Slashdot Mirror
Streaming DVD Video over the Internet
Sexy Commando writes "According to this article on ZDNet, the new codec, H.264, is able to stream DVD quality video using bandwidth as little as less than 1Mbps. The new codec requires 3 to 4 times as much CPU power than MPEG-2 to process the video. Now we can have two movies on 1 CD. Cheers."
That's a LOT of pr0n!
3cx.org - A truly bad website.
Well, I've got no phd in DVD technology, but the AC3 sound alone would take up far more than 1mbit all by itself right?
One of the reasons im not into watching movies on my PC is that I cannot take advantage of my DTS gizmos.
If this is just for video quality - Count me out.....
True ravers don't need drugs
Quote:
[..] making the size of video files a top hindrance to Hollywood's Internet video-distribution plans.
Yeah Right. Just like the Music Industry's plans for Internet music-distribution...dream on.
Moderation: +4. Modded 70% Funny and 30% Overrated. 100% Saturated.
I'm reasonably sure that I just heard Jack Valenti spinning in his grave. The MPAA thought they had problems before...
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
Average film > 2 hours = 7200 seconds; assuming constant bandwidth @ 1Mbps gives a size >=~ 858 MeB per film. I suppose you could go lower than DVD quality, but personally I just dump VOBs to my harddisc, as ripping to a compression algorithm like DiVX takes far too long, so 'two movies on 1 CD' sounds, well, a bit far fetched...
James F.
Here is the VideoLocus press release for H.264/MPEG-4 AVC.
I once shot a man who posted too many, "Imagine a beowulf cluster of these"
Finally a consumer need for CPU horsepower !
he: Hey babe, wanna watch a movie ??
she: sure
he: wait till i boot the player
she: ??????
he: here we go...
she: is it me, or is it getting hotter in here??
he: thats just my dual XEON box chewing....
Why are other peoples sig's always more witty ???
how much processing power did divx and mpeg2 require compared to good old mpeg, anyway?
the ones encoding always have time to wait a little longer, and mpeg2/divx already requires quite a lot of processing power for playback...
what people want is a quality:space ratio as high as possible, and considering the fact that very many actually have a new pentium 4 or athlon xp anyway, they don't mind watching good quality movies.
remember moore's law...if this codec is getting widespread adoption 3 years after mpeg2, the processing power availible is already 4x what it was...
H.264 exists as MPEG-4 part 10, basically using the AVC rather than the ASP profile for encoding.
Supposedly, it offers up to 2-4x size reduction over the MPEG-4 ASP.
However...
For anyone who has extensively played with the existing ASP codecs available (basically XVID, DIVX, RV9, and WM-whatever), the quality matters a *lot* based on the implementation. And not in any consistent way, letting you pick "codec X does the best job". Nope, more like "on low-motion sequences, codec X does best. For detail, codec Y. For minimal artifacts but some bluring, codec Z", and so on.
I see no reason to expect H.264 will follow any substantially different path. In another 5 years, it might well let us get a DVD quality movie onto 1 CD. For now, don't hold your breath about this changing the scene overnight. By the time this really does make good on its potential, we'll have the bandwidth and storage to make it unnecessary.
Although two movies on a cd sounds farfetched, even a single dvd-quality movie on a cd would be a big jump. Yes there have been lots of improvements in Divx, but on single-disc movies it's still quite clear at times that you're watching a divx and not a dvd.
The way I see it, Divx needs 3 things before it becomes a major threat to DVD.
1-Players capable of playing multiple soundtracks, for multiple languages and/or commentary.
2-Componant Divx Players, or more likely DVD players that can also play DIVX content. People want to watch movies on their tv, not their computer, and only geeks have good tv-output capabilities.
3-Able to fit even longer movies on a single cd with near dvd-quality. No one like changing (or flipping) disks in the middle of a movie.
Meet these demands and allow even a layman to pop a DIVX disk into their dvd player and sit back with a bowl of popcorn, and the MPA has a major problem on their hands.
-Chris
--an unbreakable toy is useful for breaking other toys--
It claims this new codec can get the same quality at 33% lower filesizes than other MPEG-4 codecs, but it doesn't say WHAT MPEG-4 codec. There is more than a 33% difference between existing MPEG-4 codecs alone! Are they comparing this to DivX 5.x, arguably the current leader in quality? Or are they comparing it to Microsoft's ISO MPEG-4 encoder, with it's horrid quality?
Regards, Guspaz.
I like watching dvd's when i want and how i want, and they're already an affordable 9 bucks at alot of stores.
What i want to see is Launch.com use this for high quality VIDEOS as i'm sick of vivendi pushing the crap they want us to see and luanch.com is an awesome place to see videos of the songs we love.
The new codec requires 3 to 4 times as much CPU power than MPEG-2 to process the video.
Long ago, in the before time, when I had an Athlon XP 2100+ (1.73 GHz, before I fried it and got thrown back to a 1.4 GHz athlon and then I fried that and got thrown back to a 600 PIII) I was able to rip DVD's and convert them to DivX in real time (a little faster actually, around 34 fps.) Now I don't know the differences between MPEG-4 and MPEG-2 but 3 to 4 times as much CPU power doesn't sound too pleasing. Right now I'm riping a DVD, err wait no, I don't do things like that it's illegal. Hypothetically speaking, if I were ripping a DVD right now, there would be 20 hours left because on a 600P III DVD's take a long time to convert to DivX (or so I'm told.) It takes all day for me... err not me, it takes all day for a person with a 600 PIII to convert a DVD to DivX. *shudders thinking about when that person ripped the Matrix for 30 hours and had 3 files, 2 700 meg files and one 50 meg file*
I'm going out of a limb here, but I'm guessing that they really aren't all that concerned with the ripping/P2P crowd. Instead this is intended for media companies that'll make one master digital MPEG4p10 stream, and cable companies can use it for PPV streams/movies on demand, etc. Digital boxes right now are being sent dozens of streams in, I believe, MPEG2, and the bitrate cannot keep up with fast action, and there are significant artifacts in parts. By going to the new codec, not only will they improve the video, they'll lower the bitrate as well, allowing them to fit more "channels" in a given frequenecy.
The economist has a great article reviewing the latest codec offerings from different players. Specifically DivX 5.0 "is said to be particularly good at preventing tearing, a playback error that occurs when the software cannot render the video for display at the same pace that it is being decompressed and fed into the media player. And a new codec from supersecretive Pulsent claims to be object rather than block based. Whereas block-based compression and object-oriented codecs slice up backgrounds and foregrounds into grids, the Pulsent approach actually pinpoints real-world items in the frame--such as a person, tree or building--and processes each element separately. story here
This is exactly what intel and AMD need. A real reason for people to upgrade their hardware.
For most people even a 400Mhz system is enough.
Simply writing bigger and clunkier apps (a la microshaft) is not a good reason for me to dump[ my hardware.
It seems to me that the limits of compression technology are self inflicted. We don't do better compression because it takes too long to compress/decompress. However, with the improved speeds capacities of new hardware we can break those barriers.
When will we see this compression to allow more bandwidth down a dialup line?
Send me that a pair of 1Thz AMD CPUs!!!
comment directly in my journal
There has been a lot of research in deriving 3d models from motion video. This would of course lead to dramatic reduction in bandwidth requirements by sending down a 3d model of the set to a renderer and then transmitting only motion through the set along with variations from the set projected to 2d. This requires huge amounts of processing up front but very little at the decompression/rendering end compared to a lot of other methods. The MPEG4 3d modeling codecs seem to be an after-thought based on provision of manually constructed 3d models (often the examples given are of rendering human faces from 3d models which is almost the opposite of what should be going on with motion video compression -- the sets should be 3d modeled leaving more problematic features like faces to the residue ) not a fundamental aspect of automatically constructed 3d models during compression.
Seastead this.