[Flightgear-devel] valgrind: diff no 1

2008-01-18 Thread till busch
hi all, i've started to run fg through valgrind. i found this to be a nice option for getting an overview over the code in flightgear. i plan to prepare some patches to fix various issues reported by valgrind. this is the first in (i hope) a series of several patches yet to follow. i couldn't

Re: [Flightgear-devel] valgrind: diff no 1

2008-01-19 Thread till busch
hi melchior, On Saturday 19 January 2008, Melchior FRANZ wrote: * [EMAIL PROTECTED] -- Saturday 19 January 2008: and i am starting to do general clean-ups and optimizations (e.g. replacing serveral if - else if - else if... with switch) Don't do that! Switch *may* sometimes be the better

[Flightgear-devel] valgrind diff no 2 and 3

2008-01-21 Thread till busch
hi all, i have continued my random optimizations to flightgear (and simgear, this time). changes in the attached diffs = in simgear: * removed _tex_cache from SGMaterial (osg takes care of texture caching) * made textures load on demand when SGMaterial::get_state() is first

Re: [Flightgear-devel] valgrind diff no 2 and 3

2008-01-22 Thread till busch
4:03 PM, till busch wrote: hi all, Hi Till, I don't know all the nuances of the OSG branch and OSG usage, but originally the common material library textures were all loaded at start time so they'd be available at any time and not hit the fps like they would if they needed to be loaded

Re: [Flightgear-devel] valgrind diff no 2 and 3

2008-01-23 Thread till busch
On Tuesday 22 January 2008, Andy Ross wrote: till busch wrote: * f_interpolate in NasalSys was leaky (valgrind) This leak is real, but the patch isn't legal C++, at least as of the last time I read the standard. You can't initialize a stack array with a dynamic value, it has

Re: [Flightgear-devel] valgrind diff no 2 and 3

2008-01-24 Thread till busch
* that the new code doesn't impact performance. if you have a good idea, could you give me a hint? otherwise we might just put it on cvs and see if people notice performance loss? thanks, -till On Tuesday 22 January 2008, till busch wrote: hi curt, why i think that this shouldn't impact performance

Re: [Flightgear-devel] trees

2008-01-28 Thread till busch
hi since i'm already much into memory-stuff i'll add my 2 cents... On Sunday 27 January 2008, Curtis Olson wrote: ... Would it be possible to compute the locations of the trees when the tile is loaded (in the load thread) and always have these structures available when needed? Now that we

Re: [Flightgear-devel] trees

2008-01-30 Thread till busch
hi stuart, i have missed you on irc today. and i wanted to tell you that indeed the second patch is much better in terms of memory usage. it saves 80 megs of allocations compared to your first patch :-) and considering it looks even prettier than the first one... congratulations! -till On

[Flightgear-devel] apt.dat runways, taxiways and atc

2008-02-01 Thread till busch
hi, i have reworked the apt.dat-loading code. everything works fine so far, but i have two questions: 1. we currently have runways and taxiways in one FGRunwayList. imho these are very different things. i'd like to make two separate lists. are the taxiways from apt.dat used at all? maybe for

Re: [Flightgear-devel] apt.dat runways, taxiways and atc

2008-02-02 Thread till busch
On Saturday 02 February 2008, Thomas Förster wrote: Am Freitag 01 Februar 2008 schrieb till busch: hi, i have reworked the apt.dat-loading code. everything works fine so far, but i have two questions: 1. we currently have runways and taxiways in one FGRunwayList. imho these are very

[Flightgear-devel] [patch] fix FGNewCache to really delete old tiles

2008-02-21 Thread till busch
hi, as discussed with tim on irc. here comes a quick fix for memory-hungry tile manager. due to bugs in FGNewCache old tiles where never deleted. i left the timestamp-updates in the cull-traversal. but imho things work just as well when timestamps are updated in FGNewCache::insert_tile() and

Re: [Flightgear-devel] [patch] fix FGNewCache to really delete old tiles

2008-02-21 Thread till busch
-traversal way. (i have tested this locally and it worked fine) regards, -till ... and i thought people here would read code ... On Thursday 21 February 2008, Curtis Olson wrote: On Thu, Feb 21, 2008 at 11:11 AM, till busch [EMAIL PROTECTED] wrote: hi, as discussed with tim on irc. here

Re: [Flightgear-devel] [patch] fix FGNewCache to really delete old tiles

2008-02-21 Thread till busch
hi curt, sorry for being rude. your asking (though not specific enough, imho) had a valid point as andy pointed out (in irc). the variable naming was bad. apparently in the transition from plib the wrong branch of the condition got in. this patch now fixes the variable names. (and adds

Re: [Flightgear-devel] [patch] fix FGNewCache to really delete old tiles

2008-02-21 Thread till busch
hi curt, i fear i'm not the right person to answer this. (at least not without doing some time-consuming research). i don't *believe* the problem you describe could be related to the deletion of old tiles. what i can tell for sure is that the correct tiles get deleted. i had lots of debugging

Re: [Flightgear-devel] Frankfurt (EDDF) scenery SVN now available

2008-03-10 Thread till busch
hi christian, sorry. that was my fault -- skimming instead of reading. the scenery looks very nice to me. i think i miss a building with a pyramid-shaped roof: messeturm?. (though my memory may not be the best, since i was in frankfurt only twice). cheers, -till On Monday 10 March 2008,

Re: [Flightgear-devel] Frankfurt (EDDF) scenery SVN now available

2008-03-10 Thread till busch
On Monday 10 March 2008, Frederic Bouvier wrote: Quoting till busch : hi christian, sorry. that was my fault -- skimming instead of reading. the scenery looks very nice to me. i think i miss a building with a pyramid-shaped roof: messeturm?. (though my memory may not be the best

Re: [Flightgear-devel] Frankfurt (EDDF) scenery SVN now available

2008-03-10 Thread till busch
hi, would you mind adding a .tar.gz or a zipfile for easier download? thanks, -till On Sunday 09 March 2008, Christian Schmitt wrote: Hi guys, As some of you might already know, we are working on improving the scenery of Germany. Right now we are 3 people and I want to announce the first

Re: [Flightgear-devel] MP forest fire model!

2008-03-15 Thread till busch
hi georg, i am currently working on a patch that adds model-paging. models (ai, multiplayer and, to some degree, scenery) will be loaded in a different thread. this is more or less what you describe. some successful testing has been done already. so i hope to get it integrated into cvs soon.

[Flightgear-devel] model-paging patch - testers wanted

2008-03-19 Thread till busch
hi all, as noted earlier in another thread, i am working on a model-paging patch. i need testers! if you want flightgear multiplayer to not pause when other pilots join, you are welcome to test my patch. some people are running flightgear with my patch alredy -- feedback is very positive so

Re: [Flightgear-devel] model-paging patch - testers wanted

2008-03-22 Thread till busch
-in. i hope the changes don't break anything! cheers, -till On Saturday 22 March 2008, Tim Moore wrote: till busch wrote: | hi all, | | as noted earlier in another thread, i am working on a model-paging patch. I've checked this in, along with Till's further changes based on my feedback. Thank

[Flightgear-devel] elevation tool

2008-04-01 Thread till busch
hi, At Jon's request (on irc) i have hacked up an elevation tool. It consists of a stock osgviewer and the FlightGear scenery system. Basically you feed it lat/lon pairs on stdin and it will output terrain (+model) elevation at the given point. Right now this is very ugly, since i had to pull

[Flightgear-devel] make osgviewer the default choice in configure

2008-04-01 Thread till busch
hi all, since we want to get rid of other viewer types in near future, i patched configure.ac and src/Main/Makefile.am to make osgviewer the default. please double-check and commit. http://flight.bux.at/0001-make-osgviewer-the-default-choice-in-configure.patch when this is in, the same change

Re: [Flightgear-devel] AI Aircraft Models

2008-04-10 Thread till busch
On Wednesday 09 April 2008, Stuart Buchanan wrote: --- On Wed, 9/4/08, Melchior FRANZ wrote: Do we really want MP support for all aircraft in the base package, at a cost of an extra 200 MB of data? Wrappers are fine (like Vivian described), but do we want a complete concorde.ac with all

Re: [Flightgear-devel] Message time and fonts

2008-04-17 Thread till busch
hi adam, this is currently not controlled by a property. you can find the timeout in Nasal/screen.nas (line 184). log = window.new(nil, -30, 10, 10); you could also specify the font there (untested): log.font = font; i think that changing the font size is not easily possible. cheers, -till

[Flightgear-devel] [patch] fix pick animations

2008-04-23 Thread till busch
hi, this is a small (-1/+3) patch to fix pick animations on scenery objects. since picking apparently doesn't care for polygon offsets, the objects got into the picklist in the wrong way. now, no matter if the highlight group or the normal group gets hit first, the callback will fire. ... i

[Flightgear-devel] [patch] database pager calm-down

2008-04-25 Thread till busch
hi, tim recently noticed the database pager was repeatedly loading and unloading the same objects. he also tracked down the problem to missing bounding sphere information in osgDB::PagedLOD. this is a simplicistic approach to fix this: SGPagedLOD will now remember whatever value it sees for

Re: [Flightgear-devel] [patch] database pager calm-down

2008-04-25 Thread till busch
indeed this was hardly noticable (in terms of stuttering) before robert made the deletion part of the database pager more aggressive. so we didn't discover it during initial testing of my model-paging patch. in the second patch i had aeroto and his movies in mind. from a fixed position, after

Re: [Flightgear-devel] [PATCH] fix fixes ;)

2008-04-28 Thread till busch
hi, this is a good thing. it was on the todo for a long time. please someone commit this. - till On Saturday 26 April 2008, Csaba Halász wrote: Hi! Attached patch adds support for multiple FIXes with the same name. Applies to both branches. New functionality is in query_and_offset, which

Re: [Flightgear-devel] [patch] Get rid of pthread

2008-04-30 Thread till busch
salut Benoît, your approach sounds reasonable to me. having all threads-related functionality in OpenThreads is the way to go. imho there is no good reason for maintaining another thread-wrapper in sg. a quick scan shows following classes in sg: class SGThread class SGMutex class SGPthreadCond

[Flightgear-devel] [patch] move most scenery-related code to simgear

2008-05-08 Thread till busch
hi, i have prepared a patch that moves most scenery-related code to simgear. namely following classes are moved to simgear/scene/tgdb/: ReaderWriterSTG TileEntry TileCache (renamed from FGNewCache) this move was needed to allow independent apps to use flightgear scenery. the app i made this

Re: [Flightgear-devel] [patch] move most scenery-related code to simgear

2008-05-15 Thread till busch
On Thursday 15 May 2008, Tim Moore wrote: I've committed this. Thanks. thanks a lot. small clean-ups will probably follow. 0002-fg-add-an-elevation-util-0.0.4.patch I haven't committed this yet but will look at it soon. don't look to closely, the code is quite ugly. also i have a little

[Flightgear-devel] DatabasePager changes

2008-05-23 Thread till busch
hi all, robert has done some major changes to osgDB::DatabasePager. as he correctly noted this affects us. for now, please use an osg-version from before 2008-05-22. from robert on osg-users: For the vast majority of users this won't be any issue as subclassing from the DatabasePager, and

Re: [Flightgear-devel] Shaders (Mac, nVidia 7300GT, latest CVS)

2009-08-31 Thread till busch
hi james, this is not at all how it should look like. i believe that either something is wrong with lighting values in gl_FrontLightModelProduct / gl_LightSource[0] or smoothstep() does not work correctly on your system. i'm working on improving the shaders. i'll try to make a shader without