vicx

The feeling of walking beside the nose, looking closely at the rivet lines, crouching to get a closer look at ordinance, craning ones neck to see over a wing ... all this can be done in your room. A hanger would be ideal but a hanger is not needed to give you the FEELINGS. I know because I have tested it. DK2 was not suited to the testing (I ran out of cable before using 1/4 of the room and lost tracking a lot) but Rift will have a larger camera volume and longer cable - it might be better.

You have seen SteamVR results table for yourself so you know the 980TI dominates as it should. What we need to keep in mind is that the SteamVR test uses either the Source2 engine or Unity engine with a custom SteamVR renderer. The results are only useful as an indication of the performance you can expect from SteamVR launch titles or games from teams that have worked closely with Valve The SteamVR test is a best case scenario test and your hardware will probably not run as well for games from other developers. I suggest DCS falls into this category. I have no idea what CPU or GPU configuration combo will give you 90fps with no dropped frames while flying over Vegas in DCS. This is probably why DCS can't be used as a VR poster-child.

The person on the nets complaining that their TitanZ should be rated higher for VR is dreaming. Sorry guy but the TitanZ only has 48ROPS which is the same as my poor GTX780. Sadly the TitanZ gets hit with the extra whammy of also having a rather slow core clock. The reason these high end cards are so poor for VR is that they were made when VR wasn't on the radar. Nvidia dedicated all the silicon in these beasts into shader processors and texturing units. These cards are total beasts for compute and they can do amazing effects in real-time ... but what they can't do is consistently render 4K worth of pixels under 10ms. It is telling that the heavy hitters in the SteamVR test all have 96ROPS and pixel fillrates to match. Also doing well are AMD cards with 64ROPS. It is pretty cool how well the R9 290 does. It really is the best "cheap" VR card you can get right now ... if you can find one still for sale. The GTX970 is not as good as the 290 with only 56ROPS but it does have a pretty high core clock which is why it does so much better than the Nvidia 700 series cards which are ~200-300Mhz slower.

Yep but noone here really knows how VR performance is going to pan out in DCS for Vive OR Rift. I'm not in the situation of buying a first gen HMD just to play DCS so I won't be too worried about DCS performance in the short term. It might be the case that 2nd generation HMDs are available by the time that DCS performance becomes truly viable for VR. This might be something to consider if you were planning to buy a HMD "exclusively" for DCS.

It is my experience that most dials and switches in the cockpit are rendered with sufficient detail to be readable and usable. The navigational MFDs "can" be harder to read small text on so that is the weakest point in the VR cockpit ... IMO. For these reasons I found the Mi-8 to be the most VR ready cockpit. All the dials and switches I can be read and you can fly all non-combat missions without any problems and "many" combat missions can be flown too. It is a shame that the Kneeboard does not work in VR. We really NEED the kneeboard for VR.

Youtube will be filled with videos of people playing DCS in VR and then the VR horde will descend over these boards.

Run DCS repair. Missing textures can do this.

Ahh you guys talk about sound. A thought experiment ... how would YOU go about modeling very realistic sound in DCS ... in VR! I think you have to extend the present model to enable worthwhile VR sound. The present model is good and being improved but AFAIK it is based on stereo separation of a single point sound which uses single point position, orientation, velocity, and acceleration relative to world sound sources to build a stereo matrix. This model has worked well but with VR HMD tracking it is now possible to do so much more with spatialised sound. Now ... how to do it. *AND EARN BONUS POINTS for non-VR users.* I would not take the existing model and just add VR orientation to it. I think a MORE logical model extension would generate a matrix based around the cockpit and then place the VR HMD inside that space. This is because for VR purposes it would be nice for the sound to attenuate more realistically to all the orientations a head can have inside the cockpit. Example tilt head on side. Left ear facing the seat, right ear facing the canopy - you would expect to hear heavily attenuated sounds from the left ear (seat direction) and less attenuated and more heavily directional cues from the right ear (canopy side). In all honesty mostly wind noise and engine noises right? Also sounds would be attenuated based on frequency and impulse energy ... more attenuated higher frequencies through the seat ... basic stuff like that would suffice. So ... logically we need a cockpit matrix for the VR head to use and we need a minimum amount of points to add some complexity beyond the existing L and R side of the cockpit. The number of points required depends on how complex you want the cockpit model to be BUT I think a cockpit has logical directions for sound to come from and that provide distinct characteristics. A common cubic model can be used. - Front Cockpit - Rear Cockpit - Left Cockpit - Right Cockpit - Top Cockpit - Bottom Cockpit A very nuanced model could be built using an actual cockpit and recording frequency sweeps and impulse noises to build a cockpit specific matrix but I think a hand made model could be just as convincing. No need to add difficulty to a big task. Now with FC,RC,LC,RC,TC,BC you have a six channel cockpit sound matrix to feed a stereo model representing the head. The head will have a position and orientation inside the cockpit matrix. And now we finally reach the part you guys have been arguing about. Q. Do you use a HRTF model based on the physiology of human skulls and mix the cockpit matrix using this model or do you use a hand made matrix which approximates the effects you would expect? A. It does not matter that much because the more significant work is done elsewhere. The HRTF model kind of has to be a head in helmet type HRTF model. Which is probably available in certain parts. Otherwise you can approximate the effects you would expect to hear through the helmet and skull and sounds you would expect from the front/back/top/bottom as heard by ears, in a head, in a helmet. Personally I think that the HRTF vs Hand-made stereo model is a premature argument. Other things are more important. First we will need a cockpit sound matrix with extra dimensions so that that sound differences based on head orientation are accurate enough to care about AND rich enough that it is ACTUALLY worth feeding into another matrix (HRTF or otherwise). The good thing about this approach is that ... a cockpit matrix would lend itself to be used by non-VR users with 5.1 sound systems and physical cockpits. This way we please VR people and non-Vr people at the same time. It could be LOTS of people happy. These are just my own ideas. You will have your own.

Really disappointed to see how CPU bound frame-rates are. This does not bode well for VR performance where keeping frames over 75fps is crucial. Is there any testing you can do to see what settings can raise the lower CPU bound. Everything set to LOW and Traffic:OFF maybe.

When people post Tacview playbacks of simulation gone bad ... it isn't the aerodynamic perf that has people face palming. It seems like the modelling for guidance and counter-measures is not heading in the right direction.

Extra features would be nice but I'd be happy to see the existing DCS features work in VR. Kneeboard but also the optical weapon sights in Combined Arms do not work in VR. No zoom and the overlay does not stay locked to the weapon station orientation. If I was to prioritise ... I would say fixing the CA sights is the more technically challenging task and work there could unlock a LOT of potentially awesome VR features. Collimated optical telescopic sights in VR. A difficult task but once it is worked out ... lots of applications.

Those hover junkers videos are awesome. Thanks for posting that one DE, I was just laughing pretty much the whole time. It is so impressive how using even very primitive IK on just three inputs can be so effective. Sometimes comical but really adding something special. I think the low latency tracking is what makes this work. We have had hand and head tracking before but it never worked this well. I was struck by that watching this video a few days ago DCS HMD tracking shows in the external view. I love that but you have to get pretty close to see it.

Looks like we got the seeds of a VR squadron happening right here. Yeah I feel this way but I completely understand that it isn't for everyone. In a lot of ways playing DCS on a 4K TV is probably how you should be playing Nevada. You can't fully appreciate the level of detail and in Vegas in a HMD. VR players are not mainstream and won't be for a while. I have this idea that maybe instead of making noises about Vegas and it's lack of suitability for VR .. we could focus on the Black Sea Update instead. It would be great if the Black Sea Update could be undertaken with VR compatibility in mind. I'd be totally happy with a constant 75fps in the Caucasus Map with optimised assets and draw distance for VR HMDS.

In a DK2 reading instruments is doable ... I expect the optical improvments and the 25% better resolution will sufficiently improve instruments readability in the CV1 and Vive. IMO there is no better way to experience a DCS cockpit than in VR. Even in the DK2 it is absurdly awesome. The spotting will remain a problem and this is why in another thread I mentioned that certain experiences in DCS lend themselves to VR better than others at the moment. Formation flying ... awesome. The presence of other aircraft flying close is stunning. Low level helicopter flying ... awesome. Landing and slingloading are easier and more fun. Some combat works just OK ... I have found that using the Ka-50 with the schval is pretty good. Using model enlargement with Mig15/F86 is OK but you can't see when the enemy aircraft changes direction that easily. I can't see it straight away and that means I lose my position. I'm a complete novice so perhaps others have an easier time of it in VR than I do. So yeah combat is going to lag some of the other uses ... but the other uses are going to grow in popularity. I can see formation flying and aerobatic flying in VR being a BIG thing. And then there is other stuff like multicrew and CA vehicles in VR. It is going to be interesting.

Aka a fingertip :)

I think there are some experiences in DCS which work really well in VR even with the limitations of the first generation hardware. - all of the cockpits feel completely amazing in VR. There is no better way to experience DCS cockpits than in VR. The cockpits are very compelling and are completely usable. Some switch labels are harder to read but moving your head closer solves this in most cases. - the compromised resolution of VR means that spotting without using sensors is much harder. ED have provided a sprite system for targets that stops targets disappearing completely when they should be in visual range. It works but you can't ell what the target is until you get pretty close. This is a definite downside. - spatial awareness for objects 0-500m away is extremely compelling. Lots of activities in DCS are benefit a LOT from this; formation flying is amazing. Flying choppers in general is really great especially low altitude flight and slingrope operations. :thumbup: - CA feels amazing but only driving vehicles really works at the moment because the sights in CA are not yet VR compatible. If CA vehicles were given VR compatible sights ... CA in VR would make many people VERY happy. I could write more but that is enough for now.

Oculus Runtime 0.8.0.0

Watermanpc, you have several forests in that view and I think this is why you are only getting 22fps. I have done a HEAP of testing of 1.5 and one thing I can say without any reservations is that TREES KILL the framerate like nothing else can. And forests are the worst.

Great work Why :thumbup:

You aren't making ANY valid points about the controllers at all. Both the Vive and Oculus touch controllers require microcontrollers, as well as the standard 'instantaneous' motion sensors, batteries, battery charging circuits, wireless transceivers and haptics drivers. The cost of a dozen or more photo-diodes is a matter of a few cents.

You have mixed up a few concepts in that statement. These videos should help you out. How the lasers scan. How the electronics work. The sensors on the Lighthouse controllers and HMD are very simple components; no more complex or expensive than a basic photo-transistor. The sensor fusion on the controllers can be done by cheap micro-controllers. IMO the haptic part of the controllers is a way more complex than the positioning component and yet it too can be produced cheaply. The Lighthouse emitters are not necessarily expensive to produce. The emitters I expect will have undergone serious industrial redesign and this is one area where I am very curious to see what the HTC engineers have come up with. A Rift with one camera and xbone controller is going to be cheaper than a Vive with two lighthouse controllers and two lighthouse emitters. BUT a Rift with two cameras and two constellation touch controllers approaches parity with the Vive in terms of cost.

Targeting radars are defined the same way as turreted guns WS WeaponStations? in a unit's lua definition. Most of the naval units have a damage model with hit boxes for the areas like the turrets, towers and decks above the hull. WeaponStations can be assigned to a hitbox zone so that if that zone loses all it's hitpoints then the WS will be disabled. Gun turrets are disabled with hits to the turret hitboxes. I think targeting radars are assigned to the tower hitboxes if they don't have their own hitbox. The main search radar is not a WS so not sure if it can be disabled.

Yeah I was expecting people to nominate some first generation LGB platforms but I didn't want to steer responses. Have done reading and found it interesting that the first designating laser systems had optical sights and not TV displays. That was pretty surprising.

LN Please be testing the VR performance for this map. Maybe we can have a special setting in graphics menu graphic.options:VISIB DISTANCE = VR | Low | Medium | High | Ultra | Extreme I can see in terrain.cfg.lua that you can load separate object sets based on this graphical setting.

I was not expecting the M2000 as an answer but it does fit. It ends up being way cooler than almost every other possible candidate. Very unexpected.