Don't we sometimes go too far in subjectivism?

[esldude]

The author addressed this issue in an article, explaining why the scale is dynamic and should be looked with some reserve and some care. Unfortunately every time some one googles his name and scale he just gets the scales and naively interprets them.

Well using his own explanation:

http://www.hificritic.com/colloms/sound.aspx

http://www.hificritic.com/colloms/ratings.aspx

This still doesn't really support his scale very well. His initial assessments were 0-10 with 10 being perfect. In time many improved products began reach 8 and then 9 regularly. So he expanded the scale linearly on the perceived improvement. Reaches near 100 now. Okay, so was his assessment accurate? Did so many products get to 9 of 10? If so how is the linear adjustment in relative terms now up to 10 times that amount. And if true as described are products better than perfect before, or is the scale now so divided we are talking improvements between 90% perfect and perfect expanded greatly. It would appear he is claiming to rate things in these percents or tenths of a percent even. I don't think the finest human ears are that stable from one day to the next. Sorry, not buying it.

MC was doing this in good faith, but I believe he has lead himself down the garden path quite a ways. And I still don't know if he means today's 99 is equivalent to yesterdays 9.9 or whether today's 50 is equal to the old 5 or is it halfway between 9 and 10 on the old scale. Again, there are simply too many problems with his scale. I don't know whether it is just a numerical version of saying I slightly prefer product A to B or whether it is an attempt to make his subjective assessments appear reliable on a genuine scale of some false precision. In years past I would have said the former though his writing in the last decade makes me think the latter.

 

[microstrip]

Well using his own explanation:

http://www.hificritic.com/colloms/sound.aspx

http://www.hificritic.com/colloms/ratings.aspx

This still doesn't really support his scale very well. His initial assessments were 0-10 with 10 being perfect. In time many improved products began reach 8 and then 9 regularly. So he expanded the scale linearly on the perceived improvement. Reaches near 100 now. Okay, so was his assessment accurate? Did so many products get to 9 of 10? If so how is the linear adjustment in relative terms now up to 10 times that amount. And if true as described are products better than perfect before, or is the scale now so divided we are talking improvements between 90% perfect and perfect expanded greatly. It would appear he is claiming to rate things in these percents or tenths of a percent even. I don't think the finest human ears are that stable from one day to the next. Sorry, not buying it.

MC was doing this in good faith, but I believe he has lead himself down the garden path quite a ways. And I still don't know if he means today's 99 is equivalent to yesterdays 9.9 or whether today's 50 is equal to the old 5 or is it halfway between 9 and 10 on the old scale. Again, there are simply too many problems with his scale. I don't know whether it is just a numerical version of saying I slightly prefer product A to B or whether it is an attempt to make his subjective assessments appear reliable on a genuine scale of some false precision. In years past I would have said the former though his writing in the last decade makes me think the latter.

The scales must be used WITH the reviews. Unfortunately people take them as ratings that can be used as absolutes. IMHO Hifi Critic releases them and the short information about them for advertising the magazine - Hifi Critic would be ignored by forums if it was not for such lists. The best of magazine is only available by paid subscription.

As we say your heart does not feel what your eyes do not see - some German and French audio magazines have similar, but much more detailed subjective global rating systems that are used to create their recommendation lists.

BTW, I would advise those wanting to get a fuller understanding of the article Amir referred to read "A SIMPLE NETWORK MUSIC SYSTEM BY NAIM" Martin Colloms gets to grips with Naim's UnitiServe/DAC combination - HFC Vol 5 no 2, and many other views of other authors showing different views on music servers in this magazine.

 

[Phelonious Ponk]

Jeez. Wouldn't it have been a whole lot easier to just say "the improvement in sub micro dynamic granular detail in the soundstage/time continuum rates a +3 on the Collums scale," or "oh this is unfortunate, I consulted my cat and the extension beyond 20 khz has taken on a disturbing opacity with this mains cable, -4 on the Collums scale?"

Yes. It would have been easier and it would have made no less sense, if that's even possible. But it would clearly ID the rating system as no more than one man's personal opinion, and would have implied nothing more objective.

Tim

 

[Atmasphere]

They are easy to get into a room WAF and you can get some really good sound from them. Add a sub and now you have almost full range performance in a very space friendly package. Perfect for a living room or a smaller second system if you don't have a dedicated room.

Rob

I knew that.

 

[MylesBAstor]

Fortunately we don't have to gauge such things, as this can't possibly be a problem. In high-end audio, bigger is better, the best speakers have tweeters and/or mids 3 feet above the seating position, and because small, more modest speakers can't possibly be better than big impressive ones, we have decided, against all reason, that stereo images vertically.

There. A perfect example of subjectivism (a very nice word for it) going too far; so far that it exits reality altogether.

Tim

I'm sure Alon Wolf will be very to explain to you Tim why the tweeter needs needs to be 48 inches off the floor.

 

[microstrip]

Jeez. Wouldn't it have been a whole lot easier to just say "the improvement in sub micro dynamic granular detail in the soundstage/time continuum rates a +3 on the Collums scale," or "oh this is unfortunate, I consulted my cat and the extension beyond 20 khz has taken on a disturbing opacity with this mains cable, -4 on the Collums scale?"

Yes. It would have been easier and it would have made no less sense, if that's even possible. But it would clearly ID the rating system as no more than one man's personal opinion, and would have implied nothing more objective.

Tim

Tim,

As far as I remember the reviewer with audiophile cats in Myles. And he prefers writing reviews about turntables, not music servers.

 

[PeterA]

I once started a thread on Audiogon asking "how close are we to the sound of the real thing"? I asked the readers to remember the best audio system that they had ever heard or one that they presume to be the best in the world, and then to assign a percentage to that sound relative to a live, unamplified musical performance. Sure they are very different, and almost no one would mistake one for the other, but how close are they, subjectively? Is it 90% close to the real thing? 50%, 5%. My buddy said that he thought the best system in existence is not even close to 5%, and he used the sound of a piano in a room which also had a stereo in it as an example. Clearly, he had not heard a really good system before. I then asked him if the best system in the world is only 5%, where does that leave a typical audiophile's system, a Bose wave radio, a typical car radio, a cheap transistor AM radio? We can listen to those and understand what we are hearing.

Many of the responders to that thread were hesitant to put a percentage on it. I contend that the mere fact that we can recognize that a recording of a piano on a cheap transistor radio is in fact a piano and not a harpsichord, guitar or bulldozer means that the SOTA systems are much closer than 5%.

I remember listening to a demo of a solo guitar playing at a dealership that was startlingly real sounding. It was not a full symphony, but still. It was there playing in front of me. Basis turntable, Clearaudio cartridge, Spectral electronics, MIT cables, Magico Q3 speakers. Surely not the best system in the world, but boy did it sound surprisingly real to me - tone, dynamics, presence, scale. If that was 5% like the real thing, I shudder to think what 10% (or twice as good) would sound like.

This reminds me of another question I asked a well known reviewer for TAS. I noted that I had read recent reviews in which the authors wrote that the sound of such and such tube amplifier was sounding much more real than previous generations. And that other recent reviews described the same thing happening with solid state amplifiers. So I asked quite simply that if that was in fact the case, and both types of amps were sounding more like the real thing, could I then not conclude that tube and SS amplifiers were sounding more and more alike? He wrote back with a simple one word response: NO.

 

[esldude]

This reminds me of another question I asked a well known reviewer for TAS. I noted that I had read recent reviews in which the authors wrote that the sound of such and such tube amplifier was sounding much more real than previous generations. And that other recent reviews described the same thing happening with solid state amplifiers. So I asked quite simply that if that was in fact the case, and both types of amps were sounding more like the real thing, could I then not conclude that tube and SS amplifiers were sounding more and more alike? He wrote back with a simple one word response: NO.

And of course the reviewer's answer makes little sense. Even if tubes and SS are approaching from vastly different directions compared to real, if each is getting more real sounding, then each must in some way be getting more alike at some point. Don't know the reviewer's opinion, but plenty have opined the two are getting further apart and better. Makes me think their grasp of where real is vs fake real is in error.

 

[DaveyF]

I have done this test a few times....we listen to a solo guitar recording, typically a steel string-- on an a'phile's system. I then play my steel string Dreadnought unamplified. The result has ALWAYS been the same...the listener cannot believe the HUGE difference between the 'live' and the reproduced.
Although I don't think that the reproduced is only 5% of the real, it's not that close in reality.....perhaps 30%.

 

[katylied]

Isn't the Linn table 734X better?

I remember that was the Oracle Delphi 734 (or something like that) times better than the Linn Sondek LP12 (IAR, end of 70s or early 80s).

 

[RogerD]

Remembering how Miles Davis sounded in a outdoor concert years ago and playing a cut from one of his live concerts,brought me back to that time. I wouldn't say that a 2 speajker system couldn't do it,recreate the ambience because some can,but with psychoacoustics no problem.

 

[microstrip]

I remember that was the Oracle Delphi 734 (or something like that) times better than the Linn Sondek LP12 (IAR, end of 70s or early 80s).

No, it was 634 times better , but is was not subjective data - it was really objective, established with measurements comparing the impulse response of the damped platter to the undamped felt mat platter of the Linn Sondek. This comparison in IAR was a key point in the success of the Oracle at that time.
We can read the Oracle history at http://www.oracle-audio.com/imagesup/event/35_delphi_audio_icons_hfn_april_2013.pdf

 

[MylesBAstor]

No, it was 634 times better , but is was not subjective data - it was really objective, established with measurements comparing the impulse response of the damped platter to the undamped felt mat platter of the Linn Sondek. This comparison in IAR was a key point in the success of the Oracle at that time.
We can read the Oracle history at http://www.oracle-audio.com/imagesup/event/35_delphi_audio_icons_hfn_april_2013.pdf

That's what I was thinking of. Thought it was 734 times.

 

[microstrip]

That's what I was thinking of. Thought it was 734 times.

I only posted because a 15% error would be adequate in a subjective appraisal, but is unacceptable in an objective statement.

 

[Atmasphere]

I asked the readers to remember the best audio system that they had ever heard or one that they presume to be the best in the world, and then to assign a percentage to that sound relative to a live, unamplified musical performance. Sure they are very different, and almost no one would mistake one for the other, but how close are they, subjectively? Is it 90% close to the real thing? 50%, 5%.

Many of the responders to that thread were hesitant to put a percentage on it. I contend that the mere fact that we can recognize that a recording of a piano on a cheap transistor radio is in fact a piano and not a harpsichord, guitar or bulldozer means that the SOTA systems are much closer than 5%.

I remember listening to a demo of a solo guitar playing at a dealership that was startlingly real sounding. It was not a full symphony, but still. It was there playing in front of me. Basis turntable, Clearaudio cartridge, Spectral electronics, MIT cables, Magico Q3 speakers. Surely not the best system in the world, but boy did it sound surprisingly real to me - tone, dynamics, presence, scale. If that was 5% like the real thing, I shudder to think what 10% (or twice as good) would sound like.

This reminds me of another question I asked a well known reviewer for TAS. I noted that I had read recent reviews in which the authors wrote that the sound of such and such tube amplifier was sounding much more real than previous generations. And that other recent reviews described the same thing happening with solid state amplifiers. So I asked quite simply that if that was in fact the case, and both types of amps were sounding more like the real thing, could I then not conclude that tube and SS amplifiers were sounding more and more alike? He wrote back with a simple one word response: NO.

Real music being the definition, of course the better tubes and transistors get the more they will sound alike, just as diverse speaker technologies will sound more and more alike the more they sound like real music instead of a reproduction.

Occasionally we get a phone call from a customer where they say they were at home playing their stereo and suddenly had the realization that someone was in the room with them, maybe someone that was taunting them by singing along with the stereo. When they started with trepidation and then looked where the singing was coming from they realized that no-one was there- that the stereo had fooled them. I've had this happen myself and its a little unnerving...

 

[Phelonious Ponk]

Remembering how Miles Davis sounded in a outdoor concert years ago and playing a cut from one of his live concerts,brought me back to that time. I wouldn't say that a 2 speajker system couldn't do it,recreate the ambience because some can,but with psychoacoustics no problem.

That's because Miles was playing through a sound system. If it was enough years ago, it was probably mono. If your system couldn't re-create a decent memory of that, that would be sad.

Tim

 

[Robh3606]

I have done this test a few times....we listen to a solo guitar recording, typically a steel string-- on an a'phile's system. I then play my steel string Dreadnought unamplified. The result has ALWAYS been the same...the listener cannot believe the HUGE difference between the 'live' and the reproduced. Although I don't think that the reproduced is only 5% of the real, it's not that close in reality.....perhaps 30%.

Hello Davey

OK but are you really surprised by that?? How was the recording made, close mic and how dry was it?? Any reverb added and god knows what else as far as EQ and what have you.

Rob

 

[DaveyF]

Hello Davey

OK but are you really surprised by that?? How was the recording made, close mic and how dry was it?? Any reverb added and god knows what else as far as EQ and what have you.

Rob

I'm not surprised at all. The fact is that regardless of recording technique ( and we have listened with various recordings, some done of my actual guitar)...the result has always been the same.

 

[Phelonious Ponk]

I'm not surprised at all. The fact is that regardless of recording technique ( and we have listened with various recordings, some done of my actual guitar)...the result has always been the same.

You shouldn't be at all surprised. A guitar doesn't radiate sound away from itself into a room like any speaker. The high-frequency information - pick attack, a string snapping against a fret, etc. does not come from one point in a vertical plane while the rest of the note comes from another point in a vertical plane. Its sound radiates from the location of the instrument itself, not from two locations to the left and right of where the instrument would be. There are a lot of very difficult challenges there that exist regardless of how good the reproduction system is, and reproducing an acoustic guitar is probably one of the easier challenges a system will face. Reproducing a jazz quartet, with three instruments and a drum kit -- another half dozen instruments -- all of which radiate differently than each other, and from any speaker....

Our perceptions are very forgiving.

Tim