Measurements and the Correlation of What We Hear
- Thread starter mep
- Start date
If A then B then C. If Don agrees with Greg and Tim agrees Don, then that means Tim agrees with Greg. lol.
Although this thread is about electronics measurements, not speaker measurements, an interesting anecdotal fact is the famous myth that the QuadESL63 loudspeaker was developed by Peter Walker without listening to it. PW created a mathematical model of the wave pressure created by the electrostatic speaker mylar foil and all developments were carried from this model – including the concentric progressive stator sections and the delay line crossover. Never during design the speaker was measured as PW had no access to an anechoic chamber. But, although the prototype was created this way, critical developments such as the resistive cloth that dampens the mylar were carried by hearing – it is reported that the introduction of the speaker was delayed because of fine tune of some subjective aspects of sound carried by listening tests.
In the production line the speakers were never measured – their phase is so linear that the only test besides listening was acoustically comparing them with a reference pair playing a 400 Hz square wave in opposite phase to create a null in a microphone placed halfway between them. No measurement numbers were generated.
BTW, although people often attribute the sentence "if it can be heard it can be measured" to PW, he never said it, but he said something like "if a difference can be heard, it is our responsibility to develop a measurement technique that shows why it sounds different".
Do you think audiophiles should be sentenced to listen just to to good midfi until some one develops such measurement techniques?
Or should they join the high-end manufacturers in exile?
In the production line the speakers were never measured – their phase is so linear that the only test besides listening was acoustically comparing them with a reference pair playing a 400 Hz square wave in opposite phase to create a null in a microphone placed halfway between them. No measurement numbers were generated.
BTW, although people often attribute the sentence "if it can be heard it can be measured" to PW, he never said it, but he said something like "if a difference can be heard, it is our responsibility to develop a measurement technique that shows why it sounds different".
Do you think audiophiles should be sentenced to listen just to to good midfi until some one develops such measurement techniques?
Or should they join the high-end manufacturers in exile?
Maybe this is like most on-line debates a matter of semantics, maybe it is more than that. But IMO the objective of an audio system is to reproduce audio (which may or may not be music), which is not per se subjective. Perception may be subjective, but reproducing audio most certainly is not. The audio system does not know one way or the other if a living creature even is listening, let alone whether the sound is of a character some will enjoy, find fatiguing, or used for the purpose of, as the 2006 Isley Brothers' recent album was entitled, "Baby Making Music".
Then there is that pesky little term again, "better". Who gets to decide? What are the grounds for such a decision? What if someone disagrees?
It is one thing to argue that there are insufficient measurements which can be used to predict whether one will care for the sound of a particular component but quite another to assert that all will care for that sound as well as agree that that sound is *better*. Even Harman's speaker tests don't go that far.
The strawman argument apparently knows no bounds.
You know the world better than me Greg!
I should have guessed that what I wrote as a sarcastic joke could be taken seriously by some people.
Last edited:
Ron,
Not everyone accepts your perspective of what is an audio system.
I hope that everyone knows that what I write is IMHO. Can I humbly suggest that you read what F. Toole has written about the human capacity to perceive "what is good"?
Only this way I can be sure that you will not suggest that I am using your favorite "strawman argument" just because of the divergence in the definition of what means "better", or the way I express it.
Toole's words keep getting referenced around here in the context of anti-mesurement arguments. I don't doubt that he believes in listening, or that he believes that many things audio come down to choices. But I do not believe he is a subjectivist in the sense that the members here who are bringing him up believe he is. Could someone making this questionable connection please quote him, in context?
Tim
Tim, he is a staunch advocate of blind testing. What you've been reading is the cherry picking of language out of context.
This might be interesting if somebody would compile a database using this software.
"CRC has developed a novel and proprietary software called "Perceval" to objectively assess and measure the quality of a processed audio sequence relative to an original unprocessed version of the same sequence."
http://www.crc.gc.ca/en/html/crc/home/tech_transfer/10079
"CRC has developed a novel and proprietary software called "Perceval" to objectively assess and measure the quality of a processed audio sequence relative to an original unprocessed version of the same sequence."
http://www.crc.gc.ca/en/html/crc/home/tech_transfer/10079
This is a statement that you have repeated often, Tim, and I, and I think others have trouble with it. The history of science is littered with such bold assertions by those at the top of their fields, who have the highest level of recognition by others, as a reaction to some new bold concept proposed by some generally unrecognised, irritating whippersnapper. In fact, a chap called Einstein was virtually completely ignored for about a decade, after his most famous papers were published.
Frank
Looks interesting, Roger, except I note this point: "It is assumed that both versions are simultaneously available in computer files and that they are synchronized in time". This was the problem for Gary's experiments with TT sound, getting perfect synchronisation.
Unless this time matching is perfectly done, then any results will be very misleading ...
Frank
Tim-I don't know who you are referring to as anti-measurement around here. I for one don't consider myself anti-measurement. I'm trying to shine the light on the fact that audiophiles don't have many measurements to look at before they purchase a piece of gear. OEMs aren't providing measurements to their customers. And again, I'm saying the standard measurements that have been adopted and used since the 1960s for amps and preamps only show gross problems in designs and provide no guidance on how the gear will sound.
It's great to say you love measurements because it's motherhood and apple pie. It's great to say, "if you can hear it, it can be measured." The truth is you really don't have any measurements unless the gear you own was reviewed by a rag that takes measurements and stating something can be done is not the same as showing it has been done. It's all a bit of puffery on the part of those who proclaim to so dearly love measurements they don't have.
At the end of the day, subjectivists and objectivists are no different when it comes time to buy a piece of gear. We both have access to specification sheets the manufacturers provide. Some of us are lucky enough to live close to a high-end store that carries the gear we want to buy and we can hear it first. The rest of us have to rely on reviews, word of mouth or buzz from the net, liberal return policies, and faith.
Last edited:
Tim,
I hope someone is taking note of your straw-man use.
I was the only one who quoted Toole, not as an anti-measurent person, but as someone with an open view on the role of emotions in sound reproduction. I do not know of his opinions on electronic measurements as he only covers loudspeakers measurements and my post ended with a clear sentence:
"Disclaimer: as a few people have seen, my argumentation line is based in the writings of Floyd E. Toole about the musical experience and sound reproduction. This does not mean he agrees with many audiophile views, and his approach to maximize the quality of pleasure of the listening experience is different from most audiophile conceptions. By no means when I use his ideas about sound reproduction I am claiming support for my amateur views."
http://www.whatsbestforum.com/showthread.php?4149-Measurements-and-the-Correlation-of-What-We-Hear&p=69152
Anyway, the subject is closed to me - I feel very honored that my posts debating your views can be so interesting as they manage to attract the rage of two supermoderators, ignoring the main ideas and looking for second order details to weaken them. Anyway if you want to know more details read the F. Toole book "Sound Reproduction" , it is not expensive and you can get it at Amazon easily. He explains clearly what is in HIS VIEW the relationship between measurements (in loudspeakers in rooms) and emotions.
Last edited:
"It is one thing to argue that there are insufficient measurements which can be used to predict whether one will care for the sound of a particular component but quite another to assert that all will care for that sound as well as agree that that sound is *better*."
Maybe I'm just pulling at a thread(pun intended) but I certainly agree with that. Let's go a little further with the argument that the component " measures adequately to to produce a reasonably perfect replica of it's input.
Maybe I'm just pulling at a thread(pun intended) but I certainly agree with that. Let's go a little further with the argument that the component " measures adequately to to produce a reasonably perfect replica of it's input.
His best known book - Sound Reproduction has 522 pages. At an average of 30 lines per page and 3 lines per quote you can make about 522x30/3 ~ 5000 posts quoting it. Enjoy ...
Not having a go at anyone, but that's a bit of a weasel phrase. You can stretch the word "reasonably" great distances in many directions ...
Frank
Similar threads
