===================================================
= NB: Over 50% of this message is QUOTED, please =
= be more selective when quoting text =
===================================================
> Rick Pali wrote:
>
> From: Simon Barnes
>
> > The original CD is coded 16 bit, so any bits added by
> > the player have been made up by guessology.
>
> The *solution* was to run a higher bitlevel so that the inaccurate least
> significant bits were stripped before output. That is, an 18 bit DAC was
> used and the 'top' 16 bits did all the work. Most of the errors occurred
> in
> the 'bottom' two bits which were not used in the audible output anyway so
> it
> was a huge boost in the accuracy of the output signal.
>
> For all intents and purposes, it solved the problem of low-level linearity
> errors in the 16 bit output.
>
I have no problem with this, but don't find it relevant. The CD only
contains 16 bit data. When this is transferred to the MD input, it is not
converted back to analog, so D/A linearity is not a factor. The ATRAC
algorithm will apply all sorts of hoopla to the digital data stream, and it
is conceiveable that the result might be WORSE if you tell it to use all 18
bits of input data if the bottom 2 bits are just the result of some
guesswork by the player.
simon
-----------------------------------------------------------------
To stop getting this list send a message containing just the word
"unsubscribe" to [EMAIL PROTECTED]
