Jump to content

beanbag

  • Posts

    378
  • Joined

  • Last visited

  • Country

    United States

Retained

  • Member Title
    I left

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Asking for DBT's on an internet forum has never silenced anybody ever - rather quite the opposite.
  2. I prefer my speakers to remain magnetized, thanks.
  3. I was in the middle of writing a reply to merrill.
  4. DBT tests are a tool of Science, just like voltmeters are a tool of science. And the statistical analysis and studies behind them are Science with a big S. I already gave two links and a book. Here is another link showing some of the maths and statistics behind this kind of testing: Receiver operating characteristic - Wikipedia, the free encyclopedia The short version is that in discrimination tests, there is going to be a tradeoff between avoiding false positives and false negatives. If you are out to prove or show something exists, you have to use a test with a low false positive rate, such at the "null-result-inducing" blinded tests. You cannot use a rotten useless test like "listening to two different people give their sighted evaluations", hoping that they don't all have the same biases. So both of you, please stop complaining that there is insufficient study or characterization behind these types of tests. For an example, read the recent AES paper by the Meridian people and how they conducted their tests to eek out a positive result. I also suggest that you go ask your negative control questions on avsforums where arnyk hangs out, and has patiently and repeatedly answered these types of questions many many times. You can also ask him about how he is able to do an A/B comparison in one second.
  5. I wouldn't readily agree with the "proven to be worse than CD", but look at the stats: 12723 likes 187 dislikes
  6. beanbag

    The Prot

    Let's exacerbate the drama by teaming up against one person and offering our personal commentary!
  7. I got this joke the first time, in case anybody else is gathering statistics on when somebody is being funny or not.
  8. I think it's called "having high standards of evidence". I noticed that among the "subjectivists" (whoever they are) there are the more mellow ones that go "All I care about is my own experiences; I don't care about your dumb scientific theories la la la..", and then there are the meaner ones that go on personal attacks. Now while I do appreciate the free psychological assessment, I should point out that my life is filled with happiness and roses. esldude is like Mr. Spock - truly a paragon of rationality.
  9. Sensitivity and specificity - Wikipedia, the free encyclopedia en.wikipedia.org/wiki/Discrimination_testing Macmillan, Neil A.; Creelman, C. Douglas (15 September 2004). Detection Theory: A User's Guide. Psychology Press. p. 7. ISBN 978-1-4106-1114-7. Science has all the answers. Have faith in science.
  10. and right after that highlighted sentence he wrote: "Discusing their technical merits and limitations is already much better than the ouright bashing and the my-ears-only atitudes. Why not just keep the thread there..."
  11. Based on my blinded assessment of this situation, I declare scientifically that "A" is the humorless one.
  12. That means instead of having to buy and hold the music yourself, you pay a monthly fee and get it from the internet onto your computer. Millions of songs on demand. Check out Spotify or Qobuz. Spotify has a free version with ads, so check it out. What I mean is spend most of your budget on speakers. It will be a lot of fun to listen to all different kinds when shopping.
  13. Fremer's article can be summarized as follows: Mario is an idiot. Oh yeah? Well _I_ heard it! And so did my friend. Meyer's system sux, LOL. I like records. The filtersssss! They rinnnngggg!
  14. Just a minor correction: the 56% number is the 95% confidence threshold. On average, the users reported correctly about 60% of the time. On some particular "high yield" segments, they got it right about 70% of the time. That also means on some segments they got lower (duh) Combined with the above, this just shows that the longer you listen, the more likely you are going to hit the "high yield" segments, that tells your brain (whether consciously or subconsciously) that things are different. The subconscious part is especially relevant if you don't know quite what you are listening for, but just have a "general impression". The purpose of the short segments is for when you know specifically what to listen for, e.g. the echo or decay of a sound. So the moral of the story is that sometimes shorter segments helps, and sometimes it doesn't. Doesn't matter if your expectation bias goes the wrong way. With sighted listening, there are also subconscious biases, such as fatter cables giving better bass, thin flexible cables have a more refined high end, Gold plated is richer, silver plated is brighter, etc. The moral of the story is sighted comparisons are mostly useless for small differences. Meridian paper shows ringing structure due to FIR linear phase with 459Hz transition band is barely detectable for "normal people" with some training. I bet Peter and Miska can do better, but they are super human, and if you too want to be super human like them, then go ahead and practice. P.S. If you want to ask questions about listening test methodologies, you should ask on Hydrogen Audio, the premier internet forum for objective audibility testing, and support forum for the best free software player, Foobar2000.
  15. Yes, you would think that this would be a site in which people learn about how computers can make their audio better, not all the ways that computers can make their audio worse. In that sense, all the other forums besides the general forum have this kind of useful info, e.g. DSP, room correction, music servers, etc. Yes, that would be me. I would like measurements to be all and end all so we can stop having this drama about DBT and he said she said.
×
×
  • Create New...