Chapa

@Voight: Thanks you for your feedback. Effectively, we should say which plane we use to compare our settings because there are specific optimisation for VR to reduce bad effect like shimmering. (For example, adding around all textures a layer of full transparent pixels to reduce Aliasing). In my case, I test with Huey, P51 and F18. I bought yesterday the MiG21 and also noticed that shimmering is more noticeable in this plane and I pretty sure that I will reconsider my settings if I focus on this plane. Concerning the AA settings in Nvidia Inspector, from my experience, they are not taken into account in DCS. One simple test is to select the more costly AA in Inspector, run DCS in no VR mode and see that the FPS is exactly the same than with just MSAAx2 or no MSAA. But if you have really the feeling that some AA settings works, please share with us your settings because it is a debate on other thread with nobody being able to prove it works. At the time of DX9, the idea to improve some games was in a nutshell to: 1) Add negative LOD bias to increase sharpness but also shimmering unfortunately 2) Add transparency Sparse Grid SSAA to remove shimmering 3) Enhence/Override the application AA settings by adding MSSA from the Inspector with the same multiplicator than the Sparse Grid AA to remove the blur induced by the whole process As you can see, surprisingly, the steps 1),2) is what we did. Keeping the LOD Bias Driver to On allow the driver to apply negative LOD bias. But because applying AA settings from the inspector seems to have no impact we can’t test the full approach . P.S the AA settings that has no impact I am speaking about would be: AA mode. I tried to pair it with enhence and override application and try to apply the AA compatibility DX1x flags (all the possible combinations) with no success. P.S.2 However FXAA works and have impact on the monitor. But in the HMD... I am not convinced FXAA works

@Wicked: I would like to thank you warmly, Wicked, to have share your settings and offer the opportunity to investigate about it. I was already very happy with my Rift S but now it just a lot better. And there is no placebo effect :)

After some investigations about the meaning and history about the settings in NVidia Inspector (and some performances issues), the following settings seems to do the job well: - Antialiasing - Transparency Supersampling: AA_MODE_REPLAY_MODE_ALL - Texture filtring - Driver Controlled LOD Bias: Off (very important to set to Off to force the LOD bias setting below) - Texture filtring - LOD Bias (DX): -0.625 - Texture filtring - LOD Bias (OGL): +0.000 - Texture filtring - Negative LOD Bias: Allow MSAAx2 is counterproductive for clarity. It blurs a lot the picture... (but remove aliasing for sure) So, I put MSAA to Off and added some PD at 1.1 (sweet spot for my 1080ti to stay at 80Hz and have some margin) The gain in clarity is huge and the counterpart acceptable for me

If you followed my advice by setting the 8x Sparse Grid Supersampling, it effectively tax a lot on performances. I did know when I wrote my post but I did some measures this afternoon and I discovered that this setting cost (all other things equal) about 30 FPS. These tests was done in no VR mode. Using 2x Sparse Grid SS doesn't change a lot the cost but reduce the effect, so I think the All-in is the best choice. Now, here is the dilemna: As suggested in the picture page 3, you can go with the setting AA_MODE_REPLAY_MODE_ALL and a negative LOD Bias (DX). My test didn't show any cost in FPS for that. But the price to pay is in shimmering. Applying negative LOD Bias increase the shimmering. Personnaly, I find shimmering as annoying as ghosting effect and I decided to stay with the Sparse Grid SS setting. I don't see now any reason to increase the PD for clarity and prefere to reduce at 1.0 or even 0.9 to increase other settings and stay at 80Hz. If you go with the AA_MODE_REPLAY_MODE_ALL setting, I found that a LOD Bias of -0.5 was good enough and do not introduce exagerated amount of shimmering. All of that tends to prove that it exists or will exist solutions to have good clarity without the need of the brute force solution called Pixel Density.

Hi, the ED Teams, First of all, thanks all for your amazing work. I recently discovered DCS but I am really impressed and have already spent a tons of money for very good stuff and started to play VR. Waow !!!! My wish would be to have more granular graphic settings in general but to keep my wish focussed, it would be more granularity about shadows : At least, we should have: 1) Cockpit shadows: Off|Flat|Low|Medium|High 2) Vehicules/Planes shadows: Off|Flat|Low|Medium|High 3) Terrain shadows: Off|Flat|Low|Medium|High Today, the 1) and 2) categories are merged and it is the issue, specially in VR. Cockpit shadows (when airborn) don't have a huge impact on FPS but as soon as there are other planes (on the ground), the FPS drops drasticly. To play in VR at native frequency, without AWS, shadows are the big ennemi. It could be solved by the proposed granularity. A cockpit without shadows are not as immersive as it could be. Vehicules and terrain shadows are more nice to have visual suggar and I am pretty sure that people prefer to set more essential settings like MSAA, Distance View, Pixel Density that are not so easy to handle well. I hope it will be possible to have this feature implemented. Thanks again for all the work and pleasure provided !!! Best regards, Chapa P.S I want to precise that I have found acceptable compromise to play with a Rift S at 80Hz, I have spent hours and hours to tune and understand the settings in DCS but also in NVidia Inspector. For some reason, vehicules/planes shadows kills the FPS and I tend to think it could be avoided , with the result to have cockpit shadows with AWS Off, even on the ground.

Yes, these settings have definitively positive impact with Rift S. However, I have better results with: - Antialiasing - Transparency Supersampling: 8x Sparse Grid Supersampling - Texture filtring - Driver Controlled LOD Bias: On - Texture filtring - LOD Bias (DX): +0.000 / Automatic - Texture filtring - LOD Bias (OGL): +0.000 - Texture filtring - Negative LOD Bias: Allow I have also set the following anisotropic filtering settings and disable the AF in DCS (nvidia settings are taken into account): - Anisotropic filtering mode: User-defined/Off - Anisotropic filtering setting: 16x The 16x AF from Nvidia inspector is better than the 16x provided by DCS. And finally MSAAx2 in DCS settings. With these settings it start to make no sense to use VR supersampling at all. PD 1.2 stay better than PD 1 but.... with the transparency supersampling settings, it is at least as clear as it was with PD 1.4 or more without... At this point, using PD to reduce aliasing becomes very costly in performances. Even PD 0.9 becomes acceptable for my point oof view (note that I want to play at 80Hz so I accept some compromises). P.S. The new nvidia drivers from today or yesterday add support for sharpening to OpenGL and Vulkan. We will probably have good news for VR in futur driver release.

The 1080ti has alteady hard time to handle the latest HMDs. I don’t think that you can expect even a 2080ti to be able to handle well the future HMDs but if there is a card that could, it is this one at the moment. I would not buy a new GPU without knowing the HMD that will go with it. In your place, I would postpone the GPU buy when you decide to change your CV1 or buy the cheapest GPU that match the best your CV1 (to play at 90Hz)

I have a setup quiet similar to yours and I also started by the setup proposed in the link you provide. Honestly, after spending few days to try the settings, understand their impact and experiment how bad AWS can be, I don't recommend to follow this link... The goal is clearly to play at 80Hz and to accept that during taxi, take-off and landing to have ASW on but depending the scene, not always... First, your SS is way too high. Just to have an idea, a SS of 1.4 feed the GPU with more than 11'000'000 pixels, it is 1.4x the number of pixel of a 4K screen. I found the sweet spot to be 1.2 with my 1080ti. And I easily believe it is 1.3 for a 2080ti. For info I don't use Oculus Tray Tool because the same issues you describe and the PD setting in DCS is exactly the same because it is a native implementation of the Occulus SDK behind. In DCS with the ctrl + num pad keys you can change the ASW mode so you have everything included, except the HUDs for monitoring your FPS but you can use the oculus debug tool in the oculus support installation folder. Important note, changing the MSAA setting require to quit and relaunch DCS to take effect, even if the game don't ask for. I lost a lot of time, thinking that this setting didn't have real impact and was a placebo. But It has a nice impact but also on performances. For me the mandatory settings are: PD 1.2 Texture: High (nice cockpit first) MSAAx2 (to reduce aliasing and shimmering) Filter Anisotropic x 16 (I find it to much important to accept comprise and it don't tax to much the system) Terrain shadows: OFF (not worth the price in perf) Global illumination: OFF (not worth the price in perf) Then I start the optimization with all the setting at the lowest possible value. And increase at much as possible the ones that are the most important for me. Overall, it is a trade-off between Distance view and Shadows to keep the 80Hz. Last but not least, I found that the following setting in the Nvidia control panel gives clearer image: Filtering quality : Best performance (yes best performance, try it!) Sorry for the long post and the approximative English. Hope it will help you to find your settings but I think you are not far.

Thanks you for the info and the link. You are right the Anton has also the same behavior with AWS on. I also use the Rift S with AWS disabled because the ghosting effect in genera. I prefer to reduce some graphical settings and have stable 80Hz.

The base comes from factory with the soft springs and cams (with a bump) mounted. I changed both. The other cams in package don’t have the bumb and it gives a more continuous movement feeling when you cross the center (no bumb). However, the cam change will not solve the issue, I just find the feeling more natural with TM sticks and I think it is more adapted to their weight, especially after setting 5 deadzone on both axis. Imagine that your plane will start to move only when you will feel resistance from the stick. Here is a video explaining how to change the cams: and here some other Virpil tutorials that could interest you: P.S don’t be sorry to ask question ;)

I use the TM F18 stick with the WarBRD base and the hard springs and no center cams. I put 5 deadzone for pitch and roll to solve the issue you describe. Alternative is to increase the deadzone in the VPC software that is set at 2 or 3 by default, probably adapted for the VPC sticks height/weight.

For sure, just no way to fly the Spit in VR with AWS on because the artifacts caused by the propeller. There are also artifact with other warbirds but not with the same magnitude.

The Nvidia release note says : scaling will not work with VR. (https://us.download.nvidia.com/Windows/441.12/441.12-win10-win8-win7-release-notes.pdf)

I am really sad to read this forum and all these angry posts in the different sections (not only about Razbam). I am also happy to see that even when people are very angry about an helicopter but the next one is announced (by the same 3rd party), the are forced to stay suspicious but are very excited by the news anyway. Why ? because you are passionate and it’s great but... your are not fair, you are capricious and honestly I will stay polite - You are not working for the 3rd party companies so don’t ask to be part of their professional processes. Don’t ask to these niche market companies more than you ask to your banker, supermarket, boss, insurer, TEACHER maybe etc you get the idea... - I also hate Facebook but not following some channels is your choice at the end. I also think that 3rd Parties should include news in the forum but I will not complain if they don’t. Adapt yourself, it helps a lot - Complain when you think it can help the community but don’t insist when you are not heard. It appears you have some time to spend and do again and again and again the same posts, complain without success. Maybe you are right and it is justified to be angry but what’s the point ? Revenge ? In fact, if you are right and you want revenge the best way is to use your time to help other 3rd parties and by the same occasion the community again. I could continue the list but it will convinced nobody how not fair you are.... But concerning Razbam and what I learnt by reading this forum, youtube and unfortunately facebook (no friends only for DCS news and not only Razbam), there are some interesting facts: - Razbam do serious stuff with AdA and we profit some derived products and features. Tomorrow, it could be different, who knows ? - Razbam with the help of Baltic Dragon provide serious educational resources to start playing and enjoying DCS - Razbam provides enough information from the different channels to have an idea about what’s is coming - Razbam do some mistakes for sure like monthly video that is clearly very difficult to provide long-term and bug trakers. Don’t ask DCS 3rd Parties to be able to do the same as some very succesfull startups in the Sillicon Valley that marketed their work processes as the recipe of their success. Please stop bashing Razbam, you have the right to be angry but in this case accept to have loose your money and use your energy for helping the community by other way. I worked as a developer in small companies and It is clear that your expectations/requirements are totally crazy and unrealistic. I don’t work for Razbam but for me Razbam provides very special value to DCS that even Heatblur don’t provide nor ED. Razbam needs more time but this time is required to provide this special value. Up to you but I am convinced that you have more to provide to the community that always the same kind of posts...

Well said, I perfectly agree with that and even add to not forget from where come your frustration where you are not happy: from passion and dream and wish to have the game (sim sorry) of your dream !!!! Don’t bite the hand that feed you and remembre also your voice is never the voice of all other passionate people. Keep going on Razbam ! And for sure, it just make more sense to make video about some New features than a monthly video. I am very excited to see the new mirage 2000 cockpit.

With +30 curve for sure. Try with negative curve instead.

If I am right with this intermediate curve that map the virtual deflection of the stick to a pression level, people with longer stick and/or extension will have stronger sluggish feeling than people with shorter sticks. To solve the issue but keeping this special pression feeling of the F16, a Special setting would be requiered: Max real roll deflection, max real pitch deflection (in cm or inch). With these information, the curve could be adapted to give to every people the same pression level for the same real deflection, so the same feeling. Of course if my supposition is right and I let you judge about the probability :)

Same for me with a Virpil WarBRD base and TM F18 stick. I have to put negative curve or be very agressive on the stick to roll fast. Maybe its right because the special pression stick of the F16. On some posts, people complain about large deadzone but I checked and it is not the case because there are some very light output from a small stick deflection. But I have the feeling that an intermediate curve map the virtual deflection of the stick to a pression level. This pression level is then translated to aileron movement with the fly by wire in the middle and some delay time constant. It could make sense and explain why we have to be really agressive on the stick and the very different feeling from other jets. Personnaly I prefer without negative curve and live with the différence until it is confirmed as normal or a bug