Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

YouTube Now Transcoding All New Uploads To WebM

timothy posted more than 3 years ago | from the rising-flood dept.

Media 267

theweatherelectric writes "According to the YouTube blog, YouTube is now transcoding all new uploads to WebM, whereas previously the focus was on 720p and 1080p video. Google's James Zern writes, 'Transcoding all new video uploads into WebM is an important first step, and we're also working to transcode our entire video catalog to WebM. Given the massive size of our catalog — nearly 6 years of video is uploaded to YouTube every day — this is quite the undertaking. So far we've already transcoded videos that make up 99% of views on the site or nearly 30% of all videos into WebM. We're focusing first on the most viewed videos on the site, and we've made great progress here through our cloud-based video processing infrastructure that maximizes the efficiency of processing and transcoding without stopping. It works like this: at busy upload times, our processing power is dedicated to new uploads, and at less busy times, our cloud will automatically switch some of our processing to encode older videos into WebM. As we continue to transcode the remaining inventory, we'll keep you posted on our progress.'"

cancel ×

267 comments

Sorry! There are no comments related to the filter you selected.

Open Standards Fanboy (-1)

Anonymous Coward | more than 3 years ago | (#35878460)

Go WebM!

Re:Open Standards Fanboy (0, Interesting)

Anonymous Coward | more than 3 years ago | (#35878560)

H.264 is produced, managed and licenced by a consortium of companies with excellent documentation and a low barrier to entry of said consortium. Patent liabilities are well-known.

WebM is produced by one firm, controlled by one firm, has had no real determination of patent liability, and is documented well by... no-one.

In organisational terms, WebM is closer to Flash and H.264 to HTML. In patent terms, WebM is undefined: a single corporation's promise is meaningless (going to hand over control of WebM to a non-profit consortium to be developed by a working group or some such? thought not) and there has been little to no effort to determine who may be owed what (legally speaking) for its implementation or deployment.

tl;dr H.264 is far more open than WebM.

Of course, the "open" solution is allowing lots of competing plug-in technologies rather than dropping support for everything which doesn't support your desire for control and resultant bottom line. Google, as the new Microsoft, are learning to take the latter approach.

Re:Open Standards Fanboy (5, Interesting)

walshy007 (906710) | more than 3 years ago | (#35878620)

H.264 is produced, managed and licenced by a consortium of companies with excellent documentation and a low barrier to entry of said consortium. Patent liabilities are well-known.

At any point, someone not part of the group could pipe up and sue h264 for patent infringement, sure it's the same with webm, but to pretend that h264's patent liabilities are 'well known' is a farce. Sure some known patents are covered for, but there is no denying the possibility that there are submarine patents somewhere for it, just like there could be for webm.

That is the crux of it. All the people who made mpeg would have to do to get everyone on the h264 bandwagon is to say, unlimited royalty free redistributable license for all forever, and there would be no issue, since they won't do that, it's being worked around.

In other words, wait until the law suits start flying before you say webm is a patent minefield, or instead name some yourself that it breaks that it is liable for.

tl;dr H.264 is far more open than WebM.

If that were the case, there would be no issue shipping implementations of it with free operating systems.

Of course, the "open" solution is allowing lots of competing plug-in technologies rather than dropping support for everything which doesn't support your desire for control and resultant bottom line.

Last I checked people can make plugins for both firefox and chromium, what is your issue here? they have to ship in-built support for every third party format now? no, they can support what they want to support and others are free to implement plugins that add extra.

Google might very well be becoming a skynet equivalent, but that doesn't mean you have to hate the nicer things they do for us. Their goal is for an open internet that is completely platform agnostic, it gives them more eyeballs which is what they sell. That it is in googles best interest to provide us with an open internet is convenient and you should never look a gift horse in the mouth.

Re:Open Standards Fanboy (4, Informative)

Anonymous Coward | more than 3 years ago | (#35878878)

Utter rubbish.

http://www.osnews.com/permalink?470666 [osnews.com]

tl;dr "Google hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer implementations of this specification"

IOW: Anyone may use, anyone may implement, full permission is granted irrevocably and in perpetuity (as long as you don't sue Google).

Specification is documented and submitted to the ITEF.

An independent implementation is here:
http://x264dev.multimedia.cx/archives/499 [multimedia.cx]

Your claim "H.264 is far more open than WebM" couldn't be more wrong if you tried for millennia to make it more incorrect.

Re:Open Standards Fanboy (2)

mug funky (910186) | more than 3 years ago | (#35879200)

webm = matroska(VP8,vorbis). there's some kinda subtitles in there too.

matroska is patent free and open standard

vorbis is patent free (by design) and has been around and in pretty wide use long enough that if there were submarines they'd have sunk by now.

VP8 was the last work of On2 tech, who famously donated the patents and source for VP3 to xiph for use in theora. there have been no challenges in court to this, and VP3 was used for years in flash video and youtube itself.

now, the technical problems:

matroska doesn't actually support standard frame rates - it has a frame length stored on every frame, set in nanoseconds. this mauls standard rates like 24000/1001 and 30000/1001, but it isn't a huge issue as a competent splitter will know what to do.

vorbis doesn't really have a lot of problems, though computational complexity (and hence battery time) used to be an issue. not sure if it is still (i think it used to be float only, but i've no idea).

VP8 is a big mess and has some roadblocks to quality. these appear to have been a consequence of on2 consciously avoiding as many patents as possible. the x264 crew appear to be working on a VP8 encoder, so we'll see what happens there.

Open Standards Troll (4, Funny)

Anonymous Coward | more than 3 years ago | (#35878564)

Help! Help! Someone is trying to give me something for free!

Re:Open Standards Troll (-1)

Anonymous Coward | more than 3 years ago | (#35878580)

Like you can get AIDS for free? Laugh all you want, I'm not laughing.

Now what about 3d? (1)

elucido (870205) | more than 3 years ago | (#35878464)

When are we going to get YouTube in 3d?

Re:Now what about 3d? (5, Funny)

Jappus (1177563) | more than 3 years ago | (#35878474)

When are we going to get YouTube in 3d?

If I had to venture a guess, somewhere around April 1st next year.

Re:Now what about 3d? (5, Informative)

Danieljury3 (1809634) | more than 3 years ago | (#35878544)

Actually its been around for a while While [youtube.com]

Re:Now what about 3d? (1)

johnsnails (1715452) | more than 3 years ago | (#35878654)

Wished I knew how t o/ if i could mod you up.

Re:Now what about 3d? (5, Informative)

macemoneta (154740) | more than 3 years ago | (#35878546)

When are we going to get YouTube in 3d?

Youtube is already in 3D, and has been for some time. You can find 3D videos with this search:

http://www.youtube.com/results?search_query=yt3d%3Aenable%3Dtrue&search=tag [youtube.com]

3D videos have an additional '3D' menu at the bottom, to select the type of 3D output preferred.

Quality? (0)

Anonymous Coward | more than 3 years ago | (#35878476)

No doubt at the lowest possible bit-rate giving even worse video quality than they already do.

It works like this (0)

Anonymous Coward | more than 3 years ago | (#35878490)

"It works like this: at busy upload times, our processing power is dedicated to new uploads, and at less busy times, our cloud will automatically switch some of our processing to encode older videos into WebM."

OMG! Quick, someone implement this ground breaking technique into EVERYTHING!

Dear Slashdot, what point in the past do you formally recognize 'yourself' as having become redundant / post-shark jumper?

Anon

Google/Youtube learning from Microsoft (5, Insightful)

erroneus (253617) | more than 3 years ago | (#35878494)

When you have critical mass, use it. Microsoft and others can bitch about their patent encumbered format 'til they are blue in the face, but Google knows when it comes to video on the web, Youtube is the first thing people think of and the first place they look.

If no other move makes a difference in this html5 format war, this move is the blitzkrieg that will pretty much end it quickly and definitely.

Re:Google/Youtube learning from Microsoft (2)

Spad (470073) | more than 3 years ago | (#35878506)

Step 1: All videos available as WebM
Step 2: HD videos only available as WebM
Step 3: All videos only available as WebM

Re:Google/Youtube learning from Microsoft (0)

Anonymous Coward | more than 3 years ago | (#35878514)

Step 4: Massive anti-trust fines!

Re:Google/Youtube learning from Microsoft (3, Insightful)

erroneus (253617) | more than 3 years ago | (#35878556)

In this case, Google isn't selling a product -- just making content available. (They do sell advertising, not that I see much if any of it) It's "their content" and they can put it into any format they want and make it available to anyone who wants to see it. They will just need a browser with support for WebM... whether that is in the form of a plug-in or native to the application. It will work for everyone and will cost the users nothing.

Antitrust cannot really be used to require the use of proprietary or patent encumbered stuff. Well, it "can" but I don't think it would fly.

Re:Google/Youtube learning from Microsoft (1)

mjwx (966435) | more than 3 years ago | (#35878836)

Step 4: Massive anti-trust fines!

Step 5: here you go Mr DOJ, the full specifications for the WebM format, in exactly the way we've implanted them, oh and have some source code too.

I've got enough for you too Mr EU, dont you worry.

Re:Google/Youtube learning from Microsoft (0)

lxs (131946) | more than 3 years ago | (#35878802)

Step 4: all WebM videos transcoded in the next codec they adopt.
Repeat until nothing is left of grandpa's videos.

Take some responsability for your own content (2, Insightful)

Anonymous Coward | more than 3 years ago | (#35878916)

Grandpa can always re-upload his videos (for free!) if he's not satisfied by the quality of the (free!) transcoding.

You aren't using a (free!) web service without keeping a local copy, now are you?

Re:Google/Youtube learning from Microsoft (1)

tepples (727027) | more than 3 years ago | (#35879238)

What do you mean? YouTube saves the original uploaded video. When it added MPEG-4 AVC to available formats, it didn't go back to the FLVs; it went back to the uploaded video. Likewise, I assume, with these WebM videos.

Re:Google/Youtube learning from Microsoft (2)

mjwx (966435) | more than 3 years ago | (#35878824)

Step 1: All videos available as WebM
Step 2: HD videos only available as WebM
Step 3: All videos only available as WebM

Step 4: Profit (cmon, this is the one time this meme is appropriate, Google want to make a profit from YouTube).

Re:Google/Youtube learning from Microsoft (0)

Anonymous Coward | more than 3 years ago | (#35878924)

Step 1: All videos available as WebM
Step 2: HD videos only available as WebM
Step 3: All videos only available as WebM

Step 4: No longer use YouTube.

More seriously... (2)

DrYak (748999) | more than 3 years ago | (#35879162)

Indeed. But, for it to work, there's also another needed step :

Step "1 1/2" : Widespread hardware availability.

It's already on the way.
WebM is basically H264 with the patented bit swapped out, so just like lots of prior knowledge could be leveraged to code a WebM codec, lots of prior hardware blocks in dedicated decoders could be leveraged to make WebM hardware support.
Also, lots of modern embed platforms feature much more than just a RISC CPU core : vector units, DSPs, and Compute-capable graphic cores are the norm.
Thus, one can already find on the web proof-of-concept code for WebM (and for Theora, for that matters).

Though I don't know yet how much actual usage in end user product it has seen as of yet. (Probably, Android will provide some vector- / DSP- / OpenCL- accelerated support on compatible platforms, soonish)

Re:Google/Youtube learning from Microsoft (1)

mjwx (966435) | more than 3 years ago | (#35878804)

If no other move makes a difference in this html5 format war, this move is the blitzkrieg that will pretty much end it quickly and definitely.

The format troops will be home by Christmas.

Unfortunately I dont share your optimism here. Google may have launched a veritable operation Overlord with WebM but the Axis of MPEG wont give up that easily.

Apple and MS will fight this tooth and nail on the mobile front. Lets just face it, not being able to watch a video in the browser and having to open a separate application is just a pain in the butt, even on Android with supports true multitasking. Apple wont permit WebM to be in the browser and I'm not sure if MS will permit an alternate browser on WP7.

The desktop battle is trivial, the allies of Firefox and Chrome have already got dominance on the desktop, it's on the phone that the battle will be waged.

This is not the end, it is not even the beginning of the end but perhaps it is the end of the beginning.

I've carried on the war analogy a bit far haven't I?

Re:Google/Youtube learning from Microsoft (1)

Bert64 (520050) | more than 3 years ago | (#35878852)

MS are not really relevant in the mobile market right now, meaning it's pretty much a battle between Google and Apple.

Re:Google/Youtube learning from Microsoft (0)

Anonymous Coward | more than 3 years ago | (#35879038)

Uhm, videos in the iPhone aren't viewed in the browser anyway. YouTube videos actually open up in the dedicated YouTube app, and other media types open up in a QuickTime process; they're already doing the "app switching" you're talking about.

Apple would need to adopt the codec in QuickTime, and would have to update the YouTube app to be able to leverage the updates in QuickTime, which may not require any coding (perhaps only a recompilation). Whether this actually happens is up for debate.

As for WP7, they'll need a large enough market share to be able to matter in this debate. That may happen in the future, but it's not the case right now.

All I need to know... (0, Funny)

Anonymous Coward | more than 3 years ago | (#35878508)

is "Friday" converted yet?

WebM is too "geeky"; too "open/free" (0, Insightful)

Anonymous Coward | more than 3 years ago | (#35878518)

"Geeks" have the problem of not knowing how to market things properly. Let's take two examples everyone knows: OGG/Vorbis. What's the penetration of this open and free format out in the music player industry? Zero. Another example: Theora. Penetration? Zero.

Vote troll all you want, but these are facts - as much facts as the reality that current WebM encoders do a worse job in terms of video quality than x264 does for H.264. End-users' experience doesn't matter, I take it.

Re:WebM is too "geeky"; too "open/free" (-1)

Anonymous Coward | more than 3 years ago | (#35878552)

Yeah, you take it in your patent controlling hole.

Nice troll. (0)

nu1x (992092) | more than 3 years ago | (#35878562)

Nice troll.

Re:WebM is too "geeky"; too "open/free" (3, Insightful)

some_guy_88 (1306769) | more than 3 years ago | (#35878574)

Fact: Google is a huge company whose services are used by MANY people

Highly likely: Whatever format Google choose for YouTube will become extremely popular.

Re:WebM is too "geeky"; too "open/free" (0)

Anonymous Coward | more than 3 years ago | (#35878612)

Highly likely: Whatever format Google choose for YouTube will become extremely popular.

Good. If that happens, then the more open standard wins and it's better for everyone.

Re:WebM is too "geeky"; too "open/free" (0)

Anonymous Coward | more than 3 years ago | (#35878686)

They chose MPEG-4 ASP first. It was already popular, and has remained popular for many other reasons outside of YT. Then they introduced MPEG-4 AVC (h.264) on YT. It is incredibly popular right now. Now they switch again, to WebM, but I'm not sure the -commercial market- will follow suit with it.

Re:WebM is too "geeky"; too "open/free" (0)

Anonymous Coward | more than 3 years ago | (#35878632)

You mean except for the fact, that WebM is a crippled Matroska container format with H.264 video inside?

Also, the problem with geeks is, that they have zero self-confidence. They are the smartest people out there, yet fold like chickens to every drooling retard who raises his voice.
And retards are known to act more secure in themselves, since they lack the intelligence to even come up with way they could be wrong. They also don't know if they are right. They have to believe it. Which makes it worse.
It's known as the Dunning-Kruger effect.

If you manage one thing, just work on associating geeks with being cool and dominant. If you're secure enough in in, you can turn all cattle around. I know this from personal experience.

Re:WebM is too "geeky"; too "open/free" (0)

Anonymous Coward | more than 3 years ago | (#35878894)

WebM is a crippled Matroska container format with H.264 video inside

WebP is VP8, not H.264.

Re:WebM is too "geeky"; too "open/free" (1)

Anonymous Coward | more than 3 years ago | (#35878928)

> You mean except for the fact, that WebM is a crippled Matroska container format with H.264 video inside?

This is not a fact.

In ACTUAL fact, WebM is a Matroska container format with VP8 video and Vorbis audio inside.

http://www.webmproject.org/about/ [webmproject.org]

Each technology within WebM: VP8, Vorbis and Matroska, is royalty-free for anyone to implement.

Re:WebM is too "geeky"; too "open/free" (1)

BisexualPuppy (914772) | more than 3 years ago | (#35878672)

" Let's take two examples everyone knows: OGG/Vorbis. What's the penetration of this open and free format out in the music player industry? Zero"
I know you are just trolling, but just getting the fact straight : ogg/vorbis was read by my samsung 'mp3' player in ~2005/2006. It is also the format of Spotify streams. That's the two examples I've got right now, personnal experience, and it was not a choice of my own at all.
Another example, nearly every single professionnal (AAA etc etc.) video game you buy today uses ogg/vorbis (that's ... free).
You yes, you are indeed trolling. Or dumb. Or both.

Re:WebM is too "geeky"; too "open/free" (1)

ElusiveJoe (1716808) | more than 3 years ago | (#35878764)

I use music player with OGG/Vorbis support. I've got a feeling that by "music player" you mean only iPod. But even in this case, you could install Rockbox firmware and play Vorbis.

Re:WebM is too "geeky"; too "open/free" (0)

Anonymous Coward | more than 3 years ago | (#35878904)

No. No, you couldn't. Because RockBox doesn't install on recent iPods. I can't even install it on my 120gig Classic. You can't install it on any iPod Classic. Neither can you install it on any iPod Touch. Looking at the page now, you also can't install it on iPod Nanos past the second generation and it doesn't look like you could install it on a Shuffle, not that that's any great surprise.

So actually, no, for *every single iPod you can buy on the market today and have been able to buy for the last couple of years* RockBox is totally useless.

This isn't an "Oooh iPods are great!" post, it's not an "Oooh Apple are assholes!" post, it's simply pointing out that you've got no clue what you're talking about.

Fortunately I don't give a toss if my music is encoded with MP3, AAC or Vorbis, so I reripped the stuff that was in Vorbis to AAC and just loaded it all on. That's because I like to listen to my music rather than wank over how brilliant I am to only use Vorbis, much as I like Vorbis.

Re:WebM is too "geeky"; too "open/free" (5, Informative)

amolapacificapaloma (1000830) | more than 3 years ago | (#35878888)

Some brands that include the OGG playback feature in their products: SanDisk, Cowon, Trekstor, HTC, Archos, Grundig, iRiver, Philips, Samsung... Pretty neat for a "zero penetration" format ;) BTW, many of them also support FLAC.

Re:WebM is too "geeky"; too "open/free" (1)

sakti (16411) | more than 3 years ago | (#35878900)

Let's take two examples everyone knows: OGG/Vorbis. What's the penetration of this open and free format out in the music player industry?

Not true. It is easy to find music players that support not only ogg/vorbis but flac as well. The only player I've seen recently that doesn't support them are ipods, which isn't surprising given that it is primarily a vehicle to promote itunes.

Re:WebM is too "geeky"; too "open/free" (1)

grumbel (592662) | more than 3 years ago | (#35879190)

What's the penetration of this open and free format out in the music player industry? Zero.

My player can play it out of the box just fine. You also have to look beyond the music industry, it might not have killed MP3 there, but when it comes to commercial computer games for example the penetration of OGG is extremely high, I see it used quite a bit more these days then MP3. I have even seen Theora being used (that one however is pretty rare). So while OGG isn't exactly an MP3 killer everywhere, it certainly has found a few niches where it is extremely popular.

One missing browser tho (1)

nu1x (992092) | more than 3 years ago | (#35878548)

Not for a long time, but still, Seamonkey stable still does not support WebM, it is in upcoming 2.1 as I understand. Seamonkey does not represent a large portion of clients, of course.

Waste of energy... (0, Insightful)

Anonymous Coward | more than 3 years ago | (#35878550)

Until computers/phones have hardware to decode it this will just result in shorter battery life for everyone.

Re:Waste of energy... (0)

Anonymous Coward | more than 3 years ago | (#35878644)

The codec is already supported in hardware in new ICs, just because Apple chose not to, doesn't change the what the rest can pick from the shelves.

Re:Waste of energy... (3, Informative)

monkeythug (875071) | more than 3 years ago | (#35878714)

Seriously, why does this meme just keep going round and round?

computers/phones already have hardware capable of decoding WebM - it's the same hardware used to decode h.264! In most cases all that is/will be needed is a firmware update.

Android phones will obviously be there first - it's already available in Gingerbread. Apple will follow suit eventually, they might resist for a while but with Android's rising market share and Google controlling Youtube, they're caught between a rock and a hard place and I'm sure they know it.

Or (1)

Mateo_LeFou (859634) | more than 3 years ago | (#35878776)

"Apple will follow suit eventually, they might resist for a while but with Android's rising market share and Google controlling Youtube, they're caught between a rock and a hard place and I'm sure they know it."

Or...

Watch this space for iTube?

Re:Waste of energy... (1)

Overzeetop (214511) | more than 3 years ago | (#35878892)

Really? Everything I see on the web indicates that WebM is a container for VP8 video, not h.264. Now, I'm not intimately aware of the details of the two codecs, and they may likely use similar operations, but for them to be patent independent would require significant differences in their implementation. Purpose built hardware for h.264 would have to be exceptionally flexible to have firmware that could be rewritten to process VP8. Call me skeptical.

Re:Waste of energy... (2)

keeperofdakeys (1596273) | more than 3 years ago | (#35879272)

Actually, VP8 uses many of the same techniques to encode video as h264, it is just implemented in a different way. The techniques themselves aren't patented, just the ways of using them, which is firmware not hardware trouble (well, depending how the h264 acceleration hardware was written).

Re:Waste of energy... (2)

tepples (727027) | more than 3 years ago | (#35879280)

Now, I'm not intimately aware of the details of the two codecs, and they may likely use similar operations, but for them to be patent independent would require significant differences in their implementation.

The infamous article 377 [google.com] shows that VP8 is just MPEG-4 AVC with the patented parts ripped out. So yes, any DSP that can handle MPEG-4 AVC should be able to handle VP8 with a minor rewrite of the bitstream parser.

Re:Waste of energy... (0)

Anonymous Coward | more than 3 years ago | (#35878740)

Another ignorant idiot. / Troll.

WebM is crippled Matroska. Matroska is a container format. Not a codec. It is EBML, which is basically binary XML. Very efficient and flexible. And WebM limits the video codec inside it to H.264.

All smartphones and computers I know have hardware H.264 acceleration.

Re:Waste of energy... (2)

amn108 (1231606) | more than 3 years ago | (#35878882)

You seem to be somewhat uninformed as well. WebM indeed is a "narrowed" Matroska container format. The video stream however is NOT a H.264 but VP8, and ONLY VP8. Google chose to narrow down the Matroska and call it WebM precisely because they wanted to avoid having a format on the loose on the Web that could include any type of video stream. And so chose to limit video to VP8 and audio to Ogg Vorbis. Basically if you have a .webm file, the video (if any) it carries MUST be a VP8 stream, and the audio (if any, again) MUST be a Vorbis stream.

Google bought On2 Technologies which developed VP8. The latter is comparable to H.264 with pretty much any kind of motion and bitrates. There are subjective perception tests on the Web dating back at least two years, when the debate on VP8 vs. H.264 and open video was heating up.

Re:Waste of energy... (1)

greg23s (1864340) | more than 3 years ago | (#35878920)

You mean VP8, not H.264

Free speech (0)

Anonymous Coward | more than 3 years ago | (#35878598)

Fine. Now that we have secured our ability to post and view unprovocative videos of kittens and whatnot, how about solving the problem of free speech on the internet?

Youtube can and do limit speech by removing videos and suspending users. We need a free speech tube! This probably means that it can't be financed by ads. (Or at least not financed by any old ad. Maybe there are companies who's CEO:s and boards are hard-core free-speechists who would finance such a site.) Remember, free speech is not the same thing as good speech. The ultimate litmus test for a free speech video site would be whether someone could post a video denying the holocaust and promoting Nazism.

Of course, any free speech video site would have to remove illegal content such as incitement and defamation.

I'm curious about codec efficiency (1)

PhrostyMcByte (589271) | more than 3 years ago | (#35878604)

It's been established that WebM's only real advantage is in being supposedly patent-free, with H.264 still offering significantly more room for higher quality at lower bitrates.

But YouTube doesn't care about efficiency, really. They care about speed and compatibility, which significantly reduces their options. I wonder how x264 fairs compression-wise against YouTube's WebM encoder when tuned to run at the same speed. I'd guess probably still better, but I haven't seen anyone do this sort of test.

Based on their graphs, a 3min video takes them about 1min 45sec to finish encoding -- about 85fps. Unfortunately they don't list what resolution that's in, or what encoders/settings they use.

Re:I'm curious about codec efficiency (0)

Anonymous Coward | more than 3 years ago | (#35878630)

It's been established that WebM's only real advantage is in being supposedly patent-free

Citation needed.

Oh, and nice use of the scare word "supposedly" by the way.

Re:I'm curious about codec efficiency (1)

carlhaagen (1021273) | more than 3 years ago | (#35878744)

Why don't you just google and enjoy the results of the many decent tests of WebM vs x264's H.264 output that you can find out there? So far, the current best (only?) WebM encoder hasn't proven itself to give better quality at same bitrate as what x264 puts out as H.264 in any test; x264's results consistently bests WebM at same (and sometimes even lower) bitrate. The bonus? H.264 decoding is adopted in hardware all over the place.

Re:I'm curious about codec efficiency (0)

Anonymous Coward | more than 3 years ago | (#35879014)

Why don't you just google and enjoy the results of the many decent tests of WebM vs x264's H.264 output that you can find out there?

No, how about the GPP backs up his claims with some evidence? There are many tests out there that compare WebM and H.264. Most of them are flawed in one or more of the following ways:

  • They are out of date
  • They compare poorly encoded WebP files to optimally encoded H.264 files
  • They use poor source material
  • They transcode from one lossy source to another.
  • They use still shots of moving video to prove a pre-conceived notion that one is "better" than the other.
  • They use pretty PSNI graphs that are meaningless in real life

The only thing that anyone has ever conclusively proven with any comparison tests ever done to date is that the current WebP encoder is slower than x264.

Transcode from what? (1)

Anonymous Coward | more than 3 years ago | (#35878608)

Are they transcoding from the original upload materia going back to 2005, or are they transcoding from 240p .flv in many cases?

Re:Transcode from what? (1)

slim (1652) | more than 3 years ago | (#35878880)

Are they transcoding from the original upload materia going back to 2005, or are they transcoding from 240p .flv in many cases?

Presumably, they transcode from the best version they have of any given video. I doubt they've been discarding the original upload for a very long time. Google doesn't treat storage as scarce.

Currently, if I upload an HD video, they transcode it into multiple formats at multiple resolutions; WebM is just another format to add.

WebM WTF? (0)

Anonymous Coward | more than 3 years ago | (#35878610)

WebMediocrity?
WebMonopoly?

Re:WebM WTF? (0)

Anonymous Coward | more than 3 years ago | (#35879182)

WebMedia

This will only hurt the users (0)

Anonymous Coward | more than 3 years ago | (#35878622)

h.264 hardware support is in my TiVo, my point-and-shoot camera, my dedicated video camera, my Apple products, my Android phone ... the list goes on. WebM is useless to me. It locks me back to a software codec. It requires a transcode. If YouTube doesn't work for me, I'll just stop using it. No great loss. Plenty of websites out there will stream video for me, use h.264, not require Flash, and even not decimate my soundtrack. I've already pulled out of Google Calendar and reduced GMail to just bills/online shopping/online signups with mandatory email addresses, looks like I'll be pulling out of YouTube, too.

Re:This will only hurt the users (0)

Anonymous Coward | more than 3 years ago | (#35878688)

I wasn't aware that Google Calendar and Gmail used either H.264 *or* WebM. You learn something every day!

Re:This will only hurt the users (0)

Anonymous Coward | more than 3 years ago | (#35878702)

So long, you'll be greatly missed.

Re:This will only hurt the users (2)

slim (1652) | more than 3 years ago | (#35878902)

Where does it say they're dropping H.264 from YouTube?

YouTube stores videos in a bunch of formats, and the client negotiates the best format for the current situation. It will keep delivering h.264 to you. It will have the option of delivering WebM to clients that are better at displaying that. Everybody's happy.

Does WebM take more processing power to decode? (1)

Posting=!Working (197779) | more than 3 years ago | (#35878678)

I noticed this week that YouTube videos will now make my old laptop overheat and shut down. I can't get through a 4 minute video anymore. I took it apart, cleaned the fans/heat sinks, made sure the fans still ran, and tried a few different video sites, but YouTube seems to be the only one with a problem.

Is this a freak coincidence (or not so freak, it is a 4 year old laptop and my test was far from scientific), or is WebM more processor intensive to decode than the old encoding?

Re:Does WebM take more processing power to decode? (2)

carlhaagen (1021273) | more than 3 years ago | (#35878816)

It's more intensive in terms of software decoding than the current best H.264 decoders. I am not aware of any software that can do hardware decoding of VP8 video on current, common GPUs.

Re:Does WebM take more processing power to decode? (1)

amn108 (1231606) | more than 3 years ago | (#35878844)

On my setup the situation is opposite - Adobe Flash heats my laptop (IBM Thinkpad T43) to the point where even with the highest fan speed the CPU temperature climbs to 70C degrees and I either have to lower the frequency ceiling or pause/stop playback. Fullscreen is infinitely worse, if there is such a thing. And an interesting thing: even with videos that don't have much motion or use a static image, it'sWebM is much less resource intensive there.

Re:Does WebM take more processing power to decode? (0)

Anonymous Coward | more than 3 years ago | (#35878934)

I had that problem with my 5 year old T43. I took it apart and blew all the dust out with compressed air. Now it runs significantly cooler.

Re:Does WebM take more processing power to decode? (1)

amn108 (1231606) | more than 3 years ago | (#35878972)

Yes, I did that once too, a couple of years ago. But somehow, afterwards, I came to detest the idea that I have to unscrew the poor thing whenever it stops being able to ventilate itself. I repaired four or so Thinkpads and some other models - call me a bad nerd, but I am really tired of fiddling with computer internals, especially when it's a laptop (everything is tiny and more fragile). Yes, I know - it's the nature of having moving parts and being cooled by air and so on and so on, but if what you suggest is a common thing to do (like it is with say vacuum cleaners) - don't you wish they had some spring which you push and the entire ventilation system of your Thinkpad just pops out, which you i dunno, wash or rinse or blow clean and reinsert? :-) Basically, a reusable filter. Yeah, i am creative today ;-) And if someone does it first, it must be Lenovo, they got so much in the R&D dep. going on, it's a wonder they haven't done it already.

Re:Does WebM take more processing power to decode? (0)

Anonymous Coward | more than 3 years ago | (#35878866)

My crystal ball needs a polish so it might help to know if you were actually watching WebM encoded video or not.

Re:Does WebM take more processing power to decode? (1)

Skuto (171945) | more than 3 years ago | (#35879000)

It's quite comparable in complexity, but it's less likely to have (full) hardware acceleration support for the decode. Some parts of H.264 decoders would be reusable, though.

Then again, if your laptop is so old that decoding a video overheats it, it probably didn't have it for H.264 either.

Re:Does WebM take more processing power to decode? (0)

Anonymous Coward | more than 3 years ago | (#35879056)

Unless you deliberately CHOSE to watch the webm format, then you are seeing the good ol' flash based video.

The age of a laptop does not tells us much anyway.
You could've bought the worst shit ever made 4 years ago, just like you could've bought a decent core 2 quad which is still powerful enough to run crysis.

Interesting statements (0, Insightful)

Anonymous Coward | more than 3 years ago | (#35878700)

From TFA ... let's translate.

one of our key aims is to deliver great content to you wherever you are - regardless of device, browser or other technical specification

So let's take a step backwards here from the ubiquitous, standards-backed h.264 to something that currently exists only from us and only in battery-sucking Flash format.

Its openness allows anyone to improve the format and its integrations, resulting in a better experience for you in the long-term.

It will be no time at all before people "improve" the codec by adding things to it that won't work on your particular player or device.

Re:Interesting statements (1)

Jesus_666 (702802) | more than 3 years ago | (#35878884)

Except that WebM doesn't have anything to do with Flash and, in fact, can't be played back using Flash. And it's already supported by Firefox 4, even though Fx's implementation seems rough around the edges at the moment. And every file format you have an encoder's source code for (like, say, h.264) can be turned into a proprietary version with little effort.

But yes, apart from that you nailed it.

Re:Interesting statements (1)

slim (1652) | more than 3 years ago | (#35879126)

So let's take a step backwards here from the ubiquitous, standards-backed h.264

Where does it say they're abandoning H.264?

Shame on you, made me read the article (1)

Anonymous Coward | more than 3 years ago | (#35878718)

"What is WebM?

WebM is an open, royalty-free, media file format designed for the web.

WebM defines the file container structure, video and audio formats. WebM files consist of video streams compressed with the VP8 video codec and audio streams compressed with the Vorbis audio codec. The WebM file structure is based on the Matroska container."

Copyright issues? (2)

amn108 (1231606) | more than 3 years ago | (#35878724)

One thing I've been thinking ever since I joined YouTube HTML5 preview, is: do they know how much easier it is to download their videos when playing them back in HTML5? I know that one can also extract Flash video in one way or another, but with HTML5, at least on my setup - Firefox 4 on Ubuntu 9.10 - all it takes is choosing "Save Video" in context-menu. Voila - you can now have whatever you like on YouTube for your own private viewing.

The definite advantage to this, is that one can skip the page parsings and renderings, and instead simply use say mplayer to launch and watch or listen to your favs. Let's face it - the cloud or web 2.0 applications are too slow, at least for me there is noticeable delay. mplayer handles webm videos in much better way than even Firefox 4, not to mention the monstrocity that is Adobe Flash. I simply download anything I watch more than 5 times in a month to the local storage.

Possible solution? (1)

Mateo_LeFou (859634) | more than 3 years ago | (#35878796)

I just got this crazy idea for dealing with this problem:
When people make unauthorized copies of non-free material, prosecute them for doing that.
I know this goes against the legal mainstream (viz. find out what they used to do that and ban it); I'm just thinking out loud.

Re:Copyright issues? (1)

TyIzaeL (1203354) | more than 3 years ago | (#35878828)

People expecting any sort of copy-protection from YouTube are Doing It Wrong. The plethora of re-uploaded videos (with their source being other YouTube videos) should be evidence enough of that.

Re:Copyright issues? (1)

amn108 (1231606) | more than 3 years ago | (#35878944)

I am not talking about the legality of downloading copyrighted content as such, but about how it is far easier doing so with HTML5+WebM than it ever was (at least for me, and I can write C/C++ software _AND_ Adobe Flash Player applications!) with Adobe Flash video player they have. Simply because nobody ever made it big with a sensible, easy-to use thing that gave you a "Download this FLV" button. I've used some Firefox plugins but first, they tended to break whenever YT made changes to their website code and second, just getting to the point when you could click the button was a pain in the ass because Flash sucked the lifeblood of my laptop. I still get nerves when I have to click on a YT link, expecting massive heat issues and fan hitting 50 dB levels.

The WebM however works pretty good. I can watch stuff and still manage to do other work, and I can download not because someone retrofitted a plugin to a browser, but because nothing is actually hidden - there is a .webm stream playing and the browser can download it as well as try to render it. Simple and straightforward.

Re:Copyright issues? (1)

Bert64 (520050) | more than 3 years ago | (#35878922)

If your watching the flash version on linux, look at the files created in /tmp... Flash downloads the video file into /tmp and gives it a random name, but its there ready for you to copy.

Re:Copyright issues? (1)

amn108 (1231606) | more than 3 years ago | (#35878992)

Thanks for the tip, i'll check it out. Still, you got to admit - there is no "Download video" button, is there, as is the case with HTML5 video. It's the small and simple things... And yes, one can do a plugin or two, or a script or a launcher or what not, but it's already there with HTML5, from day one. That's the important difference.

Re:Copyright issues? (1)

amn108 (1231606) | more than 3 years ago | (#35879020)

And of course, for the completely clueless, they won't ever know or care about /tmp, but they WILL (sooner or later) discover the context menu and choose the convenient "Save video" option. And it will spread like wildfire in dry grass :-) Before you know it, it is a fact that YouTube is essentially a video and music installment that lets customers walk out with the content they (YT) put up, without the customers paying a dime for it. Will we get the sort of witchhunts for the average consumers that the BitTorrent freaks tend to get?

batch processing system? (2)

vlm (69642) | more than 3 years ago | (#35878734)

cloud-based video processing infrastructure that maximizes the efficiency of processing and transcoding without stopping. It works like this: at busy upload times, our processing power is dedicated to new uploads, and at less busy times, our cloud will automatically switch some of our processing to encode older videos

Finally, a clear and concise explanation of "the cloud". Its batch processing just like JCL on MVS/360. And to think people thought it was something new...

Re:batch processing system? (1)

jank1887 (815982) | more than 3 years ago | (#35879268)

"our cloud"

doesn't that sort of make it not "the cloud"? I thought the whole point of "cloud computing" was that you're using other people's hardware to do your work. If it's your hardware doing your work on your network, that's now called working 'in a cloud'? So the next time Pixar uses their Renderfarm to produce a movie, they can call it their "RenderCloud"? Heck, I scripted my computer downstairs chop up my audiobook downloads so they play nicely on my older MP3 player. Then it dumps it to a shared drive where I pull it to my device through my laptop. Is that my itty bitty home cloud? Is all network/server based computing now some "cloud something"?

Gah... lets take bets on the next "we don't know what's actually new, so we'll just make up new buzzwords" word. Web 2.0, Cloud computing, ____next?___

The only question I have (1)

devent (1627873) | more than 3 years ago | (#35878810)

The only question I have is does it affect me in any way? I'm using Fedora 14 with FF3. It would be very nice to ditch the flash plugin, which I'm only using for Youtube and other video content.

Re:The only question I have (0)

Anonymous Coward | more than 3 years ago | (#35878930)

The only question I have is does it affect me in any way? I'm using Fedora 14 with FF3. It would be very nice to ditch the flash plugin, which I'm only using for Youtube and other video content.

no, it does not. maybe if you have opted in for html5-mode.

Re:The only question I have (1)

amn108 (1231606) | more than 3 years ago | (#35879142)

As far as I know, FF3 doesn't do HTML5. FF4 does, and does so quite adequately. You won't be ditching the dreadful Adobe Flash pluging just yet though - last time I checked (this morning), a substantial share of YT content is still not available in WebM. Also, if you use Flash for other websites as well, then obviously nothing has changed there. I am a Flash Player developer on occasion, and I also wish i'd disappear. There are some things there is no alternative (HTML5 including) for though - camera and microphone access and publishing to name one.

Where are the MPEG LA VP8 patents? (0)

Anonymous Coward | more than 3 years ago | (#35878910)

The MPEG LA announced their call for VP8 video patents and then they spent a month looking for them. It's now over a month since their patent search ended, so do they have any patents relevant to VP8 or not? I guess not:

http://www.mpegla.com/main/pid/vp8/default.aspx [mpegla.com]

C'mon, at least TRY to sound informed (0)

Anonymous Coward | more than 3 years ago | (#35878918)

"According to the YouTube blog, YouTube is now transcoding all new uploads to WebM, whereas previously the focus was on 720p and 1080p video.

WebM is a file format. 720p and 1080p are resolutions. They are not mutually exclusive. This is like saying "Ford are now making black cars, whereas previously the focus was on cars with round wheels."

Re:C'mon, at least TRY to sound informed (0)

Anonymous Coward | more than 3 years ago | (#35878976)

> They are not mutually exclusive

What IS mutually exclusive, however, is that for WebM Google are now transcoding all videos contrasting to previously only 720p and 1080p videos.

Re:C'mon, at least TRY to sound informed (1)

chocapix (1595613) | more than 3 years ago | (#35879010)

This is like saying "Ford are now making black cars, whereas previously the focus was on cars with round wheels."

Additionally, Ford puts the round wheels on cars named Focus.

Re:C'mon, at least TRY to sound informed (1)

slim (1652) | more than 3 years ago | (#35879160)

It's like saying "Ford are now making Focus cars in black, whereas previously only Fiestas were available in that colour".

libvorbis (0)

Anonymous Coward | more than 3 years ago | (#35878966)

I hope they use libvorbis instead of ffvorbis for the vorbis audio otherwise it will be a big messup.

Too bad it doesn't work with firefox nightlies (0)

Anonymous Coward | more than 3 years ago | (#35879172)

F#$@ useragent sniffing.

Goodbye Flash! (1)

oh_my_080980980 (773867) | more than 3 years ago | (#35879252)

About damn time!!
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>