Dan Dugan wrote:
>Walter Knapp, you wrote,
>
> >What you are also saying is that a DAT recorded original transferred to
> >CD is changed by the process. I would agree, everything we do changes
> >the signal some. Even simple digital transfers are not 100% accurate.
> >That's why there are error correction routines in the software.
>
>Not so. Simple digital transfers (like CD or DAT into DAW, or
>vice-versa) are normally perfect, that is, bit-for-bit identical
>files are produced.
The output signal from a DAW is usually dithered on conversion to
16-bit with a digital transfer to CD or DAT or MD, and can also be
dithered to, for example, 24-bit for DVD where the source is greater
than 24-bit. This reduction of the wordlength means that the least
significant bit is in fact changed. Just what implications this
(among all the other processing tasks that may occur within the DAW)
has for an audiofile that is then used to supposedly evaluate the
effect of data compression at point of recording is the question I
was raising in my previous post. If processing causes a change in a
file, how do we know what effects can be ascribed to data
compression? If these variables are omitted, by only comparing raw
recordings with and without data compression, we might have some hope
of enlightenment.
>When a DAT tape or CD is being played, it's
>normal for error correction to reconstruct data that can't be read.
>But that correction results in just that--correction of the read
>errors, and the output of the player is the original data unless
>gross errors bring the process up past the threshold of correction
>into the region of error concealment.
>
>and further on,
>
> >I realize there is this belief that something must have changed and if
>...
>An excellent, fresh description of the encoding process. One must
>also consider how much damage analog recording does to an audio
>signal!
One can consider that if one wishes, but this is another subject, and
is not relevent to examination of ATRAC. One may claim that analogue
(shall we say 30ips with Dolby SR on one-inch tape, perhaps?) is a
better alternative than MiniDisc. Yes/no/maybe - who cares? Well,
it seems from the discussion here that some are using analogue
formats because of various constraints, and they may care. But can we
focus on the central issue, which is to evaluate the effects of data
compression. The only relevent comparisons are between compressed and
uncompressed digital data.
John Campbell
________________________________________________________________________
________________________________________________________________________
|