We Will Have An Answer To Youtube On Pandora -- Webm.


Dead1nside

Well-Known Member
Joined
Sep 20, 2007
Messages
1,057
Location
UK
Website
www.jonathanpritchard.com
This is very important for the wider web, and quite significant for Pandora as well.

http://webmproject.blogspot.com/

Edit: A more critical look at VP8, http://x264dev.multimedia.cx/?p=377
 
Tripmonkey_uk said:
Just been reading that myself, very good news.
Just hope that it's taken up by the industry :)

ARM and TI are both on the list of supporters :) Apparently it's less processor-intensive in decoding, which can only be a plus for us.
 
Last edited by a moderator:
Already available on Youtube HTML5.
That's a good thing, we will finally get rid of those friggin' Flash video players!
 
I downloaded ff3.7a5pre, headed to youtube and followed instructions, but couldn't get the html5-webm emblem. In some cases there was the html5 emblem, but then the video would just error on missing plugin :(

anyone?

EDIT: weird html-tags?

EDIT2: Just tried the latest chromium snapshot, no luck.
 
B-ZaR said:
I downloaded ff3.7a5pre, headed to youtube and followed instructions, but couldn't get the html5-webm emblem. In some cases there was the html5 emblem, but then the video would just error on missing plugin :(

anyone?

EDIT: weird html-tags?

EDIT2: Just tried the latest chromium snapshot, no luck.
Did you append &webm=1 to the URL of your search? Took me a couple of tries to find a WebM video, but it worked pretty amazingly, actually.
 
Last edited by a moderator:
limetang said:
B-ZaR said:
I downloaded ff3.7a5pre, headed to youtube and followed instructions, but couldn't get the html5-webm emblem. In some cases there was the html5 emblem, but then the video would just error on missing plugin :(

anyone?

EDIT: weird html-tags?

EDIT2: Just tried the latest chromium snapshot, no luck.
Did you append &webm=1 to the URL of your search? Took me a couple of tries to find a WebM video, but it worked pretty amazingly, actually.

How will I know if I am watching a WebM video? I added &webm=1 to my search for openpandora and found nothing, but when I searched for 'smashing pumpkins' I found a few results.

For example, I chose this video, which is in HD.... http://www.youtube.com/watch?v=TqWtI9OQ8Sw

Any ideas? I'm running Chrome (5.0.375.38 beta) on Win7
 
Last edited by a moderator:
limetang said:
Did you append &webm=1 to the URL of your search? Took me a couple of tries to find a WebM video, but it worked pretty amazingly, actually.

Yes. Could you post an example of a working one? Thanks :)
 
Last edited by a moderator:
B-ZaR said:
limetang said:
Did you append &webm=1 to the URL of your search? Took me a couple of tries to find a WebM video, but it worked pretty amazingly, actually.

Yes. Could you post an example of a working one? Thanks :)
Right, what I did was:
  • Download one of these nightly builds (the Windows one, in my case).
  • Start up the developer preview and enroll in the Youtube HTML5 beta here.
  • Search trailers, add &webm=1 to the URL, then select the Avatar trailer.
And you should see something like this (unfortunately, the video didn't appear in the screenshot...):


It appears that Google are only enabling WebM for 720p videos at the moment, so it's probably best to look for them.

Edit: didn't realise how big the image was...
 
Last edited by a moderator:
While it's good to have some more or less open video codec, the only thing I can think of right now is "missed chance". Even though it's not bad, it's not good either. I would have expected better from Google and On2 considering how flawed the code is after so many years.

Oh and the name... WebM... wtf were they smoking?

edit:
If someone is porting this to devices such as the Pandora, brace yourself for some optimising rework:
Before I go into specific components, a general note on code quality. The code quality is much better than VP3, though there’s still tons of typos in the comments. They also appear to be using comments as a form of version control system, which is a bit bizarre. The assembly code is much worse, with staggering levels of copy-paste coding, some completely useless instructions that do nothing at all, unaligned loads/stores to what-should-be aligned data structures, and a few functions that are simply written in unfathomably roundabout (and slower) ways. While the C code isn’t half bad, the assembly is clearly written by retarded monkeys.
 
Zotty said:
While it's good to have some more or less open video codec, the only thing I can think of right now is "missed chance". Even though it's not bad, it's not good either. I would have expected better from Google and On2 considering how flawed the code is after so many years.

Oh and the name... WebM... wtf were they smoking?

edit:
If someone is porting this to devices such as the Pandora, brace yourself for some optimising rework:
Before I go into specific components, a general note on code quality. The code quality is much better than VP3, though there’s still tons of typos in the comments. They also appear to be using comments as a form of version control system, which is a bit bizarre. The assembly code is much worse, with staggering levels of copy-paste coding, some completely useless instructions that do nothing at all, unaligned loads/stores to what-should-be aligned data structures, and a few functions that are simply written in unfathomably roundabout (and slower) ways. While the C code isn’t half bad, the assembly is clearly written by retarded monkeys.

I'm hoping that TI and other large companies are now going to have an interest in developing software decoders for DSPs like that in the OMAP3 because VP8 and WebM are going to become big, big news.
 
Last edited by a moderator:
Zotty said:
While it's good to have some more or less open video codec, the only thing I can think of right now is "missed chance". Even though it's not bad, it's not good either. I would have expected better from Google and On2 considering how flawed the code is after so many years.

Oh and the name... WebM... wtf were they smoking?
The problem is, regardless of quality, this is still likely to be full of patented tech, and we're not talking weird submarine patents like those that might be in Theora. As DarkShikari puts it "a pithy, if slightly inaccurate, description of VP8 would be H.264 Baseline Profile with a better entropy coder".

So basically this is a mediocre codec that probably isn't any freer than H264.
 
Last edited by a moderator:
Before I go into specific components, a general note on code quality. The code quality is much better than VP3, though there’s still tons of typos in the comments. They also appear to be using comments as a form of version control system, which is a bit bizarre. The assembly code is much worse, with staggering levels of copy-paste coding, some completely useless instructions that do nothing at all, unaligned loads/stores to what-should-be aligned data structures, and a few functions that are simply written in unfathomably roundabout (and slower) ways. While the C code isn’t half bad, the assembly is clearly written by retarded monkeys.
:lol: Retarded monkeys, I bet someone feels good about themself.
 
limetang said:
Right, what I did was:
  • Download one of these nightly builds (the Windows one, in my case).
  • Start up the developer preview and enroll in the Youtube HTML5 beta here.
  • Search trailers, add &webm=1 to the URL, then select the Avatar trailer.
And you should see something like this (unfortunately, the video didn't appear in the screenshot...):


It appears that Google are only enabling WebM for 720p videos at the moment, so it's probably best to look for them.

Edit: didn't realise how big the image was...

Thanks, that version did the trick. The nightly that was linked from the webm website did not show that same trailer.
 
Last edited by a moderator:
Yeah I read about this just now and am highly disappointed. Not just in the quality but at the continuing patent threats.

Either way, there's nothing stopping us from having x264/ffmpeg h.264 on the pandora, I just hope Google and other don't prematurely drop H.264 videos before he patent issues are cleared up and they can come up with a free codec which is competitive.
 
Mr.Confuzed said:
:lol: Retarded monkeys, I bet someone feels good about themself.
Imho he's allowed in this case. I know him (not personally though) through some 'less legal places on the web' and the guy really knows his stuff. It's scary :lol:
 
Last edited by a moderator:
Yeah I read about this just now and am highly disappointed. Not just in the quality but at the continuing patent threats.
The quality sounds ok to me. It may or may not be as good as h264, but in my opinion it doesn't really need to be.

The patent threats are a shame though. Also, Microsoft will not provide any native support (they said: use a plugin).

So instead of two formats, we now have three formats :(
 
Caine said:
Also, Microsoft will not provide any native support (they said: use a plugin).
Personally I've always thought this is the way it should be done.
I don't understand the rush to embed multiple decoders into every single application that needs to play video.
Every OS has a perfectly good multimedia backend (Directshow/Gstreamer/Quicktime?), just use that.

I realise there has been an issue with codecs in the past, but that should be a problem for OS vendors to solve.

If it wasn't for stupid software patents everyone could just stick ffmpeg on the backend, and everything would work.
It would no longer matter what codecs people used, and we could get on with having a decent HTML5 spec.
 
Last edited by a moderator:
Well, the issue in the past was everybody wanted to push their pay/nag/ad-ware shitty players. It seems the philosophy for html5 video is that it'll work standalone with no extra downloads required. Unfortunately, because of the patent issues, this is still hard to come by. :/
 
Back
Top