[Flightgear-devel] Shader performance and multiple GPU?
Hi, as pointed out in other threads, I am currently assembling a machine with five nVidia cards. The machine and the cards are fast enough to render a steady 60fps, synced to vblank. But this is only with shaders disabled. The moment, I enable shaders and 3d- clouds, the frame rate drops to 20fps and sometimes even less. This does not happen when I limit the display to just one single GPU. One GPU with dual screen ("TwinView") runs at framerates beyond limits even with all shaders on and quality level to highest. The moment I add a second card to the setup, it slows down significantly. I know that shaders are little program fragments running on the GPU itself. Now, what happens, if a shader effect spawns multiple GPU's, e.g. a cloud which is visible across the entire window? Cheers, Torsten -- Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] Shader performance and multiple GPU?
This mirrors comments to the OSG mailing list, where people have reported performance issues with multiple GPUs and screens. There hasn't been much of a resolution there; some people have the problem, others don't. I've put together a multi-GPU machine to explore these issues; now I just need to buy another screen :) What threading mode are using? Could you post a screenshot of the OSG statistics (cycle debug statistics twice) and your X configuration file? Tim On Mon, Nov 1, 2010 at 9:06 AM, Torsten Dreyer wrote: > Hi, > > as pointed out in other threads, I am currently assembling a machine with > five > nVidia cards. The machine and the cards are fast enough to render a steady > 60fps, synced to vblank. > But this is only with shaders disabled. The moment, I enable shaders and > 3d- > clouds, the frame rate drops to 20fps and sometimes even less. This does > not > happen when I limit the display to just one single GPU. One GPU with dual > screen ("TwinView") runs at framerates beyond limits even with all shaders > on > and quality level to highest. The moment I add a second card to the setup, > it > slows down significantly. > I know that shaders are little program fragments running on the GPU itself. > Now, what happens, if a shader effect spawns multiple GPU's, e.g. a cloud > which is visible across the entire window? > > Cheers, Torsten > > > -- > Nokia and AT&T present the 2010 Calling All Innovators-North America > contest > Create new apps & games for the Nokia N8 for consumers in U.S. and Canada > $10 million total in prizes - $4M cash, 500 devices, nearly $6M in > marketing > Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store > http://p.sf.net/sfu/nokia-dev2dev > ___ > Flightgear-devel mailing list > Flightgear-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/flightgear-devel > -- Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] Shader performance and multiple GPU?
> This mirrors comments to the OSG mailing list, where people have reported > performance issues with multiple GPUs and screens. There hasn't been much > of a resolution there; some people have the problem, others don't. I've > put together a multi-GPU machine to explore these issues; now I just need > to buy another screen :) > > What threading mode are using? Could you post a screenshot of the OSG > statistics (cycle debug statistics twice) and your X configuration file You can find the configuration files here http://wiki.flightgear.org/index.php/FSweekend_2010#Configuration_files I'll try to post the statistics and later today. FWIW, I get occasional hanging fgfs with a complete black screen very soon during startup. Here is the output of --log-level=debug (not very informative) Processing command line arguments Finished command line arguments Initializing splash screen found path:Aircraft/SenecaII/splash.png via /sim/aircraft-dir: /home/flightgear/fg-osg/data/Aircraft/SenecaII/splash.png Unknown font name '', defaulting to Helvetica Unknown font name '', defaulting to Helvetica Unknown font name '', defaulting to Helvetica It ends up in an unkillable fgfs (not even kill -9 from root kills) and I have to reboot the machine. Looks like a deadlock in the driver :-( Cheers, Torsten -- Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
[Flightgear-devel] Fwd: in air starts and all preset control/trim settings broke
I am passing this along from David Megginson, who used to be one of our core developers and became a pilot through the process of participating in FlightGear, (but more recently has backed off helping with FlightGear to concentrate more on developing his own consulting business.) One important use of FlightGear is as a pilot training aid, and one huge thing that instructors like to do is setup in-air starts so you can practice a lot of approaches in a short amount of time (that's one of the key benefits of a simulator over a real airplane.) This capability seems to be broken now. David did some investigation and here is his email (attached). Also, here is a follow up email he just sent me: *Any control-related property **specified in the aircraft *-set.xml files, .fgfsrc, or the command **line is blown away. Many of the *-set.xml files set default fixed **aileron and rudder trim to counteract torque etc. in cruise, just like **some planes have tabs that can be physically bent on the ground (most **planes are designed so that torque/slipstream/etc. effects are **negligible in level cruise).* * * *I'm sure this was just an accident during some code refactoring.* -- Forwarded message -- From: David Megginson Date: Sun, Oct 31, 2010 at 5:50 PM I've been trying to use FlightGear for instrument approach practice, which means I want to start in the air, which means I want the engine already running. This, however, no longer works: fgfs --prop:/controls/engines/engine/magnetos=3 The value was set correctly at first, but then would be blown away during later initialization. I did a bit of digging in the code, and discovered the culprit: #0 FGControls::reset_all (this=0xe005b0) at controls.cxx:132 #1 0x00446538 in fgReInitSubsystems () at fg_init.cxx:1534 #2 0x0043e879 in do_presets_commit (arg=) at fg_commands.cxx:1236 #3 0x00a0d66e in SGCommandMgr::execute (this=, name=..., arg=0x13f94700) at commands.cxx:91 do_presets_commit calls fgReInitSubsystems(), which calls FGControls::reset_all(), which blows away any control presets because it reinitializes everything to default values. I can see the value of that when resetting the plane to a default position in an airport, but it's a bit of a disaster for any saved or canned scenarios. It might be a good idea either to remove the call to FGControls::reset_all() from fgReInitSubsystems, or to create a function other than do_presets_commit for the program's initialization sequence. For now, I'm just commenting out the call to FGControls::reset_all in fg_init.cxx. As a side note, this also blows away all the aileron- and rudder-trim adjustments used to make planes fly straightish in cruise -- for example, if you try flying the Cub now, you'll find that it pulls hard to the side even in level flight. All the best, David -- Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
[Flightgear-devel] FlightGear Newsletter - Octob er 2010
Hi all, time for yet another newsletter! The October edition is slightly less filled as last month, but we still collected a bunch of interesting updates/news items. Once again I would like to emphasize the need for contributions from all of you. Even if it is "just" adding one screenshot, or correcting someone else's English... Well, here it is: http://wiki.flightgear.org/index.php/FlightGear_Newsletter_October_2010 Contributions for next month's edition are welcome at this page. Cheers, Gijs -- Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] Fwd: in air starts and all preset control/trim settings broke
On 1 Nov 2010, at 16:10, Curtis Olson wrote: > This capability seems to be broken now. David did some investigation and > here is his email (attached). > > Also, here is a follow up email he just sent me: > > Any control-related property specified in the aircraft *-set.xml files, > .fgfsrc, or the command line is blown away. Many of the *-set.xml files set > default fixed aileron and rudder trim to counteract torque etc. in cruise, > just like some planes have tabs that can be physically bent on the ground > (most planes are designed so that torque/slipstream/etc. effects are > negligible in level cruise). > > I'm sure this was just an accident during some code refactoring. This is likely my fault - as David notes, accidental and during some re-factoring. The problem is, as he mentioned in his original email, there's no straightforward fix: reset clearly should reset controls (magnetos, flaps, stick input) to a default state. The definition of 'default' being the issue. 'preset-commit' should possibly not trigger a full-reset - but I'd worry about introducing subtle bugs in re-position from the GUI, if we change that behaviour now. On the other hand it is fairly testable, if breakage does occur. My 'least bad' option is to save control state (from the -set.xml, or config, or command line) as part of the initial state, so that it's re-applied after the controls are reset. That seems to most robust to me, but I need to check if there's knock on effects (possibly rather subtle ones) from saving and then restoring more properties during the reset process. Other suggestions most welcome! James -- Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] Shader performance and multiple GPU?
Looks like this is a hardware problem. Looking at the kernel log, I found irq 16: nobody cared (try booting with the "irqpoll" option) irq 16 is being used by nvidia and the onboard USB ports. After disabling USB in the BIOS, it seems like everything is running smooth. Because all my control devices are connected by usb, I re-enabled USB and started the kernel with acpi=debug. So far it's running - now I'll check shader-performance... Torsten > This mirrors comments to the OSG mailing list, where people have reported > performance issues with multiple GPUs and screens. There hasn't been much > of a resolution there; some people have the problem, others don't. I've > put together a multi-GPU machine to explore these issues; now I just need > to buy another screen :) > > What threading mode are using? Could you post a screenshot of the OSG > statistics (cycle debug statistics twice) and your X configuration file? > > Tim > > On Mon, Nov 1, 2010 at 9:06 AM, Torsten Dreyer wrote: > > Hi, > > > > as pointed out in other threads, I am currently assembling a machine with > > five > > nVidia cards. The machine and the cards are fast enough to render a > > steady 60fps, synced to vblank. > > But this is only with shaders disabled. The moment, I enable shaders and > > 3d- > > clouds, the frame rate drops to 20fps and sometimes even less. This does > > not > > happen when I limit the display to just one single GPU. One GPU with dual > > screen ("TwinView") runs at framerates beyond limits even with all > > shaders on > > and quality level to highest. The moment I add a second card to the > > setup, it > > slows down significantly. > > I know that shaders are little program fragments running on the GPU > > itself. Now, what happens, if a shader effect spawns multiple GPU's, e.g. > > a cloud which is visible across the entire window? > > > > Cheers, Torsten -- Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel