INTERNET BLIND TEST: MQA Core Decoding vs. Standard Hi-Res (24/88 or 24/96)

Discussion in 'Audio Hardware' started by Archimago, Jul 15, 2017.

Thread Status:
Not open for further replies.
  1. Pastafarian

    Pastafarian Forum Resident

    Your wrong we can't discuss methodology and if people want to trust this, so be it, as I've said statistics can't support this and it's likely people will have voted without listening.

    Enjoy your MQA music.:righton:
     
    Last edited: Dec 14, 2017
  2. Musicisthebest

    Musicisthebest Exiled Yorkshireman

    Location:
    Manchester, UK
    Why is is likely that people will have voted without listening?
     
  3. Pastafarian

    Pastafarian Forum Resident

    As I did Psychology at Liverpool uni, our infamous hero was Cyril Burt and his twin studies, so one reason could be they'd had a dem at a dealers and they want the truth to be known.
     
  4. Musicisthebest

    Musicisthebest Exiled Yorkshireman

    Location:
    Manchester, UK
    As it happens I also went to Liverpool Uni but not to study Psychology so I'll google Cyril Burt & his twin studies.

    I don't understand the second part of your comment. If someone has had MQA demoed to them (& I haven't) why would this mean that they would want the truth to be known (whatever that may be) by randomly inputting answers (which is what people would be doing if they haven't listened to the files first)?
     
  5. testikoff

    testikoff Seasoned n00b

    Nobody's discussing the values of ABX testing here. All I'm asking is proof that one can indeed reliably discern the difference between two sources in a level-matched blind comparison (i.e. a successful ABX test report).
     
  6. Pastafarian

    Pastafarian Forum Resident

    OK but I'm saying there is no proof to be found here, so I can't help with that one, over to otherso_O
     
  7. Pastafarian

    Pastafarian Forum Resident

    I should have said more accurately, their truth and it's just human nature. Putting that aside if anyone here thinks about all the things members say effect their system and we've no idea about anything apart from what the files are, I wouldn't sell all my CDs and buy MQA on that basis.
     
  8. Encore

    Encore Forum Resident

    I still don't see what motivation people in Archimego's study could have had not to listen before casting their vote. Sure, an oddball might be so opposed to the idea of a blind test that they would sabotage it to the extent they can (which is to input random data points from one subject--i.e. themselves), but again, given the clientele at this site, I think that is highly unlikely.
     
    Musicisthebest likes this.
  9. testikoff

    testikoff Seasoned n00b

    For those willing to run ABX comparisons with excerpts of Archimago's samples, described in my previous post, the ZIP file can be found here... Good luck! :)
     
  10. Higlander

    Higlander Well-Known Member

    Location:
    Florida, Central
    This was a great idea, a great test, and frankly separates the claims from the chaff.

    After seeing this test implemented in this thread, I wish it would be implemented to other comparisons, that often end in arguing.
    Perhaps seeing something of substance would help nullify the bickering.
     
    Leonthepro and basie-fan like this.
  11. ralf11

    ralf11 Forum Resident

    Location:
    Earth
    that reminds me... what kind of HiFi system do they have on the Orville?
     
    Higlander likes this.
  12. Edgard Varese

    Edgard Varese Royale with Cheese

    Location:
    Te Wai Pounamu
    I've got three degrees in Psych and I know all about Cyril Burt. That particular (spectacular) case of research fraud has absolutely nothing to do with @Archimago's test.
     
    Chooke likes this.
  13. lechiffre

    lechiffre Forum Resident

    Location:
    phoenix
    One would have to have some pretty boss ears & gears to hear a difference.
     
  14. Pastafarian

    Pastafarian Forum Resident

    It may have if wanting to reveal the truth caused someone to provide some false evidence, that's the problem we don't know. Anyway back to HI FI, perhaps you need a £2000 mains lead to show the amazing sound of MQA
     
  15. Edgard Varese

    Edgard Varese Royale with Cheese

    Location:
    Te Wai Pounamu
    The difficulty is that, under that logic, all science is suspect (or worthless) because there might be fraud taking place. I'm a scientist, and I can't really take that position.
     
    wgriel and Musicisthebest like this.
  16. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    I do. Thanks. :righton:

    I would hope that people here have the gears and ears to hear a difference between these samples. Cause the samples do sound different.
     
    Musicisthebest likes this.
  17. Pastafarian

    Pastafarian Forum Resident

    As we're parking in MQA's way, this will be my last input on scientific inquiry but yes all studies are subject to peer review, fraud itself will be rare but confirmation bias is a very real danger and I'm presuming the majority taking part in this study aren't trustworthy scientists. The answer to has it been a genuine oversight or something akin to fraud, takes us to a possible legal defense.
     
    Last edited: Dec 15, 2017
  18. testikoff

    testikoff Seasoned n00b

    Then make a little effort to post your ABX test result(s)... ;)
     
    Chooke and Brother_Rael like this.
  19. ralf11

    ralf11 Forum Resident

    Location:
    Earth
    how could confirmation bias be present at all here?

    the subjects (here the listeners or listening panel) are normally not the scientists when a study involves human subjects
     
    Musicisthebest likes this.
  20. Pastafarian

    Pastafarian Forum Resident

    Easiest way not to be wrong, or intentionally skew results, vote no difference.:shh:oops, I said no more of my cr*p.
     
  21. Encore

    Encore Forum Resident

    Maybe I'm missing something here, but wasn't the test made without the subjects knowing which files were the MQA versions and which were the PCM versions? If not, then I would agree that the design looses some of its strength. I still think there's very little risk that the subjects have deliberately cheated but confirmation bias is something that we can't be sure to conquer, even if we want to. Especially not in AB tests with quick shifts, which is why I don't believe in them as a method.

    But in the end, science does ultimately rely on honesty. Or on the fear of being caught cheating. The peer reviewer typically doesn't suspect the authors of a scientific manuscript of lying or having cheated, only of misinterpreting data or having been sloppy.
     
  22. Pastafarian

    Pastafarian Forum Resident

    My incoherent input has come from me not wanting but failing to get in to the methodology yet answering questions as they've arose. I think the basic problems are unknown variables, some participants will have had a belief about MQA's merits, this can be seen on this and other sites and we don't know what the results would be if the files were identical. Of course I could be wrong but it's possible I'm right so a study should try and address the issues raised. I don't think honesty is the issue when it's confirmation bias.

    In fact Archimago addresses a lot of these problems on the blog, so I don't think I'm saying anything that's particularity contentious.
     
    Last edited: Dec 16, 2017
  23. Ric-Tic

    Ric-Tic Forum Resident

    Location:
    Stockholm
    What about your (first) comments about Archimago's design being bad, and not *science* enough for your taste/liking/ whatever?
     
  24. Claude Benshaul

    Claude Benshaul Forum Resident

    He does and there might be some serious problems. But as long as no other scientific test are performed, and that's pretty much what Bob Stuart admit, his test is as much valid as any other and is head and shoulder above the presentations, nonsensical articles, controlled demo and marketing Newspeak we got so far.
     
    LarryP, Robert C, Chooke and 3 others like this.
  25. Encore

    Encore Forum Resident

    Fully agree. But I just checked again--the subjects didn't know which files were which, so I don't how confirmation bias could have been at play here.
     
    Chooke and Musicisthebest like this.
Thread Status:
Not open for further replies.

Share This Page

molar-endocrine