 |
DTS v. Lexicon question
Posted by Heinz Seltmann on October 01, 1998 at 15:48:42:
In Reply to: DTS v. Lexicon question posted by SMR [WizOp] on October 01, 1998 at 14:13:52:
>
> : DTS like AC-3 does have flags in its bitstream. Otherwise you
>would never be able to decode and detect what you have.
>
> I don't believe DTS does have bitstream flags which identify it
>as DTS, and yes, it would be possible to decode the format without them.
>DTS decoders contain a model which enables them to recognise and resolve
>the syntax, the incoming data is then compared with is model and resolved
I guess its a matter of what we are calling a valid DTS bitstream.....
When I think of a DTS/AC-3 bitstream I am thinking of the digital data
once it has been extracted from the IEC958 format. IEC958 is simply
an encoding and transfer format, but you know this....
The extracted DTS bitstream then has a certain format which include flags,
CRC, etc to it which would correspond to the "model" that you are refering
to.
I guess some boring background would be useful. CD's and PCM LD info is
transfered using IEC958/AES3-1992. This was really only set up with
digital audio in mind. Compressed/non PCM formats were not really
thought of at this point. But AES did leave in a valid audio flag which
could be used to signal that non-PCM data is present.
Then came IEC1398(or IEC1938 I can never remember). This format
says that non-PCM data can be tranfered over IEC958 when the audio
valid flag is invalid. IEC1398 has a set of flags which indicates
which bitstream is encoded. Which is where we get the idea of flags
in the bitstream. But the problem is that CD and PCM LD tech. never
had non-PCM audio in mind when the std. came out.....
>as DTS - the reason we hear a "fzztt" - that occurs when the decoder is
>recognising the data.
This is again true......but the reason that we are hearing a "fzztt" is
because DTS encodes CD's and LD's using what they call pseudo-PCM. And
the reason that its pseudo-PCM is that it has a valid audio flag set
when it comes out of the CD player and LD, per the IEC958 or AES3-1992.
This tells/confuses a processor that the audio stream is valid and
playable PCM audio data. The processor is then responcible for deciding
to wait to try to decode DTS or just drive it out to the DAC's.
So ultimatly the reason for the trouble is that DTS wanted to be able to
use older equipment which assumes that anything on a CD/LD is valid audio.
It would not be difficult to encode AC-3 on a CD and play it. But you
would find that an AC-3 encoded CD would behave the same way as a DTS
encoded CD. Notice that on a LD you have a usually seperate box which
does the Rf to IEC958 encoding. I bet that this box actually encodes the
AC-3 using IEC1398(? or is it 1938, non-audio over iec958). Which sets
the audio valid flag as false which tells a processor that it should not
play this data as PCM but should instead check for encoded data.Heinz
Follow Ups:
Return to the new SMR Forums Menu
Design
& HTML © SMR Home Theatre, Images © SMR Home Theatre cannot be reproduced without
permission. The images on this page are digitally watermarked. New forum
messages should be posted into SMR Forums v2 - http://www.smr-forums.com/
|
|
