Themis;373413 Wrote: > I understand the tests, don't worry. ;) You say so, but then you say...
> But the way the tests are made is only valid if (and only if) the > differences are equally distributed along the examined samples. which is completely wrong. With sufficient statistics, these tests will find -any- ability to discern differences, no matter how small or uneven, even if the experimenter is too rigid or dumb to adapt the test to focus on the interesting cases. For example even if 9 out of 10 pairs of samples were absolutely identical, so that only the 10th pair had audible differences, a test which equally treated all 10 pairs would -still- find a statistically significant difference if you took enough data. In a more realistic case, where the differences were easier to hear in some samples and harder in others, the test would do just fine. Moreover the experimenter has the discretion to "zoom in" on segments and/or listeners that look promising, so if there was one "good" sample it would be trivial to focus on it and exclude the rest. The facts are that these tests are skewed very much in favor of the listener - they will detect even a very poor ability to hear something. And yet, time after time after time, audiophiles that claim to be able to hear "night and day" differences in (say) cables fail miserably when blind. It's -totally- obvious (based on countless such experiments and the last century of science on hearing and perception and psychology) to any sensible person why this is, and yet audiophiles continue to ignore the elephant in the room. -- opaqueice ------------------------------------------------------------------------ opaqueice's Profile: http://forums.slimdevices.com/member.php?userid=4234 View this thread: http://forums.slimdevices.com/showthread.php?t=56425 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/lists/listinfo/audiophiles