Frusheen

The new slider with deferred lighting is definitely an improvement. Normandy looks very good. Unfortunately we still need to turn off MSAA to maintain at least 45fps so the jaggies and shimmering is pretty bad. Hopefully a fix for that will come sooner rather than later as the alpha progresses. Overall it is definitely an improvement in VR.

That is the only viable solution. None of these vaporware motion platforms that keep appearing are doing anything to cancel platform motion from head motion. It would require custom code in the headset SDK. It may be possible with a vive tracker puck at some point but is currently not possible at all on rift without additional hardware. The rift camera only updates at 60hz which is nowhere near fast enough to eliminate unwanted view changes when mounted on a platform. Recent Oculus updates have changed the tracking software to such an extent that the current runtime performs worse on a platform than it did at rift launch. I'm hoping Oculus will use some of the samsung inner ear tech in the future. I think vestibular stimulation would work well with my G-Seat.

Once again they are using a dcs cutscene. DCS output.lua does not support 360 degree rotation. The video you are watching is not a live recording of motion output from DCS it is just marketing hype and false advertising. EDIT: as I've previously posted they have also not shown any form of motion cancelation for the headset. No placing the camera on the platform does NOT solve the problem.

That's a fair point. In my case, with my setup, I still have very poor performance. Point being, it's less about hardware and more about poorly optimized alpha software. Hopefully there's plenty of room for improvement and now that 2.1 is out the door ED will start focusing on optimization. I agree there is not point comparing FPS given the amount of variables.

I don't think the single core CPU performance is the bottleneck in 2.1. I never see any core go over 40% yet my GPU easily hits 100%.

I have it set to high on my setup. It doesn't make much of a difference. I can't believe the tree are meant to look the colour they do. I'm hoping it's an issue with the new lighting engine that will be fixed but that doesn't help with level of detail problem.

This isn't a rant. I appreciate we are dealing with alpha code. I'm just posting to highlight the issue and perhaps others will agree and it may be addressed at some point. The trees in 2.1 appear to be an almost fluorescent green. The colour is completely different from the underlying ground textures. Forests look absolutely terrible in VR. I'm not sure if this is a texture or lighting issue but I hope it is something that can be addressed. In addition, the level of detail increase causes the trees to change shape and colour as you get closer. The difference in the LOD is so dramatic that it is really noticeable. Flying a helicopter over forestry you see a cascading wave sweep over the forest just below the helicopter as the tree detail is increased. Is there any setting which can cause this to happen further out so as not to be so jarring?

That makes sense. Thanks

Does anyone know the reason for this? Full post is here: https://forums.eagle.ru/showthread.php?p=3150975#post3150975

Love it Trip. Looking forward to more updates.

Looking forward to more news on this. I wonder where the Airbus arrangement stands now? does the authorization move with Oliver to Polydynamics? I presume so seeing as they will be doing the 105 but that raises the question of the continued development of the Gazelle and its flight model! Anyway, rhetorical questions that I don't expect will be answered here so just wishing good luck to both dev houses.

I would say no also. I doubt you will get high enough framerates.

It's the same module and does not need to be purchased twice. The installer for the module will recognize which version of dcs it is being installed too and will download and install the files appropriate to that version via the module manager.

For those who believe the Gazelle is modeled correctly and asked for specific flight model issues to be listed rather than 'issues with feel' this video from MAD_MIKE is a prime example and is a specific issue which I have listed in several previous threads. That being, the lack of interaction between torque and cyclic pitch. I don't care how many real world hours anyone has on the Gazelle, stating the flight model is accurate while this issue is present is a clear indication they are either not taking the testing seriously or choose to ignore the obvious flaws. Probably unknown to many (as I never saw it listed in a changelog) Polychop did include a flight model update in 2.0 which has the helicopter hover one skid low. If modeled correctly then this behavior should not be possible in 2.0 so it is worth testing for comparison. I think it only fair to point out that the FM is obviously still receiving attention. Credit to Polychop where it's due. It's high time the 'real pilots signed off on it' argument be put to bed though as the above video clearly illustrates more work is needed.

Yes wipers are the best solution just make sure you isolate them properly as they normally use the housing for ground. Not good when you use them as we do to run in both directions. I'll update this build thread when I find time. The project had to go on hold for a while due to other commitments but I hope to complete it over the next few months. Brassem has a great build thread on his website. He is a faster worker than I am lol. http://www.brass-em.com/portal/work-log/136-g-seat-brassem-version-category As does Bergison on his website http://bergisons.simpit.info/g-seat-how-to

checked the output: velocity vector gives Velocity.x Velocity.y and Velocity.z angular velocity gives RotVel.x RotVel.y and RotVel.z (expressed as radians per second). I would love to express the angular velocity as a g-force and try it with my g-seat build in place of roll, pitch and yaw. Will see what we come up with.

This is progress guys. I have also discovered from Ian's BFF lua code that we can get velocity vector, angular velocity and vertical velocity with: local velocityvector = LoGetVectorVelocity() local RotationalVelocity = LoGetAngularVelocity() local VerticalVelocity = LoGetVerticalVelocity()

No need to upload Trip. I found it on his website. :thumbup:

I'm familiar with Ian's work (I purchased his shaker card and built one of his bass shakers for x-plane). I'd love to see the export lua as if it is the same/similar to that which the rest of us use for x-sim or simtools then he has figured out a way to solve the issues we are all having within his own 6dof driver in which case I'll give him a call. He didn't support dcs the last time I had contact with him but that's a few years ago now. The current lua output most of us are using outputs -180 to 0 to +180 on yaw. Once you pirouette to +180 degrees and continue to turn the output flips to a value of -180 and goes back to zero as you complete the turn. The result on a 6dof Stewart would be an abrupt deflection from full right yaw to full left yaw after turning 180degrees. If you continue turning it would go back to the neutral position then the process repeats. Please do share the file when you get a chance.

That's interesting. If you preform a pirouette in a helicopter do you not get a sharp reverse in direction after the yaw goes through 180degrees? Could you share your export.lua file?

Trip the problem, in a nutshell, is unwanted interaction between axis. For those with full motion rigs the output doesn't support 360degree rotations. For conventional platforms, be they 2,3...6dof once you pitch past vertical yaw and roll jump/flip 180degrees (instantly). If you do a roll, once passing fully inverted the angle output jumps from +180 to -180. This causes the platform to violently switch sides. Do you have all 6dof implemented in DCS on your rig or just the three acceleration forces of surge, heave and sway?

Any reply from Andy? I've decided to revisit this now that I have finished building a new PC and will return to working on my G-Seat off and on in the coming months as time permits (catch 22 of needing to work long hours to pay for expensive hobbies and then having no time to enjoy the expensive hobbies due to working long hours lol). I'm planning to just use pitch, heave and sway until we sort this out. I can't use surge as I only fly helicopters and as you pitch nose down to accelerate having a surge force pushing against your back feels all wrong. It feels better when I reverse the output but that's not ideal and causes issues during other phases of flight. I also need to start looking at simtools v2 as well...

Just got mine installed. Delighted with the performance in DCS. Hawkeye60 driver 381.65 not showing any issues.

It could be worse. A lot worse! Just fired up X-Plane11 with FlyinsideXP after receiving my new graphics card and I still can't achieve a stable 45fps for ASW to work. I'm done with X-Plane until they release their version of native support. In fairness to ED they have done a great job on the implementation of VR given that DCS is primarily a monitor title. You're right though to get the best experience you do need a good system.

My Asus ROG strix 1080ti OC shipped yesterday. I doubt I'll see it before Monday though. Let's hope they make a big enough difference to justify the cost. I'm officially broke! Lol