Quote:
I don't think i may able to notice a difference between 2 identical FLAC, if its using the same system and source. If you burnt it into CD-R and playing one from there that's another case. I would like to know what audio system you're using and will it differ or not..
Anyway, like I said, the difference its obvious like you're listening 320kbps MP3 compare to CD quality music. furthermore, I usually blindfold my friend and they prefer CD-R than harddisk too...
These results are based on non-scientific testing (I know you said "blindfold", but that's not sufficient). They technically are subjective opinion, and may be the result of influence. By the way, I never consider things like this "invalid" for the individual, just not scientifically and repeatably valid in general. That means, if you hear something that makes you happy, go with it.
Quote:
Okie... hmmm... I don't know what's your experience in CD audio... when listening to CD, you guys ever notice that playing the same track again and again makes it sound better? and usually it will maintain the same from the third time onwards. Now don't tell me its a mind thing, many of my known audiophile knows about this.
Now, this may sound personal, I don't mean it that way. "Now don't tell me its a mind thing," indicates strong bias. Bias strongly affects perception. In order to test perception on it's own, you have to completely eliminate bias. The test for the repeated CD playing effect is complex enough that it will probably never be done, which means this effect will always remain in its current status of unverifiable results.
To test for the change in sound with repeated playings of the same CD you'd have to have a stack of identical, fresh CDs, and a group of test listeners. There would be two identical players, probably the highest-end units possible. The system would be a high resolution audiophile-accepted system in a well treated room. They would be presented with the choices of A, B, and X. In the test, A is a well-played CD. B is a brand new unplayed CD, and X is randomly assigned to be either A or B, unknown to the listener. Their task is to listen, as long as they like, then match X to A or B. Each would have between 16 and 20 trials. Then you'd need about 20 or 30 listeners. Their scores are then compiled, and analyzed. The degree to which the average score deviates from 50% (guessing) would indicate the degree of audibility of the effect. The reason for all the trials and all the listeners is to improve the test resolution around the 50% point to be able to discern a real trend from guessing. It may also be good to swap which player has the fresh vs well-played CD half way through to spot a player-specific bias.
Now the point is, not that the test would or could be done, but that any other test methodology will have its results skewed by bias.
When someone presents you with a listening evaluation test of any kind, but tells you in advance what to expect to hear, you can stop right there. The results are biased by expectations. And, the person introducing the bias could be yourself, expecting the difference and fulfilling that expectation. This is the same problem you're experiencing with your CD vs Lossless comparison. The primary issue is pre-existing bias, and secondarily a somewhat different audio path. If you could eliminate those two issues, then and only then will you be making a true comparison between CD and FLAC. But, to actually prove the difference, it has to be a double-blind test with 20 trials, and hopefully, more than one listener.
How about an intermediate step: make some audio measurements on each source all the way through to the output to your headphones or speakers? FR and distortion measurements might just tell you something.
By the way, your susceptibility to bias doesn't in any way denigrate you. It means you're human, and all humans are susceptible to bias, often especially the ones who believe they aren't, which is in itself a bias.
