Re: [SurroundSound] Re: Bit Rate Resolution, Sampling Rate, Upsample, and Lossless vs. Lossy ....

> I singled out redbook as it is the most common source for digital
> conversion to M:P3/AC3. Indeed the article does state the only noticible
> sound degradation noticed by the subject group was the 128k compression.
>
> It is also true the job of most compression codecs is to preserve as much
> of the original source as possible while reducing file and data size. The
> argument is that in that process detail and transparency is lost. I
> believe
> this is true starting with the highest transparent and detailed source
> which could be anything from a high data uncompressed digital file to an
> analog studio tape master.

Certainly data are lost in lossy compression, hence the name; the argument
is under what conditions this becomes audibly noticeable and
objectionable.

(Data --- 'details' -- are lost in all forms of audio recording too,
including analog taping. That's just the laws of physics at work. Yet
many pleasant recordings exist. So the fact of data loss by itself is not
an indicator of bad sound.)


> On the other hand, starting with a 44.1/16 data source already compressed
> and already having lost a fair amount of detail and transparency (redbook
> CD standard)

And here we go. You state as fact what has not been demonstrated, and i
know where this is goign to go 'well, I hear it'. So maybe best to stop
right here.

However. Redbook audio is not ;'compressed' by the common usage or meaning
of the word -- unless you accede that all recording methods involve
'compression' (data loss) compared to the live acoustic event. But no one
calls that 'compression' because it's so universally true. 'Compression'
is reserved for intentional file size reduction.

Also of course Redbook and high rez and analog recordings can have their
dynamic range reduced (the other meaning of 'compressed').


> by compressing it more that will result in very little
> audiable change to most listeners except the most discriminating who have
> heard the source material in at it's best whether that is vinyl or source
> analog tape or digital high sample rate audio files. Those original
> sources would show significant audiable loss identifiable to even the
> casual subject listener at 320k, 128k, and 160k I theorize.

The accepted wisdom among lossy codec developers and mavens (e.g.
Hydrogenaudio) is that if you take an mp3 (something that is already
'compressed' by the standard definition of lossy data compression) you run
a much greater chance of it sounding worse if you compress it again, than
if you start with a CD (or a hi rez file). (This is why mp3 is not
recommended for archiving audio -- you are pretty much limited yourself
from ever using another lossy compression method on those files)

By your theory, one should find the opposite. The further degradation to
a lossy source should be 'very little audible change', whereas the
lossless source should take a big hit.



>

--
You received this message because you are subscribed to the Google Groups "SurroundSound" group.
To post to this group, send email to SurroundSound@googlegroups.com
To unsubscribe from this group, send email to SurroundSound-unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/SurroundSound