2017-12-29 6:48 GMT+01:00 Luke Kenneth Casson Leighton lkcl@lkcl.net:
On Thu, Dec 28, 2017 at 9:56 PM, Bill Kontos vkontogpls@gmail.com wrote:
If this goes well it will go down in history as one of those famous conversations in the early days of tech we read about.
yehyeh!
I think so too. And India is very keen on becoming the next China/Taiwan/Etc. Which I think is a good thing.
This is something that I don't get with the shakti project. How are they planning to tackle the 2d/3d/vpu problem?
also on opencores there's a series of hard macros with basic video primitives, including cabac decode and many more: dropping lots and lots of those in will go a long *long* way towards being able to tackle 2D, 3D and a VPU.
the key here though is: it is *really* necessary to find a full software team to get the userspace stuff done *at the same time* and on emulated FPGAs so that performance can be verified / estimated.
2D: Skip. AMD and Vivante already do so, NVIDIA will too IIRC. The 2D accelerators were mostly for windowing systems now replaced by composting systems, including MS Windows, and other means, Androids SurfaceFlinger, etc. The missing functions are now done on 3D or CPU.
VPU: That would require also licenses from the format owners. That's going to be difficult. And the new, open, video formats are not ready. Daala, Thor, NETVC, AV1.
Something generic to offload parts of the decoding/encoding would be the best bet I guess. Avoids licenses, single format isolation. IIRC most codecs share techniques. I might be talking jibberisch, or it might be to impractical.
3D. Wasn't there a PoC from some students in the open macro's? Perhaps those guys can be hired to refine their work?