michaelk@azure.UUCP (Michael Kersenbrock) (10/16/84)
<> This talk of controlled compression/expansion of CDs is interesting. I assume it is to "correct" a golden-ear detectable flaw in CD recording. I therefore assume that compression/expansion is undetectable by golden ears. Is that correct? Or is the solution worse than the problem being fixed? Would we THEN have to do something about "Those crummy CD's that expand/compress the WRONG way (no matter what way,if at all)"? I'd like to hear a golden-ear on this. . . I'm not one, I LIKE my plain old Sony CDP-101 CD player. I must be crude. Mike Kersenbrock Tektronix Microcomputer Development Products Aloha, Oregon (P.S.- I even think the "Fresh Aire II" and "Oliver" CD's are very enjoyable (I just bought them recently)). How gauche!
shauns@vice.UUCP (Shaun Simpkins) (10/17/84)
Controlled compression of CDs was proposed as a means of getting 20-bit
effective dynamic range out of CDs without building a 20-bit digitizer, which
is very difficult to build. I stated that CDs have a noise floor that is
below the average listening room's but above hearing threshold. By increasing
the dynamic range of the system we improve the reproduction of near-threshold
level signals. IF the ear can discern 8 or 10 bit quantization at such low
levels (which I don't think it can) compansion will shove the signal level up
to a point where the resolution is higher. The use of a control track
for gain riding eliminates the envelope tracking error of present compansion
systems. I agree that preceeding a digital recorder with a dbx unit is
pointless from a distortion point of view.
I think the real issue in this conversation was not `is the CD inferior' which
I think it isn't for home use, but `what should be the professional quality
level', over which there is some reasonable concern. I am concerned over
the blanket approval of digital techniques as a panacea to the problems of
multi-track mixdowns. It isn't. Let's take an example - we have two 16bit
digital words. The LSB of both fluctuates randomly (i.e., the noise floor
of the digitizer is equal to 1 LSB). We now add them together. Oops! The
2nd LSB is dithering, too. The noise floor has moved up from 1/2 LSB average
to 1 LSB average. Do this a few times and you've lost a lot of your
resolution to noise.
My point? If recordists are not to sacrifice their beloved multitrack
techniques for minimalist recording methods (I doubt that they will) the
recorders they use must be better than our players. This means more bits or
a noise floor of the recording system before the sampler of much, much less
than the digitizer's resolution. Compansion is a way of achieving this
with present systems.
The wandering squash,
--
Shaun Simpkins
uucp: {ucbvax,decvax,chico,pur-ee,cbosg,ihnss}!teklabs!tekcad!vice!shauns
CSnet: shauns@tek
ARPAnet:shauns.tek@rand-relayherbie@watdcsu.UUCP (Herb Chong, Computing Services) (10/19/84)
I guess somewhere along the way, I never mentioned that the compression
system was being researched for professional digital recorders. CD's
have, as pointed out by many, more than enough dynamic range for home use.
However, transfering the master to the CD deals with signal levels that
are known and can be previewed. Professional systems need more dynamic
range than a home system to make sure that saturation of a digital system
NEVER occurs during a recording session. It also guarantees that any
added noise, mixing tracks in the presence of a dither signal, does not
lead to audible noise in the final product.
Herb...
I'm user-friendly -- I don't byte, I nybble....
UUCP: {decvax|utzoo|ihnp4|allegra|clyde}!watmath!watdcsu!herbie
CSNET: herbie%watdcsu@waterloo.csnet
ARPA: herbie%watdcsu%waterloo.csnet@csnet-relay.arpa
BITNET: herbie at watdcs,herbie at watdcsufudd@fluke.UUCP (Mark Freeman) (10/23/84)
--------- Shaun Simpkins' analysis of the noise problem is not quite right: > Let's take an example - we have two 16bit > digital words. The LSB of both fluctuates randomly (i.e., the noise floor > of the digitizer is equal to 1 LSB). We now add them together. Oops! The > 2nd LSB is dithering, too. The noise floor has moved up from 1/2 LSB average > to 1 LSB average. Do this a few times and you've lost a lot of your > resolution to noise. If the dither noise is uncorrelated, then the addition of the noise is an RMS process, so the noise floor does not grow quite so fast. Also, the signal may have to be scaled down after mixing, to keep the signal from limiting, sending the noise floor back down. So, how many bits do studio recorders need? Mark Freeman John Fluke Manufacturing Co., Inc. {allegra,lbl-csam,microsoft,ssc-vax,sun,telematic,teltone,uw-beaver, uw-vlsi}!fluke!fudd