RealDCSpilot

Just for fun, that's my theory...

Maybe you should give lessons as the BRU whisperer

I think it's the opposite. DCS just doesn't do the full calculation for the BRU sensor atm. They took a shortcut by assuming every head position is in the same spot, especially when it comes to height. That's why monitor users have no issue in 2D.

ED is working on it already. There are enough other threads about troubles that are connected to boresighting, even for TrackIR users.

No, i just want it aligned when i look straight forward centered at horizon level on ground.

We are running in circles here (not concentric ), ED has it on the list. Let's see what they will do about it.

That's not what he meant. Try boresighting with looking way off the BRU and check how your PNVS looks after that.

Yeah, makes all sense if you get it in the right ears. Of course the PNVS will match less when you are looking more to each side or up and down. The mismatch will grow the more you look away from the center, that's the paralax between your eyes (head) position and the sensor package on the nose. But if you are on the ground and look straight forward center, reference points like horizon line and objects should match almost perfectly. An offset of the PNVS in left or right -> after boresighting, WHEN LOOKING STRAIGHT FORWARD <- is really bad because your IHADSS will always aim with that offset later when you start attacking a target in flight. An up and down offset would be much easier to compensate. Checking your PNVS against the front center view is just a method to verify the result of boresighting in VR, because you can easily mess it up there when only using the currently provided state of the concentric circles method.

I only see this as a workaround for now. JSpidey's method tries to find the position where monitor players have there fixed head position. My method is only the refinement for whatever everyone elses head position might be in VR.

The FLIR image comes directly from the sights (PNVS and TADS), it's center points are the primary targeting point which is symbolized with a crosshair on all the internal monitors. The helmet monocle is nothing else than an augmented input device that needs to be aligned as best as possible because it's crosshair is just an overlay and does not represent the actual targeting point as long as you don't align it correctly. That the FLIR does not need to line up when you are straight looking forward is nonsense. Try it for yourself, an offset like in the first my screenshots (a bit to the left and down) will always lead to missing the target by the same offset when shooting the gun. @key_strokedI also renamed the thread to make it more clear that the problem is only affecting VR as it seems. It's also a bit logical because the current version of the concentric circles seems to be made for monitor usage only. For monitor players the head position is always on the same spot, a fixed known position in 3D space that can be taken easily into a triangulation equation. In VR it differs like in real life, every pilot comes in a slightly different size and that results in a different head position.

As long as ALT hold mode is not implemented, i would say how the FCS works isn't finished anyway. We have to re-evaluate the handling again at a later point...

See, this is the problem, there should be no "my way" or "your way". We need one exact method that works the same for everybody without the fiddling.

Listen more closely to him, he is talking about the parallax offset as soon as you start looking more far left or right... This is how i fix the problem:

PNVS/TADS video feed is the direct sensor targeting feed, boresighting makes your helmet sight synchronize and match with it. With some additional adjustments it is possible right now, but the realistic concentric circles method doesn't provide this in it's current state.

Sorry, but you get this wrong. The whole sensor pack on the front ist 100% aligned to your gun and the other weapon systems. Your helmet sight needs to be aligned with them when looking straight forward. That's the whole purpose of boresighting. The only acceptable offset, which comes naturally, is the parallax offset the more you are looking left or right and up or down, because of the distance between the point of view from the nose and the point of view of your head.

The problem with your video is the lack of checking the result at the end. By simply bringing up the PNVS FLIR in the helmet, you can check if the whole procedure was a success. Your method is not working on my end by the way, the PNVS always shows that it's off by 10-20 meters for a target that is 100 meters away for instance, where it should match almost perfectly (checked against buildings, vehicles, horizon line in front of me).

The problem is not how the procedure works, the result is the problem. If i do it like in the video my FLIR and reticle alignment are way off, if i order George to look at a target with this alignment he starts looking 100 meters away from target and also the gun aims at the same offset. This is not how it is intended to work and we need ED to investigate this issue. Right now i can only correct this with having the FLIR active and try several boresighting clicks somewhere at the boresighting unit to find the best match. Good thing that a bindable boresighting button is available.

I don't think that's a good idea. They would need to add a triangulation loop that is constantly running every frame to check for the VR camera rig's position and rotation to make sure the alignment is always spot on. I would like to have this extra workload limited and only applied during the boresighting procedure. For me, it looks like the source of the problem has a pretty simple reason. At the moment the "calibration" is only done against screenspace, a simple 2D parallax matching process, sufficient enough for monitor use and only mimicking the real procedure. But to work correctly in VR it needs to be applied to world space in three dimensions and has to work much more like the real thing. They have to emulate a virtual TrackIR system in the cockpit so to say...

Yep, i'm using basically the same procedure. With the fixed mouse pointer crosshair you get the absolute center between your eyes, first boresighting calibration pass goes with crosshair over center of the boresighting unit, second step is refining with PNVS FLIR overlay. Two or three B/S clicks with little offset from the boresighting unit later the FLIR overlay is aligned almost perfectly.

@BIGNEWY Can you please remove the "solution" mark from the thread? We are not there yet. Thanks!

My guess is that the current state of boresighting is not "calibrating in 3D space" like IRL. It's simply using a paralax method which will only work correctly in 2D on a monitor.

@comcat You can exit the target list without confirming a target by pressing "left" again.

After boresighting check the result with PNVS FLIR overlay in the IHADSS. By straight looking forward the FLIR image should fit exactly over the outside world. Only then, the gun and everything else related to the crosshair in the IHADSS will be as accurate as possible.

Best way for me now, to get pinpoint accuracy, is bringing up the PNVS after startup and check it's alignment against the world at the front outside. If it doesn't match (like a bit to much up or down or right or left) i look at the boresight reticle unit and press the B/S button (bindable) and do little corrections with my head repeating using the B/S button until the PNVS image fits perfectly. From this point the aiming with the gun works pretty well (it also effects how precise George is looking at a specific point you command him to look at). About all the fuzz here in the thread, it's pretty obvious that the concentric circles method is only working for monitor players where the head position is fixed and correctly lined up. ED needs to enhance the calibration functionality to make it fully work in 3D space for VR.