Slashdot Mirror
Sun, Philips Push MPEG-4 Up Steep Hill
Kellym writes: "Sun Microsystems and Philips Digital Networks are putting their chips on MPEG-4 in the battle to determine the streaming media standard of the future. The companies have agreed to expand their year-long relationship to promote and develop MPEG-4 technology for broadband and wireless markets. The companies have partnered on marketing and have agreed to share technologies. In the most recent deal, Philips licensed Sun's StorEdge Media Central server technology. Philips said it will include the technology in a WebCine Server MPEG-4 system it is developing to run on Sun's Solaris Operating Environment and Sun Cobalt servers."
I think MPEG-4 (read "DivX :)") is cool... but, i'm afraid of market domination by companies like Phillips and Sun...
are there any other companies involved in MPEG-4 that have competing products?
I have mpg1, mpeg2 and mpeg2.5(mp3) hardware in my house and car, I'm pretty sure mpeg4 will be there shortly.
People are trading VCD because they play on newer dvd drives. If they come out with a DVD player that plays some Mpg4 format, everyone will jump all over it. (IMHO)
I have a Dazzle and 2. I started encoding home movies on VCD and then migrated to SVCD for higher res. DVD-R is still a little pricey. If they come out with a Dazzle type of encoder with Mpeg4, I can keep using cheap CD's and make a "mini-dvd" type of disc. (Also DVD-Rs dont have burn proof yet, 10 dollar coasters, oh boy.)
I can't help but wonder where this leaves me. Some (few ?) of us are still using dialup 56K (or lower) connections. All this wonderful new content is great, but remember it takes some of use a bit longer to download stuff. Yep, you could say "Get with the times" and "Get Cable or ASDL or something" but it's not exactly affordable in some places... sigh
I just hope that Sun & Co. will not try to go for their profits immediately. It'd be better to lower the prices, perhaps sponsor some Open-Source work, make it a popular thing among the consumer. Otherwise, it'll all be crushed by M$'s "we do it for free" strategy.
I want everyone and their mothers to support MPEG 4. As a web developer I am simply SICK of having to support 3 different media players and a bunch of different OSs. I want to be able to stream 1 format that can be played on every media player on every OS.
It's time we stopped tring to one up each other with new codecs and media players. We need to seriously PICK SOMETHING. This is the only way technology gets adopted by the masses.
I don't have to upgrade my CD player, DVD player, radio, or microwave every 6 months, why should I have to update my stupid computer's media players.
Consumers HATE adding plugins, codecs, and players...and I hate developing for 1 million different things.
MPEG4 is dynamic, auto-upgradable, and will make me a muuuuuch happier camper.
"Things are more moderner than before- bigger, and yet smaller- it's computers-- San Dimas High School football RULES!"
Is this the same "MPEG-4" as used by the DiVX codec? Will we see standalone DiVX players possibly, like we have MP3 players for audio?
There's 10 types of people in this world, those who understand binary and those who don't.
Then watch us build your 'high-speed distributed video serving network' in our spare time at no cost to you, while MS, Apple, Real and AOL fuck around for years trying to court 'industry players' and 'content copyright holders'.
If the whole 'Napster' thing proved anything it's that there are a shitload of people out there desparate for content they don't get supplied through 'mainstream media', but nobody wants to pay the same people who have been screwing them down at the record store for years.
The Linux community is crying out for decent video tools, and none of the other players except maybe Real seem particularly interested in providing them.
To beat M$ in this area, Sun and Philips are going to need some serious help, and the only place they're likely to find it these days is with the Open-Source/Free Software community.
We have more clout with M$ than the US Justice department does, anyway.
I gots ta ding a ding dang my dang a long ling long
DivX (from Project Mayo) is based on MPEG-4 and will at some point be fully MPEG-4 compliant. It still needs more features but the existing features are according to MPEG-4. .avi
What is DivX
Forum discussion about
MS-MPEG4 is - of course - a different take on MPEG-4 with a different feature set.
Btw, is there hardware available *now* for reasonable cost that would allow me to speed-up the ProjectMajo/DivX/MPEG-4 encoding process on my home PC?
------------------
You may like my a cappella music
Many people don't really understand what DivX is. There's the DivX
I've been looking into streaming MPEG 4 video off a Linux server and it's still rather immature. FFMpeg looks like it might be getting there, but I quote from the FAQ: "New developments broke ffserver, so don't expect it to work correctly. It is planned to fix it ASAP."
It would be nice to find a good OpenSource (pref. Linux) solution for streaming MPEG 4 content (from a Video4Linux BTTV device). Does anyone know of one?
An AMD Athlon should do you quite nicely :)
Dave
I write a blog now, you should be afraid.
An AMD Athlon should do you quite nicely
My last round of encoding on an Athlon 1200 took more than a day (120 min movie, dual pass).
Currently, my home PC (Duron 800) is busy encoding another movie, encoding time is estimated at 30 hours.
No, just a CPU doesn't do quite nicely. That was the point of my original question. I want a dedicated DSP to speed-up things.
------------------
You may like my a cappella music
See for example: http://www.e-vue.com/about/may072001.cfm
One of the ways in which the MP4 standard is quite dumb is that the "security" features are an optional extra. You don't have to have lots of ornate key management policies and encrpytion schemes in order to enjoy the benefits of increased compression/versatility/whatever.
But (as we all know, I guess) that'll never work the way they think it should. This is simply because so long as there is an "insecure" standard for exchanging content (alongside the secured version), people who rip stuff off and share it with their friends will use it. The only ways you can stop that are: (a) pass an unenforcable law like the DMCA, or (b) get rid of all "insecure" standards. Solution b is not workable because everyday life would grind to a halt if everything had to be authenticated with military-grade encrpytion. So we're stuck with the laws (which, incidentally, don't necessarily go away once the companies which bought them go bust).
My conclusion is, therefore, by all means adopt MPEG-4 because in almost every other way, it rocks. Don't be scared by the "rights management" bullshit, because as long it's optional, it's worthless.
--
anonymous CVS: geeks check in - they don't check out
These sigs are more interesting tha
There is a player for Linux [...] which gives excellent hardware video acceleration on Matrox G200 [...]
Uhm, thanks, but I am looking for hardware acceleration of the encoding process.
------------------
You may like my a cappella music
I assume that when talking about MPEG-4, we're talking about the whole standard, and not just the MPEG-4 Video codec.
I can really see the MPEG-4 Video codec taking off, as it offers superior video quality for low-bandwidth connections, but the MPEG-4 standard as a whole...hmmm I don't know
Admitedly, I've stopped following the developments of the MPEG-4 standard closely, but the last I saw it was quite a bloated standard that incorporated the video codec, much of VRML and some Java scripting. All these parts of the standard are necessary for things like scene graph rendering of video objects (turing off backgrounds in video etc.), and interactivity.
Unless a subset of this functionality (profile) is decided on for internet use, I can't see the whole standard taking off. However, I think that the video codec on its own has a lot of potential.
------------ jay*arr*tee
MPEG-4 is not the panecea everyone seems to think it is. Currently MPEG-4 is heavily patent encumbered ( see http://www.m4if.org/patents/ ). The result is I doubt you will find it possible to produce a legal open source MPEG-4 codec.
The standard is also being put forth by ISO, a notoriusly shitty standards body. Do you want to pony up more than $1,000 to get a copy of the standard so you can begin making a standards compliant implementation? That's roughly what the MPEG-4 standards docs cost. Even if we disregard the patent concerns, this represents a serius barrier to entree for anyone wanting to do an open source implementation of the codec.
ISO ( and it's child the ITU-T ) are designed to be used as weapons by corporate players against each other, not to produce good clean standards that can be used by all.
Try looking at the ogg tarkin project http://www.xiph.org/ogg/index.html as a group trying to pursue a non-patents encumbered video codec with a truely open standard ( I don't consider ISO standards to be open because of the intense barriers to entree like the expense of the standards docs).
This is a very, very funny post in the context of a discussion about digital video. You clearly lack anything resembling a clue. Apple is one of the major backers of MPEG-4 and one of the major contributors to the standard. The MPEG-4 file format is, in fact, based on QuickTime's. QuickTime and Apple's open source streaming server will no doubt fully support MPEG-4 before too long. And I'm sure they'll really support it, unlike some of the half-assed attempts we've seen. Final Cut Pro is making serious inroads into the low-end and mid-range (read: under $25K) of the video editing market. Throw in iMovie and DVD Studio Pro, and Apple is doing more interesting things with digital video than any other single company I can think of.
(Oh, and I doubt Apple Expo was canceled for the reason you give; the most recent Macworld Expo set attendance records. Again.)
This space unintentionally left unblank.
sounds like you weren't using the encoder optimsed for the Athlon SIMD instruction set
There is a specific encoder for AMD? Afaik, the DivX codec (the Windows binary) is compiled for both platforms and is partly optimized for both.
I am also doing some additional filters on the material in VirtualDub - e.g. resizing and subtitling.
It all adds up, making the whole process slow.
------------------
You may like my a cappella music
You can encode a 90min action dvd into 700mb with sound(160kb mp3) and afterwards watch it on a computer without almost any loss in perceptable picture quality(interestingly sound suffers almost more).
Downside is that 2-pass variable bitrate encoding for 90min movie takes roughly 13 hours on a 900mhz athlon and setting it up is not exactly trivial..
Check out www.divx.com for more info.
The wavelet compression of MPEG-4 offers better quality than JPEG with file sizes approximately 25 percent of the size for Web quality. Wavelets dynamically allow servers to reduce bitmap file sizes (which also affect quality) when working with lower bandwidths, reducing the need to create different presentations to account for a variety of connection speeds.
For audio, MPEG-4 offers a wide variety of features, such as codecs for low-bitrate speech and general purpose audio. For servers, the audio component offers several quality layers which, based on bandwidth, can be dynamically adjusted. Given how MP3 became a popular music file format MPEG-4 could well follow the same trend.
For Rich Media, MPEG-4 constructs everything out of media objects, such as video/audio streams, stills, text, etc. Further, these media objects can be mapped to a scene as opposed to simply working within a rectangle. Also, MPEG-4 can blend the capabilities of Flash, VRML, Shockwave and digital video into a single file format, making it easier to deliver content over slower connection.
MPEG-4 Variations Version 1 of MPEG-4 offered nine video and four audio profiles. Version 2 added seven more video and four audio profiles. These profiles create subsets for different marketing options. Profiles, or features, are designed to work on different platforms. An example would be cell phones and on the other end of technology, HDTV. Into the Future Among other things, MPEG-4 has been slated to replace the current MPEG-2 standard in the cable industry, meaning among other things, that the companies could triple the number of channels available and could implement interactive capabilities.
MPEG-4 also offers MPEG-J, a Java library for controlling MPEG-4. Combining the two would let developers embed a Java applet in the MPEG stream, making possible such innovative cable options as interactive advertisements, home shopping capabilities and more. Other possibilities include videoconferencing, security observation, etc.
A potential barrier to widespread MPEG-4 use are the licensing and fees issues, due to several companies having patents that apply to aspects of MPEG-4. According to Shelly: "There is a group known as MPEG LA, based out of Los Angeles, that are working with a number of people who hold patents. They are attempting to speak for the entire industry, but not everyone who owns a patent for MPEG is a part of that group." The challenge is to combine the patents into one licensing fee, which is still in process.
The preceding is from: http://streamingmediaworld.com/
Curious George
***General Consultant to the Human Race*** My opinions are free. You get what you pay for.
Unfortunatly even 90 minutes at 700MB still means a maximum realistic limit of arround 20-50 users simultaneously streaming from one server on a hundered megabit fully switched super high powered network. And forget trying to do real time encoding at the same time.
As an aside:
What is the best cross platform (client), linux based, open source (or at least free) solution for streaming 5 videos at arround 200-500kbits/second (preferably running on a low end server)?
It is from Sigma Designs. They make PC cards, too, but none yet with this particular chip. And, oh, yes, they support Linux. Read the specs, though.
I contacted MPEGLA because I'm the author of an MPEG-1/2 encoder (http://mjpeg.sourceforge.net) and I wanted to lay the groundwork for an official Debian distro. The official response was "everyone who *distributes* an MPEG encoder
must obtain a license from MPEGLA and pay $4 unit". This is fine and actually quite reasonable(ish) for a hardware vendor. For an open-source project
Since M$ bundling of their codec more or less precludes any commercially viable closed-source MPEG-4 codecs I think we can safely conclude MPEG-4 is dead dead dead as a mainstream platform in the PC space. Informal derivatives (the DivXes) of course will carry on, but I think its safe to assume no-one will be broadcasting or pressing disks in those formats.
I am sorry to be a wet blanket, but I don't see how streaming video has much of a future for mass acceptance. There is an architectual problem with this approach. True enough...broadband rollouts have made it possible to distribute video with relative ease, but the transmission requirements are just so great that it is really difficult to justify the bandwidth investment, especially on the server end. More than that, I don't see how any streaming media company can try to provide robustness of service.
My view is that there are some applications that are well suited for point-to-point communication mechanisms such as IP. If we were discussing the possibility of using this technology to enable video phone or other video conferencing applications, I would be a bit less pessimistic. But, we need to recognize the fact that some transmission modes are inherently broadcast: one source, many many listeners. We can talk about implementations of IP broadcast to save upstream bandwidth, etc, but the fundamental scaling problems are still there. Many networks need to carry identical copies of the same data.
Last Tuesday, we witnessed the fragile nature of current servers/onramps in dealing only with high levels of http traffic. How many of us got anything more than a server timeout from cnn.com last Tuesday? But it wasn't very hard to just punch 204 in the DirecTV remote and there it is. Streaming anything over IP has a long way to go to catch up with truly broadcast mechanisms.
If such streaming applications are going to be attempted, the entire process needs to be decentralized. Video-on-demand needs to stream from many servers at once to improve robustness. It needs to automatically replicate popular data to servers in different parts of the Internet, etc. The current work in P2P networks is focusing on just this type of scheme. Of course, doing so flies in the face of DMCA and the media wonks who want paid. Centralization provides control and a single point of failure. Decentralization provides robustness and loss of control.
I question whether or not streaming media will ever become the service that Sun, Sony, and MS are envisioning. The only way to make it work is by taking the P2P route and most of those approaches are "pirate" in nature. It may come to fruition with P2P swarmcast/distributed-caching schemes, but I doubt that using it will be legal.
The group involved are likely to allow free use for a good while, until it attains an MP3-like level of pervasiveness- then they'll pull a Fraunhofer.
Remember, "standard" doesn't mean "Free"(as in freedom,) it just means that everyone uses it.
Look to Ogg Vorbis as an example of what could be a Free standard.
What we call folk wisdom is often no more than a kind of expedient stupidity.-Edward Abbey
It's the insecure format that becomes the true (de facto) standard, since everyone's using it to pass things around while the bigwigs are fighting amongst themselves, trying to make the de jure standard.
What we call folk wisdom is often no more than a kind of expedient stupidity.-Edward Abbey
An interesting step forward. Since MPEG4 is the protocol behind video transmission over bluetooth links, can bluetooth-enabled HUDs be very far away?
:-)
Therein lies the potential for reasonable quality video from you cellphone/PDA/mobile-device...
I'm certainly looking forward to it.
--The more you know, the less you know.
Ah.. But my point with 700mb was that it fits on a cd. Resolution in question is the full dvd-resolution with black borders cut off.. If you're willing to halve the resolution the filesize will probably roughly be halved too(need better quality for those fewer pixels)..
This isn't DivX's fault, you must have made a mistake somewhere along the line. 120 MB should be more than enough to watch. I have a lot of Cowboy Bebop episodes (which are about 5 minutes longer then american shows) that are usually about 60 megs and the quality is pretty good. Well past acceptable.
This Wiki Feeds You TV and Anime - vidwiki.org
The answer to bandwidth issues is smart multi-casting in the pipe. The MBone experimental system showed this was quite feasible. I am unclear as to how much of this has gotten/will get into the actual backbones the majority of traffic goes through but if the demand is there then there is a proven solution as long as you group users together into shared time-slots.
The next step beyond that is smart content caching in the network. I had no problem with CNN on Tuesday because I share a caching server with a large number of users. One person gets through and we all get the results. The next level is caching in the back bone so someoen who subnscribes "late" to a channel gets the "old" data first. Think of it as a TIVO in the router.
In re reliability, the solution again is caching, in this case local caching to cover any reasonable sized "hiccups" in transmission.
The solutions arent rocket science, they just require enough financial icnetive to make them worth developing and installing.
Specifically, the newest incarnation of DivX ;-) is compliant with MPEG-4 simple profile video. For an explanation of MPEG-4 profiles, see the MPEG 4 standard website.