[vorbis] about vorbis encoder

[Leena]

>I think (I haven't checked, this is from distant memory) the PACKETBLOBS
value
>is used for bitrate management - the vorbis encoder will do partial encodes
>of the data multiple times, then select between them.

What is meant by partial encode .? In bi rate manegmnet I have found that
they
keep on adjusting noise offset depending on bit arte ( less noise  high bit
rate )
So data is encoded multiple times is fine . But what is the idea behind
partial encode .

>I'm not sure quite what you're asking here.

>Is it:
>a) How are the floor points calculated (i.e. how do we fit line segments to
>the original data)? This is a pretty straightfoward line-fitting algorithm
>minimising local error in the fit.
or
>b) How is the floor data calculated before the line fit and quantisation?
This
>is very complex - this is where the psychoacoustic model comes in. A simple
>explanation would not suffice.

<p>  Ya , I have both the questions --- how floor points are calculated and how
  they are fitted in line . Here floor 1 is used . I can't map the procedure
given in spect
  with that of code .

  Bye the way , so sight only decoder spects are avialabe . What about
encoder ...?
  NO document is there ..
  By chance do any one has that ... Reading a code without block diagam ,
doc is
  puttig me in more trouble

<p>  Regards ,

  Leena

<p><p><p>-----Original Message-----
From: owner-vorbis at xiph.org [mailto:owner-vorbis at xiph.org]On Behalf Of
Michael Smith
Sent: Wednesday, March 10, 2004 5:34 AM
To: vorbis at xiph.org
Subject: Re: [vorbis] about vorbis encoder

<p>On Tuesday 09 March 2004 22:27, Leena wrote:
> Hello !!
>
>    Can anyone tell me what is this PACKETBLOBS in vorbis encoder .?
> The value of this is 15 . So is that complete packet is divided into 15
> parts .

I think (I haven't checked, this is from distant memory) the PACKETBLOBS
value
is used for bitrate management - the vorbis encoder will do partial encodes
of the data multiple times, then select between them.

>
>   I am not that way clear with function floor1_fit .  Each blob is divided
> into
> posts . ( floor points ) Those points are quntised and fitted into line
fit
> function .
> Can anyone tell me how this line fit function is calculated ..
>

I'm not sure quite what you're asking here.

Is it:
a) How are the floor points calculated (i.e. how do we fit line segments to
the original data)? This is a pretty straightfoward line-fitting algorithm
minimising local error in the fit.
or
b) How is the floor data calculated before the line fit and quantisation?
This
is very complex - this is where the psychoacoustic model comes in. A simple
explanation would not suffice.

Mike

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.