Slashdot Mirror
Cable Industry backs Mpeg-4 for Streaming Video
Greyfox writes "This techweb story informs us that the Cable Industry has thrown their chips in with Mpeg4 and will probably want to tweak the codec for streaming video. I'm all for it, I'm sick of QuickTime movies I can't view in Linux and RealVideo movies I'd prefer not to download the player for. "
The MPEG-4 Industry Forum:
http://www.m4if.org
Is working to that you can conveniently obtain a licence to the whole MPEG-4 patent bundle (over 200 patents), in the same way as you can for MPEG-2.
A free software imnplementation of an MPEG-4 CODEDC is available from ISO, which is linked from the M4IF site. Note that while the software (i.e. the implementation) is free, you still need to licence the patents to use it.
Still, we're all using free MPEG-2 players and encoders under Linux, so it sounds pretty good to me!
I was an early adopter of DSS; in fact, the DSS reciever I still use is the original first-generation RCA unit. Having used DSS for years I can say that TCI (ie AT&T) digital cable is inferior.
Specifically, newer (anything in the past 3 years) DSS units do not have the slow-channel-switching problems. Also the channel guides are very sophisticated and morever very useful. The image quality is like S-VHS, and the sound quality is AWESOME. I only notice MPEG artifacts when there is a subtle gradient or a large dark area on the screen.
My DSS freaks out maybe three times a month. By freaks out I mean the MPEG signal goes crazy, the screen becomes corrupted (sound is fine) and it takes about 5 seconds to return to normal.
With digital cable, this is a MUCH more frequent occurance. And, like you mentioned, channel switching is painful. Not only does it take a second to tune, but then it seems like it starts building the image before it gets a keyframe; ie you will see a bunch of boxes on the screen that eventually resolve into a normal image.
Whats more is the cost -- "digital cable" here in Berkeley, CA means 40 channels of analog cable at $25/mo + $10/mo for TEN channels of digital (+PPV). Uhmm... no thanks, I'll stick with DSS, which is technically better, as well as the fact that not only do I get ALL the cable channels in digital, I also get the local networks digital for an extra $5. Thats ABC, NBC, CBS, and Fox.
The streaming server never decodes or encodes a movie file. And a player for the Quicktime file format is not a problem either. xanim can access them just fine. The problem is that nearly all QT4 videos use the Sorenson codec. This codec is a proprietary product of Sorenson. Apple has a license to encode and decode Sorenson in their player and encoder. Sorenson and Apple even refuse to allow a decoder to be written under an NDA. The xanim author was able to write Indeo and Cinepak decoders under NDA, and then distribute binary modules. We can't even do this!
What do we do if the companies decide to get greedy like Fraunhoefer and clamp down on Open-Source players (as FhG is planning to do with AAC) as well as encoders (as FhG does with mp3)?
This is why I think ISO shouldn't be accepting patent-encumbered stuff as open standards. Companies do have a right to their work and their intellectual property. But they shouldn't be allowed to hold that out to the public as an "open standard" then force people to jump through hoops. You want to patent stuff, fine. You want to open it as an indistry standard, fine. But you shouldn't be able to have your cake and eat it too. Nowhere else in the universe do things work that way. Why, then, should standards?
Broadband internet is already very suitable for streaming high quality video. I fooled around with some divx videos yesterday (matrix and starwars). I was very much impressed with both. The matrix had been ripped straight of a dvd and the starwars video was ripped from a video tape.
The matrix at some points approached dvd quality (nice sharp images and a decent framerate). It then occured to me that it only took me about 3 hours to download the thing. With a little boost in bandwidth (the download was not maxing out my connection), it would have been possible to stream the movie.
Jilles
First off, this sounds a lot like the New Coke fiasco to me. The audience is going to reject digital movies if the MPEG artifacts become visible. And what about when there's a storm, and the signal drops? Is the theater going to want to reimburse 400 people? I don't think so.
Second, is the price of a movie ticket going to drop? Of course not! The money savings will be pocketed by the theater or the production company.
And the men who hold high places must be the ones who start
To mold a new reality... closer to the heart
I don't think Apple insistantly promote proprietary format.
The QuickTime is just an architecture or wrapper or accessor to many codecs.
It gives "Same access method on different codecs"
So, if you make a codec as a plugin for a QuickTime, it works under QuickTime without significant modification of your existing source codes.
And the MPEG4 is just standardized.
Because the Apple is in multimedia industry not just for playing but also for editing media, QuickTime would be better editing format than just a MPEG S/W.
Because there are many tracks, Text, Sound, Video, etc, it would be easier to edit with QuickTime movie files than MPEG movie files. For example, if there are 2 scens which are transited by dissolve effect, the cut detection could be done without considering "slow transition" or "abrupt transition". But with MPEG, although when a media is created, there would be no problem, when handling existing MPEG file, that kind of manipulation would be difficult.
I still study the MPEG and QuickTime things.
I have why MPEG could be bettern than the QuickTime or the AVI or vice versa in question for long time.
Probably as a distribution media, MPEG could be better, but.. you know..
And the MPEG4 is not what you get from the MPEG1( video CD ) and the MPEG2 (DVD). The purpose is quite different.
And the QuickTime already has the foundation and Apple's own solution to the MPEG4 issue already.
And the Apple is a strong supporter of the MPEG4.
According to the MPEG group leader's own explanation in ETRI,located in Taejon, Korea.
Sooner or later, somebody is going to have to accept that in the near future, computers will be dealing with high-quality video in a similar way that today's computers deal with pictures and sound.
;-), watching the trailer for the Matrix. The video quality for a small file was astounding - we need this thing for Linux...
Ten years ago, just displaying a high quality image was a novelty, today your computer is doing it all the time with banner advertisments. In ten years time? Perhaps broadband internet will be prevalent, and banner advertisments will be full motion video. And in another ten years? Who knows...
What the industry really needs is a standardised, open multimedia format, much like the JPEG is a standard format for pictures today. Just imagine how the WWW would have developed if anyone wanting to write a program that could read or write JPEGs had to pay large royalties - this is what's happening today with systems such as MPEGs 2 and 4. The technologies may be esoteric now, but soon they'll become commonplace, and it would be best if they weren't encumbered by such restrictions.
The main reason for this post is because I've just been playing with DivX
Ford Prefect
Tedious Bloggy Stuff - hooray?
The stream format in MPEG-4 is the QT streaming protocol. You can also use the QuickTime file format over those streams if you want to. However, the codecs for whatever goes over the stream is not the Sorenson codec, which is the most used of the QT options in QT players.
MPEG-4 is a huge beast of many different layers. There's the streams, then the synchronisation protocols, then you have audio, video and geometry over the top of that. In the Version 2 spec and MPEG/J there are even whole APIs defined for interacting with the stream (They basically ripped my entire work that I did on the VRML External Authoring Interface and put it in there without so much as a single credit!)
Life is complete only for brief intervals in between toys or projects -- John Dalton
To the moderator:
/extremely/ annoying having to deal with this stupid nag screen if you have to watch multiple QuickTime movies.
No, the original comment is not flamebait. It is
Normally, commercial software has some sort of remind-me-later checkbox that will make the buy-me window pop up only at some rare occasions. Nobody will buy the software only because they don't like those screens, people just get a bad opinion of QuickTime. That's where MS Media Player wins -- no problems, no price, just works.
DISH Network, my man.
Digital signal, I've had it for three months now. Razor sharp images, and I've never once noticed any artifacts.
Of course, due to stupid legislation, I don't get NBC or CBS, and you know what? I don't miss them. What's on CBS anyways? Nightline, or 20/20, or some stupid pseudo-news show doing documentaries on the heartbreak of psoriasis.
Now I have Cartoon Network. Johnny Bravo dude. And my son can watch like 10 episodes of Pokemon every Saturday. I get like 20 channels of sports I don't ever, ever watch. And best of all, NASA channel, where they show the award-winning Earth Views - cataloged video from shuttle missions. This rocks. Cable sucks.
I just remembered this old Metallica song. . .
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
Of course, the whole idea of International Standards that can't be used by anyone is patently (heh) ridiculous. How to deal with this, I have no idea, short of creating a new codec that rivals MPEG-4, which is no easy task. Perhaps we can find a way to make our voice heard to the various lawyers deciding licensing, but I'd rate that as a rather low chance of success.
If truly open standards are to persist much into the next century, it is critical that patented methodologies be explicitly excluded from consideration at the get-go. Furthermore, methodologies which become patented down the road (for which there is no prior art - e.g. there is a patent application pending which no one is aware of) would be immediately disqualified or marked as "tainted" and replaced with an unencumbered equivelent.
To achieve this we need to dump the ISO as a standards body. This wouldn't be as difficult or traumatic as one might think - they are not know for creating great technical standards (remember the virtually unimplimentable, horribly ineffecient, and none too soon defunct OSI "seven layers"?).
I would submit that we should use and expand the auspices of the Internet Engineering Task Force, with a specific addendum to the charter prohibiting proprietary and patent-encumbered technologies from being considered, much less defined into, any open standard whatsoever.
More to the immediate point, we do need to develope our own standards and codecs (see OggVorbis and the nascient Open Video Disc mailing list for examples.
We need to take back our standards on both the implementation (defacto) front and the institutional (IETF vs. ISO) front.
The Future of Human Evolution: Autonomy
I'm sick of QuickTime movies I can't view in Linux and RealVideo movies I'd prefer not to download the player for
How the hell can anyone be "for" anything that you don't have a clue about? If this shapes up to be like 90% of all other big commercial ventures, it'll be Windows only. Maybe the Mac, and probably not Linux. Second, it's a custom codec, so if you're too lazy to DL realplayer, MPEG4 isn't going to be any different.
Don't hold your breath expecting any open source details on the custom codec. In fact, the mere mention of any customization of a standard (oxymoron), should flash huge warning signs.
Same ol same ol, is about the best I'm hoping for.
No sig is worth reading.
The only question now is, when are we going to see an actual standard released? All I've seen so far is Microsoft's proprietary implementations of MPEG4 drafts, and those certainly couldn't be called cross-platform. DivX is just a hack of the Microsoft codec, so that's just as closed in the end (though I do hope that contest will help speed things up as concerns platform adoption, particularly since that implementation is to be Open-Source).
One thing that windows has that Linux doesn't have is "standard" component based media libraries. I Say "standard" in the sense that you don't have 4 different system mpeg API's that all work in different ways.
I like Linux but it sure sucks in the "wow" department for digital entertainment and production.
I also came across the MPEG4, Web3D collaboration 'movement'. Check it out here
For all of you who are confused about MPEG (Moving Pictures Experts Group), check here .
You hate a format because a viewer compatible with your OS of choice is unavailable?!
That's a pretty narrow minded view of the world. The push is on for Apple to either release a Linux viewer and/or release the codec allowing a viewer to be developed.
Free distribution has one goal and many methods. Release the codec and allow many implementations, develop for all platforms, focus on one or two and screw the rest. All have advantages and disadvantages, but this is not our decision as much as it is our responsibility to convince towards the option of our choice.
I happen to love QT over Mpeg and I'm also a Linux supporter. Doesn't mean I stay up at night fretting over it. I don't bother with AVI's as my main machine is a PowerBook and my Linux, Irix, WinNT, and W2k boxes are secondary minimally powered systems for testing and learning. I don't for a second feel deprived of content by not bothering with formats such as AVI.
video on linux is poor at best. I have xanim, and every other player I could find, but I still have had spotty luck at best getting anything to play under linux.
windows media player is still many, many times better than anything available for linux.
If I could just view asf and DivX in linux, I'd be happy. So far RealPlayer is looking like the only hope.
________
1995: Microsoft - "Resistance is futile"
Keep in mind that even though MPEG-4 is an ISO standard, that doesn't mean it's free. MPEG-2 requires a $4 license every time anything MPEG-2 related is transfered to a consumer. That means no free players. We have no idea what MPEG-4 licensing is going to look like, but I'm not holding my breath. I ran across a document showing that all the big patent players explicitly objected to free implementations of BIFS, which AFAICT is the equivalent of Program/System Streams for MPEG-4. If we can't even get a free implementation of the transport stream code, how are we going to get a free implementation of the codec itself?
While anyone with a brain can plainly see that free decoders are critical to any kind of market share on the Internet, these are lawyers making these licensing decisions. We all know that lawyers live in their own little litigious world, and can't generally be counted on to have any connection to reality. I wouldn't at all be surprised to find MPEG-4 requires a $N license just to get a player.
Of course, the whole idea of International Standards that can't be used by anyone is patently (heh) ridiculous. How to deal with this, I have no idea, short of creating a new codec that rivals MPEG-4, which is no easy task. Perhaps we can find a way to make our voice heard to the various lawyers deciding licensing, but I'd rate that as a rather low chance of success.
Suggestions as to how we can deal with this problem would be welcome....
GStreamer - The only way to stream!
This appears to be the original, geekier version of the story before TechWeb's editors got to it. I'm not sure I like the idea of inserting personalized advertising objects into video streams.
We already have 200 channels with MPEG-2; how many do they want to give us with MPEG-4?
From : http://www.vxm.com/MPEG-4.html
The QuickTime file format will be used to store digital video, audio, and other types of content displayed using MPEG-4. According to Apple, the creator of the QuickTime technology, MPEG-4 will also contain things like MIDI, animation and 3D worlds. QuickTime will be used to store all of these things. The fact that QuickTime is being used for these kinds of things today is one of the reasons it was such a compelling choice for the MPEG-4 efforts.
No doubt, this QuickTime victory will likely ruffle some feathers at Microsoft, as its adoption by MPEG was actively "supported" by several of its most notable arch rivals -- Sun, Oracle, and Netscape (SGI and IBM also pushed for its adoption). But apart from smarting over the obvious NIH factor, if Microsoft wants to be MPEG-4 compliant, it must now incorporate the Quicktime format into its multimedia applications, like NetShow, NetMeeting, ActiveMovie, and Interactive Music. ActiveX Controls may also have to make some sort of accommodation with Quicktime. How likely and how soon Microsoft will deliver all these modifications after MPEG-4 debuts is anybody's guess.
Other resources:
http://www.internetwk.com/news/news0211-15.htm
http://www.cselt.it/mpeg/faq/faq-systems.htm#MP4-M PEG_4_is_based_on
This past summer, I had the misfortune of trying out digital cable. Our cable provider, Time Warner, is promoting "digital cable" as a superior alternative to the standard analog cable. With digital cable, you get:
I hated digital cable. The MPEG artifacts drove me crazy. I had technicians come out to my house FIVE times, and they couldn't do a damn thing. Most of them didn't even know what I was talking about when I pointed out the MPEG artifacts (they called it "macro blocking").
The problem was not my signal. They rewired my apartment, and it didn't help at all. The signal was coming in 100%, no lost packets or frames. It turns out that the real problem was the broadcast. It was being transmitted that way. Depending on the channel, it was possible for the signal to go through multiple D/A and A/D converstions before it got to my TV. The ironic part was that that channels I wanted the most (SciFi, Showtime, etc) were the ones that had the worst problems.
After about a month, I went back to analog. The picture is super-sharp in comparison, and I never experience any image degradation.
So what's this got to do with anything? The cable industry thinks that digital video is the way of the future. But MPEG only works when the bandwith is high enough, and the cable industry is not interested in providing enough bandwith to make the picture 100% sharp. We've all seen DVD's that have MPEG artifacts, and it really sucks. They say that streaming media over the Internet is bad. Well, streaming media from Time Warner over my cable TV is not much better.
And the men who hold high places must be the ones who start
To mold a new reality... closer to the heart
If it is an ISO standard (like the other MPEG compression schemes) that means anyone can get all the specs to it.
I'm a loner Dottie, a Rebel.
The nag movie in QT 3 & 4 does suck, but it's fairly easy to hack around it. It doesn't actually appear every time you invoke Quicktime, but the first time you do each day. Logic indicates that it needs to store the date when it does this, so you can banish it (virtually) forever by setting your computer's clock to some date far in the future (2019, say) and then opening a Quicktime movie. You'll see the nag one more time, and when you set your clock back to the present you won't see it again (until 2019 ;). This works on a Mac or a PC (and presumably on a Unix box when they finally get around to porting it).
I can lay claim to having come up with this idea on my own, but it's also been independently discovered by a number of people. I needed it when I was running a lab with 30 Mac & Windoze boxen, all using QT3 heavily and all nagging my users when they had better things to do.
It'd be nice if Apple would just get the message and lose the annoying advertising, of course. I'd be content if it played it one or a few times after QT was just installed, because that'd be of genuine utility to people who might want to know about the option to upgrade. Showing it every day is just dumb.
-- Life is short. Forgive quickly. Kiss slowly. ~ Robert Doisneau
I too hate Apple *and* Quicktime for their insistence on and promotion of proprietary codecs. The same goes for Microsoft *and* AVI files. Microsoft at least tried to make their NetShow player run under Linux, and maybe they'll port Media Player. (maybe their applications people will do it once they realize the potential captive audience involved :)
.ASF files under Linux....
What's the point of making a format if it's just a container for other, new, unsupported codecs? (okay, this can have *some* value, but then why make more than one? How about a generic container format, if we have to have them...) Then you can't reliably even make a player and say that it "plays AVIs" or "plays MOVs".
You can make an mp3 player and say it "plays MP3s". That's because the file format specifies this. The same goes for the other MPEG variants. I'd be happy as a clam if everyone would just ditch their proprietary video formats and collaborate on one. Then websites wouldn't have two or three copies of different video formats, (that all seem to play fine on Mac and Windows anyhow, and *still* don't often work on Linux) and everyone would have a better file format, at both low and high bitrates...
Of course, this is never going to happen as the situation is now. Hopefully free alternatives will eventually win out, but I doubt it. Incidentally, mtvp is a pretty good, free-beer player, and RealPlayer 7 really sucks on my Linux box, and I'd love to be able to play
---
pb Reply or e-mail; don't vaguely moderate.
pb Reply or e-mail; don't vaguely moderate.
There must be some ISO source code for MPEG-4, just like there is some for an MP3 encoder. Patching the ISO code and get a free player would be a similar thing like some people did it with LAME for MP3 encoding.
/content/ differ from raw MPEG-4? The parsing of the chunked ASF / AVI structure is very easy and well-documented, I've done that part myself for another app.
How much does Microsofts ASF's and AVI/DivX's
Heya. . .
.supposedly the release of Sorenson II will fix this.
.its quality is generally far poorer than either ASF or Quicktime, but its available on just about every OS and has a good compression ratio. Its also waaaay easier to set up than Quicktime or ASF! Working with a technologically-braindead instructor, I could set up a realplayer streaming lecture in under 20 minutes. Thats part of its alure. . .but real might be heading towards extinction if they dont improve image quality.
;o) ruling supreme with this format.
.from the streaming server setup end of things, all three are equally a pain in the ass. This article was interesting: a new player to the field, eh? We'll see how this all works out, but I think streaming content will have a big role in the direction the internet takes for the future.
I'm project staff at a development lab in a major University (U of Washington, Seattle). I have spent a lot of time working with different streaming formats and wanted to toss my two cents into the middle, since here at the U we're working in making streaming video a viable classroom tool for instructers.
Quicktime is an interesting streaming option simply because it can do so much. It plays extremely well with others--you can stream everything from old Autodesk Animator files to Flash with it. You can also control web pages from within quicktime streaming moviess, as well as doing internal menus and interactive movies (using something like Flash in conjuntion). The primary codec used with Quicktime is the Sorenson codec, which is great for low-motion videos but often breaks up on anything else. .
Real is kinda the everyman of streaming formats. .
ASF has me impressed!! Great image quality: however, the media player for mac just plain sucks, not to mention linux. If these two aspects of the format get fixed, I could easily see Microsoft (yeah, yeah, this is Slashdot
So there's a bit of information and my humble opinions. .
Thanks
Steve Martin
CTLT
steve0@u.washington.edu
http://students.washington.edu/steve0/
steve0@u.washington.edu
- - - - - - - -
Don't worry, being eaten by a crocodile is just like going to sleep in a giant blender.
I'm sick of QuickTime movies I can't view in Linux and RealVideo movies I'd prefer not to download the player for
Apple released the source to their Quicktime Streaming Server as soon as it came out last year as part of the Darwin project. And this year Apple ported Darwin to X86 since no one else would take the source and "do it."
All someone has to do is write the player, I am surprised one is not already there, or at least a mozilla plug-in.
For the interested, the applicable necessary source should be here, and there may be further stuff being worked on with MkLinux or LinuxPPC, where these things were ported to the Mac Linux distros.
I challenge academics to produce a free and open standard the kicks the hell out of all these lame commercial money grabs.
How we know is more important than what we know.
ASF -- Mpeg4 wrapper which improves streaming support for the format. The data which represents 99% of the file is just Mpeg4 standard.
MOV -- Quicktime -- Pretty much just an MpegX file with a little extra data wrapping it up. Same story as ASF really.
Real -- Pretty much useless.
DivX -- Mpeg4 wrapped in a AVI and called DivX. The codec itself just makes encoding simpler.
the real reason all of these exist is not because M$ thought they would be able to improve anything with ASF or anything like that, its because when they make it propreitary, they FORCE users to use thier player. Using thier player is just like advertising. That is why players are Free (usually) and they all have the same data in a slightly different package.
I might be wrong about the Real Movie thing, but it seems that way. They play really shitty, even if they are already downloaded. Someone needed to smoke more crack when they made that piece of crap.
Spring is here. Don't believe me, look outside!