News for hackers. Stuff that matters.
Re:An update on GNU performance by ilaksh on Sunday October 21, 2018 @07:46PM (#18271048)
I thought that media codecs were handled outside of general purpose code. Like maybe a special chip or on a graphics processor.
Re:An update on GNU performance by Oowee1Ee on Monday October 22, 2018 @10:45AM (#18275038)
It depends on the hardware. There may be offloading for h.264, VP8, for different profiles and for encoding or decoding. But not all combinations are available on all machines. So software decoding may still be needed.

Now software decoders are often written with some hand-optimized assembly loops, but those optimized variants don't cover all possible target platforms and not every single loop is hand-optimized. So everything that's either a plain-C fallback or just hasn't been optimized can benefit from compiler improvements.

Re:An update on GNU performance by richardwhiuk on Sunday October 21, 2018 @09:39PM (#18271530)
Nope. They could be GPGPU accelerated, but if you aren't rendering to screen then it's not typically graphics processor territory.

Plus modern codecs are quite linearly data dependent which is a bad for for a GPU.

Re:An update on GNU performance by ilaksh on Sunday October 21, 2018 @09:47PM (#18271570)
But aren't movies rendered to the screen?

My understanding is most smartphones at least have hardware decoding for H264.

Re:An update on GNU performance by semi-extrinsic on Monday October 22, 2018 @12:59AM (#18272288)
Exactly, a GPU needs to have specific hardware (a dedicated ASIC) for H264 acceleration. It's not accelerated by the usual GPU hardware, which is sufficiently general-purpose that an accelerator running there would not be codec specific.
Re:An update on GNU performance by TazeTSchnitzel on Monday October 22, 2018 @05:15AM (#18273220)
Arm even sells such a solution, there is a line of Mali video processing units.