor hearing aids until they adapt to them. They just don't sound right until your ears are retrained for the new signal processing. This is also why once you start wearing hearing aids, you pretty much have to wear them all the time. There are not like glasses where you can take them off and put them back on.

If you have a hearing loss that is so bad that it affects your perception of music, then the only solution is hearing aids. Its ludicrous to even consider changing the audio system.

And thats not to mention that hearing loss is not simply a change in frequency response, which is a common misunderstanding. Its a highly nonlinear effect that really cannot be corrected in the audio system.

I'm sorry, but I really had to respond to your post because it had so much that I objected to. No offense intended.
 
Re: Re: There is one thing to be said for the subjective

gedlee said:


I couldn't disagree more. "Must [be]" should be changed to "shouldn't be"

I think that you may have taken some of what I said differently than I intended and I should have been more clear. I say "must be" for me, since I am designing for me and as I said, I cannot do a side-by-side comparison,to verify the absolute accuracy, neither by measurements nor by ear. I do 95% by measurement and CAD, but in the end I and I alone have to judge whether or not I've done it well. Quite often I find that for some combination of drivers and a "flat FR", the sound is definitely off, likely power response or pushing a driver to far at one end or the other, so I re-do it to correct and improve upon it. The gross errors don't require extensive measurements, the ear can and does pick out certain problems, especially in the early stages.

So you are going to take it upon yourself to "correct it" in your own system? And this makes sense to you?
Yes, it does as a matter of fact, but only when the problems are minor because the reproduction chain is not perfect. Nevertheless, no reproduction system is ever perfect beginning with the recording side, so it may be that a minor alteration in the playback side will produce something closer to the original for the majority of recordings to which one listens. That last caveat should not be glossed over. I can only guess what the original is, so I base my decisions on my personal designs (i.e. not for sale) using the recordings that are my preference.

Any other approach is senseless (except the "to my ears" part, which is wrong). You cannot correct a bad recording, but you can heighten the enjoyment of a good one.
Correct a bad one, no, improve the enjoyment by some slight adjustment, absolutely. Is it true to the recording? Not at all, but then why say that bad recordings must be heard that way if there is some way to improve upon them?

I would never compromise the playback of a great recording to accomodate a poorly engineered one. The facts are that not all recordings are that good - thats life.
As I said, my tendency is to try to make my system as good as it can be with good recordings, but there are times when I'd like to "improve" the playback of good performances that were unfortunately badly recorded and/or engineered. Were I to listen most often to a set of recordings that would be somewhat enhanced by some slight bit of EQ, then I think that I'd to that. Context of the discussion is important.

I do have a hearing loss (heck, I 56 and I played in R&R bands for years). One gets used to their loss and any attempt to change the playback makes the music sound unnatural. Its a fact that people with hearing loses don't realize it - because they have adapted to it. To try and "reverse" the situation is simply folly.
That is your opinion and I respect that. You seem not to respect the opinion of others in this regard and simply dismiss them. I'm 52 and my rolloff is somewhere around 16K. It may be that most people don't realize when their hearing acuity is declining, though I hope that I do recognize this. The thought of loss helps to keep my attention on preservation. My left and right ears were noticeably different to me when I was 12 and remain that way to this day. That may be why I'm more acutely aware of certain differences in systems and therefore focus more of my attention on the perceived response. Were I to suddenly lose hearing in the ear I deem to be more "accurate" or "natural" in sound I think that I'd be doing a bit of equalization because I do recognize the difference. It might prove futile in the end, but I would likely attempt it anyway.

If you have a hearing loss that is so bad that it affects your perception of music, then the only solution is hearing aids. Its ludicrous to even consider changing the audio system.
I was not implying severe hearing loss, I'm referring to age-induced frequency extension loss not of the gross variety. Maybe I'll not notice loss over time, but given that I was concerned enough at 12 to tell my parents that I heard slightly differently between left and right I suspect that I may be one who will recognize age-induced loss over time.

And thats not to mention that hearing loss is not simply a change in frequency response, which is a common misunderstanding. Its a highly nonlinear effect that really cannot be corrected in the audio system.
Fully correctable, I expect not. Improve upon the perceived response to some limited extent, I don't see why not. The biggest issue there is, of course, the individual nature of it. For the DIYer, it's certainly within the realm of possibility, even if only to a limited degree.

I'm sorry, but I really had to respond to your post because it had so much that I objected to. No offense intended.
None taken, I don't entirely agree with your positions. A forum such as this is intended for debate, is it not?

Dave
 
Re: Re: Re: There is one thing to be said for the subjective

dlr said:
Quite often I find that for some combination of drivers and a "flat FR", the sound is definitely off, likely power response or pushing a driver to far at one end or the other, so I re-do it to correct and improve upon it. The gross errors don't require extensive measurements, the ear can and does pick out certain problems, especially in the early stages.


Again, my experience in designing loudspeakers does not agree with yours. If you want to do things "your way" then there is nothing to talk about, but I think that my 35 odd years of experience might be worth considering.

... no reproduction system is ever perfect beginning with the recording side, ... I can only guess what the original is, so I base my decisions on my personal designs (i.e. not for sale) using the recordings that are my preference.

For all practical purposes, the reproduction chain up to the loudspeakers is perfect, or it can certainly be made so. The signal integrity prior to the loudspeakers is vastly better that it is after the loudspeaker making the degradation insignificant.

And I CAN use original material for comparisons, just not live performances. But the vast majority of recordings are not live performances but recordings meant to create an image in the listening space. Here accuracy is more easily defined and obtainable and it doesn't require the ear as confirmation.

Is it true to the recording? Not at all, but then why say that bad recordings must be heard that way if there is some way to improve upon them?

It its not "true to the recording" then its not an accurate reproduction - anything that is not accurate reproduction is personal opinion and not Hi-Fi.

That is your opinion and I respect that. You seem not to respect the opinion of others in this regard and simply dismiss them. I'm 52 and my rolloff is somewhere around 16K.

Well it really isn't my opinion, its well know in the hearing health care industry (my wife is Dept. Chair in Audiology at the local U). I respect others opinions, but I don't allow misinformation to be propagated as true either.

If we define "hearing bandwidth" as the 3 dB down points or more realistically the 10 dB down points then there is no way that you hear to 16k - that would simply be a phenomina in nature Most people are more than 10 dB down at 16k in their 20's or 30's if they were ever 0 dB at 16k. I would bet that you are down about 20 dB at 8k - that would be the norm. People think because they can perceive a 16k signal that there is no loss - this is not the case. Turn 30 k up loud enough and you will perceive something - probably a tweeter going into overload.

Fully correctable, I expect not. Improve upon the perceived response to some limited extent, I don't see why not. The biggest issue there is, of course, the individual nature of it. For the DIYer, it's certainly within the realm of possibility, even if only to a limited degree.

I continue to contend that nothing can improve upon accuracy - if I hear a bad recording as bad then thats what it is. My job is to create an accurate reproduction of what the producer - whose job it is to create the art - intended. I'll do my job and leave his job to him.
 
Hello Lynn,

Now that my issues of listening height have been addressed :D , I see that you are brainstorming some different driver configurations. Just as you got me talked out of 12+18 into 15x2, you change to 8+12+18 :headbash: (I don't think the aesthetics of an array of 12s will ever appeal to me). In any event, I have a couple of questions.

1. What thoughts have led you away from the "nothing compares to midrange from a 12" camp to an 8" for mids?

2. I know you are looking at pro-sound hi-efficiency drivers for (let's go back to the 15's for a moment) bass. Do you think the high Q and low effciency of something like Hawthorne's Augie would be worse than trying to equalize up a low Q pro sound driver? It would certainly make a passive solution easier to implement. I know Linkwitz's answer to this, I'm just not sure if somewhere in the last 100 pages, you've addressed it.

Thanks,
(the other) John L.
 
Re: Re: Re: Re: There is one thing to be said for the subjective

gedlee said:


For all practical purposes, the reproduction chain up to the loudspeakers is perfect, or it can certainly be made so. The signal integrity prior to the loudspeakers is vastly better that it is after the loudspeaker making the degradation insignificant.


I should qualify my statements a bit more. By reproduction chain, my intent was to include the recording, not just the playback, since it is all involved in arriving at a reproduction of the live event. In that regard, it is far from perfect. The chain from mic to speaker is, I will grant, much closer to perfect. The mic chosen and it's use as well as the speakers still are too far from perfect. This is of course, separate from the extra influence of the mixing.

And I CAN use original material for comparisons, just not live performances. But the vast majority of recordings are not live performances but recordings meant to create an image in the listening space. Here accuracy is more easily defined and obtainable and it doesn't require the ear as confirmation.

Seems to me that "an image in space" being a term with respect to perception by humans using their ears, it can then only be assessed with the ears. Use a mono recording and a mono playback system, then I might agree with you, but there wouldn't be much of "an image in space".

It its not "true to the recording" then its not an accurate reproduction - anything that is not accurate reproduction is personal opinion and not Hi-Fi.

Without the ability to compare the recorded material in the environment used to record and engineer, I would contend that it's not possible to accurately assess the recording for comparison with another playback system. But then the recording itself, due to the non-perfect nature, may itself not be fully representative of the original, original meaning live event, even if in a studio. In that regard, it may be possible to "reverse engineer" (so-to-speak), meaning some engineering choice, such as some excess rolloff in the bass region or over-emphasis in the highs, might be partially corrected. This is of course one area where you'd say leave it as it is. Others would ask why not try to make some improvement if it's possible. This doesn't begin to address the needs of any individual for limitations due to rooms that may not be correctable via room treatments or system/listener placement.

Well it really isn't my opinion, its well know in the hearing health care industry (my wife is Dept. Chair in Audiology at the local U). I respect others opinions, but I don't allow misinformation to be propagated as true either.

If we define "hearing bandwidth" as the 3 dB down points or more realistically the 10 dB down points then there is no way that you hear to 16k - that would simply be a phenomina in nature Most people are more than 10 dB down at 16k in their 20's or 30's if they were ever 0 dB at 16k. I would bet that you are down about 20 dB at 8k - that would be the norm. People think because they can perceive a 16k signal that there is no loss - this is not the case. Turn 30 k up loud enough and you will perceive something - probably a tweeter going into overload.

I'm not sure, but that's probably more accurately stated as being for people in a developed or developing country. Those in primitive environments not subject to the noise as we are do have better hearing into older age, if I recall some of the shows I've seen from researchers in those environments. If correct, that supports the benefit of taking great care, as I do, to limit that exposure.

I certainly can't say with precision what mine is, it is down some level at 16K, yes, but I feel sure that it's not down 20db at 8K. I've always gone to extremes to protect my hearing since my college days. But it's no use debating my personal hearing. That's a bit irrelevant to the discussion anyway.

I continue to contend that nothing can improve upon accuracy - if I hear a bad recording as bad then thats what it is. My job is to create an accurate reproduction of what the producer - whose job it is to create the art - intended. I'll do my job and leave his job to him.

The recording itself will certainly limit. However, I would contend that certain deficiencies in the recording that exist in all recordings, such as those that are related primarily to bandwidth limitations of the original hardware, those poorly recorded or engineered that have a similar result or mixed using playback systems that were themselves colored, are to some degree correctable in playback.

But to each his own.

Dave
 
I would bet that you are down about 20 dB at 8k - that would be the norm.

That's a very interesting statement. Could you elaborate on this a bit more (i.e., give a short overview of the typical hearing performance of the average population)? Typically only Fletcher Munson gets a quote in these forums.

I am asking because I strongly suspect that the 20-20k standard is grossly optimistic. I for myself can hear to 14k at best, and that's easily 20 dB down. And this has been the case since age 17 when I first did a test, before exposure to any extreme clubbing/motorcycling/etc. On the low side I suspect 40 Hz in my case (hard to assess due to woofer harmonics blending into the picture). I always wondered about the claims of many, to hear up to 16-18k and down to 20 Hz.
 
MBK said:


That's a very interesting statement. Could you elaborate on this a bit more (i.e., give a short overview of the typical hearing performance of the average population)?


I am asking because I strongly suspect that the 20-20k standard is grossly optimistic.


Hearing loss is inevitable as we age - the hair cells get brittle and break reducing our hearing acuity. The problem is that the variance of this effect is very wide. But typically we would expect a 50-60 year old to have a loss of about 20 dB at 8 kHz. This, by the way, is insignificant and would not be noticable by the subject or anyone around them. A real hearing loss just begins when you loose about 60 dB at 8 kHz. I once saw a guy who could converse with his hearing aids on that had about 100 dB loss across the band. The last time that I check I was down about 30 dB at 4k, but it came back up sligtly at 8 k. This is not typical - usually you loose from the top down.

Speech intelligability is not affected at all by a loss at 8k. Its not until you are 20-30 dB down at 4 k that you have trouble understanding speech.

My opinion is that anything above about 10k has no effect on audio quality (note the start of the flame wars on that statement). There is lots a real data to support this, not the least of which is MP3 and the like which basically throw away anything above about 8 kHz.
 
gedlee said:


...

My opinion is that anything above about 10k has no effect on audio quality (note the start of the flame wars on that statement). There is lots a real data to support this, not the least of which is MP3 and the like which basically throw away anything above about 8 kHz.

I've been spending so much time on that region 10KHz~20+KHz just because it makes a significant difference to my ears.:smash: The difference in transients and detail are quite audible at the age of 51 with quite a few db inbalance of hearing.
 
Earl,

thanks, interesting indeed, I wouldn't have put the performance that low.

Now, the thing is I can still perceive what I feel are HF artefacts, but I highly suspect that these are audible elements falling in the 2-6k band (where the ears are most sensitive). Any time I do sweeps I find myself surprised again at how shrill and high pitched even 6k sounds to me.

Soongsc,

Well, maybe individual differences are at play, or changes in phase if the system has a low cutoff, and especially, the wider dispersion of a driver that goes higher (as opposed to its on axis frequency response per se).
 
JohnL said:
Hello Lynn,

Now that my issues of listening height have been addressed :D , I see that you are brainstorming some different driver configurations. Just as you got me talked out of 12+18 into 15x2, you change to 8+12+18.

In any event, I have a couple of questions.

1. What thoughts have led you away from the "nothing compares to midrange from a 12" camp to an 8" for mids?

2. I know you are looking at pro-sound hi-efficiency drivers for (let's go back to the 15's for a moment) bass. Do you think the high Q and low effciency of something like Hawthorne's Augie would be worse than trying to equalize up a low Q pro sound driver? It would certainly make a passive solution easier to implement. I know Linkwitz's answer to this, I'm just not sure if somewhere in the last 100 pages, you've addressed it.

Thanks,
(the other) John L.

This might sound a little stupid, but the choice of any of these configurations depends on the sonics (and measurements) of the drivers. My own personal set of priorities are low subjective colorations in the 1 to 5 kHz region (requiring a minimum of equalization), traded off against in-band headroom. The 12" driver is almost certainly going to have an insane amount of headroom, but the subjective quality of 1 to 5 kHz region is probably not going to be as good, and there are secondary (to me) issues about dispersion vs crossover frequency.

The most annoying downside of the 8" driver is the reference efficiency, which is 3 dB less than the other drivers I'm looking at. In Class A triode-land, that makes a substantial difference for the required power amplifier.

In terms of "feel", esthetics, and a sense of "rightness", I admit the 8+12+18 layout with a 6-degree slanted baffle wins the prize. But - I really do have some serious questions about those 18-inch drivers - most have really horrendous things happening above 1 kHz, completely unlike their 15-inch brothers, which are readily available as midbass drivers.

I think part of this is the sheer weight of tradition, going right back to the Altec/WE Iconic, with its 15-inch driver and small multicell horn sitting at the top - the prototype for all 2-way studio monitors that followed. This "made" the 15-inch midbass driver, while by contrast modern 18-inchers seem to date back to the "Earthquake" movie and are seen as subwoofers. Thus, an implied 24 dB/oct lowpass crossover at 80~100 Hz, and too bad about any higher frequencies.

You bring up a good point about magnet size in the LF augmentation driver(s). A driver with a small magnet will have the "correct" high Qts, but an overall lower efficiency. The exact same driver (same cone mass and compliance) with a bigger magnet will have substantially lower QTs, require equalization, but have much higher efficiency. All of my instincts tell me to choose the driver with the highest in-band efficiency, and only equalize where necessary.

The same applies to raising Qts with a series resistor, which has precisely the same effect as reducing magnet size - in effect, you are throwing away efficiency of a device that already has abysmally low efficiency to start with. (Direct-radiator audiophile reference efficiency is typically 0.3 to 1.0%, direct-radiator prosound reference efficiency is typically 3 to 8%)

I'm quite OK with throwing away efficiency to improve damping and control resonance - back when audiophile drivers were smoother, and prosound drivers were really rough (Altec 604, anyone?), this choice made a lot of sense. Now that modern audiophile drivers have both low efficiency and rough response, they seem like a poor choice, considering the strides made in the prosound world.

I'd like to return to the larger point in the previous post comparing the level of engineering in the automotive world to audio - looking also at the level of knowledge of consumers, magazine reviewers, and the target markets the manufacturers are aiming for.

In the auto world, $100,000 gets you a very fast car with superb engineering, or a very extreme off-road vehicle. The claim to do both is ridiculous, and enthusiasts know this - but if you know cars, you can get an awful lot of vehicle if you're willing to pay.

Enthusiasts and magazines are aware of the subjective feel of solid-axle, trailing-link/swing-arm, and double-wishbone suspensions, of the torque curves and subjective acceleration qualities of small displacement/large displacement motors, and the feel of turbos and variable valve timing. A few minutes at the wheel, and you can feel these things immediately - scaring the salesman maybe (never as much fun being the passenger) - but the technical characteristics are immediately evident to an enthusiast.

Even the glossy magazines - with all the questions about corruption - continue to describe the technical highlights and road feel of the vehicle under review. They might minimize reliability issues - that's the domain of Consumer Reports with their user database - but they'll still bitterly complain about the BMW iDriveYouNuts interface or the uncertain future of the Jaguar/Ford cars.

Compare this to the abysmal writing in the Big Two audio magazines or the Web equivalents - the review begins with many paragraphs of an excruciatingly dull navel-gazing psychodrama centering around the reviewer's appalling taste in music, two or three paragraphs lifted from the manufacturer's "white paper" and slightly rephrased, and a hastily-written conclusion that either damns with faint praise - the stab in the back - or puts the component in the "must-buy" category.

You can read several pages of this bilge before you get to any description of sound at all - and then, the description is cursory in the extreme. The meaningless weasel-words "accurate" or "euphonic" almost always make an appearance. If you're really lucky, you'll see wacko reviewer-speak words like "chocolate midrange" or other, equally disastrous metaphors. Discussion of the underlying technology is typically absent, or at best, a direct repetition of the half-truths or outright falsehoods of the marketing literature.

You can complain all you want about auto magazines, but this level of writing would never be acceptable in the auto world - the readers would laugh it out of existence. So what's happened to the readers in the audio world? Why do they tolerate this?