André
I admire your tenacity André. But I think you are making this measurement thing more complicated than it is. All the wonderful things you experience whilst listening to music are all derived from an electrical signal which varies over time and very slowly as electrical signals go. If you look at a vynl record, the sound of every instrument with all their nuances has been reduced down to one "wiggley line" as Andy calls it. The shape of this one line is all that determines the sound you hear. Alter (distort) the shape and you hear something different from the original. We don't need to worry about trying to measure things in the sound that make us feel good, they vary from person to person. You say you don't believe in magic, I don't either.
I admire your tenacity André. But I think you are making this measurement thing more complicated than it is. All the wonderful things you experience whilst listening to music are all derived from an electrical signal which varies over time and very slowly as electrical signals go. If you look at a vynl record, the sound of every instrument with all their nuances has been reduced down to one "wiggley line" as Andy calls it. The shape of this one line is all that determines the sound you hear. Alter (distort) the shape and you hear something different from the original. We don't need to worry about trying to measure things in the sound that make us feel good, they vary from person to person. You say you don't believe in magic, I don't either.
GlidingDutchman said:
The ear itself is only part of hearing, the brain do all the complicated work and the brain can be trained. (Contrary to popular believe😀)
I'm certain that trained ears / brain can detect detail that will be hard to measure with current methods. Perhaps two mics and complicated software will help.
André
fredex said:
Yes, I agree with the one (two) "wiggley line(s)". Unless we try and preserve all of the detail and phase between the two channels, up to reaching our ears, we loose information needed by the brain to accurately recreate the "picture" (sorry) of the soundfield.
Not exactly what is attempted to be measured at this stage.
André
Andre Visser said:
Ah, your long term audio memory must be better than your short term lingual memory 😀 You wrote something about differences in cables that will never be measured.
Details, low level, ambience and everything else audiophiles seem to recognize in a piece of music is recorded and reproduced with equipment which is designed, built and tested using off-the-shelf technology.
There's no technical reason to assume that this technology would not be sufficient to fully characterize a simple piece of wire.
Perhaps for audiophiles it's easy to think of a zillion of important non-technical reasons which would require the invention of new measurement techniques to measure a cable. But as long as audiophiles prefer claiming over explaining and as long as they buy the stuff, the audio industry won't care.
Ravon, you read the way you listen. 😀
Here is what I said:
Not all equipment are created equal, not all can do it. Measuring and reproducing is not the same.
I can't help if ears / brain are better than current measurements. 😀
These measurements (if possible) will make for far more meaningfull specifications on a hi-fi system that are currently achieved. Measuring cables is only part of it.
Remember "audiophiles" are not necesarilly scientists, but if we can help, we will try. 😉 😀
André
@ fredex,
Andre Vissers comments are related to the pov expressed by R-Carpenter. In the same way you´ve argued you could draw quite the opposite conclusion.
If measurements are ´oversimplified´it may occur that any differences seems neglectible, but if you keep in mind that our understanding of the way our brain is processing the information catched by our ears is really limited, than you´ll have to very conservative in regard to dismiss any side effect as valueless.
As measurement by using specialized (and quite often simplified) test signal is needed for some reasons it is otherwise a bit dangerous to speculate about audibility with music signals based on findings with test signals.
Therefore difference measurements are a nice way to overcome this limitation in a way. It is possible to use normal music for difference measurement and even to listen to the difference signal to get an impression of its sonic nature.
Thats the reason why i often point to the freeware named diffmaker , which helps a lot in generating difference files.
An interesting statement from the mentioned Paul Frindle paper is:
´We have found that the only fundamental difference between people, of normal hearing faculties, in their ability to detect audio artefacts is experience. Even a trained professional may not be sensitive to an artefact he has not heard before.
And in addition a comment from Dave Moulton:
` It boils down to this obvious but inescapable fact: it is harder to correctly answer questions whose answers we don’t know than questions whose answers we do know. Setting aside the obvious issues of prejudice, bias and cheating for a moment, we will get “correct” answers more often when we “know” the answers than when we don’t. I’ve seen this effect a lot when doing my Golden Ears seminars (I publish a set of audio ear training CDs called “Golden Ears,” and often present ear-training seminars using them). Listeners asked to identify the difference between two versions of the same recorded excerpt will have real trouble, at first, hearing that one version is 3 dB louder than the other. Once they are told and shown that such a difference exists, they find it “obvious.” `
source is:
http://www.moultonlabs.com/more/wacky_world_of_blind_testing/
Andre Vissers comments are related to the pov expressed by R-Carpenter. In the same way you´ve argued you could draw quite the opposite conclusion.
If measurements are ´oversimplified´it may occur that any differences seems neglectible, but if you keep in mind that our understanding of the way our brain is processing the information catched by our ears is really limited, than you´ll have to very conservative in regard to dismiss any side effect as valueless.
As measurement by using specialized (and quite often simplified) test signal is needed for some reasons it is otherwise a bit dangerous to speculate about audibility with music signals based on findings with test signals.
Therefore difference measurements are a nice way to overcome this limitation in a way. It is possible to use normal music for difference measurement and even to listen to the difference signal to get an impression of its sonic nature.
Thats the reason why i often point to the freeware named diffmaker , which helps a lot in generating difference files.
An interesting statement from the mentioned Paul Frindle paper is:
´We have found that the only fundamental difference between people, of normal hearing faculties, in their ability to detect audio artefacts is experience. Even a trained professional may not be sensitive to an artefact he has not heard before.
And in addition a comment from Dave Moulton:
` It boils down to this obvious but inescapable fact: it is harder to correctly answer questions whose answers we don’t know than questions whose answers we do know. Setting aside the obvious issues of prejudice, bias and cheating for a moment, we will get “correct” answers more often when we “know” the answers than when we don’t. I’ve seen this effect a lot when doing my Golden Ears seminars (I publish a set of audio ear training CDs called “Golden Ears,” and often present ear-training seminars using them). Listeners asked to identify the difference between two versions of the same recorded excerpt will have real trouble, at first, hearing that one version is 3 dB louder than the other. Once they are told and shown that such a difference exists, they find it “obvious.” `
source is:
http://www.moultonlabs.com/more/wacky_world_of_blind_testing/
I believe this would be an easier way to look for the differences, just keep in mind that the differences we look for in cables will still be in the smaller signals, perhaps making it more difficult. I would like to see the results.
Very true, learning to listen to compare and evaluate, took me years and I'm still learning new things. The problem is we can't learn to listen if we do not concentrate and analyse what we hear and also the system must be able to reproduce these "things".
André
Andre Visser said:
Please feel free to think that the fact that I don't care quoting what you exactly wrote down is an indication for my reading or listening capabilities.
You don't seem to understand that I'm referring to a certain standard of technology.
No. Everything you can really hear can be measured using off-the-shelf technology.
Perhaps you could help me with this one: Suppose those measurements which produce meaningful specifications will exist this afternoon. And these measurements show clearly that there is no difference between two cables in between you claim to hear differences.
Would you accept that result? Or would you think that the measurement technique needs further improvement because what you believe is reference?
Andre Visser said:
Yes and peoples ability to run faster differs too, remember? Own back yard, etc?
So the psychogenic crowd believe that the phenomena they "hear" exists in physical reality, but "it's" properties cannot be measured eh?
Pray tell, how does one design and engineer a widget to have these properties?
cheers,
AJ
AJinFLA said:
Today is your lucky day. I just read that audiophiles are always willing to help.
ravon said:
Hey, that was meant as a joke.
Leaving out half of the sentence could lead to misunderstandings later on.
ravon said:
Can they really measure the way the brain hear already?
ravon said:
Ravon, believe is usefull untill you know, then you have no need for that believe anymore. I've done enough testing (blind and sighted) with consistent results that I have no doubt in the differences in sound between cables, on a good system.
Now shoot me and get it over with. 😀
André
@ ravon & AJinFLA,
i´m a bit wondering about your reactions to Andre Visser´s comments.
He stated clearly not to express a scientific validated opinion just some thoughts about apparent (more or less) differences between measurements and subjective listening experience.
Down to earth, if a complex acoustical stimulus is given, then our ear-brain ´tag team´is able to analyze it and draw conclusions about the number of instruments involved and the placement of these musicians in a threedimensioal space, along with paying attention to spoken/sung text passages, melodies played and so on.
Is there really a measurement tool that could supply the same informations, just from analyzing this complex signal?
BTW, there is a project for automatic evaluation of percepted audio quality :
http://www.peaq.org/
i´m a bit wondering about your reactions to Andre Visser´s comments.
He stated clearly not to express a scientific validated opinion just some thoughts about apparent (more or less) differences between measurements and subjective listening experience.
Down to earth, if a complex acoustical stimulus is given, then our ear-brain ´tag team´is able to analyze it and draw conclusions about the number of instruments involved and the placement of these musicians in a threedimensioal space, along with paying attention to spoken/sung text passages, melodies played and so on.
Is there really a measurement tool that could supply the same informations, just from analyzing this complex signal?
BTW, there is a project for automatic evaluation of percepted audio quality :
http://www.peaq.org/
Sorry, I didn't recognize it as such (not even with the smiley taken in account).Andre Visser said:
Is that what I said? Please keep in mind that there is a difference between measuring a signal and interpreting a signal.
What you really hear (my exact words) may not be what your brain makes of it.
@Jakob2: I don't know such a system. If you ask me it seems a bit out of scope for this discussion.
If the belief is strong enough all measurement techniques will be disqualified by believers.
Alan Hope said:
1. Is the amp two prong or three.
2. Where have you tied the safety ground?
3. Is there any line cord filtering, common mode or diff?
4. If there is any ground loop induced shield current, what does your amp do with this?
How would one null a stereo signal? Running a mono signal is not the same. We have to run a correlation between the two channels, looking for significant interchannel errors for specific signals.Conrad Hoffman said:
I'd build a contrived signal (actually, been there done that)...
1. Female vocals on center channel, in mono format. (sibilance required).
2. Null the TWEETER signals. Deep as possible.
3. Drive a single channel bass sine.
4. Check for null of the vocal.
The tweeter signal should not change...at all.
Any loss of null means the system reacted in a fashion not expected, and must be further investigated.
Culprit candidates:
1. Tweeter differences.
2. Crossover differences.
3. Connection resistance
4. Ground loop error signals coupled into one or both input signals.
5. Amplifier internal coupling (stereo amp)
6. Supply rail layout (error in damping factor through all four quadrants of operation)
7 Mutual inductance of supply capacitor runs.
audio-kraut said:
I know. You've claimed the room (et al) renders this moot.
I disagree. First arrival still occurs, humans key off this for localization (image placement).
audio-kraut As anecdotal evidence said:
Quite well stated. To date, most recordings use only the pan pot to locate images, so we are indeed fooling the brain. But that is only a contrived stimulus..and one which individuals will perceive slightly differently.
Cheers, John
ravon said:
Sorry then, I didn't mean to offend you.
ravon said:
That is exactly the point I want to make and the reason for saying that hearing is a training process. I believe that what we really hear depends on the analysing ability of that persons brain.
If it is possible to hear differences consistantly, then there must be some change somewhere in the signal.
ravon said:
No, I truly belief we will be able to measure the differences if we do it right.
André
ravon said:
Hmm..ok.
How does one measure the image placement stability of a female vocal located 25 degrees to the right of center, 12 feet from the listener, with a bass guitar playing center 9 feet out, low note organ tones at full power 35 degrees to the left 8 feet out, lead guitar 45 degrees to the right of center 6 feet out, t-line based cabinets for woofer reactance..
IOW, how do we measure ITD and IID for the first arrival specific signal (female vocal from lowest harmonic to 10Khz) buried within the entire soundfield of a room?
Microphones/spectral analysis? Doomed..
Remember, this requires finding the specified signal within the mud to within about 2 uSec and .2 to .5 dB accuracy, across the full frequency range of the vocal. Correlation is required.
Cheers, John
John, my mono test was only to address the original question of cable differences. Your test is excellent and covers a lot more potential error sources. I sometimes forget how easily we can now generate almost any test signal imaginable 😎
@ ravon,
we all have strong beliefs up to a certain degree. 🙂
In your case it seems that your strong belief is, that indeed we knew enough about our hearing ability to rely on pure (but somewhat) simple measurements.
In the end it is just a belief in psychoacoustical issues not so much in technical things.
Using the tool diffmaker it is quite easy to generate difference files using normal music samples. Of course, as mentioned before it should be done channelwise, because interchannel differences of the equipment will most likely set a lower border.
Try it for various interconnect and speaker cables and for sources or amplifiers as well.
You´ll find it hard to produce files containing just pure signal noise and even that could be further analyzed to ensure that no patterns are buried in the noise floor.
But if the difference signal contains something else than signal noise than would you ´believe´it is audible? 🙂
we all have strong beliefs up to a certain degree. 🙂
In your case it seems that your strong belief is, that indeed we knew enough about our hearing ability to rely on pure (but somewhat) simple measurements.
In the end it is just a belief in psychoacoustical issues not so much in technical things.
Using the tool diffmaker it is quite easy to generate difference files using normal music samples. Of course, as mentioned before it should be done channelwise, because interchannel differences of the equipment will most likely set a lower border.
Try it for various interconnect and speaker cables and for sources or amplifiers as well.
You´ll find it hard to produce files containing just pure signal noise and even that could be further analyzed to ensure that no patterns are buried in the noise floor.
But if the difference signal contains something else than signal noise than would you ´believe´it is audible? 🙂
I know. And, I liked the test..Conrad Hoffman said:
But think about what is happening at the front end.
Any ground loop caused by the IC's will intercept fields (and cause currents and voltages) as a result of the capture of stray magnetic fields. So, what fields generated by the system are we talking about that could couple to the ground loop??
1. The amp haversines.
2. The amp diode switching transients.
3. The amp audio leakage as a result of filter cap mutual inductances caused by layout..
4. Supply rail layout (or more accurately, lack of layout)...the bowl of spaghetti syndrome.
A single ended amp input will generate current based on the trapped field, the loop geometry, and the rate of change of the field, and will distribute the induced voltages based on the loop resistive and inductive distribution.
My test examines the effect of this coupling.
It also finds quite a few other things, so assignment of effect requires we investigate further to identify cause and effect. Finding a difference does not necessarily mean the IC's were the primary culprit.
Cheers, John
- Status
- Not open for further replies.