On Wed 28 Dec 2016 at 13:18:35 -0500, Mouse wrote: > >>> Is NetBSD going to play with Wayland? 'Cause X.org seems to be in > >>> a bit shaky and captured by Linux-droids. > > Conceptually, KMS strikes me as the right way to do things. I don't > know enough about the current design/implementation to comment on it. Except that it seems that in the Linux way of doing things, you need support for your graphics card both in the kernel AND in X. Which definitely seems wrong to me. The things in Wayland I strongly object to are that all rendering is done client-side (it ships bitmaps to the display server), and that the network transparency is considered useless. I use remote X clients "all the time". Which I find very useful; much more useful than a "remote desktop". So now the applications have to render everything themselves into some local bitmap. Which is slow, of course, if you do it with the CPU rather than the GPU. So now multiple applications need to access the GPU for their acceleration. And the display server too. Which gives you annoying resource sharing problems. Usually they are best solved with a client/server architecture (and modern written-from-scratch GUIs like BeOS did that too, so that idea isn't suddenly discredited). But... X is already a client/server model, so replacing it with another one is nonsensical. I would have no issues with an X server that runs completely on the GPU. That only requires inventing a transport for the requests and responses, but that is proably trivial. As I understand, GPUs run using their own operating systems already anyway, so adding an application to that should be no problem. -Olaf. -- ___ Olaf 'Rhialto' Seibert -- Wayland: Those who don't understand X \X/ rhialto/at/xs4all.nl -- are condemned to reinvent it. Poorly.
Attachment:
signature.asc
Description: PGP signature