Question re problems with .wmv for Macs

Dan Euritt · December 10th, 2004, 12:22 PM

<<<-- Originally posted by Mark Sloan : You own M$ stock Dan?
Again, the point earlier is that in the US 15%-25% of HOME, internet users are on Macs. Businesses are more like 99% PCs, Universities 75%-90% PC, creative firms 50% PC.-->>>

mark, you haven't shown any reason why the physical location of the computer is relevant to this thread.

what marcia DID say was "I was going to throw up a Quicktime version, but I've had trouble getting one that was a small enough file size and didn't look horrible." ...which is exactly what i've said all along about the lousy quality of qt video codecs.

hey ernest, how come she can't get decent video out of your qt h.264? how come you guys can't help her with that, lol?

"H.264 is a high compression digital video codec standard written by the ITU-T Video Coding Experts Group (VCEG) together with the ISO/IEC Moving Picture Experts Group (MPEG) as the product of a collective effort known as the Joint Video Team (JVT). This standard is identical to ISO MPEG-4 part 10, also known as AVC, for Advanced Video Coding. " ...h.264 is indeed mpeg4, just as i stated earlier... geez!

one of the reasons why the qt h.264 mpeg4 has zero market penetration is because the qt player apparently doesn't work right on 96% of the desktop computers on the internet... i tried to look at your peter jackson clip, but i kept repeatedly getting this qt pop-up window telling me i had to install mpeg4, then reboot the computer because i was running active desktop... i've never had to reboot a computer because of a wmp codec update!

so we have crappy qt video quality, a dysfunctional qt player install procedure, and poor qt player market penetration... hence my recommendation to always use wmp9, lol.

fyi ernst, sorensen 3 is nothing more than a crappy tweaked version of h.264... the nero implementation of h.264 is far superior in picture quality to anything that sorensen has ever done, and the nero encoder costs a lot less $$$, doesn't it??... that killer nero h.264 will soon be playable in the vlc media player, which is a 6.7mb download.

Mark Sloan · December 10th, 2004, 01:12 PM

Not really, as your quote points out:
"This standard is identical to ISO MPEG-4 part 10, also known as AVC, for Advanced Video Coding. "

Both AVC and H.264 are simply codecs and MPEG-4 is really just a wrapper with a set of standards for your video package: "This is important to note: the standard does not specify the encoding process." So the different ways of encoding the file to a compliant MPEG-4 file has nothing to do with the codecs used to compress the video data.

A decent write up FAQ: http://www.m4if.org/resources/mpeg4userfaq.php

MPEG-4 was designed with the default codec H.263, but H.264/AVC are new additions that were ratified as part of the specification in June or so. They are finishing the work on adding sampling structures known as YUV 4:2:2 and YUV 4:4:4 for better color, and 10 and 12 bit sampling. The codec H.264/AVC has been adopted as part of the HD DVD forum as I believe WM9 has as well (or WM9 might be Blu-Ray).

As for Marcia, if she is using QT Pro to encode her MPEG-4 file then she isn't using H.264/AVC, she is using H.263 essentially, and the encoder that comes with QT is not very good, which is really sad because you have to pay for it!! QT with H.264/AVC debuts with Mac OS 10.4... but I don't know if Apple will be providing it as a separate download or not.

As for a Codec's marketshare? Again, I don't think you understand what you are saying... a lot of content is encoded in MPEG-4, not tons as of yet, but it is fairly new. What you should care about is what marketshare can PLAY your MPEG-4 video... and with Real Player and QT you have a huge installed base. WM9 has what installed base that can play it? It isn't all windows users just like not all QT users or Real Player users can play MPEG-4. In both cases, WM9 and MPEG4, you have to have a fairly RECENT player in order to play them.

As you pointed out, VLC is only 6.7 MB and can support MPEG-4, so why doesn't MS support an open standard? That seems silly to me.

Dan Euritt · December 11th, 2004, 11:40 AM

mark, i just proved that h.264 is iso mpeg4, despite your earlier denial of that fact.

>>>the encoder that comes with QT is not very good<<<

lol... how many times have we heard that already?

>>>What you should care about is what marketshare can PLAY your MPEG-4 video<<<

what you should care about is what market share can play any qt files, period!

wmp9 should always be the main choice for 'net video, because it has the best picture quality and the biggest market share.

are you aware that the chinese just bought out the pc division of ibm? you won't see any macs in china, but you will see 200 million more pc's located in china, all on the 'net within the next few years... probably all of 'em running hacked windows o.s.'s.

that european lawsuit against microsoft is further proof that microsoft has taken over the 'net video market... just like they did with internet browsers and desktop computers.

resistance is futile, you will be assimilated :-)

Mark Sloan · December 11th, 2004, 12:47 PM

Dan, H.264/AVC is NOT MPEG-4, it is simply supported by the spec as of June this year. H.264/AVC support by things like HD-DVD players does not guarantee MPEG-4 ISO level 0, 1 or 2 playback, although most will probably support it as it is trivial to support at that point because it is simply a wrapper. It was based on the wrapper architecture of QT so they are similar in the way they are set up, but are completely different specs. H.264/AVC can be delivered in many different ways, one of the choices is using MPEG-4.

From internet news:
"A recent report on media player market share sales for client and enterprise applications by research firm Frost & Sullivan showed Apple moving into second position with 36.8 percent of the market, after running a distant third a few years ago. That puts Apple above Real, which controls 24.9 percent and not too far off from Microsoft and its share of 38.2 percent."
http://www.internetnews.com/infra/article.php/3366831

Which is exactly why stats are not the only thing you should go by. Obviously more than 38.2% of users have WMP, but not WMP9. That piece of software has only been around for what 2 years?

"what you should care about is what market share can play any qt files, period!"
And so with any file format you need to worry not just about the file type you use, but also the codec. Encoding something in WMP9 only in theory hits 60% of the windows base using XP, although they did make a version for 98 SE, ME, and 2000 that users could update to. So what raw percentage of users can view WMP9 files without updating? It is the same as thinking that any QT player can play any QT file... it simply isn't true. You want better compatability, go with Sorenson codec. Want better file size and quality, go with Sorensen3 codec but leave out all QT users before 4.x... you have more to take into consideration than OS numbers.

China... it is a really interesting marketplace (it was a Chineese company, not the country of China that bought out IBMs PC division). The infrastructure is so weird that now they are building a middle class they skipped land lines altogether! They are also similar to Korea and Japan in that the middle class proves how well off it is by buying more expensive status symbols like BMWs instead of just cars. What does that mean? Well, no one knows yet, but if the trend with iPods is an indication then Apple as an overall company will do well. Maybe more like 6-8% as in Japan than the 2% overall share in the US.

As for QT... Some facts: Apple has iTunes and QT pre-installed by China's #2 manufacturer (Founder) of PC computers, so QT already has a good chunk of installed base in the country. MPEG-4 and QT are both cheap standards targeted at infrastructure such as cellphones and high-bandwidth deals for delivering content. iTunes is a great Trojan Horse for getting QT onto computers and ensuring greater marketplace.

For China the biggest thing will be delivery of content on cell-phones and who pushes into that market the most. If Apple works out a deal over there for iTMS sales then you'll see even more.

As you pointed out, there are already tons of PCs there (many running Linux) and so they have WMP of some sort there. But a lot of the base is pre-XP still as their machines are generally slower... but look for that to change dramatically every day.

QT and WMP aren't going anywhere. They will both be here for a long time, and have substantial installed bases. Real Player is the one that I could see dying off.

Les Wilson · December 11th, 2004, 10:30 PM

Marcia had two problems.

1) The WMV she produced played poorly on some of her customer's computers.
2) She couldn't produce QT files she liked at a size she liked.

I and some others suggested she use different tools (Sorenson Squeeze or Flash) to produce better quality QT or Flash. While no specifics were given by Marcia, that probably would do the trick. I use them to produce nice looking QuickTime and Flash all the time..

Your postings and proposal to use WMV 9 as the "main choice for the net" did nothing to help either problem.

I understand your frustration. It seems people don't agree with your opinions, believe what you say or take your advice. The emotional rants, blind MSFT bias, circular arguments, strawman tactics and general lack of reasoning probably don't help.

Mark Sloan · December 12th, 2004, 12:04 AM

Maybe she is taking his advice. I think she got wise that this thread had little to do with her actual question anymore and ran for the hills! Sorry Marcia! I hope you answer was questioned somewhere along the way. (and yes, i do mean it that way ;-)

Dan Euritt · December 12th, 2004, 02:06 PM

>>>H.264/AVC support by things like HD-DVD players does not guarantee MPEG-4 ISO level 0, 1 or 2 playback<<<

mark, i just proved that h.264 IS mpeg4, why are you confusing yourself with this player compatibility b.s.?

i don't blame her for bailing out, lol! who could stand to listen to these mac fanatics twisting the stats around?

the 'net user base stats i posted prove that 59% of the people browsing the 'net have winxp, and 15% have win2k... so we have at least 75% of the internet that is wmp9-capable.

that is far more wmp9 media players than all of the qt players combined together... you guys have not posted ANY stats that prove how many qt players there are on the internet!

poor ernest... you posted the eu lawsuit that proves microsoft's total dominance of internet video players, but your feeble attempt at logic backfired in your face, lol... so of course you are upset and embarrassed.

without an accurate count of total qt players on the 'net, there is little reason to recommend it for internet use... especially in light of the horrible video quality!

Mark Sloan · December 12th, 2004, 02:50 PM

Dan, go look at the actual spec for MPEG-4 and learn what a spec is about. Support for a codec does not mean the codec is that spec. You don't know what you are talking about when it comes to this... so just stop. Seriously, its ridiculous.

If you want me to pull out the white paper and show you the diagrams of what the MPEG-4 spec actually specifies and what H.264/AVC actually specifies I can, but what you are trying to say is similar to saying that QT is Sorensen3... simply because it supports it. You can have MPEG-4 without H.264, so how are the two the same thing?

The MPEG-4 spec was updated to support H.264... that is all. There are 9 other parts including 2-d mesh objects that have nothing to do with AVC/H.264... Anyone can implement AVC/H.264 (with a license) but that doesn't make it MPEG-4. Hell, H.264 only exists because a bunch of people working on MPEG-4 AVC codec basically remade it freely as H.264. So while AVC and H.264 are identical (they were made by the same people) they are only codecs.

Dan Euritt · December 13th, 2004, 02:09 AM

sorenson 3 is a proprietary codec(tweaked h.264, actually) that exists nowhere else.

h.264 is open standard iso mpeg-4, part 10, period... sorenson 3 is not a standard for anything.

i might be impressed, tho, if you could pull your "white paper" on sorenson 3 and show it to us.

and i'd be even more impressed if you could actually find some real stats on how many mac players there are on the 'net ;-)

Mark Sloan · December 13th, 2004, 03:54 PM

From everything I have seen, Sorenson 3 has nothing to do with H.264. I think you meant the new Sorenson codec, 4, that is similar to H.264.

White paper for MPEG-4 can be found here: http://www.m4if.org/public/documents/vault/m4-out-20027.pdf
or some more detailed info:
http://www.chiariglione.org/mpeg/standards/mpeg-4/mpeg-4.htm

Sorenson is a codec, so I don't really know what their white paper would contain that is anything more than what their website advertises. Besides, they are a private company, I don't see them publishing the details of how they do what they do for free. :-)

H.264 is a copy of the ISO MPEG-4 part 10 called AVC, so they are the same thing, but have different names to show that one was published by the ISO and the other by ITUC. It can be used for anything. All making it MPEG-4 part 10 means is that you can say that you are MPEG-4 part 10 compliant so that people know what will play what. It also means that they have adopted AVC/H.264 as part of that specification, but AVC/H.264 can be licensed by anyone for anything, hence HD DVD is using AVC/H.264 for encoding HD quality video. But they are not trying to be MPEG-4 part 10 compliant so they are not constrained by the bit rate limits, the frame sizing, or whatever else.

To use a different example, to say you are Level 0 Simple Profile compatible you cannot have more than 15 frames per second. They even put a limit on the luminance.

More stats: http://www.creativepro.com/story/news/16921.html
http://www.itfacts.biz/index.php?id=P471

Here is an example of H.264 without MPEG-4:
http://www.eetimes.com/sys/news/showArticle.jhtml?articleID=54200808

MPEG-4 is more like quicktime and MPEG-1 and MPEG-2 are more restricted in that they are pretty strictly defined. MPEG-4 is a wrapper with different levels/profiles that you can be compliant too, which helps make it easier to implement and support without constraining it. You can see how the "wrapper" or containers work with this image: http://www.chiariglione.org/mpeg/standards/mpeg-4/mpeg-429.gif

From a programming perspective it is great because they have abstracted the data into distinct objects that allows for specialization or generalization. The programmer gets to decide. It also has some great stuff for facial animations and other goodies....

As a whole, it is nice because unlike MPEG-1 or MPEG-2 it is only a wrapper so when better stuff comes along like H.264 and it can be adopted as a supported format, and new Profiles can be added to standardize its use. With each addition, there is also always backwards compatability.

Dan Euritt · December 13th, 2004, 05:40 PM

most people don't know that sorenson 3 is strongly rumored to be nothing more than just hacked h.264, which is why i keep pointing it out :-) google it for yourself.

it's sorta similar to what happened with divx and the original microsoft-hacked mpeg-4 codec... it all started somewhere, then people just tweaked it so that they could legally sell it without paying royalties... that's why you won't find a "white paper" on sorensen 3.

by comparison, ask yourself what microsoft had to do to get the wmp format accepted as part of both hd dvd specs.

more h.264 info: http://forum.doom9.org/showthread.php?s=&threadid=73022

you guys STILL haven't provided any evidence of the number of qt players on the internet today.

Mark Sloan · December 13th, 2004, 06:55 PM

Sorenson 3 is missing a lot of features of AVC/H.264, so while it may have some similar structures and features... eh, who cares... Go with the codec you like. Sorenson 3 came out in 2001 though, and AVC/H.264 finished in 2003, so while it might have some similar features, it isn't AVC/H.264... and then, how you implement AVC/H.264 is going to determine the quality of it... how much processing is required... etc.

In the end, they are all pretty much trying to accomplish the same things and there are only so many different ways of compressing video with negligible loss of quality. No one, MPEG, MS, Apple, is doing anything that is technologically special really. No breakthroughs or anything.

MS wants the licensing for the next generation as much as anyone, they aren't stupid. And their X-Box isn't doing so hot, using a format that dies like Betamax would hurt them even more. I'd say MS has a lot at stake. What is interesting is that both groups will end up supporting both WMV9 AND AVC/H.264 and leave it up to the producers of content to choose. Kind of like which audio to use on a DVD...

Doom9 link was a rehash of everything from the sites I listed, so I guess we now agree that MPEG-4 != AVC/H.264? But that MPEG-4 part 10 provides support for it?

They can't track the numbers of players the same way as WMP9 player (you know the number of XP users appoximately, add more for downloads)... its an anonymous download for QT. So most of the stats are how many content providers are streaming QT, WMV, and Real... which is what I provided before... The stats for number of downloads of QT is over 250 million... add in that HP will be preloading it (or have they already started?) and that every iTunes download includes it, and you have a very large installed base. Is each of the 250 million downloads unique? No. But they do know that 90% of those downloads were for PCs. Do your best guessing from there.

Here is PC World with streaming stats from June:
http://www.pcworld.com/news/article/0,aid,116589,00.asp

Les Wilson · December 13th, 2004, 08:46 PM

Dan,

Bored would be a better description.

There is a gap between data and conclusion that is filled by something called analysis.

You repeatedly state your conclusion that people should produce WMV because Windows is the dominant OS and, in turn, there are more WMP players. If you follow this logic, the conclusion is to use Flash but you don't follow your own analysis.

To date, you have only been able to count WMP players based on OS and not on usage. That is why your analysis and conclusions are faulty and debated. Tracking players based on downloads is subject to double counting. AFAIK, there isn't an accurate way to count players. You have no solid data to argue a format based on the number of installs. Therefore, arguing format based on something you cannot quantify is has led to a poor conclusion.

Voices in this thread have presented more factual and reasoned approaches based on a broader basis and data from industry analysts. That discussion lead to the merits of standards because they render the platform neutral and create an open market with the broadest opportunity for all members of the ecosystem (producers, consumers, technology, players etc). It's open because all platforms have equal opportunity to participate.

Ironically, we are exchanging these posts because of standards. If the standards for discussion boards and email were controlled by single vendors then market fragmentation occurs and the market is constrained. This is comparable to the early days of networking where you had little fifedoms of Compuserv, AOL, MSN, and thousands of billboards each with their own phone number. The internet blew them wide open. Standards and competition, not abusive monopolies drive innovation. The road to your beloved Nero codec for H.264 was paved by MPEG4 and the implementors like QuickTime, Real and others to innovate beyond the prior standard H.263 and ultimately to H.264. H.264 and Nero are the very things threatened by a single vendor control of the format. As a producer, you should be infavor of standards not a single vendor.

The EU found MSFT guilty of abusing their monopoly powers to gain unfair advantage over competition. The US convicted them of the same. The $600 million EU was punishment not proof of dominance. The requirement that MSFT ship a version of Windows without WMP was an attempt to prevent the defacto domination of WMP. WMV still only has 38% market share and Nero, QuickTime and Real are still alive so....

In your December 8th post your analysis concluded "what happened with microsoft in europe is irrelevant".

In your December 12th post your analysis concluded "the eu lawsuit proves microsoft's total dominance of internet video players".

Which is it?

My mention of the EU lawsuit was a backhand remark while showing flaws in your counting and I also used it to make the point about the value of standards vs single vendor controlled. I made no claim of relevance to domination as you erroneously imply and conclude after you reversed yourself.

Yawn.

Mark Sloan · December 13th, 2004, 11:43 PM

As a vendor do really care if it is an open standard or not? In the end, all you really care about is: do your users get your product and is it quality... right? Do you really care HOW it is done? Of course, you don't want a monopoly because then you end up paying too much and have no innovation, but as it is today, use what makes sense for you.

Les Wilson · December 14th, 2004, 06:34 AM

That's a good question. I think evaluating applicable standards are input to making the choice. I favor a broad approach that looks at reliability, availability, ecosystem (support for the format from other vendors) as well as the efficiency/cost of producing the format. In my opinion, availability and reliability of a format are important to determining if users can get the product. Standards favor availability, reliability and ecosystem across the widest audience.

I wouldn't pick a format that's available on a subset of computers and not made available by the format developer for years on others.

So, I think there's a point at which a producer should care about what standards are applicable to the task. Not at the DCT transform algorithm level but at the player availability, reliability, tools cost, quality... level. Right now, when I'm on a Mac, I cannot get to MSNBC Basketbrawl video as it dissallows access based on OS (not on installed format support). So in MSNBC's case, they've actually picked an OS (not a format) and conciously limit their audience. The video was easily available on ESPN in Flash.

December 10th, 2004, 12:22 PM	#31
Dan Euritt Trustee Join Date: Mar 2004 Location: Carlsbad CA Posts: 1,132	<<<-- Originally posted by Mark Sloan : You own M$ stock Dan? Again, the point earlier is that in the US 15%-25% of HOME, internet users are on Macs. Businesses are more like 99% PCs, Universities 75%-90% PC, creative firms 50% PC.-->>> mark, you haven't shown any reason why the physical location of the computer is relevant to this thread. what marcia DID say was "I was going to throw up a Quicktime version, but I've had trouble getting one that was a small enough file size and didn't look horrible." ...which is exactly what i've said all along about the lousy quality of qt video codecs. hey ernest, how come she can't get decent video out of your qt h.264? how come you guys can't help her with that, lol? "H.264 is a high compression digital video codec standard written by the ITU-T Video Coding Experts Group (VCEG) together with the ISO/IEC Moving Picture Experts Group (MPEG) as the product of a collective effort known as the Joint Video Team (JVT). This standard is identical to ISO MPEG-4 part 10, also known as AVC, for Advanced Video Coding. " ...h.264 is indeed mpeg4, just as i stated earlier... geez! one of the reasons why the qt h.264 mpeg4 has zero market penetration is because the qt player apparently doesn't work right on 96% of the desktop computers on the internet... i tried to look at your peter jackson clip, but i kept repeatedly getting this qt pop-up window telling me i had to install mpeg4, then reboot the computer because i was running active desktop... i've never had to reboot a computer because of a wmp codec update! so we have crappy qt video quality, a dysfunctional qt player install procedure, and poor qt player market penetration... hence my recommendation to always use wmp9, lol. fyi ernst, sorensen 3 is nothing more than a crappy tweaked version of h.264... the nero implementation of h.264 is far superior in picture quality to anything that sorensen has ever done, and the nero encoder costs a lot less $$$, doesn't it??... that killer nero h.264 will soon be playable in the vlc media player, which is a 6.7mb download.

December 10th, 2004, 01:12 PM	#32
Mark Sloan Major Player Join Date: Jun 2002 Location: Venice, CA Posts: 346	Not really, as your quote points out: "This standard is identical to ISO MPEG-4 part 10, also known as AVC, for Advanced Video Coding. " Both AVC and H.264 are simply codecs and MPEG-4 is really just a wrapper with a set of standards for your video package: "This is important to note: the standard does not specify the encoding process." So the different ways of encoding the file to a compliant MPEG-4 file has nothing to do with the codecs used to compress the video data. A decent write up FAQ: http://www.m4if.org/resources/mpeg4userfaq.php MPEG-4 was designed with the default codec H.263, but H.264/AVC are new additions that were ratified as part of the specification in June or so. They are finishing the work on adding sampling structures known as YUV 4:2:2 and YUV 4:4:4 for better color, and 10 and 12 bit sampling. The codec H.264/AVC has been adopted as part of the HD DVD forum as I believe WM9 has as well (or WM9 might be Blu-Ray). As for Marcia, if she is using QT Pro to encode her MPEG-4 file then she isn't using H.264/AVC, she is using H.263 essentially, and the encoder that comes with QT is not very good, which is really sad because you have to pay for it!! QT with H.264/AVC debuts with Mac OS 10.4... but I don't know if Apple will be providing it as a separate download or not. As for a Codec's marketshare? Again, I don't think you understand what you are saying... a lot of content is encoded in MPEG-4, not tons as of yet, but it is fairly new. What you should care about is what marketshare can PLAY your MPEG-4 video... and with Real Player and QT you have a huge installed base. WM9 has what installed base that can play it? It isn't all windows users just like not all QT users or Real Player users can play MPEG-4. In both cases, WM9 and MPEG4, you have to have a fairly RECENT player in order to play them. As you pointed out, VLC is only 6.7 MB and can support MPEG-4, so why doesn't MS support an open standard? That seems silly to me.

December 11th, 2004, 11:40 AM	#33
Dan Euritt Trustee Join Date: Mar 2004 Location: Carlsbad CA Posts: 1,132	mark, i just proved that h.264 is iso mpeg4, despite your earlier denial of that fact. >>>the encoder that comes with QT is not very good<<< lol... how many times have we heard that already? >>>What you should care about is what marketshare can PLAY your MPEG-4 video<<< what you should care about is what market share can play any qt files, period! wmp9 should always be the main choice for 'net video, because it has the best picture quality and the biggest market share. are you aware that the chinese just bought out the pc division of ibm? you won't see any macs in china, but you will see 200 million more pc's located in china, all on the 'net within the next few years... probably all of 'em running hacked windows o.s.'s. that european lawsuit against microsoft is further proof that microsoft has taken over the 'net video market... just like they did with internet browsers and desktop computers. resistance is futile, you will be assimilated :-)

December 11th, 2004, 12:47 PM	#34
Mark Sloan Major Player Join Date: Jun 2002 Location: Venice, CA Posts: 346	Dan, H.264/AVC is NOT MPEG-4, it is simply supported by the spec as of June this year. H.264/AVC support by things like HD-DVD players does not guarantee MPEG-4 ISO level 0, 1 or 2 playback, although most will probably support it as it is trivial to support at that point because it is simply a wrapper. It was based on the wrapper architecture of QT so they are similar in the way they are set up, but are completely different specs. H.264/AVC can be delivered in many different ways, one of the choices is using MPEG-4. From internet news: "A recent report on media player market share sales for client and enterprise applications by research firm Frost & Sullivan showed Apple moving into second position with 36.8 percent of the market, after running a distant third a few years ago. That puts Apple above Real, which controls 24.9 percent and not too far off from Microsoft and its share of 38.2 percent." http://www.internetnews.com/infra/article.php/3366831 Which is exactly why stats are not the only thing you should go by. Obviously more than 38.2% of users have WMP, but not WMP9. That piece of software has only been around for what 2 years? "what you should care about is what market share can play any qt files, period!" And so with any file format you need to worry not just about the file type you use, but also the codec. Encoding something in WMP9 only in theory hits 60% of the windows base using XP, although they did make a version for 98 SE, ME, and 2000 that users could update to. So what raw percentage of users can view WMP9 files without updating? It is the same as thinking that any QT player can play any QT file... it simply isn't true. You want better compatability, go with Sorenson codec. Want better file size and quality, go with Sorensen3 codec but leave out all QT users before 4.x... you have more to take into consideration than OS numbers. China... it is a really interesting marketplace (it was a Chineese company, not the country of China that bought out IBMs PC division). The infrastructure is so weird that now they are building a middle class they skipped land lines altogether! They are also similar to Korea and Japan in that the middle class proves how well off it is by buying more expensive status symbols like BMWs instead of just cars. What does that mean? Well, no one knows yet, but if the trend with iPods is an indication then Apple as an overall company will do well. Maybe more like 6-8% as in Japan than the 2% overall share in the US. As for QT... Some facts: Apple has iTunes and QT pre-installed by China's #2 manufacturer (Founder) of PC computers, so QT already has a good chunk of installed base in the country. MPEG-4 and QT are both cheap standards targeted at infrastructure such as cellphones and high-bandwidth deals for delivering content. iTunes is a great Trojan Horse for getting QT onto computers and ensuring greater marketplace. For China the biggest thing will be delivery of content on cell-phones and who pushes into that market the most. If Apple works out a deal over there for iTMS sales then you'll see even more. As you pointed out, there are already tons of PCs there (many running Linux) and so they have WMP of some sort there. But a lot of the base is pre-XP still as their machines are generally slower... but look for that to change dramatically every day. QT and WMP aren't going anywhere. They will both be here for a long time, and have substantial installed bases. Real Player is the one that I could see dying off.

December 11th, 2004, 10:30 PM	#35
Les Wilson Inner Circle Join Date: Jan 2004 Location: Boca Raton, FL Posts: 3,014	Marcia had two problems. 1) The WMV she produced played poorly on some of her customer's computers. 2) She couldn't produce QT files she liked at a size she liked. I and some others suggested she use different tools (Sorenson Squeeze or Flash) to produce better quality QT or Flash. While no specifics were given by Marcia, that probably would do the trick. I use them to produce nice looking QuickTime and Flash all the time.. Your postings and proposal to use WMV 9 as the "main choice for the net" did nothing to help either problem. I understand your frustration. It seems people don't agree with your opinions, believe what you say or take your advice. The emotional rants, blind MSFT bias, circular arguments, strawman tactics and general lack of reasoning probably don't help.

December 12th, 2004, 12:04 AM	#36
Mark Sloan Major Player Join Date: Jun 2002 Location: Venice, CA Posts: 346	Maybe she is taking his advice. I think she got wise that this thread had little to do with her actual question anymore and ran for the hills! Sorry Marcia! I hope you answer was questioned somewhere along the way. (and yes, i do mean it that way ;-)

December 12th, 2004, 02:06 PM	#37
Dan Euritt Trustee Join Date: Mar 2004 Location: Carlsbad CA Posts: 1,132	>>>H.264/AVC support by things like HD-DVD players does not guarantee MPEG-4 ISO level 0, 1 or 2 playback<<< mark, i just proved that h.264 IS mpeg4, why are you confusing yourself with this player compatibility b.s.? i don't blame her for bailing out, lol! who could stand to listen to these mac fanatics twisting the stats around? the 'net user base stats i posted prove that 59% of the people browsing the 'net have winxp, and 15% have win2k... so we have at least 75% of the internet that is wmp9-capable. that is far more wmp9 media players than all of the qt players combined together... you guys have not posted ANY stats that prove how many qt players there are on the internet! poor ernest... you posted the eu lawsuit that proves microsoft's total dominance of internet video players, but your feeble attempt at logic backfired in your face, lol... so of course you are upset and embarrassed. without an accurate count of total qt players on the 'net, there is little reason to recommend it for internet use... especially in light of the horrible video quality!

December 12th, 2004, 02:50 PM	#38
Mark Sloan Major Player Join Date: Jun 2002 Location: Venice, CA Posts: 346	Dan, go look at the actual spec for MPEG-4 and learn what a spec is about. Support for a codec does not mean the codec is that spec. You don't know what you are talking about when it comes to this... so just stop. Seriously, its ridiculous. If you want me to pull out the white paper and show you the diagrams of what the MPEG-4 spec actually specifies and what H.264/AVC actually specifies I can, but what you are trying to say is similar to saying that QT is Sorensen3... simply because it supports it. You can have MPEG-4 without H.264, so how are the two the same thing? The MPEG-4 spec was updated to support H.264... that is all. There are 9 other parts including 2-d mesh objects that have nothing to do with AVC/H.264... Anyone can implement AVC/H.264 (with a license) but that doesn't make it MPEG-4. Hell, H.264 only exists because a bunch of people working on MPEG-4 AVC codec basically remade it freely as H.264. So while AVC and H.264 are identical (they were made by the same people) they are only codecs.

December 13th, 2004, 02:09 AM	#39
Dan Euritt Trustee Join Date: Mar 2004 Location: Carlsbad CA Posts: 1,132	sorenson 3 is a proprietary codec(tweaked h.264, actually) that exists nowhere else. h.264 is open standard iso mpeg-4, part 10, period... sorenson 3 is not a standard for anything. i might be impressed, tho, if you could pull your "white paper" on sorenson 3 and show it to us. and i'd be even more impressed if you could actually find some real stats on how many mac players there are on the 'net ;-)

December 13th, 2004, 03:54 PM	#40
Mark Sloan Major Player Join Date: Jun 2002 Location: Venice, CA Posts: 346	From everything I have seen, Sorenson 3 has nothing to do with H.264. I think you meant the new Sorenson codec, 4, that is similar to H.264. White paper for MPEG-4 can be found here: http://www.m4if.org/public/documents/vault/m4-out-20027.pdf or some more detailed info: http://www.chiariglione.org/mpeg/standards/mpeg-4/mpeg-4.htm Sorenson is a codec, so I don't really know what their white paper would contain that is anything more than what their website advertises. Besides, they are a private company, I don't see them publishing the details of how they do what they do for free. :-) H.264 is a copy of the ISO MPEG-4 part 10 called AVC, so they are the same thing, but have different names to show that one was published by the ISO and the other by ITUC. It can be used for anything. All making it MPEG-4 part 10 means is that you can say that you are MPEG-4 part 10 compliant so that people know what will play what. It also means that they have adopted AVC/H.264 as part of that specification, but AVC/H.264 can be licensed by anyone for anything, hence HD DVD is using AVC/H.264 for encoding HD quality video. But they are not trying to be MPEG-4 part 10 compliant so they are not constrained by the bit rate limits, the frame sizing, or whatever else. To use a different example, to say you are Level 0 Simple Profile compatible you cannot have more than 15 frames per second. They even put a limit on the luminance. More stats: http://www.creativepro.com/story/news/16921.html http://www.itfacts.biz/index.php?id=P471 Here is an example of H.264 without MPEG-4: http://www.eetimes.com/sys/news/showArticle.jhtml?articleID=54200808 MPEG-4 is more like quicktime and MPEG-1 and MPEG-2 are more restricted in that they are pretty strictly defined. MPEG-4 is a wrapper with different levels/profiles that you can be compliant too, which helps make it easier to implement and support without constraining it. You can see how the "wrapper" or containers work with this image: http://www.chiariglione.org/mpeg/standards/mpeg-4/mpeg-429.gif From a programming perspective it is great because they have abstracted the data into distinct objects that allows for specialization or generalization. The programmer gets to decide. It also has some great stuff for facial animations and other goodies.... As a whole, it is nice because unlike MPEG-1 or MPEG-2 it is only a wrapper so when better stuff comes along like H.264 and it can be adopted as a supported format, and new Profiles can be added to standardize its use. With each addition, there is also always backwards compatability.

December 13th, 2004, 05:40 PM	#41
Dan Euritt Trustee Join Date: Mar 2004 Location: Carlsbad CA Posts: 1,132	most people don't know that sorenson 3 is strongly rumored to be nothing more than just hacked h.264, which is why i keep pointing it out :-) google it for yourself. it's sorta similar to what happened with divx and the original microsoft-hacked mpeg-4 codec... it all started somewhere, then people just tweaked it so that they could legally sell it without paying royalties... that's why you won't find a "white paper" on sorensen 3. by comparison, ask yourself what microsoft had to do to get the wmp format accepted as part of both hd dvd specs. more h.264 info: http://forum.doom9.org/showthread.php?s=&threadid=73022 you guys STILL haven't provided any evidence of the number of qt players on the internet today.

December 13th, 2004, 06:55 PM	#42
Mark Sloan Major Player Join Date: Jun 2002 Location: Venice, CA Posts: 346	Sorenson 3 is missing a lot of features of AVC/H.264, so while it may have some similar structures and features... eh, who cares... Go with the codec you like. Sorenson 3 came out in 2001 though, and AVC/H.264 finished in 2003, so while it might have some similar features, it isn't AVC/H.264... and then, how you implement AVC/H.264 is going to determine the quality of it... how much processing is required... etc. In the end, they are all pretty much trying to accomplish the same things and there are only so many different ways of compressing video with negligible loss of quality. No one, MPEG, MS, Apple, is doing anything that is technologically special really. No breakthroughs or anything. MS wants the licensing for the next generation as much as anyone, they aren't stupid. And their X-Box isn't doing so hot, using a format that dies like Betamax would hurt them even more. I'd say MS has a lot at stake. What is interesting is that both groups will end up supporting both WMV9 AND AVC/H.264 and leave it up to the producers of content to choose. Kind of like which audio to use on a DVD... Doom9 link was a rehash of everything from the sites I listed, so I guess we now agree that MPEG-4 != AVC/H.264? But that MPEG-4 part 10 provides support for it? They can't track the numbers of players the same way as WMP9 player (you know the number of XP users appoximately, add more for downloads)... its an anonymous download for QT. So most of the stats are how many content providers are streaming QT, WMV, and Real... which is what I provided before... The stats for number of downloads of QT is over 250 million... add in that HP will be preloading it (or have they already started?) and that every iTunes download includes it, and you have a very large installed base. Is each of the 250 million downloads unique? No. But they do know that 90% of those downloads were for PCs. Do your best guessing from there. Here is PC World with streaming stats from June: http://www.pcworld.com/news/article/0,aid,116589,00.asp

December 13th, 2004, 08:46 PM	#43
Les Wilson Inner Circle Join Date: Jan 2004 Location: Boca Raton, FL Posts: 3,014	Dan, Bored would be a better description. There is a gap between data and conclusion that is filled by something called analysis. You repeatedly state your conclusion that people should produce WMV because Windows is the dominant OS and, in turn, there are more WMP players. If you follow this logic, the conclusion is to use Flash but you don't follow your own analysis. To date, you have only been able to count WMP players based on OS and not on usage. That is why your analysis and conclusions are faulty and debated. Tracking players based on downloads is subject to double counting. AFAIK, there isn't an accurate way to count players. You have no solid data to argue a format based on the number of installs. Therefore, arguing format based on something you cannot quantify is has led to a poor conclusion. Voices in this thread have presented more factual and reasoned approaches based on a broader basis and data from industry analysts. That discussion lead to the merits of standards because they render the platform neutral and create an open market with the broadest opportunity for all members of the ecosystem (producers, consumers, technology, players etc). It's open because all platforms have equal opportunity to participate. Ironically, we are exchanging these posts because of standards. If the standards for discussion boards and email were controlled by single vendors then market fragmentation occurs and the market is constrained. This is comparable to the early days of networking where you had little fifedoms of Compuserv, AOL, MSN, and thousands of billboards each with their own phone number. The internet blew them wide open. Standards and competition, not abusive monopolies drive innovation. The road to your beloved Nero codec for H.264 was paved by MPEG4 and the implementors like QuickTime, Real and others to innovate beyond the prior standard H.263 and ultimately to H.264. H.264 and Nero are the very things threatened by a single vendor control of the format. As a producer, you should be infavor of standards not a single vendor. The EU found MSFT guilty of abusing their monopoly powers to gain unfair advantage over competition. The US convicted them of the same. The $600 million EU was punishment not proof of dominance. The requirement that MSFT ship a version of Windows without WMP was an attempt to prevent the defacto domination of WMP. WMV still only has 38% market share and Nero, QuickTime and Real are still alive so.... In your December 8th post your analysis concluded "what happened with microsoft in europe is irrelevant". In your December 12th post your analysis concluded "the eu lawsuit proves microsoft's total dominance of internet video players". Which is it? My mention of the EU lawsuit was a backhand remark while showing flaws in your counting and I also used it to make the point about the value of standards vs single vendor controlled. I made no claim of relevance to domination as you erroneously imply and conclude after you reversed yourself. Yawn.

December 13th, 2004, 11:43 PM	#44
Mark Sloan Major Player Join Date: Jun 2002 Location: Venice, CA Posts: 346	As a vendor do really care if it is an open standard or not? In the end, all you really care about is: do your users get your product and is it quality... right? Do you really care HOW it is done? Of course, you don't want a monopoly because then you end up paying too much and have no innovation, but as it is today, use what makes sense for you.

December 14th, 2004, 06:34 AM	#45
Les Wilson Inner Circle Join Date: Jan 2004 Location: Boca Raton, FL Posts: 3,014	That's a good question. I think evaluating applicable standards are input to making the choice. I favor a broad approach that looks at reliability, availability, ecosystem (support for the format from other vendors) as well as the efficiency/cost of producing the format. In my opinion, availability and reliability of a format are important to determining if users can get the product. Standards favor availability, reliability and ecosystem across the widest audience. I wouldn't pick a format that's available on a subset of computers and not made available by the format developer for years on others. So, I think there's a point at which a producer should care about what standards are applicable to the task. Not at the DCT transform algorithm level but at the player availability, reliability, tools cost, quality... level. Right now, when I'm on a Mac, I cannot get to MSNBC Basketbrawl video as it dissallows access based on OS (not on installed format support). So in MSNBC's case, they've actually picked an OS (not a format) and conciously limit their audience. The video was easily available on ESPN in Flash.