Re: What's the use of a 192 kHz sample rate?

On Fri, 16 May 2008 16:29:18 -0400, "Arny Krueger" <arnyk@[EMAIL PROTECTED]
>
wrote:

>"rajesh" <getraje****n@[EMAIL PROTECTED]
> wrote in message
>news:819396c9-ece7-4bf4-94bf-398c07bddeaf@[EMAIL PROTECTED]
>
>> What ECC do they use on DVD's.
>
>The short answer is lots. Enough so that DVD playback, whether data,
audio 
>or video,  is reliably and repeatably perfect.  The perfect playback
allows 
>audio and video data to be enciphered. If data is enciphered, data errors
in 
>the media are usually highly destructive to the integrity of user data.

What do you mean by "perfect"?   Do you mean completely error-free? Or
do you mean below some accepted threshold error rate?

If you mean completely error-free then I think that doesn't agree with
the engineering behind the system.  There are no error-free storage
techniques or transmission techniques.  The media, even in pristine
condition, has a read error probability greater than zero, and the ECC
has, like all ECCs, limited correction capability.   Analytically that
means that the output error rate is not zero.  It's good enough,
though, and that's all one really needs.

To answer rajesh's question properly, it's a Reed-Solomon Product
Code.   The dimensions are, as near as I've been able to sort out,
(182, 172) for the rows, which is the inner code, and (208,192) for
the columns, which is the outer code.  That's an overall rate of
33024/37856, so that R = 0.8724 (there is parity on the parity symbols
for the overall code rate).

That's a fairly high code rate (less than 13% redundancy), although
not unusual for storage media.   In other words, it's not "lots",
especially compared to low-code-rate codes used in comm systems.  But
since it's a product code it allows interesting tricks in the decoder,
like iteration, that can help reduce the error rate in trade for
higher complexity, latency, and power consumption, should the vendor
decide to employ them.

It's still not going to be error-free on the output.  

>> What is the redundancy introduced.(bits/ sample) by ECC in DVD's.
>
>Enough so that DVDs are generally perfect  reproducers of user data,
whether 
>audio, video or data.

See above.

>> How does it  compare with that on CD's.
>
>A major difference between CDs and DVDs is that at the user data level
there 
>are two kinds of CDs - audio CDs and data CDs. At the user data level
there 
>is only one kind of DVD.
>
>Audio CDs hold more user data bits and have a reduced level of
redundancy, 
>error checking, and correction. Data CDs have enough redundancy, error 
>checking and correction such that they reliably and repeatably provide 
>absolutely perfect replicas of the user data.
>
>Audio  CDs may have a few undetected/uncorrected errors, which is why
error 
>concealment is part of their operation. In practice  audio CD playback
may 
>be very close to perfect as there still is considerable redundancy, error

>checking and correction in place. The benefit of the reduced redundancy, 
>error checking and correction of audio CDs is that they hold more user
data.
>
>DVDs are different in that there is no distinction in terms of error 
>checking and correction between audio, video and data DVDs.  Audio and
video 
>DVDs are essentially data DVDs with a predetermined file structure.
>
Eric Jacobsen
Minister of Algorithms
Abineau Communications
http://www.ericjacobsen.org

Blog: http://www.dsprelated.com/blogs-1/hf/Eric_Jacobsen.php