Hi,
I would agree that you can accummulate sessions.
The ABX/DBT as proof of no audible difference crowd would disagree. I wonder why?
I've done some playing with PCABX, 16 is tedious. It is however interesting to listen to level changes, bandwidth differences, distortion etc.
Rest assured, 8 trials is tedious. I did this as part of a challenge on a on-line (german) discussion board.
I provided a set of three eac ripped wav files, which the challenger burned to CDR as originals and MP3 reconverted versions and snail mailed to me (8 trials in total per track).
When I declared that NON of the files matched my original reference (recordings) CD's but ranked them in order of preference the challenger unilaterally declared a victory.
(as in "he [that is me] cannot tell the original")
I had tested using a Pioneer DV-505 DVD Player (Audiocom Superclocked) into a Philips LHH-1000 DAC (read Marantz DA-12 - the DAC part of the CD-12 Combo) fully restored and rebuild, using latest high quality components and op-amps but not changing the fundamental circuit and design.
I subsequently tested said CD on a CD-Player capable of HDCD decoding (I should note that all tracks where from HDCD encoded CD's and that HDCD encoding survives neither Sample Rate Conversion nor MP3 and back conversion) and found several interesting things:
1) Tracks that "the challenger" identified in his posts as original, which I had rated lowest in sound quality failed to light the HDCD Light, thus confirming digital manipulation.
2) Tracks I had rated "closest to my CD's" did turn on the HDCD Light. The "challenger" had insisted these where actually low rate MP3's converted back.
So, this DBT was first of all one I actually passed (I did rate the bit identical tracks as best in sound quality) and secondly, it illustrated (to me) that even a CD Player that reads the CD asyncronously (with a massive ram buffer - in fact that particular DVD Player treated the CD as DVD and was working in DATA MODE extraction into RAM) seems to show, for some reason a reliable difference between an original pressing and a CDR.
Past that, it illustrated to me that "debunkers" are, if called upon, quite willing to use outright fraud to suppress evidence they find inconvenient.
In the case of the particular discussion board the "challenger" was also moderator, resulting in me being banned, after having called his bluff and further having all posts in which I exposed the fraud deleted, while the earlier discussions where retained.
Talk about "chinese democracy".
Anyway, bottom line, good testing limits "trials" to five or six (six allows easier balancing of "different/same" conditions as each session can be balanced) per session, tests the actual test setup including listeners for sensitivity to known audible phenomenae before testing ones being disputed and finally, instead of a sensationlist "everything sounds the same" conclusion presents findings as a confidence interval, as in "we can be sure to the following degree we have not missed things and we can be sure to the following degree that we found is real" and not as an abstract pass/fail test (which quite frankly is the kind of statistics only Tony Blair's, Gordon Brown's, Maggie Thatcher's, George Shrub's (Jun & Sen) and Ronald Reagan's Governments would dare to employ to deceive the gullible and which the chinese communist party would never bother with....).
Ciao T
.
