A lot was said lately about the Vorbis/Theora vs h.264/AAC situation on the draft of the HTML5. As some of you know, video is my main hobby these days (I care not about operating systems anymore), so I have gain some experience on the field lately, and at the same time this has made me more demanding about video quality. Read on for a head to head test: OGG Theora/Vorbis vs MP4 h.264/AAC. Yup, with videos. And pictures.First things first: the disclaimers.
1. This is not a scientific test. Both sides will be arguing about this arcane switch, and the other arcane switch, and “oh, you should have used another encoder”, this or that. I don’t wanna hear that. I use the most popular and free front-ends for these encoders: ffmpeg with x264/faac and ffmpeg2theora. FFmpeg is a command line front-end to many encoders and itself is used by many GUI front-ends too, so it’s one of the best utilities for the job. Ffmpeg2theora on the other hand, is the most popular Theora front-end too. I didn’t use Ffmpeg for Theora because that implementation is still buggy. FFmpeg2theora does a better job.
2. Tests were done on Windows XP 32bit because the Ubuntu version of ffmpeg does not bundle AAC for legal reasons. I used a May 2007 version of Ffmpeg (the latest version crashes repeatedly on Windows when libx264 is used on HT Intel machines), and for ffmpeg2theora I used the latest version, 0.20.
3. I used an HDV .m2t file as source file. My first idea was to use the lossless intermediate Huffyuv format, but it would be problematic trying to upload this to the osnews server, because the file size would be huge. So, I went with .m2t, I put two clips side by side on Sony Vegas Pro 8, and then I exported again as .m2t 1080/60i (interlaced). Vegas does not re-encode clips that were not edited, so this is as close to “straight out of the camera” as possible (only a few frames at the beginning and the end of each clip were re-encoded by Vegas). Download the source clip here (unzip it).
I did two tests: one for a widescreen QVGA format, and one for a 720p HD one. I chose these two sizes on purpose. One format will show us how the codecs perform in low bitrate, which is super-important for web usage, and the other one will show us how well the codecs perform with HD.
QVGA test
FFmpeg command line:
ffmpeg -deinterlace -y -i "orignal.m2t" -threads 2 -f mp4 -title "h.264 QVGA test" -vcodec h264 -level 41 -refs 2 -loop 1 -deblockalpha 0 -deblockbeta 0 -parti4x4 1 -partp8x8 1 -partb8x8 1 -me full -coder ac -subq 6 -brdo 1 -me_range 21 -s 320x180 -r 30000/1001 -b 320k -maxrate 450k -bufsize 5000k -g 300 -acodec aac -ac 2 -ab 128k "qvga.mp4"
Ffmpeg2theora command line:
ffmpeg2theora --optimize --deinterlace --title "Theora QVGA test" -f mpg -v 4.2 -F 30 -x 320 -y 176 -A 128 -c 2 -H 44100 -o "qvga.ogg" "orignal.m2t"
Download the h.264 video version here, Theora video here (right click to save).
– 
Two things of note here: Ffmpeg2theora would not let me use the resolution that keeps the correct aspect ratio when downscaled from 1920×1080 (anamorphic 1.3333). I had to use 320×176 (direct multiple of 8), instead of the correct 320×180. FFmpeg/x264 had no problem with that.
The second thing is something that really made me angry: Ffmpeg2theora will upsample the NTSC frame rate of 29.97 to 30fps. Well, I don’t want 30fps, I want 29.97. I am accustomed to view lots of my videos on my TV (usually by using a PS3), so should a Theora2TV device comes out, I want to be able to have smooth frame rate, without adding pulldown and all the problems that brings. No matter what trick I tried to do, Theora would keep adding pulldown to 30fps. Update: You can get to 29.97 only by either not specifying it at all, or via 30000:1001 (instead of 30000/1001 or 29.97).
Now, you probably noticed that I gave a “videoquality” of “4.2” to Theora, while I gave FFmpeg 320kbps video with a 450kbps peak rate (VBR). I did this because Theora would simply NOT do what I asked it to. It would not stay close to the requested 320kbps, and so it would create a 550kb file instead of a 640kb one. So I changed the Theora command line, by trying different VBR quality numbers, until I got an output file size that matched that of h.264’s. Remember, the important thing here to remember is THE WEB. Bandwidth matters. So it’s important to know what kind of quality you get from each codec, for the same file size (not necessarily the same bitrate).
As you can see from the tests, h.264 whooped Theora pretty bad on this test. Even when I gave the benefit of the doubt to Theora and re-encoded it in 400kbps, h.264 was still visibly much better.
720p
FFmpeg command line:
ffmpeg -deinterlace -y -i "orignal.m2t" -threads 2 -f mp4 -title "h.264 720p test" -vcodec h264 -level 41 -refs 2 -loop 1 -deblockalpha 0 -deblockbeta 0 -parti4x4 1 -partp8x8 1 -partb8x8 1 -me full -coder ac -subq 6 -brdo 1 -me_range 21 -s 1280x720 -r 30000/1001 -b 5000k -maxrate 6000k -bufsize 15000k -g 300 -acodec aac -ac 2 -ab 128k "720p.mp4"
Ffmpeg2theora command line:
ffmpeg2theora --optimize --deinterlace --title "Theora 720p test" -f mpg -v 4.3 -F 30 -x 1280 -y 720 -A 128 -c 2 -H 44100 -o "720p-theora.ogg" "orignal.m2t"
Download the h.264 cabac video version here, Theora video here (right click to save).
– 
Again, a quality number was used for Theora to match h.264’s file size at the requested 5 Mbps (with 6mbps peak rate). I did two ffmpeg passes, one with the CABAC entropy coding, and one with the CAVLC one (just change the “-coder ac” to “-coder vlc” in the ffmpeg code above). CABAC creates better looking video than CAVLC, but it’s harder to decode (Quicktime chokes good on it, for example, on my P4 3Ghz). I usually encode my stuff in CAVLC simply because I might find myself sharing my own videos to people with AppleTVs or slower PCs, so I am trying to play nice.
HD is HD, so even if Theora again performs worse than h.264, there is enough resolution that the differences might not as obvious as in the QVGA example. For example, I have an old CRT HDTV, which barely manages well 1080i, so the difference would not be much visible (heck, on my TV I barely see the difference between Comcast’s HD channels and progressive DVDs). However, people with big LCDs or Plasmas will be able to see more artifacts — should a device that playbacks Theora on TVs comes to market.
Another interesting tidbit about the 720p test is the encoding times! H.264 did 1′.15″ for CAVLC and 1′.20″ for CABAC on my PC. Theora took its sweet time and finished at 3′.14″. Granted, I used two threads on the FFmpeg command line, while it was obvious to me by monitoring my Hyperthreaded CPU that FFmpeg2theora was single-threaded. But even if you divide the 3′.14″ by 2, you get 1′.37″, which is still slower than h.264’s encoding times.
Now, a last point, CPU utilization during decoding. Of course, this is implementation-specific, but you know, you gotta test with whatever is popular, no matter if their implementation sucks or not. So I used VLC and a new version of Media Player Classic. VLC used on average 65% of my hyperthraded CPU on CAVLC, 75% on CABAC and 75% on Theora. MPC used 34% on CAVLC, 43% on CABAC and 33% on Theora. These were rough averages during multiple times of running each video.
Conclusion
I prefer h.264. Goodnight everyone.
Ok, ok, not yet goodnight. Truth is, I prefer open standards and open source and Free Software and what not, over proprietary royalty/patent-ridden solutions. But pictures (and videos) don’t lie. Theora is simply not at the same league as h.264. Vorbis fairs well compared to the big 3 audio formats, but Theora is visibly worse in comparison. By their own admission in the past, the Theora developers had to use less efficient algorithms to go around patents. This sucks of course, but it all comes down to this: do you want an open FOSS solution, or you want the best solution? Many readers on OSNews will choose the first option, but their teenager brother would probably choose the second.
Some say that Dirac is better, and it’s free/open. But until their developers actually do take the time and commit their implementation to ffmpeg (instead of only offering patchsets), it just doesn’t exist for me. FFmpeg — with all its problems… Come on, no automatic letterboxing??– is used by many GUI front-ends, so if they want market penetration, they gotta add their encoder to ffmpeg. Then, I will give them a whirl too.
I can understand why they would want an open standard for video on the web. However things change way too much.
Today the most common formats are wmv and flash for streaming and divx for downloading.
A few years ago it was .wmv and .mov for streaming and divx for download
Ten years ago it was .mov and .rm for streaming and .mpg for download
Setting a standard no one uses is just going to end up a joke in a few years when people are wondering why Firefox 5.0 is installing Theora codecs just to have full HTML 5.0 support when only a hand full of websites used the codec
to clear it up, the flash players can/are using h264, the current wmv was microsoft’s attempt at mp4, but h264 beat it, of course microsoft didnt embrace it and are doing a good job and messing up the media distribution online.
also theora as i understand was another competing codec that lost to h264 and only opened up after.
anyways, i don’t even really understand why there is debate for an official codec or object wrapper for HTML5, but i’d prefer it would be something standard instead of some obsolete and technically inferior codec that lost the mp4 race.
Although MPEG 4 ASP is by the far the most popular format for ‘downloading’. I dare say this is going to change soon and MPEG 4 ASP as whole has past is past its peak. If you look at any of the HD content on the web today it will all be in H264. Even SD DVD rips are starting to use H264.
P.S. Doesn’t .mov also use H264?
Depends how you export. You can put a lot of different formats on .mov, but h.264 is one of them, yes. Apple is pushing h.264.
MPEG4-SP/ASP has indeed passed its peak. Even the XDiV team is currently working on MPEG4-Part10 (which is nothing but another name for h.264), and the DivX Corp *purchased* MainConcept who have an h.264 implementation. As you understand, the main MPEG4-SP/ASP providers are moving to an h.264-like implementation too.
Microsoft’s WMV and VC-1 are h.264-like in many ways too.
Even on the mobile space, Nokia now has h.264 support on their Symbian S60 3.1 phones, while in the past they would only use h.263 or MPEG4-SP.
Edited 2007-12-12 07:16
it seems even today people are still confusing codecs with containers.
for instance, WMV is not a codec but a container the holds a video and audio codec inside.
also to be clear ,it needs stressing it seems:
Mpeg4-part2 is AKA divX/Xvid/ASP
(the old hat one)
Mpeg4-part10 is AKA AVC/H264
(the one EVERYTHINGS is using today or will be soon, from HD-DVD, HD-BR, HD/SD STBs, DVB, IPTV, web streaming, and personal end user multicast streaming, just use it….)
VC-1 infact has its roots in DivX and is just an extension of that, it is not good for low bitrates and it DOES NOT have a lossless MODE like AVC/H.264 does.
see this chart for more info
http://img297.echo.cx/img297/8742/compchart2fa.png
as you can see only AVC/H.264 and Snow are have lossless modes.
a container can be any of the common formats and you put your code(s) inside that, containers include anything from
.AVI, .MPEG, .TS (transport stream is a good choice), .VOB, .mov, .qt .WMV* .mkv .asf and more…
also dont get the options confused with the profiles levels, for instance keep in mind the Broadcasters wouldn’t use anything other than the Main or High Profiles at Level 3 (SD) or Level 4 (HD).
so its best if you too only use main and high today and into the future.
if your confused , just remember that AVC/H.264 (and its AAC audio equiv) is becoming the world standard and for a container your best bets are .mpeg, .TS , i favour this one as you can have many streams inside it and the editing/cutting tools are becoming better as time passes for it.
you could for instance do 2 or 3 encodes/transcodes
and include them all inside the one file, one an HD version, one an SD version and a PIP version for your wireless DVB VLC multicast LAN and ISP if they finally turn on the generic multicast for ipv6 and even the old ipv4 kit.
btw , can anyone remind me of the right sizes to use for CIF/PIP format but widescreen PAL AVC, i always forget that and is that size best/optimal at 500kbit/s or less?.
Edited 2007-12-17 07:56
H264 is the standard format for exporting videos in iMovie atleast, thought you can choose like 20-30 others aswell.
Sadly you are right…
Its nice to see some constructive criticism, maybe they can improve on the things that annoyed you.
Yes, h264 is superior to Theora. No doubt about it.
The patent-free alternatives to h264 would be Dirac (as mentioned in the article) or Snow, but they are not ready yet. Snow had a very promising start, but then it stalled. Dirac is progressing slowly. Currently it works, and has good quality, but it used way too much CPU the last time I tried it (a few months ago). It needs to be optimized before it’s really usable.
For anyone wanting to try Dirac there are gstreamer plugins for a Dirac implementation called Schrödinger:
http://schrodinger.sourceforge.net/
##EDIT##
There was a SoC project this year to implement a Dirac encoder and decoder for ffmpeg. The decoder is done and the encoder is on its way, so hopefully they will be merged soon.
Edited 2007-12-12 07:39
x264 is genius, everyone should be using it for everything with enough decoding power
Why does the HTML 5 standard say you must use this Codec or that Codec? Won’t the web in say 10 or 15 years be forced into using an old standard? Why can’t they just say this is a video stream and let the page author choose the Codec just like currently you can choose the embedded picture type?
It doesn’t (any longer): http://www.w3.org/html/wg/html5/#video0
Theora is already old and outdated today.
“must” was never used, “should” on the other had.
basically ogg for both audio and video was put forward as a kind of baseline. without it the media tags becomes a kind of joke, as one just moves the video controls out of the flash file and the plugins are still needed for the codecs.
but, as anyone could expect that ogg should at least be supported, everyone would be using it unless they had some kind of specific reason not to.
as for choosing? i would go for open vs quality any time…
hell, i suspect that most users would not care as long as they had a simple way to put that family movie online. or maybe that shaky party movie, taken using a low quality mobile phone or camera while drunk.
only issue then is that you really need to look around to find a chip that can encode ogg native. but you will find plenty of chips that will do mpeg/h264 in one form or other…
and i think thats whats getting the codec companies and hardware companies up in arms. they either loose market share (cant sell their IP to others), or have to find new chips to put into designs. designs that maybe have been on the drawing board for a year or more, waiting for the steam to run out of the existing ones.
ogg is a wild card in all this, and was suggested by the wild card in the browser field, opera.
thing is, they all know ogg would be acceptable for most users. just look at how much high def formats in audio and video stand vs divx and mp3 (to use known examples). i suspect that the non-technical consumer is tired of formats that give no no practical value vs the old ones. the cd is dead, long live the audio file, and there the mp3 have become ubiquitous (but i cant help wonder what would happen if you could get ogg support in those small flash devices from sandisk, or heaven forbid, the ipod. this not because of quality, but because you didnt have to run into alternative os’s that could not support the format because of some vague patent issues).
the dvd is dying a similar death. and for most, hd-dvd or blu-ray isnt interesting as its still a physical format. and one that requires the user to replace both player and display for it to be of any use. hell, is there not a physical audio media based on dvd? does anyone own, or know of someone that do, any copies in that format?
all im saying is that for most users quality comes second to ease of use. and whatever format that would be the base line for streamed media on the net would be king of ease.
and introducing a wild card, by someone that have no vested interest in either hardware or codec. they only produce a browser, for many platforms.
hmm, when i think about it i can suspect one reason for opera suggesting ogg, licencing. as it stands, they have to licence flash, windows media and maybe quicktime to make sure that they are covered when supplying a third party (like say nintendo) with a browser for their product.
with ogg they could probably show the middle finger to all of those, and keep the difference.
so i guess everyone have some kind of interest in the choice of codec chosen, and i fear that quality is the last of those.
‘thing is, they all know ogg would be acceptable for most users. just look at how much high def formats in audio and video stand vs divx and mp3 (to use known examples). i suspect that the non-technical consumer is tired of formats that give no no practical value vs the old ones. ‘
I think most users have no idea what formats are or care – nor should they have to. They just watch films, there not interested in OSS values, they just want to be able to view their films in what ever media player comes with their computers.
‘all im saying is that for most users quality comes second to ease of use.’
What research have you done on the subject?. Ease of use for most people would probably be that it runs in Windows Media Player or iTunes and that they don’t need to download or install anything else just to see the ‘shaky party movie, taken using a low quality mobile phone or camera while drunk.’
bingo, except here we are talking about a web format, and more and more people upload their small videos to pages like youtube for no other reason then to share it with friends and family via im, community sites or their blog/webjournal.
i see it first hand every day.
when the file is in the cloud, they dont care about windows media player or itunes, its in their browser it needs to work.
and right now, to get anything like this working there, every browser needs a plugin (unless your a IE user in windows or a safari user in osx, and you only browse pages that use a matching format).
thing is that the most repeated issue with ubuntu and similar is that one have to download some codec pack or other to get media going.
if ogg got a foothold on the web, then it may get the leverage needed to be used in larger numbers elsewhere.
at that point the codec argument on alternative os’s, while not going away fully, becomes greatly diminished.
hell, one can even start to ship computers that use “alternative” cpu’s. a arm laptop anyone? where the biggest drain on battery is the back light of the lcd?
yes, people dont care about oss. but they care about usability, and they care about cost. and right now oss has a advantage in the latter. but because of copyright and patents, not so on the former. if the former was to change, things could get highly interesting.
hell, many have moved their digital life into the clouds, or more specifically that google branded one. think about the impact if those low cost “gos” computers could “just work” with anything the web had to offer?
HTML 4.x is more or less agnostic towards picture formats [1] :
This laissez faire approach has caused some medium to major problems for web developers, customers and the software vendors behind browsers in the past ( GIF not unanimously acceptable due to software patent baggage, no png support in IE < 7, etc. ) and it is likely, that the W3C aims at a bit more specificity for future standards / recommendations, especially since the potential for troubles inherent in the current situation with several competing, incompatible ( and differently licensed) video codecs is magnitudes higher than for picture formats.
Note, that – to my knowledge – the current draft speaks of “SHOULD” and not of “MUST” wrt these multimedia codecs, e.g. vendors would be able to claim compability with HTML5 without supporting Theora ( So much for being more specific this time around 🙂
).
The situation is a royal mess, agreed. The strict guidelines of the W3C and their aim at vendor independence, cross platform and FOSS compability rule most established (properitary) codecs out. In the light of this commitment (and since it is safe to assume that the bulk of tomorrows video content will be delivered
via what we call now webpages while the current trend towards a higher diversification of the consuming software platforms is likely to continue) the HTML5 spec can not neglect the discussion of video formats without risking a fragmentation of the web along vendor lines (or facing little adoption, which is imho not unlikely).
On the other hand, we have the situation that the big boys (Microsoft, Apple, Nokia et al) have invested considerable amount of time, development man-hours, money and patent cross licensing deals to keep the
legal mine field underneath their favoured formats clean. And while it is imho likely, that Ogg-Theora is the best researched contemporary video format when it comes to “IP” risks, even I can understand why in this age of patent trolls the big vendors have reservations about introducing additional (e.g. additional to their own, already established formats) risks of submarine patents. This seems to be at least the major, official argument against Ogg-Theora in this discussion (although I’m pretty sure, that other aspects of Ogg-Theora – no DRM underpinnings, no properitary vendor backing it 100%, being truly cross platform and therefore a potential lock-in-breaker besides its – agreed – lackluster performance ) contribute to the reservations.
[1]http://www.w3.org/TR/html4/struct/objects.html#edef-IMG
Does this guy even know what he is doing? FFMPEG is a really unstable and rather random project, it is great for playing absolutely any format in the world, but for quality? and especially encoding which it is very rarely used for? Sounds wrong, very wrong.
That said, sure MPEG-4 is going to win no matter what, it is a newer generation of codec, but also a much more expensive one.
This “guy” knows what she’s doing. FFmpeg is used by MANY (and I mean, MOST) GUI front-ends out there. Just because PSP3Video9 or PSPVideo9 or SUPER or WinFF or FFmpegX or VisualHub have nice GUIs, it does not mean they don’t use ffmpeg underneath, because they DO. FFmpeg, stable or unstable, random or not random, it is the *most used* utility for conversions. Even more than mencoder, I would say.
Even vimeo.com, uses ffmpeg to decode the files its users upload on the site.
mencoder uses ffmpeg too (among other things), doesn’t it?
Well put “guy” 😉
i think you dont know what you are doing.
ffmpeg is very stable, and its the best codec package that exists. If you cannot handle it, well, less power to you!
The USA patent system is holding the web back. And please don’t get overpatriotic, I’m not saying the USA is any more shit that any other country, I’m just talking about that stupid patent system.
The solution is not to include that or this codec in HTML 5, the straight answer is to get rid of that shit all together.
I agree entirely with the post, the US patent system needs to be dumped, however, you could have put it on page without the language. -1 to you, now go and get some soap to wash your mouth out.
For me the big news is an OSnews editor stating: “I care not about operating systems anymore”.
I am not part of the main team anymore. I only contribute occasionally. It has been like that since 2005.
If that is correct, then why can we not vote your posts down ?
Because I am still a moderator here, even if I rarely moderate anymore.
Which of these reasons did you try to vote her down for?
Yes, this comment includes personal attacks/offensive language
Yes, this comment is off-topic
Yes, this comment is spam or includes advertisements
Yes, I think I am on Digg and am big & clever when I mod down opinions I don’t like
I was not talking about her last post, I was talking about generally.
What is XDiV also mentioned in Eugenias Ipod Touch review ? Is there meant Xvid projects effort ?
Erm, do you really have to come here, be off topic, for a simple typo? Mention it on the story that’s relevant, ask for a fix over the typo, and move on. No reason to be off topic here.
Who cares about VP3? It’s inferior even to Mpeg4. HTML5 with technology from the past? Wtf …
Is it only me or isn’t Eugenia ranting about the FFmpeg-implementations of the formats rather than the actual formats?
It’s just you. Your tool is only as good as it is, and the tool picked is the most popular one.
Edited 2007-12-12 10:05
She talks about the quality of the resulting video files too, and AFAIK that is independent of the implementation of FFMPEG.
People don’t care about the quality. Look at Youtube, it’s video quality is bad, but everybody uses it. They just want something that works. OGG/Theora is the best way to make sure it *works*, and that it can be supported on every operating system (including mobile phones) and every browser.
Besides, it isn’t that bad, it can beat Youtube’s video quality easily.
Edited 2007-12-12 10:08
This is not always true. If you want to do video as videography, like myself and my buddies do, you go to places where quality is good. For example, look at my youtube page: http://www.youtube.com/EugeniaLoli
I only have TESTS there, it’s like a bag of socks.
My MAIN videography site, for me and most of the Canon HV20 enthusiasts, is Vimeo which supports 1280x720p:
http://vimeo.com/eugenia/videos
See the difference?
in the same way that any enthusiast will not settle with budget gear, big woop…
sad to say eugenia, but your coming of as just another mediaphile. as in, the same kind of people that rant and rave about flac, pcm or whatever being the lowest acceptable format for their media of choice.
or to bring in that sad old car, the people that will not settle for a 4-cylinder stationwagon, if they can get their hands on something with a big more “meat”.
what we are talking about here is the common man format, the one that joe will use for his family video. the same joe that will film his newborn using a cheap camera or maybe even his phone. the joe that dont care that upping the megapixels on his compact camera isnt equal to upping the quality of the photos taken.
the term i guess im looking for is “good enough”. and to me thats a term that the people you seem to personify right now, just dont grasp. it seems that unless its perfection, its no point in doing it at all…
Edited 2007-12-12 10:37
> your coming of as just another mediaphile. as in, the same kind of people that rant and rave about flac, pcm
Not really:
http://eugenia.gnomefiles.org/2007/08/13/producers-howl-over-sound-…
think about it for a moment…
There is nothing to think about. Youtube is big because it’s big, the rest have to compete by upping the quality. So quality does matter. And companies like apple and nokia do care about quality, because their customers do, so they won’t allow to their browsers to use a lesser codec.
I don’t even know why we talk about this. Theora is by their own admission not as good as MP4-Part10. If you care about patents and royalties, use that, but if you are an average consumer, you don’t care about it. You are going to pay for it in the price of the product, a few pennies.
i give up, you would not see it if it was 10 meters tall and flashing green…
that you even get to use osnews as a extension of your own blog is just…
Edited 2007-12-12 11:01 UTC
Why is osnews an extension of my blog? I made a test which shows clearly what is what, and given the amount of comments here, people are interested in it. That’s what we do here, we post interesting stuff. If they are LESS interesting, then they only get posted on our blogs. So, stop trying to make this about me, it ain’t. OSNews does not get its stories from sponsors, we just write whatever we find interesting and we think others will too. There is nothing more or less into it. Get over the “become CNN” syndrome.
Edited 2007-12-12 11:07
ok, so let me put it this way. if quality is as important as you say, why isnt people moving en mass away from youtube?
Because better quality would take longer to load. G4tv.com’s Flash videos are very good quality, but they take way too long to load on a 768kbps DSL connection.
People like YouTube since it loads relatively fast and there’s just so many videos uploaded.
so in the end its a known brand, and one get a kind of mental inertia or “black hole”?
because youtube says you need to use this codec, its clear its not good for the future requirements, as can be seen by apple contracting them to re-encode all that youtube content to the AVC codec for their new venture.
if youtube allowed bog standard AVC CIF/pip/sd main profile encoding then people would use that instead,
as AVC gives a far better quality for a far smaller filesize, and thats got to be a good thing to save bandwidth for everyone, with the added bonus its better than your old good enough, and sets a new better ‘good enough for TODAY and the future.
youtube dont allow AVC/H.264 on their site (yet?) perhaps in time they will.
dont equate AVC/H.264 to only HD 720 and 1080I/P content, its not.
infact its very good for your average 500mbit/s SD content, and even lower, make your encoed/trancode an even smaller screen size and you can lower your bitrates and still end up with a FAR better filesize to quality ratio than any of the other codecs today and that includes DivX.
the one time your old-hat DivX is good for you is for play back video on the old 500mhz laptops you might have around doing nothing useful today, AVC and the coreAVC codec is all you need for anything faster if your justencoding to CIF/PIP type screensizes and most SD content.
dont just sit there, go get a copy of the x264 if your on windows take any mpeg2 DVB-t/-s TV capture you might have laying around and encode it to SD or below size and at a bitrate of 500kbit/s to se for yourself.
Edited 2007-12-17 09:30
I this, actually, that most people want ‘adequate’ quality.
Honestly, quality means nothing if I can’t play the stupid thing.
]{
There is nothing to think about. Youtube is big because it’s big, the rest have to compete by upping the quality. So quality does matter.
This doesn’t make any sense whatsoever.
If YouTube doesn’t have to care about quality, and everyone else does because that’s the only selling point they have, and people still use YouTube, then quality simply doesn’t matter otherwise everyone would be using something else.
Aaah, logic ftw!
We are talking about video codecs here. Quality is just about the only thing that matters.
So here I am, I am writing an article to show that theora is not as good, and all I get is the free software apologists saying that “quality does not matter”.
This is ridiculous. I am not going to reply here again. Thank you for reminding me why I don’t do much osnews anymore.
I would say quality is not the only thing that matters. It is the balance of Quality at any given file size and cost. Note I am not necessarily saying that a codec needs to be completely unencumbered, but if it costs the browser distributor money to be in compliance with the HTML5 spec then this cost should be taken into account. I honestly like h264, and use it for all my video encoding. However, based on what others have said here it does not seem like a practical choice for inclusion in HTML5 due to licensing costs.
The commentary can get very polarized, I mostly lurk now. But OSNews still has an interesting mix of articles (including yours when you submit them) and a few posters worth reading.
Edited 2007-12-12 21:44
“””
“””
That is a very absolutist statement on a very subjective topic. With respect to a web standard, I don’t really see how issues outside of that one very narrow one can be ignored.
“””
“””
Will all due respect, that is not accurate. There have been some good things said here which cannot be dismissed as apologist rhetoric. I do not have a strong opinion on the relative qualities of the two codecs. I do have a somewhat stronger opinion on the relative long-term benefits to “the people” of more open and unencumbered standards. Accepting your opinion of their relative qualities, which I neither challenge nor endorse here, I would say that it is unfortunate that theora cannot claim technical superiority in this area. But that’s not the whole ball game. That’s my position, and I think it is a reasonable one.
Edited 2007-12-12 22:30
“We are talking about video codecs here. Quality is just about the only thing that matters.”
Quality is vital but not the only thing that matters. This reminds me of the Firewire vs USB wars.
“This is ridiculous. I am not going to reply here again. Thank you for reminding me why I don’t do much osnews anymore.”
Please don’t be upset. I found your article very informative. Articles like this are one of the reasons I come here.
We are talking about video codecs here. Quality is just about the only thing that matters.
I’m not saying that quality doesn’t matter. It obviously does to audiophiles, videophiles and people who want to do a lot of professional stuff with multimedia, but clearly quality is a trade-off. It’s a trade-off between quality, convenience and stuff like bandwidth in day-to-day use. In terms of HTML it’s also a case of having something good enough (not fantastic) that everyone can implement. How far would we get if HTML was proprietary and had some severe patent issues?
So here I am, I am writing an article to show that theora is not as good, and all I get is the free software apologists saying that “quality does not matter”.
There are plenty of formats that are not as good as others, but they get more widely used for a wider variety of reasons other than video and audio quality. Not everyone has to have optimum quality, and I dare say Ogg Theora will still improve and get better and be something everyone can use everywhere for video – if Nokia, Apple and Microsoft leave the bloody thing alone and actually get involved positively. Sadly, all we have to look forward to is the status quo and continued stalemate we have today, because that isn’t going to happen. Do we really want YouTube, via Flash, to be the only means of getting video content piped reliably over the web? It’s certainly not going to be MP4, despite many years of posturing.
You know what? What we really need is less committees, less companies squabbling over their own petty self-interests and a bunch of meritocratic people, particularly on the desktop, saying “This is what we’re going to do”.
We are talking about video codecs here. Quality is just about the only thing that matters.
You’re wrong. We’re talking about the HTML5 spec and what the best choice would be for the spec. This includes many factors, not just quality.
So here I am, I am writing an article to show that theora is not as good, and all I get is the free software apologists saying that “quality does not matter”.
It would help if you didn’t insult people and their choices and actually tried to understand why people want a free and unencumbered codec for standarization. You don’t even attempt to address the issues.
This is ridiculous. I am not going to reply here again. Thank you for reminding me why I don’t do much osnews anymore.
Well I say if you can’t take the heat…
You failed to see his point.
Some anal people are always going to pick the better format over the “right” format, so you and your buddies can go play in the h264 park over at noteveryonecanviewme.com and the rest of the planet who just wants to share videos of their dog falling over or the most recent happy slapping plot on sites like youtube.com blip.tv etc. (Not everyone can view YouTubes either, but again, that’s a format issue)
You want to share exceptionally bandwidth wasting video? Fine, just dont spoil the party for everyone who doesn’t.
PS, also nice to see the only reason you’re posting the article is because of replies on:
http://www.j5live.com/?p=421
Nice to see OSNews being an unbiased source of information
Sorry, but you are talking about of your a$$ (sorry for the language, but you ACCUSED me here).
So, the reason I wrote this article is because I had it on the back of my mind for 2 days now. I wrote a blog post about all this LONG BEFORE I got into that discussion on J5’s site. My proof: http://eugenia.gnomefiles.org/2007/12/09/the-next-web-media-format/
>Fine, just dont spoil the party for everyone who doesn’t.
Most people want GOOD quality. So you are in the minority my friend.
Edited 2007-12-12 10:46
say “good enough” and we agree.
there is no such thing as good in this debate, only good enough for the use its intended for.
and this is not about streaming the latest from hollywood onto that HD tv in the living room, its about putting a already “crappy” recording onto a public page for ease of distribution.
I agree with you that GOOD ENOUGH is good enough for most cases. But when there is so much difference in quality, like in this test, the consumer will ask “why not the other fruit”? If the quality was less apparent, then sure, it’s good enough.
is there really that much difference?
to me both is lesser then seeing it with my own eyeballs so…
Yes, to me, this is quite some difference, and I am sure that content providers would agree. For example, if youtube was to use theora, and would have bleed an average of 200 KBs per video in order to achieve the same quality as if they were to use h.264 at a lower bitrate, this is a lot of money per day.
And since youtube is the No1 video provider right now, I am pretty sure they would care about this…
I don’t see people moving en mass away from youtube btw.
ok, I am off to sleep, it’s 3:22 AM here.
Edited 2007-12-12 11:20
i would hazard the guess that if youtube was so interested in quality pr bits, they would not have gone for flash in the first place…
They had no alternative. Flash has market penetration. Ok, off to sleep now. This has become silly.
heh, good night to you.
as for market penetration, they could just as well have selected wmv then, as it was pre-installed on every windows machine out there.
And of course no one uses Macs or Linux.
not enough to shift the percentages in a meaningful way imo. and on at least one, i would think that you can legally install a wmv decoder (osx).
not enough to shift the percentages in a meaningful way imo. and on at least one, i would think that you can legally install a wmv decoder (osx).
It’s never been a question of percentages. HTML has always been developed as something that different platforms can implement. The worrying thing is that the many companies who now make lots of money and have markets that wouldn’t otherwise exist off the back of the web and HTML now want to hold it back. We then will get Windows Media as the effective de facto standard, and Microsoft will then grab Nokia and Apple by the balls and give them a nice squeeze every time they want some entertainment.
i really need to use smilies more…
sarcasm anyone?
WMV is a container not a codec, but perhaps you ment the latest MS VC-1 codec you find in WMV* today perhaps?.
that being to case you should know (after reading my above comments)that VC-1 just does not scale well to low bitrates when compared to AVC/H.264, theres just no contest, if your ‘good enough’ includes the far smaller file sizes and bitrates you get for the same video quality screensize of the youtube generation.
I accused you of posting a blog-style rant on a news site, I feel I am correct. If you want to push your own agenda, keep it to your own site. You might want to keep the ass calling there too.
However, I shall correct the speaking-from-ones-arse comment:
Your other blog entry again repeats the same thing. You’ve said W3C has done the wrong thing which is incorrect, the spec is at the WHATWG atm. However, it cannot progress to the W3 with any technology that isn’t royalty-free, so however much you wave the h.264 flag, they can’t accept it.
If, in the discussions, they can convince whomever that the 264 baseline spec would be RFed, then we all win. Except we still might not; who’s to say another vendor wont ignore the spec, promote their format, and we’re back to the codec/browser/os wars. Lovely.
Picking a reasonable baseline codec for the video&audio tags is the best possible idea, even if we compromise marginally on quality.
“Most people want GOOD quality. So you are in the minority my friend.”
YouTube isn’t good quality, it is however used by a lot of people. I am not in this category though; I am just arguing with your bullshit.
This IS our own site. WE decide what gets posted here. If you want control, start your own website or go to Digg.com.
It’s quite silly to tell us what to do with OUR website. I won’t tell you what to do with yours either.
Edited 2007-12-12 12:24 UTC
Depends on the purpose, what is youtube used for mostly? music videos, perhaps to watch an anime series, and to show off some work. To get the high res stuf people go elsewhere, think thepiratebay.org.
Besides, youtube does use the better codec, why? better compression = faster download of the complete file. Which is also one of the reasons people use youtube. To have the same bad quality on youtube in ogg/theora it would take longer to load.
Ah well, hopefulle dirac gets finished before the official html5 draft finishes (expected in like euh… 2015 or so ) then there’s no need to argue since that one is specifically created for streaming and from the looks of it it’s supposed to surpass h264 in quality as well.
“””
Most people want GOOD quality. So you are in the minority my friend.
“””
FWIW. I desire adequate quality. Something I can watch comfortably. Beyond that, it’s pretty much just refinement which I would likely not notice unless I were obsessing on the quality issue. The difference is simply not that valuable to me. I strongly prefer an open standard, and the benefits that is likely to bring both now and in the future.
Of course, higher quality is an easier sell to the general internet-browsing public. It would have been nice if Ogg could have claimed a victory. Oh, well…
Edit: I should add that I am simply accepting Eugenia’s conclusions as to quality, for the sake of argument, because:
1. I have not paid enough attention to notice a difference in quality between the two.
2. It’s unproductive to argue about something so subjective as that.
Edited 2007-12-12 16:03
This is not always true. If you want to do video as videography, like myself and my buddies do, you go to places where quality is good.
Well, whatever. People still use MP3, which has been inferior sound-wise to the original CD for some time, but by God, people still use it and refuse to use anything else.
Think worldwide including Africa, Central and South America and several part of Asia. Not many people have the luxury to get high quality movies. With the project like OLPC taking effects, using another proprietary format won’t be a viable option.
AFAIR, Theora is still open to the public so you are welcome to bring suggestion or join them to improve it. It is always easy to complain yet somehow, it is hard to not participate to the development.
Edited 2007-12-12 20:24
Post deleted by author
Edited 2007-12-12 11:44
That H.264 will be patent-unencumbered by the 2015-2020 timeframe envisioned for the rollout of HTML5. All the relevant patents were filed in the late 1980s through the mid 1990s. A few have already expired. Thus, if you figure the HTML5 isn’t really likely to come online until the mid-2010s at the very earliest, chances are that MPEG-4 will be patent-free by the time it’s officially embedded into any “final” internet standard.
Many readers on OSNews will choose the first option, but their teenager brother would probably choose the second.
What exactly do you get out of insulting your readers? Does it make you feel better about yourself? Openess is important for web standards. It doesn’t mean you cannot use H.264 if you want to, but Theora should be the standard because it is open and patent-free. For something like the web being open is more important than quality. People don’t mind less quality in return for standarization. Just look at MP3, while it is only a quasi-standard it is pretty crappy in comparison to other codecs out there but a lot of people prefer it because it is compatible with just about everything.
I think it’s funny that Eugenia claims that people want quality over open standards yet she is rebuked by just about everyone here yet still wants to tell us that we want quality over open standards. As it has been said before, mediaphiles might want better quality but everyone else just wants something they can use everywhere without worrying about patent issues and DRM.
And Theora is more widely available than H.264? Seriously?
And Theora is more widely available than H.264? Seriously?
You’re missing the point. If video content is to become as ubiquitous and easy to get and use as web content, which is what people have been angling for with HTML 5, then we need a reasonable baseline that everyone can implement themselves – free software, open source software proprietary etc. – with a reasonably small risk.
It’s not a question of being ‘widely available’, as Nokia and some others are pathetically trying to say. I dare say that HTML was a useless format to many in the early nineties. I also dare say that people would have said “Why should I convert my CD tracks into MP3 when MP3 isn’t widely used and it has a somewhat inferior quality?”
It doesn’t work like that.
And Theora is more widely available than H.264? Seriously?
You misinterpreted what I said. I wasn’t saying it is widely available now, but it has the ability to be widely available because it isn’t patent encumbered, which means anyone can implement it in any piece of software or hardware without patent or DRM issues.
First off I agree with the conclusion. I am a video systems engineer and working on IPTV and trust me, H.264 quality in terms of utilizing all the tool sets in the standard and it is the best out there right now.
Now, I am trying to findout if there are any H.264 decoders in the Freesoftware world that can decode H.264 video encoded with MBAFF. All 1080i broadcast signal (ATSC in the US or DVB in Europe) use the MBAFF tool set because it provideds for improved quality on interlaced signals (like I said so many tool sets in the standards and only about 65% have bene implemented). I was able to sucessfully decode MBAFF content using Nero Digital 8 on windows but no such luck with any of the free software decoders yet. Anyone else have any luck?
“I am a video systems engineer and working on IPTV and trust me”
as in coder?, yes there was a patch in 2006 but by all acounts it still needs work, perhaps you and your mates can take it and improve the toolsets/codec options.
http://download.videolan.org/pub/testing/freehd/
http://forum.doom9.org/showthread.php?t=111620
http://search.virginmedia.com/results/?channel=homepage&vml=ntl&vmt…
Eugenia, though your posts are generally excellent from a technical standpoint, I find your writing generally difficult to read, because of the many grammatical errors and odd sentence constructs. I would suggest that you find a native English speaker to edit and review your posts.
when you say, “do you want the foss, or the best”, you are sorely missing a point.
for some purposes, such as the web standard, its simply not an option choosing something non free like h264, so its not a question of good, theora is the ONLY option, and thus far superior to h264, which is USELESS, utterly useless, in this regard.
either way, i think theora does quite well. its certainly a hell of alot better than the codecs 99% of all flash video sites use(since none of the popular ones have went to h264 yet), and flash certainly has been judged good enough.
OC thers always SNOW, as that is the only other codec other than AVC/H.264 to have a lossless option, shame its encodes are to large when compared to AVC though.
There seem to be two main camps here: 1) the ones who want technical excellence and 2) the ones who want open standards. Each camp admits that the other is important but disagree on which is more important.
Just about everybody here makes fun of Microsoft for having niether technical excellence or open standards. It certainly makes them an easy target.
On the other hand people argue a lot about Apple who have technical excellence but don’t have open standards.
Now we have a review that pits a better product (h264) against an open product (Ogg theora). Eugenia picks h264 because it has better quality and is easier to work with. Other pick theora because it’s open and “good enough”. Unfortunately there’s no concrete right and wrong here. It’s all subjective.
The right answer is that HTML5 shouldn’t be based around a codec that cannot be legally used without paying royalties and licensing fees. That completely flies in the face of how the ‘Net works.
I’ve never heard anyone say Theora has better quality than h264.
It is shame that SNOW codec develop is stalled, it is very feature rich open source wavelet codec, it work but still lack a lot of features that make it compete with MPEG4 AVC.
Dreaming of open formats for everyone is nice, but how realistic is it? How many people use Theora? How many devices support Theora? Not many.
Compare that to H.264: Most modern HD-Camcorders support it. Many MP3/Video-Players. Every blu-ray, and every HD-DVD-Player supports it. Mobile phones do it. And many more devices. It is a huge advantage for a format that there are already many chips that encode and decode H.264 in hardware. And don’t forget: The most recent version of the Flash webbrowser-plugin supports it, too!
Youtube has already mentioned that they are going to update the video quality on their site soon. I don’t think they will go to H.264 directly because too many people still have old PCs and H.264 decoding is very demanding – but sooner or later, they will.
I doubt that sites like YouTube would switch to Theora even if it became HTML5-standard and would be integrated into webbrowsers some years from now (don’t forget that using HTML5 is not an option before at least Internet Explorer and Firefox have implemented it). H.264 has not only superior video quality, but it already works *today*, even within webbrowsers.
Don’t get me wrong. I like the idea of a patent-free video format as the standard. But I just don’t think it is realistic.
Is there any substantial reason why you are no longer interested in operating systems any more? Some major incident or just gradually fading away?
Well keep up the good work you are doing, anyway.
We’re talking about a WEB standard here. When I view video on the web, it’s usually in a Flash, WMV or RM stream.
How does either fare well in a streaming format vs. the top three dominant video/audio web formats?
“However, people with big LCDs or Plasmas will be able to see more artifacts — should a device that playbacks Theora on TVs comes to market.”
Er, I’ve got one. It’s called a “computer”…
The real point of this was licensing issues. It costs around $2Million a year to legally include H.264 in a product, no problem for Microsoft, more of one for Mozilla, and a huge pain for independent Linux distributors.
Just an FYI: The AVC Patent licensing terms are available here: http://www.mpegla.com/avc/AVC_TermsSummary.pdf .
A quick read of that would indicate that MozCorp would have to put up 4.25 -5 million dollars per year (and this would likely continue to go up for 2011+) to be able to ship an AVC decoder integrated into the browser. And that wouldn’t even cover the linux distros that ship their own build of Firefox.
Also, AAC isn’t covered by this either The costs for that run .20 per channel per decoder shipped (so a stereo decoder would run $0.40 each) when shipping over 1,000,000 decoders.
Now my opinion on the matter is that since encoding tech moves at an incredibly fast pace, HTML 5 should not specify a specific codec for a standard that will be around for probably another 10 years. But there should be a mechanism for ensuring that the codecs used are available on a patent royalty free basis. Theora is outdated, there is no doubt of that, but there is no reason that snow or dirac development can’t be picked up and moved ahead.
The web version of all YouTube videos currently show the low-quality streams… however, YouTube DOES have a library of H.264 videos that they will be streaming shortly. Here’s the nifty part, you can see those high quality videos NOW as sort of a preview if:
(1) You are viewing them from the Apple iPhone’s YouTube application WHILE USING A WI-FI CONNECTION (not EDGE, which defaults to the blurry typical YouTube video)
(2) AppleTV
The difference is stunning, though I can’t wait til the H.264 versions are released for general Web viewers (imagine the bandwidth usage then).
“(1) You are viewing them from the Apple iPhone’s YouTube application WHILE USING A WI-FI CONNECTION (not EDGE, which defaults to the blurry typical YouTube video)
(2) AppleTV ”
Is there not some simple way to access these on my Mac?
BTW: LOVE H264– converting my “iMovies” for CD distribution to my family in this format. Great compression. Less flaky than ripping DVD’s.
Forgive my ignorance because I don’t follow the video format scene. What’s the problem with VC-1? Does it give significantly worse file sizes or quality than H.264? I have read that it does take significantly less processing power on the decoding end. So what’s the problem with it?
I’ve personally settled on H.264, but that’s more because this is an Apple house (I’ve about 200GB of H.264-encoded video, largely because it’s easily transportable between iPods, Apple TVs and so forth).
By the looks of things, everything’s moving to H.264, until the next big thing comes along. Flash now officially supports H.264 (which means eventually FLV will die) and if I remember Adobe’s developers words correctly it’s now the preferred video format. Apple has been pushing it since Tiger was released, and of course all of their devices support it. Sony love MPEG4 in general, and H.264 specifically. (all of their phones, not to mention the PSP—maybe the PS3 too?—support it), as do the 3GPP and all of the people building next-generation broadcast systems. It was designed to be the universal video codec, and it seems to be finding its footing being just that.
I’m not too keen on the patent side of it, but in all honesty ubiquity wins out.
To the earlier commenter regarding why YouTube didn’t just encode everything as WMV: Flash’s browser’s penetration hovers around 92-95% in general; WMV support is lower down the list, especially in browsers other than IE (although it’s better now, playing Windows Media in Firefox used to be a bitch). Also, using Windows Media doesn’t give you any provision for advertisements and overlays and such, whereas with Flash you get the control.
As it turns out, as others have said, YouTube/Google’s deal with Apple means that lots of H.264 video is being created in the wild. Now that Flash supports it natively, I doubt it’ll be too long before we see it as desktop (rather than Apple device) users.
Edited 2007-12-12 19:27
For that matter, even the percentage of IE users is lower than Flash.
I’d love to see a change in the approach to patents; sure, I can understand that the companies which spent R&D developing the technology behind it, want to be paid, but the same time, it stifles the movement forward of internet technology.
I’ve floated this idea before; why don’t they allow open source software projects to ship binaries of the implementation? or even a limited version – say, those who hold the h264 video patents give individual exceptions, “opensource web browsers implementing h264” etc. etc.
It’s quite simple. Many managers are stuck in their ways and their thinking. We gotta keep our secrets and dont give anything away for free. If it costs money it must be good.
Definitely; take Oracle; the Windows equivalent in the database world when you consider security vulnerabilities – and the amount of time taken to patch them. I’m always confused about how willing people are to suck down the the kool aide knowing full well there are better alternatives which would require minimal retraining of IT staff (or in my mind, the IT employee’s should be on the ball rather than having to be told when to start learning).
Rather what they should do, is standardize on a media platform, and say that what it supports is the standard. Gstreamer’s base and good would be a decent candidate, as those only include well supported freely available codecs that aren’t patent encumbered.
Ofcourse quality matters but you can have all the quality
in the world and it wont do you any good if you cant play the file.
streaming wma is crapp on osX and streamin quicktime is crap in windows.
there is a reason why flash is widley used today and it is not becus of the good quality it provides to it’s viewers. it is becus it works.
If ogg was accepted and free to implement with out royalties. both apple and microsoft and all phone providers would implement it to follow html5 (maby not all but most)
then we would have somthing better than flash (for streaming video) and people trying out other operating systems would not be
so botherd with non working media.
It’s not like h264 and other codecs just will disapear. content providers will be able to chose if they wish to pay royalties to be able to stream in better quality or be able to stream for free.
and quality is not always all that hell most people wont se the difference betwen 480p and 1080p on a 42″
tv from a distance of 2meters.
And bandwith is not that big of a problem atleast not where i live, among my friends i have the slowest since i only have 10mbit
Hello Eugenia!
The comparison you conducted is a nice thing, but I knew Theora is worse than h.264 already, because Monty himself told it in public (an interesting read itself):
http://web.mit.edu/xiphmont/Public/theora/demo.html
So – it’s nothing new to me. What I’m more interested in is how dirac compares to h.264.
I know it doesn’t exist for you. ;-} I think it really isn’t serious thing right now, but not because of problems with integrating it with ffmpeg, but rather because its specification is not set in stone yet:
http://sourceforge.net/mailarchive/forum.php?thread_name=1194958954…
Still, it is close to be one, and if you – or anybody else here – try to look how they compete, I would be very glad, since I’m not the codec junkie and don’t know how to make good test myself. =}
Edited 2007-12-13 02:51
While I can see how patent licensing would apply for closed standards like WMV, what I don’t get is how it applies in this case.
h.264 IS an open ISO standard AFAIK:
http://forum.doom9.org/showthread.php?t=73022
http://en.wikipedia.org/wiki/H.264
IANAL but I though patents applied to processes/methods/algorithms and not to specifications themselves, for which there are many ways of implementing them.
e.g. you can’t have a patent that just covers a house of a certain dimension/appearance, but you can patent specific methods involved in its construction.
Edited 2007-12-13 09:16
While I generally like h.264 myself (generally use it in a Quicktime container), the samples you post of frames from the encoded videos show that the big difference is sharpness due to local decreases in contrast. Presumably this would be corrected by using a different DCT matrix (an option to FFMPEG).
In this case, I’d be inclined to conclude that you’d probably get encodings that were functionally identical if you further adjusted the Theora encoding parameters (which may take considerable trial and error — but that was true of h264 encoding at one time too).
Using Theora as a baseline probably makes very good sense. However, Quicktime is also open and free from patents (the container format, not the popular codecs used with it), and it might make sense to use Theora in a quicktime container to leverage the additional features of that which including hinting, references, chapters, captioning, etc.
… as I’m not aware of this stuff.
As much as I would have prefered the open solution to come on top I have to agree with you.
Huuuge difference in cpu consumtion of VLC and Media player classic. Are there any “better” alternative for OS X out there aswell? What is the recommended solutions for OS X at all? What codec packs and players are the best ones? And mod players aswell I guess.
Since this thread is rather huge if you wanna answer please do so at [email protected] aswell.
Edited 2007-12-14 11:14
I know we techies care alot about quality and filesize.. and if we get better quality from one codec over another and end up having the same filesize (h264 > theora in this case). There really is no question which one we’d go with. But regular web site users don’t give a damn about quality. Look at how famous sites like break.com or youtube have become. I have never seen 1 quality encoded video on any of those (and still i stay, because what’s the alternative?). My point is. I think w3 is doing the right thing trying to set a baseline SHOULD support for a media codec. With the poor <object>/png/css/you-name-it support coming out of Redmond, there won’t be any guarentee’s we’ll be better off in the future, even if the Theora support was to be a MUST for html5. The browser should still be media agnostic and let players/addons/extensions/plugins (user chosen) decide who get’s to play it.
Eugenia if you udate your editorial to clarify the codec and container relationship, make sure to use the chart i pointed you to in the update to help educate and inform readers.
http://img297.echo.cx/img297/8742/compchart2fa.png
BTW AFAIC the x264 encoder is using a patent-free implimentation of AVC/H.264 so use that as all the pros at doom9 seem to do.
Edited 2007-12-17 08:14