My views on audio reviewers

[fas42]

Tom, the only way the recording engineer knows what's encoded on the mix at any point in his activities is by listening to the tracks, singly, in any combination, and in total. And what he hears is always molded by the monitoring headphones, or speakers driven by the studio electronics ... the actual material on the tracks is always 'true' - he, doing his work is hearing just one version of that truth; imagine feeding the output of his mixing tool at any moment while he works, adjusting whatever, simultaneously to multiple audio rigs outside the studio by some means - and have various audiophiles, enthusiasts listen to each of those 'versions'. You would say the only valid version is that experienced by the engineer, which is correct in itself, but that version is impossible to replicate, in the consumer's environment.

The only way out, that I can see, is to be true to what is actually the data, the 'truth' of the recording - and that is what I try to do. Luckily , that's pretty damn impressive, and I'm quite content to stop at that point - it's giving me satisfaction in the hearing, so how the engineer wanted it to sound is not really interesting, for me.

 

[NorthSky]

If I may Tom, because it's a great subject, and part of the reviews by audio reviewers using music recordings with drum kits.
Get a drum kit right and you win a big chunk of the audio reviewer's 'sound' in his review.

• http://www.soundonsound.com/sos/feb03/articles/drummiking.asp
• http://productionadvice.co.uk/drum-mix-record/
________

Bonus:

 

[tomelex]

Tom, the only way the recording engineer knows what's encoded on the mix at any point in his activities is by listening to the tracks, singly, in any combination, and in total. And what he hears is always molded by the monitoring headphones, or speakers driven by the studio electronics ... the actual material on the tracks is always 'true' - he, doing his work is hearing just one version of that truth; imagine feeding the output of his mixing tool at any moment while he works, adjusting whatever, simultaneously to multiple audio rigs outside the studio by some means - and have various audiophiles, enthusiasts listen to each of those 'versions'. You would say the only valid version is that experienced by the engineer, which is correct in itself, but that version is impossible to replicate, in the consumer's environment.

The only way out, that I can see, is to be true to what is actually the data, the 'truth' of the recording - and that is what I try to do. Luckily , that's pretty damn impressive, and I'm quite content to stop at that point - it's giving me satisfaction in the hearing, so how the engineer wanted it to sound is not really interesting, for me.

Well, there is nothing wrong in my book for trying to amplify an exact replica of the recorded medium. Nor is there anything wrong with processing it additionally in a way that meets our preferences, if we so desire. The point of adding to OP the #13 is that the reference can only be the recording, as the original performance, or original panning for most newer stuff, is not even original, as a mic does not hear like an ear does. And if the recording is not much of a facsimile of the original event, then there is no absolute sound. As Tim has often said, the only "sound" is what is on the recording, and that is the only original, yet, it is NOT what the mix and master engineer heard, unless you have their ears and their gear. Your gear and my gear simply amplify the sound with the added distortions along the way and then our own unique ear/brain processes that, and the OP is about why audio reviewers are of almost no value to me, for all the reasons listed in the OP. Simply, there is no truth OR absolute sound, the only truth is for that last guy who did the mastering. Like Amir has pointed out numerous times, there are standards for video, but none for audio.

 

[fas42]

And I would agree with everything you said there, Tom.

 

[Cosmik]

Nor is there anything wrong with processing it additionally in a way that meets our preferences, if we so desire.

This is an idea that I find unconvincing i.e. that we audiophiles are so omnipotent that we can modify a recording to meet our arbitrary preferences without 'breaking' it.

There's an article I read a while ago that talked about how once we muck up some aspects of the sound, it then becomes impossible to determine whether other aspects of the sound are 'better' or 'worse', or 'correct' or 'incorrect'. Audiophiles could spend from here to eternity attempting to make a recording meet their (supposed) preferences, but the only one that 'hangs together' is the one that has not been messed about at all. As a simple example, with your tone controls you can't boost the level of the bass guitar without also boosting the bass on the vocals and everything else, because you don't have access to the individual mic feeds/recorded tracks. A better strategy would be for the audiophile to accept the recording as it is. He might find his "preferences" were not what he thought they were, anyway.

 

[RayDunzl]

As a simple example, with your tone controls you can't boost the level of the bass guitar without also boosting the bass on the vocals and everything else, because you don't have access to the individual mic feeds/recorded tracks. A better strategy would be for the audiophile to accept the recording as it is.

I accept all my (commercial) recordings at face value, and prefer to adjust some global deficiencies caused by the room.

 

[tomelex]

This is an idea that I find unconvincing i.e. that we audiophiles are so omnipotent that we can modify a recording to meet our arbitrary preferences without 'breaking' it.

There's an article I read a while ago that talked about how once we muck up some aspects of the sound, it then becomes impossible to determine whether other aspects of the sound are 'better' or 'worse', or 'correct' or 'incorrect'. Audiophiles could spend from here to eternity attempting to make a recording meet their (supposed) preferences, but the only one that 'hangs together' is the one that has not been messed about at all. As a simple example, with your tone controls you can't boost the level of the bass guitar without also boosting the bass on the vocals and everything else, because you don't have access to the individual mic feeds/recorded tracks. A better strategy would be for the audiophile to accept the recording as it is. He might find his "preferences" were not what he thought they were, anyway.

I am perhaps particularily sensitive to screeching (to me) treble. At first, I turned down the level on my tweeters, but that meant that everything I played was subsequently turned down, so I decided to use my tone controls for that. I am simply not a slave to the mix or mastering engineers idea of frequency balance. I don't always grab for the tone controls, but I have them and consider any pre-amp worth its salt to have those controls, bypassable of course. There is only so much I can do with my given room setup or my various headphone responses. My first priority is to enjoy the music not accept it, I am a picky son of a gun you could say!

 

[Fitzcaraldo215]

I accept all my (commercial) recordings at face value, and prefer to adjust some global deficiencies caused by the room.

Fully agreed. Part of the issue is what repertoire and implicitly what engineering practices are employed in those we tend to buy and to listen to. I may be fortunate in preferring classical hi rez Mch recordings from a small list of very intensely dedicated European labels. So, my chosen sample tends to be consistently rather good. I never feel the need to fiddle. Others might not be so lucky.

I think it may also be true that the raw response of many speakers with beautifully "flat" anechoic measurements are just too perceptually "hot" in the highs in most rooms. This also emphasizes any tendency toward "digititis" noticeable in the top octave. My room EQ used full range employs a downward sloping target curve with increasing frequency. I think that coupled with my choice of hi rez avoids much of the problem of hot highs and the "digititis" we know from CD. Plus, to me the highs, and everything else, sound more like the real thing live.

 

[AJ Soundfield]

I think it may also be true that the raw response of many speakers with beautifully "flat" anechoic measurements are just too perceptually "hot" in the highs in most rooms.

Based on what scientific evidence?

 

[Cosmik]

Based on what scientific evidence?

Surely this is related to your hero's circle of confusion. What is the scientific evidence for "the circle of confusion"?

 

[AJ Soundfield]

Surely this is related to your hero's circle of confusion.

Wrong. It's related to his objective claim, quoted.

What is the scientific evidence for "the circle of confusion"?

The light waves from your mirror?

 

[Cosmik]

Wrong. It's related to his objective claim, quoted.


The light waves from your mirror?

I am saying that if your audio science hero can make the assertion that there is such a thing as "the circle of confusion", then anyone else is entitled to assert that "perceptually", flat speakers sound "too hot in the highs in most rooms" using an equal amount of scientific evidence for their claim. To quote the article "You needn't look any further..." i.e. there is no scientific evidence for this, just 'common sense'.

The "circle of confusion" is the concept that shows up all audio science claims based on DBTs to be just as useless as all other audiophile claims. If the DBTs are made using pseudoscientifically-approved recordings (approved by arbitrary criteria suggested by the experimenter) then they do not have any direct relevance to real recordings used by real listeners that have been created on the subjective whims of real recording/mastering operatives in real rooms monitored on real equipment. It is a tough idea to swallow, but 'audio science' based on DBTs is referenced entirely to material created as 'art' by artists with the assistance of technicians who don't have PhDs. The criteria for selecting the material are arbitrary and subjective. Your cry of "Obtain your own data then!" is just another layer being balanced precariously on top of the shifting heap.

In audio, we just need common sense.

 

[AJ Soundfield]

I am saying that if your audio science hero

Yes, believer IQ types refer to Olive, Toole et al this way.

anyone else is entitled to assert that "perceptually", flat speakers sound "too hot in the highs in most rooms" using an equal amount of scientific evidence for their claim.

the raw response of many speakers with beautifully "flat" anechoic measurements are just too perceptually "hot" in the highs in most rooms.

Yes, severe reading comprehension maladies are part of the Dunning-Kruger package. Along with basic word meaning illiteracy. Thanks for illustrating.

In audio, we just need common sense.

Along with IQ, education, reading comprehension, etc, etc. sure. Self awareness is a good start, bravo.

 

[Purité Audio]

Floyd is my hero, certainly one of them, he is just so sensible,reasonable without any hint of condescension or aggrandisement , so very different from some of the puffed up ,conceited so called 'designers' who are floating around.
Keith.

 

[Thomas savage]

Floyd is my hero, certainly one of them, he is just so sensible,reasonable without any hint of condescension or aggrandisement , so very different from some of the puffed up ,conceited so called 'designers' who are floating around.
Keith.

Shame his disciples don't follow this as religiously as his teachings in audio.

 

[krabapple]

I'm with Floyd. Most audio reviewers and reviews are useless, being based on biased, sighted impressions.

Toole early on even showed that Harman's own in-house designers, who were convinced of their own exquisite aural discrimination, were prone the same sighted biases.

 

[krabapple]

Shame his disciples don't follow this as religiously as his teachings in audio.

You should watch the video I posted, Mr. Mod. The mostly genial Dr. Toole gets a bit tart and dismissive at times, giving the impression that he doesn't suffer fools gladly.

 

[Fitzcaraldo215]

Based on what scientific evidence?

Alas, I have none to present for your ever inquiring mind. Remember, this is NOT Hydrogenaudio, so we are free here to say "I think" something is true without catcalls from the Peanut Gallery.

However, what I think is based on measurements I myself or friends have made in a number of rooms with high quality speakers in the process of calibrating and applying DSP room EQ. The pre-EQ measurements tended to be more towards "flat" in the highs +- a few dB, often with a resonance peak in the 20k region. But, the sound was always preferable to all of us with the gently downward sloping final response, usually with a more rapid decline above about 15K Hz, usually 6 or more dB down at 20k. But, that is, of course, merely a small sample, and it is therefore merely anecdotal.

The response measurements I am talking about are of smoothed average response in the highs, which avoid getting hung up on all the narrow band comb filtering that occurs naturally in rooms. Amir has an excellent article here on that somewhere in the library, if you would like to dig further.

A number of other commentators with considerable credibility have also commented on the psychoacoustic advantages of smooth downward sloping measured response vs. measured response that was closer to flat in the highs. Sean Olive is one of them, and I am sure you can find his study results on various EQ tools, probably even somewhere on this site. In fact, I have never seen anyone with any credibility recommend "flat" measured in-room response with increasing frequency vs. gently downward sloping.

There is also an AES paper somewhere that I have seen by Tomlison Holman, et al. which summarized the before/after results of thousands of Audyssey Pro calibrations in various systems/rooms. I no longer use Audyssey, and I am no longer a fan of the tool compared to others. Audyssey Pro included a central data collection process for individual calibration results via the Internet, including the measurements "before". And, that paper, as I recall, confirms my impressions, that is, the in-room measurements pre-EQ tended to be closer to flat in the highs (often with resonance peaks) rather than downward sloping. Sorry if I cannot locate it for you.

So, I am a believer. What is your belief?

 

[Thomas savage]

I'm with Floyd. Most audio reviewers and reviews are useless, being based on biased, sighted impressions.

Toole early on even showed that Harman's own in-house designers, who were convinced of their own exquisite aural discrimination, were prone the same sighted biases.

I agree..

 

[AJ Soundfield]

Alas, I have none to present for your ever inquiring mind.

I know.

A number of other commentators with considerable credibility have also commented on the psychoacoustic advantages of smooth downward sloping measured response vs. measured response that was closer to flat in the highs.

Which confirms your state of utter confusion along with all others who repeat the same nonsense. Not one single technically literate sane person advocates for flat far field power. Zero. None.
It is the onset, direct on axis, near field (1m or so) frequency response that should, like your amplifier or CD player, be free of amplitude distortion. That, should be "flat".
But because loudspeakers tend to be directional at high frequencies (forward radiating narrow beam) and non-directional at low frequencies (omni) and the room provides far absorption at HF and loading at LF and there will be more HF propagation losses in air vs LF vs distance....the sum total of FRs/sound power at the listening position in the mid-far field, will indeed exhibit a downward slope...even when the onset free field response is "flat".
The exact slope is a matter of taste, but once again, no technically literate sane person would advocate that it be flat, which would mandate an extremely tilted up onset and be unbearably bright, even to those as deaf as audiophiles.