1
0
Fork 0
flightgear/src/Scenery/tilemgr.cxx

529 lines
17 KiB
C++
Raw Normal View History

// tilemgr.cxx -- routines to handle dynamic management of scenery tiles
//
// Written by Curtis Olson, started January 1998.
//
// Copyright (C) 1997 Curtis L. Olson - http://www.flightgear.org/~curt
//
// 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
2006-02-21 01:16:04 +00:00
// Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
//
// $Id$
1998-01-07 23:50:01 +00:00
#ifdef HAVE_CONFIG_H
# include <config.h>
#endif
1998-04-03 22:09:02 +00:00
#include <plib/ssg.h>
1998-01-07 23:50:01 +00:00
2000-02-15 03:30:01 +00:00
#include <simgear/constants.h>
2000-02-16 23:01:03 +00:00
#include <simgear/debug/logstream.hxx>
#include <simgear/math/point3d.hxx>
#include <simgear/math/polar3d.hxx>
2000-09-27 20:16:22 +00:00
#include <simgear/math/sg_geodesy.hxx>
2000-02-16 23:01:03 +00:00
#include <simgear/math/vector.hxx>
#include <simgear/structure/exception.hxx>
2003-05-15 21:35:51 +00:00
#include <simgear/scene/model/modellib.hxx>
#include <simgear/scene/model/shadowvolume.hxx>
#include <Main/globals.hxx>
#include <Main/fg_props.hxx>
#include <Main/viewer.hxx>
#include <Scripting/NasalSys.hxx>
#include "newcache.hxx"
#include "scenery.hxx"
#include "tilemgr.hxx"
#define TEST_LAST_HIT_CACHE
1998-09-08 15:05:10 +00:00
#if defined(ENABLE_THREADS)
SGLockedQueue<FGTileEntry *> FGTileMgr::attach_queue;
SGLockedQueue<FGDeferredModel *> FGTileMgr::model_queue;
#else
queue<FGTileEntry *> FGTileMgr::attach_queue;
2001-05-21 20:44:59 +00:00
queue<FGDeferredModel *> FGTileMgr::model_queue;
#endif // ENABLE_THREADS
queue<FGTileEntry *> FGTileMgr::delete_queue;
bool FGTileMgr::tile_filter = true;
extern SGShadowVolume *shadows;
// Constructor
FGTileMgr::FGTileMgr():
state( Start ),
current_tile( NULL ),
vis( 16000 )
{
}
1998-01-24 00:03:27 +00:00
// Destructor
FGTileMgr::~FGTileMgr() {
}
1998-01-07 23:50:01 +00:00
// Initialize the Tile Manager subsystem
int FGTileMgr::init() {
2001-03-24 06:03:11 +00:00
SG_LOG( SG_TERRAIN, SG_INFO, "Initializing Tile Manager subsystem." );
2001-05-21 20:44:59 +00:00
tile_cache.init();
#if 0
// instead it's just a lot easier to let any pending work flush
// through, rather than trying to arrest the queue and nuke all
// the various work at all the various stages and get everything
// cleaned up properly.
2001-05-21 20:44:59 +00:00
while ( ! attach_queue.empty() ) {
attach_queue.pop();
}
while ( ! model_queue.empty() ) {
#if defined(ENABLE_THREADS)
2002-12-07 02:26:50 +00:00
FGDeferredModel* dm = model_queue.pop();
2001-05-21 20:44:59 +00:00
#else
FGDeferredModel* dm = model_queue.front();
model_queue.pop();
#endif
delete dm;
}
loader.reinit();
#endif
state = Inited;
previous_bucket.make_bad();
current_bucket.make_bad();
longitude = latitude = -1000.0;
2002-12-07 02:26:50 +00:00
return 1;
1998-01-07 23:50:01 +00:00
}
// schedule a tile for loading
void FGTileMgr::sched_tile( const SGBucket& b, const bool is_inner_ring ) {
// see if tile already exists in the cache
FGTileEntry *t = tile_cache.get_tile( b );
if ( t == NULL ) {
// make space in the cache
while ( (int)tile_cache.get_size() > tile_cache.get_max_cache_size() ) {
long index = tile_cache.get_oldest_tile();
if ( index >= 0 ) {
FGTileEntry *old = tile_cache.get_tile( index );
shadows->deleteOccluderFromTile( (ssgBranch *) old->get_terra_transform() );
old->disconnect_ssg_nodes();
delete_queue.push( old );
tile_cache.clear_entry( index );
} else {
// nothing to free ?!? forge ahead
break;
}
}
// create a new entry
FGTileEntry *e = new FGTileEntry( b );
// insert the tile into the cache
2002-12-07 02:26:50 +00:00
if ( tile_cache.insert_tile( e ) ) {
// Schedule tile for loading
loader.add( e );
} else {
// insert failed (cache full with no available entries to
// delete.) Try again later
delete e;
}
} else {
t->set_inner_ring( is_inner_ring );
}
}
// schedule a needed buckets for loading
void FGTileMgr::schedule_needed( double vis, const SGBucket& curr_bucket) {
// sanity check (unfortunately needed!)
if ( longitude < -180.0 || longitude > 180.0
|| latitude < -90.0 || latitude > 90.0 )
{
SG_LOG( SG_TERRAIN, SG_ALERT,
"Attempting to schedule tiles for bogus lon and lat = ("
<< longitude << "," << latitude << ")" );
SG_LOG( SG_TERRAIN, SG_ALERT,
"This is a FATAL error. Exiting!" );
exit(-1);
}
SG_LOG( SG_TERRAIN, SG_INFO,
"scheduling needed tiles for " << longitude << " " << latitude );
// vis = 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
double tile_width = curr_bucket.get_width_m();
double tile_height = curr_bucket.get_height_m();
// cout << "tile width = " << tile_width << " tile_height = "
// << tile_height << endl;
xrange = (int)(vis / tile_width) + 1;
yrange = (int)(vis / tile_height) + 1;
2004-01-26 22:44:27 +00:00
if ( xrange < 1 ) { xrange = 1; }
if ( yrange < 1 ) { yrange = 1; }
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
// note * 2 at end doubles cache size (for fdm and viewer)
tile_cache.set_max_cache_size( (2*xrange + 2) * (2*yrange + 2) * 2 );
// cout << "xrange = " << xrange << " yrange = " << yrange << endl;
// cout << "max cache size = " << tile_cache.get_max_cache_size()
// << " current cache size = " << tile_cache.get_size() << endl;
// clear the inner ring flags so we can set them below. This
// prevents us from having "true" entries we aren't able to find
// to get rid of if we teleport a long ways away from the current
// location.
tile_cache.clear_inner_ring_flags();
2001-01-17 23:30:35 +00:00
SGBucket b;
// schedule center tile first so it can be loaded first
b = sgBucketOffset( longitude, latitude, 0, 0 );
sched_tile( b, true );
2001-01-17 23:30:35 +00:00
2001-03-21 23:10:15 +00:00
int x, y;
2001-01-17 23:30:35 +00:00
// schedule next ring of 8 tiles
2001-03-21 23:10:15 +00:00
for ( x = -1; x <= 1; ++x ) {
2002-12-07 02:26:50 +00:00
for ( y = -1; y <= 1; ++y ) {
if ( x != 0 || y != 0 ) {
b = sgBucketOffset( longitude, latitude, x, y );
sched_tile( b, true );
2002-12-07 02:26:50 +00:00
}
}
2001-01-17 23:30:35 +00:00
}
2002-12-07 02:26:50 +00:00
2001-01-17 23:30:35 +00:00
// schedule remaining tiles
2001-03-21 23:10:15 +00:00
for ( x = -xrange; x <= xrange; ++x ) {
2002-12-07 02:26:50 +00:00
for ( y = -yrange; y <= yrange; ++y ) {
if ( x < -1 || x > 1 || y < -1 || y > 1 ) {
SGBucket b = sgBucketOffset( longitude, latitude, x, y );
sched_tile( b, false );
2002-12-07 02:26:50 +00:00
}
}
}
}
void FGTileMgr::initialize_queue()
{
// First time through or we have teleported, initialize the
// system and load all relavant tiles
SG_LOG( SG_TERRAIN, SG_INFO, "Initialize_queue(): Updating Tile list for "
<< current_bucket );
// cout << "tile cache size = " << tile_cache.get_size() << endl;
// wipe/initialize tile cache
// tile_cache.init();
previous_bucket.make_bad();
// build the local area list and schedule tiles for loading
// start with the center tile and work out in concentric
// "rings"
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
schedule_needed(visibility_meters, current_bucket);
// do we really want to lose this? CLO
#if 0
// Now force a load of the center tile and inner ring so we
// have something to see in our first frame.
int i;
for ( i = 0; i < 9; ++i ) {
if ( load_queue.size() ) {
2001-03-24 06:03:11 +00:00
SG_LOG( SG_TERRAIN, SG_DEBUG,
"Load queue not empty, loading a tile" );
2000-12-13 20:36:04 +00:00
SGBucket pending = load_queue.front();
load_queue.pop_front();
load_tile( pending );
}
}
#endif
}
/**
* return current status of queues
*
*/
bool FGTileMgr::all_queues_empty() {
return attach_queue.empty() && model_queue.empty();
}
/**
* Update the various queues maintained by the tilemagr (private
* internal function, do not call directly.)
*/
void FGTileMgr::update_queues()
{
// load the next model in the load queue. Currently this must
// happen in the render thread because model loading can trigger
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
// texture loading which involves use of the opengl api. Skip any
// models belonging to not loaded tiles (i.e. the tile was removed
// before we were able to load some of the associated models.)
if ( !model_queue.empty() ) {
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
bool processed_one = false;
while ( model_queue.size() > 200 || processed_one == false ) {
processed_one = true;
if ( model_queue.size() > 200 ) {
SG_LOG( SG_TERRAIN, SG_INFO,
"Alert: catching up on model load queue" );
}
// cout << "loading next model ..." << endl;
// load the next tile in the queue
#if defined(ENABLE_THREADS)
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
FGDeferredModel* dm = model_queue.pop();
#else
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
FGDeferredModel* dm = model_queue.front();
model_queue.pop();
#endif
2002-12-07 02:26:50 +00:00
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
// only load the model if the tile still exists in the
// tile cache
FGTileEntry *t = tile_cache.get_tile( dm->get_bucket() );
if ( t != NULL ) {
ssgTexturePath( (char *)(dm->get_texture_path().c_str()) );
try {
ssgEntity *obj_model =
globals->get_model_lib()->load_model( ".",
dm->get_model_path(),
globals->get_props(),
globals->get_sim_time_sec(),
dm->get_cache_state(),
new FGNasalModelData );
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
if ( obj_model != NULL ) {
dm->get_obj_trans()->addKid( obj_model );
shadows->addOccluder( (ssgBranch *) obj_model->getParent(0),
SGShadowVolume::occluderTypeTileObject,
(ssgBranch *) dm->get_tile()->get_terra_transform());
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
}
} catch (const sg_io_exception& exc) {
string m(exc.getMessage());
m += " ";
m += exc.getLocation().asString();
SG_LOG( SG_ALL, SG_ALERT, m );
} catch (const sg_exception& exc) { // XXX may be redundant
SG_LOG( SG_ALL, SG_ALERT, exc.getMessage());
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
}
dm->get_tile()->dec_pending_models();
2002-12-07 02:26:50 +00:00
}
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
delete dm;
2002-12-07 02:26:50 +00:00
}
}
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
// Notify the tile loader that it can load another tile
loader.update();
if ( !attach_queue.empty() ) {
#if defined(ENABLE_THREADS)
2002-12-07 02:26:50 +00:00
FGTileEntry* e = attach_queue.pop();
#else
2002-12-07 02:26:50 +00:00
FGTileEntry* e = attach_queue.front();
attach_queue.pop();
#endif
2002-12-07 02:26:50 +00:00
e->add_ssg_nodes( globals->get_scenery()->get_terrain_branch(),
globals->get_scenery()->get_gnd_lights_root(),
globals->get_scenery()->get_vasi_lights_root(),
2002-12-07 02:26:50 +00:00
globals->get_scenery()->get_rwy_lights_root(),
globals->get_scenery()->get_taxi_lights_root() );
// cout << "Adding ssg nodes for "
}
if ( !delete_queue.empty() ) {
// cout << "delete queue = " << delete_queue.size() << endl;
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
bool processed_one = false;
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
while ( delete_queue.size() > 30 || processed_one == false ) {
processed_one = true;
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
if ( delete_queue.size() > 30 ) {
// uh oh, delete queue is blowing up, we aren't clearing
// it fast enough. Let's just panic, well not panic, but
// get real serious and agressively free up some tiles so
// we don't explode our memory usage.
SG_LOG( SG_TERRAIN, SG_ALERT,
"Warning: catching up on tile delete queue" );
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
}
Make a subtle change to tile loading/unloading policy in order to make the tile paging system much more robust when position change is very rapid and sporadic. Recall that we must load 3d models in the main render thread because model loading can trigger opengl calls (i.e. with texture loading) and all opengl calls *must* happen in the main render thread. To accomplish this we load the base tile in the pager thread and build a work queue of external models that need to be loaded. We never allow a tile to be paged out of the tile cache until all it's pending model loads are complete. However, when changing position very rapidly, we can quickly create a huge backlog of pending model loads because we are changing positions faster than we can load the associated models for the existing tiles. The end result is that tiles that are long out of range can't be removed because there is still a huge backlog of pending model load requests and memory blows up. This change being committed allows the tile paging system to remove tiles if they are out of range, even when there are pending models to load. The model loading code in the render thread can now check to see if the tile exists and discard any model load request for tiles that no longer exist. This situation should never occur in normal operation, but could occur in "contrived" situations where an external script was rapidly changing the simulator position to then be able to query FG terrain height, and doing this for a large number of points that are distributed across a large area.
2004-09-15 15:52:05 +00:00
FGTileEntry* e = delete_queue.front();
if ( e->free_tile() ) {
delete_queue.pop();
delete e;
}
}
}
}
// given the current lon/lat (in degrees), fill in the array of local
// chunks. If the chunk isn't already in the cache, then read it from
// disk.
int FGTileMgr::update( double visibility_meters )
{
SGLocation *location = globals->get_current_view()->getSGLocation();
return update( location, visibility_meters );
}
int FGTileMgr::update( SGLocation *location, double visibility_meters )
{
SG_LOG( SG_TERRAIN, SG_DEBUG, "FGTileMgr::update()" );
longitude = location->getLongitude_deg();
latitude = location->getLatitude_deg();
// add 1.0m to the max altitude to give a little leeway to the
// ground reaction code.
altitude_m = location->getAltitudeASL_ft() * SG_FEET_TO_METER + 1.0;
// if current altitude is apparently not initialized, set max
// altitude to something big.
if ( altitude_m < -1000 ) {
altitude_m = 10000;
}
// SG_LOG( SG_TERRAIN, SG_DEBUG, "FGTileMgr::update() for "
// << longitude << " " << latatitude );
current_bucket.set_bucket( longitude, latitude );
// SG_LOG( SG_TERRAIN, SG_DEBUG, "Updating tile list for "
// << current_bucket );
fgSetInt( "/environment/current-tile-id", current_bucket.gen_index() );
// do tile load scheduling.
// Note that we need keep track of both viewer buckets and fdm buckets.
if ( state == Running ) {
2002-12-07 02:26:50 +00:00
SG_LOG( SG_TERRAIN, SG_DEBUG, "State == Running" );
if (!(current_bucket == previous_bucket )) {
// We've moved to a new bucket, we need to schedule any
// needed tiles for loading.
SG_LOG( SG_TERRAIN, SG_INFO, "FGTileMgr::update()" );
2002-12-07 02:26:50 +00:00
schedule_needed(visibility_meters, current_bucket);
}
} else if ( state == Start || state == Inited ) {
2002-12-07 02:26:50 +00:00
SG_LOG( SG_TERRAIN, SG_INFO, "State == Start || Inited" );
// initialize_queue();
2002-12-07 02:26:50 +00:00
state = Running;
2002-12-07 02:26:50 +00:00
// load the next tile in the load queue (or authorize the next
// load in the case of the threaded tile pager)
loader.update();
}
update_queues();
// save bucket...
previous_bucket = current_bucket;
return 1;
1998-09-08 15:05:10 +00:00
}
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
// timer event driven call to scheduler for the purpose of refreshing the tile timestamps
void FGTileMgr::refresh_view_timestamps() {
SG_LOG( SG_TERRAIN, SG_INFO,
2002-12-07 02:26:50 +00:00
"Refreshing timestamps for " << current_bucket.get_center_lon()
<< " " << current_bucket.get_center_lat() );
if ( longitude >= -180.0 && longitude <= 180.0
&& latitude >= -90.0 && latitude <= 90.0 )
{
schedule_needed(fgGetDouble("/environment/visibility-m"), current_bucket);
}
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
}
void FGTileMgr::prep_ssg_nodes( SGLocation *location, float vis ) {
// traverse the potentially viewable tile list and update range
// selector and transform
float *up = location->get_world_up();
FGTileEntry *e;
tile_cache.reset_traversal();
const double *vp = location->get_absolute_view_pos();
Point3D viewpos(vp[0], vp[1], vp[2]);
while ( ! tile_cache.at_end() ) {
// cout << "processing a tile" << endl;
2002-12-07 02:26:50 +00:00
if ( (e = tile_cache.get_current()) ) {
e->prep_ssg_node( viewpos, up, vis);
} else {
2002-12-07 02:26:50 +00:00
SG_LOG(SG_INPUT, SG_ALERT, "warning ... empty tile in cache");
}
2002-12-07 02:26:50 +00:00
tile_cache.next();
}
}
Major viewer-code overhaul from Jim Wilson: Description: This update includes the new viewer interface as proposed by David M. and a first pass at cleaning up the viewer/view manager code by Jim W. Note that I have dropped Main/viewer_lookat.?xx and Main/viewer_rph.?xx and modified the Makefile.am accordingly. Detail of work: Overall: The code reads a little easier. There are still some unnecessary bits in there and I'd like to supplement the comments in the viewer.hxx with a tiny bit on each interface group and what the groupings mean (similar but briefer than what you emailed me the other day). I tried not to mess up the style, but there is an occasional inconsistency. In general I wouldn't call it done (especially since there's no tower yet! :)), but I'd like to get this out there so others can comment, and test. In Viewer: The interface as you suggested has been implemented. Basically everything seems to work as it did visually. There is no difference that I can see in performance, although some things might be a tiny bit faster. I've merged the lookat and rph (pilot view) code into the recalc for the viewer. There is still some redundancy between the two, but a lot has been removed. In some cases I've taken some code that we'd likely want to inline anyway and left it in there in duplicate. You'll see that the code for both looks a little cleaner. I need to take a closer look at the rotations in particular. I've cleaned up a little there, but I suspect more can be done to streamline this. The external declaration to the Quat_mat in mouse.cxx has been removed. IMHO the quat doesn't serve any intrinsic purpose in mouse.cxx, but I'm not about to rip it out. It would seem that there more conventional ways to get spherical data that are just as fast. In any case all the viewer was pulling from the quat matrix was the pitch value so I modified mouse.cxx to output to our pitchOffset input and that works fine. I've changed the native values to degrees from radians where appropriate. This required a conversion from degrees to radians in a couple modules that access the interface. Perhaps we should add interface calls that do the conversion, e.g. a getHeadingOffset_rad() to go along with the getHeadingOffset_deg(). On the view_offset (now headingOffset) thing there are two entry points because of the ability to instantly switch views or to scroll to a new view angle (by hitting the numeric keys for example). This leaves an anomaly in the interface which should be resolved by adding "goal" settings to the interface, e.g. a setGoalHeadingOffset_deg(), setGoalPitchOffset_deg(), etc. Other than these two issues, the next step here will be to look at some further optimizations, and to write support code for a tower view. That should be fairly simple at this point. I was considering creating a "simulated tower view" or "pedestrian view" that defaulted to a position off to the right of whereever the plane is at the moment you switch to the tower view. This could be a fall back when we don't have an actual tower location at hand (as would be the case with rural airports). ViewManager: Basically all I did here was neaten things up by ripping out excess crap and made it compatible as is with the new interface. The result is that viewmanager is now ready to be developed. The two preexisting views are still hardcoded into the view manager. The next step would be to design configuration xml (eg /sim/view[x]/config/blahblah) that could be used to set up as many views as we want. If we want to take the easy way out, we might want to insist that view[0] be a pilot-view and have viewmanager check for that.
2002-03-20 17:43:28 +00:00
bool FGTileMgr::set_tile_filter( bool f ) {
bool old = tile_filter;
tile_filter = f;
return old;
}
int FGTileMgr::tile_filter_cb( ssgEntity *, int )
{
return tile_filter ? 1 : 0;
}
bool FGTileMgr::scenery_available(double lat, double lon, double range_m)
{
// sanity check (unfortunately needed!)
if ( lon < -180.0 || lon > 180.0 || lat < -90.0 || lat > 90.0 )
return false;
SGBucket bucket(lon, lat);
FGTileEntry *te = tile_cache.get_tile(bucket);
if (!te || !te->is_loaded())
return false;
// Traverse all tiles required to be there for the given visibility.
// This uses exactly the same algorithm like the tile scheduler.
double tile_width = bucket.get_width_m();
double tile_height = bucket.get_height_m();
int xrange = (int)fabs(range_m / tile_width) + 1;
int yrange = (int)fabs(range_m / tile_height) + 1;
for ( int x = -xrange; x <= xrange; ++x ) {
for ( int y = -yrange; y <= yrange; ++y ) {
// We have already checked for the center tile.
if ( x != 0 || y != 0 ) {
SGBucket b = sgBucketOffset( lon, lat, x, y );
FGTileEntry *te = tile_cache.get_tile(b);
if (!te || !te->is_loaded())
return false;
}
}
}
// Survived all tests.
return true;
}