Re: GSK review and ideas


On 16 December 2016 at 15:08, Owen Taylor <otaylor redhat com> wrote:

First of all, generally you have to supply entire buffers that have
valid content everywhere. If you're lucky you can use extensions like
buffer age so that you can track which part of the back buffer is up-
to-date, but that requires double or tripple buffering, which itself
brings up the memory use and possibly the the power use.

I don't think we need to count on luck to have the buffer age extension
- it is widely available on Xorg and Wayland and we have the ability to
fix cases where it is missing.

That's, sadly, not true when you start moving on embedded hardware,
especially on the low end of the price point spectrum — which is where
switching to GL for rendering would benefit us.

P.S. - when thinking about power consumption, not only does clipping
allow actually not processing pixels, if you combine it with culling,
you greatly reduce the number of objects you are walking over and the
amount of setup that is being sent to the rendering API.

I agree: culling at the scene graph level — i.e. before reaching the
render tree — is probably where we're going to win the most, because
drawing widgets — especially with CSS — is still the expensive bit,
compared to submitting pixels to the screen.


[@] ebassi []

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]