One of the advantages of this driver is that with 2.2+ kernels, you
can run fullscreen applications from the Linux console without root
permissions, on any Linux platform (only tested on x86)
I have always been a little confused as to all different types of frame
buffer access in Linux. You have X11, DGI, GGI, fbcon, etc. I thought that
even in X11 you had direct framebuffer access if you wanted. Does using
fbcon negate using any X11 setup/calls? What would using fbcon buy you
(speed? resolution/pixel depth changes?)? Does all the talk about the DRI
(Direct Rendering Infrastructure by Precision Insight and SGI I think)
apply to this, or to only OpenGL/3D rendering?
Could someone elaborate a little?
Thanks,
Kelly
(Game prog. hacker who is only used to Win32/DX programming, but am now
moving to Linux as a dev. platform)>
One of the advantages of this driver is that with 2.2+ kernels, you
can run fullscreen applications from the Linux console without root
permissions, on any Linux platform (only tested on x86)
I have always been a little confused as to all different types of frame
buffer access in Linux. You have X11, DGI, GGI, fbcon, etc. I thought that
even in X11 you had direct framebuffer access if you wanted. Does using
fbcon negate using any X11 setup/calls? What would using fbcon buy you
(speed? resolution/pixel depth changes?)? Does all the talk about the DRI
(Direct Rendering Infrastructure by Precision Insight and SGI I think)
apply to this, or to only OpenGL/3D rendering?
Could someone elaborate a little?
I’m only going to post my opinion, and I’m sure others have more to add…
I think the main reason for so many different fullscreen APIs is that
none of them (yet) have a good API and lots of hardware support.
Eventually it will settle down to one or two mainstream APIs. My vote is
some marriage of fbcon with the GGI hardware acceleration architecture.
All of the APIs are young, and still developing.
fbcon buys you fullscreen access at the Linux console without root access,
and without X11 overhead. It pretty much as close to the hardware as you
can get, but there is no hardware acceleration available yet.
The DRI infrastructure has nothing to do with the fullscreen APIs, it
is a way for OpenGL to talk to the hardware (input mostly I think) with
much less X transaction overhead.
-Sam Lantinga (slouken at devolution.com)
Lead Programmer, Loki Entertainment Software–
“Any sufficiently advanced bug is indistinguishable from a feature”
– Rich Kulawiec