Re: [Flightgear-devel] CVS - Frame Rates under Windows XP - plus movie comments AJ's Car
Dear Vivian, Your figures sound about right as I also get very frustrated with this also, in fact when I am filming making the movies I have found something interesting! When I use mibs for example and set it on a fixed heading to do straight in line pan scans for panoramic shots such as moving in a straight line from point A to point B the first time it stutters like hell, then I use the reverse button (Brake) and pan back out (Reversing along the same path) and do this several times forward and back then for some odd reason maybe due to caching it becomes smooth and no stutter, this is a technique I now use all the time when I am making the OLeg Movies, but unlike you I can't afford to Buy expensive new graphics cards so YES I agree the frame rates do need more work, and based on the method I tried in the point A - B scenario maybe something in the caching needs looking at, maybe even some sort of dynamic pre-processing of scenery just ahead might solve these issues or using a dynamic quality resolution technique to reduce the scenery quality being displayed based on cpu usage might help? Regards, Aerotro PS, I am planning and currently working on part 4 of OLeg's Adventures some scenarios require video using AJ's Duplo car and I want to simulate a speeding car with smoke from the tires / exhaust and so I will most likely be adding special effects for this, unless someone wants to make the job easier by implementing a smoking exhaust to the car? Every bit of help counts :oD Vivian Meazza Wrote: Message: 8 Date: Mon, 31 Mar 2008 19:43:18 +0100 From: "Vivian Meazza" Subject: [Flightgear-devel] CVS - Frame Rates under Windows XP To: "'FlightGear developers discussions'" Message-ID: <[EMAIL PROTECTED]> Content-Type: text/plain; charset="us-ascii" Hi About 10 days ago I got fed up with low frame rates using OSG around KSFO, so I went out and bought the best nVidia AGP card that I could find - 7600gs with 512Mb of VRAM. I fitted it with great anticipation to my machine - P4 2.8Ghz/800Mhz FSB with 1.5 Gb of RAM running XP - and, precisely NOTHING. No change in frame rate between the new card and the old - a FX6200 with 256Mb VRAM - for FG-OSG-HEAD. On the other hand plib-HEAD flies. Here are some comparative results: snip- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Register now and save $200. Hurry, offer ends at 11:59 p.m., Monday, April 7! Use priority code J8TLD2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
On Wednesday 02 April 2008 12:14, Frederic Bouvier wrote: > According to this chart, it seems you are right : > http://cvs.flightgear.org/cgi-bin/viewvc/viewvc.cgi/source/src/Aircraft/rep >lay.cxx?view=graph&sortby=file&pathrev=PRE_OSG_PLIB_20061029 > > -Fred Okay, I've just committed the port from CVS/PLIB. Cheers, Durk - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
Quoting Durk Talsma : > On Tuesday 01 April 2008 22:39, Vivian Meazza wrote: > > > > So I conclude that: > > 1. There is a problem with replay, > > 2. MinGW has about the same performance gap between OSG and plib on XP as > > gcc does on Linux. > > 3. MinGW performance is probably as good as it gets. > > 4. Either MSVC8 doesn't compile such fast code as MinGW, or there is a > > setting wrong somewhere. > > > It's been a while ago since I last looked at at, but back in November I put a > lot of effort into optimizing the replay system's memory usage. At the time, > this was one of our prime candidates for the infamous periodic stutter bug. > Because we were aiming for a release, I placed the job of forward porting > these changes to CVS/HEAD rather low on my TODO list. I thought I had already > finished most of these ports, but it is possible this one still remains to be > done. I will have a look. > > Anyways, it looks like it is possible that CVS/HEAD has a much less efficient > replay system than CVS/PLIB. According to this chart, it seems you are right : http://cvs.flightgear.org/cgi-bin/viewvc/viewvc.cgi/source/src/Aircraft/replay.cxx?view=graph&sortby=file&pathrev=PRE_OSG_PLIB_20061029 -Fred -- Frédéric Bouvier http://my.fotolia.com/frfoto/ Photo gallery - album photo http://fgsd.sourceforge.net/ FlightGear Scenery Designer - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
On Tuesday 01 April 2008 22:39, Vivian Meazza wrote: > > So I conclude that: > 1. There is a problem with replay, > 2. MinGW has about the same performance gap between OSG and plib on XP as > gcc does on Linux. > 3. MinGW performance is probably as good as it gets. > 4. Either MSVC8 doesn't compile such fast code as MinGW, or there is a > setting wrong somewhere. > It's been a while ago since I last looked at at, but back in November I put a lot of effort into optimizing the replay system's memory usage. At the time, this was one of our prime candidates for the infamous periodic stutter bug. Because we were aiming for a release, I placed the job of forward porting these changes to CVS/HEAD rather low on my TODO list. I thought I had already finished most of these ports, but it is possible this one still remains to be done. I will have a look. Anyways, it looks like it is possible that CVS/HEAD has a much less efficient replay system than CVS/PLIB. Cheers, Durk - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
Tim Moore > Sent: 01 April 2008 12:35 > To: FlightGear developers discussions > Subject: Re: [Flightgear-devel] CVS - Frame Rates under Windows XP > > > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Heiko Schulz wrote: > | > |> AND, with Linux and the same Graphics Card 7800 GS > |> 512 MB, i can notice the > |> same decrease of performance from FG-PLIB to FG-OSG > |> , > |> I ever had about 20% less performance with OSG. > |> > |> I am running FG on AMD ATHLON 3200 (32 bit) with > |> AGP mothercard. > |> > |> May be OSG is more accurate with modern (recent) > |> CPU. (i must test it). > |> > |> > |> Cheers > |> > |> > |> -- > | I notice a difference between OSG and Plib too- but > | this is known! > | But we can discuss and find out, how to make it > | faster- time for Tim to answer! > > I suggested that Vivian post his experiences here. I've been > working with him for several months on trying to resolve > these problems. I'm convinced that his machine is cursed :) I > don't develop on Windows and so have run out of ideas other > than general suggestions. I'm hoping that more experienced > Windows developers will have some ideas for how to proceed. > > One thing seems clear: the OSG version uses more memory than > the plib version does. Therefore there is a more memory > "pressure;" the entire system needs to deal with this. Memory > allocation, especially on Windows, seems to be expensive. ~ > Vivian had some improved results today turning off the replay > system, which may give us a clue. I think it will only help > performance to, in general, optimize the memory usage of FlightGear. > Just an update. Using --prop:sim/replay/disable=true Makes FG-OSG as smooth as plib is with --prop:sim/replay/disable=false I tried Csaba's MinGW build - frame rates are improved by 20% across the board, and only 20% below plib (which of course might see further improvement with a MinGW build. So I conclude that: 1. There is a problem with replay, 2. MinGW has about the same performance gap between OSG and plib on XP as gcc does on Linux. 3. MinGW performance is probably as good as it gets. 4. Either MSVC8 doesn't compile such fast code as MinGW, or there is a setting wrong somewhere. Vivian - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
Hi Frederic Bouvier schrieb am 01.04.2008 08:52: > I have a Core2 Duo 2.66 ( E6600 ) and a 7600GT. I always saw the greatest fps > increase after upgrading CPU and was disappointed by several GPU-only upgrade. > > All I can tell is that with the Seahawk, at KSFO, I have 75hz steady ( with > vsync on ) if I wait for 2 minutes. In the meantime, fps vary greatly from 40 > to 75, during the threaded model loading process. And this is done with a 50% > CPU usage. > > Here (WinXP, Core2 Duo 3Ghz, 8800GT) FG consumes more than 50% CPU. Directly after starting (when the 3D scenery is displayed for the first time) the CPU usage is >90%. Therefore the loader-thread seem to run on a different core than the rest (or is there another thread?). > -Fred > > Maik - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
On Tuesday 01 April 2008 13:10, Anders Gidenstam wrote: > On Tue, 1 Apr 2008, LeeE wrote: > > Hmm... [looks at watch and wonders if it's time to post > > another missive about the _need_ for a redesign of FG to run on > > MPP systems as it gets ever clearer that significant increases > > in computing power have more or less stalled in terms of height > > (cpu speed) and in future, will come instead from width > > (parallel processing) - FG's current design is effectively > > obsolete] > > Yes, it might be time for that. However, the recent work on model > loading is certainly a step in the right direction. > > One problem is to identify parts that we will gain anything from > moving to a separate thread. I have seen the FDM suggested in the > past, but even on my (ancient) system JSBSim corresponds to about > 1-5% of the CPU usage (estimated by looking at the rate sim time > progresses in the standalone version of JSBSim). Andy has told me > YASim is more expensive (it does more at runtime) but it is > probably at most 20-30% of the CPU usage (guesstimate :). So, the > prospective gains there do not look that large. > Doing some profiling might make the picture clearer. > > I think the main targets for parallelization are the rendering > pipeline and various "add-on" systems, like the traffic manager. > Personally, I'd like to have threads (possibly with very limited > interaction abilities) available in Nasal for isolated and > computation intensive tasks (e.g. fast forwarding my fire > cellular automaton :). > > Just my 2 (euro) cents.. > > Cheers, > > Anders Without a fairly deep understanding of how the various subsystems within FG have been implemented and work it's difficult to make worthwhile suggestions, especially while the developers are still getting their heads around the intricacies of OSG... ...but fwiw:) I think the single most important step would be to run the graphics subsystem in it's own process, splitting it from everything else. On multi-core systems this would mean that the graphics subsystem gets the resources freed by the 'everything else' and the 'everything else' gets the resources freed by the graphics subsystem. This would be a relatively small gain for the graphics subsystem and a much bigger gain for everything else, where it's arguable that it's needed, but it would allow much higher and more consistent rates in the FDMs, autopilot controllers & filters and Nasal. The thing is though, that the graphics subsystem needs a lot of data and it's questionable that it could be transferred quickly enough. Therefore it's likely that the scenery & model loaders would need to be included in the graphics subsystem so once it's told what data it needs it can fetch it itself. With a core to itself, the 'everything else' part of FG would benefit less by further splitting but if it was well designed it should make plug-ins much easier to implement and maintain. In the longer term, thought needs to be given to 'box-rendering' the graphics - splitting the scene into several regions and processing them in parallel - but this is much easier said than done, especially as rendering is h/w based. Still, this is the sort of thing that newer versions of OGL/OSG will _have_ to address in the future, if they haven't already got some features in this area. LeeE - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
On Tue, 1 Apr 2008, LeeE wrote: > Hmm... [looks at watch and wonders if it's time to post another > missive about the _need_ for a redesign of FG to run on MPP systems > as it gets ever clearer that significant increases in computing > power have more or less stalled in terms of height (cpu speed) and > in future, will come instead from width (parallel processing) - > FG's current design is effectively obsolete] Yes, it might be time for that. However, the recent work on model loading is certainly a step in the right direction. One problem is to identify parts that we will gain anything from moving to a separate thread. I have seen the FDM suggested in the past, but even on my (ancient) system JSBSim corresponds to about 1-5% of the CPU usage (estimated by looking at the rate sim time progresses in the standalone version of JSBSim). Andy has told me YASim is more expensive (it does more at runtime) but it is probably at most 20-30% of the CPU usage (guesstimate :). So, the prospective gains there do not look that large. Doing some profiling might make the picture clearer. I think the main targets for parallelization are the rendering pipeline and various "add-on" systems, like the traffic manager. Personally, I'd like to have threads (possibly with very limited interaction abilities) available in Nasal for isolated and computation intensive tasks (e.g. fast forwarding my fire cellular automaton :). Just my 2 (euro) cents.. Cheers, Anders -- --- Anders Gidenstam mail: anders(at)gidenstam.org WWW: http://www.gidenstam.org/FlightGear/ - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
On Tuesday 01 April 2008 07:52, Frederic Bouvier wrote: > Selon Vivian Meazza : > > CoreDuo 2,6 Ghz and a Gainward 8800GT. Not surprised it runs > > well!!! In particular I think the CoreDuo does threading better > > than the P4. In case you haven't noticed, the 7600gs is coping > > easily with the output from FG-OSG - that's why the frame rates > > didn't increase. > > [...] > > > Could we have some _real_ numbers to compare instead of > > hearsay. > > > > This is not a OSG versus plib discussion - it's a why OSG is so > > poor on XP discussion > > I have a Core2 Duo 2.66 ( E6600 ) and a 7600GT. I always saw the > greatest fps increase after upgrading CPU and was disappointed by > several GPU-only upgrade. > > All I can tell is that with the Seahawk, at KSFO, I have 75hz > steady ( with vsync on ) if I wait for 2 minutes. In the > meantime, fps vary greatly from 40 to 75, during the threaded > model loading process. And this is done with a 50% CPU usage. > > -Fred That 50% CPU usage will be one of your cores running flat out while the other one is idling:) Hmm... [looks at watch and wonders if it's time to post another missive about the _need_ for a redesign of FG to run on MPP systems as it gets ever clearer that significant increases in computing power have more or less stalled in terms of height (cpu speed) and in future, will come instead from width (parallel processing) - FG's current design is effectively obsolete] LeeE - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Heiko Schulz wrote: | |> AND, with Linux and the same Graphics Card 7800 GS |> 512 MB, i can notice the |> same decrease of performance from FG-PLIB to FG-OSG |> , |> I ever had about 20% less performance with OSG. |> |> I am running FG on AMD ATHLON 3200 (32 bit) with |> AGP mothercard. |> |> May be OSG is more accurate with modern (recent) |> CPU. (i must test it). |> |> |> Cheers |> |> |> -- | I notice a difference between OSG and Plib too- but | this is known! | But we can discuss and find out, how to make it | faster- time for Tim to answer! I suggested that Vivian post his experiences here. I've been working with him for several months on trying to resolve these problems. I'm convinced that his machine is cursed :) I don't develop on Windows and so have run out of ideas other than general suggestions. I'm hoping that more experienced Windows developers will have some ideas for how to proceed. One thing seems clear: the OSG version uses more memory than the plib version does. Therefore there is a more memory "pressure;" the entire system needs to deal with this. Memory allocation, especially on Windows, seems to be expensive. ~ Vivian had some improved results today turning off the replay system, which may give us a clue. I think it will only help performance to, in general, optimize the memory usage of FlightGear. Tim -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFH8h3jeDhWHdXrDRURArN1AKCvTYByAdaiWoizFreC2ueU3W2/OgCfS14j 6nNqt9a0Vq6VS11PzbD53BU= =OhsK -END PGP SIGNATURE- - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
--- Frederic Bouvier <[EMAIL PROTECTED]> schrieb: > Selon Vivian Meazza : > > > CoreDuo 2,6 Ghz and a Gainward 8800GT. Not > surprised it runs well!!! In > > particular I think the CoreDuo does threading > better than the P4. In case > > you haven't noticed, the 7600gs is coping easily > with the output from FG-OSG > > - that's why the frame rates didn't increase. > > > [...] > > > > Could we have some _real_ numbers to compare > instead of hearsay. > > > > This is not a OSG versus plib discussion - it's a > why OSG is so poor on XP > > discussion > > I have a Core2 Duo 2.66 ( E6600 ) and a 7600GT. I > always saw the greatest fps > increase after upgrading CPU and was disappointed by > several GPU-only upgrade. > > All I can tell is that with the Seahawk, at KSFO, I > have 75hz steady ( with > vsync on ) if I wait for 2 minutes. In the > meantime, fps vary greatly from 40 > to 75, during the threaded model loading process. > And this is done with a 50% > CPU usage. > > -Fred Noticed too, taht the fps the very first seconds are not stable- but after that there is fps variation to see. still in work: http://www.hoerbird.net/galerie.html But already done: http://www.hoerbird.net/reisen.html E-Mails jetzt auf Ihrem Handy. www.yahoo.de/go - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
Selon Vivian Meazza : > CoreDuo 2,6 Ghz and a Gainward 8800GT. Not surprised it runs well!!! In > particular I think the CoreDuo does threading better than the P4. In case > you haven't noticed, the 7600gs is coping easily with the output from FG-OSG > - that's why the frame rates didn't increase. > [...] > > Could we have some _real_ numbers to compare instead of hearsay. > > This is not a OSG versus plib discussion - it's a why OSG is so poor on XP > discussion I have a Core2 Duo 2.66 ( E6600 ) and a 7600GT. I always saw the greatest fps increase after upgrading CPU and was disappointed by several GPU-only upgrade. All I can tell is that with the Seahawk, at KSFO, I have 75hz steady ( with vsync on ) if I wait for 2 minutes. In the meantime, fps vary greatly from 40 to 75, during the threaded model loading process. And this is done with a 50% CPU usage. -Fred -- Frédéric Bouvier http://my.fotolia.com/frfoto/ Photo gallery - album photo http://fgsd.sourceforge.net/ FlightGear Scenery Designer - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
On Mon, Mar 31, 2008 at 8:43 PM, Vivian Meazza <[EMAIL PROTECTED]> wrote: > > I've tried both executables generated here using > MSVC8, and Fred's pre-cooked binaries. So far as I can see the results are > identical. Here is a mingw32 build made with gcc 4.3.0 if you want to give it a try: http://www.youshare.com/view.php?file=fg-mingw.zip I can only say it starts and the menu works (tested using a virtual machine through remote desktop :) No warranties, might have left out a dozen files from the archive... -- Csaba/Jester - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
> CoreDuo 2,6 Ghz and a Gainward 8800GT. Not surprised > it runs well!!! If you read carefully, I said that even on my old pc there wasn't any trouble! >In particular I think the CoreDuo does threading better > than the P4. In case > you haven't noticed, the 7600gs is coping easily > with the output from FG-OSG > - that's why the frame rates didn't increase. > > And what would Vsync do? The frame rate never gets > anywhere near the point > at which it might operate. Really? I don't believe that if I look at your PC - it is better than my old one > I've tried > frame-rate-throttle - it just makes it > worse. The graphics are absolutely no different here > between OSG or plib > that I can see, and why should they be? You are the first and only one yet having problems with that- the kids in our kids corner havn't this problem yet. > Could we have some _real_ numbers to compare instead > of hearsay. > > This is not a OSG versus plib discussion - it's a > why OSG is so poor on XP > discussion > It isn't poor on XP- I had framerates about 25-37 on my old 2.8HG single core and Nvidea FX5200 - with the tree shader I had about 17-25fps- absolut usuable. (KSFO, 28R, Noon, 1024x768, all boxes are checked) Note: the FX5200 is the lowest card. And: Windows XP needs a lot of perfomance - linux not! And that's the fact why a lot of things are running faster on Linux systems! But we can always talk about how to make FGFS faster! I think you have to check your pc first- still in work: http://www.hoerbird.net/galerie.html But already done: http://www.hoerbird.net/reisen.html Machen Sie Yahoo! zu Ihrer Startseite. Los geht's: http://de.yahoo.com/set - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
> > AND, with Linux and the same Graphics Card 7800 GS > 512 MB, i can notice the > same decrease of performance from FG-PLIB to FG-OSG > , > I ever had about 20% less performance with OSG. > > I am running FG on AMD ATHLON 3200 (32 bit) with > AGP mothercard. > > May be OSG is more accurate with modern (recent) > CPU. (i must test it). > > > Cheers > > > -- I notice a difference between OSG and Plib too- but this is known! But we can discuss and find out, how to make it faster- time for Tim to answer! still in work: http://www.hoerbird.net/galerie.html But already done: http://www.hoerbird.net/reisen.html Lesen Sie Ihre E-Mails jetzt einfach von unterwegs. www.yahoo.de/go - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
On lun 31 mars 2008, Vivian Meazza wrote: SNIP > > > > Even on my old PC with a FX5200 OSG run very pretty at > > least. > > It is true that OSG is slower than plib- but the > > graphic is much better (exxept the 3D-clouds and the > > missing shadows)- please not another discussion about > > OSG vs Plib! > > CoreDuo 2,6 Ghz and a Gainward 8800GT. Not surprised it runs well!!! In > particular I think the CoreDuo does threading better than the P4. In case > you haven't noticed, the 7600gs is coping easily with the output from > FG-OSG - that's why the frame rates didn't increase. > > And what would Vsync do? The frame rate never gets anywhere near the point > at which it might operate. I've tried frame-rate-throttle - it just makes > it worse. The graphics are absolutely no different here between OSG or plib > that I can see, and why should they be? > > Could we have some _real_ numbers to compare instead of hearsay. > > This is not a OSG versus plib discussion - it's a why OSG is so poor on XP > discussion > > Vivian > > > Hello, Vivian You are right :) AND, with Linux and the same Graphics Card 7800 GS 512 MB, i can notice the same decrease of performance from FG-PLIB to FG-OSG , I ever had about 20% less performance with OSG. I am running FG on AMD ATHLON 3200 (32 bit) with AGP mothercard. May be OSG is more accurate with modern (recent) CPU. (i must test it). Cheers -- Gérard http://pagesperso-orange.fr/GRTux/ - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
Heiko Schulz wrote > Sent: 31 March 2008 20:45 > To: FlightGear developers discussions > Subject: Re: [Flightgear-devel] CVS - Frame Rates under Windows XP > > > Vivian Meazza schrieb: > > > Hi > > > > About 10 days ago I got fed up with low frame rates > > using OSG around KSFO, > > so I went out and bought the best nVidia AGP card > > that I could find - 7600gs > > with 512Mb of VRAM. I fitted it with great > > anticipation to my machine - P4 > > 2.8Ghz/800Mhz FSB with 1.5 Gb of RAM running XP - > > and, precisely NOTHING. No > > change in frame rate between the new card and the > > old - a FX6200 with 256Mb > > VRAM - for FG-OSG-HEAD. On the other hand plib-HEAD > > flies. Here are some > > comparative results: > > > > Aircraft: Seahawk > > OSG > > PLIB > >KSFOGeneral Ocean > > KSFOGeneral > > Ocean > > every check box turned to off: > > 25 30 60 > > 4580 > > 117 > > with everything checked, traffic manager on: > >15 25 36 > > 2550 > > 80 > > as above with shadows and 3d clouds: > > > > 1840 > > 65 > > as above but with trees and precipitation: > >10 15 40 (but the rain > > turned itself off!) > > > > Aircraft: c172p > > every check box turned to off: > > 3340 60 > > 4590 > > 130 > > with everything checked, traffic manager on: > > 2733 40 > > 3050 > > 75 > > as above with shadows and 3d clouds: > > > > 3050 > > 65 > > as above but with trees and precipitation: > > 1525 30 > > > > Aircraft: Buccaneer (particles) > > as above but with trees and precipitation: > >818 20 > > > > In addition to the difference in frame rate, OSG > > also stutters, while plib > > is commendably smooth. I've tried both executables > > generated here using > > MSVC8, and Fred's pre-cooked binaries. So far as I > > can see the results are > > identical. I've profiled both OSG and plib - for > > anyone interested, some of > > the results are here: > > > > ftp://abbeytheatre2.org.uk/fgfs/OSG/ > > > > While I am no expert in profiling code, the results > > seem to me to show that > > we are getting stuck in the bowels of OSG somewhere, > > while plib is > > well-ordered, with no particular cpu-hog, which is > > pretty much what the > > above table indicates. > > > > My principle concern is that there appears to be > > very little headroom for > > future developments in OSG for shadows, or 3d > > clouds, or landing lights, on > > what is a not-very-old and pretty capable machine. I > > would be grateful if > > some Windows user(s) could confirm at least the > > shape of these comparative > > results. I understand that Linux users get much > > better results than these. > > We might be getting towards the point when Windows > > users are stuck with plib > > (and that's not all bad) while Linux users get to > > play with all the new > > goodies. We might be drifting away from our cross > > platform ethic. > > > > > > Vivian > > > > Luckily the new card wasn't terribly expensive > > otherwise I would be quite > > p'd off > > > Hi, > > With the recent built by Fred I found the old stutters > again- but I could sovle this for me. > > But I can't see any of those problems you have- I can > run FGFS with all features we have - plus heavy > self-written interactive traffic and mp. And the best- > I can run other programs beside too without any > problems (blender as an graphic example!) > > You should know, that you have to set some things on > your pc, Nvidea and FGFS: > > - switch Vsync in the Nvidea settings > - use the frame-rate-throttle! > > I use a CoreDuo 2,6 Ghz and a Gainward 8800GT - but > there were people in the german forum and the official > forum (kid's playground) having no problems with the > recen
Re: [Flightgear-devel] CVS - Frame Rates under Windows XP
--- Vivian Meazza <[EMAIL PROTECTED]> schrieb: > Hi > > About 10 days ago I got fed up with low frame rates > using OSG around KSFO, > so I went out and bought the best nVidia AGP card > that I could find - 7600gs > with 512Mb of VRAM. I fitted it with great > anticipation to my machine - P4 > 2.8Ghz/800Mhz FSB with 1.5 Gb of RAM running XP - > and, precisely NOTHING. No > change in frame rate between the new card and the > old - a FX6200 with 256Mb > VRAM - for FG-OSG-HEAD. On the other hand plib-HEAD > flies. Here are some > comparative results: > > Aircraft: Seahawk > OSG > PLIB >KSFOGeneral Ocean > KSFOGeneral > Ocean > every check box turned to off: > 25 30 60 > 4580 > 117 > with everything checked, traffic manager on: >15 25 36 > 2550 > 80 > as above with shadows and 3d clouds: > > 1840 > 65 > as above but with trees and precipitation: >10 15 40 (but the rain > turned itself off!) > > Aircraft: c172p > every check box turned to off: > 3340 60 > 4590 > 130 > with everything checked, traffic manager on: > 2733 40 > 3050 > 75 > as above with shadows and 3d clouds: > > 3050 > 65 > as above but with trees and precipitation: > 1525 30 > > Aircraft: Buccaneer (particles) > as above but with trees and precipitation: >818 20 > > In addition to the difference in frame rate, OSG > also stutters, while plib > is commendably smooth. I've tried both executables > generated here using > MSVC8, and Fred's pre-cooked binaries. So far as I > can see the results are > identical. I've profiled both OSG and plib - for > anyone interested, some of > the results are here: > > ftp://abbeytheatre2.org.uk/fgfs/OSG/ > > While I am no expert in profiling code, the results > seem to me to show that > we are getting stuck in the bowels of OSG somewhere, > while plib is > well-ordered, with no particular cpu-hog, which is > pretty much what the > above table indicates. > > My principle concern is that there appears to be > very little headroom for > future developments in OSG for shadows, or 3d > clouds, or landing lights, on > what is a not-very-old and pretty capable machine. I > would be grateful if > some Windows user(s) could confirm at least the > shape of these comparative > results. I understand that Linux users get much > better results than these. > We might be getting towards the point when Windows > users are stuck with plib > (and that's not all bad) while Linux users get to > play with all the new > goodies. We might be drifting away from our cross > platform ethic. > > > Vivian > > Luckily the new card wasn't terribly expensive > otherwise I would be quite > p'd off > Hi, With the recent built by Fred I found the old stutters again- but I could sovle this for me. But I can't see any of those problems you have- I can run FGFS with all features we have - plus heavy self-written interactive traffic and mp. And the best- I can run other programs beside too without any problems (blender as an graphic example!) You should know, that you have to set some things on your pc, Nvidea and FGFS: - switch Vsync in the Nvidea settings - use the frame-rate-throttle! I use a CoreDuo 2,6 Ghz and a Gainward 8800GT - but there were people in the german forum and the official forum (kid's playground) having no problems with the recent OSG-built on lower pc. And the difference between a 8800 and 7600 isn't really far. Even on my old PC with a FX5200 OSG run very pretty at least. It is true that OSG is slower than plib- but the graphic is much better (exxept the 3D-clouds and the missing shadows)- please not another discussion about OSG vs Plib! Regards HHS still in work: http://www.hoerbird.net/galerie.html But already done: http://www.hoerbird.net/reisen.html E-Mails jetzt auf Ihrem Handy. www.yahoo.de/go - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
