Aural hallucinations can occur for many reasons. It's a very easy thing to go a little bit off kilter. Usually it's voices, and the voices usually are very critical or threatening. Why? Some psychologists think its because brains are very attuned to detecting impending threats from other people. There is measurably more sensitivity to hearing threatening sounds than non-threatening. Apparently, its a kind of pattern recognition that when triggered gets forwarded to conscious awareness. False triggering is what seems to occur fairly easily.
EDIT: I have some tinnitus too. Do my best not to never listen to it or think about it. Not focusing attention on it help makes the brain think it is unimportant and can be safely ignored, at least if its mild enough to be ignorable.
EDIT: I have some tinnitus too. Do my best not to never listen to it or think about it. Not focusing attention on it help makes the brain think it is unimportant and can be safely ignored, at least if its mild enough to be ignorable.
Last edited:
Thankfully I don't, but a neighbour does, it varys but she says sometimes she hears what sounds like low flying jet aircraft that takes her so much by surprise it can make her duck.
Edit: That's interesting Mark, her jets could certainly be perceived as a threat
Edit: That's interesting Mark, her jets could certainly be perceived as a threat
Last edited:
It is so easy to go down that path... but "experienced" A/B-ers tend to resist...
It's not possible to concurrently listen 'analytically' to more than one musical stream in a piece of music - in other words if you are focused on trying to find a specific difference between two replays of the same piece of music, you focus on an element you believe is or should be different. This focus excludes other sounds - it's a well known auditory phenomena - even given a name - "inattentional deafness".
Now, it is possible to not do this & to de-focus as I believe you are suggesting - not listening in a focused, analytic way but instead absorb the way each playback effects you or holds your attention or presents a more natural illusion but this is not what is going on here as is evidenced by the o/p's response to Abaraxalito on page one about doing this. And from one of the participants in the test posting here, this certainly does not seem to be the approach taken
Last edited:
When a friend of mine first started getting stress-related tinnitus, he would halucinate conversations with people he'd had the previous day - his brain would recall the memory and recreate the neural activity to the effect as if he was hearing the voices in real time.
You can't prove a negative.
You can only say that none were noticed, you can never say that it proved they weren't there.
But you could be using the word "proved" to mean "convinced us", which would be true but completely different.
33 pages of replies, advice, personal views and opinion.... Wow.
Sometime more than 25 years ago I tried to sell a well designed UK DAC, simply because audio was a passion and because I believed in its superior sound. I thought lots of people would hear the dacs sonic superiority and magazine HFNRR had given it a fine review. It was expensive at the time and the only real competition was Arcam, Pink Triangle and DPA form the UK, Monarchy and Audio Alchemy spring to mind from across the pond. Because of its expense, I tried to do home dems that allowed potential customers to listen to it it at home for several days to get a handle on its performance.
I didn't sell one, and during one demonstration the DPA little bit Mk? dac at maybe half or a third of the cost sounded to my ears pretty much the equal of the dac I was trying to sell, a fact which I couldn't really argue and had to agree with the customer. The customer had a well put together system with good synergy with his existing dac. Further, the dac I was trying to sell went through a few filter subtle filter iterations over the months, which were probably less than 1dB variations towards the last octave, maybe some midrange dip for subtle voicing. I mis-wired one of these filter mods and was unhappy with the resulting sound so it had to go back for some checks.Two wires were swapped which effected the earthing of the filter stage. On the AP test set my mistake had worsened the linearity below 30dB's Full scale signal levels.
Consequently, taking the point that the majority of dacs now have very similar numbers to at least -100 dBs levels I am not surprised by the current outcome of your testing.
Could I detect the difference in your tests, I doubt it. Could I detect the differences here in my own system maybe?
Raspberry Pi and add on HAT dacs with suitable streaming software offer more than competent, sometimes stunning audio for very little outlay, less than a Friday night down the pub?
Finally, if you deduce that I was a crap salesman you may have a point....
Sometime more than 25 years ago I tried to sell a well designed UK DAC, simply because audio was a passion and because I believed in its superior sound. I thought lots of people would hear the dacs sonic superiority and magazine HFNRR had given it a fine review. It was expensive at the time and the only real competition was Arcam, Pink Triangle and DPA form the UK, Monarchy and Audio Alchemy spring to mind from across the pond. Because of its expense, I tried to do home dems that allowed potential customers to listen to it it at home for several days to get a handle on its performance.
I didn't sell one, and during one demonstration the DPA little bit Mk? dac at maybe half or a third of the cost sounded to my ears pretty much the equal of the dac I was trying to sell, a fact which I couldn't really argue and had to agree with the customer. The customer had a well put together system with good synergy with his existing dac. Further, the dac I was trying to sell went through a few filter subtle filter iterations over the months, which were probably less than 1dB variations towards the last octave, maybe some midrange dip for subtle voicing. I mis-wired one of these filter mods and was unhappy with the resulting sound so it had to go back for some checks.Two wires were swapped which effected the earthing of the filter stage. On the AP test set my mistake had worsened the linearity below 30dB's Full scale signal levels.
Consequently, taking the point that the majority of dacs now have very similar numbers to at least -100 dBs levels I am not surprised by the current outcome of your testing.
Could I detect the difference in your tests, I doubt it. Could I detect the differences here in my own system maybe?
Raspberry Pi and add on HAT dacs with suitable streaming software offer more than competent, sometimes stunning audio for very little outlay, less than a Friday night down the pub?
Finally, if you deduce that I was a crap salesman you may have a point....
The complaint isn’t based on disappointment in the results of the “test”. The complaint is that the outcome was certain, given the approach used. It provides no information at all. If you wanted to “prove” that differences between two things were inaudible, you would employ ABX testing. Like a game of Three Card Monte, the results are preordained.
It offends me that the OP’s beliefs about the audio industry, which depend on the existence of a rather large conspiracy, are being touted as fact through the employment of “science”.
This sort of approach is not a new phenomenon. Long before the OP’s began his ABX "testing", there was David L. Clark. In a 1987 Stereo Review article, Mr. Clark undertook an examination of the sound of amplifiers using ABX “testing”. From the February 1, 1987 New York Times:
“The amplifiers compared in this way ranged all the way from the budget-priced Pioneer SX-1500 receiver, rated at 45 watts per channel (wpc) and listing at $219.95, to the ultra-deluxe Futterman OTL-1, which costs $12,000 in its stereo version. Between these extremes, listeners heard the Mark Levinson ML-11 (50 wpc, $2,000), the NAD 2200 (100 wpc, $548) and Hafler DH-120 (62 wpc, $320). These were chosen to represent typical price points in the long ladder from the bargain basement to the stratosphere, and despite their divergence in power ratings, they were all adjusted to be heard at precisely the same volume level.”
Any guesses as to the outcome?
The “startling” conclusion was that there were no audible differences between any of the amplifiers!
I am guessing that on a site like DIYAudio, there are not many who believe that the sound of a Futterman OTL tube amplifier is indistinguishable from that of a cheap, mid-80s Pioneer receiver.
The results of the OP’s DAC “test” are just as meaningful as Mr. Clark’s. I believe one should be suspicious of a something characterized as a “test” when, in fact, the outcome is always the same.
It is like a Salem witch detector that amazingly, always determines that the subject is a witch.
Last edited:
"The startling conclusion was that there were no audible differences between any of the amplifiers! I am guessing that on a site like DIYAudio, there are not many who believe that the sound of a Futterman OTL tube amplifier is indistinguishable from that of a cheap, mid-80s Pioneer receiver".
If it was as simple as "That" and they were pieces of straight wire with gain and vanishingly small output impedance with sensible open loop gains and sensible feedback networks and for good measure a beefy psu, that would probably be true. If not designed correctly they will have at least subtle frequency response aberrations within their differing power output limitations.
As to the Salem witches wasn't that fuelled by lack of due process, family rivalries, social media and good home baking.
More cake pastor!!!!
I guess we will never know.
If it was as simple as "That" and they were pieces of straight wire with gain and vanishingly small output impedance with sensible open loop gains and sensible feedback networks and for good measure a beefy psu, that would probably be true. If not designed correctly they will have at least subtle frequency response aberrations within their differing power output limitations.
As to the Salem witches wasn't that fuelled by lack of due process, family rivalries, social media and good home baking.
More cake pastor!!!!
I guess we will never know.
It goes both ways. When somebody claims to hear something very tiny on the one hand, on the other hand claims not to hear something that shouldn't be tiny, lots of people get excited about it and start blaming each other for defects of character, defects of hearing, lying, etc.
Who really knows what happened somewhere when you weren't present. It's unlikely the DACs tested sound exactly the same to some people. It's also likely they do sound the same to other people. So relax, please.
If it gets heated enough the moderators will shut down the thread. Sometimes they do.
For this particular thread, it doesn't sound like the OP is asking for help to figure out why the results come out that way they do. Also, he's not providing especially detailed information about how the tests are conducted for anyone reading to duplicate with great precision. That being so, there is nothing to verify, and the results as reported could possible under some circumstances.
Also, the results of 4 or 5 people doing an undocumented test don't count for anything except something to argue about on a forum. It has no scientific standing at all. Not even AES would publish something like that, and sometimes they can be pretty lax.
Who really knows what happened somewhere when you weren't present. It's unlikely the DACs tested sound exactly the same to some people. It's also likely they do sound the same to other people. So relax, please.
If it gets heated enough the moderators will shut down the thread. Sometimes they do.
For this particular thread, it doesn't sound like the OP is asking for help to figure out why the results come out that way they do. Also, he's not providing especially detailed information about how the tests are conducted for anyone reading to duplicate with great precision. That being so, there is nothing to verify, and the results as reported could possible under some circumstances.
Also, the results of 4 or 5 people doing an undocumented test don't count for anything except something to argue about on a forum. It has no scientific standing at all. Not even AES would publish something like that, and sometimes they can be pretty lax.
That does indeed seem to be the case...
Apart from trying to avoid the "inattentional deafness", once we think we know which source sounds good, we always leave it to play the music at low levels, in the background, while enjoying a cup of coffee and having a chat about the meaning of life. It's only then that we can really start to believe that the A/B listening impression was truly "the correct" one.
But, what is "correct"?
For the professional musicians, it's the studio equipment-like sound that they prefer (this equipment is usually designed and built to the highest audio standards...). However (and it is a big however), this studio equipment may sound too sterile and analytical to others...
It could be argued that correct is either what measures best or what sounds best to you. Equipment designers often like to use both methods, go for good measurements and nothing objectionable in listening tests. Listening tests in that case may consist of a couple of days of critical listening with a variety of sources. It's a personal choice.
Last edited:
So all ABX tests give a null result, however different the items actually sound?Doppler9000 said:
That may be true: both tests could be equally meaningful. It is necessarily true if in both cases the equipment was good enough; all 'good enough' should be indistinguishable from each other. The only other way to get indistinguishable items is to have the same error in all of them - which is unlikely if they have very different internal architectures.
Does your 'offence' imply that you are part of the audio industry? If so, that might explain your concern. Certainly there are sections of the audio industry which depend for their existence on a group of people (journalists, customers) believing things which are not true. The only unanswered question is to what extent the industry itself knows these things are untrue; I suspect the truth is mixed on this. Some people are knowingly selling snake oil, while others are ignorantly selling snake oil.
Yes, that is wise but as is evidenced here (and in the example I gave in my link) these blind ABX tests are biased towards a null result because of the nature of the test, the nature of auditory perception & the fact that amateurs don;t understand these factors in their testing.
So there usually is no such preference to test further - the conclusion, as seen here, is that both devices sound the same just because the participants haven't yet found any element in the sound which differentiates them i.e they can't "spot-the-difference".
Which is why I suggest doing a blind preference test which is really what you are also suggesting i.e forget about trying to spot-the-difference & just focus on which you prefer without trying to analyze why.
Sure, it gets back to your question above - what is 'correct'. If talking about commercial recordings, we have no access to the original, pre-microphone sound so we don't have this reference point to judge the replay sound by. Second, the replay we are listening to is an artistic interpretation which has been modified through the various stages of production to produce the final version we are listening to. Again, we have no way of evaluating if what we are hearing is what was intended. So "correct" is impossible because of what Toole calls "cone of confusion" as there is no way we can audibly verify what we hear is what was intended.
Referring back to something I mentioned earlier, Harmon's professionally run (as opposed to amateur) blind preference test of speakers indicates that trained, untrained, musicians, non-musicians, etc. all gravitate to the same preference in sound (for this specific setup) & the good news is that this preference correlates with better measurements (of the set of measurements they have done). The point being that there is some internal reference that we all seem to use to judge sound
Last edited:
Why not do a distinguishability test where you can forget about spot-the-difference but simply decide same or different? Surely same or different is easier than better or worse?mmerrill99 said:
If so, then all we need to do is measure and achieve the best measurements possible at each price point. Wait a moment - isn't that what dastardly engineers are accused of doing by warm-hearted audiophiles (who much prefer sighted listening preference tests)?
For all the reasons I already pointed out that using an ABX test requires one to identify X as either A or B.
No, it's exactly the opposite - we can have a preference for one device over another just on gut feel alone without resorting to any analytic listening. Same or different usually requires analytic listening & all that implies, as I've already outlined.
Why not do both tests - do the preference test first & then ABX test - what's the problem if searching for truth in this matter?
We all know that measurements do not define how something will sound - listening is needed. I was careful to word that text in my post to show that " same preference in sound (for this specific setup) & the good news is that this preference correlates with better measurements (of the set of measurements they have done)."
You need to read the description of the Harmon test to ascertain what I'm referring to in my carefully chosen provisos in my text. But you are aware that correlation does not equal to causation, right?
Last edited:
Analytic listening would seem to require identifying the difference - but as soon as you have done that there is a risk of preference affecting the result. If it is a small difference than surely gut feeling can play a role too?mmerrill99 said:
If no difference is perceptible then there can be no preference. Why not save time by checking first?
Of course.
Last edited:
The question would be which part of the mind is mostly doing the perceiving and prefering, System 1 or System 2?
If System 1, there may be no recognition of difference available to System 2, but System 1 may send a "liking" or "disliking" type of emotion to some degree or intensity to System 2.
If System 2 is doing most of the prefering, it is influenced by what it is aware of, and also by what it is not aware of in the way of leaning or nudging from System 1.
That would be a System 1 influence. If you think it might be a confounder you could try repeating the test over multiple days.