Khronos group, OpenGL ES/miniglx/EGL?

Dave Airlie airlied at
Wed Apr 18 21:10:29 PDT 2007

> why do i feel im not welcome here?
> do you suggest that my questions are not appropriate here?
> i maybe have failed to read some 'rules', please advise?

> >"..also look at all previous attempts to get rid off X, i am confident we
> can say they all failed."
> im sorry, you dont seem to understand, its about *embedded* window system,
> there is no embedded X, even less OpenGL based one, so its clear - i want to
> produce, not replace or 'get rid off'

There is lots of embedded X, what do you think the N800, back to the
Digital IPAQ are running?.. there is also the twin project which is
being used to do a bootloader which is an embedded windowing system...

Miniglx and EGL are the two project that are trying to do a HW
accelerated open source GL stack with out X, we don't have driver for
many cards that support these, and there is a lot of other work to be
done around card setup before this can happen. Miniglx is considered a
hack and *embedded* people use it where they know what graphics chip
they want to use...

> how do i:
> - talk to this Linux community?
> - find if im not trying to do what already has been done?
> - gather people to work on a project?
> - find about project and join/help some project?

You are talking, listening you need to do more off, it hasn't been
done, people won't just work on your project because you ask people
will work on what they want to work on and if benefits you so be

> now, if you would rather people use EGL that has ".. been in limbo for a
> while"
> or
> use miniglx which ".. is a hack.  It filled a specific need when it was
> created.."
> than ok, but if you dont want to re-write what has already been
> written(possibly better) than im happy i could have helped by making it EASY
> for others to SEE, experiment and then decide if its worth using or not
> i can not SEE how EGL, miniglx, fbdev/dri works, it has not been made easy,
> and therefore i've come here to ask for help, which i, kind of regret, as
> people are so keen to jump and say how my project sucks and is not "real
> work"
> if someone did this 3 years ago, EVERYTHING could have been different, maybe
> quite a bit actually, just how many people "wasted" their time to "come up"
> with something that was long ago written and was only a mouse click away?
> i've learned this lesson long time ago - 1st find whats best on the net,
> then try to do it better, otherwise you would likely be wasting your time..

Scitech isn't the solution to anyone else's problem it may be the
solution to your problem in which case well done you've found a
solution but I don't think anyone here is going to put resources into
Scitech software which from the source I can read is sw rendered GL,
I've even downloaded it to find that out. EGL is something that will
probably be off use in the future but it will not run on all the GPUs
available each open source supported GPU driver needs to be ported to
the EGL system, and this requires major changes to the Linux kernel
and Mesa, until these changes are done you probably cannot use EGL for
your project, these changes are not being done to any timescale,
people work on them whenever they want and your project will not hurry
those ppl along..

Is that clear enough?


More information about the xorg mailing list