Why H.264 Has Won And What It Will Take To Change

H.264 has become the dominant codec for video on the Internet, according to recent data. It has won the browser race with IE9, Safari, and Chrome throwing their weight behind it. The lone major browser hold out is Firefox. H.264 is a proprietary codec and Mozilla is putting their weight behind Theora, an open codec.

There are a lot of people who want to switch to an open codec. Beyond the fact that H.264 is proprietary, the free use for end users on the Internet is only set to run until December 31, 2015 (This could change). But, changing to something other than H.264 is not a simple task. It is not as easy as deciding on a new codec and then having software make updates. Let's take a look why.

Theora Is Not The Answer

When the discussion around codecs comes out one of the first rebuttals I hear is that everyone should switch to Theora. Problem is, this just isn't going to work. Putting aside all the other issues, Theora is an older codec. It is based on VP3. The most recent version of that (a proprietary codec itself) is VP8. Theora is older with newer and better technology.

I tend to think of Theora like I think of IE6. It's there, it works, in it's early years it was pretty good, but now it's an old timer with better options available. So, there is no good reason to use it.

Why H.264 Is So Dominant

H.264 is dominant well beyond the web. Every video recording device in my home uses H.264 and the playback devices, like my DVD player, use H.264 as well. It's everywhere. A mass number of devices all around us and professional video producers use H.264. The inroads are built.

Making A Change

Changing is going to be hard for two very important reasons.

  1. H.264 or the codec of use is often powered by hardware in devices. Hardware needs to be transitioned out of use rather than updated.
  2. The general public (the masses) want something that just works. They don't hold the ideals of using something open. A change in cost, making it more expensive, is about all that would get them riled up. So, a change would likely not be supported by the masses.

If a change from H.264 is going to happen it has to have more than a few loud voices behind it.

A Changed Hardware

Changing from H.264 is about hardware more than software. To make video work well on the Internet and be in so many devices H.264 has been built into the chips powering many devices. And, hardware is harder to switch from than software. Hardware goes out of use rather than being updated.

The transition to a different codec would have to happen slowly over 3 - 5 years to account for the hardware changes. This, also, means that the new codec and H.264 will likely need to have hardware support in new devices for at least this time period.

But, it doesn't stop there. H.264 already has lots of chip designs. It should be easy to incorporate into a device. A different codec would have a lot of up front development for the chips to encode and decode material. Companies need to be willing to suck this up or, in the spirit of openness, some open designs could help move this along.

This is why simply pushing people to use VP8 (if Google Open Sources it) is not a fast solution to change to.

Managed Change

Just because hardware is available doesn't mean a change can happen. Just look at what happened with Betamax. The change needs to be managed including a healthy transition from H.264 to something else and the change needs to be sold to just about everyone.

Through a managed change that will take years the supporters will have to have patience for years. This is going to be a tough egg to crack. The toughest change may be to remain patient through a long hardware change.