1
0
Fork 0
flightgear/src/Main/main.cxx

963 lines
33 KiB
C++
Raw Normal View History

// main.cxx -- top level sim routines
1998-04-21 17:02:27 +00:00
//
2002-04-05 00:38:55 +00:00
// Written by Curtis Olson, started May 1997.
1998-04-21 17:02:27 +00:00
//
// Copyright (C) 1997 - 2002 Curtis L. Olson - http://www.flightgear.org/~curt
1998-04-21 17:02:27 +00:00
//
// This program is free software; you can redistribute it and/or
// modify it under the terms of the GNU General Public License as
// published by the Free Software Foundation; either version 2 of the
// License, or (at your option) any later version.
//
// This program is distributed in the hope that it will be useful, but
// WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
// General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program; if not, write to the Free Software
// Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
//
// $Id$
#ifdef HAVE_CONFIG_H
# include <config.h>
#endif
1998-04-21 17:02:27 +00:00
2003-08-17 09:54:41 +00:00
#include <simgear/compiler.h>
#if defined(__linux__) && defined(__i386__)
# include <fpu_control.h>
# include <signal.h>
#endif
2001-03-23 22:42:49 +00:00
#ifdef SG_MATH_EXCEPTION_CLASH
# include <math.h>
#endif
1998-04-21 17:02:27 +00:00
#ifdef HAVE_WINDOWS_H
# include <windows.h>
# include <float.h>
1998-04-21 17:02:27 +00:00
#endif
#include <plib/netSocket.h>
2003-08-17 09:54:41 +00:00
#include <simgear/props/props.hxx>
#include <simgear/timing/sg_time.hxx>
2004-09-19 17:59:40 +00:00
#include <simgear/math/sg_random.h>
// Class refferences
#include <simgear/ephemeris/ephemeris.hxx>
2003-05-15 21:35:51 +00:00
#include <simgear/scene/model/modellib.hxx>
#include <simgear/scene/material/matlib.hxx>
#include <simgear/scene/model/animation.hxx>
#include <simgear/scene/sky/sky.hxx>
#include <Time/light.hxx>
2003-08-17 09:54:41 +00:00
2000-02-15 03:30:01 +00:00
#include <Include/general.hxx>
2003-08-17 09:54:41 +00:00
#include <Cockpit/cockpit.hxx>
#include <Cockpit/hud.hxx>
#include <Model/panelnode.hxx>
#include <Model/modelmgr.hxx>
#include <Model/acmodel.hxx>
#include <Scenery/scenery.hxx>
#include <Scenery/tilemgr.hxx>
#include <Sound/beacon.hxx>
#include <Sound/morse.hxx>
2003-08-17 09:54:41 +00:00
#include <FDM/flight.hxx>
#include <FDM/UIUCModel/uiuc_aircraftdir.h>
// #include <FDM/ADA.hxx>
#include <ATC/ATCdisplay.hxx>
2003-08-17 09:54:41 +00:00
#include <ATC/ATCmgr.hxx>
#include <ATC/AIMgr.hxx>
2003-08-17 09:54:41 +00:00
#include <Replay/replay.hxx>
#include <Time/tmp.hxx>
#include <Time/fg_timer.hxx>
#include <Environment/environment_mgr.hxx>
#include <GUI/new_gui.hxx>
2004-09-19 17:59:40 +00:00
#ifdef FG_MPLAYER_AS
#include <MultiPlayer/multiplaytxmgr.hxx>
#include <MultiPlayer/multiplayrxmgr.hxx>
#endif
2002-10-29 19:44:03 +00:00
#include "fg_commands.hxx"
2003-08-17 09:54:41 +00:00
#include "fg_io.hxx"
#include "renderer.hxx"
#include "splash.hxx"
#include "main.hxx"
2002-10-16 03:44:44 +00:00
2004-09-19 17:59:40 +00:00
static double real_delta_time_sec = 0.0;
double delta_time_sec = 0.0;
extern float init_volume;
2000-09-10 00:04:50 +00:00
#ifdef macintosh
# include <console.h> // -dw- for command line dialog
#endif
// This is a record containing a bit of global housekeeping information
FGGeneral general;
1998-04-21 17:02:27 +00:00
// Specify our current idle function state. This is used to run all
// our initializations out of the idle callback so that we can get a
// splash screen up and running right away.
int idle_state = 0;
long global_multi_loop;
SGTimeStamp last_time_stamp;
SGTimeStamp current_time_stamp;
// The atexit() functio handler should know when the graphical subsystem
// is initialized.
extern int _bootstrap_OSInit;
2003-03-22 10:38:15 +00:00
1998-04-21 17:02:27 +00:00
// Update internal time dependent calculations (i.e. flight model)
2002-05-11 21:10:11 +00:00
// FIXME: this distinction is obsolete; all subsystems now get delta
// time on update.
void fgUpdateTimeDepCalcs() {
2001-01-17 23:30:35 +00:00
static bool inited = false;
static const SGPropertyNode *replay_master
= fgGetNode( "/sim/freeze/replay", true );
static SGPropertyNode *replay_time
= fgGetNode( "/sim/replay/time", true );
// static const SGPropertyNode *replay_end_time
// = fgGetNode( "/sim/replay/end-time", true );
//SG_LOG(SG_FLIGHT,SG_INFO, "Updating time dep calcs()");
// Initialize the FDM here if it hasn't been and if we have a
// scenery elevation hit.
// cout << "cur_fdm_state->get_inited() = " << cur_fdm_state->get_inited()
// << " cur_elev = " << scenery.get_cur_elev() << endl;
if ( !cur_fdm_state->get_inited() &&
globals->get_scenery()->get_cur_elev() > -9990 )
{
SG_LOG(SG_FLIGHT,SG_INFO, "Finally initializing fdm");
cur_fdm_state->init();
if ( cur_fdm_state->get_bound() ) {
cur_fdm_state->unbind();
}
cur_fdm_state->bind();
}
// conceptually, the following block could be done for each fdm
// instance ...
if ( cur_fdm_state->get_inited() ) {
// we have been inited, and we are good to go ...
2001-01-17 23:30:35 +00:00
if ( !inited ) {
inited = true;
}
2001-01-17 23:30:35 +00:00
if ( ! replay_master->getBoolValue() ) {
cur_fdm_state->update( delta_time_sec );
} else {
FGReplay *r = (FGReplay *)(globals->get_subsystem( "replay" ));
r->replay( replay_time->getDoubleValue() );
replay_time->setDoubleValue( replay_time->getDoubleValue()
+ ( delta_time_sec
* fgGetInt("/sim/speed-up") ) );
}
} else {
// do nothing, fdm isn't inited yet
}
globals->get_model_mgr()->update(delta_time_sec);
globals->get_aircraft_model()->update(delta_time_sec);
// update the view angle
globals->get_viewmgr()->update(delta_time_sec);
// Update solar system
2000-07-08 05:09:24 +00:00
globals->get_ephem()->update( globals->get_time_params()->getMjd(),
globals->get_time_params()->getLst(),
cur_fdm_state->get_Latitude() );
1998-04-21 17:02:27 +00:00
}
void fgInitTimeDepCalcs( void ) {
2002-04-05 00:38:55 +00:00
// noop for now
1998-04-21 17:02:27 +00:00
}
1999-09-01 20:24:54 +00:00
static const double alt_adjust_ft = 3.758099;
2001-03-24 04:56:46 +00:00
static const double alt_adjust_m = alt_adjust_ft * SG_FEET_TO_METER;
1998-04-21 17:02:27 +00:00
1999-09-01 20:24:54 +00:00
// What should we do when we have nothing else to do? Let's get ready
// for the next move and update the display?
1998-04-21 17:02:27 +00:00
static void fgMainLoop( void ) {
int model_hz = fgGetInt("/sim/model-hz");
static const SGPropertyNode *longitude
= fgGetNode("/position/longitude-deg");
static const SGPropertyNode *latitude
= fgGetNode("/position/latitude-deg");
static const SGPropertyNode *altitude
= fgGetNode("/position/altitude-ft");
static const SGPropertyNode *clock_freeze
= fgGetNode("/sim/freeze/clock", true);
static const SGPropertyNode *cur_time_override
= fgGetNode("/sim/time/cur-time-override", true);
// static const SGPropertyNode *replay_master
// = fgGetNode("/sim/freeze/replay", true);
SGCloudLayer::enable_bump_mapping = fgGetBool("/sim/rendering/bump-mapping");
// Update the elapsed time.
static bool first_time = true;
if ( first_time ) {
last_time_stamp.stamp();
first_time = false;
}
double throttle_hz = fgGetDouble("/sim/frame-rate-throttle-hz", 0.0);
if ( throttle_hz > 0.0 ) {
// simple frame rate throttle
double dt = 1000000.0 / throttle_hz;
current_time_stamp.stamp();
while ( current_time_stamp - last_time_stamp < dt ) {
current_time_stamp.stamp();
}
} else {
// run as fast as the app will go
current_time_stamp.stamp();
}
real_delta_time_sec
= double(current_time_stamp - last_time_stamp) / 1000000.0;
// round the real time down to a multiple of 1/model-hz.
// this way all systems are updated the _same_ amount of dt.
{
static double rem = 0.0;
real_delta_time_sec += rem;
double hz = model_hz;
double nit = floor(real_delta_time_sec*hz);
rem = real_delta_time_sec - nit/hz;
real_delta_time_sec = nit/hz;
}
if ( clock_freeze->getBoolValue() ) {
delta_time_sec = 0;
} else {
delta_time_sec = real_delta_time_sec;
}
last_time_stamp = current_time_stamp;
globals->inc_sim_time_sec( delta_time_sec );
SGAnimation::set_sim_time_sec( globals->get_sim_time_sec() );
// These are useful, especially for Nasal scripts.
fgSetDouble("/sim/time/delta-realtime-sec", real_delta_time_sec);
fgSetDouble("/sim/time/delta-sec", delta_time_sec);
static long remainder = 0;
long elapsed;
#ifdef FANCY_FRAME_COUNTER
int i;
double accum;
#else
1998-12-18 23:40:55 +00:00
static time_t last_time = 0;
static int frames = 0;
#endif // FANCY_FRAME_COUNTER
1998-04-21 17:02:27 +00:00
2000-07-07 20:28:51 +00:00
SGTime *t = globals->get_time_params();
1998-04-21 17:02:27 +00:00
globals->get_event_mgr()->update(delta_time_sec);
2001-03-24 06:03:11 +00:00
SG_LOG( SG_ALL, SG_DEBUG, "Running Main Loop");
SG_LOG( SG_ALL, SG_DEBUG, "======= ==== ====");
#if defined( ENABLE_PLIB_JOYSTICK )
1998-12-18 23:40:55 +00:00
// Read joystick and update control settings
// if ( fgGetString("/sim/control-mode") == "joystick" )
// {
// fgJoystickRead();
// }
1998-12-18 23:40:55 +00:00
#endif
// Fix elevation. I'm just sticking this here for now, it should
// probably move eventually
/* printf("Before - ground = %.2f runway = %.2f alt = %.2f\n",
scenery.get_cur_elev(),
cur_fdm_state->get_Runway_altitude() * SG_FEET_TO_METER,
cur_fdm_state->get_Altitude() * SG_FEET_TO_METER); */
/* printf("Adjustment - ground = %.2f runway = %.2f alt = %.2f\n",
scenery.get_cur_elev(),
cur_fdm_state->get_Runway_altitude() * SG_FEET_TO_METER,
cur_fdm_state->get_Altitude() * SG_FEET_TO_METER); */
// cout << "Warp = " << globals->get_warp() << endl;
1998-04-21 17:02:27 +00:00
// update "time"
static bool last_clock_freeze = false;
if ( clock_freeze->getBoolValue() ) {
// clock freeze requested
if ( cur_time_override->getLongValue() == 0 ) {
fgSetLong( "/sim/time/cur-time-override", t->get_cur_time() );
globals->set_warp( 0 );
}
} else {
// no clock freeze requested
if ( last_clock_freeze == true ) {
// clock just unfroze, let's set warp as the difference
// between frozen time and current time so we don't get a
// time jump (and corresponding sky object and lighting
// jump.)
globals->set_warp( cur_time_override->getLongValue() - time(NULL) );
fgSetLong( "/sim/time/cur-time-override", 0 );
}
if ( globals->get_warp_delta() != 0 ) {
globals->inc_warp( globals->get_warp_delta() );
}
}
last_clock_freeze = clock_freeze->getBoolValue();
t->update( longitude->getDoubleValue() * SGD_DEGREES_TO_RADIANS,
latitude->getDoubleValue() * SGD_DEGREES_TO_RADIANS,
cur_time_override->getLongValue(),
globals->get_warp() );
1998-04-21 17:02:27 +00:00
if (globals->get_warp_delta() != 0) {
FGLight *l = (FGLight *)(globals->get_subsystem("lighting"));
l->update( 0.5 );
}
// update magvar model
globals->get_mag()->update( longitude->getDoubleValue()
* SGD_DEGREES_TO_RADIANS,
latitude->getDoubleValue()
* SGD_DEGREES_TO_RADIANS,
altitude->getDoubleValue() * SG_FEET_TO_METER,
globals->get_time_params()->getJD() );
// Get elapsed time (in usec) for this past frame
1998-04-21 17:02:27 +00:00
elapsed = fgGetTimeInterval();
2001-03-24 06:03:11 +00:00
SG_LOG( SG_ALL, SG_DEBUG,
"Elapsed time interval is = " << elapsed
<< ", previous remainder is = " << remainder );
// Calculate frame rate average
#ifdef FANCY_FRAME_COUNTER
/* old fps calculation */
if ( elapsed > 0 ) {
double tmp;
accum = 0.0;
for ( i = FG_FRAME_RATE_HISTORY - 2; i >= 0; i-- ) {
tmp = general.get_frame(i);
accum += tmp;
// printf("frame[%d] = %.2f\n", i, g->frames[i]);
general.set_frame(i+1,tmp);
}
tmp = 1000000.0 / (float)elapsed;
general.set_frame(0,tmp);
// printf("frame[0] = %.2f\n", general.frames[0]);
accum += tmp;
general.set_frame_rate(accum / (float)FG_FRAME_RATE_HISTORY);
// printf("ave = %.2f\n", general.frame_rate);
}
#else
if ( (t->get_cur_time() != last_time) && (last_time > 0) ) {
general.set_frame_rate( frames );
fgSetInt("/sim/frame-rate", frames);
SG_LOG( SG_ALL, SG_DEBUG,
"--> Frame rate is = " << general.get_frame_rate() );
frames = 0;
1998-12-18 23:40:55 +00:00
}
last_time = t->get_cur_time();
1998-12-18 23:40:55 +00:00
++frames;
#endif
// Run ATC subsystem
if (fgGetBool("/sim/ATC/enabled"))
globals->get_ATC_mgr()->update(delta_time_sec);
// Run the AI subsystem
if (fgGetBool("/sim/ai-traffic/enabled"))
globals->get_AI_mgr()->update(delta_time_sec);
#ifdef FG_MPLAYER_AS
// Update any multiplayer models
globals->get_multiplayer_rx_mgr()->Update();
#endif
1998-04-21 17:02:27 +00:00
// Run flight model
// Calculate model iterations needed for next frame
elapsed += remainder;
global_multi_loop = (long)(((double)elapsed * 0.000001) * model_hz );
remainder = elapsed - ( (global_multi_loop*1000000) / model_hz );
2001-03-24 06:03:11 +00:00
SG_LOG( SG_ALL, SG_DEBUG,
"Model iterations needed = " << global_multi_loop
<< ", new remainder = " << remainder );
2001-01-17 23:30:35 +00:00
// chop max interations to something reasonable if the sim was
// delayed for an excesive amount of time
if ( global_multi_loop > 2.0 * model_hz ) {
global_multi_loop = (int)(2.0 * model_hz );
remainder = 0;
2001-01-17 23:30:35 +00:00
}
bool scenery_loaded = fgGetBool("sim/sceneryloaded") || fgGetBool("sim/sceneryloaded-override");
// flight model
if ( global_multi_loop > 0) {
// first run the flight model each frame until it is intialized
// then continue running each frame only after initial scenery load is complete.
if (!cur_fdm_state->get_inited() || scenery_loaded) {
fgUpdateTimeDepCalcs();
}
} else {
SG_LOG( SG_ALL, SG_DEBUG,
"Elapsed time is zero ... we're zinging" );
1998-04-21 17:02:27 +00:00
}
1999-11-19 02:10:24 +00:00
// Do any I/O channel work that might need to be done
globals->get_io()->update( real_delta_time_sec );
1998-04-21 17:02:27 +00:00
// see if we need to load any deferred-load textures
globals->get_matlib()->load_next_deferred();
// Run audio scheduler
#ifdef ENABLE_AUDIO_SUPPORT
if ( globals->get_soundmgr()->is_working() ) {
globals->get_soundmgr()->update( delta_time_sec );
}
#endif
globals->get_subsystem_mgr()->update(delta_time_sec);
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
//
// Tile Manager updates - see if we need to load any new scenery tiles.
// this code ties together the fdm, viewer and scenery classes...
// we may want to move this to it's own class at some point
//
double visibility_meters = fgGetDouble("/environment/visibility-m");
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
FGViewer *current_view = globals->get_current_view();
// get the location data for the primary FDM (now hardcoded to ac model)...
SGLocation *acmodel_loc = NULL;
acmodel_loc = (SGLocation *)globals->
get_aircraft_model()->get3DModel()->getSGLocation();
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
// update tile manager for FDM...
// ...only if location is different than the current-view location
// (to avoid duplicating effort)
if( !fgGetBool("/sim/current-view/config/from-model") ) {
if( acmodel_loc != NULL ) {
globals->get_tile_mgr()->prep_ssg_nodes( acmodel_loc,
visibility_meters );
globals->get_tile_mgr()->
update( acmodel_loc, visibility_meters,
acmodel_loc->
get_absolute_view_pos(globals->
get_scenery()->get_center()) );
// save results of update in SGLocation for fdm...
if ( globals->get_scenery()->get_cur_elev() > -9990 ) {
acmodel_loc->
set_cur_elev_m( globals->get_scenery()->get_cur_elev() );
fgSetDouble("/position/ground-elev-m",
globals->get_scenery()->get_cur_elev());
}
acmodel_loc->
set_tile_center( globals->get_scenery()->get_next_center() );
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
}
}
globals->get_tile_mgr()->prep_ssg_nodes( current_view->getSGLocation(),
visibility_meters );
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
// update tile manager for view...
// IMPORTANT!!! the tilemgr update for view location _must_ be
// done last after the FDM's until all of Flight Gear code
// references the viewer's location for elevation instead of the
// "scenery's" current elevation.
SGLocation *view_location = globals->get_current_view()->getSGLocation();
globals->get_tile_mgr()->update( view_location, visibility_meters,
current_view->get_absolute_view_pos() );
// save results of update in SGLocation for fdm...
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
if ( globals->get_scenery()->get_cur_elev() > -9990 ) {
current_view->getSGLocation()->
set_cur_elev_m( globals->get_scenery()->get_cur_elev() );
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
}
current_view->getSGLocation()->
set_tile_center( globals->get_scenery()->get_next_center() );
#ifdef ENABLE_AUDIO_SUPPORT
// Right now we make a simplifying assumption that the primary
// aircraft is the source of all sounds and that all sounds are
// positioned relative to the current view position.
static sgVec3 last_pos_offset;
// set positional offset for sources
sgVec3 source_pos_offset;
sgSubVec3( source_pos_offset,
view_location->get_view_pos(), acmodel_loc->get_view_pos() );
// cout << "pos all = " << source_pos_offset[0] << " " << source_pos_offset[1] << " " << source_pos_offset[2] << endl;
globals->get_soundmgr()->set_source_pos_all( source_pos_offset );
// set the velocity
sgVec3 source_vel;
sgSubVec3( source_vel, source_pos_offset, last_pos_offset );
sgScaleVec3( source_vel, delta_time_sec );
sgCopyVec3( last_pos_offset, source_pos_offset );
// cout << "vel = " << source_vel[0] << " " << source_vel[1] << " " << source_vel[2] << endl;
globals->get_soundmgr()->set_source_vel_all( source_vel );
// Right now we make a simplifying assumption that the listener is
// always positioned at the origin.
sgVec3 listener_pos;
sgSetVec3( listener_pos, 0.0, 0.0, 0.0 );
// cout << "listener = " << listener_pos[0] << " " << listener_pos[1] << " " << listener_pos[2] << endl;
globals->get_soundmgr()->set_listener_pos( listener_pos );
#endif
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
// If fdm location is same as viewer's then we didn't do the
// update for fdm location above so we need to save the viewer
// results in the fdm SGLocation as well...
if( fgGetBool("/sim/current-view/config/from-model") ) {
if( acmodel_loc != 0 ) {
if ( globals->get_scenery()->get_cur_elev() > -9990 ) {
acmodel_loc->set_cur_elev_m( globals->get_scenery()->
get_cur_elev() );
fgSetDouble("/position/ground-elev-m",
globals->get_scenery()->get_cur_elev());
}
acmodel_loc->set_tile_center( globals->get_scenery()->
get_next_center() );
From: "Jim Wilson" <jimw@kelcomaine.com> This is a new improved patch for the previous tile manager fixes. Rather than building dependencies between FGlocation or the viewer or fdm with tilemgr what I ended up doing was linking the pieces together in the Mainloop in main.cxx. You'll see what I mean...it's been commented fairly well. More than likely we should move that chunk somewhere...just not sure where yet. The changes seem clean now. As I get more ideas there could be some further improvement in organizing the update in tilemgr. You'll note that I left an override in there for the tilemgr::update() function to preserve earlier functionality if someone needs it (e.g. usage independent of an fdm or viewer), not to mention there are a few places in flightgear that call it directly that have not been changed to the new interface (and may not need to be). The code has been optimized to avoid duplicate traversals and seems to run generally quite well. Note that there can be a short delay reloading tiles that have been dropped from static views. We could call the tile scheduler on a view switch, but it's not a big deal and at the moment I'd like to get this in so people can try it and comment on it as it is. Everything has been resycned with CVS tonight and I've included the description submitted earlier (below). Best, Jim Changes synced with CVS approx 20:30EDT 2002-05-09 (after this evenings updates). Files: http://www.spiderbark.com/fgfs/viewer-update-20020516.tar.gz or http://www.spiderbark.com/fgfs/viewer-update-20020516.diffs.gz Description: In a nutshell, these patches begin to take what was one value for ground elevation and calculate ground elevation values seperately for the FDM and the viewer (eye position). Several outstanding view related bugs have been fixed. With the introduction of the new viewer code a lot of that Flight Gear code broke related to use of a global variable called "scenery.cur_elev". Therefore the ground_elevation and other associated items (like the current tile bucket) is maintained per FDM instance and per View. Each of these has a "point" or location that can be identified. See changes to FGLocation class and main.cxx. Most of the problems related to the new viewer in terms of sky, ground and runway lights, and tower views are fixed. There are four minor problems remaining. 1) The sun/moon spins when you pan the "lookat" tower view only (view #3). 2) Under stress (esp. magic carpet full speed with max visibility), there is a memory leak in the tile caching that was not introduced with these changes. 3) I have not tested these changes or made corrections to the ADA or External FDM interfaces. 4) The change view function doesn't call the time/light update (not a problem unless a tower is very far away). Details: FDM/flight.cxx, flight.hxx - FGInterface ties to FGAircraftModel so that it's location data can be accessed for runway (ground elevation under aircraft) elevation. FDM/larsim.cxx, larcsim.hxx - gets runway elevation from FGInterface now. Commented out function that is causing a namespace conflict, hasn't been called with recent code anyway. FDM/JSBSim/JSBSim.cxx, YASim/YASim.cxx - gets runway elevation from FGInterface now. Scenery/newcache.cxx, newcache.hxx - changed caching scheme to time based (oldest tiles discard). Scenery/tileentry.cxx, tileentry.hxx - added place to record time, changed rendering to reference viewer altitude in order to fix a problem with ground and runway lights. Scenery/tilemgr.cxx, tilemgr.hxx - Modified update() to accept values for multiple locations. Refresh function added in order to periodically make the tiles current for a non-moving view (like a tower). Main/fg_init.cxx - register event for making tiles current in a non-moving view (like a tower). Main/location.hxx - added support for current ground elevation data. Main/main.cxx - added second tilemgr call for fdm, fixed places where viewer position data was required for correct sky rendering. Main/options.cxx - fixed segfault reported by Curtis when using --view-offset command line parameter. Main/viewer.cxx, viewer.hxx - removed fudging of view position. Fixed numerous bugs that were causing eye and target values to get mixed up.
2002-05-17 17:25:28 +00:00
}
}
// END Tile Manager udpates
if (!scenery_loaded && globals->get_tile_mgr()->all_queues_empty() && cur_fdm_state->get_inited()) {
fgSetBool("sim/sceneryloaded",true);
fgSetFloat("/sim/sound/volume", init_volume);
globals->get_soundmgr()->set_volume(init_volume);
}
if (fgGetBool("/sim/rendering/specular-highlight")) {
glLightModeli(GL_LIGHT_MODEL_COLOR_CONTROL, GL_SEPARATE_SPECULAR_COLOR);
// glLightModeli(GL_LIGHT_MODEL_TWO_SIDE, GL_TRUE);
} else {
glLightModeli(GL_LIGHT_MODEL_COLOR_CONTROL, GL_SINGLE_COLOR);
// glLightModeli(GL_LIGHT_MODEL_TWO_SIDE, GL_FALSE);
}
fgRequestRedraw();
2001-03-24 06:03:11 +00:00
SG_LOG( SG_ALL, SG_DEBUG, "" );
1998-04-21 17:02:27 +00:00
}
// This is the top level master main function that is registered as
// our idle funciton
// The first few passes take care of initialization things (a couple
// per pass) and once everything has been initialized fgMainLoop from
// then on.
static void fgIdleFunction ( void ) {
// printf("idle state == %d\n", idle_state);
if ( idle_state == 0 ) {
// Initialize the splash screen right away
if ( fgGetBool("/sim/startup/splash-screen") ) {
fgSplashInit(fgGetString("/sim/startup/splash-texture"));
}
idle_state++;
} else if ( idle_state == 1 ) {
// Initialize audio support
#ifdef ENABLE_AUDIO_SUPPORT
// Start the intro music
if ( fgGetBool("/sim/startup/intro-music") ) {
SGPath mp3file( globals->get_fg_root() );
mp3file.append( "Sounds/intro.mp3" );
SG_LOG( SG_GENERAL, SG_INFO,
"Starting intro music: " << mp3file.str() );
#if defined( __CYGWIN__ )
string command = "start /m `cygpath -w " + mp3file.str() + "`";
#elif defined( WIN32 )
string command = "start /m " + mp3file.str();
#else
string command = "mpg123 " + mp3file.str() + "> /dev/null 2>&1";
#endif
system ( command.c_str() );
}
#endif
idle_state++;
} else if ( idle_state == 2 ) {
// These are a few miscellaneous things that aren't really
// "subsystems" but still need to be initialized.
#ifdef USE_GLIDE
if ( strstr ( general.get_glRenderer(), "Glide" ) ) {
grTexLodBiasValue ( GR_TMU0, 1.0 ) ;
}
#endif
idle_state++;
} else if ( idle_state == 3 ) {
// This is the top level init routine which calls all the
// other subsystem initialization routines. If you are adding
// a subsystem to flightgear, its initialization call should
// located in this routine.
if( !fgInitSubsystems()) {
SG_LOG( SG_GENERAL, SG_ALERT,
"Subsystem initializations failed ..." );
exit(-1);
}
idle_state++;
} else if ( idle_state == 4 ) {
// Initialize the time offset (warp) after fgInitSubsystem
// (which initializes the lighting interpolation tables.)
fgInitTimeOffset();
// setup OpenGL view parameters
globals->get_renderer()->init();
// Read the list of available aircrafts
fgReadAircraft();
idle_state++;
} else if ( idle_state == 5 ) {
idle_state++;
} else if ( idle_state == 6 ) {
// sleep(1);
idle_state = 1000;
SG_LOG( SG_GENERAL, SG_INFO, "Panel visible = " << fgPanelVisible() );
globals->get_renderer()->resize( fgGetInt("/sim/startup/xsize"),
fgGetInt("/sim/startup/ysize") );
}
if ( idle_state == 1000 ) {
// We've finished all our initialization steps, from now on we
// run the main loop.
fgSetBool("sim/sceneryloaded",false);
fgRegisterIdleHandler(fgMainLoop);
} else {
if ( fgGetBool("/sim/startup/splash-screen") ) {
fgSplashUpdate(0.0, 1.0);
}
}
}
// Main top level initialization
2003-08-17 09:54:41 +00:00
bool fgMainInit( int argc, char **argv ) {
2000-09-10 00:04:50 +00:00
#if defined( macintosh )
2000-04-27 21:57:08 +00:00
freopen ("stdout.txt", "w", stdout );
freopen ("stderr.txt", "w", stderr );
argc = ccommand( &argv );
#endif
// set default log levels
sglog().setLogLevels( SG_ALL, SG_ALERT );
string version;
#ifdef FLIGHTGEAR_VERSION
version = FLIGHTGEAR_VERSION;
#else
version = "unknown version";
#endif
2001-03-24 06:03:11 +00:00
SG_LOG( SG_GENERAL, SG_INFO, "FlightGear: Version "
<< version );
SG_LOG( SG_GENERAL, SG_INFO, "Built with " << SG_COMPILER_STR << endl );
// Allocate global data structures. This needs to happen before
// we parse command line options
globals = new FGGlobals;
// seed the random number generater
sg_srandom_time();
FGControls *controls = new FGControls;
globals->set_controls( controls );
string_list *col = new string_list;
globals->set_channel_options_list( col );
// Scan the config file(s) and command line options to see if
// fg_root was specified (ignore all other options for now)
fgInitFGRoot(argc, argv);
// Check for the correct base package version
static char required_version[] = "0.9.8";
string base_version = fgBasePackageVersion();
if ( !(base_version == required_version) ) {
// tell the operator how to use this application
SG_LOG( SG_GENERAL, SG_ALERT, "" ); // To popup the console on windows
cerr << endl << "Base package check failed ... " \
<< "Found version " << base_version << " at: " \
<< globals->get_fg_root() << endl;
cerr << "Please upgrade to version: " << required_version << endl;
#ifdef _MSC_VER
cerr << "Hit a key to continue..." << endl;
cin.get();
#endif
exit(-1);
}
sgUseDisplayList = fgGetBool( "/sim/rendering/use-display-list", true );
// Initialize the Aircraft directory to "" (UIUC)
aircraft_dir = "";
// Load the configuration parameters. (Command line options
// overrides config file options. Config file options override
// defaults.)
if ( !fgInitConfig(argc, argv) ) {
SG_LOG( SG_GENERAL, SG_ALERT, "Config option parsing failed ..." );
exit(-1);
}
1998-08-20 15:10:33 +00:00
// Initialize the Window/Graphics environment.
#if !defined(__APPLE__) || defined(OSX_BUNDLE)
// Mac OS X command line ("non-bundle") applications call this
// from main(), in bootstrap.cxx. Andy doesn't know why, someone
// feel free to add comments...
fgOSInit(&argc, argv);
_bootstrap_OSInit++;
#endif
1998-08-20 15:10:33 +00:00
2004-12-20 08:36:56 +00:00
fgRegisterWindowResizeHandler( &FGRenderer::resize );
fgRegisterIdleHandler( &fgIdleFunction );
fgRegisterDrawHandler( &FGRenderer::update );
Frederic Bouvier: FG_ENABLE_MULTIPASS_CLOUDS must be defined to enable the algorithm. I made this because the stencil buffer must be initialized at the beginning of the program and OpenGL can fallback to software rendering if it can't find a visual with stencil buffer. I didn't touch the configure script, so CXXFLAGS=-DFG_ENABLE_MULTIPASS_CLOUDS must be set before running ./configure. If FG_ENABLE_MULTIPASS_CLOUDS is defined, the main render loop begins by reading the /sim/rendering/multi-pass-clouds property. It is a boolean property so there are only two quality levels. false means no multi pass and no use of the stencil buffer, true means an additionnal pass for both upper and lower cloud layers. The algorithms are as follow : /sim/rendering/multi-pass-clouds=false 1. draw sky dome 2. draw terrain only 3. draw clouds above the viewer 4. draw models except the aircraft 5. draw clouds below the viewer 6. draw the aircraft. The cloud rendering doesn't update the depth buffer. This means that models overwrite clouds above the viewer. This is only noticeable for tall buildings and when flying very low. Also, drawing low clouds after models means that they are not blended with models' translucent surfaces. Large transparent area require alpha test enabled and AI aircraft canopy are making holes. The pilot's aircraft being rendered at the end, there is no problem with canopy or prop disc. /sim/rendering/multi-pass-clouds=true 1. draw the sky dome 2. draw the terrain only 3. draw all clouds 4. draw models except the aircraft 5. redraw the clouds where the models where drawn ( stencil test on ) 6. draw the aircraft The assumptions made by this algoritm are that the terrain is not transparent ( should be true in all cases and that there are no clouds between the aircraft and the viewer. Assuming these facts, there should be no blending bugs. The screenshot rendering is not updated yet.
2004-04-02 14:40:54 +00:00
#ifdef FG_ENABLE_MULTIPASS_CLOUDS
bool get_stencil_buffer = true;
#else
bool get_stencil_buffer = false;
#endif
// Clouds3D requires an alpha channel
// clouds may require stencil buffer
fgOSOpenWindow( fgGetInt("/sim/startup/xsize"),
fgGetInt("/sim/startup/ysize"),
2004-04-01 15:44:13 +00:00
fgGetInt("/sim/rendering/bits-per-pixel"),
Frederic Bouvier: FG_ENABLE_MULTIPASS_CLOUDS must be defined to enable the algorithm. I made this because the stencil buffer must be initialized at the beginning of the program and OpenGL can fallback to software rendering if it can't find a visual with stencil buffer. I didn't touch the configure script, so CXXFLAGS=-DFG_ENABLE_MULTIPASS_CLOUDS must be set before running ./configure. If FG_ENABLE_MULTIPASS_CLOUDS is defined, the main render loop begins by reading the /sim/rendering/multi-pass-clouds property. It is a boolean property so there are only two quality levels. false means no multi pass and no use of the stencil buffer, true means an additionnal pass for both upper and lower cloud layers. The algorithms are as follow : /sim/rendering/multi-pass-clouds=false 1. draw sky dome 2. draw terrain only 3. draw clouds above the viewer 4. draw models except the aircraft 5. draw clouds below the viewer 6. draw the aircraft. The cloud rendering doesn't update the depth buffer. This means that models overwrite clouds above the viewer. This is only noticeable for tall buildings and when flying very low. Also, drawing low clouds after models means that they are not blended with models' translucent surfaces. Large transparent area require alpha test enabled and AI aircraft canopy are making holes. The pilot's aircraft being rendered at the end, there is no problem with canopy or prop disc. /sim/rendering/multi-pass-clouds=true 1. draw the sky dome 2. draw the terrain only 3. draw all clouds 4. draw models except the aircraft 5. redraw the clouds where the models where drawn ( stencil test on ) 6. draw the aircraft The assumptions made by this algoritm are that the terrain is not transparent ( should be true in all cases and that there are no clouds between the aircraft and the viewer. Assuming these facts, there should be no blending bugs. The screenshot rendering is not updated yet.
2004-04-02 14:40:54 +00:00
fgGetBool("/sim/rendering/clouds3d"),
get_stencil_buffer,
fgGetBool("/sim/startup/fullscreen") );
// This seems to be the absolute earliest in the init sequence
// that these calls will return valid info. Too bad it's after
// we've already created and sized out window. :-(
general.set_glVendor( (char *)glGetString ( GL_VENDOR ) );
general.set_glRenderer( (char *)glGetString ( GL_RENDERER ) );
general.set_glVersion( (char *)glGetString ( GL_VERSION ) );
SG_LOG( SG_GENERAL, SG_INFO, general.get_glRenderer() );
GLint tmp;
glGetIntegerv( GL_MAX_TEXTURE_SIZE, &tmp );
general.set_glMaxTexSize( tmp );
SG_LOG ( SG_GENERAL, SG_INFO, "Max texture size = " << tmp );
glGetIntegerv( GL_DEPTH_BITS, &tmp );
general.set_glDepthBits( tmp );
SG_LOG ( SG_GENERAL, SG_INFO, "Depth buffer bits = " << tmp );
// Initialize plib net interface
netInit( &argc, argv );
// Initialize ssg (from plib). Needs to come before we do any
// other ssg stuff, but after opengl has been initialized.
ssgInit();
1998-08-20 15:10:33 +00:00
// Initialize the user interface (we need to do this before
// passing off control to the OS main loop and before
// fgInitGeneral to get our fonts !!!
guiInit();
// Read the list of available aircrafts
fgReadAircraft();
2001-05-23 20:54:51 +00:00
#ifdef GL_EXT_texture_lod_bias
glTexEnvf( GL_TEXTURE_FILTER_CONTROL_EXT, GL_TEXTURE_LOD_BIAS_EXT, -0.5 ) ;
#endif
2002-10-17 04:34:32 +00:00
// get the address of our OpenGL extensions
if ( fgGetBool("/sim/rendering/distance-attenuation") )
{
2003-06-20 20:18:40 +00:00
if (SGIsOpenGLExtensionSupported("GL_EXT_point_parameters") ) {
glPointParameterIsSupported = true;
2003-06-20 20:18:40 +00:00
glPointParameterfPtr = (glPointParameterfProc)
SGLookupFunction("glPointParameterfEXT");
glPointParameterfvPtr = (glPointParameterfvProc)
SGLookupFunction("glPointParameterfvEXT");
} else if ( SGIsOpenGLExtensionSupported("GL_ARB_point_parameters") ) {
glPointParameterIsSupported = true;
2003-06-20 20:18:40 +00:00
glPointParameterfPtr = (glPointParameterfProc)
SGLookupFunction("glPointParameterfARB");
glPointParameterfvPtr = (glPointParameterfvProc)
SGLookupFunction("glPointParameterfvARB");
} else
glPointParameterIsSupported = false;
}
2002-10-17 04:34:32 +00:00
// based on the requested presets, calculate the true starting
// lon, lat
2003-02-21 02:46:07 +00:00
fgInitNav();
fgInitPosition();
SGTime *t = fgInitTime();
2000-07-07 20:28:51 +00:00
globals->set_time_params( t );
// Do some quick general initializations
if( !fgInitGeneral()) {
SG_LOG( SG_GENERAL, SG_ALERT,
"General initializations failed ..." );
exit(-1);
}
2002-10-29 19:44:03 +00:00
////////////////////////////////////////////////////////////////////
// Initialize the property-based built-in commands
////////////////////////////////////////////////////////////////////
fgInitCommands();
////////////////////////////////////////////////////////////////////
// Initialize the material manager
////////////////////////////////////////////////////////////////////
globals->set_matlib( new SGMaterialLib );
2003-05-15 21:35:51 +00:00
globals->set_model_lib(new SGModelLib);
////////////////////////////////////////////////////////////////////
// Initialize the TG scenery subsystem.
////////////////////////////////////////////////////////////////////
globals->set_scenery( new FGScenery );
globals->get_scenery()->init();
globals->get_scenery()->bind();
globals->set_tile_mgr( new FGTileMgr );
////////////////////////////////////////////////////////////////////
// Initialize the general model subsystem.
////////////////////////////////////////////////////////////////////
globals->set_model_mgr(new FGModelMgr);
globals->get_model_mgr()->init();
globals->get_model_mgr()->bind();
////////////////////////////////////////////////////////////////////
// Initialize the 3D aircraft model subsystem (has a dependency on
// the scenery subsystem.)
////////////////////////////////////////////////////////////////////
globals->set_aircraft_model(new FGAircraftModel);
globals->get_aircraft_model()->init();
globals->get_aircraft_model()->bind();
////////////////////////////////////////////////////////////////////
// Initialize the view manager subsystem.
////////////////////////////////////////////////////////////////////
FGViewMgr *viewmgr = new FGViewMgr;
globals->set_viewmgr( viewmgr );
viewmgr->init();
viewmgr->bind();
// Initialize the sky
SGPath ephem_data_path( globals->get_fg_root() );
ephem_data_path.append( "Astro" );
2000-07-08 05:09:24 +00:00
SGEphemeris *ephem = new SGEphemeris( ephem_data_path.c_str() );
ephem->update( globals->get_time_params()->getMjd(),
globals->get_time_params()->getLst(),
0.0 );
2000-07-08 05:09:24 +00:00
globals->set_ephem( ephem );
// TODO: move to environment mgr
thesky = new SGSky;
SGPath texture_path(globals->get_fg_root());
texture_path.append("Textures");
texture_path.append("Sky");
for (int i = 0; i < FGEnvironmentMgr::MAX_CLOUD_LAYERS; i++) {
2002-12-21 05:01:51 +00:00
SGCloudLayer * layer = new SGCloudLayer(texture_path.str());
thesky->add_cloud_layer(layer);
}
SGPath sky_tex_path( globals->get_fg_root() );
2000-03-06 23:28:43 +00:00
sky_tex_path.append( "Textures" );
sky_tex_path.append( "Sky" );
thesky->texture_path( sky_tex_path.str() );
// The sun and moon diameters are scaled down numbers of the
// actual diameters. This was needed to fit bot the sun and the
// moon within the distance to the far clip plane.
// Moon diameter: 3,476 kilometers
// Sun diameter: 1,390,000 kilometers
thesky->build( 80000.0, 80000.0,
463.3, 361.8,
2003-06-09 20:24:50 +00:00
globals->get_ephem()->getNumPlanets(),
globals->get_ephem()->getPlanets(),
2003-06-09 20:24:50 +00:00
globals->get_ephem()->getNumStars(),
globals->get_ephem()->getStars() );
// Initialize MagVar model
SGMagVar *magvar = new SGMagVar();
globals->set_mag( magvar );
// kludge to initialize mag compass
// (should only be done for in-flight
// startup)
// update magvar model
globals->get_mag()->update( fgGetDouble("/position/longitude-deg")
* SGD_DEGREES_TO_RADIANS,
fgGetDouble("/position/latitude-deg")
* SGD_DEGREES_TO_RADIANS,
fgGetDouble("/position/altitude-ft")
* SG_FEET_TO_METER,
globals->get_time_params()->getJD() );
double var = globals->get_mag()->get_magvar() * SGD_RADIANS_TO_DEGREES;
fgSetDouble("/instrumentation/heading-indicator/offset-deg", -var);
// airport = new ssgBranch;
// airport->setName( "Airport Lighting" );
// lighting->addKid( airport );
// build our custom render states
globals->get_renderer()->build_states();
// pass control off to the master event handler
fgOSMainLoop();
// we never actually get here ... but to avoid compiler warnings,
// etc.
return false;
1998-04-21 17:02:27 +00:00
}
// end of main.cxx