Editor Tools Programmer Project (split)
- Tab.
- Joined: Tue May 13, 2003 10:36 pm
- Status: SLP
- Location: gayville
Bad news guys :|
The recent ffdshow/ffvfw developments seem to contain completely working huffyuv yv12 code. Project has really improved since Milan came back into the picture. The median prediction has some small issues, but plane prediction did pretty damn well in my tests anyway, median having an advantage of only 97-98%. So.. yeah.
The recent ffdshow/ffvfw developments seem to contain completely working huffyuv yv12 code. Project has really improved since Milan came back into the picture. The median prediction has some small issues, but plane prediction did pretty damn well in my tests anyway, median having an advantage of only 97-98%. So.. yeah.
- Tab.
- Joined: Tue May 13, 2003 10:36 pm
- Status: SLP
- Location: gayville
- Zarxrax
- Joined: Sun Apr 01, 2001 6:37 pm
- Contact:
- Kalium
- Sir Bugsalot
- Joined: Fri Oct 03, 2003 11:17 pm
- Location: Plymouth, Michigan
-
- is
- Joined: Tue Jul 23, 2002 5:54 am
- Status: N͋̀͒̆ͣ͋ͤ̍ͮ͌ͭ̔̊͒ͧ̿
- Location: N????????????????
Well, one of the reasons I want to do this is so that I can learn more about how video storage works at the computer's level.Tab. wrote:Zarxrax wrote:Thats bad news?apparentlykmv wrote:The core of the project will be a cross platform lossless video codec based upon the (windows only) huffyuv codec and enhanced to support YV12 colourspace (the colourspace used by DVDs)
Producing a useful product is obviously a goal, but I also have educational motives.
- AbsoluteDestiny
- Joined: Wed Aug 15, 2001 1:56 pm
- Location: Oxford, UK
- Contact:
It doesn't matter where ffvfw is with the code. It's a project that is under constant change and the multifaceted nature of the codec makes it unstable and not appropriate for storage use.
No offense to Milan but the Huffyuv codec needs to be stable so that people can trust that the captures they make with it are going to be decodable in future iterations. When you have a codec where different parts of the code are constantly changing then it threatens its stability.
If the HuffYV12 code works then that's great, it can be separated from ffvfw with credit given where it is due to keep it in a stable dll release. There are still many more goals to achieve - a quicktime version for example - so I don't see this as being in any way a negative thing for this particular project.
No offense to Milan but the Huffyuv codec needs to be stable so that people can trust that the captures they make with it are going to be decodable in future iterations. When you have a codec where different parts of the code are constantly changing then it threatens its stability.
If the HuffYV12 code works then that's great, it can be separated from ffvfw with credit given where it is due to keep it in a stable dll release. There are still many more goals to achieve - a quicktime version for example - so I don't see this as being in any way a negative thing for this particular project.
- AbsoluteDestiny
- Joined: Wed Aug 15, 2001 1:56 pm
- Location: Oxford, UK
- Contact:
- AbsoluteDestiny
- Joined: Wed Aug 15, 2001 1:56 pm
- Location: Oxford, UK
- Contact:
What's the news kmv?kmv wrote:Don't worry though I should be "with it" by this time tomorrow.
Things to do (in no particular order):
- Setup CVS for the code that is already available
- Notify owners of existing code, tell them what is happening
- Create a project to do list
- Configure the SourceForge project
- Have a call for coders on the above list
- Create initial builds for existing code
- Sort out some kind of roadmap for the bigger items
- Kick off design
- Kick off coding
- Build some web pages for the SourceForge project
A lot of this any one of us could do if you would just PLEASE make on of us an admin.
I appreciate that people become busy but nothing can be done with this project when there is only a single admin who isn't around. There's a lot of work to be done and it would be great if we could get started.