Linked by Thom Holwerda on Thu 20th May 2010 23:22 UTC
Multimedia, AV There's an incredible amount of momentum behind Google's WebM Project. Opera, Mozilla, and of course Google will all include it in their browsers by default, meaning about 35% of web users will be able to use it with a minimal amount of fuss. On top of that, Microsoft has changed its previously announced plans to make HTML5 video in Internet Explorer 9 H264-only to include VP8 as well. Only Apple's opinion was unclear - until now.
Order by: Score:
Fettarme H-Milch
Member since:
2010-02-16

Nonono. VP8 proponents tell us since two years (VP8 was releases back then) that it blows the entire h.264 standard out of the water.

Now a mere name change and a quality-unaffecting container switch to the proven and robust Matroska format makes VP8 "very raw, hardly out of development" and excuses for rather mediocre performance against h.264???

Make up your mind.

Edited 2010-05-20 23:49 UTC

Reply Score: 11

Thom_Holwerda Member since:
2005-06-29

...compared to H264, it's new. This isn't rocket science.

Reply Score: 2

tf123 Member since:
2010-01-28

But it was still supposed to be twice as good as h.264 at half the bandwidth at conception 2 years ago. On2 wasn't saying, "In a decade, VP8 will be twice as good as h.264 at half the bandwidth was last decade." That doesn't make for good marketing.

Reply Score: 5

diegocg Member since:
2005-07-08

Well, it isn't. Who cares, anyway? I don't mind if it's a bit worse if it's open. We were willing to use Theora instead of H.264, and we are going to ditch VP8?

Keep in mind that Google is going to invest insane amounts of money on this and that long-term it may even develop entirely new codec techniques.

Reply Score: 2

MollyC Member since:
2006-07-04

So the public, which doesn't give a fig about "open source", is forced to settle for an inferior codec for the sake of a small minority's ideology?

It's fine for you to choose OSS stuff even if it's inferior. My beef is that that choice is being shoved down the public's throat as well. So much for letting competition allow the better technology to win. Advancind a group's ideology trumps such quaint notions, I guess.

Reply Score: 2

diegocg Member since:
2005-07-08

It turns out it's the public who doesn't give a fig about inferior codecs. It's the geeks who care about that.

Reply Score: 2

PresentIt Member since:
2010-02-10

Molly, you are seriously not up to date.

Was HTML and CSS "forced" on you? No, they are just standards. The web needs free and open standards as a foundation.

Standards aren't "forced down the public's throat". The public doesn't make browsers or sites.

Yes, a free and open codec is needed for HTML5. A closed codec will damage the market like IE6 damaged the market.

Reply Score: 1

Fettarme H-Milch Member since:
2010-02-16

...compared to H264, it's new. This isn't rocket science.

VP8 is -- as the name suggests -- On2's 8th generation codec. Its first generation (under the name TrueMotion-S) was released 15 years ago and supposedly development started "in the early 1990's". See http://en.wikipedia.org/wiki/TrueMotion
You'd expect that On2 learned to program a competitive codec during that time frame.
Or do those almost 20 years still qualify as "very raw, hardly out of development"?

Reply Score: 6

xmv_ Member since:
2006-06-09

"...compared to H264, it's new. This isn't rocket science.

VP8 is -- as the name suggests -- On2's 8th generation codec. Its first generation (under the name TrueMotion-S) was released 15 years ago and supposedly development started "in the early 1990's". See http://en.wikipedia.org/wiki/TrueMotion
You'd expect that On2 learned to program a competitive codec during that time frame.
Or do those almost 20 years still qualify as "very raw, hardly out of development"?
"
obvious troll. do you think h264 has no previous codecs? that's right h263. thats right all the previous too.

VP8 has never been seen outside on2 before this week and claims were on2 claims alone.
The codec does better than h264 baseline and worse than main and high profile (which are a good step up)
even x264 author claims that

additionally x264 has had a tremendous work done compared to VP8.

so.. VP8 isn't bad really. in fact, it's pretty good.

Reply Score: 2

Stratoukos Member since:
2009-02-11

From the analysis you linked:

Initially I was intending to go easy on On2 here; I assumed that this encoder was in fact new for VP8 and thus they wouldn’t necessarily have time to make the code high-quality and improve its algorithms. However, as I read through the encoder, it became clear that this was not at all true; there were comments describing bugfixes dating as far back as early 2004. That’s right: this software is even older than x264! I’m guessing that the current VP8 software simply evolved from the original VP7 software. Anyways, this means that I’m not going to go easy on On2; they’ve had (at least) 6 years to work on VP8, and a much larger dev team than x264’s to boot.

Of course this refers to the encoder and not the spec, but the spec has to predate the encoder (or at least parts of it). Certainly not "very raw, hardly out of development".

Reply Score: 3

Karitku Member since:
2006-01-12

...compared to H264, it's new. This isn't rocket science.

VP8 is older than H264! It has rather shitty specs with too many "look the code" style marks. When Theora came it was pretty much crap and not until lot of development it has become quite good but you can't turn shit to diamonds in decade.

Main reason why Microsoft isn't including VP8 on Windows or IE is due patent threat and it's real. You seriously think that closed source spec codec made by few coders is completely unique and that they never perhaps looked competition and asked "hmm that looks nice let's use it ourself"?

Reply Score: 1

PresentIt Member since:
2010-02-10

VP8 is not older than H264. It's almost as old, but the fact is that it is not as mature because H264 has had a lot more eyeballs and resources poured into it.

Now VP8 is out in the open, which means that it too will benefit from more eyeballs and resources.

It isn't the age, but the maturity. VP8 is simply not as mature. Yet.

Reply Score: 2

PresentIt Member since:
2010-02-10

Nonono. VP8 proponents tell us since two years (VP8 was releases back then) that it blows the entire h.264 standard out of the water.

Really? When and where?

Now a mere name change and a quality-unaffecting container switch to the proven and robust Matroska format makes VP8 "very raw, hardly out of development" and excuses for rather mediocre performance against h.264???

No, VP8 is only out there now, which means that there's a lot of development remaining. A lot of improvements can be made because it hasn't been optimized yet.

And the performance is not mediocre. It's actually better than H.264 baseline, which is what matters today.

Seriously, how about reading the damn article before whining?

Make up your mind.

The mind is made up already. You are just being dishonest and misrepresenting people's position.

Reply Score: 3

Radio Member since:
2009-06-20

http://img243.imageshack.us/img243/985/youtubevergleich.png

Top: WebM youtube video with the Opera nightly build.
Bottom: h264 youtube video with flash in firefox.

Platform: Win 7 64bits

Filesize: 35407kbyte for flash, 35183kbyte for webm.

See http://x264dev.multimedia.cx/?p=377#comment-4433

And all is said on *true* WebM quality.

Reply Score: 4

Neolander Member since:
2010-03-08

-WebM +VP8 =p

Reply Score: 1

1c3d0g Member since:
2005-07-06

See? Now we can debate, with hard FACTS such as these great screenshots. Every h264/x264 proponent keeps telling people WebM sucks, but ALL the evidence so far points to the contrary...my, my...perhaps a little scared, are we? :-P

Reply Score: 4

kaiwai Member since:
2005-07-06

See? Now we can debate, with hard FACTS such as these great screenshots. Every h264/x264 proponent keeps telling people WebM sucks, but ALL the evidence so far points to the contrary...my, my...perhaps a little scared, are we? :-P


I wouldn't be surprised; people sometimes mark themselves up as rational and reasonable people but ignore that emotions drive people behaviour where a choice is made then rationalising that choice comes secondary. They have a vested interest in x264 because they've poured their heart and soul into the project. That being said, they shouldn't need to view it as a threat but instead an opportunity to as end users to have more choice and as developers to learn from what Google and friends will do to VP8 to improve it for the long term.

Reply Score: 2

Won't matter
by darknexus on Thu 20th May 2010 23:56 UTC
darknexus
Member since:
2008-07-15

In the end, with every other browser vendor backing it, it won't matter how many hissy fits His Steveness wants to throw. If every other vendor backs it, Apple will either have to do the same or they'll lose what little marketshare they have. This is why Microsoft backed it, they had no real choice. As goes Youtube, so go the browser vendors.
It doesn't have to be the best, it just has to be good enough, and it is. Good enough + everyone else backing it == Apple in a corner. Mmmm, where's my popcorn and beer? I'm going to enjoy this.

Reply Score: 4

RE: Won't matter
by Fettarme H-Milch on Fri 21st May 2010 00:31 UTC in reply to "Won't matter"
Fettarme H-Milch Member since:
2010-02-16

In the end, with every other browser vendor backing it, it won't matter how many hissy fits His Steveness wants to throw. If every other vendor backs it, Apple will either have to do the same or they'll lose what little marketshare they have. This is why Microsoft backed it, they had no real choice.

Dude, get a reality check.
Microsoft said that IE9 will play VP8 if a codec for it is installed. MS did not say that VP8 will be shipped with IE9.

This is exactly the same approach Apple takes since years with Safari: Safari plays back everything inside the HTML5 video tag as long as a codec for it is installed.

Reply Score: 2

RE[2]: Won't matter
by chandler on Fri 21st May 2010 01:52 UTC in reply to "RE: Won't matter"
chandler Member since:
2006-08-29

It's not the same situation. Previously Microsoft had stated that only H.264 would be supported. System video codecs for other formats (like VC-1 and Theora) would not be used even if installed. This is a prudent posture since not all of these codecs have been screened to the same level for security.

Now Microsoft is offering another option provided you bring the codec yourself. Of course we can expect Google to be pushing the WebM codecs on as many computers as possible using Chrome and their other software.

Reply Score: 3

RE[3]: Won't matter
by Fettarme H-Milch on Fri 21st May 2010 09:28 UTC in reply to "RE[2]: Won't matter"
Fettarme H-Milch Member since:
2010-02-16

Of course we can expect Google to be pushing the WebM codecs on as many computers as possible using Chrome and their other software.

That's what sane people said to Mozilla since years: Use the system's media framework, but install the Ogg's family of codecs along with Firefox.

Reply Score: 2

RE[3]: Won't matter
by kaiwai on Sat 22nd May 2010 02:49 UTC in reply to "RE[2]: Won't matter"
kaiwai Member since:
2005-07-06

It's not the same situation. Previously Microsoft had stated that only H.264 would be supported. System video codecs for other formats (like VC-1 and Theora) would not be used even if installed. This is a prudent posture since not all of these codecs have been screened to the same level for security.

Now Microsoft is offering another option provided you bring the codec yourself. Of course we can expect Google to be pushing the WebM codecs on as many computers as possible using Chrome and their other software.


What a load of bull crap; Microsoft said that they wouldn't be supporting more than h264 but other codecs could be supported via the Windows Media Player activeX plugin. The situation hasn't changed on iota except for people like you, clueless as to the situation, crowing about the supposed change by the prodigal son named Microsoft.

Reply Score: 2

RE: Won't matter
by MollyC on Fri 21st May 2010 05:15 UTC in reply to "Won't matter"
MollyC Member since:
2006-07-04

I agree. But this is a clear instance of a company using its monopoly power to push a codec that isn't as good as the already accepted standard. Hmmm....

Reply Score: 0

RE[2]: Won't matter
by Valhalla on Fri 21st May 2010 05:25 UTC in reply to "RE: Won't matter"
Valhalla Member since:
2006-01-24

I agree. But this is a clear instance of a company using its monopoly power to push a codec that isn't as good as the already accepted standard. Hmmm....


When did 'we' accept that standard? I must have missed that vote. Hasn't this whole debate of late been because alot of people does not accept h.264 to be the web standard?

Reply Score: 4

RE[3]: Won't matter
by MollyC on Sun 23rd May 2010 00:22 UTC in reply to "RE[2]: Won't matter"
MollyC Member since:
2006-07-04

Maybe you didn't accept H.264 as the standard (nor did I, nor did the general public), but the industry had*, until Google used it's monopoly power to change that.

In theory, a patent holder in the H.264 patent pool (I don't refer to Microsoft or Apple, but one of the many others in that patent pool) could sue Google for abusing its monopoly in the web video market to stamp out usage of H.264 on the web, thus denying that patent holder the income it would otherwise have received. I don't support that, but I could see someone making that case in theory.

If VP8 were clearly superior, that would be one thing. But apparently H.264 is superior, yet is being denied its rightful place in the market due to Google's monopoly power. I don't see how anyone could really deny that that's what's happenning here.



* Generally speaking, H.264 had been accepted by Flash (I think Flash's HD vids are H.264), the professional content creation industry (because it has DRM support), Apple (which controls the mobile space), Blu Ray (H.264 is one of the three standard Blu Ray codecs (along with MPEG2 and VC-1), so one could use the same tools to encode a Blu Ray video and a web video), Silverlight (which uses VC-1 and H.264 as both are suitable to be packaged for adaptive streaming and have DRM support), etc. As I said above, the public didn't really choose H.264 (they don't even know what a codec is, let alone H.264 specifically), but lots of industry was already using H.264.

Edited 2010-05-23 00:27 UTC

Reply Score: 2

RE[4]: Won't matter
by lemur2 on Sun 23rd May 2010 10:50 UTC in reply to "RE[3]: Won't matter"
lemur2 Member since:
2007-02-17

Maybe you didn't accept H.264 as the standard (nor did I, nor did the general public), but the industry had*, until Google used it's monopoly power to change that.


Err, no. The standard we are talking about is HTML5, in particular the video tag within HTML5, and the codec to be used for that video.

Like most web standards, HTML5 is a W3C standard.

W3C policy is that unless there is absolutely no alternative, then only royalty-free technologies will be specified by W3C standards.

http://www.w3.org/Consortium/Patent-Policy-20040205/
Abstract
The W3C Patent Policy governs the handling of patents in the process of producing Web standards. The goal of this policy is to assure that Recommendations produced under this policy can be implemented on a Royalty-Free (RF) basis.


In the case of the HTML5 specification, in line with their policy, W3C originally stipulated Theora as the video codec. At the time, Theora was appreciably inferior to H.264, and there was sufficient dissent that W3C had to remove Theora from the HTML5 specification because of this lack of consensus.

However, Theora was NOT replaced by H.264. Currently, the HTML5 specification makes no recommendation about the video codec to be used. So H.264 was never in the HTML5 standard. It still isn't.

In theory, a patent holder in the H.264 patent pool (I don't refer to Microsoft or Apple, but one of the many others in that patent pool) could sue Google for abusing its monopoly in the web video market to stamp out usage of H.264 on the web, thus denying that patent holder the income it would otherwise have received. I don't support that, but I could see someone making that case in theory.


Hardly. H.264 doesn't meet the primary policy requirement of a W3C standard, in that it is not royalty-free.

If VP8 were clearly superior, that would be one thing. But apparently H.264 is superior, yet is being denied its rightful place in the market due to Google's monopoly power. I don't see how anyone could really deny that that's what's happenning here.


I'll deny it, most definitely.

H.264 isn't superior to VP8, BTW.

http://www.osnews.com/story/23342/First_Look_VP8_vs_H264

* Generally speaking, H.264 had been accepted by Flash (I think Flash's HD vids are H.264), the professional content creation industry (because it has DRM support), Apple (which controls the mobile space), Blu Ray (H.264 is one of the three standard Blu Ray codecs (along with MPEG2 and VC-1), so one could use the same tools to encode a Blu Ray video and a web video), Silverlight (which uses VC-1 and H.264 as both are suitable to be packaged for adaptive streaming and have DRM support), etc. As I said above, the public didn't really choose H.264 (they don't even know what a codec is, let alone H.264 specifically), but lots of industry was already using H.264.


They were using it (perhaps short-sightedly enjoying the royalty amnesty period the MPEG LA had allowed for H.264), but it wasn't the standard. Until MPEG LA promise that it can be used royalty-free by all parties forever more (as Google have recently declared for VP8), then H.264 isn't even eligible as the web video codec standard.

Edited 2010-05-23 10:52 UTC

Reply Score: 2

RE[4]: Won't matter
by PresentIt on Sun 23rd May 2010 13:39 UTC in reply to "RE[3]: Won't matter"
PresentIt Member since:
2010-02-10

Maybe you didn't accept H.264 as the standard (nor did I, nor did the general public), but the industry had*, until Google used it's monopoly power to change that.

Monopolies aren't automatically bad, especially in cases like this where the "monopoly" (you are probably referring to YouTube) is used to create a free and open standard, which benefits the entire market.

In theory, a patent holder in the H.264 patent pool (I don't refer to Microsoft or Apple, but one of the many others in that patent pool) could sue Google for abusing its monopoly in the web video market to stamp out usage of H.264 on the web, thus denying that patent holder the income it would otherwise have received. I don't support that, but I could see someone making that case in theory.

Monopoly is only abused when it's to the detriment of the market. Google's use of its monopoly in this case is clearly beneficial for the market as whole.

If VP8 were clearly superior, that would be one thing. But apparently H.264 is superior, yet is being denied its rightful place in the market due to Google's monopoly power. I don't see how anyone could really deny that that's what's happenning here.

I deny that. VP8 is better than H.264 baseline, which is all that matters (because baseline is the only thing all those devices out there support).

And even if it wasn't, Google has not abused its monopoly power. Instead, it has used it to strengthen free and open standards. The very opposite of abuse.

Reply Score: 1

RE[2]: Won't matter
by lemur2 on Fri 21st May 2010 05:36 UTC in reply to "RE: Won't matter"
lemur2 Member since:
2007-02-17

I agree. But this is a clear instance of a company using its monopoly power to push a codec that isn't as good as the already accepted standard. Hmmm....


There is no evidence to suggest that VP8 isn't as good as H.264 (as used by some on the web). Despite its use on the web, H.264 is NOT the standard for the video codec for use on the web. (W3C originally proposed Theora, Apple refused to accept that, Microsoft had nothing to say on the matter, and so right now there is no standard codec for the web).

Edited 2010-05-21 05:36 UTC

Reply Score: 4

RE[3]: Won't matter
by hornett on Fri 21st May 2010 08:42 UTC in reply to "RE[2]: Won't matter"
hornett Member since:
2005-09-19

There is no evidence to suggest that VP8 isn't as good as H.264 (as used by some on the web).


Yes there is. It's linked from the article.

X264 in hq mode and baseline:
http://doom10.org/compare/x264.png
http://doom10.org/compare/x264baseline.png

VP8:
http://doom10.org/compare/vp8.png

That's not to say that it cannot be improved, of course.

My major concern is that the codec seems to be very poorly specced.

Reply Score: 2

RE[2]: Won't matter
by chris_l on Sat 22nd May 2010 10:48 UTC in reply to "RE: Won't matter"
chris_l Member since:
2010-02-14

I agree. But this is a clear instance of a company using its monopoly power to push a codec that isn't as good as the already accepted standard. Hmmm....


Accepted by who? Certainly not by Firefox or any of the major players in the Free Software/Open Source World, and quite frankly that's the only thing that really matters here.

Nobody here really cares what Apple does or doesn't do concerning the piece-of-shit called the Ipad.

Same concerning Microsoft and their crap.

Reply Score: 1

RE[2]: Won't matter
by Soulbender on Sat 22nd May 2010 13:15 UTC in reply to "RE: Won't matter"
Soulbender Member since:
2005-08-18

Yeah, isn't it disgusting how Apple is pushing for H264 instead of the standard, open VP8. That's what you mean, right?

Reply Score: 2

Not an opinion
by umccullough on Fri 21st May 2010 00:07 UTC
umccullough
Member since:
2006-01-26

Bah, Steve's response is hardly an opinion... Linking someone to a 3rd party review of the codec doesn't say much other than: "We saw this review."

What it does suggest is that Apple is actually interested in how it stacks up against h.264 - probably before they will make any real public statements about it.

Reply Score: 1

Patents not quality
by VistaUser on Fri 21st May 2010 00:27 UTC
VistaUser
Member since:
2008-03-08

While people seem to be commenting on the quality aspect of the link that Steve Jobs linked to, this may be missing the bigger menace.

It may be more a veiled threat about patents which the blog suggested that VP8 could potentially infringe.

I would love to see a good quality comparison between the codece though - kind of like the theora "improvements" that monty from Xiph did to show the improvements.

Edited 2010-05-21 00:28 UTC

Reply Score: 1

v RE: Patents not quality
by Fettarme H-Milch on Fri 21st May 2010 00:37 UTC in reply to "Patents not quality"
RE[2]: Patents not quality
by galvanash on Fri 21st May 2010 01:00 UTC in reply to "RE: Patents not quality"
galvanash Member since:
2006-01-25

I dont know if you have any clue what you are talking about, but if you do - that is simply a completely untrue statement. While some aspects of VP8 do have some similarities to h.264 baseline (as pointed out in the technical analysis) there are other things it does completely differently. Even the similar parts appear to be different enough to avoid patent issues - Google did study this thing before pulling the trigger...

VP8 is VP7 with some refinements. Before that you had VP6, VP5, VP4, VP3, etc. and so on. The fact that this is the last in a line of codecs that stretches back FAR before h.264 came into existence is relevent, and the fact that MPEG-LA has never attempted to collect royalties on any of On2's previous work is going to work against them to at least some degree if they try to litigate now...

Granted patent law is not like trademarks, but ignoring something for this long and then suddenly giving a shit is not going to work in their favor.

Reply Score: 5

RE[3]: Patents not quality
by lemur2 on Fri 21st May 2010 01:26 UTC in reply to "RE[2]: Patents not quality"
lemur2 Member since:
2007-02-17

I dont know if you have any clue what you are talking about, but if you do - that is simply a completely untrue statement. While some aspects of VP8 do have some similarities to h.264 baseline (as pointed out in the technical analysis) there are other things it does completely differently. Even the similar parts appear to be different enough to avoid patent issues - Google did study this thing before pulling the trigger... VP8 is VP7 with some refinements. Before that you had VP6, VP5, VP4, VP3, etc. and so on. The fact that this is the last in a line of codecs that stretches back FAR before h.264 came into existence is relevent, and the fact that MPEG-LA has never attempted to collect royalties on any of On2's previous work is going to work against them to at least some degree if they try to litigate now... Granted patent law is not like trademarks, but ignoring something for this long and then suddenly giving a shit is not going to work in their favor.


You don't know for a fact that MPEG LA have ignored the On2 lineage of codecs. It may simply be the case that MPEG LA have scrutinised the whole series of On2's codecs, and have found no infringements via which to attack (just as Google has done now).

Edited 2010-05-21 01:27 UTC

Reply Score: 2

RE[4]: Patents not quality
by galvanash on Fri 21st May 2010 02:02 UTC in reply to "RE[3]: Patents not quality"
galvanash Member since:
2006-01-25

I find that _highly_ doubtful since they claim to be putting together a patent pool in an attempt to license/litigate VP8 as we speak (there is a link to the article with the quote from MPEG-LA in this thread).

So either they know there are no patent issues and they are lying in an attempt to spread FUD, or they do think there are patent issues and have simply not bothered to pursue them because On2 wasn't a particularly wealthy target. Either way they are just posturing assholes.

Edited 2010-05-21 02:07 UTC

Reply Score: 2

RE[5]: Patents not quality
by lemur2 on Fri 21st May 2010 02:17 UTC in reply to "RE[4]: Patents not quality"
lemur2 Member since:
2007-02-17

I find that _highly_ doubtful since they claim to be putting together a patent pool in an attempt to license/litigate VP8 as we speak (there is a link to the article with the quote from MPEG-LA in this thread). So either they know there are no patent issues and they are lying in an attempt to spread FUD, or they do think there are patent issues and have simply not bothered to pursue them because On2 wasn't a particularly wealthy target. Either way they are just posturing assholes.


The only part that is not doubtful is that Google have done a thorough patent search viz a via VP8, and they believe it to be not infringing of any other patents, and they have stated this position, and indeed they only released VP8 as open source and royalty-free because they believe it to be not infringing.

http://www.osnews.com/permalink?425732

Edited 2010-05-21 02:22 UTC

Reply Score: 2

RE[6]: Patents not quality
by tf123 on Fri 21st May 2010 02:21 UTC in reply to "RE[5]: Patents not quality"
tf123 Member since:
2010-01-28

Again, Microsoft said the exact same thing about VC-1; where is VC-1 now? Oh, that's right: it's an MPEGLA patent pool that has a small fraction of contributions from Microsoft and is hardly indistinguishable from the main MPEG-4 and AVC patents but for a couple of algorithms.

Why would any company say anything but, "We reviewed it and are confident it is all ours"?

Reply Score: 2

RE[7]: Patents not quality
by lemur2 on Fri 21st May 2010 02:41 UTC in reply to "RE[6]: Patents not quality"
lemur2 Member since:
2007-02-17

Again, Microsoft said the exact same thing about VC-1; where is VC-1 now? Oh, that's right: it's an MPEGLA patent pool that has a small fraction of contributions from Microsoft and is hardly indistinguishable from the main MPEG-4 and AVC patents but for a couple of algorithms. Why would any company say anything but, "We reviewed it and are confident it is all ours"?


Think about this for a second.

1. Microsoft release VC1 and claim it to be patent free. Good PR.
2. Microsoft include VC1 in Silverlight.
3. A short while later, a patent troll pops out of the woodwork, and sues Microsoft.
4. Microsoft immediately caves without a fight, and pays money to the patent troll.
5. Microsoft say sorry everyone, but in order to use VC1 you are going to have to get a licensed binary from us. No independent implementations allowed.
6. Microsoft/Novell include a provision in Moonlight so that Linux systems running Moonlight now download a binary dependency on Microsoft.
7. Silverlight/Moonlight is constrained to run only on x86 systems (bad luck, ARM).

All done while Microsoft appear (PR-wise) to be trying to do the right thing.

Quite neat, really. A pity (from a Microsoft POV) that it didn't work.

Reply Score: 0

RE[6]: Patents not quality
by galvanash on Fri 21st May 2010 02:31 UTC in reply to "RE[5]: Patents not quality"
galvanash Member since:
2006-01-25

I agree. But they didn't take the last step off the ledge and indemnify their users. That tells me they are at the very least hedging their bets a wee bit.

I think Google is taking a page from the Bush administration - they are attempting to create a "coalition of the willing" so to speak ;) Hope it works out better for them. Anyway, between Google and everyone else who has thrown in with them so far, they may have enough clout in silicon valley to simply make MPEG-LA shut up and leave them alone, or at least scare the companies who actually hold the patents into careful consideration before setting off the nuke that a patent litigation case over this would be.

In a nutshell I would wager Google is about 95% confident they can defend VP8 in patent litigation, but if it comes to that they are making it very clear that their will be hell to pay for going down that road...

Reply Score: 2

RE[7]: Patents not quality
by cycoj on Fri 21st May 2010 17:00 UTC in reply to "RE[6]: Patents not quality"
cycoj Member since:
2007-11-04

I agree. But they didn't take the last step off the ledge and indemnify their users. That tells me they are at the very least hedging their bets a wee bit.

I think Google is taking a page from the Bush administration - they are attempting to create a "coalition of the willing" so to speak ;) Hope it works out better for them. Anyway, between Google and everyone else who has thrown in with them so far, they may have enough clout in silicon valley to simply make MPEG-LA shut up and leave them alone, or at least scare the companies who actually hold the patents into careful consideration before setting off the nuke that a patent litigation case over this would be.

In a nutshell I would wager Google is about 95% confident they can defend VP8 in patent litigation, but if it comes to that they are making it very clear that their will be hell to pay for going down that road...


Well none of the other players are offering indemnification either. If you get a license from the MPEG-LA that does not protect you from someone else who believes he has patents which cover h264 so you're not any safer than going with VP8. That's something everyone is constantly ignoring.

Reply Score: 2

RE[6]: Patents not quality
by tf123 on Fri 21st May 2010 02:35 UTC in reply to "RE[5]: Patents not quality"
tf123 Member since:
2010-01-28

Moreover, they created a new license where the only new provision was: if you engage in patent litigation against VP8, the license is no longer valid because they are very confident it is free of patents. Riiiiighht.

Reply Score: 1

RE[7]: Patents not quality
by lemur2 on Fri 21st May 2010 02:46 UTC in reply to "RE[6]: Patents not quality"
lemur2 Member since:
2007-02-17

Moreover, they created a new license where the only new provision was: if you engage in patent litigation against VP8, the license is no longer valid because they are very confident it is free of patents. Riiiiighht.


I think you may have missed the gist of this. The open license from Google to use the patents covering VP8 becomes no longer valid for the party who engaged in the patent litigation against VP8.

Therefore, if MPEG LA engage in patent litigation against VP8, then no-one using H.264 (or indeed any MPEG LA product) after that point will be licensed any longer to use any method of H.264 that happens to be covered by any ex-On2 (now Google) patent.

Edited 2010-05-21 02:55 UTC

Reply Score: 2

RE[3]: Patents not quality
by Fettarme H-Milch on Fri 21st May 2010 10:01 UTC in reply to "RE[2]: Patents not quality"
Fettarme H-Milch Member since:
2010-02-16

MPEG-LA has never attempted to collect royalties on any of On2's previous work

Maybe that's because the source code was not as easily accessible as it's now.

Just because VP8 does some things different, it does not mean that the aspects it does the same are not patented.

Reply Score: 2

RE[2]: Patents not quality
by Valhalla on Fri 21st May 2010 01:37 UTC in reply to "RE: Patents not quality"
Valhalla Member since:
2006-01-24


Didn't you get the memo? When Google releases the source code to a modified H264 Baseline codec, it does not infringe patents, because a.) the codec has a different name and b.) Google possesses magical fairy dust that makes all patent claims disappear. As a side result, btw, x264 is now also free of any patents.


Don't make things up, using Garrett-Glaser's comment of -"With regard to patents, VP8 copies way too much from H.264 for anyone sane to be comfortable with it, no matter whose word is behind the claim of being patent-free." and try to turn it into vp8 being 'a modified h264 baseline codec' comes of as quite desperate. As for the patents, Garrett-Glaser ignores patents so he is hardly qualified to make that judgement. Now he can definately make code comparisons between vp8 and x264 and make speculations, but I doubt he has any idea of wether the techniques in question are patented, or if they should hold up in court. Howerver, I think it's pretty reasonable to believe that Google themselves has a pretty good idea of exactly what is patented or not, it's sort of in their interest.

Reply Score: 3

RE[3]: Patents not quality
by Fettarme H-Milch on Fri 21st May 2010 10:06 UTC in reply to "RE[2]: Patents not quality"
Fettarme H-Milch Member since:
2010-02-16

As for the patents, Garrett-Glaser ignores patents so he is hardly qualified to make that judgement.

Why does implementing patented techniques result in incapability to identify patents? After all, he works with them.

He's likely more qualified to comment on patents than some Wikipedia dude or an Opera guy who both do not a relationship as deep with multimedia codecs as the x264 guy has.

Reply Score: 2

RE[4]: Patents not quality
by lemur2 on Fri 21st May 2010 10:20 UTC in reply to "RE[3]: Patents not quality"
lemur2 Member since:
2007-02-17

"As for the patents, Garrett-Glaser ignores patents so he is hardly qualified to make that judgement.

Why does implementing patented techniques result in incapability to identify patents? After all, he works with them.
"

If you willingly infringe on a patent, the court will award triple damages against you. This is a very, very good reason to avoid even reading someone else's patents.

He's likely more qualified to comment on patents than some Wikipedia dude or an Opera guy who both do not a relationship as deep with multimedia codecs as the x264 guy has.


However, he is nowhere near as qualified to comment on patents as a party that was developing codecs themselves at the same time as MPEG LA, and applied for patents themselves, and had some of their applications granted and others rejected. Hello, On2.

He is nowhere near as qualified to comment on patents as a party that was awarding/rejecting patents, and informing applicants of the reasons (if any) for rejection. Hello, USPTO.

Reply Score: 2

RE[5]: Patents not quality
by Fettarme H-Milch on Fri 21st May 2010 10:34 UTC in reply to "RE[4]: Patents not quality"
Fettarme H-Milch Member since:
2010-02-16

If you willingly infringe on a patent, the court will award triple damages against you. This is a very, very good reason to avoid even reading someone else's patents.

x264 is hosted in France, not the US. US courts have no authority there.

Reply Score: 3

RE[6]: Patents not quality
by lemur2 on Fri 21st May 2010 10:57 UTC in reply to "RE[5]: Patents not quality"
lemur2 Member since:
2007-02-17

"If you willingly infringe on a patent, the court will award triple damages against you. This is a very, very good reason to avoid even reading someone else's patents.

x264 is hosted in France, not the US. US courts have no authority there.
"

x264 developers are reportedly trying to convert their product to closed source, and start to charge people for it.

If they wish to sell x264 to anyone who does business in the US, then they will have to address the issue.

Reply Score: 2

RE[7]: Patents not quality
by Fettarme H-Milch on Sat 22nd May 2010 00:07 UTC in reply to "RE[6]: Patents not quality"
Fettarme H-Milch Member since:
2010-02-16

x264 developers are reportedly trying to convert their product to closed source, and start to charge people for it.

Any proof of that?

Reply Score: 2

RE: Patents not quality
by lemur2 on Fri 21st May 2010 01:13 UTC in reply to "Patents not quality"
lemur2 Member since:
2007-02-17

While people seem to be commenting on the quality aspect of the link that Steve Jobs linked to, this may be missing the bigger menace. It may be more a veiled threat about patents which the blog suggested that VP8 could potentially infringe.


After doing an apparently thorough search, Google are saying they are very confident that it doesn't infringe.

http://www.theregister.co.uk/2010/05/20/google_confident_on_vp8_and...
In a private email, Apple chief executive Steve Jobs even went so far as to say that unnamed forces were putting together a patent pool to "go after" Ogg Theora. Today, when The Reg asked if VP8 was vulnerable to patent attack, Google product manager Mike Jazayeri indicated this isn't a big concern for the company.

"We have done a pretty through analysis of VP8 and On2 Technologies prior to the acquisition and since then, and we are very confident with the technology and that's why we're open sourcing," he said.


My bold.

I would love to see a good quality comparison between the codece though - kind of like the theora "improvements" that monty from Xiph did to show the improvements.


There is no reason why the kind of optimisations that Xiph have applied to Theora for Theora 1.2 (Ptalarbvorm), which are still ongoing, cannot be applied to WebM also.

http://people.xiph.org/~xiphmont/demo/theora/demo9.html

In particular, Activity masking, Altered Skip weighting, quantization matrix adjustments and Temporal RDO techniques might be useful.

Because of their nature, I'd imagine that the earlier improvements Xiph made to Theora (for Thusnelda), however, probably don't apply to WebM.

Likewise, since VP8 patents are released royalty-free by Google, there is a possibility that some of those techniques could be back-ported to Theora (as long as the file structure supports it).

Edited 2010-05-21 01:15 UTC

Reply Score: 2

RE[2]: Patents not quality
by Fettarme H-Milch on Fri 21st May 2010 10:11 UTC in reply to "RE: Patents not quality"
Fettarme H-Milch Member since:
2010-02-16

"We have done a pretty through analysis of VP8 and On2 Technologies prior to the acquisition and since then, and we are very confident with the technology and that's why we're open sourcing," he said.

My bold.

Microsoft said the same about VC-1 a few years ago when they opened up Windows Media to the ITU consortium as competing standard against MPEG-4.
Turned out that Microsoft is not the sole patent holder. VC-1 is covered by fewer patents than MPEG-4 AVC, but it's still covered by 3rd parties' patents.

Reply Score: 2

RE[3]: Patents not quality
by lemur2 on Fri 21st May 2010 10:14 UTC in reply to "RE[2]: Patents not quality"
lemur2 Member since:
2007-02-17

""We have done a pretty through analysis of VP8 and On2 Technologies prior to the acquisition and since then, and we are very confident with the technology and that's why we're open sourcing," he said.

My bold.

Microsoft said the same about VC-1 a few years ago when they opened up Windows Media to the ITU consortium as competing standard against MPEG-4.
Turned out that Microsoft is not the sole patent holder. VC-1 is covered by fewer patents than MPEG-4 AVC, but it's still covered by 3rd parties' patents.
"

There is a train of thought that leads one to speculate that Microsoft might have wanted this to be the outcome.

Reply Score: 2

RE[4]: Patents not quality
by Fettarme H-Milch on Fri 21st May 2010 10:41 UTC in reply to "RE[3]: Patents not quality"
Fettarme H-Milch Member since:
2010-02-16

There is a train of thought that leads one to speculate that Microsoft might have wanted this to be the outcome.

Losing against MPEG is a desired outcome for Microsoft??? What?!?!
If VC-1 had won, Microsoft would be the premier provider of encoding tools that -- of course -- only work on Windows, hence pushing Macs out of the media creator's desks. Microsoft would have the premier streaming server lineup.
Ect. pp.

Reply Score: 2

RE[5]: Patents not quality
by lemur2 on Fri 21st May 2010 11:01 UTC in reply to "RE[4]: Patents not quality"
lemur2 Member since:
2007-02-17

"There is a train of thought that leads one to speculate that Microsoft might have wanted this to be the outcome.

Losing against MPEG is a desired outcome for Microsoft??? What?!?!
If VC-1 had won, Microsoft would be the premier provider of encoding tools that -- of course -- only work on Windows, hence pushing Macs out of the media creator's desks. Microsoft would have the premier streaming server lineup.
Ect. pp.
"

Microsoft got good PR for trying to release VC1 as open, and yet now it is somehow not open and it cannot be used by any open source competitor.

Microsoft get their video codec lock-in, they get no bad PR for having that lock-in, and Macs can easily be killed as competition later on.

Reply Score: 2

RE[5]: Patents not quality
by tf123 on Fri 21st May 2010 12:11 UTC in reply to "RE[4]: Patents not quality"
tf123 Member since:
2010-01-28

There was no "LOSING". Microsoft voluntarily submitted the format to the ISO themselves which they knew would involve a patent review. Either way, whether or not they intended it or not -- it would appear they were very stupid about their misappropriation of patents. The point being: too many people hear are claiming there is no way this could happen because Google "is smarter than that." Well, if Microsoft didn't intend for VC-1 to be overtaken by MPEGLA, they were VERY, very, very stupid. And they're pretty smart too.

Reply Score: 2

RE[5]: Patents not quality
by MissTJones on Fri 21st May 2010 12:40 UTC in reply to "RE[4]: Patents not quality"
MissTJones Member since:
2010-03-25

How exactly would creating a royalty-free, re-implementable standard from VC-1 have locked people into Windows?

They might have had a head start since they already had an implementation, but assuming it wasn't a joke of a standard (like their OOXML) then it would have been easily re-implementable by others. And the best version would have likely been an open source rewrite, just like it is for H.264 and x264, and probably will be the case for VP8 and some as yet unlaunched project.

Reply Score: 2

RE[6]: Patents not quality
by lemur2 on Fri 21st May 2010 13:00 UTC in reply to "RE[5]: Patents not quality"
lemur2 Member since:
2007-02-17

How exactly would creating a royalty-free, re-implementable standard from VC-1 have locked people into Windows?


Official Microsoft PR spin (translated): Microsoft alone can supply VC1 with all the required permissions (licenses), unfortunately <sarcasm>because of that nasty patent troll that attacked VC1</sarcasm>, and unfortunately the trolls insist it must be closed with no open implementation. Because it is mostly Microsoft IP, with a bit of IP owned by that nasty troll, the combination is only legally available from Microsoft. This is clearly not Microsoft's fault. And oh, BTW, Microsoft offers it to you only if you have one of these Microsoft-endorsed platforms to run it on ... <list of platforms that excludes whatever Microsoft feels like excluding>.

Edited 2010-05-21 13:03 UTC

Reply Score: 2

RE: Patents not quality
by Luis on Fri 21st May 2010 12:16 UTC in reply to "Patents not quality"
Luis Member since:
2006-04-28

While people seem to be commenting on the quality aspect of the link that Steve Jobs linked to, this may be missing the bigger menace.

It may be more a veiled threat about patents which the blog suggested that VP8 could potentially infringe.

I would love to see a good quality comparison between the codece though - kind of like the theora "improvements" that monty from Xiph did to show the improvements.


He did comment a few things in his blog. I asked about the quality of VP8 and this is what he said:

Quality of the format: Theora was good enough and VP8 is solidly better than Theora (even current Ptalar). So, awesome, it's a pretty pure win.

I know everyone wants a 'but how does it compare to h.264' answer, or more specifically 'how does it compare to x264'. A: Total red herring; none of the orgs with money involved care much about that answer. It's only good for trash talking, not actually winning a fight. B: I don't actually know for sure yet. We've been concentrating on evaluating the spec and the codebase quality rather than drag racing it, because pinning VP8's success on a size contest completely misses the point.

Theora was already beating some h.264 encoders out there (no, not x.264) and VP8 at very least already beats more. As far as the sourcebase goes, it raises the ceiling over Theora on how far we can improve things. Neither h.264 nor VP8 will be holding still.

...but once I have a clearer picture of the answer to B, I'll still fess up ;-) Chances are the answer is 'about the same', at least right now. We'll see.


He also has comments about the spec and the patents:
http://xiphmont.livejournal.com/50239.html

Reply Score: 2

Why am I being stuck with Baseline?
by tf123 on Fri 21st May 2010 00:28 UTC
tf123
Member since:
2010-01-28

What is Maxwell's and Haavard's rationale for claiming that no internet video should ever target anything greater than Baseline? Google is telling me I can have fullscreen Flash video on my 50 inch tv today; I watch HD-quality web video at fullscreen on my PC every day. Why is anything superior to AVC Baseline out of the question?

Reply Score: 2

Delgarde Member since:
2008-08-19

Why is anything superior to AVC Baseline out of the question?


It's not that it's out of the question, but H264 is a fairly complicated spec, and right now, almost nothing implements more than just the baseline. Which is important when doing comparisons - there's little point in criticising VP8/WebM for not having all of the H264 features, when in practice, H264 implementations don't have those features either.

Reply Score: 5

tf123 Member since:
2010-01-28

Baloney. As I said, I watch AVC Main video on the internet on a daily basis ... all the way on up to High 4:4:4: Predictive.

If VP8 is intended to replace or supplant .flv, it is certainly fair to compare it to any of the Profiles and Levels supported by Flash.

Reply Score: 1

PresentIt Member since:
2010-02-10

What is Maxwell's and Haavard's rationale for claiming that no internet video should ever target anything greater than Baseline?

Did they make that claim?

No.

They pointed out that if you wanted to target all those devices and mobile phones out there with H.264, you would be stuck with the baseline profile.

That's because that's what the hardware supports.

And lack of hardware support was one of the main arguments against Theora, remember?

Reply Score: 1

A raw IMPLEMENTATION of a mature codec.
by tomz on Fri 21st May 2010 00:37 UTC
tomz
Member since:
2010-05-06

The VP8 codec as shipped was probably better with all the positive comparisons. But that as such - for whatever reason - is not what Google released. I suspect there were some deep optimizations that were backed out to make it portable C for the spec and for the library.

Even Theora has an optimized ARM version.

It appears that WebM is VP8, two steps back to make it open and portable. Those steps should be retraced quickly.

Reply Score: 1

galvanash Member since:
2006-01-25

The implementation of VP8 in WebM did not get worse than at release, h.264 implementations simply got better...

Reply Score: 2

Royalty Free?
by tuzor on Fri 21st May 2010 00:42 UTC
tuzor
Member since:
2007-08-07
RE: Royalty Free?
by lemur2 on Fri 21st May 2010 01:22 UTC in reply to "Royalty Free?"
lemur2 Member since:
2007-02-17



I'll match your link to mere speculation, and raise you this link which includes a direct quote from a Google rep that Google have done a thorough investigation, and are confident that WebM does not infringe any other patents.

http://www.theregister.co.uk/2010/05/20/google_confident_on_vp8_and...

Royalty-free? Yes indeed.

Edited 2010-05-21 01:28 UTC

Reply Score: 4

RE[2]: Royalty Free?
by tf123 on Fri 21st May 2010 01:35 UTC in reply to "RE: Royalty Free?"
tf123 Member since:
2010-01-28

Microsoft said WMV9/VC-1 was solely their own creation as well. What's your point? Am I supposed to expect Google to say, "Yeah, there's definitely some bits that are really damn iffy and some parts we aren't sure about, but hey, we're Google so we hope to slip this past everyone." No, I do not.

Reply Score: 2

RE[3]: Royalty Free?
by Lazarus on Fri 21st May 2010 05:24 UTC in reply to "RE[2]: Royalty Free?"
Lazarus Member since:
2005-08-10

Microsoft said WMV9/VC-1 was solely their own creation as well. What's your point? Am I supposed to expect Google to say, "Yeah, there's definitely some bits that are really damn iffy and some parts we aren't sure about, but hey, we're Google so we hope to slip this past everyone." No, I do not.


My half-joking response follows:

If my experience with Microsoft's search offerings are any indication, I'm not surprised they weren't able to find any information regarding patents held by other companies that may have covered portions of VC-1.

Poor bastards probably just got ads for Patent offices and porn.

Reply Score: 3

VP8 Analysis
by galvanash on Fri 21st May 2010 00:43 UTC
galvanash
Member since:
2006-01-25

I read the technical analysis of VP8. Although the article certainly carries quite a bit of bias towards h.264 considering who wrote it, the technical points are probably not far from the truth. By that I mean that I seriously doubt VP8 will become _better_ than h.264 in the near future, if ever. But it is still pretty good, even the writer of the article essentially admitted it was pretty descent.

Regardless, it is definitely good enough. Hell, Theora was good enough. I don't much care that it is not technically superior to h.264 - as I said many times before being royalty free counts for a lot more than technical capability when it comes to the web.

I personally don't expect Apple to come onboard with WebM though - they are too heavily vested in h.264. They may add support for it, but they are never likely to put any weight behind it. I would assume Desktop Safari on OSX will be able to support it as soon as a QT plugin i available for it, but their mobile stuff is a different matter...

It isn't just a matter of hardware support - they also have the issue of HTML live streaming which Apple has put a lot of effort into. That is the only method of streaming Apple currently promotes for sending video over 3G networks, and it is currently defined only for h.264/aac streams in MPEG2-TS containers. You might be able to use VP8 in an HTML live stream, but it would have to be in an MPEG2-TS container to be valid as per the current spec, which makes it essentially something other than WebM which defeats the purpose.

The point is this streaming method is _already_ pretty much a defacto standard within Apple's infrastructure - and no one else currently supports it. They don't seem to care much about broad industry support at this point, they are content to solve their own problems in their own way. It is published as an open spec, but the fact that it is so narrowly defined and requires h.264 makes it unlikely to be picked up by the rest of the world. I realize the same can be said for WebM, but being royalty free makes the fact that it is narrowly defined kind of a moot point...

Whatever, Apple always does their own thing - and the rest of the world moves on.

Reply Score: 3

couple of points (vs theora)
by bnolsen on Fri 21st May 2010 01:24 UTC
bnolsen
Member since:
2006-01-06

theora's big advantage was that it decodes with 30% less cpu resources than h264. vp8 is apparently the same as h264. People may comment about mobile dedicated hardware. My take on those responses isn't *if* theora/vp8 can be hardware accelerated but by who (advantage vp8 here).

google having frozen vp8's features (b frames, etc) makes me wonder how it compares feature wise, and complexity wise for implementation. Guess have to wait for the theora guys analysis of vp8 next.

Reply Score: 1

RE: couple of points (vs theora)
by bert64 on Fri 21st May 2010 12:21 UTC in reply to "couple of points (vs theora)"
bert64 Member since:
2007-04-23

theora's big advantage was that it decodes with 30% less cpu resources than h264. vp8 is apparently the same as h264. People may comment about mobile dedicated hardware.


Theora and h.264 have also had plenty of time for optimized decoders to be written, in time similar improvements will come to VP8 as well...

Reply Score: 3

RE[2]: couple of points (vs theora)
by kaiwai on Sat 22nd May 2010 02:57 UTC in reply to "RE: couple of points (vs theora)"
kaiwai Member since:
2005-07-06

Theora and h.264 have also had plenty of time for optimized decoders to be written, in time similar improvements will come to VP8 as well...


Out of curiosity, how much of the optimisations in VP3/Theora can be ported to VP8?

Reply Score: 2

I have solved all of our problems
by thesunnyk on Fri 21st May 2010 04:47 UTC
thesunnyk
Member since:
2010-05-21

My friends. I have a solution we can all agree on: http://blog.quaddmg.com/2010/05/20/the-solutions-to-all-our-problem...

Take a look and tell me what you think. This will stop all the arguing and whining. I am holding the goddamn joker here.

I'm getting a little over-excited, but it's so hard to stop myself.

Edited 2010-05-21 04:48 UTC

Reply Score: 0

rexstuff Member since:
2007-04-06

So, if I understand you right, you're suggesting that we create video streams out of high-level programming language, such as how documents are rendered out of something like postscript or LaTeX, or how video games are rendered using GL shaders?

If so, then I must regretfully say: Sorry dude, I don't think that will work.

Sure, glorious HD scenes can be rendered in as little as 64k, but a) it would be incredibly lossy in the sense that rendering engine shortcomings would make the difference between the originally captured stream and the rendered video as seen by the viewer more than a little... significant. And b) a realistic encoder (converting a stream to textures and geometry) is virtually impossible, barring major paradigm shifts in current video processing

Reply Score: 2

voracity Member since:
2010-05-22

Actually, I think something like this would work. It wouldn't solve any patent issues, but it would mean that 'freezing the bitstream' wouldn't be such a big deal.

You just need to make the bitstream turing complete, efficient, and have access to GL primitives and the like (maybe like WebGL). The idea is that you would embed the codec into the bitstream --- which is really easy to do, since codecs are pure functions and not that big code-wise.

Like I said, doesn't solve patent problems, but (like any turing machine/VM) solves all flexibility issues.

Reply Score: 1

thesunnyk Member since:
2010-05-21

I think it definitely shifts the patent issues. It's no longer something the spec nor the decoder has to worry about. If someone wants to build an encoder using these techniques and pay MPEG-LA, then sure, but that's where this stuff is relegated to. I'd be interested to hear more on why you think it doesn't solve the patent problems.

There are actually simpler machines than Turing machines (and I've no idea what they are) which cannot do everything but for the purposes of "things that can be put on a DSP" are actually what we're looking for. The spec itself would be working to that machine instead, then downconverting to GL. Sort of like the distinction between Java Applets and Java Applications (but obviously both are real Turing machines, just that one is a limited form of the other).

Like you say, part of the intent here is to make the codec really flexible and still hardware supported.

Reply Score: 1

thesunnyk Member since:
2010-05-21

I don't think that's true. All I was saying about the 64k thing is that there's no floor to the theoretical efficiency of the codec. In addition, if you can make a demo in 64k, then theoretically you can make a video of the demo in 64k. If you were capturing a movie, then sure you wouldn't be able to do it, but then again, maybe one day some guy would build an incredible encoder to read the data straight from the matrix. In either case your efficiency can be incredibly good, which is one of the touted features of h264.

The paradigm shift is actually exactly what I'm suggesting, but like I said, if you squint it's not a paradigm shift at all. h264 is a step in this direction anyway when compared to MPEG-2. All we're doing is coming at it from completely the other end, which means no patent issues. Frankly if you don't shift the paradigm some guy is going to come up and tell you you're infringing on their patent.

Reply Score: 1

i don't care which is better
by stabbyjones on Fri 21st May 2010 05:30 UTC
stabbyjones
Member since:
2008-04-15

H264 could make you sandwiches while VP8 punches you in the face while watching a video but i'd still pick VP8.

Technical specs shouldn't matter when MPEG-LA wants to own everything we watch in every single level of production and consumption.

Open and royalty free all the way.

Reply Score: 2

Apple is still unclear...
by fithisux on Fri 21st May 2010 06:37 UTC
fithisux
Member since:
2006-01-22

"appears to be significantly weaker"

I have not yet seen an optimized implementation for x86/x86-64 or even ARM (and my favorite MIPS64). So, how can they claim it?

On the contrary it seems to me that a proper hardware documentation-specification (and not how the HW was created) like SSE3 / Neon instruction sets and unlike current gfx cards which lack such a polished exposed API at the HW level can lead us into the future.

To me the benefits and possibilities are tremendous and certainly not "significantly weaker". Hats off to ARM/Google and Intel.

Reply Score: 2

RE: Apple is still unclear...
by tf123 on Fri 21st May 2010 06:58 UTC in reply to "Apple is still unclear..."
tf123 Member since:
2010-01-28

It is very much possible, within a degree of error, to simulate potential optimizations. When H.264 was first released all of the implementations sucked but within months numerous companies had projections for what would ultimately be accomplished in software and in hardware -- without having produced the software or the hardware. Those predictions have borne out over the last several years. One man's opinion is not enough but it is a start; he cannot be dismissed outright.

Reply Score: 1

OSNews?
by dbolgheroni on Fri 21st May 2010 16:42 UTC
dbolgheroni
Member since:
2007-01-18

No. VideoNews? PhoneNews? GadgetNews?

Reply Score: 1

project_2501
Member since:
2006-03-20

As devices and bandwidth get ever faster and bigger, hardware acceleration and quality per bitrate will matter less and less.

Leaving openness as the major benefit.

Reply Score: 2