emacs-devel
[Top][All Lists]
Advanced

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

Re: Rethinking the design of xwidgets


From: Kai Ma
Subject: Re: Rethinking the design of xwidgets
Date: Sat, 17 Oct 2020 12:26:55 +0800

Arthur Miller <arthur.miller@live.com> writes:

> Cairo is a nice 2D api, and has opengl backend too. I dont' know how
> effective the backend is, but sure it could work. I dislike all the
> dependencies it pulls in, but in a typical GNU/Linux system they are
> anyway installed. Can you divorce it from Gtk requirement? Or it is
> already possible to use Cairo in Emacs without rest of Gtk?
>
> I proposed egl, as a platform independent surface drawing that enables
> use of OpenGL. Someone might one day create a scene graph (isnt't it
> just a list of lists) and maybe do some cool stuff in ELisp that we are
> not really thinking of now.

Maybe I'm diverting a little but I would like to add that if we have
OpenGL (or generally, hardware acceleration), we can enable more
"fashionable eye-candies".

I once worked on a shadowy text patch [1], which was discussed here some
months ago.  Though implementing the feature for Cocoa is exceptionally
easy, I couldn't get it fast enough on Linux systems with Cairo.  I
don't know if Cairo really has an OpenGL backend, but even if it has,
the blur API can't make use of it -- because Cairo doesn't even have
this API.

And yes, I do think Emacs for Linux deserves proper hardware
acceleration. (I don't really mind if it's OpenGL or something else, though.)

[1] https://github.com/ksqsf/emacsmoe/pull/1

Regards




reply via email to

[Prev in Thread] Current Thread [Next in Thread]