Slashdot Mirror
MPEG Continues With Royalty-free MPEG Video Codec Plans
yuhong writes "From the press release: 'In recognition of the growing importance that the Internet plays in the generation and consumption of video content, MPEG intends to develop a new video compression standard in line with the expected usage models of the Internet. The new standard is intended to achieve substantially better compression performance than that offered by MPEG-2 and possibly comparable to that offered by the AVC Baseline Profile. MPEG will issue a call for proposals on video compression technology at the end of its upcoming meeting in March 2011 that is expected to lead to a standard falling under ISO/IEC "Type-1 licensing", i.e. intended to be "royalty free."'"
I think I can save MPEG a lot of time. I've found a royalty-free container, a video codec and an audio codec we can all use:
http://www.webmproject.org/
So we won't find any videos of Charles, Camilla, William and Kate, Harry and the rest of the family in that format then
Better than Mpeg 2 they say? Well I should hope so. And AVC Baseline isn't great. They're clearly making some crap/free encoder so that they can start charging more $$$ for their good ones. The only issue for them is that Google/Xiph have good ones that will always be free. If MPEG tries to force this new standard people will move to VC8 which has been around for some time.
Probably just another knee-jerk reaction to VP8/WebM. And you can bet this "royalty free standard" will still be protected by tons of patents. It just keeps getting more interesting all the time. Just what we need, though, yet another video standard.
I happen to know that H.264 was _also_ supposed to be royalty free, with certain patents being reverse-engineered around in the standards development. MPEG-LA had different ideas, and they may have different ideas about this new work as well.
MPEG wants to merely merely standardize things, ending the problem of searching for a royalty free codec. Mozilla and Google both simply want a royalty free standard that is Good enough. VP8 seems like one possibility, but if something even slightly better than VP8 is standardized, both should be quite willing to implement it.
MPEG LA on the other hand actively does not want any codec better than say Microsoft Video 1 (the format most classic AVI files used) available on royalty free terms. They would lose out on a substantial amount of royalties if devices like phones or low end Digital Cameras used such a format rather than one of their formats. This is why they so actively fear projects like WebM. They make a substantial portion of their royalties from Cell phones, low end cameras, and similar devices.
Stylish sheet to fix many problems in Slashdot's D3: https://gist.github.com/801524
Really, I think everyone should care about this issue. It all boils down to device ownership. Say you buy a decent prosumer camcorder with the intent of maybe shooting your own low budget film. You purchased the camera so you own the device and therefore should not have to pay any additional royalties for using it in a way you desire. Under the MPEG-LA licensing agreement, you will have to pay royalties for each copy of the film you distribute to the MPEG-LA. This could get quite expensive and, in effect, creates a legal racketeering operation. You as the filmmaker are threatened with high punitive fines making it even more costly to try your own film out. Ignorance is what allows corporations (and government, too) to get away with such actions. This is where VP8 comes into play. Imagine if you had a prosumer camcorder with the VP8 capabilities - you would not worry about creativity and artistic innovation.
Not necessarily. They could decide to adopt Theora as the basis of the new standard, and see if they can get royalty free patent licenses for possible improvements.
Keep in mind that MPEG has little issue with standardizing something that already exists, like how the MOV container format was standardized as the MP4 container format, how they standardized Adobe and Microsoft's OpenType as MPEG4 Part 22: Open Font Format, or how they standarized a slight modification to ASPEC as MPEG-1 Layer 3 Audio.
Stylish sheet to fix many problems in Slashdot's D3: https://gist.github.com/801524
It is free for now ONLY for internet videos offered for no charge.
Products and services other than Internet Broadcast AVC Video continue to be royalty-bearing.
http://www.mpegla.com/Lists/MPEG%20LA%20News%20List/Attachments/231/n-10-08-26.pdf
I am a free slashdotter. I will not be modded, blogged, DRM'd, patented, podcasted or RFID'd. My life is my own.
Dirac is meant to be a high-quality codec, period; it was largely designed for archival work. It makes no particularly strong effort to be low-bitrate in the process.
The result is that at high bitrates it's pretty good (and even offers lossless compression, etc). At the bitrates at which people normally serve internet video today, it's worse quality than Theora, I'm told. But this is third-hand, so don't take my word for it.
As bandwidth goes up, Dirac might find a place on the web, but we're not there yet.
Dirac was designed for the future. Whilst it can achieve the same bitrate/quality as H.264 it is much more computationally expensive. It is also too different from H.264 to reuse the existing hardware acceleration. It was designed (by the BBC) for distribution of broadcast streams, where the required hardware is irrelevant, its better performance at high bit rates on super-HD content, optional lossless-compression, and ability to down-sample without re-compressing are all more important. In a decade Moore's Law may make it the obvious choice for all content. By acting as prior-art for most wavelet encoders it may be very important indeed.