Vorbis dev - Sep 2002 - More errors in the file format specification Was: Test files for decoder implementation

Hi, 

I've found some more errors in the file format specification (or 
at least points, where the specification and actual libvorbis implementation 
mismatch):

- Floor 1 / curve computation / step 1: amplitude value synthesis

21) vector [floor1_final_Y] element [i] = 
  [predicted] - (([val] - 1) divided by  2 using integer division)

hould be:

21) vector [floor1_final_Y] element [i] = 
  [predicted] - (([val] + 1) divided by  2 using integer division)

<p>- in the function render_line

11) [ady] = [ady] - [base] * [adx]

hould be:

11) [ady] = [ady] - absolute value of [base] * [adx]

<p>Also, the parameter order of render_point(x0,x1,y0,y1,X) and
render_line(x0,
y0, x1, y1, v) is not consistent, and the references to the render_line 
function use different ordering:

- Floor 1 / curve computation / step 2: curve synthesis

8) render_line( [lx], [hx], [ly], [hy], [floor] )
12) render_line( [hx], [hy], [n], [hy], [floor] )

<p><p>Tor

<p><p>==================================================================EASY
and FREE access to your email anywhere: http://Mailreader.com/
==================================================================
<p>--- >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-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.

And right away a few more:

When filling the vector [floor1_final_Y] in floor 1 / amplitude value 
synthesis, libvorbis is just replacing the values in the vector [floor1_Y] 
instead of filling a new vector. If there is a reason for allocating 
and using a new vector instead of just modifying [floor1_Y] the step

5) [predicted] = render_point( 
  vector [floor1_X_list] element [low_neighbor_offset],
  vector [floor1_X_list] element [high_neighbor_offset],
  vector [floor1_Y] element [low_neighbor_offset],
  vector [floor1_Y] element [high_neighbor_offset],
  vector [floor1_X_list] element [i] )

must read:

5) [predicted] = render_point( 
  vector [floor1_X_list] element [low_neighbor_offset],
  vector [floor1_X_list] element [high_neighbor_offset],
  vector [floor1_final_Y] element [low_neighbor_offset],
  vector [floor1_final_Y] element [high_neighbor_offset],
  vector [floor1_X_list] element [i] )

<p>Also, in the same block the following step:

19) vector [floor1_final_Y] element [i] = 
  [predicted] - ([val] - [lowroom]) - 1

must read:

19) vector [floor1_final_Y] element [i] = 
  [predicted] - [val] + [hiroom] - 1

<p><p>Tor

<p><p><p>==================================================================EASY
and FREE access to your email anywhere: http://Mailreader.com/
==================================================================
<p>--- >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-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.

I'll reply to myself once more:

In the floor 1 curve synthesis, I see no need for step 11 and 12 
to be performed for each iteration. Wouldn't it be enough to do that 
once before step 15?

And is the condition in step 13 (hx > n) at all possible?

Tor

<p><p><p>==================================================================EASY
and FREE access to your email anywhere: http://Mailreader.com/
==================================================================
<p>--- >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-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.