Is it curious that none of the companies who profit from small allegedly hard to hear differences in sound have not clubbed together to fund the few thousands of dollars mentioned before more than once to test Earl Gedde's 5% question? If they did so, and it showed something, it would surely be a great aid to sales? Ie, worth the investment?
Worth to think about it again......
Ask yourself, what kind of corrobation/evidence would you accept and further would you buy a new expensive (more or less) audio gear just because somebody else could somewhere, somehow detect a difference?
In a broader sense, why do you think are results from controlled experiments are used so rarely in the marketing of the luxury segment (audio is luxury)?
It could be that the imagined "great aid to sales" does not work so well in reality.
Last edited:
A friend of mine shall be coming round to my workshop in a few days to listen to my hornspeaker build. Where I live is a wild land, and the weather is shall we say, variable. So for the past couple of busy months, I have had the horn mothballed under a tarpaulin snuggled up to a kilo bag of silica gel. Today, I set everything up, switched on, and from stone cold it sounded terrible.
For the next hour, I cleaned and tidied up the workshop, made more listening space in front of the horn, all while listening to music. It sounded laboured, as if the eq settings were trying to make the system sound like I wanted to hear in terms of my previous expectations. So, using my ears, I reset the eq to represent what I was hearing in the here and now - and to heck with the smiley eq face setting I had before.
And I was very pleased by what I could hear as a presentation of music. After everything had warmed up, and with a bit of flexing of the speaker cone, I was able (with a degree of objectivity inspired by an enforced rest) to get back into my zone of musical appreciation, simply because I used my ears as trustworthy arbiters of sonical taste.
I know you people are clever, and have all the gear, but some of us have just two ears to work with. Shall we say my hearing is way less than magical.
I'm not, and I don't, this shouldn't come as a surprise to the many well endowed audiofreaks here (I'm not sure what Jakob2 means by audiophile). If you have ears to hear and you love music and can read you can achieve a lot with the basic tools available to you and by following best proven electrical and acoustical practice, the rest is for fun and laughs, oh, and in some cases for extracting wads of cash out of people with more money than sense.
Over the decades, I have used the ears of people who could clearly hear better than me. I would still consider myself one of the 5%, but these people sometimes have amazed me with what they can differentiate in their hearing. Three of these people were: My ex, a former hi fi dealer and who was asked to be a reviewer by both TAS and Stereophile; Enid Lumley, formerly a reviewer for TAS and IAR, who was truly amazing with what she could do with my own hi fi system in getting it to sound its best; and a black friend that I met about 50 years ago, who has taught more people than me how to listen for differences in audio quality. All of these people are unavailable today, two are deceased, one just doesn't talk to me, but they each and all could hear better than me, and I thank them for their past input that helped me become a better audio designer.
Good attribution, John!
The very best ears I have known belonged to a 25 year old female mastering engineer named Ellen Threatt who could pick small eq and distortion artifacts out of a mix seemingly effortlessly. Once she heard them and described what she was hearing and where it was I could usually pick them out as well. Sometimes she didn't know what the thing she was hearing was due to but pointed it out to me and often I could identify it as a specific problem with a machine or process. She became my final arbiter of quality when releasing a master. Repetition was key with this, in that a casual once-through would sometimes miss the sound, but the more replays and the more I concentrated on what she was pointing out, the more likely I was to hear it. An interesting point is, after that I couldn't un-hear it!
I miss spending time listening critically with her, like you with your golden-ears I learned a lot and became a more critical listener and engineer. We took advantage of her talent and she trained all of our mastering and QC technicians, and there are thousands of Sheffield, Telarc, DMP and other titles out there that owe part of their sonic excellence to her hearing and judgment.
Given the nature of our hearing, I believe it is a truism that teaching your brain to listen critically is essential to hearing subtle sounds...perhaps in addition to having our ears tested (verifying the sensory apparatus), courses in critical listening or the equivalent hours of experience (educating the auditory cortex) are essential.
Cheers!
Howie
Scott Joplin, John Curl, I agree with both of you. This is why I have begun to invite friends and other intelligent folk round to listen to what I have created. In some ways it is not so much about how good their hearing is, more their intelligence and musical sensibility. It is very isolated here, and both friends and acquaintanceships tend to be quite serious. Once you trust someone, you may not know them very well, or see all that much of them - but they become friends through mutual respect. The familiarity of actually knowing someone comes much later.
I have spent quite a few days now putting together a playlist of music, of many genres, for my friend(s) to listen to. With an emphasis upon outstanding vocal performances, beautiful thoughtful lyrics, and excellent musicianship. If my friends are moved, and can articulate why in a way quite original, then that is good enough for me to continue in confidence with my audio projects. I expect (I hope!) we will talk about music. If the conversation is merely about audio, then I know I have failed.
As an aside, I have two artist painter friends who are a married couple. The wife has the keenest, most perfect emotional sense of riotous colour I have ever encountered, while her husband limits himself to a palette of just a few muted colours. I believe hearing can be very similar. It is one thing to hear the notes, quite another to appreciate how they interact.
I am beginning to believe that a good audio playback can, like a good visual painting, become a portal into another reality. This is what keeps me going.
I have spent quite a few days now putting together a playlist of music, of many genres, for my friend(s) to listen to. With an emphasis upon outstanding vocal performances, beautiful thoughtful lyrics, and excellent musicianship. If my friends are moved, and can articulate why in a way quite original, then that is good enough for me to continue in confidence with my audio projects. I expect (I hope!) we will talk about music. If the conversation is merely about audio, then I know I have failed.
As an aside, I have two artist painter friends who are a married couple. The wife has the keenest, most perfect emotional sense of riotous colour I have ever encountered, while her husband limits himself to a palette of just a few muted colours. I believe hearing can be very similar. It is one thing to hear the notes, quite another to appreciate how they interact.
I am beginning to believe that a good audio playback can, like a good visual painting, become a portal into another reality. This is what keeps me going.
Last edited:
I wrote a long reply to this but my connection failed & I lost it
So here goes on a shortened second attempt
The objective of our sense of hearing is to make sense of the world & it does this by attempting to map these signals to sound objects & generally to a soundscape that makes sense.
The problem is that there is a poverty of data in these signals - there is not enough data for the processing to solve the task to a unique solution (what is called an ill-posed problem in maths, I believe?) - in the field of linguistics the term "poverty of data" also suits but it normally is used in linguistics.
So auditory processing holds a number of possible solutions at any point in time with weightings associated with each solution from favored on down. As new data comes in at every moment these weightings can change & our perception shift.
Given this uncertainty, our auditory perception uses all sorts of techniques to try to settle on a unique solution. One of the main ways of improving this situation is to find auxiliary data in the form of nerve signals from our sense of sight. So now the processing engine has another set of data coming from the same physical objects some of which can be correlated to the sound objects. This helps greatly in removing the uncertainty in the process.
Remove this auxiliary data & we are back to a more uncertain determination. What mostly happens in Foobar style ABX listening is that we can't decide if A sound the same as X or is it B. We think we have it, only to second guess ourselves a little later. In the end we have no real certainty & it comes out no better than guess work.
Hence blind tests generally need well trained people & carefully controlled environments to deal with the natural uncertainty of the task. Look at that series of ABX tests I posted a link to & recognize that this guy is a recording engineer & yet it still takes great perseverance, flexibility, confidence in one's own hearing & tenacity to be able to achieve positive results with small audible differences
Your question assumes that Foobar ABX delivers the true hearing result & I'm saying that it is an uncontrolled listening which has little to no validity. The fact that there are hundreds of null results doesn't indicate to me that this is correct, it simply signifies to me that the test focus is exclusively to eliminate false positives & as a result there's a high preponderance of false negatives - this was already clearly shown statistically by Levinthal. It also exclusively ignores the necessary advice & conditions necessary for doing perceptual testing if it's meant to be something other than a bit-o-fun - people here are certainly not treating it as a bit of fun
So here goes on a shortened second attempt
Yes, your misunderstanding/misconception is that auditory perception is a brain function - the ear acts as the converter of sound pressure waves into nerve signals.
The objective of our sense of hearing is to make sense of the world & it does this by attempting to map these signals to sound objects & generally to a soundscape that makes sense.
The problem is that there is a poverty of data in these signals - there is not enough data for the processing to solve the task to a unique solution (what is called an ill-posed problem in maths, I believe?) - in the field of linguistics the term "poverty of data" also suits but it normally is used in linguistics.
So auditory processing holds a number of possible solutions at any point in time with weightings associated with each solution from favored on down. As new data comes in at every moment these weightings can change & our perception shift.
Given this uncertainty, our auditory perception uses all sorts of techniques to try to settle on a unique solution. One of the main ways of improving this situation is to find auxiliary data in the form of nerve signals from our sense of sight. So now the processing engine has another set of data coming from the same physical objects some of which can be correlated to the sound objects. This helps greatly in removing the uncertainty in the process.
Remove this auxiliary data & we are back to a more uncertain determination. What mostly happens in Foobar style ABX listening is that we can't decide if A sound the same as X or is it B. We think we have it, only to second guess ourselves a little later. In the end we have no real certainty & it comes out no better than guess work.
Hence blind tests generally need well trained people & carefully controlled environments to deal with the natural uncertainty of the task. Look at that series of ABX tests I posted a link to & recognize that this guy is a recording engineer & yet it still takes great perseverance, flexibility, confidence in one's own hearing & tenacity to be able to achieve positive results with small audible differences
Based on what I said above, I would expect any serious physicist to know the area that he is running tests in & known the scope & pitfalls in the test procedure itself.
Your question assumes that Foobar ABX delivers the true hearing result & I'm saying that it is an uncontrolled listening which has little to no validity. The fact that there are hundreds of null results doesn't indicate to me that this is correct, it simply signifies to me that the test focus is exclusively to eliminate false positives & as a result there's a high preponderance of false negatives - this was already clearly shown statistically by Levinthal. It also exclusively ignores the necessary advice & conditions necessary for doing perceptual testing if it's meant to be something other than a bit-o-fun - people here are certainly not treating it as a bit of fun
See my above simplistic explanation for the workings of auditory perception to see answer this
Last edited:
No, as I said, it takes a great deal of perseverance, flexibility, confidence in one's own hearing & tenacity to be able to achieve positive results with small audible differences - as will be seen in the link I gave to positive ABX results for small differences (differences that some on this thread & that thread say are impossible to hear)
The skew is that Foobar ABX is exclusively designed to eliminate false positives & as a result there's a high preponderance of false negatives i.e. real differences go statistically unrecognized
Last edited:
@ tapestryofsound. I get a little concerned when people talk about critical listening and training themselves to hear distortions, in the past when I've mentioned this to people they've assured me that it doesn't detract from their pleasure in the music, sometimes the opposite, I hope so. I suppose if your motivation is to reproduce music in a believable way then it follows.
I tend to agree with Earl Geddes insomuch as speakers and room is where the magic happens, DACs and amps etc are pretty much sewn up I think, although our personal taste in speakers is very different, he likes high directivity, I like more of a compromise that dipoles provide, which just goes to show how personal the whole matter of perception really is.
I tend to agree with Earl Geddes insomuch as speakers and room is where the magic happens, DACs and amps etc are pretty much sewn up I think, although our personal taste in speakers is very different, he likes high directivity, I like more of a compromise that dipoles provide, which just goes to show how personal the whole matter of perception really is.
Yes, although he did say to me he likes the small degree of spaciousness that he still gets from the late reflections. I think I probably like a little more, hard to say since the dipoles have a degree of controlled directivity too and I listen nearfield I'm under the impression I like the reflections off the front wall
Isn't this also the problem that Jakob2 was describing (the gorilla in the video) - what is "obvious" often means what is consciously registered & can be pointed to. Your friend could do this & when she identified the anomaly to you, you were then able to pick out the same, independently. Was this now "obvious" to you? When people can't do this, it's assumed they can't hear a difference & yet they may consistently prefer one playback over another
Yes, it was now obvious - which shows that auditory processing is changed by focus - we don't 'hear' everything at once, just as we don't see everything at once. If you had done any number of blind tests prior to this anomaly being pointed out to you, I suspect you would have not stumbled on this & your results would have reported no difference to be heard.
It also may signify something which I mentioned before - one of the auditory system techniques that seems to be used to store sounds is a statistical pattern of the sound from an analysis over time. I'm fairly convinced that we need repetition to accumulate & solidify such internal statistical models - I would go so far as to suggest that we have such models stored for how our playback system sounds based on this repetitive listening to it
I agree - learning to be able to spot the gorilla in the room in the video goes a long way towards being able to spot any gorilla in any room, at any time
Last edited:
If you like to check the numbers for "30% of the people on here" (even if strechted far out in using all the members) and the for "5%" of the population it really could be true ......
If I didn't have 2 small children and lots of new music to listen to I might try and do some filtering on posts by active members to try and work out how many do think they can hear tiny change from rolling wires/capacitors/resitors/opamps etc. It's certainly a significant number.
Haaaang on there Howie. Are you saying that my 6 children means that my perception is to warped to be able to do any critical audio comparisons
Good excuse...I will use it one day
I think I understand what you are saying ....... I really like listening to music through my hornspeaker, as it reveals everything, and feel that distortion is part of the music - except excessive compression and clipping, but that's personal.
I have some very early live recordings of Maria Callas on CD that are so mashed up that when she hits the high notes you can hear the oxide dropouts on the tape. It just makes it all the more real.
What I also really enjoy is hearing Transglobal Underground's use of boxy 10bit mono drum loops mixed with beautifully crisp atelier quality stereo miked vocals. Portishead's 'Dummy' album is famous for a similar kind of sonic mashup. And I truly love Joni Mitchell's 'Blue' for no apparent distortion of any kind at all.
So yes, by all means train the ear to appreciate music, and that becomes the definitive arbitrator as to what high fidelity is, or is meant to be. I'm still learning.
These days, what I want isn't sold so I have to make it myself. Or in a few cases is made but never sold for replay use. But I am aware that honest audio companies don't tend to stay in business these days.
As an aside discovered the annual london hifi show on this weeked and close enough that I could go. Looked at the list of exhibitors and absolutely nothing that would interest me enough to go there. Mainstream 'hifi' now has zero interest for me (or in me) as the goals are so different. I think I am one birthday from official curmudgeon status
Found this quote:
We must look at the mechanisms involved in hearing, and attempt to understand them. Through that understanding we can develop a model of the capabilities of the transduction and analysis systems in human audition and work toward new and better standards for audio system design.
- David Blackmer, founder of dbx
One must verify the test equipment surely (in this case our ears), to have results of any validity.
Also, he claims the time resolution of human hearing is 5 microseconds or better, which would correspond to a frequency of 200 kHz.
Thoughts?
Article here: https://www.earthworksaudio.com/wp-content/uploads/2012/07/The-world-beyond-20kHz.pdf
Touches on many of the themes currently being discussed.
Still completely baffled that our own hearing is the only thing not being measured in an effort to evaluate what people can or cannot hear.
We must look at the mechanisms involved in hearing, and attempt to understand them. Through that understanding we can develop a model of the capabilities of the transduction and analysis systems in human audition and work toward new and better standards for audio system design.
- David Blackmer, founder of dbx
One must verify the test equipment surely (in this case our ears), to have results of any validity.
Also, he claims the time resolution of human hearing is 5 microseconds or better, which would correspond to a frequency of 200 kHz.
Thoughts?
Article here: https://www.earthworksaudio.com/wp-content/uploads/2012/07/The-world-beyond-20kHz.pdf
Touches on many of the themes currently being discussed.
Still completely baffled that our own hearing is the only thing not being measured in an effort to evaluate what people can or cannot hear.
Last edited:
- Status
- Not open for further replies.