WebM vs H264?

[CarlJ]

The pro-WebM camp keeps pointing out, over and over, that IE will support WebM for the <video> tag because they'll support any video codec that's installed on the system. It's a shame that Mozilla and Google won't take this same position.

The problem here is not whether or not WebM is a good thing, it's that Google has announced that they're going to go out of their way to block access to H.264 in Chrome (to drum up support for WebM). They're actually going to spend extra time ripping out existing working code, rather than just leaving it untouched.

Why not just have everyone (Microsoft/Mozilla/Google/Apple/Opera) agree to take the IE approach, make use of any codecs already on the system and present those in their existing HTTP request headers as types they can support "natively", thus, "video/h264, video/ogg, video/webm", etc. Then sites that provide video (whether one or thousands) can send back an appropriate response that uses the <video> tag to offer one of the supported formats, or failing a common format between the site and browser, falls back to (H.264 or WebM) over Flash. If IE can do it, why not everyone else? Note that I'm not saying that just Apple should do this -- Mozilla, Google, and Opera must do it too. If you support Microsoft doing this, surely you would support Mozilla and Google doing it too, right?

For the record, let me state that, I do not like software patents at all, but the reality is that the highly effective and efficient codec that is H.264 could not exist in the current software (and patent) landscape without the MPEG-LA or something like it, because otherwise the companies involved would be spending all their time suing each other over patent violations. The MPEG-LA website currently lists 29 companies with patents in the AVC/H.264 pool, and I don't expect that Apple is the top dog there, just one of the many players. The existence of the MPEG-LA is like Mutually Assured Destruction for all these companies, "I won't sue you in exchange for you not suing me, and we'll charge each other and everyone else a nominal fee to use our pool of patents." Without this, each company would only have it's little piece of the patent pie, and nobody would want to license to anyone else (or only for exorbitant fees).

WebM likely either already violates patents in the pool, and if not, it is even more likely to violate the pool when Google starts trying to improve WebM's performance. This is not FUD, this is an observation that is very likely to be correct. Shouting it down does not make it less true.

If WebM performed at least as well as H.264, and was proven to not be in violation of existing patents (Google could go a long way in this area by agreeing to indemnify users of the code against any patents it violates; they keep sidestepping this), and it was supported by current hardware decoding, I'd be 100% in favor of it, rather than H.264. I prefer free (and Free) software. But I prefer software that actually works, even more. And "works" includes not draining the batteries on the current crop of devices (not some imagined future devices) in the burgeoning mobile field.

The pro-WebM camp keeps saying, "MPEG-LA has been spreading FUD about VP3 [/VP8/etc.] for years, but they haven't sued yet, so they won't sue over WebM". The answer to this is, until now (Google), there hasn't been anybody worth suing. It isn't worth their time (and money) to take a few hundred users to court ... but Google is a competitor with deep pockets who is looking to run roughshod over their patents. It's like the hackintosh scene -- Apple can't get too worked up over a few thousand people building their own clones from scratch and (in violation of the license) loading Mac OS X on the systems... but when companies start trying to do the same thing on a large scale, they get taken down. If WebM comes into really wide use, expect MPEG-LA to (quite reasonably) come knocking on Google's door.

 

[Rodimus Prime]

The pro-WebM camp keeps pointing out, over and over, that IE will support WebM for the <video> tag because they'll support any video codec that's installed on the system. It's a shame that Mozilla and Google won't take this same position.

here is kicker. Mozilla and Google can not access system codec like IE can. That is kind of an IE only system.

problem with h.264 being the default is the cost of entry for new browsers becomes impossibly high at 5 mil+ per year. Now that is chump change for Apple, MS and Google but is a huge chunk of money for Mozilla, Opera and any other new start up. That makes it impossible for new browsers to enter the market.

I would also argue that h.264 violates web standard in html in the since that html is supposed to be 100% free to implement. Now until the video part that was a none issue there is nothing in HTML that not open source. With HTML5 video and Apple and MS trying to force h.264 to keep others out of the market.

What is best for HTML5 is not have a 100% free video standard and h.264 is not free. It is very costly.

 

[KnightWRX]

here is kicker. Mozilla and Google can not access system codec like IE can. That is kind of an IE only system.

Actually, that's false. They can. The problem with that is obviously that Chrome and Firefox are portable applications that work on many different OSes and using the system provided video/audio stacks would result in all these portable applications to basically support multiple software stacks that essentially do the same thing. Whereas if they roll their own (they don't, both use FFmpeg if I'm not mistaken, a very portable and efficient video decoding library that supports multiple codecs) they only need to support 1 stack making porting easier and evening out the experience on multiple platforms.

And yes, there is a big difference between an Internet Explorer plug-in and a system codec. If you don't understand it, I'm sorry you need to read up on both before we continue any kind of debate.

 

[Rodimus Prime]

Actually, that's false. They can. The problem with that is obviously that Chrome and Firefox are portable applications that work on many different OSes and using the system provided video/audio stacks would result in all these portable applications to basically support multiple software stacks that essentially do the same thing. Whereas if they roll their own (they don't, both use FFmpeg if I'm not mistaken, a very portable and efficient video decoding library that supports multiple codecs) they only need to support 1 stack making porting easier and evening out the experience on multiple platforms.

And yes, there is a big difference between an Internet Explorer plug-in and a system codec. If you don't understand it, I'm sorry you need to read up on both before we continue any kind of debate.

did not know that but that does make sense because if they wanted to do that it would require a very big difference between Windows FF and other FF if I am not mistaken in the coding of it. Instead of minor changes to jump between OS it would require a much larger changed.\

Also makes sense I have seen Windows Media player get access to I know MPEG-2 after I had some media player that supported it installed on my computer.

 

[Sedulous]

It is entirely possible that the MPEG-LA consortium could undermine WebM by allowing free use in browsers. Competition is good and may force their hand to either compete or become irrelevant.

 

[Xian Zhu Xuande]

It is entirely possible that the MPEG-LA consortium could undermine WebM by allowing free use in browsers. Competition is good and may force their hand to either compete or become irrelevant.

That would be nice. There will probably be lawsuits instead, though...

 

[KnightWRX]

It is entirely possible that the MPEG-LA consortium could undermine WebM by allowing free use in browsers. Competition is good and may force their hand to either compete or become irrelevant.

The problem is not just use in browsers, it's uses on Websites. If you have a business of streaming video, you need to pay the MPEG-LA to use H.264. If you use WebM, you just need to find an encoder for it.

Someone, somewhere needs to pay for H.264 use and those people thus can't provide services that are Free and Open, they will need to be a corporate outfit that makes profits. That is the big loss of going with such a patent encumbered solution.

 

[Sedulous]

The problem is not just use in browsers, it's uses on Websites. If you have a business of streaming video, you need to pay the MPEG-LA to use H.264. If you use WebM, you just need to find an encoder for it.

Someone, somewhere needs to pay for H.264 use and those people thus can't provide services that are Free and Open, they will need to be a corporate outfit that makes profits. That is the big loss of going with such a patent encumbered solution.

Likely that media companies are not going to opt to use Google's fly-by-night undeveloped WebM over h264 to produce their product... many providers already use h264, have been using it, and likely will continue to use it.

 

[FX120]

Likely that media companies are not going to opt to use Google's fly-by-night undeveloped WebM over h264 to produce their product... many providers already use h264, have been using it, and likely will continue to use it.

Unless the majority of the public doesn't use a browser that supports the viewing of it...

 

[Rodimus Prime]

Unless the majority of the public doesn't use a browser that supports the viewing of it...

and boom there you have it. 95% will be able to use WebM compared to less than 65% that will not be able to use h.264. And I mean a far amount less than 65% will not use h.264 since the people who tend to stream tend to be more technological savy which means they tend to go away from IE and Safari going right to FF and Chrome.

I can not stand IE but I hate safari a heck of a lot more than IE. If a sight forces me to use one of those 2 browsers I will not touch them for movie streaming.

 

[Sedulous]

The problem is that right now, nothing uses WebM. h.264 momentum, production and consumption, and user expectation has been towards h.264. Regardless of how much you may want it otherwise, this is how it is. Deal with it.

As I said before, if WebM gets any traction, the MPEG LA consortium could easily pull out the rug by agreeing to free h.264 licensing for an expanded set of uses.

 

[Rodimus Prime]

The problem is that right now, nothing uses WebM. h.264 momentum, production and consumption, and user expectation has been towards h.264. Regardless of how much you may want it otherwise, this is how it is. Deal with it.

As I said before, if WebM gets any traction, the MPEG LA consortium could easily pull out the rug by agreeing to free h.264 licensing for an expanded set of uses.

h.264 has some tracktion from some people and of course Apple Fanboys because their god said so.
But fact remains that 35% of the market can not and will not be able to use h.264 in HTML5 end of story. WebM only have 5% of the market that will not be able to play it.

35% vs 5% take your pick.

 

[dolphin842]

If you have a business of streaming video, you need to pay the MPEG-LA to use H.264.

This statement is incorrect. Here are the costs associated with h264 for internet content companies (as reported by Ed Bott the other day):
- $0 for anything 12 minutes and under
- $0 for free (to end-user), ad-supported internet video
- $0.02 (at most) per title sold for video over 12 minutes
- $0 for pay-subscription-based video up to 250,000 subscribers (this is a sliding scale that goes up to $100,000 if you have more than a million subscribers)
- Possible fee increases of (at most) 10% once every five years

In other words, if your business model is based on ads, or clips under 12 minutes, or your service has fewer than 250,000 subscribers, then you're not paying MPEG-LA a dime. If you have the good fortune of running a successful a-la-carte download or pay-for-streaming service, your costs are minimal and commensurate with your success. Given that h264 is the best codec we have in the market at the moment, I would say this is a pretty appealing deal for content providers looking to publish their video in a format playable by the vast majority of the desktop and mobile browsing public (via html5 or Flash).

 

[Rodimus Prime]

This statement is incorrect. Here are the costs associated with h264 for internet content companies (as reported by Ed Bott the other day):
- $0 for anything 12 minutes and under
- $0 for free (to end-user), ad-supported internet video
- $0.02 (at most) per title sold for video over 12 minutes
- $0 for pay-subscription-based video up to 250,000 subscribers (this is a sliding scale that goes up to $100,000 if you have more than a million subscribers)
- Possible fee increases of (at most) 10% once every five years

In other words, if your business model is based on ads, or clips under 12 minutes, or your service has fewer than 250,000 subscribers, then you're not paying MPEG-LA a dime. If you have the good fortune of running a successful a-la-carte download or pay-for-streaming service, your costs are minimal and commensurate with your success. Given that h264 is the best codec we have in the market at the moment, I would say this is a pretty appealing deal for content providers looking to publish their video in a format playable by the vast majority of the desktop and mobile browsing public (via html5 or Flash).

and to encode or decode the video it become costly. For like FF it would be 5 mil a year + 10% per year increase.
Those numbers are pretty low over all btw. It does not take much to max out at 5mil per year.

Like I said 35% and rise percentage of the browsers will not support h.264. There is no getting around that fact.

 

[Sedulous]

and to encode or decode the video it become costly. For like FF it would be 5 mil a year + 10% per year increase.
Those numbers are pretty low over all btw. It does not take much to max out at 5mil per year.

Like I said 35% and rise percentage of the browsers will not support h.264. There is no getting around that fact.

Only time will tell. Other than chrome jumping off the h.264 wagon, what other developments make you so certain content providers are going to switch? That is who is going to decide what way things go. As the above post indicated, for most h.264 is very appealing and is well developed and supported by most devices. The groundwork for h.264 is already in place. It will take a lot more than YouTube to make the whole industry shift to a questionable codec.

 

[Rodimus Prime]

Only time will tell. Other than chrome jumping off the h.264 wagon, what other developments make you so certain content providers are going to switch? That is who is going to decide what way things go. As the above post indicated, for most h.264 is very appealing and is well developed and supported by most devices. The groundwork for h.264 is already in place. It will take a lot more than YouTube to make the whole industry shift to a questionable codec.

Firefox and Opera are not supporting h.264.

Opera is a minor player at best but Firefox is the big player. Firefox is what broke the IE strangle hold on the web. So between FF and Chrome that is a 35% of the browser market and that 35% is going to house a much larger chunk than 35% of the people who view video on the web.
Reason I know that the FF, Chrome house more than 35% of the web video is because people who watch video on line are going to be more tectonically savvy which goes hand in hand with people who use FF and Chrome or more techologylly savvy.

So tell me when well more than 35% of people who view videos on the web will not be able to watch them?

 

[KeithPratt]

Like I said 35% and rise percentage of the browsers will not support h.264. There is no getting around that fact.

Flash. Flash flash flash flash flash.

 

[*LTD*]

https://www.macrumors.com/2011/01/1...6-billion-profit-on-26-74-billion-in-revenue/

http://www.appleinsider.com/article...es_of_4_1m_macs_16_2m_iphones_7_3m_ipads.html

and . . .

http://www.appleinsider.com/article...evices_average_iphone_price_grows_to_625.html

As a content provider or developer interested in profitability, if your #1 priority is doing something other than supporting iOS devices, you're doing it wrong. It's just that simple.

 

[zap2]

Firefox and Opera are not supporting h.264.

Opera is a minor player at best but Firefox is the big player. Firefox is what broke the IE strangle hold on the web. So between FF and Chrome that is a 35% of the browser market and that 35% is going to house a much larger chunk than 35% of the people who view video on the web.
Reason I know that the FF, Chrome house more than 35% of the web video is because people who watch video on line are going to be more tectonically savvy which goes hand in hand with people who use FF and Chrome or more techologylly savvy.

So tell me when well more than 35% of people who view videos on the web will not be able to watch them?

Thats the web on PCs, I'm not convinced that is where we will see the future defined, because frankly while Flash isn't perfect there its well support and works well(enough). I'm betting mobile devices will play a large role in support the winning format. Does Android support H.264 still?(I assume it does)

 

[zioxide]

how many software and hardware players play this "WebM" ****?

like none?

how many software and hardware players play h.264?

all of them?


time to move along here...

 

[Xian Zhu Xuande]

how many software and hardware players play this "WebM" ****?

like none?

It's new, and we still don't quite have what we need. This isn't really a valid argument.
There are plenty of valid arguments, though.

An excellent read.

 

[Rodimus Prime]

Thats the web on PCs, I'm not convinced that is where we will see the future defined, because frankly while Flash isn't perfect there its well support and works well(enough). I'm betting mobile devices will play a large role in support the winning format. Does Android support H.264 still?(I assume it does)

While the os might I could easily see the browser in android going nope and refusing to support h.264 which would put android pushing webM for what webM is made for. Video on the web.

 

[ChazUK]

Wirelessly posted (Mozilla/5.0 (iPad; U; CPU OS 4_2_1 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8C148 Safari/6533.18.5)

Thats the web on PCs, I'm not convinced that is where we will see the future defined, because frankly while Flash isn't perfect there its well support and works well(enough). I'm betting mobile devices will play a large role in support the winning format. Does Android support H.264 still?(I assume it does)

While the os might I could easily see the browser in android going nope and refusing to support h.264 which would put android pushing webM for what webM is made for. Video on the web.

I would bet most third party manufacturers would add h.264 playback in custom Android builds and hardware. If Google make Android WebM only at the source level, there is nothing stopping oem's adding it and with so few Android handsets running "stock" android it would barely make any difference imo.

 

[Rodimus Prime]

Wirelessly posted (Mozilla/5.0 (iPad; U; CPU OS 4_2_1 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8C148 Safari/6533.18.5)



I would bet most third party manufacturers would add h.264 playback in custom Android builds and hardware. If Google make Android WebM only at the source level, there is nothing stopping oem's adding it and with so few Android handsets running "stock" android it would barely make any difference imo.

even with mobile devices those are still a small player in the terms of streaming video compared to computers. Limited bandwithe and screen size makes them a fairly small player.

No matter how you cut it going h.264 and not webM is going to cut off huge chunks of your market compared to very limited if you go with WebM.

Problem with h.264 is the cost of entry is what keeps a lot of new players from really getting into the market. 5 mil is chump changed to company like Apple and MS but for lets say Mozilla that is a huge chunk of cash.

 

[Sedulous]

Firefox and Opera are not supporting h.264.

Opera is a minor player at best but Firefox is the big player. Firefox is what broke the IE strangle hold on the web. So between FF and Chrome that is a 35% of the browser market and that 35% is going to house a much larger chunk than 35% of the people who view video on the web.
Reason I know that the FF, Chrome house more than 35% of the web video is because people who watch video on line are going to be more tectonically savvy which goes hand in hand with people who use FF and Chrome or more techologylly savvy.

So tell me when well more than 35% of people who view videos on the web will not be able to watch them?

These are very debatable assumptions. Lots of less than tech savvy people watch web videos. In fact, I would guess tech savvy people that work with computers abandon their computers when they aren't working with them.