August 8, 20187 yr 4 minutes ago, SteveW said: Gerard, are you saying opt 3, your AI Traffic is reducing your fps as you fly, when it did not in 4.2? I think he is sitting tight at runway 24 at EHAM watching. But why not test this without AI at first ? Thanks Michael Moe Michael Moe
August 8, 20187 yr Commercial Member I'll have a look at that. Steve Waite: Engineer at codelegend.com
August 8, 20187 yr Author 1 hour ago, SteveW said: Gerard, are you saying opt 3, your AI Traffic is reducing your fps as you fly, when it did not in 4.2? Yes, it is reducing the fps . In V4.2 there was a small reduction in fos, and it was immediately and it stayed that way. 5950x3d 5.4-5.7 GHz - Asus ROG 870 Crosshair Apex - GSkill Neo 2x 24 Gb 6000 mhz / cas 26 - MSI RTX 5090 Gaming Trio OC - 1x SSD M2 6000 2TB - 1x SSD M2 2800/1800 1Tb - Corsair 5400 case - Corsair 360 liquid cooling set - 3x 75’ TCL tv. 13600 6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb - 1x SSD M2 2800/1800 2TB - 2x Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - FOV : 200 degrees My flightsim vids : https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0
August 8, 20187 yr Author 57 minutes ago, Michael Moe said: I think he is sitting tight at runway 24 at EHAM watching. But why not test this without AI at first ? Thanks Michael Moe Have done that too Michael. Also then there is a reduction over time but not that much as with AI too. Without AI it stays above 25, so that is flyable enough. But I am coming from 56... 5950x3d 5.4-5.7 GHz - Asus ROG 870 Crosshair Apex - GSkill Neo 2x 24 Gb 6000 mhz / cas 26 - MSI RTX 5090 Gaming Trio OC - 1x SSD M2 6000 2TB - 1x SSD M2 2800/1800 1Tb - Corsair 5400 case - Corsair 360 liquid cooling set - 3x 75’ TCL tv. 13600 6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb - 1x SSD M2 2800/1800 2TB - 2x Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - FOV : 200 degrees My flightsim vids : https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0
August 8, 20187 yr 15 minutes ago, GSalden said: Have done that too Michael. Also then there is a reduction over time but not that much as with AI too. Without AI it stays above 25, so that is flyable enough. But I am coming from 56... Ahh okay , i will test with the same add-ons but without the AI sitting at runway 24 for 90 minutes with the NGX. Static AS16+ASCA weather with REXSF3D clouds Thanks Michael Moe Edited August 8, 20187 yr by Michael Moe Michael Moe
August 8, 20187 yr 8 hours ago, kiek said: Then PSXseeconTraffic removes the AI object from P3D via SimConnect. Hi Nico, How are you removing the AI object? Are you using managed SimConnect? SimConnect_AIReleaseControlEx and then SimConnect_AIRemoveObject? I'm going to state what might seem obvious so please don't be offended as I don't know you and I don't know your technical background ... when you cycle thru the collection of AI objects to determine which one's qualify for disposal, are you saving off the objects reference to be disposed in a separate collection and then cycling that separate collection and dispose via reference (ObjectID)? Cheers, Rob.
August 8, 20187 yr Commercial Member 4 hours ago, GSalden said: Have done that too Michael. Also then there is a reduction over time but not that much as with AI too. Without AI it stays above 25, so that is flyable enough. But I am coming from 56... On my test setup I injected lots of Ai and culled loads.(I don't use those SimConnect functions mentioned). sat and watched, collected data, checked graphs, no fps drop. Steve Waite: Engineer at codelegend.com
August 8, 20187 yr 36 minutes ago, Rob Ainscough said: Hi Nico, How are you removing the AI object? Are you using managed SimConnect? SimConnect_AIReleaseControlEx and then SimConnect_AIRemoveObject? I'm going to state what might seem obvious so please don't be offended as I don't know you and I don't know your technical background ... when you cycle thru the collection of AI objects to determine which one's qualify for disposal, are you saving off the objects reference to be disposed in a separate collection and then cycling that separate collection and dispose via reference (ObjectID)? Cheers, Rob. Hi Rob, I am not using managed SimConnect. My C++ application is a Windows Desktop (WinApi) program, not a .NET app, I'm using SimConnect_AIRemoveObject with the objectId received when the object was created in P3D as parameter. I call the SimConnect_AIReleaseControl on each AI object as soon as it is created, otherwise you cannot set location, pitch, bank and so on on the object properly. When I remove an object I receive a confirmation from SimConnect that the object was removed via an EVENT_OBJECT_REMOVED, so it all looks okay. My program runs for almost three years now. It is only lately, with P3Dv4, that we have noticed a bigger drop in fps than before when the program runs for say 1 or more hours. So that's why I suspect some issues with removing objects/garbage collection in P3D or ?. Regards, Nico Edited August 8, 20187 yr by kiek
August 8, 20187 yr @kiek Hi Nico :) remember that Rob can replicate similar behavior as well; in this instance he stated he is not using your work, it could be that all three of you are running the same principle that's causing this atm? Edited August 8, 20187 yr by Chris Bell
August 8, 20187 yr Commercial Member I've been injecting/removing thousands of simple objects along with tons of AI traffic of stock aircraft and no errors noted within P3D or SImConnect. Might just be down to a particular object rather than the insert/delete methods. Ai aircraft and other vehicles with less than perfect cfg might cause an issue where they didn't before. There's a lot of possibilities. Steve Waite: Engineer at codelegend.com
August 9, 20187 yr 53 minutes ago, Chris Bell said: remember that Rob can replicate similar behavior as well I didn't? Nico is talking about a drop in FPS over time. I think you are confusing this with the AG issue maybe? My FPS with UTLive will drop (not as much as Gerard's) when the schedules start ramping up and there are lots of departure and arrivals schedule around similar times ... depends on the time of day, airport, etc. I have noted "consistent" FPS drops at some of the mentioned airports pending the direction my aircraft is facing, but the variance is the direction I'm facing and not AI or weather. I personally don't see how Gerard can have so many variables in his testing when trying to "isolate" a problem/issue ... that will just lead to false conclusions ... for example AS4 variance in weather will play a significant roll in FPS over time. For example FlyTampa's Amsterdam Airport Schiphol has no "official" P3D V4 installer ... you install from 1.0 then apply 1.1 update then apply several other file adjustments ... not intended to be negative towards Emilios work as it is outstanding and I'm very grateful he's provided an update but I still feel the airport is perhaps still not optimized for P3D V4.x and has a few issues remaining. In other words, it's just one more variable being added into the mix. Cheers, Rob.
August 9, 20187 yr 6 hours ago, kiek said: My program runs for almost three years now. It is only lately, with P3Dv4, that we have noticed a bigger drop in fps than before when the program runs for say 1 or more hours. Are you noticing any build up of memory usage over time when sitting stationary? Have you used some other external app like LittleNavMap to confirm AI aircraft disposal? Or maybe I should ask, how was it determine they weren't being disposed? Is it possible real aircraft are being picked up instead of the light weight AI models? Copied rather than generate GUID? Have you tried using SimConnect_AIReleaseControlEx rather than SimConnect_AIReleaseControl ? Ex variant being that "Removing the AI will allow for complete manual control of a SimConnect SimObject." Cheers, Rob.
August 9, 20187 yr 1 hour ago, Rob Ainscough said: I didn't? Nico is talking about a drop in FPS over time. I think you are confusing this with the AG issue maybe? i am aware you are each exhibiting different issues all resulting in performance degradation, we can try to triangulate the common denominator all three of you could be running the same time, as all three of you exhibiting this on a none plain 4.3 platform, Edited August 9, 20187 yr by Chris Bell
August 9, 20187 yr Author 5 hours ago, Rob Ainscough said: I didn't? Nico is talking about a drop in FPS over time. I think you are confusing this with the AG issue maybe? My FPS with UTLive will drop (not as much as Gerard's) when the schedules start ramping up and there are lots of departure and arrivals schedule around similar times ... depends on the time of day, airport, etc. I have noted "consistent" FPS drops at some of the mentioned airports pending the direction my aircraft is facing, but the variance is the direction I'm facing and not AI or weather. I personally don't see how Gerard can have so many variables in his testing when trying to "isolate" a problem/issue ... that will just lead to false conclusions ... for example AS4 variance in weather will play a significant roll in FPS over time. For example FlyTampa's Amsterdam Airport Schiphol has no "official" P3D V4 installer ... you install from 1.0 then apply 1.1 update then apply several other file adjustments ... not intended to be negative towards Emilios work as it is outstanding and I'm very grateful he's provided an update but I still feel the airport is perhaps still not optimized for P3D V4.x and has a few issues remaining. In other words, it's just one more variable being added into the mix. Cheers, Rob. ASP4 is off during testing. Just fair weather, so not much of variance. Regarding FT Schiphol I don’t know, but will test that right know. Looking at Schiphol from a distance without ASP4 and without AI for 1.5h to see what the framerate does over time... 5950x3d 5.4-5.7 GHz - Asus ROG 870 Crosshair Apex - GSkill Neo 2x 24 Gb 6000 mhz / cas 26 - MSI RTX 5090 Gaming Trio OC - 1x SSD M2 6000 2TB - 1x SSD M2 2800/1800 1Tb - Corsair 5400 case - Corsair 360 liquid cooling set - 3x 75’ TCL tv. 13600 6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb - 1x SSD M2 2800/1800 2TB - 2x Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - FOV : 200 degrees My flightsim vids : https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0
August 9, 20187 yr Author Without the use of AI : Framerate at start : 54 Framerafe after 1h 40 min : 52 Will try the same but with ASP4 enabled ( no AI ) 5950x3d 5.4-5.7 GHz - Asus ROG 870 Crosshair Apex - GSkill Neo 2x 24 Gb 6000 mhz / cas 26 - MSI RTX 5090 Gaming Trio OC - 1x SSD M2 6000 2TB - 1x SSD M2 2800/1800 1Tb - Corsair 5400 case - Corsair 360 liquid cooling set - 3x 75’ TCL tv. 13600 6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb - 1x SSD M2 2800/1800 2TB - 2x Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - FOV : 200 degrees My flightsim vids : https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0
Archived
This topic is now archived and is closed to further replies.