Mozilla Donates $100K To the Ogg Project 334
LWATCDR writes "Mozilla has given the Wikimedia foundation $100,000 to fund Ogg development. The reason is simple: 'Open standards for audio and video are important because they can be used by anyone for any purpose without royalties, and can be inspected and improved by an open community. Today, video and audio on the web are dominated by proprietary technologies, most frequently patent-encumbered codecs wrapped into closed-source player widgets.' While Vorbis is a better standard than MP3, everything I have heard about Theora is that it is technically inferior to many other video codecs. I wonder if wouldn't be better to direct effort to Dirac, perhaps putting Dirac into an Ogg container. No mention was made of FLAC or Speex funding. If more media players supported Speex it would be an ideal codec for many podcasts and audio books. It really is too bad that these codecs so often get overlooked."
Re:"Better" is relative... (Score:5, Informative)
MP3 players now mostly use hardware decoders, because they are much cheaper and energy-efficient than CPU decoding.
Re:I thought Ogg was dead (Score:3, Informative)
Ogg as an audio format is pretty good. I investigated it a few years back for a voicemail application I was working on. It worked pretty well and obviously comes unencumbered by patents or copyrights, unlike MP3.
I think the main problem is the public's conception of MP3 as the gold standard for music formats. MP3 players have pretty well saturated the standalone market, though flash multimedia and other kinds of streaming formats have made inroads in connected media.
At this point, for ogg to achieve some kind of inroads in the mass audio market, the MP3 owners would need to really jack up their rates, because presently it's priced into every product already. If you save $1 by leaving MP3 out of your player, is it worth it?
But, I am glad to see Mozilla at least making a symbolic effort to keep Ogg (and Theora) alive. As a poster above points out, $100K is not much money in the grand scheme of things, but it is a lot better than nothing and it might keep Ogg on life support a while longer until that killer application comes along (e.g., support for Ogg on iPod and similar players (probably would never happen on Zune, but that particular player looks like it's circling the drain anyway....).
Re:"Better" is relative... (Score:1, Informative)
CPU time? Nah.
Many of the players that don't support Ogg/Vorbis support it just file with the rockbox firmware on them. It's not a hardware issue. Vorbis has similar CPU demands to AAC.
Re:"Better" is relative... (Score:3, Informative)
Re:"Better" is relative... (Score:3, Informative)
The overhead is not so bad if the decoders are done in an efficient manner.
I used to run around with a IPAQ 1910 (46mb available ram total, 300 MHz. It could do full screen, full motion decoding w/ OGG (128-192kbps) and MPEG-4 at the same time.
There are very few modern MP3 devices which _don't_ have sufficient horsepower to decode ogg, yet can handle MP3s.
Re:"Better" is relative... (Score:5, Informative)
Ogg might be "better" than MP3 in terms of sound quality
First: <pedantic>Ogg is the container format, like QuickTime or AVI. Vorbis is the audio codec being compared to MP3. You could, if you wanted to, put MP3 bits into an Ogg container; I guess this would be "Ogg MP3". </pedantic>.
Vorbis gives you better quality per bit than MP3. That means you can have higher quality in the same number of bits, or similar quality in fewer bits. Given that most of us aren't using modems anymore, perhaps this is only a weak selling point for Vorbis. It's still nice for small portable music players, though.
but ultimately it consumes significantly more CPU time.
As I understand it, the overhead for Vorbis isn't really that bad. The chief sticking point is that the little portable players use DSP chips, and the DSP chip vendors have excellent support for MP3 and no support for Vorbis. This means that when a project like Rockbox [rockbox.org] adds Vorbis support to a portable player, often they use the main CPU instead of the DSP chip, and that means a drastically worse power drain.
A sticking point from the past was that Vorbis was written to use floating-point math in the decoder. The Vorbis folks made an integer-math-only decoder called Tremor [wikipedia.org], which answers that point.
For a desktop computer, you would never notice the difference between a good Vorbis decoder and a good MP3 decoder.
I think the main reason for the lack of Vorbis takeup is inertia. Everyone has MP3s, so the players all support MP3s. Since the players support MP3s, only geeks like me bother with Vorbis, so the player companies don't feel motivated to support anything but MP3. I used to hope for Vorbis support everywhere, but now MP3 is just a few years away from its patents expiring, so it's going to be MP3 for the near to middle term.
I own a couple of Sansa players that can play Ogg Vorbis. They have excellent battery life, despite being tiny little things. They stand as examples that there is no inherent technical reason why Vorbis cannot work on small portable players. By the way, if you are a geek, you should consider one of these before you buy an iPod Shuffle; more features for less money, and it works as a USB storage device so it works perfectly well on Linux.
http://www.sansa.com/players/sansa_clip/tech [sansa.com]
steveha
Hey, Speex is part of the Xiph foundation (Score:2, Informative)
I do adore how the money is apparently only for "Ogg" (and the OP means Ogg Vorbis) when Ogg is the *container format*.
Speex is just as supported by the Xiph foundation as Vorbis is.
Re:I thought Ogg was dead (Score:1, Informative)
Android also uses it.
Re:Network effects keep Ogg out (Score:3, Informative)
"Where Ogg should excel is in pure software applications, especially in heavily patented areas like VoIP where there is no hardware cost, where it's trivial to add codecs, and where the current state of play penalizes cheaper solutions."
That is where SPeex really does shine. Heck Microsoft uses Speex for XBox Live. Too bad they don't support it on the Zune.
Re:I thought Ogg was dead (Score:3, Informative)
The hardware support is impressive too. Everything from Sansa and Neuros to iRiver and Cowon support both the vorbis and flac codecs. The only major missing player is Apple. Considering over half of my collection is ripped or downloaded in these formats, that is why Apple is not received a dime from me.
Re:"Better" is relative... (Score:2, Informative)
The newest firmware even plays FLAC :) I have a 4GB clip and I love it. OLED screen, mp3/wma/vorbis/flac support, usb mass storage, usb charging, fm tuner, fantastic battery life, AND it's tiny. Great stuff.
some source links and information (Score:5, Informative)
hmm not the post I would have chosen for this news... Could have pointed out some of the source [wikimedia.org] post [0xdeadbeef.com] announcements [metavid.org] and avoid perpetuating a few misconceptions.
Hence the need for funding the Thusnelda [mit.edu] enhancements. Theora is a pretty solid codec and can be greatly improved with a few enhancements on the encoder side.
Dirac is best at high resolution high bitrate video and not so good for standard definition low bitrate video, hence an enhanced theora is the optimal way to hit the low bandwidth target. Enabling theora to be competitive or better than others codecs in the low bitrate range in the intimidate future with relatively small investment.
Furthermore dirac is planed for inclusion and will be explored in the tail end of this grant. (once liboggplay is more solid). Making liboggplay playback library solid will enable Dirac support to be solid as well. Since Dirac already has a maturing decoder/encoder library (Schrodinger) and already been mapped to an ogg container (what liboggplay plays).
It's relatively easy to add in additional free codecs with ogg mappings. if( FLAC, Speex or Dirac) and will not be the primary use of the funding so its not focused in on the announcement or secondary coverage of the announcement.
More info on the announcement here [metavid.org] and the above mentioned links.
Theora's place among video codecs (Score:3, Informative)
From the summary:
everything I have heard about Theora is that it is technically inferior to many other video codecs.
I am not an expert on video codecs, but here is my understanding of the situation.
Theora is a relatively undeveloped technology in comparison with the industry standards of MPEG2 or MPEG4. There are relatively few developers working on it. Overall they have done a pretty good job of defining a standard, but they are still working on improving the encoder. The encoding format is now frozen, which means you can write a decoder and expect it to be able to decode any future Theora bitstream; but the encoders are still being improved. The earliest Theora encoders were pretty terrible, but newer ones have gotten better, to the point where Theora is now more efficient than MPEG2. ("More efficient" meaning encoding the same video at the same quality in fewer bits, or encoding better quality in the same number of bits.) MPEG4 is currently more efficient than Theora, but not free.
There is plenty of room for a clever encoder to reduce the bitstream with video. As a trivial example, suppose we are encoding a scene where a car is driving from left to right. A brain-dead encoder could simply notice that the car pixels have changed, and encode them all over again; a smarter encoder could detect that the next frame looks very much like the previous frame, except that certain pixels have slid over a bit, and instead of re-encoding every changed pixel, the clever encoder can encode "these pixels are like those older pixels, except slid to the right by X amount". It's not easy to write an encoder that can do an optimal job of figuring out the most efficient way to represent the changes between several frames of video. Many more man-years have been spent on proprietary MPEG encoders compared to the time spent on Theora so far.
It is not clear to me how much room for further improvement there might be. Can Theora ever approach MPEG4 for efficiency? My guess is that there are patented technologies in MPEG4 which allow for more efficiency than is possible with Theora, but I don't know to what degree. Note that the Theora guys are saying [theora.org] that Theora is in the same class with MPEG4.
Given that MPEG2 is considered adequate for many purposes, it seems to me that Theora should be adequate for many purposes, and it's free. I have high-speed Internet and I would love it if Youtube and such sites offered Theora video in addition to Flash; the Flash player seems to leak memory a lot and I wish I didn't need it.
I wonder if we will start to see Theora-encoded video cutscenes in video games, just as we have seen Vorbis-encoded audio in video games?
If I got anything wrong in the above, please correct me.
steveha
Re:Network effects keep Ogg out (Score:2, Informative)
Re:Network effects keep Ogg out (Score:2, Informative)
Efficiency or something like that. Playing mp3s in rock box I get about 7 hours on my player. Playing back the same as ogg and I get less than 4 hours. Now the firmware that comes with the player will let me play those mp3s for close to 10 hours.
Re:Go ahead - throw your money away (Score:5, Informative)
Uninformed troll you are, sir.
Had you checked any source - even Wikipedia - you would know that Ogg Vorbis is being used extensively in game industry, both for technical superiority (not only that of the codec itself, which could be disputed, but of the library, which is very easy to integrate and fully supported by the Miles Sound System) and legal status. There are no patents on this, so the lawyers (and, consequently, the execs) in the game development studios are happy because they don't have to worry about some random company telling them to pay up a week before release and yet, it costs nothing.
Mp2 is not compatible with mp3 (Score:4, Informative)
mpeg1 layer 2 audio and mpeg 1 layer 3 audio are not forward/backward compatible. They share some blocks in the flowchart, but mp2 is not mp3 with a few blocks stripped off.
If you have an mp3 player/decoder that plays mp2, it's because there's a separate mp2 decoder in there too.
Re:Mp2 is not compatible with mp3 (Score:2, Informative)
That's simply untrue. All MP3 players can play MP2. It's a strict subset (or rather MP3 is a strict superset of MP2) Even if it wasn't inherent in the algorithm, in practice a conforming MP3 decoder is required to decode MP2.
Comment removed (Score:5, Informative)
They are funding FLAC and Speex (Score:5, Informative)
I guess neither the submitter or the editor that approved the story knows enough about Xiph and Ogg to know that they also support FLAC and Speex so donating to Xiph is in effect donating to these as well.
And also to Vorbis, Theora, Spiff and AO.
There really isn't 100k worth of work to do on the Ogg format, its just a container, one thats been rather well defined for a while. So considering Speex and FLAC are codecs supported in an Ogg container, and all 3 of them are managed by the same organization, I think its rather stupid to say 'no meantion of FLAC or Speex' just because the submitter doesn't know anything about Xiph.
Re:Mp2 is not compatible with mp3 (Score:3, Informative)
Nope, you're wrong. Quoth Wikipedia:
MP2 is a sub-band audio encoder, which means that compression takes place in the time domain with a low-delay filter bank producing 32 frequency domain components. By comparison, MP3 is a transform audio encoder with hybrid filter bank, which means that compression takes place in the frequency domain after a hybrid (double) transformation from the time domain.
In practice MP3 decoders support MP2 too, but not because they're similar *at all*.
Re:Go ahead - throw your money away (Score:4, Informative)
It's funny you mention Miles. It's the one audio library you can use that comes with a blanket MP3 license (I think they made a very early deal with Fraunhofer before a general per-game license fee was decided on). Pretty much any other audio library, such as FMOD or wwise you'll have to pay an additional licensing fee.
The biggest reason to use Vorbis is actually technical, not the licensing agreement. It's not that expensive to license mp3 use for a game title - just a few grand per title, which is pretty insignificant compared to total development expenses for AAA title nowadays (more critical for indy developers, of course). Vorbis has support for 6-channel audio, and it has sample-accurate containers, meaning you can easily chop it up, splice it, loop it, etc... MP3 is really only good for simple playback scenarios.
Re:Mp2 is not compatible with mp3 (Score:1, Informative)
funny, how the wiki quote was used to say it's not related and the very next line on wikipedia explicitly says:
MPEG Audio Layer II is the core algorithm of the MP3 standards. All psychoacoustical characteristics and frame format structures of the MP3 codec are derived from the basic MP2 algorithm and format.
Re:But why? (Score:3, Informative)
Why do we care that the MP3 has patents and all that? What does it do?
Makes it impossible to support them directly in any open source software, for one thing -- Firefox included.
Yes, you can write a generic API, then write third party plugins that supply them, and distribute those plugins in countries which don't honor US patents, and download them illegally in the US. But then you will have no legal download of Firefox for the Windows users who just want to download it and play.
Also, I was developing a music website. It was not an ideological one, or at least, not software-wise -- though I fought it, we eventually ended up going to an all-Flash widget to play the songs. But we would have had to pay outrageous fees to re-encode all the music in MP3, and we'd suddenly have to think about licensing, whereas with other codecs, we can just fire up as many EC2 instances as we want.
Also, MP3 sound quality sucks compared to just about everything else. Since the average customer still thinks in terms of "mp3 player", moving to wma or aac is just as big a leap for them, so why not vorbis or flac instead?
Re:MP3 is irrelevant in this (Score:3, Informative)
You're not looking very hard. The main page has a 'Featured content' link in the sidebar. As I type, there are various media files features, such as Truman's announcement of the WWII surrender of the Japanese.
Unlike sites such as youtube, every piece of content on wikipedia is verified to be in the public domain. Meaning of course, anyone can access it without fear of copyright infringement. Using the above example, if a teacher Evelyn wishes to include Truman's speech for a school project, she can freely.
The use of open formats allows Evelyn's work to be redistributed, on say a school DVD without legal rigmarole concerning patents for mp3 etc.
Sure, kids may not go to wikipedia to view videos of 'Dora the Explorer' or whatever flavour of the month entertainment the whipper-snappers access these days, but they might appreciate multimedia content for school projects etc.
And as far as ogg use personally, I not infrequently click on the sound files for pronunciation of foreign names or concepts just so I don't sound like a twit who can't even get a person's name right!
Comment removed (Score:4, Informative)
Flash 10 supports Speex too! (Score:1, Informative)
Many people don't seem to know this, but Adobe actually added the Speex codec to the Flash 10 player, allowing it to be used as an alternative to the proprietary Nellymoser codec.
Considering the market share of the Flash player, I would say that's quite an endorsement. Too bad they did't include Vorbis support as well.
But at the end of the day, the one way to let these promising completely Free codecs gain more traction is by convincing an influential company to adopt, sponsor, and champion them. Dirac has the advantage of being both developed and used (albeit internally for now) by the BBC, which is quite the juggernaut in the media and broadcasting world. So high is BBC's influence, that they successfully proposed the intraframe-only variant of Dirac (named Dirac Pro) to be accepted by the SMPTE as an official codec with the designation VC-2.
Since two is higher than one, people will then automatically think "Hey, the number is higher, so VC-2 must be better than Microsoft's VC-1 codec". :) Hopefully, SMPTE will eventually also allow the fully-featured (not just intraframe-only) version of Dirac to be ratified as an extended part of that standard, perhaps under the name "VC-2 plus" or "VC-2 extended profile" or something.
We'll see how that goes. Although the new Thusnelda implementation will be improving the quality and performance of the Theora codec, I believe the more state-of-the-art Dirac codec really remains more promising and has much more room for further improvement, especially in the long term. I hope Mozilla will take Dirac video codec at least as seriously as Theora. Ideally, Firefox would be shipped with the following integrated codecs ready to be used with the video-tag: MPEG-1 (ubiquitous and all its patents seem to have expired), Theora, and Dirac. For audio, it should ship with PCM, Speex, Vorbis, and FLAC support.
Re:Eh...what? (Score:3, Informative)
The best I can give you is hear-say -- I know someone who worked at our community radio station, and they did listening tests. MP3 is definitely the worst of current-gen codecs -- mostly because it's not a current-gen codec, it's a last-gen codec that people can't seem to let go.
So, Vorbis definitely sounds better, at the same bitrate.
I can't speak to CPU. At the time it was developed, it used a lot of CPU -- the name "ogg" [xiph.org] came from the brute-force techniques that were used, which pushed that 486 hardware to the limit.
Did I mention, 486 hardware?
Oh, by the way: You're thinking Vorbis, not Ogg. Ogg is a container format. You could just as easily put an mp3 stream in there, or put a vorbis stream in an mkv.
And the donation was not just Ogg, but also Vorbis, Theora (video), and others.
I would say, if ogg became as successful as png, I'd be happy. Right now, png is successful enough that I can use it in a website, and every browser will support it. Even if plenty of people still use gif, or even jpeg inappropriately (for images that compress well). And then you have the brand-new applications -- HD-DVD seemed to prefer PNG images for the menus.
Re:"Better" is relative... (Score:2, Informative)
If you want to play Ogg audio files, I've found cheap, unbranded 2GB MP3 player/FM radio/USB stick S1MP3 [wikipedia.org] combos available on Ebay will also play them. It's usually not documented in the instructions nor in the seller's online listings, but they play Ogg just fine. I've bought three of these units so far and they all did Ogg. In Linux you just mount them like any generic thumbdrive and transfer audio files to them.