🚀 go-pugleaf

In article <1992Dec31.131316.14084@lsican.uucp>, michael@Canada.lsil.com
(Michael Smith) wrote:
> 
>I would tend to agree with you that a properly buffered data-stream (i.e. a FIFO
>of sort) should be able to compensate for any drifts in the data output rate of
>the pickup itself (due to drifts in the motor speed).  If this buffer is not
>deep enough (this would be a function of the expected drifts in the motor rate,
>number of errors per sample, time to correct errors, etc...) then at that point
>there would be a definite error in the output stage of the CD.  This would only
>be a problem if the buffering was not matched with the physical characteristics
>of the CD - i.e. nobody bothered to find out how much variation the drive motor
>would have.
> 
>On the Pioneer 701 LD player, there is a built-in Time Base Corrector to compensate
>for changes in the drive speed motor (no error correcting here!).  Since its an
>analogue medium, buffering becomes extremely difficult and expensive.  In this case,
>they do recognize that there is a problem with maintaining a 100% accurate
>spin rate, and compensate using a TBC.  (But aren't TBC's a digital buffer that
>offsets the output by one frame?)
> 

The TBC attacks speed variations with servo feedback and line-by-line
timing adjustments -- that is, buffered line data is held and released on a
precise schedule. Drift within a line is not compensated for, as far as I
know, and the TBC is only applied to the video portion of the LD signal.
Thus, extra drive stability can help video quality beyond what the TBC
corrects. But the claim was made about audio quality, and that is where I
disagree.

The digital subcarrier of a LD analog signal, once separated and
demodulated, provides binary data with a slight jitter, just like a CD. The
same buffering and clocked-output techniques apply, leaving no more jitter
than a quartz crystal produces.

-Kerry
-
-
-
-