This is a heads up to let you know of my plans to work on a senior thesis project dealing with editing Vorbis losslessly in Audacity. You can find more information about my plans and what I'm up to here: http://www.reverberate.org/computers/thesis/ "vcut" from vorbis-tools will be an excellent starting point for me. I notice that it's marked "experimental"; why is this? Are there important problems that need to be worked out before it will be reliable, or does it just need more testing and better handling of corner cases? Josh --- >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.
On Monday 08 September 2003 16:38, Joshua Haberman wrote:> This is a heads up to let you know of my plans to work on a senior > thesis project dealing with editing Vorbis losslessly in Audacity. You > can find more information about my plans and what I'm up to here: > > http://www.reverberate.org/computers/thesis/ > > "vcut" from vorbis-tools will be an excellent starting point for me. I > notice that it's marked "experimental"; why is this? Are there > important problems that need to be worked out before it will be > reliable, or does it just need more testing and better handling of > corner cases? > > JoshJosh, vcut is labelled as experimental because it's very buggy - the concept is entirely sound, and the format features it relies on (which are corner-cases) work fine, but the actual implementation has a lot of cases it just doesn't handle (or handles incorrectly). I wrote it primarily as a proof-of-concept, and also as a test tool to iron out problems with those decoder features in libvorbis (I think it did catch one), not really as a final useful tool. However, it should still be a useful starting point for your purposes. Feel free to ask me more questions about it if there are any parts that are unclear (probably "most of it" :-) 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-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.
Hi, I want to help code on a part of the ogg vorbis project and I really don't know where to start. I know C pretty well and I can use assembly every once in a while to speed things up. My first idea was to try and speed up the library with assembly (maybe SIMD) code. However, I soon realized I didn't understand the main parts of the encoder at all (I'm assuming most of the CPU time is taken up by this modified discrete cosine transform). Where is most of the help needed? A recent post said vcut needed work, should I do that? Or would I be helpful elsewhere? -- ~Mike wickedchicken@plasticpenguins.com --- >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.
On Monday 08 September 2003 18:38, you wrote:> "vcut" from vorbis-tools will be an excellent starting point for me. I > notice that it's marked "experimental"; why is this? Are there > important problems that need to be worked out before it will be > reliable, or does it just need more testing and better handling of > corner cases?Further to what Mike has said, there was some recent discussion on the particulars of it's current bugs and possible solutions in last month's vorbis list archive. Good luck - I'd be quite pleased to see it fully functional. John --- >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.