Re: [Flightgear-devel] Why CockpitDisplayManager vs. FGInstrumentMgr?

[James T. <ja...@fl...>]

> On 28 May 2018, at 12:46, Edward d'Auvergne <tru...@gm...> wrote:
> 
> That makes sense.  Then I think I'll shift the KLN89 GPS subsystem
> completely into the CockpitDisplayManager.  That subsystem probably
> needs its OSG support completely redone as well.  The Draw() function
> via DCLGPS::draw() is never called!  I guess this is why the
> instrument broke with the 1.0 FlightGear release.  One way to fix this
> might be something similar to wxRadarBg::init(), with a FGODGauge,
> osg::Geode, and osg::Geometry set up.  That would then allow the
> instrument to work with the 3D model in the Carreidas.  Well, assuming
> there aren't any other bugs that have crept in due to bit-rot.

Honestly in a world that contains Canvas, I think the KLN89 C++ code is just dead weight, doing a similar unit (eg, the KLN94) in Canvas would be much simpler than what people already did for the Garmin 480 or G1000.

Kind regards,
James