jmt [Thu, 15 Oct 2009 21:30:10 +0000 (21:30 +0000)]
Further GPS and route manager behavioural fixes
* When the nav-radio is slaved, calculated radial/target-hdg-deg
(needed by some autopilot logic)
* Handle editing (including deletion) of route waypoints correctly,
including deleting the active waypoint
* Add a signal to the route manager when the last wpt is reached, and
use it in the GPS to revert to OBS mode.
* Change the altitude handling to use the specified cruise altitude
* Fix a bug where autopilot/locks/altitude was treated as a boolean
jmt [Wed, 14 Oct 2009 06:50:06 +0000 (06:50 +0000)]
Route-manager: remove any interaction with the autopilot, or internal sequencing; assume the GPS always handles this (it should, now). Also fix waypoint altitude handling, which was broken. If not altitude is supplied for a waypoint, default to the requested cruise altitude.
jmt [Sun, 11 Oct 2009 21:29:24 +0000 (21:29 +0000)]
Fix unrealistic-but-convenient direct-drive from the route-manager to the autopilot code. A new boolean config prop, 'drive-autopilot' exists, to disable this behaviour.
jmt [Mon, 5 Oct 2009 21:09:20 +0000 (21:09 +0000)]
Land the GPS/route-manager re-write. Many things are better, many other things will be better, some things are no doubt broken. Please be patient and report problems on the mailing list.
Ron Jensen:
While troubleshooting the new environment I discovered the metar station
elevation (station_elevation_ft) is never being set because there is a
local variable that overrides it.
Tatsuhiro Nishioka:
Patches for configure.ac and Makefile.am files in FG/SG so Mac developers can build these in a unix way.
These also enables Mac developers to choose either PLIB framework or PLIB static libs.
Different approach to handling the paired-ILS-on-opposite-runways situation. This uses additional information we now store (in FGRunway) to be more explicit about the behaviour (i.e only look at paired navids), and crucially, does not misbehave near either threshold. Instead it always switches about the runway midpoint (so back-course/missed-approaches are broken as before), leading to stable behaviour when touching down.
I am pretty confident this is (was, from FS2004) also the 'solution' taken by MSFS.
Tim Moore [Sat, 12 Sep 2009 13:50:06 +0000 (15:50 +0200)]
Merge branch 'topic/atmos-merge' into next
John Denker's atmosphere changes. Original commit message:
Two-parameter physics-based model of atmosphere up to 262,467 ft i.e.
the top of the mesosphere. Correctly exhibits the HALT phenomenon.
Various LOC/GS improvements from John Denker, adapted to trunk by me (hopefully not introducing too many bugs along the way). Includes:
- false LOC courses and GS lobes
- LOC sensitivity based on runway dimensions
- GS cutoff based on range
- More accurate GS deviation computation, making final approach more stable
Handle "double" type fields correctly for ascii input data. Previously double
data was being down-cast to float and the loss of precision impacted data types like lon/lat positions dramatically, preventing smooth replay of data.
John Denker's atmosphere changes. Original commit message:
Two-parameter physics-based model of atmosphere up to 262,467 ft i.e.
the top of the mesosphere. Correctly exhibits the HALT phenomenon.
test for the presence of alut.h. if it's not found only display a warning message in gl-info and test-env-map. A clean build without glut installed is now possible.
Further re-factoring of nav-radio code, again no functionality changed should occur. Radio reception / GPS slaving / Glideslope handling and CDI updating are all separate functions now.
Tatsuhiro Nishioka:
FG crashes at exit at very high likelihood.
The cause of the crash is that some subsystems (input and gui) call get_subsystems() at their destructor.
This is very dangerous since SGSubSystemMgr::get_subsystem() can refer to already deleted subsystems.
Tatsuhiro Nishioka:
Fixed: wrong event name for abs-hat0-y
Modified: let AxisElement to generate normalized input (-1.0 to 1.0).
This can be temporal and can be removed when AxisEvent normalizes its value.
Modified: clean up code
Added: some comments so other Mac developers can see what's going on
ehofman [Mon, 31 Aug 2009 07:06:26 +0000 (07:06 +0000)]
Bertrand Coconnier:
Additionally to the original fix (problem with the Tec2b matrix initialization) I have modified the code for the "STRUCTURE" contacts in order not to generate NaNs when the z direction of the body frame is normal to the ground normal. Now there should no longer be any NaN generated by the landing gear code.
durk [Sun, 30 Aug 2009 14:39:04 +0000 (14:39 +0000)]
Two minor (codewise) fixes to ensure everything works as advertised:
- Make sure to initialize the groundnetwork after succesful loading.
- Ensure that the AI groundnetwork code proximity detection code used the
correct values for bearing. (Note that detecting the proximity of the
user aircraft appears still to be broken. I'll look into that later).