Re: [Flightgear-devel] Properly rendering terrain data from FlightGear using OpenGL

[Samuel C. <sam...@gm...>]

On 04/09/20 20:25, Sascha Reißner wrote:
> Am Freitag, den 04.09.2020, 18:54 +0200 schrieb Samuel CUELLA:
>> On Thu, 03 Sep 2020 19:40:37 +0200
>> Sascha Reißner <rei...@no...> wrote:
>>> I dont know if it calculate also the local orientation, but thats
>>> not
>>> difficult to do. I remember some matrix calculation in the code, so
>>> i
>>> have played with rotated coordinate systems.
>>>
>> I couldn't find it yet. I used a PDF document that explained the math
>> to go from a cartesian ECEF to local NED (North East Down) vectors
>> which gave me some results but not perfect.
> 
> The code need it to calculate the correct orientation of the approach
> light system if you change or build a new one.
> I think its the function 'projection' in vertex.c on line 231.
> It takes the bounding sphere as local center and calculate the X/Y/Z
> location of the vertex based on the local center and local orientation.
> This will be stored in the vertex.projection.[x|y|z].
> So vertex.projection.[x|y|z] is a localized coordinate where X and Y
> build a horizontal plane (Y point to north) and Z is straight up.

Hi list,

A big thank you for pointing me in the right direction. For the record, 
I had three issues:

-Incorrect conversions between coordinate systems, solution was to use 
logic from View::recalcLookFrom.
-Incorrect geometry: I had hiccups in my mesh-building code
-The math lib I'm using had only float matrices/quats. I added a bare
minimum support for doubles where I needed it, and voilà !

While doing the testing, I ended up writing a FGTape reader to replace
the FG instance I had running on another computer. I'll do a separate 
post for whoever could be interested in it.

Best,
Samuel