From: teunis <teunis@computersupportcentre.com>
  To  : ggi-develop@eskimo.com
  Date: Thu, 6 Aug 1998 14:37:42 -0700 (MST)

Re: Multiple boards.

On Sun, 2 Aug 1998, Jon M. Taylor wrote:

> On Sun, 2 Aug 1998, Jan Kneschke wrote:
> 
> > 
> > On Sun, 2 Aug 1998, Brian Julin wrote:
> > 
> > > On Sun, 2 Aug 1998, Wolf Shaman wrote:
> > > 
> > > > Should a single driver handle all detected instances of that board?
> > > 
> > > This would be more efficient yes.  Would be nice too if the S3 driver
> > > could be made into a core that can have code linked in from the more
> > > primitive S3 based chipsets and handle them all together.
> > 
> > do you think about a core-driver for trio+vision+virge+(savage) ?? 
> > or one core driver for every family ?? 
> 
> 	I'd say one core driver for all S3 cards.

The further you go back in time the more differences present.  IMHO I'd
prefer by family rather than one monolithic coordinator.  The S3-805 is a
-LOT- different than the S3-ViRGE.  And even the ViRGE is a lot different
than the Trio64V+ line when you factor acceleration in...

> > virge and trio are quite
> > similar. 
> 
> 	They are virtually identical.

Virge supports almost everything Trio64V+ supports.  Reverse is not true.
But accelerator-control (timing) is quite different IIRC...  Since that's
technically a driver-ONLY issue, that's not something the core driver
should care about.  Other than that, yeah, a ViRGE is a Trio with some new
toys (and different timing).

BTW - I'm interested in accel support but have not entirely figured out
how to interface it properly...  Has this been decided yet? :)

G'day, eh? :)
	- Teunis