Home Propeller Head Plaza

Technical and scientific discussion of amps, cables and other topics.

RE: If What You Claim Is True, Machines Should Be Able to Discern Jascha Heifetz from Itzhak Perlman......... (LONG)

Its very simplistic to pick away at this point by point and take things out of context. I never said machines can interpret music. What I said is an electronic device can resolve electronic values better than human hearing or other senses. We are dealing with electronics and signals, not music emanating from an instrument in an acoustic environemnt, a live sound. And this electronics is not involved in the creative process of creating or interpreting music itself. Electronics is used to record and replay music, within the degree it can do such things. However to subjectivly interpret the workings of electronics in the same way we listen to a violin in a concert hall is not only misplaced, but ineffective. If one was to "tune up" and electronic device by ear and sell it there would be so much variation from piece to piece nobody would buy it. Its function must be precisely measured by others of it kind; electronic devices. Just like only humans can interpret and apprieciate the music created by other humans.

Even piano tuners use electronic devices to get a more precise tuning initially, then and only then detune for timbral human preference. The note A is precisely defined as 440hz and only a very rare few can tune to within a cent or so of that without some basis of comparative pitch, that's why tuning forks exist. One can get close, one out of millions can get it right on without a standard comparitive pitch. Now we use electronics to have this standard. Otherwise, instruments would be out of pitch relatively. Now I have worked with symphonic orchestras as well as pop musiciains and all use some device as a standard, either a tuning fork, a piano to tune to or an electronic pitch standard. The human ear is not precise enough for the intial pitch. ALL calibrated electronic devices and tuning forks are exactly on pitch; exactly vibrating or oscillating at 440.000hz. Humans can only distinguish in mid-frequencies (specifically 440hz) about 1/12 semitone change, which is about .037hz This is known as the "Just-­Noticeable Difference". (See work by Rossing, Moore, and Wheeler) All electronic measurement devices can detect or produce frequency at resolutions as low as .001hz (or lower depending on clock accuracy). I would be more inclined to beleive a spectrum analysers output results for flat frequency response then depend on mine nor anyone elses ears to tell whether a loudspeaker has a flat response (necessary for accurate reproduction) or whether phase anomalies exist. (Let me say this, if a loudspeaker isn't flat it is adding timbral influence to the music that isn't on the recording. I want to hear the music, not the speaker)

Instruments not tuned a standard pitch reference was Mozart's biggest pet peeve. It is reported he had perfect pitch, and was always frustrated when a tuning fork wasn't used for the orchestra to pitch A to. Or if they didn't tune to the piano in place, which he always complained wasn't tuned to pitch anyway. However the person with perfect pitch is extremely rare statisically. The human senses are not that resolute or precise. A device is always more precise in this respect.

However, a device cannot interpret music or determine if it is performed well, nor can it derive pleasure or emotional response from it. But what electronics can do is measure whether another electronic device is doing its function accurately or properly. It can precisely read what the human senses cannot determine. Can your skin and sense of touch give you an accurate voltage reading? No. Can your ears and minds-eye give you and accurate picture or measurement of amplitude response. No. Can you say with all certainly what the phase response is of a certain frequency within a passage of music played through a speaker. No way. The human ear and other senses cannot do that to that level of precision. They are not evolved for that purpose. They can only interpret, not measure with certainly.

The limited abilities of human senses is well studdied, well documented and well understood. It is not a theory. It is fact. Comparatively speaking in terms of specific criteria, the human ear can only resolve to a certain degree. For instance it is well understood what the Hi-Lo frequency range limits are regarding human perception of frequency. Human hearing amplitude response is different depending on sound pressure level (munson curves). And these both vary greatly from person to person. For instance the average amplitude differnetial a human can detect is well known to be between 1 to 2dB and for some people as much as 5dB. Any modern spectrum analyser can resolve to about .1 dB or even lower in some instruments. And this level of precision has existed for some 30 years now. Human hearing cannot, has not and never will be able to resolve at that level of precision. Can a spectrum analyser tell if Itzak is playing? Of course not. That's not its function; not can it, will it nor has it ever been able to.

Interpreting music and measuring things like voltage, phase and amplitude are two entirely different things. You're comparing apples to rocks. They are two completely different unrelated things.



Edits: 11/24/12

This post is made possible by the generous support of people like you and our sponsors:
  Western Glow Tube Service  


Follow Ups Full Thread
Follow Ups

FAQ

Post a Message!

Forgot Password?
Moniker (Username):
Password (Optional):
  Remember my Moniker & Password  (What's this?)    Eat Me
E-Mail (Optional):
Subject:
Message:   (Posts are subject to Content Rules)
Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:  Automagically notify you when someone responds.