No, false dichotomy and category error. Rationalists make use of ears only for audio evaluations, using any method appropriate that eliminates non-auditory clues. Irrationalists peek and claim that this gives them some special insight into sonic differences.
Yes, excellent point. Most people can recognize a voice on the phone from someone you know before they identify themselves, even if you haven't heard from them in a long time,
even with the restricted bandwidth. Or, a singer you have heard, but with a song you haven't heard them do before. We all can do this, it's commonplace. This refutes the assertion.
Last edited:
How can we tell the difference between the sound of a wooden flute and a metal one? Some people can't, or don't care about the difference because the sound of a flute makes them want to leave the room.
So then I guess we have to ask "How can we distinguish the sound of a flute from any other sound?", but that just starts to sound like a silly question.
So then I guess we have to ask "How can we distinguish the sound of a flute from any other sound?", but that just starts to sound like a silly question.
This is the 'heart of the matter'. Can we hear subtle differences easily without always fooling ourselves? Or are we completely deluded into thinking we do hear subtle audio differences, to the point that all hi fi is essentially the same sounding unless there are GROSS imperfections in amplitude or distortion?
It appears that if the difference can be transferred easily into long term memory, then it will be easily distinguished in the future.
If the difference is very difficult to establish, the transfer does not take place . . . Other of course than the gross data about the event.
The categorial differentiation while transferring the information to the long term memory is the vital point. (Involvement of more brain areas seems to enhance the categorisation of percepted information)
As abraxalito i would be very interested to learn about any publication (with data) that answers the question where the threshold for "gross" actual is.
In research of short memory quite often random noise is used and arranged in various loop patterns with different length. Even with these stimuli some participants are able to identify quite long patterns and to remember these for weeks.
Obviously any difference only detectable with rapid switching would be of very limited practical relevance.
There is no need for rapid switching or for an "ABX" either.
The old von Recklinghausen rule comes to mind:
"If it measures bad and sounds good, it is good. If it measures good but sounds bad, you´re measuring the wrong things"
I rely on all of it. I measure, I study, I listen and I try everything... speaker design, acoustics design, video, audio recording/playback.... analog and digital. Buy test equipment for doing all these things and look/listen carefully... take specialized courses of study. Buy, build and test. Learn to listen to real acoustic sounds, record them, do anechoic chamber tests. Everything involved with the engineering of a Complete record/play 'system'. Then do it for your entire life as well as working in the field of electronics. After awhile.... you get to know a little something and can take a more holistic and integrative view.
THx-RNMarsh
To think that we can actually separate the long term memories from our instantaneous reaction is what I have a hard time understanding. I don't think we have that type of discrimination in our thought patterns where we weigh one thought vs another's ranking before we make a decision about a sound. I understand that in-order to make a judgement about certain sound qualities we need to know what a specific instrument should sound like live, this to evaluate the accuracy of reproduced sounds. At the same time we should be able to tell the subtle differences between two sets of speakers if it is subtle in reality. So using both A-B testing and actual sighted comparisons seems to be necessary as a combination, neither one alone, but both in proportion to what you are listening or looking for.
If we could design audio equipment based solely on measured response I think that most of the argument would soon leave the art of audio circuit design, but that does not seem to be the case, the final arbiter seems to be our ears, and there in our individual hearing there is no practical agreements that can answer all we have to ask.
The argument that you have to have long term memory of a live instrument to evaluate a musical piece leaves out the fact that every instrument sounds different than every other instrument, even from the same manufacturer and the same batch of production. So what is the real sound of a violin or a Cello, or even the difference between a wooden or metal flute? There is no one right answer, therein lies a big problem, how do you quantify the differences and measure them?
If we could design audio equipment based solely on measured response I think that most of the argument would soon leave the art of audio circuit design, but that does not seem to be the case, the final arbiter seems to be our ears, and there in our individual hearing there is no practical agreements that can answer all we have to ask.
The argument that you have to have long term memory of a live instrument to evaluate a musical piece leaves out the fact that every instrument sounds different than every other instrument, even from the same manufacturer and the same batch of production. So what is the real sound of a violin or a Cello, or even the difference between a wooden or metal flute? There is no one right answer, therein lies a big problem, how do you quantify the differences and measure them?
The problem is inherent in the question, which assumes a single hard threshold. Very Manichean, but that's not how brains work. I suggested an experiment you can do for the specific case of level differences. At what level difference can you distinguish things after (fill in whatever time lag makes you happy)? Will you have the same result for other material tested at other times?
For those who cant be sure enough of what they hear and which is better -- just start with a voice recording of someone you live with.... wife, GF, parent, your children..... you know their voice very well and can tell which playback system/component sounds most like that familiar person's voice. If you get that right, most of the music played will be right also.
THx-RNMarsh
THx-RNMarsh
The process that works for me, that I have mentioned many times, is that I listen only for misbehaviour, I take absolutely no notice of how well some aspect of the sound appears to be rendered. So, if the equipment is playing at a moderate level, with an extremely well recorded piece of relatively simple, untaxing music, then that is almost useless to me as a means of assessing anything - if I was forced to use that recording, I would simple raise the volume to the point where artifacts, disturbingly incorrect playback did start to intrude, and work with that.
Otherwise, I would simply pick from a huge suite of recordings that I have, that from experience will instantly highlight various types of misbehaviour - I'm not interested in listening for extended times, I want a snap of the fingers result, a "measurement" there and then. Some material I have is so taxing, that only at the highest level of optimisation is the rendition audibly acceptable - this is something that Dejan, for example, hasn't tuned into; that even the worst of the worst recordings can be brought into the light of day if enough efforts are made ...
Otherwise, I would simply pick from a huge suite of recordings that I have, that from experience will instantly highlight various types of misbehaviour - I'm not interested in listening for extended times, I want a snap of the fingers result, a "measurement" there and then. Some material I have is so taxing, that only at the highest level of optimisation is the rendition audibly acceptable - this is something that Dejan, for example, hasn't tuned into; that even the worst of the worst recordings can be brought into the light of day if enough efforts are made ...
Last edited:
Speaking of voices, the Vanguard recording of Odetta at Carnegie Hall is excellent for that aspect of fidelity - the slightest problem in the playback, and her voice has an artificial, "black woman singing caricature" quality to to it - I've used this one many, many times to evaluate the system ...
IME, the voice being familiar is not necessary, it's the quality of "realness" that counts - from another room, is it obvious that the "unknown" voice is coming from an audio system, or if you had to put money on a decision, there and then, as to whether the vocalisation was from a live individual or not - would you struggle with it?
I agree, I certainly don't have discrimination of my thought patterns in that way. Its instinctive, not based on thinking. I recall Malcolm Gladwell talking about this in his book 'Blink'. He inter alia, mentions bird lovers identifying birds from their 'giss' - Review: Blink: The Power of Thinking Without Thinking
Good point ... where I'm coming from, I prioritise realness - which is why I pay little attention to FR and similar attributes. If a person strongly focuses on accuracy, in the typically relatively easy to measure areas, then he probably wouldn't be happy with the standard of my reproduction.
- Status
- Not open for further replies.