Head-Fi.org › Forums › Equipment Forums › Sound Science › Are there actually any Scientists in the Sound SCIENCE forum?????/
New Posts  All Forums:Forum Nav:

Are there actually any Scientists in the Sound SCIENCE forum?????/ - Page 3

post #31 of 79
Quote:
Originally Posted by Drag0n View Post
This is just what ive personally heard. YMMV.
But if you don't meet a/b/c and you haven't read the papers that tell you that you couldn't have heard what you heard, then you didn't hear it. (((Yawn.)))
post #32 of 79
I don't know why everybody gets so worked up and insults the OP.
He asked a question and many people posted offensive answers. He was even called stupid.
It is good to inquire and have an open mind. I personally like that.
Of course that is my opinion and this comment is not intended to offend anybody.
The OP has a point. Give him the forum and lets be civil about it.
If you don't like his posts, just put him in your ignore list.
I do prefer to read his point of view.
post #33 of 79
Wmcmanus, those yawns are making me tired.


gilency, I wasnt mean......i stated what i ("thought" for Wmcmanus benefit) I heard when i compared cables.

I think this thread is part of a huge conspiracy to overthrow Head-Fi, put out by some other "lesser" headphone forum. **Looks behind me as i escape**
post #34 of 79
Quote:
Originally Posted by SmellyGas View Post
I haven't seen any. I'm just curious. Are there actually any SCIENTISTS or RESEARCHERS here that:

a) hold either a Ph.D. in a natural science discipline (i.e. NOT engineering and NOT in an applied science) or an M.D.

- AND -

b) regularly design or interpret scientific experiments and/or papers designed to test hypotheses and draw conclusions

- AND -

c) are personally comfortable with statistical analysis

???

If so, what do you think about the published papers and experiments related to blinded, controlled listening tests of amps/cables/dac's?

(My prediction is zero people will meet a, b, and c, but there will be a lot of hateful and not-terribly-useful replies).
Sure. I'm a rocket scientist.
post #35 of 79
Thread Starter 
I think the responses so far really explain a lot. First of all, there doesn't appear to be many active participants here in the SCIENCE forum who are experts in the area of scientific research - that is, individuals with a doctorate or equivalent in a research-field, who regularly design and interpret scientific experiments, and who are comfortable with statistical analysis.

This makes sense. To date, I have read countless "arguments" here alleging that blind listening tests that fail to show LARGE, EASILY-AUDIBLE differences in cables, DAC's, and loudspeaker amplifiers are invalid for XYZ reasons. And every time, without fail, the "XYZ reason" has been something that demonstrates a complete lack of understanding of the scientific method, experimental design, interpretation of data, and statistical analysis. My favorite has been "I personally hear differences between cable A and cable B, therefore any blind listening test that doesn't demonstrate an audible difference is obviously invalid/flawed/etc." Individuals who are genuinely qualified and experienced in interpreting scientic research/publications would NEVER disqualify a study based on such inane reasoning.

I honestly cannot imagine how ANY experienced scientific researcher can read the several published blind listening tests available, all sufficiently powered to detect LARGE/AUDIBLE differences in cables/amps/DAC's, and all using a variety of valid blind/controlled protocols with many motivated listeners, and STILL conclude that cables/DAC's/power-amps make LARGE, EASILY-AUDIBLE differences.

So far, nobody who has met the criteria in the original post has also read the above-reference articles (linked in this thread) and followed up here with a conclusion different from mine: that cables/amp/DAC's most likely do NOT result in LARGE, EASILY-AUDIBLE differences.
post #36 of 79
Quote:
Originally Posted by SmellyGas View Post
So far, nobody who has met the criteria in the original post has also read the above-reference articles (linked in this thread) and followed up here with a conclusion different from mine: that cables/amp/DAC's most likely do NOT result in LARGE, EASILY-AUDIBLE differences.
Most, if not all of us here that have been debating the innards of DBT, cables, etc, would agree with this I suspect. Ofcourse there's the question of how large is "LARGE" and how easy is "EASILY-AUDIBLE", but broadly speaking I think most familiar with the studies would agree.

The more substantive discussions I've seen have been about non-large, non-easily-audible differences.
post #37 of 79
Guys

My take on this is that, "You don't have to be a weatherman to know which way the wind blows."

USG
post #38 of 79
Quote:
Originally Posted by SmellyGas View Post
cables/amp/DAC's most likely do NOT result in LARGE, EASILY-AUDIBLE differences.
Of course not. But it does not take a Ph.D. scientist to figure that out; nor is a Ph.D. scientist immune to being ignorant or making errors in judgment.

This section of head-fi was in part a refuge or concession to those at head-fi who dare speak the basic truths of audio, or something relatively close to them. It is a good place for people like me, who want to move a little higher up the audio learning curve. Let's not pollute it with elitism or by deifying those who have certain academic and professional credentials.
post #39 of 79
Thread Starter 
Quote:
Originally Posted by Steve999 View Post
Of course not. But it does not take a Ph.D. scientist to figure that out;
True. A 12 year-old could come to the same conclusion without even reading the published evidence. But I'm talking specifically about people who dismiss blind listening tests (including ABX) for some completely invalid reasons. It absolutely DOES take an individual experienced in experimental design/methodology, interpretation of scientific data, and statistical analysis to properly analyze and draw appropriate conclusions. What people don't realize is that blind listening tests are essentially human-subject research experiments designed to test whether a sample population is capable of differentiating between two entities (cables/DAC's/etc.). Critiquing such a study might involve determining the alpha and beta of the experiment to detect a pre-defined difference in the entities should be considered. It might involve examining the acuity of the human subjects or the generalizability of the study apparatus. But there is an alarming number of people who dismiss blind listening tests with reasons like:
"I personally hear differences, therefore the blind listening tests are invalid"
"Listening tests shouldn't be blinded, otherwise people can't hear the differences we all know exist"
"Well of course they didn't hear a difference, you have to listen for hours and hours"
"Oh, the listeners were too stressed out to hear a difference once the identities were covered up"
...the list of misconceptions and illogic goes on and on

And my contention is that NONE of the individuals who dismiss the validity of blind listening tests to detect LARGE, EASILY-AUDIBLE differences in cables/dac's/power-amps are actually QUALIFIED/EXPERIENCED to make this judgement based on a scientific premise.

Quote:
nor is a Ph.D. scientist immune to being ignorant or making errors in judgment.
Which is true. However, I believe that they are far less likely to "be ignorant" or "make errors" when dealing with an area in which they are an expert and have extensive experience.

Quote:
This section of head-fi was in part a refuge or concession to those at head-fi who dare speak the basic truths of audio, or something relatively close to them. It is a good place for people who want to move a little higher up the audio learning curve. Let's not pollute it with elitism or by deifying those who have academic credentials.
I'm interested in what people with extensive experience in research design, interpretation of papers, and statistical analysis think of the validity of blind listening tests to detect LARGE, EASILY-AUDIBLE differences in cables/poweramps/dacs. Are HS students likely to fall in this category? College grads? Masters? (2 years post-college, who are not usually the head of a lab)? Not usually! What about Ph.D. or M.D.? Bingo. So, please, take the accusations about elitism elsewhere.
post #40 of 79
Quote:
Originally Posted by SmellyGas View Post
True. A 12 year-old could come to the same conclusion without even reading the published evidence. But I'm talking specifically about people who dismiss blind listening tests (including ABX) for some completely invalid reasons. It absolutely DOES take an individual experienced in experimental design/methodology, interpretation of scientific data, and statistical analysis to properly analyze and draw appropriate conclusions. What people don't realize is that blind listening tests are essentially human-subject research experiments designed to test whether a sample population is capable of differentiating between two entities (cables/DAC's/etc.). Critiquing such a study might involve determining the alpha and beta of the experiment to detect a pre-defined difference in the entities should be considered. It might involve examining the acuity of the human subjects or the generalizability of the study apparatus. But there is an alarming number of people who dismiss blind listening tests with reasons like:
"I personally hear differences, therefore the blind listening tests are invalid"
"Listening tests shouldn't be blinded, otherwise people can't hear the differences we all know exist"
"Well of course they didn't hear a difference, you have to listen for hours and hours"
"Oh, the listeners were too stressed out to hear a difference once the identities were covered up"
...the list of misconceptions and illogic goes on and on

I'm interested in what people with extensive experience in research design, interpretation of papers, and statistical analysis think of the validity of blind listening tests to detect LARGE, EASILY-AUDIBLE differences in cables/poweramps/dacs. Are HS students likely to fall in this category? College grads? Masters? (2 years post-college, who are not usually the head of a lab)? Not usually! What about Ph.D. or M.D.? Bingo. So, please, take the accusations about elitism elsewhere.
No offense intended.

I appreciate your thoughtful post.

I would have thought you knew based on the date you joined head-fi, but perhaps you didn't know--Sound Science was named as such after rejection of other not so flattering titles for this forum thought up by our kind but not so scientific moderators. Head-fi chased off (my own characterization) a few people expert in audio over the years prior to that time. Head-fi is going to be generally lean on people with legitimate high-octane expertise in the areas of audio and science, because it is has generally been grown (and very well) by subjectivists, for subjectivists. Sound Science was conceived as a little outpost for what was once completely prohibited dissent, such as, for example, using blind listening test results as evidence. I appreciate that folks such as yourself have the potential to help it grow into something much more significant.

In my view, dismissing blind listening test results based on personal experience is not logically invalid. It is ultimately incorrect to do so in my opinion, but it is not illogical.

In my view, the people who say they hear differences when changing cables, amps, or DACs are often more honest and open-minded than those who say they don't hear any difference--because even a small movement of the audio transducers on the ears (with headphones) or change in head position or location of the person (with speakers) will result in a perceptible difference in sound.

And so, this is a field where an honest and good person may be easily misled by an unscrupulous merchant or a group of other misled individuals. These are people to whom your knowledge would be very valuable, if you could convey your knowledge to them. Why are the differences they truly perceive not proof of improvement? Now that is a complicated matter that requires all of the tools to which you allude. However, the good and perfectly intelligent people asking this heartfelt question will not likely be persuaded by your insistence and evidence that others with certain academic and professional qualifications who have read certain studies will very rarely, if ever, agree with their own empirically based conclusions.

To me, at least, your conclusions about cables, DAC's and amps are extremely well-proven, and are in fact are somewhat under-stated. From a scientific point of view, it seems to me to be the equivalent of playing chopsticks on the piano. Yet the state of popular belief and the popular press on the subject is quite alarming. The perspective that I am coming from is that I would rather see minds being pried open to the truth rather than closed shut to it. I realize that your agenda is different, but given your extensive and well-written reply, I thought I should explain myself better.

If you are truly simply interested in whether others with high expertise, knowledge and education would agree with your conclusions about cables, DAC's and amps, I can tell you the answer in advance--of course they will. It is not exactly a well-kept secret. If you want to find out for sure and you want company a little closer to your own knowledge, experience and credentials, try hydrogenaudio.org. It is a great community too. They will likely answer your question efficiently (if it catches their interest and they do not think it too mundane) and most "subjectivist" dissent will be squelched by the rules of posting, just as our general point of view was once squelched by the rules of posting here. They have, among others, members who develop and improve audio codecs for major commercial enterprises. I suspect you have visited there, but if not, it is very much worth the trip, just as head-fi is.
post #41 of 79
You put too much credence into what you consider as scientific tests IMO. This is the problem with the sound science forums. There's no one that looks at the data with a proper critical mind i.e. there are no real scientists here. It's basically the polar opposite of the cables and tweaks forums.

It's quite naive to view Science as absolute truth. Look at many cosmetic ads, they make certain claims (e.g. reduces signs of aging) with a reference to a publication is some journal somewhere. Often they use ABX. Do you find there results credible? I usually don't for various reasons. If you have enough money/power etc. you can get science to "prove" whatever you want.

Btw, I satisfy at least 2 of the 3 criterion listed in your OP, possibly 3 if you believe in quasiempiricism


Quote:
Originally Posted by SmellyGas View Post
I think the responses so far really explain a lot. First of all, there doesn't appear to be many active participants here in the SCIENCE forum who are experts in the area of scientific research - that is, individuals with a doctorate or equivalent in a research-field, who regularly design and interpret scientific experiments, and who are comfortable with statistical analysis.

This makes sense. To date, I have read countless "arguments" here alleging that blind listening tests that fail to show LARGE, EASILY-AUDIBLE differences in cables, DAC's, and loudspeaker amplifiers are invalid for XYZ reasons. And every time, without fail, the "XYZ reason" has been something that demonstrates a complete lack of understanding of the scientific method, experimental design, interpretation of data, and statistical analysis. My favorite has been "I personally hear differences between cable A and cable B, therefore any blind listening test that doesn't demonstrate an audible difference is obviously invalid/flawed/etc." Individuals who are genuinely qualified and experienced in interpreting scientic research/publications would NEVER disqualify a study based on such inane reasoning.

I honestly cannot imagine how ANY experienced scientific researcher can read the several published blind listening tests available, all sufficiently powered to detect LARGE/AUDIBLE differences in cables/amps/DAC's, and all using a variety of valid blind/controlled protocols with many motivated listeners, and STILL conclude that cables/DAC's/power-amps make LARGE, EASILY-AUDIBLE differences.

So far, nobody who has met the criteria in the original post has also read the above-reference articles (linked in this thread) and followed up here with a conclusion different from mine: that cables/amp/DAC's most likely do NOT result in LARGE, EASILY-AUDIBLE differences.
post #42 of 79
Quote:
Originally Posted by XXII View Post
You put too much credence into what you consider as scientific tests IMO. This is the problem with the sound science forums. There's no one that looks at the data with a proper critical mind
I like to think that I do. My Ph.D is in Information Science but I have done a load of experimental design and much peer reviewing of experimental designs, I have also designed and run my own audio experiments using non subjective data viz analyzing measurable audio properties such as frequency, noise and amplitude for different conditions viz CD players, filters, cables. While I may not be regarded as one of the white-coated community, as my Dad used to refer to boffins, I like to think that I can detect bollocks and absurd criticisms when I see them with equanimity .

However, whether you call yourself a Scientist or not is hardly the issue. The issue as you correctly point out is the extent to which one can critically evaluate "evidence".

Quote:
It's quite naive to view Science as absolute truth. Look at many cosmetic ads, they make certain claims (e.g. reduces signs of aging) with a reference to a publication is some journal somewhere. Often they use ABX.
This is a bit of a strawman. As you well know, there are journals and journals and the standard of review varies from "we read the abstract" to proper peer review and of course each field has its own definition of data and how measurements are made...saying there is n% reduction in the appearance of wrinkles is rather different than saying that the SNR was reduced from 102db to 99db by the use of blah drone etc
post #43 of 79
Quote:
Originally Posted by nick_charles View Post
I like to think that I do.
You excepted then

Quote:
Originally Posted by nick_charles View Post
I like to think that I can detect bollocks and absurd criticisms when I see them with equanimity .
It is my experience that this is not true with the general public.

Quote:
Originally Posted by nick_charles View Post
...saying there is n% reduction in the appearance of wrinkles is rather different than saying that the SNR was reduced from 102db to 99db by the use of blah drone etc
Very true. But how about audio ABX tests? How accurate are they? If they can be used to demonstrate (for example) that amps don't make a difference which I believe is against common experience, then shouldn't we at least question the efficacy of ABX tests in this field? Shouldn't this also pose questions on the accuracy of other results based on ABX tests?

Also, I can't resist but point out that the OP intentionally excludes engineers for some reason yet the journal he keeps referring to is an engineering journal.
post #44 of 79
Quote:
Originally Posted by XXII View Post
But how about audio ABX tests? How accurate are they? If they can be used to demonstrate (for example) that amps don't make a difference which I believe is against common experience, then shouldn't we at least question the efficacy of ABX tests in this field? Shouldn't this also pose questions on the accuracy of other results based on ABX tests?
There are a few DBTs that show audible differences between amps,

ABX Double Blind Test Results: Power Amps

if these are flawed then logically these have to be disregarded as evidence of audible differences between amps.

But, I have to say I think you are using somewhat circular reasoning here. You reason that amps sound different so tests that challenge this must be flawed.

What you describe as common experience is to some extent based on sighted evaluations which are difficult to use as strong evidence since the knowledge of the amps' identities may possibly bias the results of such tests since we are all human and fallible and influenced by things other such as appearance, price and so.

In fact there have been very few DBTs on amps but the biggest one was a Masters and Clark article from the 1980s

http://bruce.coppola.name/audio/Amp_Sound.pdf

which showed that many listeners were very certain that they could hear differences "sighted" , but when the visual cues were removed nobody scored reliably at a 5% level of confidence that they could distinguish between very differently made and priced amps - same kit, same cables, same speakers, same room, same listeners, just no visual cues - an interesting data point.

Quote:
Also, I can't resist but point out that the OP intentionally excludes engineers for some reason yet the journal he keeps referring to is an engineering journal.
LOL, I am a lapsed member of the AES, in fairness while the papers are variable they mostly provide enough detail so that you can evaluate the hypotheses and findings for yourself, the AES (afaict) has no hidden agenda, sometimes I have concerns about the usage of stats but mostly I think the papers are decent quality in that most are based on objective measures of one kind or another and there are relatively few opinion pieces
post #45 of 79
Thanks for the links, the differences between "sighted" and "non-sighted" are interesting but you've missed my point. I'm not saying that the test must be flawed but rather, when a test produces a result that is somewhat counter to "common intuition" (right or wrong), then it's worth considering whether there is something wrong with the test.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Are there actually any Scientists in the Sound SCIENCE forum?????/