Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
Hi Durk Sent: 25 September 2007 19:57 To: FlightGear developers discussions Subject: Re: [Flightgear-devel] FGFS 0.9.11 release candidate two Hi Vivian, On Tuesday 25 September 2007 17:39, Vivian Meazza wrote: Here's a short burst of the output of Fred's profiling code using Windows XP, with the source code complied using MSVC8; Subsystem Timing Alert : 19000 fx Subsystem Timing Alert : 15000 fx Subsystem Timing Alert : 14000 replay Subsystem Timing Alert : 11000 replay Subsystem Timing Alert : 15000 replay Subsystem Timing Alert : 14000 replay Subsystem Timing Alert : 12000 fx Subsystem Timing Alert : 18000 replay Subsystem Timing Alert : 17000 fx Subsystem Timing Alert : 32000 input Subsystem Timing Alert : 19000 fx Subsystem Timing Alert : 11000 fx Subsystem Timing Alert : 11000 fx Subsystem Timing Alert : 11000 fx Subsystem Timing Alert : 63000 ai_model Subsystem Timing Alert : 12000 fx Subsystem Timing Alert : 16000 fx Visually, the stutters do not seem to coincide timewise with the output from this code, so I'm not _totally_ convinced that we are getting anything very useful. Is there anything else I can do to help? Thanks for the report. Looking at your report, I'm not seeing that many alarming timing problems (yet). Looks like the longest one is a 63 ms ai_model one. A few questions regarding your testing conditions: - How much time since program start had passed since taking these? - Which aircraft did you try. FWIW, I've been testing using the SenecaII-jsbsim. Initially the timing problems aren't that bad, but after about leaving FlightGear running for about 3 hours, I'm getting very significant timing problems, the last one before I killed the program being close to 45 seconds. I got the following timing errors: Subsystem Timing Alert : 12553567 replay Subsystem Timing Alert : 1953303 controller Subsystem Timing Alert : 1955229 environment Subsystem Timing Alert : 6890232 replay Subsystem Timing Alert : 39674 controller Subsystem Timing Alert : 41267 environment Subsystem Timing Alert : 108117 replay I just don't understand what's going on here... That was using the Sea Vixen, and I only ran that particular test for 5 mins or so, and the report was just an extract of more of the same. I agree with your observation that there were no particular problems shown up here, neither was there much visible on the screen. But as I said there was a little stutter visible, but didn't seem to tie up with any output Subsystem Timing Alert. I have run a much longer test, but without very different results to date. I'll test some more. Vivian - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
On mer 26 septembre 2007, Vivian Meazza wrote: With Linux Fedora 6 FG today cvs with plib Athlon AMD 32 bit GPU nvidia 7800GS I get with Crusader, during an airborne from LFTH to LFMV via LFPO (paris Orly with his huge PARIS and PARIS-ORLY Scenery) with MP connected. the following log (joined document) I do not get some significant stutters , fps from 30 to 80 and close to Paris from 10 to 20. I hope that test, will be useful. -- Gérard http://perso.orange.fr/GRTux/ Object Phalanx-3-Pos not found Object Sea-Sparrow-2-Pos not found Object Sea-Sparrow-3-Pos not found Object Vehicle-Pos not found Object Vehicle-Pos not found Chat [Aerotro] Welcome to Aerotro Chat [Aerotro] this is version v0.9.10 (Experimental features: LazyRelay) Chat [Aerotro] using protocol version v1.1 Chat [Aerotro] Crouze is now online, using Chat [Aerotro] Aircraft/F-8E-Crusader/Models/F8E-model.xml Chat [AeChat [Aerotro] Aircraft/f16/Models/f16.xml Subsystem Timing Alert : 19630 replay Initializing Nasal Electrical System Subsystem Timing Alert : 19485 instrument2 Subsystem Timing Alert : 20918 instrumentation Subsystem Timing Alert : 42271 controller Subsystem Timing Alert : 43572 environment Subsystem Timing Alert : 18352 instrument19 Subsystem Timing Alert : 19726 instrumentation Subsystem Timing Alert : 17603 replay Subsystem Timing Alert : 17193 ai_model Subsystem Timing Alert : 14016 ai_model Incidence Action Wings Up fold Action Subsystem Timing Alert : 23774 replay Chat [Aerotro] Han is now online, using Chat [Aerotro] Aircraft/f16/Models/baseline.xml Subsystem Timing Alert : 10336 replay Incidence Action Wings Down Subsystem Timing Alert : 11220 instrument19 Subsystem Timing Alert : 12787 instrumentation Subsystem Timing Alert : 10390 controller Subsystem Timing Alert : 11664 environment Subsystem Timing Alert : 10317 instrument19 Subsystem Timing Alert : 11718 instrumentation Chat [Aerotro] Lew is now online, using Chat [Aerotro] Aircraft/f16/Models/f16.xml Subsystem Timing Alert : 25479 replay Subsystem Timing Alert : 10219 ai_model Subsystem Timing Alert : 10846 replay Subsystem Timing Alert : 10716 instrument19 Subsystem Timing Alert : 12142 instrumentation Subsystem Timing Alert : 40440 system1 Subsystem Timing Alert : 41950 systems Subsystem Timing Alert : 11225 instrument1 Subsystem Timing Alert : 12729 instrumentation Chat [Aerotro] Lew is now online, using Chat [Aerotro] Aircraft/f16/Models/f16.xml Subsystem Timing Alert : 10403 gui Subsystem Timing Alert : 14744 replay Subsystem Timing Alert : 22104 controller Subsystem Timing Alert : 23345 environment Subsystem Timing Alert : 14970 replay Subsystem Timing Alert : 11605 replay Subsystem Timing Alert : 11241 ai_model Chat [Aerotro] client 'morteno' has left Subsystem Timing Alert : 12494 system1 Subsystem Timing Alert : 13833 systems Subsystem Timing Alert : 11242 systems Subsystem Timing Alert : 10216 gui Subsystem Timing Alert : 42580 replay Subsystem Timing Alert : 12246 instrument19 Subsystem Timing Alert : 13707 instrumentation Subsystem Timing Alert : 67500 replay Subsystem Timing Alert : 14983 ai_model Subsystem Timing Alert : 11859 replay Subsystem Timing Alert : 10262 gui Subsystem Timing Alert : 10353 gui Chat [Aerotro] jojo is now online, using Chat [Aerotro] Aircraft/c310u3a/Models/c310u3a.xml Subsystem Timing Alert : 11326 gui Subsystem Timing Alert : 11341 gui Subsystem Timing Alert : 11393 instrument12 Subsystem Timing Alert : 12998 instrumentation Subsystem Timing Alert : 11031 gui Subsystem Timing Alert : 11391 replay Subsystem Timing Alert : 43948 ai_model Subsystem Timing Alert : 11066 instrument20 Subsystem Timing Alert : 13392 instrumentation Subsystem Timing Alert : 12507 replay Subsystem Timing Alert : 11353 input Subsystem Timing Alert : 39008 od_gauge Subsystem Timing Alert : 40452 instrumentation Subsystem Timing Alert : 10720 gui Subsystem Timing Alert : 42512 replay Chat [Aerotro] client 'F-VEBR' has left Subsystem Timing Alert : 11179 gui Subsystem Timing Alert : 12021 Traffic Manager Subsystem Timing Alert : 45648 replay Subsystem Timing Alert : 10576 ai_model Subsystem Timing Alert : 14499 ai_model Subsystem Timing Alert : 11380 gui Subsystem Timing Alert : 21932 instrument19 Subsystem Timing Alert : 23346 instrumentation Subsystem Timing Alert : 11226 replay Subsystem Timing Alert : 28247 controller Subsystem Timing Alert : 29479 environment Subsystem Timing Alert : 12417 replay Subsystem Timing Alert : 10554 Traffic Manager Subsystem Timing Alert : 49630 replay Subsystem Timing Alert : 40208 ai_model Subsystem Timing Alert : 43546 replay Chat [Aerotro] Kailing is now online, using Chat [Aerotro] Aircraft/b29/Models/b29-model.xml Subsystem Timing Alert : 51235 ai_model Subsystem Timing Alert : 10979 instrument19 Subsystem Timing Alert : 12430 instrumentation Subsystem Timing Alert : 10133 gui Subsystem Timing Alert : 41212 instrument19 Subsystem Timing Alert : 48479 instrumentation Subsystem Timing Alert :
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
On Wednesday 26 September 2007 18:22, gh.robin wrote: With Linux Fedora 6 FG today cvs with plib Athlon AMD 32 bit GPU nvidia 7800GS I get with Crusader, during an airborne from LFTH to LFMV via LFPO (paris Orly with his huge PARIS and PARIS-ORLY Scenery) with MP connected. the following log (joined document) I do not get some significant stutters , fps from 30 to 80 and close to Paris from 10 to 20. I hope that test, will be useful. Hi Gerard, Thanks for the report. A few questions: How long did you have FlightGear running when you collected these data? Did you have the metar whether fetch enabled? Cheers, Durk - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
Hi Vivian, On Wednesday 26 September 2007 09:30, Vivian Meazza wrote: ... That was using the Sea Vixen, and I only ran that particular test for 5 mins or so, and the report was just an extract of more of the same. I agree with your observation that there were no particular problems shown up here, neither was there much visible on the screen. But as I said there was a little stutter visible, but didn't seem to tie up with any output Subsystem Timing Alert. I have run a much longer test, but without very different results to date. I'll test some more. Thanks. FWIW, I've let FlightGear run today with metar weather fetch disabled. This removes any timing errors related to environment and controller. However, the pauses caused by replay are still there. Instead, I'm now seeing occasional delays in instrumentation; as in Subsystem Timing Alert : 1160337 replay Subsystem Timing Alert : 27720732 instrument5 Subsystem Timing Alert : 27723983 instrumentation I'm hoping that timing problems in all these other subsystems are caused due to the fact that they need to catch up for time lost in replay. In other words, that replay timing errors have a cascading effect on the other system. My next test will be to disable replay and leave flightgear running some more. FWIW, this is my current test setup: .fgfsrc: --airport=EHAM --time-match-local --prop:/environment/params/real-world-weather-fetch=true --fg-scenery=/home/durk/FlightGear-Scenery-0.9.10/ --fg-root=/home/durk/src/FlightGear-0.9/data-cvs #--atlas=socket,out,1,localhost,5500,udp #--nmea=socket,out,1,42.0.1.4,5501,udp --verbose --visibility-miles=25 --geometry=1280x1024 cmd line: ./src/Main/fgfs --aircraft=SenecaII-jsbsim Cheers, Durk - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
On mer 26 septembre 2007, Durk Talsma wrote: On Wednesday 26 September 2007 18:22, gh.robin wrote: With Linux Fedora 6 FG today cvs with plib Athlon AMD 32 bit GPU nvidia 7800GS I get with Crusader, during an airborne from LFTH to LFMV via LFPO (paris Orly with his huge PARIS and PARIS-ORLY Scenery) with MP connected. the following log (joined document) I do not get some significant stutters , fps from 30 to 80 and close to Paris from 10 to 20. I hope that test, will be useful. Hi Gerard, Thanks for the report. A few questions: How long did you have FlightGear running when you collected these data? Did you have the metar whether fetch enabled? Cheers, Durk Hello Durk, The flight was about 30 minutes, first i started with Metar enabled (about 10 minutes) which was changed to scenario fair weather (20 minutes) Regards. -- Gérard http://perso.orange.fr/GRTux/ - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
On Sunday 23 September 2007 13:50, I wrote: Some observations: - Some small pauzes seem to be caused by consecutive calls to the systems controller and environment. Execution times of these systems is variable, between calls, but the execution time of environment seems to be just slightly longer than that of controller: Subsystem Timing Alert : 18150 controller Subsystem Timing Alert : 20519 environment Subsystem Timing Alert : 47917 controller Subsystem Timing Alert : 49514 environment The longer and icreasingly growing pauzes seem to be caused by replay, possibly in interaction with some of the other subsystems. I haven't really found out what is going on here. Just a quick follow-up question: Has anybody tried running FlightGear / plib with the last SimGear checkout (with Fred's profiling code)? The observed timing problems are the major show stopper for the next release (IMHO), so I think we should try to get to the bottom of it. I'm postponing putting out a release candidate until we have some better understanding as to what's going on. I might have another look this weekend, but don't have time right now. Cheers, Durk - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
Durk Talsma Sent: 25 September 2007 07:36 To: FlightGear developers discussions Subject: Re: [Flightgear-devel] FGFS 0.9.11 release candidate two On Sunday 23 September 2007 13:50, I wrote: Some observations: - Some small pauzes seem to be caused by consecutive calls to the systems controller and environment. Execution times of these systems is variable, between calls, but the execution time of environment seems to be just slightly longer than that of controller: Subsystem Timing Alert : 18150 controller Subsystem Timing Alert : 20519 environment Subsystem Timing Alert : 47917 controller Subsystem Timing Alert : 49514 environment The longer and icreasingly growing pauzes seem to be caused by replay, possibly in interaction with some of the other subsystems. I haven't really found out what is going on here. Just a quick follow-up question: Has anybody tried running FlightGear / plib with the last SimGear checkout (with Fred's profiling code)? The observed timing problems are the major show stopper for the next release (IMHO), so I think we should try to get to the bottom of it. I'm postponing putting out a release candidate until we have some better understanding as to what's going on. I might have another look this weekend, but don't have time right now. Here's a short burst of the output of Fred's profiling code using Windows XP, with the source code complied using MSVC8; Subsystem Timing Alert : 19000 fx Subsystem Timing Alert : 15000 fx Subsystem Timing Alert : 14000 replay Subsystem Timing Alert : 11000 replay Subsystem Timing Alert : 15000 replay Subsystem Timing Alert : 14000 replay Subsystem Timing Alert : 12000 fx Subsystem Timing Alert : 18000 replay Subsystem Timing Alert : 17000 fx Subsystem Timing Alert : 32000 input Subsystem Timing Alert : 19000 fx Subsystem Timing Alert : 11000 fx Subsystem Timing Alert : 11000 fx Subsystem Timing Alert : 11000 fx Subsystem Timing Alert : 63000 ai_model Subsystem Timing Alert : 12000 fx Subsystem Timing Alert : 16000 fx Visually, the stutters do not seem to coincide timewise with the output from this code, so I'm not _totally_ convinced that we are getting anything very useful. Is there anything else I can do to help? V. - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
Frederic Bouvier wrote: It appears that it is the replay subsystem that creates long pauses periodically, and sometimes the ai-model subsystem too : ( my traces : ) D : 12000 replay D : 11000 replay D : 17000 instrument20 D : 18000 instrumentation D : 12000 replay D : 12000 replay D : 14000 replay D : 19000 replay D : 16000 replay D : 18000 replay D : 11000 replay D : 22000 input D : 13000 replay D : 17000 replay D : 14000 replay D : 22000 replay D : 18000 replay D : 18000 replay D : 18000 Traffic Manager D : 17000 instrument13 D : 18000 instrumentation D : 17000 replay D : 23000 electrical0 D : 32000 systems D : 18000 replay D : 11000 replay D : 16000 replay D : 11000 replay D : 11000 replay D : 252000 input D : 11000 replay D : 12000 electrical0 D : 15000 systems D : 11000 replay D : 17000 ai_model D : 11000 instrumentation D : 11000 replay D : 14000 replay D : 18000 replay D : 11000 replay D : 12000 replay D : 19000 replay D : 15348000 replay D : 16000 ai_model D : 14000 replay D : 11000 replay D : 12000 replay D : 14000 replay D : 12000 replay D : 15000 properties D : 13000 replay Very long pauses are caused by breakpoints in the debugger regards, -Fred The replay subsystem is *very* slow, that's why there is an option to disable it. Anyway you have times from 12 to 18 ms in replay, I have times from 100 to 300 ms in the nasal code. Gc done: Tue May 01 11:50:37 2007 globals-allocCount=179842 dt=106.609360 Gc done: Tue May 01 11:50:47 2007 globals-allocCount=179842 dt=115.206542 Gc done: Tue May 01 11:50:56 2007 globals-allocCount=179842 dt=112.281589 Gc done: Tue May 01 11:51:06 2007 globals-allocCount=179842 dt=102.708584 Gc done: Tue May 01 11:51:16 2007 globals-allocCount=179842 dt=104.459645 Gc done: Tue May 01 11:51:26 2007 globals-allocCount=179842 dt=118.089031 Gc done: Tue May 01 11:51:36 2007 globals-allocCount=179842 dt=104.458248 -- Gc done: Tue May 01 12:41:50 2007 globals-allocCount=343605 dt=300.275594 Gc done: Tue May 01 12:42:11 2007 globals-allocCount=346632 dt=283.045471 Gc done: Tue May 01 12:42:32 2007 globals-allocCount=346632 dt=282.179160 Gc done: Tue May 01 12:42:53 2007 globals-allocCount=346632 dt=273.818600 Gc done: Tue May 01 12:43:14 2007 globals-allocCount=346632 dt=281.277928 -- On windoze XP, I'm affraid there is too many random factors in each report of the problem (mine included). HJ. - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
Hi Fred, On Sunday 23 September 2007 00:20, Frederic Bouvier wrote: I used a profiler of my own : Excellent! This is almost exactly the type of profiling I hinted at yesterday. I believe that this is a very debugging tool. Can we add a (slightly modified version) of this profiling system to the actual simgear code? It appears that it is the replay subsystem that creates long pauses periodically, and sometimes the ai-model subsystem too : It looks like I have some confirmatory evidence regarding the replay pauses. Yesterday I left FlightGear running overnight, and found this morning that the pauses had increased to multiple seconds. So this morning I reran FlightGear in gdb, and hit ^C when the pauses had become sufficiently long enough. Here's the stack trace: gdb) bt #0 0xb7a4f41c in _int_malloc () from /lib/libc.so.6 #1 0xb7a510c5 in malloc () from /lib/libc.so.6 #2 0xb7bd4477 in operator new () from /usr/lib/libstdc++.so.6 #3 0x0809a8b3 in std::dequeFGReplayData, std::allocatorFGReplayData ::_M_push_back_aux (this=0xfc866d0, [EMAIL PROTECTED]) at /usr/include/c++/4.1.2/ext/new_allocator.h:88 #4 0x08099796 in FGReplay::update (this=0xfc866b0, dt=0.83337) at /usr/include/c++/4.1.2/bits/stl_deque.h:1045 #5 0x08588c16 in SGSubsystemGroup::Member::update (this=0xfc81140, delta_time_sec=0.83337) at subsystem_mgr.cxx:236 #6 0x08589515 in SGSubsystemGroup::update (this=0x8a0bb74, delta_time_sec=0.83337) at subsystem_mgr.cxx:127 #7 0x08588dfc in SGSubsystemMgr::update (this=0x8a0bb58, delta_time_sec=0.83337) at subsystem_mgr.cxx:296 #8 0x0805255e in fgMainLoop () at main.cxx:509 #9 0x08087211 in GLUTidle () at fg_os.cxx:122 #10 0xb7f581df in glutMainLoop () from /usr/lib/libglut.so.3 #11 0x08056762 in fgMainInit (argc=2, argv=0xbfdcebd4) at main.cxx: #12 0x080512d1 in main (argc=894892171, argv=0x229) at bootstrap.cxx:215 (gdb) Cheers, Durk - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
Hi, had the same notice: the 787 has this stutters too, the 777 also. Both uses a lot of nasal while the 737-300 seems to be much better. Only the old nasal air-ground. In the moment I improve the 737-300 and add the system of the 777/ 787 - the stutter increased dramatically! Melchior always saying that it is not the issue with the setlistner- but I'm sure there is a problem with which causes this stutters. Maybe it seems not to be logical, but it is remarkable, that the aircrafts with this typical nasal has this problems more than other. It would be better for the project, if we could solve this problem. But this needs a objective look into the possible causes. And I don't think we could solve this until 1st octobre. Greetings HHS --- dave perry [EMAIL PROTECTED] schrieb: Durk Talsma wrote: Please let me know if there are any issues that need to be addressed. I just did a cvs up -dP for the plib branch of SimGear and then FlightGear to see if the periodic hesitation and drop in frame rate issue had been resolved. It has not been resolved. This problem is not in 0.9.10 and IMHO needs to be resolved before releasing 0.9.11. Am I missing/overlooking a fix. I have been working with Torsten Dreyer to implement the CenturyIII and a derivative AltimaticIIIc autopilot. So I have flown his SenecaII and the pa24-250 a lot while tweaking the two autopilot config files for these autopilots and aircraft. I do not see this issue at all when flying the pa24-250 and I see it every flight with the SenecaII. In the latter case, it is hardly noticable at first with only a slight variation in frame rate ( no more than 5 fps variation from 67 fps). As you continue to fly, the drop in frame rate increases (to 40+ fps after 15 minutes) and the duration of the drop increases. Also the time between drops increases. This has made the optimization of the autopilot config for the SenecaII difficult as this periodic hicup acts as an impulse response to the PID controllers. I am not changing anything but the aircraft between these flights. Two differences between these aircraft come to mind. First, the pa24-250 is a yasim model and the SenecaII is a jsbsim model. Second, the pa24-250 uses only three setlistener in the nasal I maintain and one of these goes away right away after the autopilot is powered up while the SenecaII uses a lot of setlisteners. I am not suggesting the cause, only noting the differences I am aware of between these aircraft. Regards, Dave Perry - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel __ Yahoo! Clever: Sie haben Fragen? Yahoo! Nutzer antworten Ihnen. www.yahoo.de/clever - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
Hi, I think that there are maybe some more causes than only the setlistener. And I'm sure v0.9.10 had it too. That's why I said we should look objective into that. btw. what happens, if you press Ctrl-c outside the stutters? HHS --- Melchior FRANZ [EMAIL PROTECTED] schrieb: * Heiko Schulz -- 9/22/2007 6:19 PM: Melchior always saying that it is not the issue with the setlistner- but I'm sure there is a problem with which causes this stutters. And I'm sure it's not. I had the same with the f16, which uses almost *no* Nasal, and the YF-23, which uses no Nasal at all(?). (Of course, there's always the global Nasal stuff, but there was much less at that time.) At one time when I researched the cause (without success), I ran fgfs in gdb, and always when the stutter appeared I pressed Ctrl-c. I almost always ended up in the nvidia driver code, and thought that some very expensive 3D drawing would be the cause. But that's as much guessing as the stale and as-good-as disproved setlistener() claim ... :-} m. - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel Wissenswertes zum Thema PC, Zubehör oder Programme. BE A BETTER INTERNET-GURU! www.yahoo.de/clever - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
* Heiko Schulz -- 9/22/2007 6:53 PM: I think that there are maybe some more causes than only the setlistener. While I still don't think that listeners are even one of the causes, I agree that there could be more sources. Some bad timing. And I'm sure v0.9.10 had it too. So am I. Actually, I think I was (one of) the first who ever reported that problem, and I seem to remember that it was long before Nasal listeners even existed. :-} what happens, if you press Ctrl-c outside the stutters? I end up in whatever code is currently executed. Most of the time it's outside the nVidia driver. If someone wonders how I could reliably press the key within the stutter: the stutters become longer and longer, and when they are half a second it's quite easy to hit them. ;-) (BTW: Yes, I also checked the other threads.) m. - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] FGFS 0.9.11 release candidate two
Harald JOHNSEN a écrit :
Melchior FRANZ wrote:
* Heiko Schulz -- 9/22/2007 6:19 PM:
Melchior always saying that it is not the issue with
the setlistner- but I'm sure there is a problem with
which causes this stutters.
And I'm sure it's not. I had the same with the f16,
which uses almost *no* Nasal, and the YF-23, which uses
no Nasal at all(?). (Of course, there's always the global
Nasal stuff, but there was much less at that time.)
Can someone plays a bit with a profiler ? While a listener is nothing
special, Nasal itself take a substancial part of the cpu time per frame
(of course that depends on a few random parameter but I have between 20
35 % of the cpu used in the nasal sources). And some time ago I was
refering to the garbage collector that causes mini stutters and the gc
was running on a period like 1 gc every 20 seconds at fg start and after
some time it was like 1 gc every 10 seconds, the time spent in the gc
was increasing too.
I used a profiler of my own :
cvs -z4 -w -q diff -u -wb -- simgear\structure\subsystem_mgr.cxx (in
directory I:\Devel\SimGear.plib\)
Index: simgear/structure/subsystem_mgr.cxx
===
RCS file: /var/cvs/SimGear-0.3/SimGear/simgear/structure/subsystem_mgr.cxx,v
retrieving revision 1.5
diff -u -w -b -r1.5 subsystem_mgr.cxx
--- simgear/structure/subsystem_mgr.cxx21 Feb 2006 12:59:31 -1.5
+++ simgear/structure/subsystem_mgr.cxx22 Sep 2007 22:12:50 -
@@ -4,6 +4,7 @@
#include exception.hxx
#include subsystem_mgr.hxx
+#include simgear/timing/timestamp.hxx
@@ -124,7 +125,17 @@
SGSubsystemGroup::update (double delta_time_sec)
{
for (unsigned int i = 0; i _members.size(); i++)
+{
+SGTimeStamp start, now;
+start.stamp();
_members[i]-update(delta_time_sec); // indirect call
+now.stamp();
+long b = ( now - start );
+if ( b 1 ) {
+ cout D : b _members[i]-name endl;
+ int a = 1;
+}
+}
}
void
It appears that it is the replay subsystem that creates long pauses
periodically, and sometimes the ai-model subsystem too :
( my traces : )
D : 12000 replay
D : 11000 replay
D : 17000 instrument20
D : 18000 instrumentation
D : 12000 replay
D : 12000 replay
D : 14000 replay
D : 19000 replay
D : 16000 replay
D : 18000 replay
D : 11000 replay
D : 22000 input
D : 13000 replay
D : 17000 replay
D : 14000 replay
D : 22000 replay
D : 18000 replay
D : 18000 replay
D : 18000 Traffic Manager
D : 17000 instrument13
D : 18000 instrumentation
D : 17000 replay
D : 23000 electrical0
D : 32000 systems
D : 18000 replay
D : 11000 replay
D : 16000 replay
D : 11000 replay
D : 11000 replay
D : 252000 input
D : 11000 replay
D : 12000 electrical0
D : 15000 systems
D : 11000 replay
D : 17000 ai_model
D : 11000 instrumentation
D : 11000 replay
D : 14000 replay
D : 18000 replay
D : 11000 replay
D : 12000 replay
D : 19000 replay
D : 15348000 replay
D : 16000 ai_model
D : 14000 replay
D : 11000 replay
D : 12000 replay
D : 14000 replay
D : 12000 replay
D : 15000 properties
D : 13000 replay
Very long pauses are caused by breakpoints in the debugger
regards,
-Fred
--
Frédéric Bouvier
http://frfoto.free.fr Photo gallery - album photo
http://www.fotolia.fr/p/2278 Other photo gallery
http://fgsd.sourceforge.net/ FlightGear Scenery Designer
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/
___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel
