glider1

It didn't auto update with me when I started the app v1.3. It said they are whatever working on it and closed. But after reading this thread I went into the actual directory with updater.exe and ran it separately. That picked up the new version 1.4. If it weren't for this thread I would have been sitting around waiting for the update forever.

I'm on Rift and love it especially now that we can gain access to the virtual desktop without needing flyinside. Screen door effect not a problem on strong hardware with good AA but I make a promise to myself never to try a better headset otherwise the lusting will begin again....

Wow thanks for the news. I don't want to pour cold water on it but we have to admit there is still a possibility that it is just your GPU working at a lower usage, not an actual fix for the device hung problem. Maybe just maybe it was the reseating that did it? Maybe it was a faulty .NET installation? Hard to believe that one.

The thing there is that the replacement card might still suffer the same fate it is just a matter of time if it turns out that something in the software chain stresses the GPU and degrades it over time above what it can handle long term either by a violation of software specification or unofficially by a bad board level hardware design or bad chip level design. I hear people always referring to 1000,2000 series cards and not 900 series where this happens. Is that true? If it is, we have a big clue. A big lead in this would be to research whether the bug also happens in 900 series or earlier GPU's. From word of mouth it does not. If it does not, why not in earlier series? What would P3D do differently on 1000,2000 GPU series that it doesn't do on 900 series GPU's? Even dynamic lighting can be done by a 900 series GPU can't it? If the device hung bug does happen in 900 series cards as well, then you can rule out hardware design as a cause of this crash because it is too unlikely to span three generations of cards. If it only happens on 1000,2000 series cards, I think it is unlikely that P3D with its old code base would treat 900 series differently to 1000,2000 series which would suggest that it is very likely a hardware level problem in the 1000,2000 series, EDIT or a driver level problem in the 1000,2000 series. But a driver level problem is unlikely, because P3D would be using the same function calls which would exist across all three generations of GPU driver.

Although I think you are right your reasoning is still not water tight (pun). What if P3D is over-driving the GPU beyond specification and that somewhere in the driver specs it says that it is not the driver or GPU's responsibility to handle it? So if that is true, P3D exceeds the GPU spec, the GPU crashes and P3D handles the crash. It would then still be P3D's fault wouldn't it? It would be sad engineering practice I admit but is such a scenario possible perhaps via a bug in the GPU driver that allows P3D to exceed spec?

The theory you have to discount is that something in the older drivers isn't working the card as hard as the newer drivers either in memory use or heat. Since the card is not working as hard, it doesn't corrupt the memory and there isn't a DXGI hung error. I got the dreaded DXGI_ERROR_DEVICE_HUNG error on a brand new 2080 today. It was in VR at night in the cockpit of the Flight1 King Air. At the time I was wondering why my memory usage was 7GB out of 8GB because I was at only one regional orbx airport and the card load was ~85%. The card is factory overclocked. I did not reboot and just restarted P3D then did a six hour flight in the day with scattered clouds not a problem but the 2080 was only loaded to 50-60% usage most of the time and memory usage was lower because after takeoff there was no more orbx scenery in the flight plan except for the region at FL290. I doubt it is an P3D problem.

Just an update, I got VR a lot smoother once I turned off MFAA in nvidia settings. It was a bad idea even though it is advertised as a more efficient version of MSAA. At this stage I'm seeing 4.4 is not a performance degrade from 4.3 it is at least as good except for visual PBR related oddities.

Thank you for reporting this Joe. Lets hope for the best that A2A can do something about it.

I don't understand why it is worth doing all that wonderful work for the community when the results only apply to a tiny number of possible simmers that can afford extreme hardware like that. There is nothing I can take away from these tests as information for a mid-range hardware setup.

I doubt flyinside is needed any more Jozeff. Maybe on a low end system. I'm not sure it has even been updated to 4.4 yet. I stopped using flyinside once P3D 4.3 hit the scene and LM optimised VR more. Native VR might have lower frames but it is the smoothness that matters.

Update on today's session trying to get 4.4 running as well as 4.3 in VR. Things are looking better. I made the classic mistake of not having "prefer maximum performance" turned on after I did a video driver update. Instead of using SSAA, I switched back to MSAA and teamed it up with MFAA in the video driver options. AA is good and more smooth. Carefully tuned the autogen distance slider which is a huge stutter inducer if not set appropriately. Reinstated FFTF=0.01 to lessen long frames in ORBX scenery close to the ground. ENVTEX/SHADE work very well in VR. Chaseplane works very well in VR so long as the flightsim window is not dragged to the screen edge which kills single pass rendering. Definitely the older aircraft that were not supported in P3Dv4 are no longer any good because of PBR incompatibility. Definitely some of the ORBX textures are not PBR render compatible for VR (the ones that try to simulate light on the ground around light poles at Avalon). Definitely there is a change in transparency for some cockpit windscreens that is a bit odd in VR but just take get used to. Definitely water reflections still don't render properly in VR but they never did (just turn them off but shadows work well). Definitely boat textures are slow to load but at least the wake renders correctly in both eyes. Definitely the pointer is weird in VR because it does not compensate focus to both eyes with distance (but still usable). Definitely could not get the VR menu working (but doesn't matter because Oculus home desktop is much more useful anyway). My conclusion is that all-in-all, performance is definitely on par with 4.3 which was already the equal of flyinside. A bit more work and 4.4 should be good to go in VR if you have hardware that is the equivalent power for 4K flatscreen simulation. I was almost ready to go for a fly and had powered up the Flight1 King Air B200 only to find that the G1000 wasn't working properly in 4.4 but an aircraft reinstall fixed it. No flying today.

A recent Oculus update has integrated VR access to the desktop really nicely now. When there is a need to see the desktop while flying P3D you hit the oculus home button on the remote and it switches the VR view to a depiction of the desktop in VR with the flightsim window and P2A next to each other. The clarity is pretty good too. No need to take off the headset any more or peep through the whole. Yeah! Just hit the oculus home button again and you are back in full VR.

Do you think that small low budget developers with old non-PBR textures will be able to quickly and cheaply convert over to PBR or is it technical hurdle for them?

Hi Jozeff - so far early days but it is touch and go whether it stays or back to 4.3 which was working well. It might sort itself out over time. I only did a client update. There are new issues with PBR. Transparencies on the cockpit of some older 3rd party aircraft don't render clearly any more. It seems the artificial textures that were put on to simulate reflections in the glass are now more solid textures looking wrong in 4.4. There is also issues with the mouse pointer which doesn't change size with distance any more making it hard to manipulate panels and switches on some aircraft because you have to guess where the center of the pointer is. I'm also having more troubles with performance in ORBX sceneries but performance in default scenery is working fine. There is an issue for sure with the frame rate counter which doesn't match the experience in the headset. Might have to drop back 4.3 which was working more reliably and smoothly even though it is less evolved than 4.4. Time will tell.

Is there a noticeable frame rate hit running SF and AS engine? I'm on the limit running the sim in VR and G1000 in urban environments. I'd run SF exclusively but need historical weather. Thanks.

I'm using a 2080 in VR because my 1080 is in repair under warranty. It is possible to crank up shadows and increase SSAA=4 and Oculus supersampling to 2.0 with a GPU load of roughly 75%. The image in a rift first gen headset with this heavy AA is very usable image quality for GA and good enough to wait for second generation headsets to arrive. Overall smoothness of head tracking has improved slightly but it was already very good. Downside is that most of the sim is CPU limited so there is little FPS improvement and when the bank angle is significant and the terrain is moving quickly there can still be long frames on SSAA=4 but not on MSAA=8. Overall I doubt the performance gain is worth the price because a decent 1080 drives first gen VR very well in P3D with only slight concessions when the settings are appropriately applied. EDIT: If you have a lot of money to burn an RTX card is still "nice".

I use P2A in VR inside native P3D with rift and headphones and MCE for voice commands. Framerate hit is not significant. The few times I need to interact with P2A during the flight, I look at the flatscreen through the rift peep hole and the immersion is not broken because your eyes are still inside the headset space.

I spent hours on this yesterday - sometimes you have to load the flightplan into P3D twice because it can ignore the waypoints the first time you load it. When this happens, you think that there is a problem with the flightplan created by the external app but it is just a very old bug in P3D that is still around. I proved that it is a bug by creating a flightplan inside P3D with waypoints on GPS or VOR, saving and loading it and the waypoints are sometimes still ignored.

"Recommended" is too weak and "Required" is too strong. If Orbx would have thought this through the solution is to say: "The product will not be supported unless installed with the region"

Love creating a small bonfire in FirefighterX next to the field when spending the night out bush camping in the A2A 182. Looks great in VR.

I am using all of the Aerosoft gliders in P3D 4.3 without problems although they aren't officially supported. They are the only ones that work really well in VR. EDIT: Just do a bit of research beforehand if you want to buy unsupported gliders there are a few things to know. For example the ASK21 radio needs a tweak to work.

You can do that in P3D with FireFighterX by Lorby that generates thermals over fires.

I think you should try vsync on and turn off ASW. I think you will get smoother experience. This is because in P3D 4.3 with vsync on it locks to multiples of the headset framerate so for rift on 90 that is 45, 30, 22-23. You don't need ASW because of this. The final FPS is lower but smoother. I think you should not use 2x SSAA which would very likely overdrive your GPU. MSAA is better.

Does VR behave the same way as flat panel when it comes to performance in sli though? We need someone who actually uses VR to do tests in VR P3D.

Yes blueshark is right. Don't use OOT supersampling flyinside has it built in. You will only overload the GPU.