Quote Originally Posted by RGA
You don't get more dynamic range with cd - what you get is cd manufacturers who changed the definition of dynamic range to suit a marketing campaign. CD's have Quantizing noise "an artifact of the analog to digital coversion process. If all were perfect that noise would be down where the last binary digit is. The noise figure would then be expressed by the formula:

20 log(2b-1) where b is the number of system bits. Most modern systems use 16 bits (but throw one bit away on the parity check), and so:

20 log(215 - 1) = 90.3db

Now a dynamic range over 90db is nough to make a recording engineer drool, but don't drool yet. That figure relates to the peak-to-peak value of the audio signal, rather than the usual root mean square voltage value. To convert you subtract the following from the noise value:

20 log (2 X Square root of 2) = 9.03db.

As you'll notice our dynamic range is now down to 81db. And you can't record at that level because the digital "ceiling" is far harder and more awful than that of analogdisc or tape. It would be a good idea to knock 8db off that figure. Total usable dynamic range : 73db even under ideal conditions. This isn't earth-shaking. A good analog recorder (12.5cm stereo, 76cm/sec) can boast dynamic range of some 74dB. Add dolby or dbx and ther's no comparison. Incidentally, all these figures refer to unweighted noise readings, treating noises of all frequencies equally. Weighting curves are often used by both sides to make the specs look prettier." (Paul Bergman).
Paul Bergman clearly didn't know what he was talking about. My advice is not to get your technical information from that rag, UHF.

Aside from all the other crap, he has evidently not heard of dither. With a little dither, CD players can have good linearity down to below -110 dB, which should be impossible, according to Bergman. If you had read some good reviews, you'd know that. Look at the low level linearity measurements for different players. Here's a link to a review of an old Radio Shack portable CD player in Stereophile in 1994. Check out the low level linearity graph in Figure 5, which shows its linearity was pretty good even a -100 dB (and this player is nothing special).

http://stereophile.com/digitalsource...0/index12.html

Others have reasonable linearity down to below -110 dB, as with this SimAudio product (Stereophile doesn't seem to review many reasonably priced CDPs!):

http://stereophile.com/digitalsource...io/index4.html

So much for the myth that CDs don't have more dynamic range than analog tapes!!