New VP8 Codec SDK Release Improves Performance 168
An anonymous reader writes "Google released a new version of the VP8 codec SDK on Thursday. They note a number of performance improvements over the launch release including 20-40% (average 28%) improvement in libvpx decoder speed, an over 7% overall PSNR improvement (6.3% SSIM) in VP8 'best' quality encoding mode, and up to 60% improvement on very noisy, still or slow moving source video. In other WebM news, Texas Instruments has a demo of 1080p WebM video playing on their new TI OMAP 4 processor, in both Android and Ubuntu."
Comment removed (Score:5, Insightful)
Re: (Score:2)
Hmm, google wave perhaps?
Re: (Score:2)
Cogent analysis (Score:3)
Relax (Score:2)
I hope that the era of cowboy coders isn't entirely done
Judging from the world *outside* Google, it's cowboys all the way down.
Or, you've already made a start on the path to working *inside* Google, with your degree. Good luck!
Re: (Score:2)
Google has good coders who write awesome software with exceptional APIs and they have excellent ideas (yes, ev
Re: (Score:2)
Well... sometimes it's easier to just buy existing products rather than "respond" to the market yourself.
Google Earth was bought from Keyhole and rebranded.
Youtube was bought and the brandname was kept.
Doubleclick was bought and the brandname was locked in a cupboard, and the key thrown away, the cubpoard was put in a seachest, the chest was put on a container ship and the ship towed out and scuttled
Re: (Score:2)
Google doesnt "reinvent" the wheel (Score:2)
Google Search was not simply a re-invention of AltaVista. It is a vast improvement.
GMail was not simply a re-invention of Hotmail. It was a vast improvement.
etc etc. The term "re-invent the wheel" implies you are doing something identical via a new method. That is not what Google does. They make bigger, better wheels.
Google was inevitable (Score:2)
If you think about how the 'incumbents' in the software industry work - their business models are not about technology or product quality, but about first capturing a monopoly (by any means available), then trying to hang on to it for as long as possible (by any means available).
Like a skyscraper shadowing a garden, this has the effect of making it almost impossible for small players to sprout or survive very long. The resources - sunlight, nutrition in the metaphor - just aren't enough.
However if an upstar
Re: (Score:2)
Perhaps evolution is a better word the reinventing?
Re: (Score:2)
Funny but you have described IBM and DEC as well. Back when minis and mainframes ruled the earth that is how software was written for the most part. And no the wild west has just moved to the mobile space.
But I find your comments funny about no more Myspace or geocites. Myspace came after Google's social network Orkut ay nd Geocities was replaced in large part by Blogger.
And instead of Pets.com we have Twitter and FourSquare. FourSquare btw seems to have figured out how to make money which is good because I
Re: (Score:2)
Re: (Score:2)
Minecraft.
Re: (Score:2)
Bullshit Allergy (Score:2)
Re: (Score:2)
Are you trying to say that your ass has long been a supported of the open sores community?
Re: (Score:2)
For a tech reporter, 1 month seems to be the equivalent of 1 year.
Forget VP8, I want OMAP4! (Score:2)
Nuff said.
Re: (Score:2)
true that.
Not that VP8 is bad or whatever, but, give us OMAP4. Not that it's the best or anything either but it's rather cheap, rather open and rather good. A gumstix OMAP4 would be like a little piece of hacking paradise.
Re: (Score:2)
I would prefer an OMAP4 BeagleBoard with an extension card so I can attach SATA & HDMI, but yes.
Re: (Score:2)
BeagleBoards are big. There's the BeagleBoard XM which is pretty powerful by the way.
Also the LeopardBoard which I believe has HDMI, but probably not SATA.
I want OMAP4+Gumstix for ultra light video processing at HD quality (720p)
The current common OMAP 3530 (also on the BeagleBoard - standard version)'s DSP just doesn't cut it for HD video encoding. (It's just ok for decoding)
Re: (Score:2)
Big is relative.
I want an OMAP4 as an always-on server & media system. It can do 1080p decoding while half asleep and has enough power for the odd NFS request or ssh session.
Re: (Score:2)
And in a perfect world, I want another OMAP4 in an ultra-portable Laptop. Thinkpad XO or something would be bliss.
Re: (Score:2)
Afaik the Beagle XM decodes 1080p30, you might want to look into it.
For me, Gumstix size is required to be "pocketable". The Beagle is fine for tiny computers (but that doesn't fit in the pocket).
Might end up buying one of the 720p Android phones myself however, at the next generation in 2011 these phones will be cheaper, more reliable, use less power, be smaller, and more "open source" than the various boards doing 720p today.
Re: (Score:2)
> Afaik the Beagle XM decodes 1080p30
Will do, but no HDMI is a deal breaker. And no SATA sucks for what I want to do.
Re: (Score:2)
My Samsung 7 Series TV doesn't even support vorbis. :(
Re: (Score:2, Funny)
My Samsung 7 Series TV doesn't even support vorbis. :(
Nevermind; people may call you a traitor for your support of proprietary codecs but don't listen to them. You traitor.
Re:What's the point? (Score:4, Insightful)
Re: (Score:2)
Yes mpeg4 and mp4 are both patented. Neither is free and there is a difference.
Re:What's the point? (Score:4, Informative)
Re: (Score:2)
Everyone? FireFox can't use it, because it requires a "paid" license and they're a "free" browser.
Their only substantial source of funding, AdSense.
Canonical found a way to get H.264 support into its OEM distribution - because Canonical knows that Linux is in desperate need of OEM support. Chrome supports Flash - because Google is also interested in market share this morning - and not in the nebulous WebM future.
Not all operating systems come with MPEG-4 codecs (Score:2)
Firefox can use it. They can decode it using the codec framework of the OS
What OS? Windows XP and Ubuntu don't come with MPEG-4 AVC and AAC decoders. I seem to remember that Windows Vista Home Basic, Windows Vista Business, and Windows 7 Starter don't either. Mozilla doesn't want to encourage webmasters to use "To play this video, upgrade to Genuine Windows® 7 Home Premium operating system" as the fallback content for a <video> element.
Re:What's the point? (Score:4, Insightful)
Wow.
What Microsoft is saying is that they are going to provide codec support so every application on the planet doesn't have to reinvent the wheel and that endusers don't have to download codec packs from 3rd parties!
Mozilla could use the provided codec frameworks on each platform to provide h.264 support. The reason they will not is simply one of politics.
Choice is a good thing so let the endusers decide. First time they got to play and h.264 video give them the choice of using the internal codec frame work or not. And in a security warning if you wish.
If not I see a lot of folks going with Chrome.
Re: (Score:2, Insightful)
Important politics. They want an open web. Supporting web video through a proprietary codec goes against that goal. It amazes me how many miss that point.
Re:What's the point? (Score:4, Insightful)
Firefox loads *many* proprietary DLL's on Windows systems (and the OS/X equivs) in order to render web content.
You've turned it into a religion ONLY in the case of CODECS. Why are CODECS special?
Re: (Score:2)
It doesn't matter that Firefox uses proprietary DLLs to render web content, because there are no restrictions or fees on its use.
There is no difference between FLASH doing it and a CODEC doing it. They are BOTH proprietary binary DLL's and H.264 carries the same restrictions and fees in both cases.
So when are you going to push to remove all support for FLASH in Firefox?
Re: (Score:2)
You forgot that Flash is a plug-in that doesn't ship with Firefox.
Nobody suggested that CODEC's would ship with Firefox.
Pretty much everyone already has these codecs, and the license to play back video with them.
If you want to take the high road, you've GOT to be consistent.
You hate H.264. Dont try to convince us that this is about some grand Free and Open slant.. when clearly you are just picking and choosing what non-Free and non-Open shit to ostracize.
Pick a theory and run with it. Someone obviously convinced you that H.264 is Bad but you never reconciled 'wh
Re: (Score:2)
1) Religion.
2) Wrong.
3) Subjective horseshit. You could say that anything is "underspecified"
4) Many plugins are malware too. Pick a theory and stick with it.
5) FLASH is a giant security hole. Arent you going to remove FLASH support?
6) Instead of interfacing with existing decoders, which we think is too hard, we are going to maintain our own decoder.. which any coder will note is obviously harder.. but we wo
Re:What's the point? (Score:4, Insightful)
I understand it.
But the statement that Firefox can to use h.264 is a flat out lie.
They can use it and they can use it without a paying a cent. The can just use the already installed codec system for each of the OSs. In fact that would be the correct way from the stand point of code reuse and software components.
It is silly to have 5 different programs on one OS all implanting h.264.
Firefox can implement h.264 they have chose not to to make a political statement! To put in any other way is a lie. And I do not care how important you think it is to make that statement it should not be wrapped with a lie.
What amazes me is how many people miss that point and are okay with that lie.
Re: (Score:2, Informative)
When they say they can't use it, they mean they can't ship it with their web browser.
Here you make the mistake again. There's a difference between "implement" and "support". Implementing H.264 would mean shipping code that decode
Re: (Score:2)
Mozilla has said time and time again they can not support H.264. They simply will not do it so they can make a political statement.
IMHO this will be the end of the mainstream FOSS browser.
Re: (Score:2)
And I disagree with the reasons for a number of reasons.
Number one reason is politics and I feel politics and tech make for a bad user experience.
But I will take his points one by one.
"2 Only a very small fraction of Windows users have a DirectShow codec for the most important encumbered codec, H.264. Windows 7 will be the first version of Windows to ship with H.264 by default. Even if millions of people have downloaded H.264 codecs "
Windows 7 is growing in percentage very quickly. Even so isn't giving x% o
Re: (Score:2)
Not fud at all. They said that not all Directshow codecs support everything that HTML 5 needs. They didn't give an example. That is FUD. Does the microsoft H.264 codec support what they need? Yes or no?
Not worth the problems? What a load of garbage. Mozilla doesn't have the the user base to force people to not use H.264. I do hope that WebM does work and gets wide support but I am not holding my breath.
So not pull javascript support out of Mozilla? I mean that causes a lot of bugs as well.
And let's just n
Re: (Score:2)
I don't consider Window 7 users a small installed base. It will grow to be pretty big over time. But what you are ignoring is why not allow the user to decide?
A lot of WinBoxs have Quicktime installed if so allow them to use that for h.264 and default to it on OS/X since every OS/X machine has it.
Allow them to use DirectShow if that is installed and has H.264 support.
Allow them to use FFMeg if they have that installed!
This "we are protecting you and we know what is right for you" crap sounds like the FOSS v
Re: (Score:2)
You think that taking away the option of choice because you may have unexpected bugs valid?
You think that 40% of the Windows users is too small of an installed base?
Good heavens I make of it what it is. The death of Firefox. Probably not at some point they will cave in because of user demand. Too bad since a lot of people will have left never to return.
Re: (Score:2)
Actually I am no troll. I am a long time Firefox user. You really think that the majority of Firefox users care about which codec is "free" as which is not?
We will see but history is not on your side.
Re:What's the point? (Score:4, Informative)
Re: (Score:2)
while they're not all free, the ones from google are, aren't they?
also the ones in ffmpeg
Re:What's the point? (Score:5, Informative)
MP4 is not free. Its encumbered by patents.
WebM/VP8 on the other hand, Google says its not encumbered by patents and the MPEG people say it is patent encumbered.
Until such time as the MPEG people can show proof that WebM/VP8 is in fact patent encumbered, I not inclined to believe them.
Re: (Score:2, Interesting)
The thing is, it doesn't matter what you think. MPEG LA isn't going to sue you for using VP8.
It matters what the companies that want to use it think, and they think they're going to get sued, and have actually been threatened by MPEG LA. Therefore, for them, they don't want to take the risk of getting sued (in this economy, that may well be reasonable,) so even if VP8 doesn't infringe a single patent, it might as well infringe all of them.
"...isn't going to sue you for using VP8." (Score:2)
Or at least ... not yet. Not until there's a billion users and the whole web depends on it.
Re:What's the point? (Score:5, Insightful)
Also it should be noted (Score:5, Informative)
That they didn't make the announcement on no royalties until AFTER WebM hit the scene. Before that, there weren't royalties, but it was a "grace period" thing that they could rethink the license terms every 5 years. They can still do that with regards to license costs for encoders and decoders.
That this happened after WebM came out is not a coincidence. They finally had some competition. The plan was likely to try and make AVC the one and only standard, then start charging more streaming royalties (there were streaming royalties when it first came out). However they realized if they kept that ambiguous, WebM might take over.
Also initially I think they figured they could brow beat Google in to playing along, because they are under the belief they have patents that cover all video compression. However you know Google did their homework both before they bought On2 and after they got the technology and before they released WebM. They checked, and Google is precisely the organization that is good at the data mining and searching needed to determine if any patents applied. They likely either found that none did, or that if any did they were subject to prior art, or that Google had patents that they could use against AVC.
Whatever the case, AVC is now free to stream forever, but not completely free. So now we have two choices and that isn't a bad thing. For commercial software/hardware, AVC is probably the better choice since it seems to be higher quality. You buy the license, life is good. For free software, WebM is the way to go as the license is explicit that you can do as you please, no royalties.
Comment removed (Score:4, Insightful)
Re: (Score:3, Insightful)
> free for end-users at the moment as long as they are using a licensed decoder
In other words, free as long as they have already paid for it, right? ;)
Re: (Score:2)
They are not going to sue individual users in the USA. There are no statutory damages for patent infringement: all they can get is actual damages.
Re: (Score:2)
Re: (Score:2)
Can you believe we still have 7 years before the mp3 codec becomes royalty free? Still need to pay for the content though....
Re:What's the point? (Score:4, Informative)
Re:What's the point? (Score:5, Informative)
Actually, due to a hole that existed in patent applications before 1995, some of the patents don't expire until 2017: http://www.tunequest.org/a-big-list-of-mp3-patents/20070226/ [tunequest.org]
Re: (Score:3, Insightful)
Re: (Score:3, Informative)
Re: (Score:3, Informative)
Well, MKV has been around for a while, and having an Xvid file within MKV was very common before being used to encapsulate h264. I really don't care what the public think when the discussion becomes technical. Being accurate never hurts, and if you want to look dumb when trying to have a tech conversation about digital video that's your problem...
Re:What's the point? (Score:4, Insightful)
So Android (no GNU there) is a micro blip on the radar? Where do you get your data, so I can avoid it?
Youtube isn't exactly "small" and uses MP4 with H.264, so no.
Anyone who wants to sell a decent device in the US as opposed to $5 player needs to pay royalties to the MPEG-LA, regardless of where it was built.
Re: (Score:3, Interesting)
Anyone who wants to sell a decent device in the US as opposed to $5 player needs to pay royalties to the MPEG-LA, regardless of where it was built.
Oddly enough, enforcement seems to be very lax. Until recently licensing fees for DVD players would typically add up to $20-$30 per unit, yet it wasn't that hard to find $40 DVD players for sale here. There's no way a $40 retail price could support $20 worth of licensing fees, So they were clearly ignoring them, yet you could find such products in stores like Target and Walmart, not to mention amazon and all the other big-name online places.
Re: (Score:2)
Re: (Score:2)
SP and ASP aka mpeg4 part 2. AVC/h.264 aka mpeg-4 part 10. SP and ASP are not different codecs. They are different profiles of the same codec. If a decoder supports ASP then it probably will support SP.
Re: (Score:2)
They are different codecs, just like Word 97 and Word 2007 are different file formats. They're related and often interoperable, but they're quite distinct and it's not unreasonable to address them as individual formats no matter their common naming or heritage.
Re: (Score:2)
wrapper != codec
Re: (Score:2)
Wait, what? Chrome will support both, IE only H.264.
Re:Who cares? (Score:5, Insightful)
Also may be of interest to cheap devices (Score:5, Interesting)
So you'll notice that WebM is getting built in to hardware, just like AVC. Means soon portable/embedded devices will be able to decode it too. Ok so just another format right? Well sort of. You have to pay per decoder (up to a maximum) for AVC and VC-1 and so on. You don't for WebM. So a company is developing really cheap devices, they don't want to pay that royalty. It adds unit cost. Maybe they decide not to, and instead use WebM because it doesn't cost anything. Sure it saves only a few bucks per unit in licensing but that can add up to $5-10 when you are talking sale price and that can be a big deal in cheap devices. Maybe they sell streaming kiosk/info devices that are $40 where the best a competitor does with AVC is $50 or $60.
There is no doubt AVC is here to stay. It has good quality, never mind the massive installed base and standards behind it. Professional (and consumer) cameras are using it for shooting video in the form of AVCHD and AVC-Intra. Blu-Rays are by and large encoded in it these days (you have a choice of MPEG-2, VC-1 or AVC) and so on. It isn't going to die. However WebM may become preferable when cost is key. No encoder, decoder, format, stream, or any costs of any kind ever for any application. That's worth something.
If I ran a video website, I'd seriously think of looking at moving to it once browsers got support. It would ensure that I don't get fucked with fees at some point in the future.
Re: (Score:2, Informative)
...and instead use WebM because it doesn't cost anything. Sure it saves only a few bucks per unit in licensing but that can add up to $5-10 when you are talking sale price and that can be a big deal in cheap devices. Maybe they sell streaming kiosk/info devices that are $40 where the best a competitor does with AVC is $50 or $60
That's a nice theory but it's not how stuff actually works.
Take MP3 for example. You want MP3 decoding in your device? Well, you'll have to pay licensing fees. If you want something free use Vorbis. So according to you there should have been some companies that produce cheap media players that support Vorbis and maybe WAV/PCM, FLAC, Musepack, ect., but no MP3 and no AAC. Remeber how that didn't happen?
Why? Because if your player only supports niche formats it won't sell. Sorry, you can't listen to that aud
Re: (Score:2)
Assume they don't pay, but you do (as you do pay in increased device/software/app costs, anyhows)
You can watch the VP8 video, or you can pay 2 cents per video and watch the almost-indistinguishable quality (although you've heard on
Sure, it's only 2c, but which do you choose?
Assuming you aren't the equivalent of a 'rich bastard' (you can guess which companies the 'rich bastard' represents), which do you choose?
Re:Also may be of interest to cheap devices (Score:4, Informative)
Of the 27 H.264 licensors, at least half half are global giants in manufacturing:
Apple, Cisco, JVC, Mitsubishi, LG, NTT, Philips, Samsung, Panasonic, Sony, Toshiba and so on.
The 901 H.264 licensees reads, for all practical purposes, like the Fortune 500 and Asian Fortune 500 lists in global tech. H.264 licensing for the mega-corp counts for less than your own pocket change. It's the price of a diet Cola from the vending machine downstairs.
H.264 is a professional/theatrical production standard. It is a distribution standard. It is Blu-Ray. It is important in broadcast, cable and sattelite distribution. It is deeply entrenched in security and industrial video. In mobile devices. In home video. From the $150 HD Flip pocket camcorder to the $5000 pro-sumer market.
WebM is - just WebM. The transcode from other formats you play in YouTube.
Re: (Score:2)
Being like Walmart is surely a sign of failure.
(If your post was genuine masked as sarcasm, not just plain sarcasm, then let my post be a sarcastic agreement rather than counter-sarcasm. Or whatever.)
Re:Who cares? (Score:4, Insightful)
Now considering that, why exactly should I assume that H.264 should be subject to patent trolling later and while WebM remains (patent) troll-free? Just because Google said so? Just because Google has an army of lawyers and money in the bank? Guess what, all of those arguments apply to H.264 too (the MPEG LA also says they have all the patents for it, they have lawyers and money in the bank from the license fees). And what Googles promises and their lawyers are actually worth, we will sadly see soon when the Android patent trials against Oracle gets started.
Also, since we already know that WebM and H.264 are technically very similar, I personally think that possible patent lawsuits coming from future patent trolls might be directed at both systems simultaneously, which would make any perceived advantage from WebM moot in that regard.
Now, WebM still has a lot of merit as an open and royalty-free web video codec. But as far as I am concerned, until either of them gets really tested in court against a patent troll, both codecs are still susceptible to litigation and H.264 may actually have an advantage in that regard as it has been on the market (and thus as a target for patent trolls) longer.
Not only that... (Score:3, Insightful)
The outcome of patent trials isn't 100% dependent on pesky facts, people can win and get injunctions because the other guy's lawyer was having a bad hair day or because the judge was too busy playing with his penis pump or any number of "human factors".
Re: (Score:3, Insightful)
Do I think WebM will be better able to defend itself against patent trolls just because Google has an army of lawyers and money in the bank?
Um... yes? Plus the due diligence they did to ensure it was not infringing before they bought it.
As far as I am concerned, this planet may not be habitable after an asteroid hits until it gets tested. However, if you are going to live your life in fear then you will never get anything done.
I agree with Google that the recording and playing back of moving images and soun
Re: (Score:2)
DAs far as I am concerned, this planet may not be habitable after an asteroid hits until it gets tested.
Um... it was asteroid tested.
Sincerely,
A Tyrannosaurus Rex
Re:Who cares? (Score:4, Informative)
> after H.264 became eternally free for streaming
Except it didn't, except in some limited cases. Please read http://shaver.off.net/diary/2010/08/27/free-as-in-smokescreen/ [off.net]
Re: (Score:3, Interesting)
Re: (Score:2)
I'd rather be perpetually free to encode too, and to do whatever i want with the result, even if it means 10% more band. You make your choice and hope that people who think like me save your a** by making it more dangerous to start monetizing market advantages.
Re:How about quality? (Score:5, Informative)
Re: (Score:2)
How about the total of each difference between the R, G and B values of every pixel to the source. It's not 100%, but I bet that would get pretty close.
Before we can give 'fitness' scores to the algorithm, it's really important that we define properly what 'fitness' actually means.
Re: (Score:2)
Re: (Score:2)
Surely though, one could do something like square (or even cube, or higher, perhaps using exponentiation) the results so that bigger differences are penalized even more so.
It reminds me of the problem determining the volume of a sound. If you take a minute long sample which is basically silent apart from a single loud spike lasting 1 millisecond, do you average the whole minute (very quiet), or take the maximum of the whole minute (very loud).
It probably will never reach AVC in quality (Score:5, Insightful)
But it doesn't need to. For one, all you have to do is be "good enough". This idea that every last bit has to be wrangled out of codecs is silly these days. Storage and bandwidth are cheap. So long as it is good enough, meaning performs like similar codecs (AVC, VC-1 and so on) it is fine. Remember that streaming Flash video was VP6 for a long time, and much of it still is.
However what it offers is a free option, truly free. Encoders, decoders, streaming, all have no royalties and never will. That is important. If you think AVC is free just because of x264 all that means is you aren't doing your homework. Go have a look, you have to pay to have encoders and decoders.
WebM is useful, if for no other reason than it puts pressure on MPEG-LA not to be dicks about patent licensing. However that aside, it may well be the smart choice for streaming out web based video (once it gets integrated in to browsers) since you don't have to worry about issues in the future.
Re:It probably will never reach AVC in quality (Score:5, Insightful)
I'll give you that storage is cheap. But bandwidth is definitely not, at least not where latency is concerned. If you're willing to say "I want to watch this video in 12+ hours" then bandwidth is cheap, but if you say "I want to start watching this video within 15 seconds and never stop to buffer again before the end" then bandwidth is a huge cost, and improving encoding efficiency 10% could have a significant practical difference.
Re: (Score:3, Insightful)
Yet if you want to watch that same video in six months time, because of bandwidth capacity improvements, that efficiency has become moot.
Now, that doesn't mean efficiency isn't important. The real point of efficiency comes when you consider the sheer scale of bandwidth used on video worldwide. Right now, online streaming video is probably the single largest user of bandwidth worldwide (with the possible exception of bittorrent). Gaining 10% efficiency here means a huge amount for the internet as a whole and
Re: (Score:2)
Where do you live where they're improving your available bandwidth every 6 months? Some people are still stuck on 1.5mbit/sec or less crap DSL.
Re: (Score:2, Insightful)
> WebM is useful, if for no other reason than it puts pressure on MPEG-LA not to be dicks about patent licensing. However that aside, it may well be the smart choice for streaming out web based video (once it gets integrated in to browsers) since you don't have to worry about issues in the future.
WebM is already included in Firefox, Google Chrome and Opera browsers. It will also be supported by IE9 if the user installs a codec for it, and there is absolutely no doubt that Google will offer a free codec f
Re: (Score:2)
>>>VP8 will also be supported by IE9 if the user installs a codec for it,
I'm tired of all these different standards. Why can't we just use MPEG4 and be done with it? It has been shown to be the best codec for bitrate-limited video (AVC) and audio (AAC or AAC+SBR) like you find on the web. What you're hoping we'll do is equivalent to abandoning DVDs/Blurays for some other new format, not for any additional benefit, but "just because".
Well no thanks. Since MPEG4 already produces the best qualit
Re: (Score:2)
Erm, doesn't Safari just use QuickTime for video? My understanding was that all one had to do is install the Theora QuickTime plugin if one wanted Theora H.264 in Safari. It's just that Apple has refused to bundle it in any sense.
It's Mozilla which has been the hold-out -- most of the Mozilla developers I've talked to have been openly hostile to such a design in Firefox, mostly because it would give the users the choice of installing whatever codecs they want, which might lead to H.264 winning the war.
Re:It probably will never reach AVC in quality (Score:4, Insightful)
You are several replies are all correct in many ways.
#1. Internet bandwidth increases ~50-60% year-over-year. Will 10% less bandwidth used for video stream matter?
#2. Many *customers* rarely get a 50% speed boost every 3-4 years. reducing video load time by 10% can help dramatically
#3. On a whole scale, video streaming is a large part of internet bandwidth. Even if bandwidth increases 50%, the average resolution of streamed videos will increases to consume that extra bandwidth. ie, 480p one year, 720p the next and 1080p the next. If the video quality was static, then bandwidth would make codec efficiency moot very easily, but people keep increasing the bit rate of the videos to instantly consume that extra bandwidth.
#4. If video streaming is 10% of internet bandwidth, and we're talking about many many teratibts of backbone bandwidth, then reducing that by 10% would free up 1%, which is still a large amount.
Re: (Score:2)
And it matters on today's 3G mobile phones that won't be seeing any bandwidth upgrades for the next 2 years.
Re: (Score:2)
Only problem I have with Theora is that encoding from something else to Theora turns out looking like crap. This isn't the codecs fault, it simply looses information however it's a big deal when you're running a video site like YouTube. It means anyone that uploads to your site needs the source material which hasn't already been encoded into some other format such as h264 which unfortunately almost always the case.
Re: (Score:2)
RTFA? Of course not.
VP8 is open source, which is good, but it's not really any better quality than MPEG2, which is bad. I'd rather just use the superior MPEG4 encoding, even if it meant paying an extra 10 cents per device. Besides MPEG4 will soon be public domain anyway (2015 if I recall correctly) which is just as good as open source.
Re: (Score:2)
VP8 is substantially better than MPEG-2 at the same bitrate. It's not yet demonstrably as good as MPEG-4 Part 10/H.264/AVC, but it's close. VP8 has also shown to be superior to MPEG-4 Part 2/ASP/DivX/XviD, particularly at lower, net-friendly bitrates. As well as VP3, Ogg Theora, and of course, MPEG-1 and MPEG-2... the ancestor of all of these DCT-based CODECs.
And to many people, close is close enough, given the open source nature. Google's releasing VP8 has already had a useful effect -- it was only after t