Blind test. I constantly fail...

Discussion in 'Blind Testing and Psychoacoustics' started by murphythecat, Jun 3, 2017.

  1. Thad E Ginathom

    Thad E Ginathom Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    6,475
    Dislikes Received:
    11
    Trophy Points:
    113
    Location:
    India
    Hypnosis? lol.

    DBT is, of course, the works, and touching on scientific method, but, to my mind, questions like does green ink really make a difference to CDs? just need a helpful mate who can resist saying "I'm putting the green one in now." And it is resistance to that simple sort of curiosity test, when it comes to the frankly-unlikely, that I find hard to understand. Because why wouldn't a person want to try it?
     
    crenca likes this.
  2. anetode

    anetode Moderator

    Staff Member Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    564
    Dislikes Received:
    1
    Trophy Points:
    93
    Location:
    How did I get here
    Home Page:
    Mostly listener training and psychological priming. Meditation practices might work, enabling a sort of hyperfocus and/or reduced cognitive load. Similarly, subjects could be studied while already in the state of flow, e.g. seamlessly switching gear during a musician's performance to see if they notice a difference. All of these are ultimately mitigation measures and may only succeed in simply inspiring a smaller or slightly different set of active biases.

    Then there's the can of worms of using drugs.
     
  3. murphythecat

    murphythecat Friend

    Friend
    Joined:
    Apr 4, 2016
    Likes Received:
    774
    Dislikes Received:
    18
    Trophy Points:
    93
    I agree about drugs, especially psychedelics. but It still needs to be in a controlled environment where other perceptions have been shown to affect listening experiences.

    I used to do quite often mescaline coming from real cactus chips.. its incredible how on a good dose of psy, music sounds. the question of how good my system sounded is laughable. even my small fe127en with sub sounded incredible. even a ipod sounds incredible. music just sounds incredible on a psy.
    last time I did mescaline, I compared my jbl lsr32 vs my small p3esr + sub. I was able within seconds to know which speaker was my preferred one.
    I feel for speakers/headphones, its reall possible for anyone to beat their own bias, but for very subtle differences, like line level preamps, cables, caps, resistors, a very controlled environment is necessary.

    I really dont agree with critics about DBT, its the only way known to remove some well known bias. experienced or not experienced listeners are prown to bias and this have been well demonstrated. everything tend to show that even experienced listeners cannot beat their own bias. I say DBT is the beginning of the process, you can then have long listening sessions, or short, and apply all the other method to determine which is preffered between x vs y.
     
    Last edited: Jun 8, 2017
    Wilson and a44100Hz like this.
  4. Kattefjaes

    Kattefjaes Mostly Harmless

    Friend
    Joined:
    Sep 5, 2016
    Likes Received:
    5,111
    Dislikes Received:
    41
    Trophy Points:
    113
    Location:
    London, UK
    Nope, it doesn't. At the time of the green ink hysteria, my colleagues cheerfully analysed the resultant bitstream for timing or payload variance under laboratory conditions, just to make the point. They enjoyed that a suspicious amount - I think the green pen merchants annoyed them just that little bit too much.

    (A DBT would have been completely redundant there, as there's nowhere for imaginary nuance to hide.)
     
    Last edited: Jun 9, 2017
    Thad E Ginathom and anetode like this.
  5. Xen

    Xen Friend

    Friend
    Joined:
    Sep 30, 2015
    Likes Received:
    198
    Dislikes Received:
    1
    Trophy Points:
    43
    Location:
    Houston, TX
    VeHafVays - DrStrangelove.png
    The usual limits are moral, ethical, and resource. As an example, recombinant DNA technology was initially limited through a long series of intense discussions between 1972 - 1975, focusing on moral and ethical concerns. In 1976, the NIH released their guidelines, bringing in resource limits.

    Anyways, a bit more relevant. Many biochemical pathways have negative feedback loops. This dampens constant stimulation such that only short-term increases in magnitude produce a signaling event. Not having done much training in neurobiology, I will make the hypothesis that the brain deals with constant stimuli the same way. Evidence for this is the use of White Noise or Pink Noise to calm the brain. You stop hearing the noise and anything softer than the noise unless you actively think about the noise.

    Now, feedback itself provides a signal, like raising the noise floor. Your brain makes a new zero reference point well above absolute zero. If the constant low-level stimuli is removed, the feedback itself produces a signal. In hearing, this creates the "deafening sound of silence" or that feeling that something is off if someone changes one of those stupid motivation posters that you stopped noticing due to negative feedback loops.

    I think this is what Mr. Moffat refers to when he says that the brain is an "integral" engine. Your brain, being used to how something sounds, has produced an arbitrary "zero level" at various frequencies (how many different hair cell lengths are there in the human ear and what frequencies do they cover?). When you change something that your brain has been "desensitized" to, your brain is quite good at picking out differences. One reason why those "what changed" puzzles are so popular, brains usually enjoy that kind of exercise.

    Now the big elephant in the room... how do you ascribe the change your brain just discovered? For large changes, choosing "better" is usually easier (e.g. high distortion vs lower distortion or estimating threat levels after hearing a loud noise). For small subtle changes... well, we have this never-ending discussion that we have now. Is "hearing" feedback the same as hearing the signal? Can you remember the source that produced the negative feedback with enough detail to ascertain the reason for the negative feedback?

    DBT is more controlled and can have more controls applied to the test. Long-term listening has far fewer controls. However, there are some obvious controls that should be used for both test setups.
    1. Level match - use a multimeter and match voltage at the transducer if using the same transducer. If testing transducer, use SPL or a microphone at a precise displacement (magnitude AND direction) from the transducer.
    2. Remove sighted bias - somehow hide from the listener what is active.
     
    crenca, Cryptowolf, anetode and 2 others like this.
  6. murphythecat

    murphythecat Friend

    Friend
    Joined:
    Apr 4, 2016
    Likes Received:
    774
    Dislikes Received:
    18
    Trophy Points:
    93
    I agree but some seem to think DBT is flawed.
    my test of Ic's, line stage and coupling caps are by design level match. the biased was removed in that my GF or friend wouldnt see which input was selectionned and by the turn of a switch you could select input 1 or 2.
    to my ears and everybody who did the test, there was no change at all. no subtle change. there was a very slight change with pre amp and coupling caps, but so subtle that I and everyone who did the test could have never spotted them accurately.
    thats why many have recommended to me to try longer listening test.

    Im in the process of comparing my Mosaic T vs my Yggdrasil. this is very apparent and obvious that the sound is different, but I have to level match by ears so its not conclusive enough for me and maybe the differences I ear so obviously are due to slight level change
     
    Thad E Ginathom likes this.
  7. Thad E Ginathom

    Thad E Ginathom Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    6,475
    Dislikes Received:
    11
    Trophy Points:
    113
    Location:
    India
    Surely there's nowhere for imaginary nuances to hide in many things, such as wired ethernet? But nuances are found, and cables are sold.

    (Or, I might have missed your point?)

    @Xen, that is a fascinating post. I'm sure you have lots of good stuff to share on the processing that goes on in our brains. I'm lousy at remembering the technical details, but I think it is a lecture by JJ Johnston in which he describes how the vast amount of data arriving at the ear is filtered by feedback loops as it travels from eardrum to conciousness?
     
  8. anetode

    anetode Moderator

    Staff Member Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    564
    Dislikes Received:
    1
    Trophy Points:
    93
    Location:
    How did I get here
    Home Page:
    Tinnitus is a great example of this process, it is the consequence of your brain compensating by raising the "gain" on sensorineural activity.

    There are several approaches which might prove fruitful. One way that I'd like to try is to set up a listening lounge which is conducive to the sort of critical listening audiophiles tend to prefer, relaxed and unhurried. Another element I'd like to introduce is a mirror box, the successful use of which might be able to reverse the usual visual cues in a sort of sighted test.

    Speaking of boxes, another approach involves a literal black box which would contain several pieces of gear and a randomizer. This box would then be loaned out to audiophiles to test at their leisure, provided they abide by an honor code to maintain the integrity of the test. Sadly the results wouldn't carry very much water unless there is a record which demonstrates that the box is not tampered with. While test subjects might be inclined to invite this blackbox into their home, a camera to record the listening sessions would be a tougher sell and then there's the added expense of possibly trying to render the box to be tamper-proof.
     
  9. murphythecat

    murphythecat Friend

    Friend
    Joined:
    Apr 4, 2016
    Likes Received:
    774
    Dislikes Received:
    18
    Trophy Points:
    93
    well, it seems my test ( http://superbestaudiofriends.org/in...e-rolling-adventures.2034/page-11#post-134725 ) was flawed.
    I have to retract my test and conclusions

    Ive been informed that my test was invalid due to how my headphone amp/pre amp, the ifi pro ican, works:.ifi audio confirmed " This sort of interconnect comparison is unlikely to result in an audible difference, since the switching on your headphone amp most likely only switches between the hot leads of each input. So the ground paths of both will be in parallel all the time, rendering the test invalid."

    I really feel like a dummy, ever since my test, I was certain that cables didnt make any differences, but my test was invalid since the grounds were running in parallel.
    Just compared my AQ water vs my Analysis plus oval one and I do seem to hear more information, more ambiance, better soundstage on the AQ water on my Harbeth SHL5+. the AQ water stays.
     
  10. Walderstorn

    Walderstorn Friend

    Friend
    Joined:
    Mar 20, 2016
    Likes Received:
    540
    Dislikes Received:
    3
    Trophy Points:
    93
    Location:
    Europe
    Makes sense. That's why i never perceive my own perception as definitive, since there are so many variables...
     
  11. Darren G

    Darren G Friend

    Friend
    Joined:
    Apr 4, 2017
    Likes Received:
    612
    Dislikes Received:
    8
    Trophy Points:
    93
    Location:
    Pennsylvania, PA
    There is a series on Netflix titled "Brain Games", of which a few of those include variations of "what changed" puzzles. Of those, several demonstrate how bad we can be at picking out what changed (visually) when switching images. Auditory versions are of course less presentable in a TV documentary, but I do think they demonstrated that our brains don't necessarily instantaneously pick-up on all changes in rapid switch tests. Definitely was food for thought for me.
     
  12. Kattefjaes

    Kattefjaes Mostly Harmless

    Friend
    Joined:
    Sep 5, 2016
    Likes Received:
    5,111
    Dislikes Received:
    41
    Trophy Points:
    113
    Location:
    London, UK

    ..or you could just watch this:

     
    Last edited: Feb 1, 2018
    Dee Dubblya and Darren G like this.
  13. skem

    skem Friend

    Friend
    Joined:
    Nov 1, 2017
    Likes Received:
    1,220
    Dislikes Received:
    16
    Trophy Points:
    93
    Location:
    Cambridge, MA
    I think there's a bigger issue here. Perfection at what price?

    I'm willing to accept that there are differences in A vs. B; but if I cannot readily hear a difference in A/B testing over the types of music that I listen to -- then how much money (and time) am I willing to invest for improvements that are essentially inaudible?

    I feel that A/B tests prime you to hear differences, and that if they aren't readily apparent in A/B, then they sure as @#% won't be apparent when you just listen casually.

    That said -- I can fully appreciate that the quest is part of the journey and the joy, but *personally,* I think it wise to put a budget on how much I spend for this aspect of the experience.
     
    Last edited: Feb 1, 2018
    MrTeaRex likes this.
  14. Darren G

    Darren G Friend

    Friend
    Joined:
    Apr 4, 2017
    Likes Received:
    612
    Dislikes Received:
    8
    Trophy Points:
    93
    Location:
    Pennsylvania, PA
    LOL, and I missed all of those changes. Thanks for sharing this. Book-marked to share with others.
     
  15. Kattefjaes

    Kattefjaes Mostly Harmless

    Friend
    Joined:
    Sep 5, 2016
    Likes Received:
    5,111
    Dislikes Received:
    41
    Trophy Points:
    113
    Location:
    London, UK
    Glad you got a kick out of it too, it's an old favourite.
     

Share This Page