Hi,
There seem to be 2 sets of block coding patterns in the
VP3x/Theora algorithm. Based on the comments in the code, the first
pattern is used if encoder version < 6 and the second pattern is used in
encoder version >= 6. None of the code to support the second pattern is
ever called. Is there an encoder with a version >= 6?
The code in question is QuadDecodeDisplayFragments() vs.
QuadDecodeDisplayFragments2() in CoreLibs/CDXV/Vp31/dx/generic/Frarray.c.
They follow different pattern rules. But the second version is never
called.
Thanks...
--
-Mike Melanson
--- >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
'theora-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.
yeah, I noticed that while working on the spec. It looks to me like an
experiment that never converged. However I will be looking closely at these
sorts of issues as I complete the spec. Eventually we will get rid of the
deadwood.
-----Original Message-----
From: Mike Melanson [mailto:melanson@pcisys.net]
Sent: Wed 2/19/2003 9:07 PM
To: theora-dev@xiph.org
Cc:
Subject: [theora-dev] Block Coding Pattern
Hi,
There seem to be 2 sets of block coding patterns in the
VP3x/Theora algorithm. Based on the comments in the code, the first
pattern is used if encoder version < 6 and the second pattern is used
in
encoder version >= 6. None of the code to support the second pattern
is
ever called. Is there an encoder with a version >= 6?
The code in question is QuadDecodeDisplayFragments() vs.
QuadDecodeDisplayFragments2() in
CoreLibs/CDXV/Vp31/dx/generic/Frarray.c.
They follow different pattern rules. But the second version is never
called.
Thanks...
--
-Mike Melanson
--- >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
'theora-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.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: winmail.dat
Type: application/ms-tnef
Size: 4899 bytes
Desc: winmail.dat
Url :
http://lists.xiph.org/pipermail/theora-dev/attachments/20030220/cba5ad9a/winmail-0001.bin
> Spec? I didn't know anyone else was working on a spec of the algorithm, too.A formal spec is slated for the upcoming Beta release (I'm the lucky volunteer on that). The bitstream is still in flux, so if you're working up something based on Alpha it may need tweaking. Perhaps we should discuss joining forces? - dan -----Original Message----- From: Mike Melanson [mailto:melanson@pcisys.net] Sent: Thu 2/20/2003 8:55 PM To: theora-dev@xiph.org Cc: Subject: RE: [theora-dev] Block Coding Pattern On Thu, 20 Feb 2003, Dan Miller wrote: > yeah, I noticed that while working on the spec. It looks to me like an experiment that never converged. However I will be looking closely at these sorts of issues as I complete the spec. Eventually we will get rid of the deadwood. Spec? I didn't know anyone else was working on a spec of the algorithm, too. -- -Mike Melanson --- >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 'theora-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. -------------- next part -------------- A non-text attachment was scrubbed... Name: winmail.dat Type: application/ms-tnef Size: 4911 bytes Desc: winmail.dat Url : http://lists.xiph.org/pipermail/theora-dev/attachments/20030220/52ce4ff7/winmail-0001.bin
Hi Mike -- The plan for the official Theora spec is similar to yours. Our specification document will include (actually, in a sense it will *be*) a reimplementation of the video decoder in Python, with an emphasis on clarity and simplicity over performance. At this point I think the best course is for us to continue to work independently for a while, at least until I have a first draft (or you do); then we can review & trade notes. My experience in this area on the MPEG committee is that independent review is the key to filtering out mistakes, unclear descriptions, and ambiguities. I would consider any reviewing, corrections, or additional documentation to be a serious contribution to the project. Does this make sense? - dan> -----Original Message----- > From: Mike MHielanson [mailto:melanson@pcisys.net] > Sent: Thursday, February 20, 2003 10:29 PM > To: theora-dev@xiph.org > Subject: RE: [theora-dev] Block Coding Pattern > > > On Thu, 20 Feb 2003, Dan Miller wrote: > > > > Spec? I didn't know anyone else was working on a > spec of the algorithm, too. > > > > A formal spec is slated for the upcoming Beta release (I'm > the lucky volunteer on that). The bitstream is still in > flux, so if you're working up something based on Alpha it may > need tweaking. Perhaps we should discuss joining forces? > > Sounds like a great idea. I have been studying VP3 > rather closely > the last few weeks (in between other multimedia-related > activities). I'm > quite confident that I have a good grasp on the decoding > process and have > been jotting down many notes as I have gone along. The end > goal is to come > up with a coherent document and be able to re-implement the > decoder (and > eventually the encoder) in ffmpeg based on the document. > > -- > -Mike Melanson > > > --- >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 > 'theora-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. >--- >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 'theora-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.
one thing to keep in mind is the issue of variable names. The names in the C libs are rather cryptic. I'm renaming them to be more in line with a specification document.> -----Original Message----- > From: Mike Melanson [mailto:melanson@pcisys.net] > Sent: Friday, February 21, 2003 1:07 PM > To: theora-dev@xiph.org > Subject: RE: [theora-dev] Block Coding Pattern > > > On Fri, 21 Feb 2003, Dan Miller wrote: > > > At this point I think the best course is for us to continue > to work independently for a while, at least until I have a > first draft (or you do); then we can review & trade notes. > My experience in this area on the MPEG committee is that > independent review is the key to filtering out mistakes, > unclear descriptions, and ambiguities. I would consider any > reviewing, corrections, or additional documentation to be a > serious contribution to the project. > > > > Does this make sense? - dan > > Absolutely. It is a good thing to have multiple independent > interoperable implementations (I read somewhere that that's actually a > requirement for IETF specs). Plus, I have always found that > it never hurts > to have more than one perspective on technical matters. I.e., multiple > documents and explanations from different authors helps a lot. > > Thanks... > -- > -Mike Melanson > > > --- >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 > 'theora-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. >--- >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 'theora-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.