It looks like a worthwhile test.  Some comments:

Would it not be better to use a fairly high end system?  Even high end
headphones, rather than $300 ones?

Secondly, if the ears of the listener are so good that he can note down
the differences, why not ask him to say which an unknown player is? 
This would be an ABX test - you play A and B, to the listener's
satisfaction, then ask him which X is.  If he can get this right, it
would be really strong evidence for a difference.

Thirdly, could you randomise the inputs for each test, i.e. in test 1 A
is the CD, in test 2 B is, and so on.  Then the listener could say which
he thinks is the best.  If he consistently chooses one, above
statistical likelihood, then you have evidence.  This is a lot easier
to run than ABX.

Finally, it's worth getting in someone who knows statistics to help you
to decide how many tests to run.


-- 
adamslim

MPD on Linux PC, Caiman, DIY Kondo M7 pre, 6B4G SET, Lowther monsters
SB3, Charlize 2, DIY open baffle speakers
Boom x 2, Radio
------------------------------------------------------------------------
adamslim's Profile: http://forums.slimdevices.com/member.php?userid=7355
View this thread: http://forums.slimdevices.com/showthread.php?t=88364

_______________________________________________
audiophiles mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/audiophiles

Reply via email to