jeffmeh;373372 Wrote: > I would suggest that "if they don't find any," then it shows that the > differences are not audible for that sample group. "If they do find > some," then it shows that the differences are audible. "If no one has > ever found any," then it is very likely, though not certain, that the > differences are not audible. My opinion is that if a group (or several groups) can't find the difference, the following thing happens :
Hypothesis 1: The difference is audible The difference must affect at least an equal statistical percentage of any possible audio sound, as the one we need to make the ABX test "valid". I let you calculate how much this must be to make the test valid. ;) In other words, we actually don't measure the confidence of the "difference is audible", we measure the above percentage, if and only if this percentage is greater than the confidence level. Hypothesis 2: The difference is not audible In this case the ABX test measures nothing at all. So, in any case, ABX tests don't test what they are used to. The only possible use of an ABX test would be to validate an existing specifically audible difference (let's say if someone hears an extra "pop" at a specific moment, that we know it is there). But, again, this is not the way it is used. -- Themis SB3 - North Star dac 192 - Denon 3808 - Sonus Faber Grand Piano Domus ------------------------------------------------------------------------ Themis's Profile: http://forums.slimdevices.com/member.php?userid=14700 View this thread: http://forums.slimdevices.com/showthread.php?t=56425 _______________________________________________ audiophiles mailing list [email protected] http://lists.slimdevices.com/lists/listinfo/audiophiles
