Slashdot Log In
Theora 1.0 Released, Supported By Firefox
Posted by
timothy
on Tue Nov 04, 2008 09:00 AM
from the funny-name-and-all dept.
from the funny-name-and-all dept.
YA_Python_dev writes "The Xiph.Org Foundation announced Monday the release of Theora 1.0.
Theora is a free/open source video codec with a small CPU footprint that offers easy portability and requires no patent royalties.
Upcoming versions of Firefox and Opera will play natively Ogg/Theora videos with the new HTML5 element <video src="file.ogv"></video>, and ffmpeg2theora offers an easy way to create content.
Theora developers are already working on a 1.1 encoder that offers better quality/bitrate ratio, while producing streams backward-compatible with the current decoder." Adds reader logfish: "Since its bit-stream freeze in June of 2004 there have been numerous speed-ups and bug-fixes. Although Nokia claimed it to be proprietary almost a year ago, nothing has been proven. So now it's time to help it take over the internet, and finally push for video sites filled with Theora encoded vlogs, blurts and idle nonsense."
Related Stories
[+]
Developers: Theora I Bistream Format Frozen 329 comments
p80 writes "The Xiph foundation announced today that the 'Theora I bistream format is now frozen,' even though Beta 1 is not out yet and encourage people to try it as 'there's no reason to delay adopting a free alternative any more!' Mplayer and Xine both support Theora. For Windows users, Directshow filters for Ogg Vorbis, Speex, Theora and FLAC are available here. You can get test cases here and transcode Quicktime movies to theora on that page." This freeze, as an anonymous reader puts it, "means that all future versions will support the format as it is now. It will be interesting to see if there is as much uptake for this as there was for the Vorbis sound format."
[+]
Technology: Nokia Claims Ogg Format is "Proprietary" 619 comments
a nona maus writes "Several months ago a workgroup of the W3C decided to include Ogg/Theora+Vorbis as the recommended baseline video codec standard for HTML5, against Apple's aggressive protest. Now, Nokia seems to be seeking a reversal of that decision: they have released a position paper calling Ogg 'proprietary' and citing the importance of DRM support. Nokia has historically responded to questions about Ogg on their internet tablets with strange and inconsistent answers, along with hand waving about their legal department. This latest step is enough to really make you wonder what they are really up to."
[+]
Your Rights Online: Ogg Vorbis / Theora Language Removed From HTML5 Spec 395 comments
Rudd-O writes "It's official. Ogg technology has been removed from the HTML5 spec, after Ian caved in the face of pressure from Apple and Nokia. Unless massive pressure is exerted on the HTML5 spec editing process, the Web authoring world will continue to endure our modern proprietary Tower of Babel. Note that HTML5 in no way required Ogg (as denoted by the word 'should' instead of 'must' in the earlier draft). Adding this to the fact that there are widely available patent-free implementations of Ogg technology, there is really no excuse for Apple and Nokia to say that they couldn't in good faith implement HTML5 as previously formulated."
[+]
Technology: Ogg Theora In Firefox, With Wikimedia Support 339 comments
An anonymous reader writes "Ogg Theora support for the HTML5 <video> tag is in the Firefox 3.1 nightlies. Theora is the only video format allowed on Wikimedia Commons, so Wikimedia people are pushing Wikipedia readers to download a nightly and try it out. Break it, crash it, report bugs, get it into good shape and nullify Apple and Nokia's FUD the best way possible. They may have gotten the words 'Vorbis' and 'Theora' removed from the HTML5 spec, but the market will tell them when their browsers are sucking."
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
Free Is Good, But Quality Is Lacking (Score:4, Insightful)
I really want to like Theora, but it's really, really hard to get around the quality issues. VP3, which Theora is based on, just isn't competitive these days. It was subpar back in 2001 when it was donated to Xiph, and the contrast has only gotten worse over time. H.264, VC-1/WMV9, MPEG-4 ASP, even Adobe Flash 8 (which added VP6) are clearly capable of outperforming it.
If nothing else, free is good (both in terms of speech and beer) and a royalty free standard for video would be great, but it's too hard to ignore just how inferior this standard is. I'm a pragmatic person, I can't think of any reason why I'd want to use this over a better codec; free isn't all that enticing if the video quality sucks.
Re:Free Is Good, But Quality Is Lacking (Score:5, Informative)
There's certainly better quality codecs out there, compared to 1.0. Take a look at the work happening now on 1.1, though, it gets very competitive:
http://web.mit.edu/xiphmont/Public/theora/demo5.html [mit.edu]
Parent
Re: (Score:3, Insightful)
it gets very competitive
That might be true, but it isn't demonstrated by posting a link that compares Theora and Theora.
I'd like to post a clip that compares Theora to the formats and codecs it tries to compete against. I don't have elephants dream available at the moment, and I don't want to get slashdotted, but someone could reencode the high-resolution version of it and post links.
Then we can compare Theora to its competitors, to see exactly how competitive it is.
Re:Free Is Good, But Quality Is Lacking (Score:5, Insightful)
- noise
- fire
- rain or snow
- smoke
These are the frames which have the highest amount of entropy and are easiest to visually illustrate the quality of a coder.
Parent
Re:Free Is Good, But Quality Is Lacking (Score:4, Informative)
http://web.mit.edu/xiphmont/Public/theora/demo.html
Note that this is before the major changes made this summer and the major changes still coming (in the encoder). The VP3 technology actually puts it between MPEG-4 ASP (xvid/divx) and H264 in theoretically achievable quality, it's just that the encoder has been extremely badly tuned up until this summer, because of lack of interest. If Theora can catch up to MPEG-4 ASP codecs and perhaps even close in on H264, it would make for an excellent patentfree codec.
Parent
Re: (Score:3, Insightful)
Re:Free Is Good, But Quality Is Lacking (Score:5, Insightful)
Can it though? Certainly part of the issue is definitely the encoder, but you're still constrained by the inherent limitations of the codec (and more to the point, the decoder). Theora can't be overhauled without breaking the decoder, and even if it was overhauled as Theora 2.0, it couldn't implement any of a multitude of patented video compression technologies already used in MPEG or other standards. And unless someone wants to hire a team of engineers for Xiph, the odds of someone inventing a revolutionary, non-patent-infrining video codec on their own is pretty slim.
From what I've seen with the work on 1.1, improving the encoder just isn't enough to nullify the deficiencies in the codec itself. It's like trying to improve Mac OS Classic when really you need to make a clean break and invent Mac OS X.
Parent
Re:Free Is Good, But Quality Is Lacking (Score:5, Interesting)
The fact that the quality improvements for theora 1.1 put it on par with a base mpeg4 implementation while not on par with the most recent h264 encoders is not really relevant in the larger sense.
Once a free codec becomes widely adopted the chance of some proprietary codec coming along afterwards is near zero. Its just like today we can't imagine someone coming out with a proprietary image format and expecting people to adopt it.
Its relatively easy to add in support for Dirac or some future free codec once there is support for a free codec ecosystem. No one will pay h264 licensing costs when quality free alternatives are vibrant. The entrenched proprietary systems are being pushed aside for free alternatives. This 1.0 release is a step towards that direction, not as big of a step once firefox 3.1 ships but an important step ;)
Parent
Re:Free Is Good, But Quality Is Lacking (Score:5, Insightful)
Parent
Re:Free Is Good, But Quality Is Lacking (Score:5, Informative)
Parent
Re:Free Is Good, But Quality Is Lacking (Score:4, Interesting)
Theora quality is good enough. You may be able to see a difference under a microscope but not to the human eye.
To get a good quality video, from A to Z must be quality, not just one step.
Many convert a MPEG video into Theora and say Theora video is not good quality. MPEG means video is already compressed and data is dropped as part of the encoding process. You don't take a such a video and convert the format to Theora and drop data again as part of the Theora encoding process. Its of course not good quality then, because data is dropped twice.
To get real quality video in Theora, you should get a raw video and convert to Theora. I have converted a raw video footage shot by a RED camera (http://www.red.com/) into Theora, I don't see any quality issue. Its crisp clear.
In digital camcoders this is what is done inside, ie. it first shot in raw and then convert to MPEG. I have not come across a video camera that convert to Theora natively inside the camera. If it does, there should not be any quality difference compared to MPEG.
Sagara
Parent
Re:Free Is Good, But Quality Is Lacking (Score:4, Interesting)
Meh. It ain't 1999 anymore. Forgetting Youtube? Flash IS that "magical wonderland", and it DOES just work for the majority of the population.
The question is, can free-as-in-beer, inferior open source compete against free-as-in-beer, superior closed source?
Parent
Re:Free Is Good, But Quality Is Lacking (Score:5, Insightful)
Don't forget, lots of nightmarish IE specific stuff also "Just Works" for "The Majority", And ask any 64bit linux user exactly how much they love adobe for their support. (I think they have it now, after something like 4 years of waiting or running in emulation, or running a 32bit OS on their 64bit machines)
The magical wonderland I think of is one where anyone on any system can easily watch video online, not just the majority.
Parent
Re:Free Is Good, But Quality Is Lacking (Score:4, Informative)
Parent
Re: (Score:3, Interesting)
Re: (Score:3, Informative)
The question is, can free-as-in-beer, inferior open source compete against free-as-in-beer, superior closed source?
x264 is open source and gives way better quality than Theora - but it is also a patent minefield and you will need to get in contact with MPEG-LA if you plan on doing commercial stuff etc.
Dirac (Score:3, Informative)
Re:Dirac (Score:5, Informative)
They're aimed at different markets. Dirac is a very high-quality CODEC, but it is incredibly CPU intensive. Remember what MPEG-4 was like when it was introduced? A couple of days to encode a film, and you could only just decode it in realtime on a fast computer? Dirac is like that. It will be a few years before you start getting Dirac support in something like an iPhone. If you want to stream HD content, Dirac is a good choice.
In contrast, Theora is very cheap in terms of CPU power. You can run it on very low-power devices. This makes it a good choice for Internet video, where the viewer might be using a massive desktop computer, a mobile phone, or anything in between. You wouldn't want to use Dirac here, because even fast laptops would struggle not to drop frames, and handhelds would just fail.
That said, my mobile phone now has about as much CPU power as the PC I had when I first got an MPEG-4 video, so eventually it will be feasible to run Dirac in low-power devices (sooner if they have dedicated ICs), but in the short term it's not ideal.
Parent
Uh? (Score:4, Insightful)
"Although Nokia claimed it to be proprietary almost a year ago, nothing has been proven. So now it's time to help it take over the internet"
I admit I don't know what the situation with Theora's licensing history is but this comment strikes me as rather worrying. We're being told to use it because no one's proven it's not likely to end you up with licensing troubles later on. Personally I'd rather before something "takes over the internet" that the burden of proof was on it to demonstrate that it is completely open. This should be as easy as showing use of a relevant open license no?
From what I can see it's under a BSD license and so should really be open. Is this the case? The way the article summary is written just really doesn't instil confidence in their intentions.
Giving this codec the benefit of the doubt I think the summary is just a case of carried away fanboyism having an adverse effect towards the neutral observers view of the situation much as seeing a forum war between a PS3 and a 360 fanboy might put someone off the idea of online console gaming.
Can someone a bit more grounded give us a better view of the concerns and realities of Theora licensing and it's suitability as a codec to "take over the internet"?
Re:Uh? (Score:5, Informative)
Xiph had the Software Freedom Law Center help establish that Nokia's claims were untrue. Mozilla sought counseling from lawyers before supporting Theora. Is that enough?
Parent
Re:Uh? (Score:4, Informative)
That "nothing has been proven" comment is pretty clearly tongue-in-cheek, just like the "take over the Internet" part.
The video encoding field is crowded with patents, so it's probably impossible to do something like Theora without needing a patent license. But Theora is based on some patented technology (VP3) whose patents have been donated for free use, irrevocably forever. And Theora is free, open source software with a BSD license. If you use the Ogg container format, Theora video, and Vorbis audio, you have a completely free media format.
http://en.wikipedia.org/wiki/Theora [wikipedia.org]
So, you can use Theora for any purpose, without needing to pay royalties, without needing to get permission. That's why it's so funny that Nokia claimed Theora is "proprietary"... I do not think that word means what they think it does.
steveha
Parent
Command line is not easy for most users (Score:3, Insightful)
Only for certain definitions of easy. Let me know when you have a point and click version that my non technical friends can use.
Re: (Score:3, Informative)
Re: (Score:3, Informative)
Mod me troll if you like, but I speak the truth.
Re:Containers... (Score:5, Insightful)
You've obviously never negotiated costs with MPEG-LA, or you wouldn't say that.
Parent
Re: (Score:3, Interesting)
Re:Containers... (Score:5, Insightful)
If we want h264/mpeg4 support in FF you going need about $3M+ donated per year for the license fees.
If you have ever needed to care about the licensing of things like codecs you would know the value of Theora and Dirac.
Parent
That's not what "free" means. (Score:5, Informative)
Sure, you can consider it to be free, but boy is that ever not what free means.
And a publically available spec means little or nothing. Patents are publically available, but try implementing those and see if you manage to escape the long arm of the litigator.
Parent
Re:Containers... (Score:5, Insightful)
Parent
Re:MPEG 4 LA terms (Score:4, Informative)
You are also forgetting the fees for producing content in H264 that come into effect later.
They are also leaving out a *lot* of fine print. In order to get a license you don't just have to pay, but you must agree to the license terms (aka hardware players must use zone flags, DRM etc). There is more than they tell you without NDAs. Not to mention all the lawyer fees in between.
They are not even going to let you pay a blanket fee for a product that others can use "free" in there own products. No matter what you pay. Because then there is no one else that needs a license and hence no one to tie into these extra terms.
Also what makes you think the fees won't increase at a latter date?
Encumbered means just that. Encumbered.
ps I have talked to them about a license......
Parent
Re:Containers... (Score:5, Informative)
Parent
Re:Well... (Score:5, Informative)
So is MKV, just a container.
Parent
Re: (Score:3, Insightful)
Ummm, literally true, but your comment seems mostly unrelated to the post it is in response to. There was a discussion about container formats, AVI came up in part of that discussion as a container format, and then you told them it was not a codec.
Re: (Score:3, Informative)
Re:Native Video in Firefox (Score:5, Funny)
Sounds like feature creep and bloat to me.
Don't worry, the pages that implement it will never get loaded into RAM because nobody will ever use it.
Parent
Re: (Score:3)
It's only bloat if they rewrite their own theora codec.
Re:Native Video in Firefox (Score:5, Insightful)
Yep, and Windows proves Linux is unnecessary as it is a widely accepted and usable solution for operating a computer.
Parent
Re: (Score:3, Funny)
Sounds like feature creep and bloat to me.
Are you talking Firefox or HTML5 or both? I know it would be massively awesome if the blink tag was only supported through an addon or plugin.
So you prefer Flash installed on every browser? (Score:4, Insightful)
Sounds like feature creep and bloat to me.
Why? Ins't going to affect you if you don't visit pages with videos and, unlike Flash there's a browser preference to start all videos in paused state. The Theora binary library is only 250 kB on AMD64, even smaller on x86. The Flash plugin, is much, much bigger.
Video on the internet (think youtube, movie trailers, pr0n, etc.) isn't going away any time soon.
The current state of the art is to have a proprietary Flash plugin installed in almost every browser. Switching to native support for an open format directly in the browsers seems like an improvement to me. In the good ol' days, people considered image support in browsers as bloat too..
And Firefox isn't alone here: Opera and Safari will support it too (altough Safari will not support Theora out-of-the-box).
Parent
Re:Long test cycle? (Score:5, Informative)
The bitstream format was frozen, not the code.
Parent
Re:How long until... (Score:5, Funny)
Parent
Re:Why not using the "object" tag? (Score:5, Informative)
Parent
Re: (Score:3)
What API? Tags do not have APIs, and the <object> can be extended with "params"s... This "video" tag looks a lot like an old Netscape HTML "extension" to me.
Re:Horray (Score:5, Informative)
Parent
Re:Horray (Score:4, Funny)
"smaller random nerd sites and one big nerd site"
there, fixed.
Parent
Re:Horray (Score:4, Informative)
Parent
Re:Back to the better! (Score:5, Informative)
People and companies likes flash players because it usually just works. The days of embedding video objects are dying because in practice this is what would happen:
1. WMV files would lock up or you would have to spend 20 minutes at windows update downloading the newest wmp or reinstalling the plug-in.
2. Mac users would complain that WMV files arent working.
3. Realplayer would do the same, except the install would crap up your computer and ruin all your file associations. You would also have to troubleshoot plugin issues.
4. Quicktime files would crash the browser and then you would have to install the newest version usually along with itunes in a 60+ meg download. Windows users would complain how crazppy quicktime is.
5. Someone would embed an avi and no one would be able to play it because end users have no idea what codecs are.
6. Some plugins would work in IE but not in Firefox.
What flash did is put all video in one cross-playform container and player. Turns out people like it this way.
Parent
Re: (Score:3, Informative)
Judging by google results, it sounds like "video microblogging", a la Twitter.
Re: (Score:3, Informative)
Oh, and to answer this question:
The answer is, yes, depending on the codec in question (for example, Microsoft would pay the MPEG-LA to distribute an MPEG2 video decoder). But keep in mind, a file format, in and of itself, isn't subject to patent. It's the methods used to create the file format that are the problem. So exportin
Re:Thank you very much, Mozilla Corp. (Score:5, Insightful)
As Wikipedia would say: "Citation needed".
Care to show an example of *any* MPEG-2 codec out performing the current Theora encoder on a typical web-video 500kbit/sec stream? Forget the new enhanced theora encoder, MPEG-2 can't even match the old crap. Plus mpeg-2 is patented to hell and back, you even have to pay for mpeg-2 decoding in Windows to play DVDs!
Can you cite a *single* example showing Vorbis to be glaringly inferior to AAC? At best the listening tests show AAC to edge out Vorbis only for speech samples at the lowest bitrates (where Xiph has Speex, which blows AAC away for those applications). And no multi-channel? wtf. Vorbis supports 255 channels.
I shouldn't expect better from slashdot, but could you at least find lies that are a bit less obvious.
Ogg high overhead? Okay, Ogg/Vorbis+Theora is something like 1% overhead vs a typical of 0.9% overhead for a movie in AVI. You win there. Then again, OGG provides frequent checksums so that a damaged OGG/Vorbis file will *never* break your speakers and damage your hearing. People who have had the misfortune of hitting a corrupted MP3 in their iPod playlist should be able to appreciate the advantage of this approach. What you consider a fault I consider a feature. Egads, room for design differences exists! who would have thought?
Parent