Imagination introduces VP9 acceleration for PowerVR GPUs

All aboard Google’s VP9 train; Choo choo…

Feb 26, 2014 by

Imagination logoAt Mobile World Congress Imagination announced that all PowerVR series6 GPUs would support its new software VP9 accelerator. This VP9 accelerator leverages the OpenCL supported GPU compute features of the PowerVR series6 architecture to decode YouTube videos for smooth 1080P playback. Developed by MulticoreWare in partnership with Google this VP9 accelerator will boost YouTube playback performance on chips like Intel’s Merrifield and Moorefield, and a host of other chips from MediaTek, LG, and Allwinner which all use PowerVR series6 GPUs.

Imagination posted a demo of the new accelerator in action.

[youtube_sc url=”http://www.youtube.com/watch?v=bz294jk2awY” theme=”light” autohide=”1″]

Outside of recognizing that indeed that video stream on that tablet is playing back smoothly there isn’t a whole lot to see in the video. That’s said it looks like the technology works.

It’s worth noting too that ARM announce a similar partnership with Ittiam for VP9 acceleration on Mali GPUs. Considering that this is a software solution from MulticoreWare that use OpenCL-based GPU compute there should be no doubt that we will see this VP9 acceleration solution rolled out in the coming months by other, if not all of the other, hardware vendors with GPUs that support OpenCL.

VP9 is still a relatively new open and royalty free video codec that Google is developing. Considering the percent of internet traffic that is directly attributable to YouTube traffic hardware vendors will start rolling out their own fixed-function VP9 acceleration blocks within the next two years. Of course for a video codec standard as new as VP9 its surprising to see that their are already software acceleration solutions available to consumers like this one from Imagination.S|A

The following two tabs change content below.
My Twitter profileMy Google+ profile

Thomas Ryan

Author at SemiAccurate
Thomas Ryan is a freelance technology writer and photographer from Seattle, living in Austin. You can also find his work on SemiAccurate and PCWorld. He has a BA in Geography from the University of Washington with a minor in Urban Design and Planning and specializes in geospatial data science. If you have a hardware performance question or an interesting data set Thomas has you covered.