One of the most talked-about video codecs in the last year, H.264, will make its debut in QuickTime 7 this year when Apple ships Mac OS X Tiger. One feature that makes H.264 particularly noteworthy is its ability to deliver the same quality of video as you see on a DVD, for example, at up to half the data rate. Another is H.264’s ability to smoothly scale from very limited bandwidth applications like 3G cell phones all the way up to HD-quality video, and everything in between.
Now don’t get me wrong, I know Theora has a long way to go but at some point we’re going to have enough bandwidth and open source codecs will have advanced to the point that non-free standards will disappear.
That said, I’m probably going to snag a copy of Quicktime Pro when it’s released.
Now, that’s sick. You are saying that OSS codecs will take over prioprietary codecs *when we will be having enough bandwidth*. In other words, you are relying on getting others to pay willingly (when bandwidth will be cheaper) for the lack of good engineering. That’s like saying that “I will continue to support my MagaOS open source OS that requires 4 GBs of RAM, because one day, 4 GBs of RAM will be cheaper”.
Sorry, but that’s just ain’t right. If Theora (or any other proprietary product or OSS project) wants to be adopted, I suggest its developers sit their a$$ down and do proper R&D and OFFER a BETTER alternative than the current solutions. If they can’t do that, then Theora should not be used where it counts. OSS or not, it does not matter. What it matters is if the XX tool can do better the job than the YY tool. Because computers and software are tools and nothing more.
Be very wary of any announcement that claims “up to”… as in “up to half the data rate”.
The average case may never even remotely come close to the best case.
Actually, this codec can do a lot of cool things. Steve Jobs talked about the new iSight driver for iChat which will be using the H.264 codec in Tiger, and he said that “it would have 4 times better image quality than the current codec, for the same data rate”. And his iSight demo indeed had much better quality. So I am guessing that this codec is not just hype.
as i remember, BeOS died because of lack of DVD apps/players. VLC came to late and my 2:nd hate object Palm bought it.
memoirs from sweden
H.264 is a specification, not a codec, and was not developed by Apple.
The H.264 codec in QuickTime 7 is an implementation of that specification. At this stage the new encoder looks like a big improvement over the abysmal H.263 encoder in QT6.
Pay more attention next time. The guy worked for BeIA’s DVD implementation, not BeOS’.
I think KirkH has a good point Eugenia.
Why are you attacking him for offering his opinion?
<<Steve Jobs talked about the new iSight driver for iChat which will be using the H.264 codec in Tiger, and he said that “it would have 4 times better image quality than the current codec, for the same data rate”. And his iSight demo indeed had much better quality. So I am guessing that this codec is not just hype. >>
Yeah the new codec really is incredible.
Comparison: http://images.apple.com/macosx/tiger/images/h264compare_20040628.jp…
“I suggest its developers sit their a$$ down and do proper R&D and OFFER a BETTER alternative than the current solutions. …. OSS or not, it does not matter. What it matters is if the XX tool can do better the job than the YY tool. Because computers and software are tools and nothing more.”
Exactly.
It. Just. Works.
How hard is that to grasp?
I don’t give a rat’s ass that Gnome w/Nautilus is OSS. That has *nothing* to do with what I think is insanely great about it. Microsoft could’ve created it, and I would *still* think it’s great. I wish Apple would incorporate it (or the design concepts behind it) into OS X.
It. Just. Works.
Apparently there is a new video codec being worked on by the Xiph guys called “Tarkin”, but I think it’s only at the proof of concept stage.
As for whether Theora is suitable for general use, that’s a decision for content producers and end users – each case will be decided on merit and I’m sure there’ll be many happy users.
Who cares if a video CODEC implementation is OSS or proprietary? Honestly, the OSS software is a square attempting to be pushed through a round hole in a lot of cases. One computing model has never demonstrated yet to be the be all and end all, so a lot of overly passionate OSS minds should do themselves a favour and let it go sometimes.
>I think KirkH has a good point Eugenia.
>Why are you attacking him for offering his opinion?
Because I believe that he does not have a good point. In fact, I believe that his point is anywhere between ‘terrible’ and ‘laughable’: “wait for hardware/services to get cheaper in the future instead of sit down and optimize today”. This stance *does not* help the world at large, neither the tech industry, only the lazy or non-talented programmers who sometimes use the OSS as their transporter to glory and fame.
I will say it again: Any product/tool must earn its value in the world. If it is not the best, or does not fill-in a special niche, then the world is better off without it and it should be ignored. Best tool for the job, mate, that’s what I believe in.
To be clear: Everyone is, of course, free to offer his opinion on this site, but so do I. Don’t forget that.
No one mentioned Dirac
http://sourceforge.net/projects/dirac/
I’ve been reading osnews for a few years now. I’ve come not to expect a lot of professionalism whenever anything related to Be gets posted.
This is a fine example. This should be a post and discussion about H.264 and maybe QT7/Tiger. There was no need to mention Be in the first place — especially as an aside and when she refuses to offer the name of the coder in question. It’s frankly irrelevant. 5 year old Be appliances have _nothing_ to do with H.264. At all.
http://www.videolan.org/x264.html
I was planning on buying Quicktime Pro along with Tiger and iWork so in the mean time I unlocked Quicktime 6 Pro with a key I found on the internet (I know, I’m a bad person) and I was shocked at how shoddy the product is! It’s slow, very slow and the only saving grace is how easily clips can be made.
MEncoder and the countless GUI frontends available for it are way cheaper and way faster than Quicktime Pro so why pay for it? Also, if you download the free version, you can play H.264 movies anyway. Why bother buy Quicktime Pro if you’re not going to be doing much video editing?
How is this different from a more readily available codec such as Dvix 3vix or even wmv? I can encode the same quality at up to 1/10th the size of the original Mpg2/DVD. Half the size doesn’t seem like a lot in comparison.
Not trolling, genuinely curious.
“wait for hardware/services to get cheaper instead of sit down and optimize”
I don’t think thats what he was saying. His point makes sense to me. At some point in the future, everyone will have quite a lot of bandwidth. It will not longer make sense to license out a non-open standard when the free/open one is “good enough,” because of the increase in average bandwidth.
If I remember correctly, you like .NET. That could be construed as “sick” as well. Why should someone have to buy more memory when you could just program it in C? I don’t believe that, but thats where your argument leads.
Should you choose the best tool for the job? Yes. Should you license a possibly expensive technology when you don’t need it? No.
Why bother buy Quicktime Pro if you’re not going to be doing much video editing?
You wouldn’t… to take a wild stab, I think their target market for Quicktime Pro is people doing video editing.
Personally I wouldn’t buy it anyway, I hate the free version with a passion and I don’t want to encourage them.
I found this amusing:
“In this entire world of standards, it is Microsoft versus the world and that’s not a good place to be,” said Casanova. “The world has voted for standards and we are on the side of standards.”
Ummmm…. since when did the world vote for standards? Everyone who knows about this stuff thinks they’re generally a good thing, but Windows’ 95% market share kind of indicates that most of the world couldn’t care less.
Also an interesting turn around – Quicktime wasn’t a standard before, it was the one codec I had to play in their player. Mpeg and WMV worked fine in Winamp, but not Quicktime, nooooo….
>Why should someone have to buy more memory when you could just program it in C?
Because .NET’s features when compared to C/C++ *outrun* its negatives. If .NET/Mono did not have all these features I like it for (no reason to go offtopic here), then I *would be sick indeed*. But that’s not the case. I don’t like .NET for its memory usage, I like it for its, let’s say, memory management schemes.
In our case, the Theora codec does not offer any substantial advantages over H.264 at the present time, and I don’t believe that it would do so in the future. As for bandwidth, it will always cost money, even when prices will fall to the floor. And I can promise you, companies do try to cut down costs as much as possible. So, if H.264 saves a phone carrier –overall, in a whole year– 1 million dollars in bandwith, that’s something to consider against Theora, even if the year would be 2014 and the bandwith is cheap enough.
“So, if H.264 saves a phone carrier –overall, in a whole year– 1 million dollars in bandwith, that’s something to consider against Theora, even if the year would be 2014 and the bandwith is cheap enough.”
That is very true. However, 5 years ago I would never have thought that trading memory usage for memory protection would make sense. However, there is a good argument for doing that today simply because memory is so cheap. Nobody can say what will happen with Theora in the future, but its not so far fetched to think it could close the gap enough balance out or negate the savings of bandwidth with the license savings.
It is all very arguable, I was just pointing out that the guy had a point.
It seems to be a growing trend that everything that is not OSS is looked at with distain and as only an interim hold-over until something “good-enough” and is OSS comes along. Frankly its quite inconsiderate. Proprietary software has done an extremely good job of advancing computer science and revolutionizing technology over the past 20 years. At the same time I’m not saying that OSS isn’t also helping the advancement of technology. However that blanket mindset of “good-enough” + OSS = better than proprietary isn’t going to do anything to help advance the state of the art.
Sure, but depends on the case. For example, usually a license is anywhere between $20,000 and $50,000 for standard consumer multimedia codecs.
Now, if the $50,000 licensing brings the company support and big savings on bandwidth it’s still better than an unsupported codec that happens to be Free and that would maybe require to hire an engineer or a consultant to support it if problems arise (an engineer that would cost anywhere between $80,000 and $100,000 per year here in the Bay Area).
Each company should do the math and go for the one with the best overall value. I don’t argue that Theora and other OSS projects will find their niche and will be used where it makes economical sense to be used.
However, as a geek person myself, and as someone who always tries to see things by taking a step back first, I believe that any product (including lanyards, glasses, tables, trash bins and anything else in this world) must be optimized and strive for the best. That’s how the human race can go forward. I strive for nothing less and I won’t be happy with any half-baked effort. Yes, I am a bit of a perfectionist, if you must. That’s why I get so misunderstood over here, so often.
I think, as bandwidth goes up, compression ratio matters less. Which gives relatively more space for other considerations, such as the benefits of using an open codec.
KirkH’s view of the future had two points:
1 – bandwidth will increase, and
2 – free codecs will improve to the level of proprietary ones.
I think that’s pessimistic, but close. My view?
1 – bandwidth will increase, reducing the importance of compression ratio and increasing the importance of other considerations, such as CPU usage (lower = less heat, cheaper equipment), free and open codecs (which helps avoid vendor lock-in), and so on, and…
2 – free codecs will eventually surpass proprietary ones due to the “many eyes” phenomenon and attraction of fame, but…
3 – free codecs will surpass proprietary ones _in the market_ before 2, because of 1.
Erik
Actually I’d have to disagree with your view on points 1, 2, and 3.
Point 1 as far as I can tell is a “vendor lock-in” avoidance issue. MP3, MPEG-2, ACC, VHS, CD-Audio, all of these are proprietary, all of them you could say are associated with varying degrees of vendor lock-in. However the general public doesn’t really care. People by far and away use MP3, not Ogg. They use MPEG-2 for DVD video and now will probably jump on H.264 or whatever is most convenient and gives them the quality they want – again “vendor lock-in” doesn’t even enter the minds of most consumers.
Point 2 is two assumptions; that many eyes will make codecs that are OSS better than proprietary competitors, and that quality difference will drive OSS codecs into the majority. One case that these assumptions are not holding true is in Ogg vs MP3. Ogg has better quality, is OSS, and is still much less often used than MP3. It can be argued that is because MP3 got there first, however WMA, ACC, MPEG+ are all gaining momentum while Ogg isn’t. There really is not solid proof that without proprietary support of some sort OSS will just naturally beat proprietary software in either quality or marketshare.
Point 3 basically doesn’t say anything it just reaffirms that you think points 1 and 2 are enough to drive OSS over proprietary software and that you give more weight to point 1.
H.264 is a specification, and was specced by a standards body. Not a lone Be engineer (I find that laughable). Just to ‘whisper’ that one person should get the credit is an insult to all the other participants.
Apple has implemented said specification. However they are most definitely not the first to implement it. Companies have been building H.264 encoders for well over a year now. It will not make its *debut* in QT7. If you attended NAB *last year* you could purchase H264 encoder software (and even the year before that, although in ‘best guestimate’ form, following the pending spec).
H.264 can perform better or worse than 1/2 the compression rate depending on a lot of factors, e.g. 720p HD video in 6Mbits vs 13Mbits required by MPEG2, or SD video in 2Mbits vs. 4-5Mbits MPEG2.
H.264 is not just about the specification and implementation of a ‘video codec’. Read the entire spec.
Theora et al will never be competition in the realm of large broadcasters, because it doesn’t have the marketing, engineering, or industry backing *and* participation vs. H.264.
It doesn’t matter if one technology is better than another, choosing what standard to use is always political in nature. If you worked in these circles you would know this. He who has the most money and market clout becomes the standard.
“It doesn’t matter if one technology is better than another, choosing what standard to use is always political in nature. If you worked in these circles you would know this. He who has the most money and market clout becomes the standard.”*
Let’s hear it for Microsoft!
*VHS vs Beta.
Vincent (IP: —.200-68.tampabay.rr.com) –
Ogg has better quality, is OSS, and is still much less often used than MP3. It can be argued that is because MP3 got there first, however WMA, ACC, MPEG+ are all gaining momentum while Ogg isn’t.
Ogg is a big failure because it can’t be decoded with a reasonable amout of computing power. The problem is while most codecs use a lot of resources to encode they can be decoded quite easily with OGG the encoding and decoding take about the same amount of ressources. It uses both a lot of RAM and Cpu Cycles so it makes decoder chips more expensive thus they have a very slim base of mobile DAP whic are OGG ebabled. Even some ogg fan is always nagging in diverse audio forum they have to accept that theit impemation is not so feasible for manufactors.
I think, as bandwidth goes up, compression ratio matters less. Which gives relatively more space for other considerations, such as the benefits of using an open codec.
Wrong. The CE industry leverages bandwidth these days against new product offerings. If you don’t, the competition will. Your HD set is already outdated, 1080p is on the way. For a large broadcaster an open codec really has no significance. It simply translates to ‘unsupported’ most of the time (no, people working on it on the side don’t count as engineering staff).
…
Each company should do the math and go for the one with the best overall value. I don’t argue that Theora and other OSS projects will find their niche and will be used where it makes economical sense to be used.
They will remain in a niche. Companies choose standards like MPEG2 *because they are a standard*. If you’re making an MPEG2 encoder, you know that you can use another manufacturer’s MPEG2 settopbox to view your content. $50k licensing is chump change.
People also seem to make the mistake here that all one needs is an open source codec. Thats like saying you only need a wheel on your car to get from point a to point b. You need specifications for video transport, captioning, epg, conditional access, modulation, the list goes on. Most of the world does not use IP to deliver video – get out of that mindset. In terms of being a complete and solid offering, the open source codecs mentioned pale in comparision to MPEG2 et al.
Bandwidth will increase, but that will not reduce the importance of compression. We will just demand higher quality and transfer more data…
(IMHO)
“Who cares if a video CODEC implementation is OSS or proprietary?”
I agree. I could care less if my codecs are open or closed source. I just want what will do the job well for me.
“Bandwidth will increase, but that will not reduce the importance of compression. We will just demand higher quality and transfer more data…”
Yep, and when HD DVD etc comes out, people will be even more dependent on XVID/DIVX whatever to be able to copy it (not that I encourage such practices ). IMHO, technology like this will always outpace available storage and bandwidth capacity such that we will always be wishing we had more.
Please not again with this myth.
Beta was not better than VHS.
Read this,
http://www.guardian.co.uk/online/comment/story/0,12449,881780,00.ht…
If you run Windows or MacOS X and have a JVM, you can check out http://video.vividas.com/sample/sales/page1.html for an example of an even better video format (Mac version of player still in beta).
It’s not a myth, and there is nothing in the article you linked that proves otherwise.
Beta was a technically a superior product and that is what people claim. As a “whole product” VHS eventually prevailed because of pornography being available only on VHS. Go figure.
<<Steve Jobs talked about the new iSight driver for iChat which will be using the H.264 codec in Tiger, and he said that “it would have 4 times better image quality than the current codec, for the same data rate”.
I also said we’d have 3Ghz processors over a year ago!
People are so gullible…
dude.. that is an implementation of the h.264 PLAYER not the codec.
first… Quicktime is a platform, not a codec. Quicktime uses standards and creates a groundwork for a media creation system that interacts very well making it very easy to do a lot of media production on it.
second… WMV is not even close to being a standard except in the online porn industry. just because the codec is available on 95% of the world’s computers does not mean that the people that create the media that those 95% watch use it… which they don’t.
“I agree. I could care less if my codecs are open or closed source. I just want what will do the job well for me.”
You’d think the MS File format lock-in would have taught people something. Guess not.
umm.. MS’s Locking is due to the fact that MS does not share the specs of the codec as a public standard that anyone can purchase.. being a pay to play codec does not mean that you are closed.
MS chooses who they let play and that is not open like MPEG is.
> H.264 is a specification, not a codec, and was not developed
> by Apple.
> The H.264 codec in QuickTime 7 is an implementation of that
> specification. At this stage the new encoder looks like a
> big improvement over the abysmal H.263 encoder in QT6.
First sentence you say “H.264 is a specification, not a codec,…”. Second sentence you say “The H.264 codec in…”.
Well, is it a specification or is it a codec? Can you clarify your statements please?
“Theora et al will never be competition in the realm of large broadcasters, because it doesn’t have the marketing, engineering, or industry backing *and* participation vs. H.264.”
Not entirely true. The BBC is considering either Theora or Dirac (with an obvious lean towards their home-grown Dirac). They’re looking to expect 20 million online viewers too. They’ve made it quite clear that they don’t want to license commercial technology and want to move away from reliance on Real.
See http://kamaelia.sourceforge.net/KamaeliaFirenze.sxi
Eugenia said
“I believe that his point is anywhere between ‘terrible’ and ‘laughable’: “wait for hardware/services to get cheaper in the future instead of sit down and optimize today”. This stance *does not* help the world at large, neither the tech industry, only the lazy or non-talented programmers who sometimes use the OSS as their transporter to glory and fame. “
I said:
“Now don’t get me wrong, I know Theora has a long way to go”
” I’m probably going to snag a copy of Quicktime Pro when it’s released.”
My thinking is this… Free software(not necessarily OSS) is getting better all the time. Codecs with licensing fees are rarely used in Linux distributions that are free downloads. I could go on here and tear apart your argument but it looks like you just wanted to call someone “sick” because you assumed they were an OSS zealot which I’m not. I have a business degree and write software but that’s not why I think you’re wrong here. I’ll take my ad clicking mouse elsewhere.
Since when did it become the norm that the better format won out. Someone mentioned VHS vs Beta. For example, Windows. Windows 95 and 98 crashed all the time. Were there better systems available, yes. Did that stop these systems from establishing Microsofts dominance in the computer industry, no. There are many reasons that the superior technology doesn’t win, but it often happens.
On the point of bandwidth vs compression, I agree that because people are getting access to more bandwidth than ever before, compression becomes less important. With 56k it was a big concern, but with so many people adopting broadband things become less tight in the pipe. That doesn’t mean that developers should abandon the idea of efficiency, but it does mean that it doesn’t have to be the primary concern when choosing a format if the compression ratios are close.
Ogg is not a codec: it’s a file format (container) like avi.
Vorbis, the lossy compression sound codec from xiph.org, can be decoded in hardware and software and with minimum processing power. Why didn’t it take over MP3? Market momentum. But I, for instance, use only Vorbis.
Theora is not 100% yet, and it’s not as good as WMV9 in quality, but is less intensive computationally, too. Dirac seems to be even better.
Everything I have read and seen about this Codec tells me its going to be a good one. I look forward to it and tiger.!
This article protests rather too much, doesn’t it? It’s written as if the World Dominance of H264 is a fait accompli and everyone with half a brain can see it, but it pushes the point just a little too much, which reveals it’s actually just agenda-bashing. It has competition in the next-gen DVD space – remember, *multiple* compression formats are being ratified for the new standards, and this is just *one* of them – and in the general-future-codec space from theora, dirac and other contenders. Much as Macworld would apparently like the game to be over, it ain’t.
The only thing important is that H.264 is here _now_ and [will be] widely implemented versus the OSS version, which is still the love child of a great many very talented minds, but it has not reached maturity yet and it does not run on its own [how many feet does a codec have, actually?] feet.
The practical upshot of that is: there is to all intents and purposes no workable alternative to H.264 [and I’m not saying that it wouldn’t blow the roof off of every technology loving mind on the face of the planet]. By the time this alternative comes online, in the hazy, undiscovered country, H.xxx will have made its way onto the scene and, and this is important children, the lessons learned from H.264 will [we can at least hope common sense would prevail to that extent] have been implemented into it. Which will make our OSS version, oh brilliant miracle of true generosity, limp behind as it tries to adjust to the new reality.
Unless OSS solutions reach a, not -the-, marketplace and can make themselves be recognized as a superior alternative, this entire virtuosity of arguments and great passion of deep conviction is as relevant as a discussion of who was the better captain of the Enterprise, Kirk or Picard [Picard, of course].
Whatever happens [all things being equal]: I’ll be driving Tiger, working with QT7 [love it or lump it] and I’ll be using my iSight not merely for having beautiful women folk ogle my handsome fysique over iChat, but also, and mainly, to scan my library into my spanking new Delicious-Library . I am merely a customer of Delicious Monster, but I am a happy customer.
And I hope you can say the same.
[BeOS should have made the big time]
“You need specifications for video transport, captioning, epg, conditional access, modulation, the list goes on.”
Which, in the computing world, are considered to be the job of the carrier format (e.g. AVI), not the ‘codec’. Remember, this is a computing focused site, although we respect the knowledge you’re obviously bringing from the broadcast world.
First sentence you say “H.264 is a specification, not a codec,…”. Second sentence you say “The H.264 codec in…”
Part of the process of creating a specification is implementing it. Video coding is part of the spec. Some call it a codec, where others call it a decoder or encoder.
Not entirely true. The BBC is considering either Theora or Dirac (with an obvious lean towards their home-grown Dirac). They’re looking to expect 20 million online viewers too. They’ve made it quite clear that they don’t want to license commercial technology and want to move away from reliance on Real.
As I said – large broadcaster. 28 million is a press release quote – broadcasters going online have always made such high claims, public or not (Movielink, etc.).
BBC is a *public* broadcaster. Public broadcasters are in a different boat than commercial entities in most countries, i.e. they can secure funding for ‘research’ projects.
Consider worldwide how many people use MPEG2 – not just the a/v codecs, but the transport, metadata, and other elements. Implementing MPEG2 is a multi-billion-dollar industry. The vendors interoperate (they have to to stay in business). The trend is towards H.264, and it will just snowball.
While it is cool that the BBC is doing Dirac it is a grain of sand on the beach in comparison. And most importantly – the hardware (encoder, multiplex, stb, pci board etc.) manufacturers will go (they already are) with H.264, not Dirac, *because* it is a standard and others are implementing it. There was a period of a year or two where people were non-commital over H.264 because they didn’t want to get burned like they did with the likes of the earlier MPEG4-ISMA, but that has passed.
H.264 will be widely adopted and Dirac et al will remain niche soft-codecs. How much people crow about open source has absolutely no weight in an argument where people already have business models in place to cope with licensing issues.
Jobs also lies about ‘4x the quality’. Quite blatant actually. I guess he had to one-up Microsoft, who always claim that VC1 has 3x the quality of MPEG2 (also a lie).
Actually I’m in the IP over broadcast world, which is for lack of a better term ‘the fence’ between the two.
What I’m trying to point out is that there is much more at work here than just having a codec available.
As I said – transport. AVI is basically a transport in nature, but it is no where as complete and rounded out as well-adopted transports like the MPEG/DVB/ATSC/etc. world. Just because you can put anything in the AVI transport as a container for whatever data you want, doesn’t make it a good one.
I suggest anyone really interested in standards work vs. the open source ‘codec’ world (the two *do* cross over) go over to the likes of http://www.atsc.org, http://www.dvb.org, http://www.m4if.org, etc., and take a closer look at all the requirements needed for delivering video. The open source specs have a long way to go, and this is why they will stay in the niche/IP-video arena, where one doesn’t ‘really’ need to define a transport (i.e. just use HTTP, or ip multicast, etc.).
Carrier is terminology associated with the telco world. Don’t even go there.
I know, I work for a communications company too. But you knew what I meant.
BTW, the BBC *is* a large broadcaster; due to the quality and comprehensiveness of its online media, it’s incredibly popular (check the figures on their archive radio service). When they get video going it, too, will be incredibly popular, I don’t doubt that at all. But I agree with the point about research funding – it’s the BBC’s excellent research division that’s writing dirac in the first place. It’s also why the BBC can afford to put huge reams of content up online, entirely free. Public service broadcasting is one thing Britain’s always done well . I don’t know why the whole Real thing happened in the first place, though…the research division had some excellent quality Vorbis streams of the main radio stations running back in 2001-2002, until they got closed in favour of the Real streams for some bizarre political reason. Sigh.
based on quality for high motion video (e.g. sports).
This is easily the best article I have seen explaining how h.264 works http://www.drunkenblog.com/drunkenblog-archives/000312.html and how good it is
Adamw – Political reason – you hit the nail on the head.
H.264 vs VC1 – they are very much neck and neck. However H.264 wins because it is born out of a standards body consisting of many industry players. VC1 is ofcourse.. Microsoft’s baby, and they very much like to promote it as such. Fanboy vendors like this, though. YMMV.
DirecTV is going H.264. So is Apple. As are many others. Half of the choice is quality/performance/technical merit, the other half is political.
“You’d think the MS File format lock-in would have taught people something. Guess not.”
Guess what. I need what works TODAY.
If there is a solution that works to deliver superior video quality and small filesize RIGHT NOW, that is what will be used.
Business/Mass of End Users isn’t going to wait for the OSS project which *might* show up and work at some point in the future.
Apple’s got the goodies NOW? Well, hey, the OSS solution is a day late and a dollar short and that’s just too damn’d bad.
“You’d think the MS File format lock-in would have taught people something. Guess not.”
And exactly what ‘lock in’ with respect to file format have I been miserably struck down by as a Windows user?? Are you referring to Office? There is a list of free apps that can read and write them which is as long as your arm. Also, ofice apps themselves include import and export filters allowing you to save in different formats, including HTML. In fact, office docs in particular use OLE compound storage, which isn’t too difficult to analyse using a graphical viewer. As for Windows Media, I have a 3rd party player which can view them on my PocketPC and it wasn’t written by MS.
I am constantly amazed at the lack of business acumen by the OSS brigade. I don’t sit on either side of the fence but I do laugh all the way to the bank when OSS people seem to cripple their futuress due to a stubborn religious passion to trap themselves in the 90s on occassion.
there is, and it’s been around for years. DivX. This whole thing really offers nothing to computer users. As Kon has said, the point of H.264 lies in the infrastructure, which is important to broadcasters. It’s not going to make your Naruto .avis any smaller, or look any better. The whole Apple announcement is a piece of fluff which tries to make an industry standard look like their own product and make end users believe it’s significantly better than anything they have now, which it isn’t. Standard PR practice, nothing to get excited about.
iTunes is based on quicktime yes? i would like to see true gapless playback there (no fake crossfade thing at 0 seconds)
How does Dirac differ from a traditional broadcast standard like NICAM? The real issue seems to be a political division between North America and Europe. NA went with NTSC, Closed Captioning and Dual Stereo, whereas the rest of the world went with PAL, NICAM and Teletext – all (debatedly) superior systems. You also have to understand that the BBC is in a different position to the PBS. The Beeb has a truly global outreach – pretty much all of Europe and the commonwealth countries and use BBC’s standards.
ATSC probably wont be used widely outside of America, but it does have hardware support.
Dirac and whatever transport mechanism it will eventually use will be supported in hardware as well.
At the end of the day hardware manufacturers want to be as multinational as possible.