Re: subjectivity
Wellfed said:
I will be testing for "different", not better. I am trying to eliminate every possible variable I can think of.
If you're only going to use two identical discs, then you should definitely skip any surface treatments.
CDs use error correction -- A Reed-Solomon code, I think, but it doesn't matter. There are about four times as many bits on the disc than are necessary for the audio encoding. Those redundant bits encode information about the audio bits and the other redundant bits in such a way that even with a few of them missing, you can reconstruct the missing bits This is why a little speck of dust or a minor scratch that's radial to the disc doesn't become a major problem. You lose a bit or two at widely separated places , the error correction code reconstructs the original data, no problem.
However, error correction does have limits. Lose enough bits, and you can no longer perfectly reconstruct the data. So a concentric scratch can take out enough bits close together to cause a real problem.
Now, conceivably (and I'm not claiming that it does, merely that it could), a surface treatment could add a non-uniform optical layer to the disc, diffracting the read laser, and losing bits. If it misses enough bits, the CD player can't perfectly reconstruct the data. I'm sure the CD player manufacturers try to make a missing sample as unobtrusive as possible, but if the data for a sample has been destroyed, then you can only guess at what it might have been, so the output waveform is going to be different.
Would you agree that two identical data streams played on the same audio system would sound the same?
If not, why would the same CD sound the same on two different playings?
If so, then you may be able to eliminate the manufacturing process, yourself and your sound system as variables.
Use CD ripping software and rip the image of a CD to file0.
Rename file0 to file1.
Rip the image of the same CD to file0 again.
Rename file0 to file2.
Compare file1 to file2. If the error correction worked, the files should be identical. If they're not, then this test scheme won't work.
Now apply whatever it is you're testing (surface treatment or enhancement chip) to the same disc. Rip it to file0. Rename it to file3. Compare file3 to file1 and file2.
If it's identical, then the applied process didn't actually affect the samples provided to the D/A converter.
If it's not identical, the applied process managed to destroy enough bits that the error correction failed., so you may indeed hear a difference, and I don't think anyone here will argue that it's not possible. Different data, different sound. No question.
) is that all of this can be done in a enjoyable and non-adversarial manner. I greatly appreciate your attitude.
and also some audiophiles, Im sure. In any case, it is wonderful to see people wanted to make decent tests, instead of just believing that skeptics are simply "blind", or something like that.
