In Reply to: changed the subject line..finally posted by jneutron on April 23, 2004 at 13:10:51:
I am not sure that what you are saying is the correct way of looking at it.Nordmark's experiment was performed as a first generation real-time event. It did not involve a playback event, which would include jitter and dither from the A to D process, as well as jitter and dither from the D to A playback process.
Thus, in your example, you can not just "flip a switch" during playback, the jitter would have to be applied during the recording process in order for the proper information to be recorded. Applying the jitter during playback only will not encode the same information that was present at Normark's first generation real-time event.
This would also mean that the playback event would have to avoid ADDING any jitter that could upset the encoded information.
Currently, CDPs can have as much as 10-20 nS RMS of jitter if they are not doing so well in this regard, a good machine might get as low as several hundred pS RMS. The peak levels can of course, be quite a bit higher.
CDPs with the higher levels are usually considered to sound worse than those with the lower levels, having less image depth, a narrower soundstage, and less precise instrumental location capability.
(The above is the result of many years of experience with listening to CDPs, and reading all the Stereophile reviews of CDPs that I and others have ever listened too).I am also getting the impression that properly applied dither during the A to D process, even though it is an AMPLITUDE process, would tend to provide the zero crossing information that Nordmark was theorizing.
Jon Risch
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: changed the subject line..finally - Jon Risch 22:38:46 04/23/04 (1)
- Re: changed the subject line..finally - jneutron 06:25:22 04/26/04 (0)