VP9 and Opus, Background Position Offset and Ruby Positioning

Published on in Google Chrome, Last Week, tech, WebKit. Version: Chrome 25

This update discusses the 1,522 WebKit and 2,131 Chromium changes that landed in the past two weeks, totaling up to 3,653 changes.

Chromium updated the libvpx library to include support for an early version of the VP9 decoder, the successor of the VP8 codec that’s currently used by WebM. Furthermore, though still behind a flag (also available in about:flags), WebM files containing audio streams using the Opus codec are now supported as well.

The Toolbar Icons in Web Inspector won’t be shown by default anymore. The gear icon immediately goes to the Settings dialog again. Newly added items to the Network Panel will now be applied to filtering and cookies can be sorted by their size again. The viewport-related warning messages have been cleaned up, a new warning has been added for scripts blocked by @sandbox and messages created by console.time and timeEnd() can now be filtered out.

Support has been added for the “widows”, “orphans” CSS properties, as well as for the “ruby-position” property. The text-orientation property now supports the “sideways-right” value and text decorations will now work correctly when text-combine is being used as well. Furthermore, out-of-range values will now be clamped to values the supported range. Applying SVG filters to elements through the url() syntax will now work even when the filter isn’t available yet, and CSS Exclusions’ shape-inside is now supported for multiple-segment polygons.

Elliot moved generated content in :before and :after CSS pseudo-elements to the DOM, which also means that supporting animations and transitions on them is close. The Shadow DOM’s ShadowRoot constructor is gone, XMLHttpRequests for blob data will now set their Content-Type header accordingly and status events will now be fired for <link rel=prerender> elements. The Web Audio API now supports an offline audio context and the ::cue pseudo-element for video elements now is supported as well.

Support for in-band text tracks is now available on the Mac port, and plumbing has been added to allow deferring displaying of text tracks to the embedder.

Following the final pieces of it’s implementation, Alexis enabled support for the new background-position offset syntax on several platforms before enabling it by default, and removing the feature flag altogether. Meanwhile Tab started on aligning WebKit’s gradient implementation with the specification, towards unprefixing.

As for more experimental features, work on implementing CSS Grid Layout continues with improved parsing and layouting capabilities. Rick started working on supporting -webkit-image-set() on cursors, too.

Other changes which occurred these weeks:

This will be the last update of 2012. Thank you for reading, have a Merry Christmas and a Happy New Year!!

10 Responses to “VP9 and Opus, Background Position Offset and Ruby Positioning”

Responses are currently closed, but you can trackback from your own site.


December 18, 2012 at 9:23 pm

How much better is VP9 supposed to be compared to VP8? Will it be better than h.265? Because I think it has to. If the world was already in the middle of adopting h.264 before VP8 came out, and it couldn’t “roll back” that adoption and wait for VP8 to be fully ready, then VP9 could at least have a chance when the world has to decide whether it moves to h.265 or VP9. And it would really help if VP9 was significantly better.

Also, if VP9 is to have a chance, Google needs to promote it a lot better, because some techies already know how much better h.265 is than h.264, and the news is going to spread, and then explode, thanks to their word of mouth, when h.265 will be officially announced.

Google needs to start planting the seed of VP9. Mentioning that WebM also has the benefit of supporting Opus would be of great help, too.


December 29, 2012 at 10:26 pm

I am not sure if this got thru but if you are interested in Fractal technology and want to see how it works with a video that can be compressed or then to screen size resolution size independent Look at a company TMMI They bought the video code rights from iterated and sold it to TMMI even as Alan Sloan advised against it. Sloan is now on the board of directors. They have a working coded and you will see it very soon so if you are a techie type person all i can say is seeing is believing There eniter history can be read at trudef blog Enjoy the read and if you truly understand that with there working codec and today’s computer processing speeds have made this a perfect storm for what is about to happen in 2013. I look forward to hear your thoughts Thomas


March 7, 2013 at 8:37 pm

hi, is it planned to have a transparent mode? eg.as like flash has? this is the one reason for us to stay on flash, because no other codec has alphatransparent pixels!

[...] technologies like Google’s VP9 worthy of mention, particularly since Google added VP9 decode to beta versions of Chrome in December, 2012, along with a new decoder for audio streams encoded with the Opus codec. [...]