From: Hartmut Niemann <niemann@cip.e-technik.uni-erlangen.de>
  To  : ggi-develop@eskimo.com
  Date: Tue, 7 Jul 1998 15:55:55 +0200 (MESZ)

Re: Changing the kernel makefile system...

> 
> Steffen Seeger wrote:
> > > Ok, pretty much like I thought.
> > > In other words we just need to make the multisync driver support
> > > more than one range of h/v timings and we have a unified monitor
> > > driver (Timelist would have min == max for every range, and
> > > monosync would have min=max for v ranges)
> > >
> > > //Marcus
> > 
> > Not really.
> > 
> > Timelist has fixed timings and fixed dot clock (and thus
> > fixed resolutions).
> > 
> > Monosync has in horizontal directions fixed timings, but no fixed
> > dot clock. But it does have fixed clock (the hsync frequency) and
> > timing in the vertical direction.
> > 
> > Multisync has neither fixed timings nor fixed clock rates in
> > either direction, but the ratios should somehow fit.
> 
> Well, a fixed timing is nothing but a non-fixed timing
> with an infinitely small range...
> But if we don't win anything by generalising the monitor
> stuff into a single scheme we can of course just do 
> switch (display->monitor.type) {
> 	case monosync:
> 	...
> 	case multisync:
> 	...
> 	case timelist:
> 	...
> }
> in the monitor-driver.
> 
> //Marcus
> 
> 
>From a practical standpoint, you can't make the range infinetely small, so you 
get a finitely small range. Think a pll driver that could do
28.004 MHz and 27.996, but not 28.000.

The only thing (afaik) the multisync driver would lack is support for
more than one frequency range.
Other than that a monosync mode could be dotclock=27.9MHz..28.1MHz, horizontal
31,9 kHz..32.1 kHz, vertical 59..61.
You can't make them sharp 28MHz, 32kHz, 60Hz anyway ...

Hartmut
--  
Hartmut Niemann   --   niemann(a)cip.e-technik.uni-erlangen.de
http://cip2.e-technik.uni-erlangen.de:8080/hyplan/niemann/index_en.html [/ggi]