Ogg and the multimedia container format struggle [LWN.net]

April 7, 2010

This article was contributed by Nathan Willis

Video codecs attract most of the attention in the multimedia format wars, from Theora adoption in HTML5 to debates about the subjective quality and objective technical demands of Dirac versus H.264. But the oft-overlooked container format is just as important; it adds overhead, it determines seekability, subtitle support, and other important features, and it can introduce patent-licensing issues for open source projects. Xiph.org's Ogg container format is the most well-known in open source, though as recent events show it has its critics and its competition.

Ogg has been under development since the beginnings of Xiph.org in 1994, and was originally designed for use with the Vorbis audio codec. As the Xiph project undertook additional codecs, Ogg continued to evolve to support them. FFmpeg developer Mans Rullgard posted a lengthy criticism of Ogg on his personal blog in March, accusing the format of falling short in six areas: poor generality, excessive overhead, high end-to-end latency, lack of random-access seeking, ill-defined timestamps, and unnecessary complexity. Rullgard cites examples from the specification and several "typical usage" numbers to support his claims.

Ogg debates

As the blog post was picked up, debate about the merits of the complaints quickly erupted in the comments and on web discussion forums. Several commenters chided Rullgard for claiming that Ogg's latency and seek times were unsuitably "bad" without citing any numbers from other formats for comparison, and for overstating the size of the problem (such as the overhead created by Ogg's headers at close to 1%) or its relative importance.

On some points, there was more of a genuine disagreement on principle, however. Rullgard said that Ogg wastes space by using a full byte for the "version" field, where a single flag bit would suffice. Xiph's Greg Maxwell explained in the Reddit discussion of the article that a byte is used for the field in order to keep the header structure byte-aligned for simplicity. Maxwell also disagreed with Rullgard's assertion that Ogg's 32-bit checksum was a waste of space, noting that Ogg also uses a 32-bit capture pattern at the beginning of each page, as opposed to the 64-bit capture pattern in FFmpeg's NUT format — thus using the same number of bits, but providing error-detection "for free."

Rullgard also argues that Ogg's ability to concatenate different streams one-after-another creates undue complexity for the decoder, without providing any practical benefit. But one blog commenter claimed to take advantage of this feature when ripping CDs with seamless track transitions.

Xiph's Christopher "Monty" Montgomery replied at length in the Slashdot discussion of the critique, admitting that Ogg has its flaws, and conceding that several design decisions made years ago would be made differently today, but attributing more of Rullgard's complaints to long-standing bad blood between the projects. Moreover, he said, even with its flaws, Ogg remains the best free option for important cases like streaming video. Neither the popular MP4 nor Matroska container formats are well-suited for streaming (particularly live streaming), and MP4 is also patent-encumbered. Additionally, he said, making changes to the Ogg format as suggested by Rullgard might improve performance at high-bitrate video, but would be detrimental to low-bitrate and audio payloads where Ogg excels.

Further work

Montgomery said that after the Rullgard blog post gained attention, Xiph decided that part of the problem with its reception was poor documentation on the Ogg format. He subsequently began rewriting and expanding the documentation, some of which is already available online. There are changes that Xiph would like to make, he added, as well as ongoing work in the metadata layer. "One of the legitimately weird things about Ogg is that we knew metadata was going to be a source of constant flux, so we moved as much as we possibly could out of the container itself. The Ogg container only does framing and delivery. [ ...] Most folks are used to this being part of the container, and so consider it 'part of Ogg' which it isn't really."

The Ogg Skeleton project is the primary focus in this area. Skeleton is essentially a "metadata track" that can hold information like MIME-types, protocol messages, and timestamps to allow the decoder to easily seek within the media. A Skeleton track could then be multiplexed or interleaved within an Ogg container file, alongside video and audio tracks.

Skeleton's timestamping capabilities are documented at the Ogg Index page, and are introduced in Skeleton 3.3. A sample indexer called OggIndex is available, and both the ffmpeg2theora converter and development builds of Firefox support it.

Montgomery concludes his Slashdot comments by noting that breaking compatibility with the existing hardware and software Ogg decoders (most of which see only Vorbis and Theora content) is probably not going to happen until the next major new codec release from Xiph.org.

The competition

Regardless of whether one finds any or all of Rullgard's criticisms valid, there are other container format options out there for content providers. The most popular on the Internet writ large is the .MP4 file, which is properly known as MPEG-4 Part 14, and was approved by ISO as ISO/IEC 14496-14:2003. A part of the larger MPEG-4 specification, MPEG-4 Part 14 is a revision of two earlier standards, MPEG-4 Part 1 and MPEG-4 Part 12. Part 14 is based on the QuickTime container format created by Apple and recognized by the .MOV file extension. It can hold content in any codec (including free codecs like Vorbis and Theora), although there is a "registered" codec list maintained by the MPEG.

There is a degree of uncertainty regarding the ability to write MPEG-4 Part 14 decoders, however. The rest of the MPEG-4 specification, like all MPEG standards, is patented, and implementations must adhere to the license terms made available by the MPEG-LA licensing authority. Part 14 was once available as part of the MPEG-4 Systems patent pool, which has subsequently been withdrawn. Many people on mailing lists and discussion forums assume that the format is free to implement since it is not explicitly mentioned in the remaining MPEG-4 patent pools: "MPEG-4 Visual" and "AVC/H.264," but this is not officially stated. MPEG-LA makes it difficult to find specific information about specific patents in its technologies, preferring instead to steer all customers into the "patent pool" products instead. The ISO specification, which should document specific patent claims, is only available to paying customers. When asked, MPEG-LA representatives said that they did not know the specific status of Part 14 in the current patent pools.

The Matroska format, like Ogg, was created to serve as an open, patent-unencumbered container. The two formats do differ in emphasis, a fact that both projects readily acknowledge. Whereas Ogg was designed alongside Vorbis with streaming audio as its primary use case, Matroska was designed to support as many codecs as possible. Xiph.org says that Matroska has better support for seeking, editing files, and using menus and chapter markers, while Matroska says that Ogg is superior at streaming media delivery (for example, Matroska only recently added support for interleaving frames from different tracks).

Matroska was launched in 2002 as a derivative of the older Multimedia Container Format (MCF). The copyright on the specification and the trademark of the name Matroska are both held by CoreCodec, Inc., but the specification is available free-of-charge. A reference library is available for download under the LGPL, and a "core parser" is offered upon request under the BSD license. The format is generally seen with the .MKV file extension for video content, although .MKA for audio is also valid.

The NUT file format Maxwell mentioned on Reddit was created by developers on the FFmpeg and MPlayer teams, but appears to be supported only in that project. The NUT project site is sparse, with a broken link to the actual specification, but there is a mailing list that indicates that development is still underway, albeit slowly. Montgomery described it as very Ogg-like in its design, incorporating some design choices that would improve Ogg, such as a simpler way of encoding the packet-length in each header (which was one of Rullgard's complaints).

Container formats are far less exciting than multimedia codecs, but the choice of containers has a very real impact on what a content provider can do. Quickly and accurately seeking within a file — while important — is just one example; another active topic right now is support for subtitle tracks. As multimedia content on the Internet grows, having subtitles accessible in their own track (or tracks, with multiple languages supported) has implications for accessibility, internationalization, and subtitle-based searching. For the record, Ogg, MP4, Matroska, and NUT all support subtitles.

As usual, the right choice depends on the usage. To some, non-free formats like MP4 ought to be avoided at all costs, even if MPEG-LA is not likely to request licensing fees. If streaming, audio-only, or low-bitrate performance are important, Ogg remains the simplest and probably best option. For seeking, video editing, menu/chapter support, or combining a wide array of codecs, Matroska offers functionality Ogg cannot. Moving forward, the relative weight of those factors may shift as either the codecs or the container formats evolve — but until then, choice is good.

(Log in to post comments)

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 14:57 UTC (Fri) by xiphmont (guest, #58693) [Link]

You're right, I was utterly wrong on this. I asked a developer who works on both Ogg and Matroska about the timing of when interleave was added to matroska (as it wasn't a feature in Matroska at the very very beginning; I had not followed the design closely since then) and his answer was simply incorrect. I'm not naming names becuase he didn't mean to mislead me.

>Other than that, I think the whole discussion about HTML5 video
>over-emphasizes the role of live streaming.

Actually, this is 100% of what I'm personally using it for right now. Though I agree with you it probably won't be the dominant use for other people. Live streaming is *essential* as a feature, especially as it's not actually hard and both Ogg and Matroska can do it.

> Another thing is that technically there's absolutely nothing that would
> stop people from successfully using Matroska for live streaming as well.

Correct, but does it exist right now and are people using it? The answer to both is 'yes' for Ogg, and every Ogg implementation can stream because it's impossible to implement the format without supporting streaming, both encode and decode.

I will argue that Ogg is better suited for streaming in several ways, though you're right that nothing fundamental prevents Matroska from being used this way. It will take more work, and requires the files be assembled for streaming, as in mp4.

>I honestly don't know why people would support OGG over Matroska for web
>video.

I submit it is because Ogg actually _implemented_ streaming, and it was functional, ready, and already being used in http streams when the major web browsers looked around. Folks have been streaming Ogg for ten years. It works well. It also comes in a package with codecs anyone can ship freely.

No one is streaming Matroska because no one ever implemented something that streams it (as far as I know).

In 1998 or so, Xiph had a disastrous presentation at Apple where we were showing off (at that time) our brand new next-gen codec engine 'Stormbringer'. Stormbringer was actually a big DSP virtual machine in which you could write other codecs, and the codec bytecode was packed into the stream header. Peter Hoddie (then part of the Quicktime group) asked if we could write an mp3 implementation in the Stormbringer bytecode. We said 'yes'. He asked if we had done so. We said 'no'. He asked us if any comparable codecs in Stormbringer were ready to go. We said 'No, but we can implement any codec we want." Peter said, approximately, "we're not interested in what you can do, we're interested in what you already have, and you don't have anything we can use." That was pretty much the end of the demo.

In short, don't complain, *go write code*!

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 15:22 UTC (Fri) by Daiz (guest, #65138) [Link]

>Correct, but does it exist right now and are people using it?

Yes, it exist right now, but not many people use it. Implementing it in browsers wouldn't really require more than supporting the container in the first place. VLC can already stream Matroska files just fine and technically speaking any other media player capable of playing Matroska files should be able to as well (I haven't tested this myself, but judging from the way it works there should be no problems).

>It will take more work, and requires the files be assembled for streaming

Not really. Basically it just means that your file is muxed without an index, and even if it was muxed with an index you should be able to play it just fine if it were being live streamed, while missing the index. The only extra work involved is setting up the live stream, and you have to do that with OGG as well.

In general, I'd say that the main reason why people aren't using Matroska for live streaming on the web is that web players for Matroska don't really exist. Flash doesn't support it, and there's no Java plugins or anything. Though I guess that VLC/MPlayer plugins in browsers should technically be able to handle live Matroska streams just fine. HTML5 <video> is a great opportunity to bring Matroska to the web. Just because other things have been used earlier does not mean that better (and completely free!) options couldn't be implemented now.

And to clarify what I mean by saying that Matroska is better: It's a better overall solution. OGG might be equally good if not slightly better for live streaming, but considering all user scenarios (including local file playback), Matroska wins hand down. It's features and software/hardware support simply blows OGG out of water.

Once again, live streaming is a very minimal part of video on the web, and choosing OGG over Matroska because it might be slightly better at it is simply put a stupid idea in my opinion.

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 15:55 UTC (Fri) by xiphmont (guest, #58693) [Link]

>any other media player capable of playing Matroska files should be able to
>as well (I haven't tested this myself, but judging from the way it works
>there should be no problems).

In reality many have problems because they want an index. If there's no index, the first thing they do is try to load the whole thing to build an index. It's both unfair and fair to point this out; I had predicted way back when that the same thing would happen to Ogg if we built in an optional index, and that's actually a driving reason we didn't. My predictions are often wrong (see: Vorbis and ARM DSPs) but this one I think was correct, because it played out that way for Matroska.

>The only extra work involved is setting up the live stream, and you have to
>do that with OGG as well.

Setting up an Ogg stream is as simple as an apt-get or yum invocation as icecast talks Ogg natively out of the box with no tweaking. Add one gstreamer hotplug script and all you have to do is plug in a DVcam and you're streaming without a single keypress. Gstreamer will sink anything it does to shout/icecast. There are other sources for live and canned streaming too. It could be far far easier yet (GUI apps), but Ogg has still got a substantial practical head start.

>Just because other things have been used earlier does not mean that
>better (and completely free!) options couldn't be implemented now.

Someone both has to implement it and then convince others that supporting it is worthwhile. What can Matroska do that Ogg can't? What does Matroska do right now that Ogg doesn't? Is there any benefit to adding Matroska too or is it just because we can?

The browser folks are concerned about code size, FWIW, because their big push is mobile. Implementing a larger, more complex, duplicate system is unlikely to interesting unless there's some killer feature Ogg is missing. This is, eg, the reason Real 'supported' Ogg but was never willing to ship it as a core codec; it merely duplicated what they already had.

>And to clarify what I mean by saying that Matroska is better: It's a
>better overall solution.

Is it? If we're talking only in theory, there's nothing Matroska offers as a complete system that Ogg [including the Skeleton] doesn't also offer as a complete system. If we're talking about what exists right now, Ogg has already implemented all the core uses that Web video currently cares about.

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 17:41 UTC (Fri) by Daiz (guest, #65138) [Link]

"In reality many have problems because they want an index. If there's no index, the first thing they do is try to load the whole thing to build an index."

They might want an index (because of faster seeking), but do not require it. For example, playing an index-less MKV in MPC-HC will result in no index being created until you seek. Considering that during live streaming you don't seek it makes this a complete non-issue (and makes MPC-HC perfectly ready for watching live-streamed Matroska content). The reason why for example MPC-HC creates an index upon seeking is simply because its main purpose is to act as a local media file player. If you were dealing with a streaming client, you could do things differently. Bottom line is that index in Matroska is completely optional and thus if you want to support Matroska you need to support index-less Matroska files too.

"Setting up an Ogg stream"

Setting up a MKV stream is quite simple with VLC too. Same goes for setting up an OGG stream with it. OGG might have more support within certain streaming solutions because of the amount of OGG Vorbis audio streams, but how many of these support streaming Theora video in OGG as well?

"What can Matroska do that Ogg can't?"

Are you seriously asking this? From the top of my head:

Native MPEG-4 AVC/H.264 support
Native MPEG-4 ASP support
Native AAC support
Native AC3 support
Native SRT support
Native ASS support
Standard chapter support
Advanced chapter magic (like ordered chapters and segment linking)

These are things that basically all proper software Matroska players support. Some features have varying degree of support (like advanced chapter magic) in software, but free and open solutions are available for all platforms. In terms of hardware support, advanced chapter magic tends to be mostly unsupported (though usually resulting simply in things like external segments not playing while the rest of the file plays fine), second being ASS. Some players, like the Popcorn Hour, offer limited ASS support though. No hardware player supports it fully though.

These are all formats that people actually use on the internet and for their personal media files. Besides supporting these, Matroska also has what probably equates to the best toolkit for working with a container format ever, mkvtoolnix. And it's free and open source too!

"Is there any benefit to adding Matroska"

Yeah. Supporting files people actually make and use, instead of forcing them to convert their files just to be web-compatible and nothing else. That nothing else part is pretty notable, because outside live streaming OGG doesn't really have anything to offer over Matroska for users. For example, among people who make digital copies of their DVDs/BDs/video content (aka doom9 users), only two preferred OGG as the container to use in 2008's "favorite video container" poll. In last year's poll, that number dropped to zero while Matroska's support percentage grew even larger. I think this alone shows pretty clearly that outside web usage, Matroska is the number one container of choice.

because their big push is mobile. Implementing a larger, more complex, duplicate system is unlikely to interesting unless there's some killer feature Ogg is missing.

Well, considering that there are PMPs out there that play Matroska files among other things, I doubt being mobile would be a problem. And since Matroska can do everything OGG can and then some, there'd be no real need for browser vendors to write support for OGG at all.

Also, I have a question: Is HTML5 video with Theora in OGG used anywhere for live streaming? I know that Vorbis in OGG is used a lot for audio streaming and I have absolutely nothing against that, hell I've listened to plenty of OGG live audio streams as well. However, I did none of that in my browser, I always used my audio player for that. Foobar2000, to be specific. Now, if someone were to start a Matroska audio stream in VLC and stream it over the internet, foobar2000 would play that equally well, since it's one of the few audio players to support Matroska.

But when it comes to live video streams, I honestly don't see much of it in general. So far I have never watched a live Theora&Vorbis in OGG stream. I've watched some FLV livestreams in VLC. I've also watched some livestreams in my browser using Flash. If you know of any site that would do live streaming using Theora, OGG and HTML5 please let me know. So far I've only seen HTML5 video used for the usual progressive download type videos.

Also, I have to say that in discussions like this, I'd wish that you people refrain from using comments like "OGG is better for streaming than Matroska", because it's very misleading. While technically speaking it might have some merit to it, it becomes false advertising when your average user confuses it to YouTube-like progressive downloading that is nowadays dubbed "streaming", where OGG provides absolutely no benefit over Matroska.

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 19:13 UTC (Fri) by xiphmont (guest, #58693) [Link]

> They might want an index (because of faster seeking), but do not
> require it. For example, playing an index-less MKV in MPC-HC will
> result in no index being created until you seek.

OK, using HPC as an example, you won't be able to seek in a
canned stream unless MPC downloads the whole things first and
builds an index. So, you're sort of answering a different
question, but it still makes my point.

> Bottom line is that index in Matroska is completely optional and
> thus if you want to support Matroska you need to support
> index-less Matroska files too.

You've completely missed what I'm trying to say. That point is
that we avoided having 100 optional features in Ogg for a reason.

Given that most of the Matroska spec is optional, it's a complete
guess as to what will be deployed and what won't, and plenty of
demuxers aren't even implementing the mandatory parts. I believe it
is fair to mention this because the spec is something of a kitchen
sink. It's huge. So much so that implementation in the real world
is a free-for-all. Is anyone contacting these non-compliant Matroska
implementations? This is something Xiph does with Ogg.

> OGG might have more support within certain streaming solutions
> because of the amount of OGG Vorbis audio streams, but how many
> of these support streaming Theora video in OGG as well?

Nearly all of them. Caertainly all the ones I can think of right now.

> Are you seriously asking this? From the top of my head:
>
> * Native MPEG-4 AVC/H.264 support
> * Native MPEG-4 ASP support
> * Native AAC support
> * Native AC3 support

You do understand that we do not and will not promote or advocate any
encumbered codecs as a policy that is coded into our charter, correct?
Strictly speaking, we're not allowed to.

I also don't think it's useful to try to convince Web vendors to
support unencumbered media by advertising your support for encumbered
codecs.

> * Native SRT support
> * Native ASS support
> * Standard chapter support
> * Advanced chapter magic (like ordered chapters and segment linking)

Are you seriously suggesting Ogg can't do this? Or perhaps more
relvantly, suggesting that having multiple ways to do something is
inherently superior to having one way everyone agrees on?

There is technical and usability value to vertical integration.
Just ask Apple.

> In terms of hardware support, advanced chapter magic tends to be
> mostly unsupported (though usually resulting simply in things
> like external segments not playing while the rest of the file
> plays fine), second being ASS. Some players, like the Popcorn
> Hour, offer limited ASS support though. No hardware player
> supports it fully though.

We could add a whole bunch of stuff that's only marginally
supported in the real world to our spec too. I think the fact
that we don't is an asset, not a liability.

> Yeah. Supporting files people actually make and use, instead of
> forcing them to convert their files just to be web-compatible and
> nothing else.

Yeah, that totally sunk YouTube. No one I know uses it.

But seriously, you're being thoroughly disingenuous. "And
nothing else". In what way is Ogg only deployed on the web?

I have a nice library of Ogg movies. Naturally, I rip my own DVDs to
Ogg. I'm starting to produce my own vids in Ogg. They play in all my
software players, under Linux, MacOSX and Windows. They also play in
all my browsers. They play on my phone.

Matroska has support in some DVD players, this is true... assuming you
use codecs that those players actually have available. Here's another
example of "100 ways to do something" doesn't mean any of those ways
actually work in practice. It would be like 90% of Perl's features
being optional.

And since you're advocating encumbered codecs anyway... how is using
Matroska different from just using the MP4 container or FLV? They have
far wider deployment than Matroska.

Ogg is offering a complete, integrated, fully unencumbered media stack.
There's value in 'integrated' and 'fully unencumbered' being together in
one place. Matroska does not offer that.

> That nothing else part is pretty notable

You are living with your head in the sand.

> Also, I have a question: Is HTML5 video with Theora in OGG used
> anywhere for live streaming?

Yes. For example, RedHat is using it internally for video
conferencing meetings between international offices.

> Also, I have to say that in discussions like this, I'd wish that
> you people refrain from using comments like "OGG is better for
> streaming than Matroska", because it's very misleading.
> While technically speaking it might have some merit to it, it
> becomes false advertising when your average user confuses it to
> YouTube-like progressive downloading that is nowadays
> dubbed "streaming", where OGG provides absolutely no benefit over
> Matroska.

...so you're saying "sure it might be true, but don't do it
because I don't like it"?

Ignoring everything else, Ogg streaming has one very obvious,
very visible benefit for users: It actually exists.

[Also, please, 'Ogg' is not an acronym. It's a proper name. No
allcaps :-) I'd kinda hoped you'd pick up the hint and I wouldn't
have to say anything because I always feel like a prick when I
point it out...]

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 20:07 UTC (Fri) by Daiz (guest, #65138) [Link]

>canned stream

What exactly is a canned stream? Sounds an awful lot like progressive download to me. MPC-HC might not do that fine if you try to seek to a part that hasn't downloaded yet, but then again it's purpose is not to be a web player. It'd be quite trivial to have your web player a) download the index separately b) not create an index and have a bit slower seeking. Once again, for playing actually live streams (which we have been mostly talking about) that you can't seek because they're live it'd work perfectly fine.

>You do understand that we do not and will not promote or advocate any encumbered codecs as a policy that is coded into our charter, correct?
>Strictly speaking, we're not allowed to.

Yes, and I'm saying that it's a serious disadvantage for the Ogg container. For example, I couldn't just take my DVD right here and mux all its streams into an Ogg and have it work. With Matroska, I could, and it'd be pretty darn simple to do as well.

>I also don't think it's useful to try to convince Web vendors to support unencumbered media by advertising your support for encumbered codecs.

Maybe if they used system playback capabilities to play back HTML5 video they wouldn't have to. As a result, there'd be no licensing costs to worry about and users could use the formats they normally use.

>Are you seriously suggesting Ogg can't do this? Or perhaps more relvantly, suggesting that having multiple ways to do something is inherently superior to having one way everyone agrees on?

So far I have not seen SRT or ASS muxed into Ogg and have it work. Ogg's own subtitle format Kate is about on par with SRT on a technical level right now and support is miniscule. Meanwhile, tons of hardware players support SRT subtitles, and so does basically every software player ever that has any kind of subtitle support.

AS to when it comes to chapters, I seriously doubt Ogg has anything similar to ordered chapters, segment linking and editions that Matroska has. I've used these features on some of my personal media files and in certain scenarios they are indeed useful. What's the situation with Ogg and chapters anyway? Does Ogg incorporate a standard chapter format and does anything support it?

>We could add a whole bunch of stuff that's only marginally supported in the real world to our spec too. I think the fact that we don't is an asset, not a liability.

It'd be pretty hard to do anything that wouldn't be "marginally supported" considering that Ogg is a marginally supported format to begin with. Any proper software player on every platform supports these features perfectly, the problems lie in the hardware players for these features. Since quite a lot of content is made to be consumed only on a PC, nothing would stop you from using them there successfully.

>Yeah, that totally sunk YouTube. No one I know uses it.

Except that YouTube converts everything you throw at it. My pretty MKV files too! Of course if you have a service like YouTube, it's relatively trivial what format it uses as long as users can upload anything they have to it. But with HTML5 video, the scope will obviously be larger: People will want to upload content on their own sites as well. If they'd use MKV for all their media purposes, they'd have to go the extra mile of converting their files to Theora+Vorbis in Ogg. Alternatively, if they used say H.264 & Vorbis in MKV they could potentially just directly upload it to their site, insert a <video> tag and have it work. And before you go about "H.264 licensing costs" I'd like to remind that using H.264 for freely distributed web video is completely free for the next 5½ years. Alternatively, they could use Theora & Vorbis in MKV too if they wanted, and maybe throw some SRT subtitles on top of that too. Or ASS, if system playback were to be used in the browser.

>In what way is Ogg only deployed on the web?
>I have a nice library of Ogg movies. Naturally, I rip my own DVDs to
Ogg. I'm starting to produce my own vids in Ogg.

Well gee, who would have thought? Xiph people using Ogg Theora! I'd just like to point you to the group of people known as "rest of the world" who use software like Handbrake and similar to rip their DVDs and who would have ever guessed, most of those recommend using H.264 and MKV! And software like Handbrake flat out just don't support Ogg. Fact is outside web normal people don't use Theora. They use H.264 and MKV for their personal stuff. The people who does this kind of stuff seriously also largely favor MKV, as evident by the doom9 container polls. Outside web, Ogg and Theora simply have no benefits over other, better formats.

>Matroska has support in some DVD players, this is true... assuming you use codecs that those players actually have available.

Well, guess what they usually support? High Profile H.264, AC3, MP3, maybe even Vorbis, AAC, SRT... all formats that people normally use as well. I don't think I've yet to run into any hardware player that supports Matroska yet doesn't support High Profile H.264 at Level 4.1. DivX 7 is also helping users in this aspect.

>how is using Matroska different from just using the MP4 container or FLV? They have
far wider deployment than Matroska.

Newsflash: They suck compared to Matroska. MP4's subtitle support is laughable at best (which is the biggest deal-breaker for me) and it doesn't natively support AC3 either. Chapters are also generally a nonstandard feature. FLV is even more restricted, I don't think it has any soft subtitle support at all. I recommend Matroska because it is better. Being free and open source just adds to that.

>Matroska does not offer that.

Yeah, it might not, but then again what would stop from using Theora+Vorbis in MKV? Nothing, and you could throw some widely supported (and equally patent unencumbered) chapters and softsubs on top of that too.

>You are living with your head in the sand.

I would say the same about you if you seriously suggest that people outside Xiph would use Ogg and Theora for their personal media encoding.

>...so you're saying "sure it might be true, but don't do it because I don't like it"?

No, I'm saying that it can be easily misinterpreted and in that way end up as false advertisement. And doing false advertisement certainly doesn't improve the image of Xiph.

>Ogg streaming has one very obvious, very visible benefit for users: It actually exists.

Yeah! Now I can watch 0.1% more video content on the web!

>Yes. For example, RedHat is using it internally for video conferencing meetings between international offices.

Can you point out any examples that I (or anyone else save for RedHat employees) could watch and prove to be working? Just hearing you say "yeah it totally exists" doesn't really cut it.

I'm streaming Ogg Theora very often

Posted Apr 10, 2010 9:08 UTC (Sat) by Velmont (guest, #46433) [Link]

I'm right now making my living streaming Ogg Theora using HTML5 in Firefox 3.6+.

I'm obviously not streaming 24/7, only when there is a conference, but I'll probably do it on the 15th. Yesterday I streamed the whole day to test how much load Icecast2 used for streaming Ogg Theora to a bunch of clients (it was almost nothing, so I don't need more servers :) )

Norwegian Unix User Group also streams all their member meetings. Linux Audio Conference does streaming in the same way, and a bunch of others. In the coming months the number will only increase.

Ogg and the multimedia container format struggle

Posted Apr 11, 2010 8:33 UTC (Sun) by njs (guest, #40338) [Link]

> And before you go about "H.264 licensing costs" I'd like to remind that using H.264 for freely distributed web video is completely free for the next 5½ years.

And *after* 5½ years money won't exist at all, because the Singularity will have arrived! 5½ years is so far in the future that taking anything about it into account when making decisions now would just be *irrational*!

(Out of curiosity, when is your cut-off for the future being relevant? 5 years? 3 years? 1 year? next week?)

Ogg and the multimedia container format struggle

Posted Apr 12, 2010 12:57 UTC (Mon) by nye (guest, #51576) [Link]

>For example, I couldn't just take my DVD right here and mux all its streams into an Ogg and have it work. With Matroska, I could, and it'd be pretty darn simple to do as well.

Do you mean to end up with a single mkv file that includes all the menus etc, and is essentially indistinguishable from the DVD? A quick Google search didn't come up with any instructions to do this (admittedly it was *very* quick) - do you have any pointers?

Ogg and the multimedia container format struggle

Posted Apr 13, 2010 13:09 UTC (Tue) by robux4 (guest, #65101) [Link]

> Do you mean to end up with a single mkv file that includes all the menus
> etc, and is essentially indistinguishable from the DVD? A quick Google
> search didn't come up with any instructions to do this (admittedly it was
> *very* quick) - do you have any pointers?

Look for DvdMenuXtractor. It extracts all the meaningful info from a DVD file structure and prepares scripts to merge them back in Matroska with mkvtoolnix. It's not a 100% safe tool and is unmaintained. It was mostly a proof of concept, but VLC is supposed to play the produced files, including the DVD menus.
I don't suggest playing such files on the web though ;)

Ogg and the multimedia container format struggle

Posted Apr 13, 2010 19:51 UTC (Tue) by ironiridis (guest, #60586) [Link]

It's crazy, but I just rip the ISO files instead. :)

Ogg and the multimedia container format struggle

Posted Apr 14, 2010 11:37 UTC (Wed) by nye (guest, #51576) [Link]

I'll take a look, thanks.

Handbrake default container

Posted Apr 15, 2010 10:14 UTC (Thu) by bawjaws (guest, #56952) [Link]

> I'd just like to point you to the group of people known as "rest of the world" who use software like Handbrake and similar to rip their DVDs and who would have ever guessed, most of those recommend using H.264 and MKV! And software like Handbrake flat out just don't support Ogg.

I use Handbrake. In fact I'm ripping a DVD right now as I type this. Both the Normal and High Profile settings (they ripped out all the old presets and boiled it down to two) create H.264 in an MPEG-4 container. I personally don't know anyone who's ever used it for MKV and I'm unaware of what the benefit would be over the defaults.

Also, Handbrake did support Ogg, Ogm to be precise, though not ever very well. They just removed the Ogg and Theora support recently to focus on H.264. (I don't think this was particularly a comment on Theora, they removed Xvid too as part of the same narrowing of focus).

Ogg and the multimedia container format struggle

Posted Apr 13, 2010 13:01 UTC (Tue) by robux4 (guest, #65101) [Link]

> OK, using HPC as an example, you won't be able to seek in a
> canned stream unless MPC downloads the whole things first and
> builds an index. So, you're sort of answering a different
> question, but it still makes my point.

There are 2 things here. Either MPC-HC wants to use the index in a remote stream, and it can do by simply seeking in that stream (very possible in HTTP) or it blindly seeks in the stream wait for the next 4 bytes of a Cluster startcode and plays from there (I believe that's the only way of seeking in Ogg). While the former may seem to do too many steps, it's actually way more efficient.

> Is anyone contacting these non-compliant Matroska
> implementations? This is something Xiph does with Ogg.

We sure do, that's why there is a field to indicate the muxing code and software that produced a file. It's some form of advertising for coders and allow us to find out bogus files and have their creator fix it. It happened many times in the past.

> There is technical and usability value to vertical integration.

Well, the problem is that Ogg is good and meant for streaming. But it's inferior in everything else. So people would have to trade all the features found and common in Matroska just so that it is 100% designed for streaming. Plus the extra container overhead and bandwidth wasting that comes with it.

Also Matroska has plenty of features, the main ones being implemented everywere because userspush coders to have them. We don't have a certification system yet, but we're thinking about it. DivX may also have something in place to certify software/devices with DivX 7. But in the end the same fragmentation happen with your vertical integration. Or are you expecting (and forcing?) all web browsers to have to support Spex, FLAC, Skeleton and the other technologies that are "vertically integrated" withing the Ogg sphere ?

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 19:25 UTC (Fri) by gmaxwell (subscriber, #30048) [Link]

"They might want an index (because of faster seeking), but do not require it. For example, playing an index-less MKV in MPC-HC will result in no index being created until you seek." I don't know about MPC-HC, but in mplayer for example it's unable to seek without scanning the whole file to build an index. Even deferring this, scanning the entire file to build an index when the user seeks is the kind of unacceptable half-solution that was trying to be avoided. I'm not saying that MKV is doing anything wrong, it is what it is and the tools could be better.

The most frequent argument I hear regarding Ogg not having an index is that the lack of one makes seeking more complicated. But as you said, " if you want to support Matroska you need to support index-less Matroska files too". In any case, looks like it's going to be the same story for Ogg nowish.

As far as live streams go I have no idea, and can't find any information on how you'd build a scalable MKV streaming infrastructure to support a couple of thousand users, or even a hundred or so. This might just be a documentation issue. For Ogg you simply use icecast and can scale to any number of users by adding servers and an interior relay layer (and you can purchase icecast distribution services commercially). The same (icecast based) services that work for Vorbis also work for Theora.

There are some live Ogg/Theora / Vorbis streams in the icecast directory, but a lot of usage is private. Other recent examples include the students for free culture conference or the lessig wireside chat.

(FWIW, I've never been able to get VLC to produce an HTTP stream with MKV/Vorbis+Theora it just throws "invalid chain" and "no suitable sout mux module for `http/mkv". The same configuration, switching MKV with Ogg works fine. Though if you say it works I don't doubt that my configuration is wrong)

I would propose that 99% of the group X is using MKV is related to codec support and not especially related to the container. There are no useful and mature tools for things like H264+AAC in Ogg. There certainly could be but why bother?

I'd offer that adding more encumbered formats is utterly poisonous to everything that made the web so successful.... But in any case thats really a separate argument from the containers issue. But if you really think that H.264/AAC is a reason to pick a container than MP4 is the absolutely unambiguous choice there simply due to device compatibility.

My own view, and it's one that Monty has disagreed with in the past, is that that the lack of tools for putting encumbered codecs in Ogg is an _advantage_. It's an advantage because avoiding licensed formats is important to you, you don't have to be a media rocket scientist to figure out if your files are 'pure' or not. Simply check the extension: If it says "Ogg" (or Oga, Ogv) you're likely good to go.

Containers really don't contribute much to anything most users ultimately cares about. Of course, tool support for Ogg with this formats could be written but there are two acceptable formats for the AAC+H264 mix, arguably thats already one too many.

... and "mature support for encumbered codecs" is most of your feature list.

Considering the rest we do have subtitle support in the Ogg toolset (In the form of the Kate codec, which supersets SRT as far as I know...), but it's not part of Ogg itself.

Ogg does miss out on the chapter support. You can use chaining as chapters, of course. But there really isn't application support for this, so it's a largely theoretical use case. I've never seen an MKV file using chapters myself, but I don't doubt that it's an important feature in the movie "backup" market.

Whatwg has been really conservative about adding features to video the API. It's far from exposing all the things it could expose already, I honestly don't think it's likely that it ever would expose chapters even if implementers were shipping MKV support today. I think this is unfortunate, but it is what it is that community would rather you implement fancy features using HTML+CSS+JS rather than putting it into the video files themselves.

Don't get me wrong, I'd certainly be happy to see more MKV support. But the benefits are not clear to me.... and the embedded developers/browser vendors are incredibly conservative about dependencies. The LGPL licensing on libmatroska and libebml the 'optional' dependencies on things like bzip2,lzo, and zlib (which are really not optional if you want to handle anything a user might throw at it) are considered to be real hurdles in the browser and embedded space which will probably require more justification than "Support for H.264/AAC but less compatible than MP4" and good support for chapters/subtitles.

Cheers.

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 20:25 UTC (Fri) by Daiz (guest, #65138) [Link]

>In the form of the Kate codec, which supersets SRT as far as I know...

On a theoretical level, yes, on actual technical level, no. In that aspect it's about on par with SRT, except a lot less supported.

>... and "mature support for encumbered codecs" is most of your feature list.

For your average user backing up their DVDs or encoding their shortfilms or doing whatever it is quite important, since more often than not these people want to get the best quality they can get, in which case for video they turn to x264, the best video encoder in the world.

My main argument against Ogg in the web is that it separates content to web and non-web. I'd rather have Theora+Vorbis in MKV than Theora+Vorbis in Ogg, though more preferrably I'd use H.264+Vorbis in MKV for my web video needs if HTML5 <video> supported that properly in major browsers. I'm quite sure many people would be delighted about this too, since it'd basically mean that their personal copy and the web copy can be the exact same file without compromising quality.

>I've never seen an MKV file using chapters myself

And I've seen thousands and made hundreds. Quite literally. And I gotta say that I love chapters. I backup plenty of TV series I own and with chapter support I can easily skip for example the opening and ending themes of a show. With segment linking, I can even have the opening and ending in separate files and seamlessly link them to the episode files, which saves space.

In general I hate how ignorant the whole media industry seems to be of subtitles. Subtitles in DVDs look atrocious, even Blu-ray subtitles are more limited than they could be, MP4's subtitle support is a joke... it seems like no-one in the industry cares at all, which isn't a surprise considering it's mostly based in the US where subtitles are a rarity. Around my part of the world everything that's not for very little kids and is in a foreign language is subtitled. This is the case in quite many countries, and for people living in those countries, subtitles matter as well. Subtitle support needs to be taken more seriously!

How about you Xiph guys rather spend your time trying to kill software patents completely? I bet you'd manage to get that done faster than improving Theora encoders to the point that they beat x264, and then we'd have no need to argue about whether something is patent encumbered or not anymore.

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 21:34 UTC (Fri) by xiphmont (guest, #58693) [Link]

> My main argument against Ogg in the web is that it separates
> content to web and non-web.

Hmm, OK, I see that concern. But I already need to do that in that I
record and edit at high bitrate and have to reencode for reasonable web
rates anyway. I'm not going to stream at 25Mbit.

> I'd rather have Theora+Vorbis in MKV
> than Theora+Vorbis in Ogg, though more preferrably I'd use
> H.264+Vorbis in MKV for my web video needs if HTML5 <video>
> supported that properly in major browsers.

...but only if you don't have to pay for it! Ahhh.... somebody
does. Even in France.

> I'm quite sure many
> people would be delighted about this too, since it'd basically
> mean that their personal copy and the web copy can be the exact
> same file without compromising quality.

The day that MPEG-LA renounces its h264 patents, I'll dance in
the streets. The argument will be over, and everyone including
99.99% of businesses will have won. But I expect that not only
will MPEG find a way to keep milking it way past 2028, but by
then the entire patent pool will be pushing its new thing and
you'll be here arguing that we're not right in the head for
thinking Ogg NextGen can stand up against h265, and h265 isn't
really that much money, and everyone is using h265 anyway and you
can put h265 in MKV and not Ogg so Ogg sucks.

> How about you Xiph guys rather spend your time trying to kill
> software patents completely?

We'll get right on that.

> I bet you'd manage to get that done faster than improving Theora
> encoders to the point that they beat x264, and then we'd have no
> need to argue about whether something is patent encumbered or not
> anymore.

Ah, so I was being trolled all along. Oh well.

Ogg and the multimedia container format struggle

Posted Apr 10, 2010 12:20 UTC (Sat) by Daiz (guest, #65138) [Link]

>Ah, so I was being trolled all along. Oh well.

I wasn't trolling even one bit. Are you seriously implying that Xiph (or anyone else) could ever make libtheora beat x264 in speed and quality? If you are, get real. Honestly. That's just silly. Even if x264 development stopped completely right now, it still wouldn't ever happen.

Ogg and the multimedia container format struggle

Posted Apr 10, 2010 15:46 UTC (Sat) by bronson (subscriber, #4806) [Link]

Your previous comments had some merit but this one sounds like pure troll. The only point you make is that nobody could make libtheora beat x264, then you fail to support your position even a little bit? Textbook definition of fanboi troll. Very disappointing.

> it still wouldn't ever happen.

Dude, ever is a very long time. Speaking of getting real...

Ogg and the multimedia container format struggle

Posted Apr 11, 2010 9:03 UTC (Sun) by Daiz (guest, #65138) [Link]

Look. Video compression is a field that is patented to hell and back. As long as Theora aims to remain patent unencumbered, it has to avoid so many things that could make it better that its utterly ridiculous. x264 developers have to care about absolutely none of these.

Also, as it is, x264 development is a lot more active than libtheora development is. From the beginning of 2009, Theora has received about 42 updates. In the same time, x264 has had 440 revisions.

The only way libtheora could ever beat x264 pretty much requires that software patents stop existing, and at that point we wouldn't even need it anymore.

Thus, libtheora will never beat x264 in quality in speed.

Ogg and the multimedia container format struggle

Posted Apr 11, 2010 10:03 UTC (Sun) by gmaxwell (subscriber, #30048) [Link]

As far as I can tell you're arguing, quite ferociously I might add, with yourself.

Ogg and the multimedia container format struggle

Posted Apr 11, 2010 14:06 UTC (Sun) by Daiz (guest, #65138) [Link]

Yeah, because xiphmont and bronson are totally me, right?

Seeking in a live stream

Posted Apr 11, 2010 15:12 UTC (Sun) by kleptog (subscriber, #1183) [Link]

That's something I find quite baffling actually. I've got a Tivo here that has no trouble seeking within a live stream and did this 10 years ago. Pause/Fast-Forward/Rewind all within a moving window.

I have yet to see this on a PC however. Are there any media players which support this mode of operation?

Ogg and the multimedia container format struggle

Posted Apr 13, 2010 13:15 UTC (Tue) by robux4 (guest, #65101) [Link]

> I'd offer that adding more encumbered formats is utterly poisonous to
> everything that made the web so successful...

Well, this is like saying HTML should have never evolved from version 1 or that the javascript and CSS extensions are poisoning the web. AFAIK not all browsers are passing the Acid3 test and still people don't consider that their web experience is broken because of that. Not all browsers are going to add support for all Matroska support all at once, nor will they with Ogg. That doesn't mean it's not the way to go.

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 18:28 UTC (Fri) by bronson (subscriber, #4806) [Link]

> What can Matroska do that Ogg can't?

When this happened (search for OGG):

http://handbrake.fr/?article=10

It pretty much sealed Ogg's fate on my system. Not sure quite what the backstory is there, maybe just developer apathy toward Ogg. But, personally, I tend to go where the tools go.

Ogg and the multimedia container format struggle

Posted Apr 9, 2010 19:19 UTC (Fri) by xiphmont (guest, #58693) [Link]

As far as I know, Handbrake never supported Ogg. It supported OGM, which was a different system built using a modified Ogg container. You could put Xiph codecs in it, but most people didn't. You're better off using Matroska to do that anyway, so no actual loss.

Ogg and the multimedia container format struggle

Posted Apr 10, 2010 16:15 UTC (Sat) by bronson (subscriber, #4806) [Link]

Interesting, I had thought that they were the same. Who created OGM? Someone totally unrelated to Xiph? (the wikipedia page doesn't say)

I'm using Handbrake because it produces by far the best results (auto cropping, optional noise reduction and other filters, multiple audio tracks, chapter marks, etc) from anywhere (disc, iso, DV, file, etc) with a tiny amount of setup work on my part. Good queue management, etc. Is there any equivalent for Ogg?

Transcoding is a wasteland of half-finished, horrible projects. I feel lucky to have found Handbrake.

Congrats on the TheorARM grant BTW.

http://google-opensource.blogspot.com/2010/04/interesting...

Ogg and the multimedia container format struggle

Posted Apr 13, 2010 12:35 UTC (Tue) by robux4 (guest, #65101) [Link]

> I will argue that Ogg is better suited for streaming in several ways,
> though you're right that nothing fundamental prevents Matroska from being
> used this way. It will take more work, and requires the files be assembled
> for streaming, as in mp4.

If I understand correctly, by live streaming, you mean something that has no start/end, like a radio broadcast. Not necessarily something like IP multicasting. So if we agree on that, such a Matroska stream would be produced with no seeking info. Therefore any existing player would have no way of seeking (by nature it's live, so no seeking). I can tell you that both DirectShow based players and VLC have no problems with that. They can also handle an "infinite" segment which is basically the only requirement for a player to handle Matroska "live" streaming.

Now, like I said, there might not be softwares that produce such streams. But if there are any software that can produce such livre streams (which are special streaming softwares) in Matroska they would just have to make sure they use an "infinite" segment. I suppose it shouldn't be hard to tweak any Theora live streaming server and add Matroska support for it.