Concrete Comparisons of Theora Vs. Mpeg-4 325
Posted
by
timothy
from the good-use-for-that-beta-you-downloaded dept.
from the good-use-for-that-beta-you-downloaded dept.
icknay writes "With the upcoming Firefox 3.5 and HTML5 video, there's natural interest in Theora vs. Mpeg-4, but without much evidence either way. Here's clips encoded at various rates to provide concrete comparison between Theora and Mpeg-4. Theora performs decently, but requires more bandwidth than Mpeg-4 (although this is a 1.1alpha release of Theora and Theora has a much better license than Mpeg-4). The quality comparisons are very subjective, but you can try the clips yourself and see how it breaks down. There was an earlier discussion about this, but it lacked much concrete evidence. (Disclosure: it's my page.)"
License (Score:0, Interesting)
So?
I'm serious. Can someone explain why this matters?
More to than bandwidth (Score:5, Interesting)
60% is bit of a price to pay, however IMHO the point of the video tag is tighter integration with your website than is easily achievable with flash. Hopefully theora will improve and compete with mpeg-4, but there are still many advantages to using it over flash for embedded video (for stand alone pages, it doesn't matter so much as most users have a plugin to handle mpeg-4)
*Interacts with the rest of the page easily (TBF actionscript, et al can achieve this)
*Much lower cpu usage. While flash is particularly bad, theora is particularly good
*Cross architecture. As people browse the web on phones, pdas, etc, this does actually matter
*Much less likely to be exploitable (TBF webhosts don't care, but users should)
*Open standards.
I don't think theora should be seen as simply a tool to replace flash videos but it should be seen as an opportunity to better integrate video into sites and/or make video content available to more people annoy people with video backgrounds
Re:License (Score:2, Interesting)
The license is the single most important thing. It determines whether or not you can use the software at all, or for your specific purpose, whatever that is.
Actually the license has really no effect at all for the end user in either of these cases. The only people who are effected by the license are people who are either creating H.264 encoders/decoders or those who are creating and streaming H.264 content. And even the costs of running a website with over 1 million subscribers is only $100,000 a year and if you have that many subscribers and that much traffic $100,000 is nothing to you. And for most small sites (anything with less than 100,000 subscribers) you pay no royalties at all.
Why nobody speaks about Dirac? (Score:4, Interesting)
Dirac is supposed to be a great opensource, patent-free codec, yet nobody seems to care a lot about it in all those HTML5 video talks....
Re:Help me out, please (Score:1, Interesting)
Believe me, license matters. I had to figure out the GIF mess for a small company years back... It took a lot of effort, we never knew if we were really clear legally (Unisys was inconsistent, unresponsive and just plain difficult), and to us the money did matter (small streams yada yada). I imagine there were thousands of companies like us. Those thousands of companies had millions of customers who got a more expensive product because of that crap.
Re:Seems pretty clear to me (Score:3, Interesting)
That doesn't necessarily prove that Vorbis didn't make a difference.
Re:Three "errors" in this test (Score:3, Interesting)
He claimed OGG is twice as good as MP3 is, not even 0.3%... Nevermind that MP3 doesn't even enter the equation when comparing H.264 to Theora (both because H.264 is typically paired with AAC audio and the video quality is the important question here, given the comparatively small size of audio), but that's also a blatant lie. You won't get double the quality out of OGG or AAC when compared to MP3, no way.
Re:Seems pretty clear to me (Score:3, Interesting)
Except that when a normal person uses the term "downward pressure" when referring to the cost of something it does mean that the price goes down
Really! "Downward pressure" cannot slow the rate of rise? "Downward pressure" only exists when something is actually going down? This is fascinating; I had no idea. (Because I'm not a "normal person", you see. How could I be expected to know?)
How do you find out these things? Is there a society that decides what terms a "normal person" can use? I'd like to join!
steveha
Re:Three "errors" in this test (Score:5, Interesting)
Not to mention the one that comes with hardware acceleration. even my cheapo $50 HD4650 comes with H.264, DivX, and WMV hardware acceleration out of the box with stock drivers. Does Theora even have hardware acceleration yet? With the move to netbook/Nettop and green computing hardware video acceleration is the present and the future. If Theora doesn't want to be DOA they need to get hardware acceleration for the big three (ATI, Intel, Nvidia) working NOW and get the GPU manufacturers to pack it in.
Because even on may nice super fast AMD Dual core having 1080p decoded on the GPU instead of the CPU just makes for a nicer experience overall. Not to mention the fact that GPUs use less power and create less heat than modern CPUs and with it already 100f in the shade here in AR temp matters.
Re:Surprisingly different (Score:3, Interesting)
Still, the developers of Theora found several glaring mistakes in the reference implementation of VP3 which brought it immediately to the same quality and bit rate level as MPEG-2. Everything since then has been vast improvements on both the encoder and the decoder.
It's like the LAME MP3 encoder. The vast improvements made in the encoder reaped huge benefits without even changing the decoder. With Theora, Ogg (and by extension, we) control both the encoder AND the decoder.
It's really not as bad as you think; it is actually quite better than you'd expect.
Re:Three "errors" in this test (Score:3, Interesting)
Ogg is definitely superior to those other audio codecs.
I've been playing with audio/video encoding a lot recently, at extremely low bitrates. I'm not quite ready to post my results, but here's the gist of it:
Using the Saga Frontier Intro [youtube.com] as a baseline, with lots of tingy sounds that you'd find in a PS1 game, these had the same subjective quality:
3GPP AAC+; 28kbit (But doesn't mux properly into mp4; only mkv)
CT AAC+; 36kbit (But explosions are still a little off)
FAAC; ~50kbit (But explosions are still a little dull sounding; didn't improve much with higher bitrates)
LAME MP3; 72kbit (sounds fine, but way too much bandwidth consumption)
Ogg; 32kbit (at 28kbit explosions are a bit off, but 32kbit is fine; also only muxes into mkv. A 29kbit ogg matches a 28kbit AAC in filesize)
Take it with a grain of salt, since this is only one kind of audio. No speech, and nothing too complex. My soundcard probably also affects the subjective results of the tests - Realtek integrated HD audio. :P
The only video codecs I tested were xvid, x264, and youtube H.263. Naturally anything encoded on your own computer will be better than what Youtube manages. H.264 is a very good choice for games/screencapture, once you tweak the settings a bit. xvid(and I suspect Theora) not so much.
Fiasco! The fractal codec (Score:3, Interesting)
Does anyone remember it? The author screwed up and coded parts of it on university time so had to revoke the GPL license since they could not prove which parts were or were not university property.
I spent a month compressing a highly scaled video clip and was able to put about 20 seconds on a floppy. I could compress a complex jpeg with the static compressor into 4 - 20k.
http://www.linuxjournal.com/article/4367 [linuxjournal.com]
Bandwidth wise it's marvelous, it's the number crunching to compress it that's the killer. I'm not a coder and his paper is marginally comprehensible but there is no way I could recreate the codec.
Re:Three "errors" in this test (Score:3, Interesting)
But publishers are nothing without customers, no?
And customers are nothing without food, but we ain't asking the cows on this one.
Consider the network effect - the more users there are in the network, the more valuable the network becomes, and alternatives become less valueable. Same thing here. If all the users are using the solution that is most convenient for them, regardless of license, that's the most valuable place for the publishers to be.
Users go to where the publishers are, not the other way around. UNIX have always been the superior OS, but since Windows had more apps (for purely historical reasons), guess which system people used.