Re: Re: "Drivers? We don't need no stinking..."

> Alan Cox tactfully implies that a dearth of register-level documentation is
> moot in a dearth of mentally competent programmers to use it.

It isnt lack of competence. Its that people are not doing it. Some of it
is because X is seen as scary (which for 2D is no longer true) and some of
it Dave is dead right about - its money issues. Why do the intel chipsets
have nice support - its because in that area intel are good guys. 

For older chipsets its a real problem. The most economic approach to
handling a lot of old chipsets, eg the P9100 is clear. If it was a business
you'd ship all Weitek P9100 customers a complementary PCI radeon rather than
hire staff to fix it. In open source that really means its down to someone
being annoyed enough to fix it. 

Another way to think about it from the vendor point of view is if there
are 100 users of the card running base Red Hat, and every one of them 
would not buy the product if only vesa mode was available, it wouldn't
be worth doing.

> With the Wretched Offender's 'kdrive' as an example, perhaps it is possible
> to prune the code.  As for chips, maybe I should replace my aging G400 and
> 300 MHz K6-2 with a 2 GHz CPU or two, and run dumb framebuffer mode; that was
> Intel's suggestion, several years ago.

AGP and CPU's get faster and faster. Todays CPU's run games designed for
SGI hardware rendering far too fast to play - in software. XFree86's 
unaccelerated 2D engine is very portable but not hand tuned mmx. Where 
people are using hand tuned software code (eg directfb) its fast without
the card. Not as fast as with it I grant.

Alan