Rosebud47

Great video with cool cuts made fit to music... enjoyed watching :thumbup: Split screen remembered me of Brian de Palma style :)

... eagerly awaiting MiG21´s update, especially for improved performance in VR. To adjust the scale for the MiGs cockpit in VR, I´ve set the IPD in DCS to 48, so the scale feels more right. lower DCS / VR / IPD setting --> scales up higher DCS / VR / IPD setting --> scales down.

Yeah, confusing it is... guess the function was meant to be kind of ingame IPD setting, but turned out, that it is a world scaling factor and never was corrected in its designated description. I´m always orientating on the dimension of the stick. If it looks too small, like a toy, I go down with the IPD value, if it looks unrealistically big, I go up with the value until it fits.

For most planes I used to have set IPD to 48, except for the F5 Tiger and the KA50 is set to 62 to look ´real´in dimensions.

BF 109 Kurfürst is for me the hardest to land... more difficult than the Spit.

Actually I don´t see so much of a problem with the canted display of the Pimax with DCS. What Twistking and me discussed earlier related more to the VRzoom function in DCS which is in fact broken due to the use of Pimax, which got canted display. I wonder, if or how VRzoom works with the Index or XTAL. Maybe Valve could have fixed this by software optimization for the Index, but another problem in this regard with Pimax might be, that Pimax does by default not work like HMDs with parallel projection... everyone knows, parallel projection could be activated for Pimax, but it is a massive impact on performance, which I wouldn´t trade-off to have VRzoom. So I disclaimed VRzoom meanwhile... it was a useful and comfortable option, but I can see all gauges etc. clearly with the 5k+. But MFDs or MFCD in planes are still a problem to read en detail. I lean a bit forward to read precisely, but when you get used to your plane with MFD, you could wokr with the MFDs by memory and don´t need to read the function of a button, before you press. It also helps a bit fixing your position in the cockpit - by leaning back and recenter VR position, you are a bit closer to the front panels in the cockpit. It would be nice to have back VRzoom for the Pimax, but beside of this, I don´t see the absolute need for more or special support for canted displays.

additional test with MiG21bis free flight over Caucasus , altitude 1.000m / same settings as below. STABLE DCS pixel density 1.2 35 FPS / 22ms GPU frametime in first few seconds 21ms CPU frametime in first few seconds DCS pixel density 1.0 43 FPS / 18ms GPU frametime in first few seconds 20ms GPU frametime in first few seconds Open Beta DCS pixel density 1.2 40 FPS / 20ms GPU frametime in first few seconds 14ms CPU frametime in first few seconds DCS pixel density 1.0 46 FPS / 17ms GPU frametime in first few seconds 14ms CPU frametime in first few seconds CPU frametime did increase around 50% from stable to open beta

I would run a balanced combination of DCS PD and Render Resolution. I´m still thinking, that these are two separated processes and that DCS PD at 1.0 render a frame in 1080 x 1200 resolution, from which you could scale up. The render target with Pitool/SteamVR only upsamples the prerenderd image of 1080 x 1200 ( PD 1.0 ). In effect the higher render resolution via Pitool/SteamVR sharpens the edges and reduces shimmering, but it does not add more information or detail to the image as DCS PD could do. DCS pixel density renders the original frame - the higher, the more detail and information is rendered, but on cost of performance. My subjective impression is, to get most quality if render resolution target and DCS PD is well balanced.

i7 5820k 6 cores @4.4 GHz 2080Ti 32 GB RAM Nvidia Driver 430.86 3D settings: everything default/application controled Prerendered Virtual Reality Images: 3 Pimax5k+ Pitool 1.0.1.132 90 Hz, Large FOV ( 170° ) Render Quality: 1.25 Smart Smoothing DEACTIVATED everything else DEACTIVATED SteamVR beta 1.5.11 asynchronous compositor and reprojection DEACTIVATED fixed render resolution: 3200 x 1800 VR performance measured with fpsVR 1.7.6 DCS stable/ open beta identical: ["graphics"] = { ["DOF"] = 0, ["LensEffects"] = 2, ["MSAA"] = 1, ["SSAA"] = 0, ["SSAO"] = 0, ["anisotropy"] = 3, ["aspect"] = 1.7777777777778, ["chimneySmokeDensity"] = 2, ["civTraffic"] = "", ["clouds"] = 1, ["clutterMaxDistance"] = 0, ["cockpitGI"] = 0, ["effects"] = 3, ["flatTerrainShadows"] = 1, ["forestDistanceFactor"] = 0.8, ["fullScreen"] = false, ["heatBlr"] = 1, ["height"] = 768, ["lights"] = 2, ["messagesFontScale"] = 1.25, ["motionBlur"] = 0, ["multiMonitorSetup"] = "1camera", ["outputGamma"] = 1.4, ["preloadRadius"] = 100800, ["rainDroplets"] = true, ["scaleGui"] = false, ["shadowTree"] = false, ["shadows"] = 0, ["sync"] = false, ["terrainTextures"] = "max", ["textures"] = 2, ["treesVisibility"] = 6000, ["useDeferredShading"] = 1, ["visibRange"] = "High", ["water"] = 2, ["width"] = 1280, Mirrors off STABLE DCS pixel density 1.2 45 FPS / 18ms GPU frametime in first seconds NOE flight Tiblisi: 42 FPS / 20ms GPU frametime DCS pixel density 1.0 53 FPS / 15ms GPU frametime in first seconds NOE flight Tiblisi: 49 FPS / 16,5 ms GPU frametime Open Beta DCS pixel density 1.2 47 FPS / 18ms GPU frametime in first seconds NOE flight Tiblisi: 42 FPS / 20ms GPU frametime DCS pixel density 1.0 55 FPS / 14,5 ms GPU frametime NOE flight Tiblisi: 47 FPS / 17,8 ms GPU frametime

Yeah, this automatic setting in SteamVR might work well for special VR games with usually low graphical fidelity, low textures, if there were textures at all and low geometrical details. But for DCS World it´s very much different. Actually the automatic setting thinks, that you got a super fast system with a 2080ti and increases the render target, therefore instead of lowering, what is needed to run DCS smoothly, it increases the render target resolution in SteamVR automatically. @wicked doesn´t matter, which HMD you use, none is really perfect and VR is mostly in developement in every aspect - most important is, that you are happy, with the experience.

Thanks for getting a bit more into it - it´s very interesting. But both we know, that these pictures don´t show, what you really see through the HMD. The distortion shown with the Pimax picture are not visible in this way. The image in the Pimax surely looks right in the first place and not like the picture might express. This could lead to some confusion to readers. Same for the Index hidden area mesh, shown on the picture is not perceivable in this way when looking through the HMDs and its lenses. Anways, but the pictures are very interesting to understand more , how the HMD works. With regard to Pimax and its inefficient use of panels, I think, that Pimax messed it a bit up. Not to undertsand wrongly, the image you get in the Pimax is pretty good and impressive, but totally agree, that it could be better. I think the´ve messed it up, because on the first hand, Pitool Service is meant to work with the 5k+ and the 8k, which have different LCD panels, but the Pimax compositor seems to work same for both, which is not optimum. Secondly, the Pimax HMDs provide three different modes: small, normal and large FOV, which doesn´t thinks make easier to be optimum for each. It would have been better, if Pimax API would ahve been optimized for only one kind of LCD panel and only one FOV, that´s why I totally agree with you and think that it´s messed up and far from optimum use of the hardware. But there are also many opportunities to adjust and specify the settings for Pimax. The only thing, which can´t be adjusted is the shape of the lenses, which is very special. I don´t know, if you would get the same result through the lenses, which feels btw right to me, if you change the distortion at the outer edges of the panels. There are quite interesting approaches to change this, but so far I could not have gone through everything... this will need a lot of time and enthusiasm to optimize the distortion and finally it´s the biggest challenge for the large FOV HMD. But I enjoy Pimax and that the followed this approach and they already reached a very good result, even it´s not perfect. Here are a some links and descriptions to the functions: https://forum.pimaxvr.com/t/pimax-native-projection-and-pre-lens-warp-transformation/15775 https://forum.pimaxvr.com/t/pimax-distortion-editing/11493 But it´s not an easy task ... How have your friend made the headset mirror pictures? It´s very interesting as they show the panels image and not though the lense image. How can I make these mirror images by myself? It´s also interesting, that with the distortion at the edges, there might be used for example two physical LCDs on the panel for one pixel of the rednered image, which result in the streched and less sharp image. Let me mention again, that you don´t perceive the discrepancy unless you have seen, how it should look correctly, but there is also the sweetspot of the lenses, which is alreayd huge in the 5k+, but become unsharp on the edges. Maybe if the distortion on the panels could be corrected, it might have no effect, as the perceivable area around the sweetspot will be in anyway not as sharp as the area in the sweetspot... can´t say if or if not, unless it tested with optimized settings. With reagrd to Valve´s Index, there´s no need to convince me, that Valve did a great job on this, but I think the Index´problem - only if we want to look for problems, not solutions this time-, is, that it comes a bit late on the market. At this point the Index display resolution and FOV is pretty standard, just because the competitors put out less engineered and optimized HMDs with the same display resolution and FOV. I hope Valve keep thing going and now could get out their next HMD on base of Index innovative techniques much faster to be head to head with the competitors with a next generation and finally significantly better optimzed for the needs in VR ( especially with regard to the FOV ).

Am not quite sure, if it could be expected from the Index what you described. I could follow, you´re refering to an eye-to-lense-distance effect, on how you perceive pixel density, but what is meant, as far as for my understanding, is the technical pixel density of the displays. Also I don´t believe, that the Index is sacrifying anything for a "wider FOV". It´s a bit strange, that Valve promotes the Index with a larger FOV for some(!) depending on the person, what means the Index does in fact not have a wider FOV, but depending on the person and how close to your eyes you adjust the screens, it could appear as a wider FOV. I would bet, that the image displayed through the lenses of the Index does not get wider or narrower while moving the lenses closer to the eyes. What I would expect from the Index are really optimum displays for VR, which could make an image through some optimum lenses as a perfect match for the used displays, look more good than just any displays with higher resolution through standard 1st Gen VR HMD lenses. I would expect that from the Index, that the parts used are perfectly engineered for their use in combination for the VR headset. I would say, the FOV of the Index will appear more large to the user, than with another HMD because of the HMDs design, but not because it actually uses larger FOV displays than Vive Pro. More important for me would be, if Index LCDs could switch much more faster than other used LCDs in HMDs. In aynway they are switching much more faster than OLEDs, but I think, there are also differences in quality from LCD to LCD display, which might effect more for use in VR, than used for Smartphones or tablets or anything. @Alec Delorian You really gave me a big smile today with your postings, but the picture showing the Pimax is really old. Actually it is a picture taken from one of the early prototypes, which had massive distortion on the edges, but it is not, what you get or have to expect from the final product as it is nowadays. Since this picture was taken, Pimax searched for a new lens provider for the final version and did found one and they finally could have solved the problem, which shows the picture you´ve linked. Also I have some doubts, if you could take the Index Video as an example on how the FOV or quality looks like, as it is a video grabbed from a flatscreen with a strange mask overlay, not though the lenses - that´s a common problem for getting some impressions with a HMD, that the only way to get a real impression is, to test it for real. But I´m really optimistic for what Valve have designed. Valve also stated, that there are some updates in function will be added to SteamVR before the launch of the Index, means with current SteamVR the Index is not supported to its full capabilities. It would be great, if there would be an option for setting the displays to high refresh rate, but to fix FPS at 90 for example. Let´s see, what´s coming up ... lastly I decided to skip the current new HMDs anyway, as I am quite happy with the 5k+ meanwhile. Don´t get me wrong: if I like to see a winning team over another, I´ll watch a soccer match on TV and not VR headsets, but of the current new ones, none could beat the 5k+ ...lol.

When SteamVR is not set to override in the video tab, it detects the CPU and GPU and sets with regard to your system specs automatically the ´appropiate´ render resolution. Problem is, that it does not differentiate between "Fruit Ninja" and "DCS World".

That´s good to know, that the render target in Pitool could be adjusted in smaller steps via profile.json. Glad to hear that it´s working now. I still think, that the value of -2.32 need to be more precised with more digits to exactly meet a multiple of the native resolution, when upscaling beyond *1.25 in Pitool. To finally make the image round in DCS, I´ve adjusted the aspect ratio in the stereo.lua ( located in c:\EagleDynamics\DCS World\Config\MonitorSetup\ ..) as follows: Left = { x = 0; y = 0; width = screen.width * 0.5; height = screen.height; viewDx = 0; viewDy = 0; aspect = screen.aspect / 1.777777; eye_shift = -0.032; --tans of side angles projection_bounds = { left = 0.964925826, right = 0.715263844, top = 0.889498115, bottom = 1.11092532, } }, Right = { x = screen.width * 0.5; y = 0; width = screen.width * 0.5; height = screen.height; viewDx = 0; viewDy = 0; aspect = screen.aspect / 1.777777; eye_shift = 0.032; --tans of side angles projection_bounds = { left = 0.715263844, right = 0.964925826, top = 0.889498115, bottom = 1.11092532, Now, the aspect ratio of the images should be the same as the native display resolution or respectively a multiple of the native resolution through upscaling/supersampling through out the whole rendering pipeline: Pitool --> SteamVR (openVR ) --> DCS World. In result the DCS cockpits look very right to me, neither compressed or streched in the slightest. @Wicked Jesus, check the override button in SteamVR ;) BTW: I think the overall clarity could be improved a bit by setting the Brightness in Pitool to "high" ( default is normal ). This should increase the backgroundlighting of the LCDs. If too bright, could be then adjusted by Gamma settings in DCS, while keeping some more overall clarity.

MiG21 all time favorite. Can´t say why, but always come back to her and always enjoy.

When the windows for the Mixed Reality Portal opens on the desktop, there´s a button in the left down corner ( as far as I remember ), which brings you to the WMR Settings. There you could adjust the ´Quality´ from low, med, high, experimental. In anyway experimental should be selected. You could also set the frequency there 60 or 90 HZ and some audio settings, as far as I remember. With the app Windows Mixed Reality for SteamVR you could further on set the render target by overriding option, what is recommendable. SteamVR is not too bad - in contrary, it´s very good, well maintained, updated frequently and the general GUI to interact with OpenVR, which then interacts with DCS.

@imacken have you tried in Nvidia control panel under --> "change resolution" tab under --> 3. --> set Nvidia colors ( by default it is set to standard )--> "dynamic ..." --> "full" ( by default this is set to ´limited´ ) This helps a bit in VR to improve the colors.

Hi Volk50, sorry for late reply. I´m using the latest available Pitool Version. Had a look into some files, to see what may affect the discrepancy with your setting, but not sure, what´s the cause. Is there set in the same file "steamVR.use.native.FOV" set to "1"? The -2.32 value did affect first the horizontal resolution until reached 2560, when reducing the value. Then it switches to the vertical resolution until 1440, when further reducing the value. A more precise value with 3 digit after comma should affect, when you´re at 1.5 scale or above in Pitool. Up to 1.25 the coefficient of the render resolution remains 1.7777777 , at 1.5 in Pitool it changed to different values after the 3 digit like 1.7746456 and there was a difference in the Render Resolution of 4 pixel in heights in SteamVR. If you find some time, try to change the -2.32 in very small steps and check, how it affects the render resolution in SteamVR to get closer and finally match the native aspect ratio.

It´s simple with the Pimax: With regular 110° FOV headsets you move and glance in the Cockpit, like you would drive your car with a diving goggle on your head. With the Pimax you move and glance in the cockpit, like you are used to, when driving your car.

Mostly improved is the CPU frametime and visuals. Great update today! Image received clearly more sharpness and less jaggy edges. Especiially in the distant view it´s much more sharp ( you could see the snowy mountains of the Caucasus much better ). But also in the cockpit the improvement is clearly visible. CPU frametime improvement is massive. I´m running the Pimax with 3200 x 1800 render target large FOV @90Hz MSAA 2x, PD 1.2: rocksolid 45 FPS with the Warbirds ( Mustang, Spit ) and CPU frametime of phenomenal 8 - 9 ms in low level flight over the city. But it seems that GPU frametime increased a bit, but still works good under 20ms. With the Tomcat I didn´t observe so much difference in GPU frametime ( didn´t observed the CPU frametime for long in the Tomcat ) like before the patch, but better sharpness all over for sure. Most love of my favorite planes received the MiG21 today. The flickering in the Cockpit is mostly gone and it´s all over sharpness is great. Also the little fixes, like the sound of the emergency air start or the better looking nose wheel break switch... the white light knob finally works, the backlighting in the MiG21 cockpit is much better etc., etc. When there is still coming a VR improvement ( hopefully which betters the GPU frametime ) and an overhaul Spring Update for the MiG21 still coming , I could only forward kudos and big respect to ED and Leatherneck developer for today´s patch! Great Work!

Doesn´t make much sense to me to talk on on this, unless it´s released for testing and fiddling. But me too have doubts ... lol

Could take a while as the availability is noted on their website to the end of September.

Meanwhile I´m pretty sick of testing, doubting and fiddling with Pimax. I´ll keep it for experimenting, but reserved an Index and will get one as soon as available. After three years of VR, I really appreciate the large FOV before regular HMDs, but am also very interested in 120Hz/ 144hz mode. Let´s see how this goes ... ;-)

Native support for WMR sounds really good. These days are really a good time for VR.

To get the render target resolution as noted. If you set steamVR not to override the render target, it will always adjust its settings automatically, when you change the render target in Pitool.