Baldrick33

I certainly wouldn't (and didn't) declare we have no room to complain. The issue is that OpenBeta is the de facto release. My FB feeds have a few DCS VR groups and they were quickly awash with people who had no understanding of why DCS had stopped working. They just clicked on DCS, it updated and bang it didn't work. As I said, I don't know how we got here. ED announce themselves that Open Beta is best for multiplayer on their download page for new players, so the players are hardly to blame. I guess at some time there was a tipping point when multiplayer servers needed Open Beta for essential fixes and it kind of spirals from there. Critical mass is everything in multiplayer. Single player wise I would suggest most of us want the new stuff and latest updates. So OpenBeta has long been the "normal" version for most players (both multi and single player) The problem is that it renders the very purpose of a Beta and Release version somewhat moot. Something like an initial version of OpenXR is ripe for Open Beta testing. There are a myriad of headsets, APIs, third party tools plus VR can be a very subjective experience where one person's perfect settings can feel horrible to the point of physical sickness to another. It needs a wide test group. Now of course had the original patch not defaulted the change to everyone a lot of anguish would have been avoided. We are right to complain about that, I have complained about that and ED have appeared to take that on board not to repeat it. That is the purpose of post mortems of IT issues - to ensure it doesn't happen again. I feel the message has got across, time will tell. That doesn't completely take away the risk that stuff won't get broken again in Open Beta. If the Open Beta was used by people who are prepared to help develop the product by providing a degree of field testing then it would be fine but it isn't. Far from it and that in my view is where some of the problems lie. Personally I would advocate a third branch, call it Open Alpha, whatever it doesn't matter. Get stuff tested in the wider field where we have so many more variables of hardware and software configurations - before it gets to Open Beta where the majority of normal customers exist. The multiplayer "version" of DCS should in my mind be reliable and balanced and having a constant feed of new stuff doesn't fit with that. Just my two cents!

I have repeatedly tested it with the OpenXR Toolkit disabled. It consistently crashes (hangs) when loading a mission in the terrain loading part. I note that others have it hang in the menu screen. I have removed any third party hooks like openkneeboard and jetseat and it still crashes. I changed the headset to 60Hz and it hung on the menu screen for the first time. Turning off motion reprojection and it runs fine. Reverting to the opencomposite method of enabling OpenXR also works fine with motion reprojection on.

Not for launching DCS but it should work the same.

When I update DCS I copy the \DCS World OpenBeta\bin\openvr_api.dll to a backup folder I then overwrite this with my opencomposite openvr_api.dll I have downloaded from OpenComposite | OpenXR Toolkit (mbucchia.github.io) I have added a shortcut on my desktop to DCS.exe (also in the bin folder) and added the parameters to test native (after putting the OpenBeta dll back)

To run it natively you will need to add the following to a shortcut to DCS.exe --force_enable_VR --force_OpenXR I use the per application method of opencomposite so I also restore the latest openBeta openvr_api.dll To revert to opencomposite I just copy over the dll and remove the --force_OpenXR from the shortcut. As I prefer motion reprojection the native version doesn't work (known bug), so I am using the "old" method.

I agree with the lack of communication being a mistake but I think that has been accepted as such and the point has been made. As for the player base testing it, well that is the purpose of Open Beta. There far too many variables of hardware and third party software for a small team to test everything especially for an immature and relatively complex aspect like VR. The real problem is the numbers who use Open Beta as the de facto release version and update it automatically to find things no longer work. I am not really sure how we got here but I don’t think many really sign up to the stated purpose of Open Beta shown below when they download it: the Open Beta version is for us to "field-test" new changes and check for any DCS-wide "Blocking" crashes, and the Release version is the result of the previous Open Beta that does not suffer any "Blocks." As with Early Access, Open Beta is an option and may not be for everyone.

Exactly. Now the priority has been changed DCS loading in VR will try to load in OpenVR and the opencomposite DLL will intercept it as it did before.

With reprojection on or off?

Just checked and same here. I use Numpad 5 for OpenXR Toolkit and it is now recentering VR although not bound in UI layer

I may not have had that old a version, I will try again. No, it hangs with "--force_enable_VR --force_OpenXR" part way through mission load, just like the new dll. The only way I can get reprojection to work is by using the opencomposite dll and forcing it to run SteamVR, Then it runs OpenXR fine with reprojection enabled.

I believe it is hidden so that it doesn't change the behaviour for all those people that were happily using SteamVR who suddenly found it changed. Motion reprojection was believed to be fixed but as sometimes happens not for everyone. There are a lot of variables. Hopefully ED can track down the reason it doesn't work for some of us (me included)

It doesn't work for me if I force OpenXR. If I force SteamVR it does (this is just doing it the old way and hijacking SteamVR to make it run OpenXR) I guess with today's change if you don't force OpenXR it will run SteamVR by default and then opencomposite will do its thing by having the modded dll in the bin folder. Edit: Just tested it. Today's patch makes the opencomposite dll swap work as before, by changing the priority order, (the same as adding force_steamvr yesterday)

I get the same with Windows 10

Mine hangs part way through loading a mission. Turn off reprojection and it loads fine. dcs.log

--force_enable_VR --force_OpenXR

VR is still pretty immature and yes it can be hard to keep on top of everything. IMHO it is worth it though. We are fortunate to have guys like @mbucchia @edmuss@nikoel @speed-of-heat and others who share a wealth of knowledge here. Following them is a good start! One of the biggest issues is the way we (the community) have adopted the beta version of DCS as the "normal" version. By its nature a beta version will regularly change and undo stuff we had working great! ED did mess up with not giving a heads up such a massive change was coming for VR users but the fact we typically update as soon as a new version comes out exacerbates the issue in my view.

Coming form a corporate IT world having such an array of hardware and third party software is nigh on impossible to support anything close to well. The simple answer would be a couple of mainstream headsets are supported and no third party products other than those that are supported directly by the headset (e.g. SteamVR) If we want all these devices working then we have to reach out to the community - which in fairness does a pretty good job here. It is just blips like this release that throw a spanner in the works but look how quickly the community have worked the issues it out. TL;DR be careful what you wish for!

For simplicity I would pick Joy2Key.

I have a G1 and the toolkit works (without reprojection) , are you using turbo mode?

I just dive in and try stuff - just like my approach to flying which is probably best left to simulators

Are you running with motion reprojection forced on? In which case you will need to disable it until a fix comes.

For some of us anything less than 90fps without reprojection produces unacceptable stuttering. It seems to be one of those "how you are wired" kind of things. Also... Some can handle 60Hz Some find it smooth without reprojection at 55fps Some hate reprojection

You will get double bindings - both f & g will be sent to DCS. You will need to unmap the physical stick buttons to avoid multiple actions. DCS adds some defaults to a new device like the view controls - it will do this for a virtual device too.

You can do several things. There is a shift option which is like a modifier though I haven't used this. You can also map short and long presses to do different things. So a long press on a button could do the things you suggested. You can also rotate between inputs. So a switch could do the take off actions and next press do the landing actions for example.

It is down to preference. I did hide some of the physical button box devices. I preferred to use the native axis of my stick & throttle rather than the vjoy ones, probably out of paranoia in case vjoy added any latency, so I didn't hide those and mapped the axis directly to them in DCS, then mapped the buttons to the vjoy device.