Using rapid A/B comparisons I get some info, but not a complete data set.
With the theory that difference is not always better (or worse) I find I obtain more useful info by putting a given cable into the system and living with it for a while (days). Then take it out and put in the comparator and listen over a couple of days. I keep doing this until the one I want to live with emerges. If you feel a need to do this blind get a buddy or a kindly spouse to help. (Just don't peek!)
Sometimes a difference is initially disconcerting; sometimes the unfamilar is initally attractive. Long term satisfaction is what you really want.
On the other hand, to the extent you have a 'gold standard' as a comparitor, rapid testing can help find DUTs that are simliar in sound. Otherwise your comparisons lack a control. John's method (using cable 'c') does account for this. I also agree with Sean... any change you introduce, other than the variable you want to test, may confound your results.
So I keep it simple. Out with one, in with the other. Reverse, repeat. Works well for me.
With the theory that difference is not always better (or worse) I find I obtain more useful info by putting a given cable into the system and living with it for a while (days). Then take it out and put in the comparator and listen over a couple of days. I keep doing this until the one I want to live with emerges. If you feel a need to do this blind get a buddy or a kindly spouse to help. (Just don't peek!)
Sometimes a difference is initially disconcerting; sometimes the unfamilar is initally attractive. Long term satisfaction is what you really want.
On the other hand, to the extent you have a 'gold standard' as a comparitor, rapid testing can help find DUTs that are simliar in sound. Otherwise your comparisons lack a control. John's method (using cable 'c') does account for this. I also agree with Sean... any change you introduce, other than the variable you want to test, may confound your results.
So I keep it simple. Out with one, in with the other. Reverse, repeat. Works well for me.