Hi Tim,
That's a really interesting observation. Any guess as to why that happens?
Tim
Guesses are about all I have. Most of the plants I've spoken with tell me their discs sound exactly like the masters. All I can conclude is they're either not listening or simply not hearing it.
Interestingly to me, when I was speaking with different plants prior to selecting one to make Soundkeeper Recordings CDs, all but one told me the same thing. Only one told me "discs will never sound like the masters". It turns out, they got the job.
The discs that sound most like the masters seem to have certain things in common, though I'm not sure exactly how much each factor comes into play. First, the glass masters are cut in real time (instead of the more common 4x or faster). I know different LBRs (the devices used to "cut" the glass master from the CD master) create different results.
Second, the best discs seem to use a longer injection molding cycle (~9 seconds vs. the more typical ~4 seconds). Some may say this makes for more accurately formed pits in the disc and diminishes jitter. Perhaps it does. I don't know.
Not too long ago, I auditioned some test disc pairs, one was an SHM vs. plain CD, another was an HQCD vs. plain CD. I heard differences that sounded to me like different EQ, so radical I was pretty sure different masters were used for each and that the test was therefore misleading (at best). Then I extracted the tracks from both discs to computer and performed a few null tests.
For those who may not be familiar with the idea, in a null test, two files are placed in a multitrack program and syncrhonized (to the sample level). Then, the polarity of one file is inverted and the two mixed together. The result is that everything the files have in common is cancelled (+1 added to -1 = 0). Only what is different, if anything, remains.
In the tests I did, the result was dead silence, all the way down, to the sample. To me, this proved both discs in the set used the same master; the *data* was 100% identical.
And from the hard disk, both sounded identical. Yet from the player, feeding the same DAC (a Metric Halo ULN-8), the piano, trumpet and cymbals sounded so different, I would have bet they were from different masters.
Definitive answers will be hard to come by while so many insist that identical data will produce identical sound. The assumption seems to discount that what is on the discs isn't "ones and zeros" as so many believe. The disc player must track the spiral of pits, focus the laser, read the data, decode the 8:14 modulation (which writes what are effectively nine different length pits - visible as sine waves on an oscilloscope), do redundant reads when necessary (information is written more than once on a disc), perform error correction and decode the binary information into analog, all in real time (and often with a common power supply for laser tracking, audio, etc.). In extraction to a computer, real time isn't a factor; the drive can read as many times and for as long as it has to in order to get the correct read. The read and the listening aren't occurring at the same time. Perhaps this has something to do with it.
With the plant we use for Soundkeeper Recordings CDs, the discs are very close to the master - so close, I need a direct, synchronized A/B to hear the differences. But the differences are still there, manifesting as a subtle loss of fine detail and "focus" when compared directly against the master used to create the pressings.
Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com
. OTOH, I think one has to give Frank credit for believing in himself, even if many of us are shall we say, a "little skeptical"
