Re: [MiNT] Proposed modification to k_init.c (XaAES)

[Paul Wratt <paul.wratt@gmail.com>]

On Wed, May 25, 2011 at 11:27 PM, Peter Persson <pep.fishmoose@gmail.com> wrote:
>
> Preferably it should be external to xaaes.km, somehow. But the modecode
> monkey business is what we have, today.
> In our case, we chose an approach where we didn't modify the existing
> resolutions at all. Setting any standard Falcon resolution yields the exact
> same result on the SuperVidel. However, setting bit 14 allows it to be
> modified, while still keeping the meaning of the original bits.
> This approach was chosen because it's imo the best compromise between
> original falcon behavior (which is intact), and some degree of compatibility
> with the CTPCI (which has a special meaning w. bit 14 too).
> Wether this should be put in all API functions (VsetScreen(), VsetMode(),
> Validmode(), VgetSize()... vq_opnwk()) - or - just handled when opening the
> physical workstation, I don't know.
> -- PeP
>
Is there a way to do the video card/driver check in a more generic
way, even if fvdi has to be changed in order for it to be compatible.

It would be nice to use the same mode setting with fvdi to allow
changing screen size on ARAnyM too, which already restricts fvdi to a
max mode.. even though some parts can still be overridden (color
depth) by fvdi settings

I am looking for some uniformity here

Paul