|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
In Reply to: Re: I disagree, but will AB anyway to satisfy my curiosity.. posted by RT on May 17, 1999 at 07:24:16:
Here is a simple initial test before you go get duplicates. A test disc is needed though, with either pink noise or a few frequency bands like 1khz, 100hz and perhaps a 10khz tone, if no pink noise is available. You also need an spl meter,RS has an analog meter, although this is not accurate enough to check the level of the two player to match, it will give you an indication of gross level differences. A tripod would also help to hold the meter stady and ina constant place. Play the test disk with bot players and check the level at the listening place. In a real test, one would use a volt meter to check at the speakers to 1% of voltage with both player. That gives the desired .1db level match but that is beyond the call of duty for your curiosity.
The other problem you will have with multiple discs is how will you synchronize the two so you cannot get a clue which disc is which by one being a bit ahead/behind the other, if you switch rapidly.
With this initial test of level, if they match, do a ten trial yesy on your wife, randomly played in the two players, without her knowledge which is which, and let her write down which player she thinks is playing. Correct the answer sheet after the ten trials, not after each trial. You need 9 correct answers.
Actually, statistically speaking, isn't anything more than 7/10 statistically significant? At 7/10 your confidence interval is > 90%, right?Or has it just been too long since I took statistics?
In the wast majority of stat testing, most use the 95% confidence level, not the 90%. So, the 9 of 10 is correct for the 95% level. Check out this link for a better explanation of the binomial stuff with some tables:
http://mars.acs.oakland.edu/~djcarlst/abx_p9.htm
I hope this will get you to the link. It does on myt hyperlink.
This post is made possible by the generous support of people like you and our sponsors: