home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!usc!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!att!allegra!alice!jj
- From: jj@alice.att.com (jj, curmudgeon and really disguted scientist)
- Newsgroups: rec.audio
- Subject: Re: CD SOund QUality
- Message-ID: <24473@alice.att.com>
- Date: 23 Dec 92 03:25:11 GMT
- Article-I.D.: alice.24473
- References: <1h17e4INNrkv@usenet.INS.CWRU.Edu> <24459@alice.att.com> <1992Dec21.213820.2737@cbnewsh.cb.att.com> <24463@alice.att.com> <Jonas.Palm-221292205230@fastpath-37.orgk2.lth.se>
- Reply-To: jj@alice.UUCP (jj, curmudgeon and really disguted scientist)
- Organization: NJ State Home for Bewildered Terminals
- Lines: 58
-
- In article <Jonas.Palm-221292205230@fastpath-37.orgk2.lth.se> Jonas.Palm@orgk3.lu.se (Jonas Palm) writes:
-
- >Could you please make the above statement absolutely clear.
- >Do you mean that the recording engineer should have (say)
- >18 bits at his disposal, for instance in order to be able to
- >maintain a reasonable safety margin when recording, but that
- >16 bits should then be sufficient in the actual CD-medium?
- Well, as some fellow name of Johnston said a few years ago
- (He didn't write a paper for his talk, so there's only the
- tapes, I'm sorry to say) said in a talk on the future
- of digital and compression at an AES meeting
- in Toronto, if we had 24 bits we could
- reference all of our stuff from below threshold of hearing to
- above the threshold of pain, and if we made all our recordings
- that way, we could use the recording as a DOCUMENT as well
- as a recording, i.e. we could go back and reestablish conditions
- given a little simple calibration.
-
- How does this relate to 18 bits? I'm looking at a piano recording
- right now, the largest absolute value any sample on it is about 2400.
- This is somewhere like 4 bits down TO START. Perhaps if we had two
- more bits, I wouldn't have to listen to the miserable quatization
- noise and also have to handle such stuff in encoding/decoding
- algorithms.
-
- I've seen albums with clipping on them, I've seen albums that
- just ignore the 4 MSB's, and I've got one or two that I'd
- SWEAR were only 12 bits to start with.
-
- It's this loss of resolution that I suspect is brought on by
- non-normalized digital transfers in the recording chain that
- annoy me. Two bits would make it disproportionally better,
- if you do some simple math. As far as the recording engineer
- is concerned, that's not what seems to be the problem.
- The problems seem to come later, in what I can only
- describe as bizzare levels present on the final CD, for
- the most part. The few that seem to have 12 bits (MSB used)
- might be recording level problems that somebody tried to fix,
- of course.
-
- (18 bits)
- >in the recording-playback chain would seem to imply this, but
- >you may not agree.
-
- Why would anyone care if I agree? After all, as the article
- I orginally replied to in this thread suggests, I'm only
- a crank audiophile, or so my e-mail tells me.
-
- As another matter, I can argue that in the best listening room
- I've seen, 18 bits would have been nicer than 16 bits if
- somebody had made the perfect digital recording, but of course
- until somebody invents the mike, preamp, mix and such to take care
- of such range, ...
- --
- Extremism *Copyright alice!jj 1992, all rights reserved, except transmission
- in the *by USENET and like facilities granted. Said permission is
- defense of *granted only for complete copies that include this notice.
- liberty is no vice. *Use on pay-for-read services specifically disallowed.
-