Jump to content

OpenXR Guide - Deprecated - This time for real (▀̿Ĺ̯▀̿ ̿)


nikoel

Recommended Posts

5 hours ago, DeltaMike said:

@Sr. -- what settings are you running? 

130%/100%  FSR

WMR 60Hz

DCS settings
 

DCS Settings.JPG


Edited by Sr.
Bumped up the slider to 130%

Ryzen 7 5800X3D | 64GB DDR4 3600| MSI RTX 4080 16GB Ventus 3X OC  | Samsung 970 Evo 2TB NVME | HP Reverb G2 | DIY Head Tracker Cap | Logitech X-56 throttle | VKB NXT Premium |  Win 11

"Any sufficiently advanced technology is indistinguishable from magic."

--Arthur C Clark

Link to comment
Share on other sites

I tried this today and unfortunately got quite poor results compared to steam vr.

I'm at least 10 fps down and frame time in the 30s.

I'm using a reverb G1 with an AMD 6800xt and 11700k cpu.

Strange how others are getting great performance. 

 

Link to comment
Share on other sites

Im also still trying to find the performance upgrade using OpenXR. And i cant seem to get it.

Even at 50% resolution, no motion smoothing, every ingame option at the lowest, and im still getting stutters and fps drops. Even using the 60hz doesnt make it much better.

I cant understand how there can be such a big difference of how everyone is experiencing DCS on their systems, even when systems are 99% identical.

But im also finding some weird things like looking at a direction from inside the cockpit with even not any cockpit instruments in sight and getting bad fps with stutters. While pressing F2 and looking at the same direction and having fluent motion. Like, why, what is making the fps dipping?

I really feel like ED really need to give some information soon about VR improvements and Vulkan to be able to play this game without doing all kinds of mods or whatever to make it playable without a stutterfest.

 

 

 

MSI MPG B550 Gaming Plus - Ryzen 5 5600x - RTX3070 - 32GB Ram - HP Reverb G2 - Warthog HOTAS - VKB rudder pedals

Link to comment
Share on other sites

35 minutes ago, RvL said:

Im also still trying to find the performance upgrade using OpenXR. And i cant seem to get it.

Even at 50% resolution, no motion smoothing, every ingame option at the lowest, and im still getting stutters and fps drops. Even using the 60hz doesnt make it much better.

I cant understand how there can be such a big difference of how everyone is experiencing DCS on their systems, even when systems are 99% identical.

But im also finding some weird things like looking at a direction from inside the cockpit with even not any cockpit instruments in sight and getting bad fps with stutters. While pressing F2 and looking at the same direction and having fluent motion. Like, why, what is making the fps dipping?

I really feel like ED really need to give some information soon about VR improvements and Vulkan to be able to play this game without doing all kinds of mods or whatever to make it playable without a stutterfest.

 

 

 

Are you certain the openvr_app.dll (and other files) made it in to the bin folder? Or maybe Windows prevented replacing?

Ryzen 7 5800X3D | 64GB DDR4 3600| MSI RTX 4080 16GB Ventus 3X OC  | Samsung 970 Evo 2TB NVME | HP Reverb G2 | DIY Head Tracker Cap | Logitech X-56 throttle | VKB NXT Premium |  Win 11

"Any sufficiently advanced technology is indistinguishable from magic."

--Arthur C Clark

Link to comment
Share on other sites

27 minutes ago, Sr. said:

Are you certain the openvr_app.dll (and other files) made it in to the bin folder? Or maybe Windows prevented replacing?

Been replacing them with every update of DCS and OpenXR.

MSI MPG B550 Gaming Plus - Ryzen 5 5600x - RTX3070 - 32GB Ram - HP Reverb G2 - Warthog HOTAS - VKB rudder pedals

Link to comment
Share on other sites

Note not everyone is seeing an increase in FPS.  Some improvements may be beyond our ability to measure (fpsVR can only measure up to the point where SVR passes the frame along). We are seeing consistently better loading times, improved image quality (eg less shimmering), less judder.  

First step is clean install as per instructions. Second step I think is to make sure we are comparing apples to apples, so we want to be comparing similar DCS frame sizes. So whatever mix of pixel density, global resolution, per-app resolution and scale factor we had in SVR needs to be replicated in OXR.  FSR and NIS scale differently (and not as aggressively) in OXR as I learned the hard way

Ryzen 5600X (stock), GBX570, 32Gb RAM, AMD 6900XT (reference), G2, WInwing Orion HOTAS, T-flight rudder

Link to comment
Share on other sites

13 hours ago, edmuss said:

Your postGPU shouldn't be that high if you have reprojection turned on. There is a bug in OXRDT performance overlay that reads the postGPU the same as appGPU when reprojection is turned off.

Does the red box say something like tex mapping or reproj?

it says motion reprojection?

ASRock Z590 Phantom Gaming 4/AC / Intel i7 10700K @ 5.1Ghz / Noctua DHS-14 Heatsinkw/Fan /  Samsung 970plus m.2 1TB  /  eVGA FTW3 2080Ti /  RipJaws - 64GB RAM @3200  /  SoundBlaster Z  / Reverb G2 VR /  ThrustMaster HOTAS Cougar & MFD's / Buttkicker Gamer 2

Link to comment
Share on other sites

So what's the consensus on Motion Reprojection vs Radeon Chill/standard frame locking?
I've tried both, Chill seems more stable at 45 and no jiggle/jello effect, but seems it has a bit of ghosting on fast movers in the merge
where as MR is just jello jello jello?

I don't know if there was some update in the past month or so, but I was able to hold a pretty stable 90 frames most of the time, but now my CPU is capping out (12700kf @ 5.3ghz) Not sure what's going on, and its holding me back to 60'ish.

System: Intel 14700k on MSi z690 DDR4 pro, Arctic Liquid Frozer II 360mm AIO, 64gb ddr4-3600mhz CL16, Gigabyte 4090
Winwing Orion2 F-15EX/F-16EX combo, VKB T-Rudder Mk.IV, TM Warthog (Shelved)
2x 256gb Vertex 4 SSD's, 2x 1tb Sabrent NVMe PCIe 4.0 SSD's, 1x 4.0tb Silicon Power NVMe, 1x 4tb Seagate HDD
VR: Quest 3 (Current), HP Reverb G1 (Shelved)

Link to comment
Share on other sites

54 minutes ago, zero2005 said:

So what's the consensus on Motion Reprojection vs Radeon Chill/standard frame locking?
I've tried both, Chill seems more stable at 45 and no jiggle/jello effect, but seems it has a bit of ghosting on fast movers in the merge
where as MR is just jello jello jello?

I don't know if there was some update in the past month or so, but I was able to hold a pretty stable 90 frames most of the time, but now my CPU is capping out (12700kf @ 5.3ghz) Not sure what's going on, and its holding me back to 60'ish.

I find MR much better, but I have an nvidia card so can't test chill, I don't like the visual artefacts in the terrain of the rtss approach.

I'd check that HAGS is not enabled if you are strangely on locked at 60....

SYSTEM SPECS: Hardware Intel Corei7-12700KF @ 5.1/5.3p & 3.8e GHz, 64Gb RAM, 4090 FE, Dell S2716DG, Virpil T50CM3 Throttle, WinWIng Orion 2 & F-16EX + MFG Crosswinds V2, Varjo Aero
SOFTWARE: Microsoft Windows 11, VoiceAttack & VAICOM PRO

1569924735_WildcardsBadgerFAASig.jpg.dbb8c2a337e37c2bfb12855f86d70fd5.jpg

Link to comment
Share on other sites

For me definitely Radeon Chill on my 6900xt. It is completely smooth, no jitters at all even on 3/9. I’ve done repeated back to back tests and for me no doubt. 45fps seems like a sweet spot. The nice thing about Radeon software too is u can hotkey chill so it’s easy to test it. 
Right now I’m running 120% resolution and it’s extremely good. Solid 45fps even over cities. 

Link to comment
Share on other sites

@speed-of-heat, @TED not questioning your results and impression of limiting the FPS externally, but what I don´t understand is, how 45FPS fixed by reprojection could be better than 45FPS of GPU output fixed by Chill or an Nvidia overcklocking tool? With reprojection deactivated you won´t get the re-projected additional 45FPS of fake images to get the impression of 90Hz in VR and even more while the GPU is capable of more than 45FPS without reprojection, you don´t get more, because it´s limited externally. 
Now for some time I really asked myself, what I´ve missed by limiting the FPS by the GPU driver.

Isn´t the function of limiting the GPU driver not meant to avoid unnecessary overheating of the GPU, like for older games like ´Counter Strike´where you could get easily 150FPS or more on 2D monitor screen, what is much more than counter strike needs to run smoothly. Or to free up some GPU cycling by limiting to let´s say 100FPS in ´Skyrim´ to have some headroom for other stuff going on on the GPU like TAA or Raytracing or such additional GPU processing? As we´re dealing in DCS VR anyways with only 45 or less real rendered FPS with reprojecting, shouldn´t this headroom for the GPU be there anyways? Measuring the advantage of limiting is also hard, as it is limited all the time.

I mean we´re limited by the CPU in DCS, means the GPU can´t get to its potential unless the CPU couldn´t deliver data fast enough to make use of the GPU power.
So far still the most effective way of squeezing out more performance in DCS is simply by overclocking the CPU.
I´m sure we will have some good discussion here for VR, once Multithreading is implemented in DCS and we could reconfigure the settings to make use of that for better perfromance and visual quality VR, but so far I just don´t undertsand the advantage for VR by limiting the GPU output to 45FPS externally.

 

Also would like to add an opinion on Supersampling in DCS. It´s just an opinion, nothing more...

Well, I wouldn´t make use of the supersample option in DCS ingame configuration for VR, just because we´re supersampling anyways by the appropriate VR Tools, like the slider in SteamVR for OpenVR headsets ( Index, Vive, Pimax, etc. ), or the slider in OXR developer tool for WMR/OpenXR headsets ( HP Reverb, Samsung Odyssey, etc.) or with Oculus Tool for Oculus headsets ( Rift, Quest, etc. ).

I believe with the additional image processing by activating supersampling in DCS configuration menu, we don´t do any good for the whole process by adding an additional image processing to the whole process, which is not faster, than the process for the same effect in VR by the appropriate software ( SteamVR, OXR developer tool, Oculus tool ).

... just some theoretical thoughts, maybe in practice we won´t see so much of a difference by doing this or that. My personal doctrine for VR is, keeping the process as simple as possible otherwise it would drive me crazy 😄.  

 

F-14b Tomcat   /   AV-8B Harrier   /   F-16C Viper  /   KA-50 Black Shark   /   Mi-24 Hind   /   MiG-21bis   

Link to comment
Share on other sites

@Rosebud47 Frame Limiting has many functions and uses, of which a more recent one is to limit temps and provide a "safety" in the past it was used to avoid screen tearing and any one of a number of conditions, like your screen stuttering in TrackIR which only really works properly in 60hz multiples;  though your understanding is correct, but nvidia's Frame Rate limiter does not work in VR , you have to use an external tool, like RTSS... and to be clear I get artefacts in BOTH but they are very different artefacts... 

In MR i get wing/missile blurring but the ground is perfectly clear, with RTSS I get blurring of the ground features almost like double vision, but the wings are perfectly solid... the cockpit is perfectly clear in both.  As to why that is I honestly can't say.

Otherwise in both they are smooth, with the caveat of not in the Marianas or Chanel, where I need 30.

MSAA and SS achieve similar things in very different ways and I suggest you read up on them @DeltaMikedid a good bit on them here:

 

and this isn't a bad thread either Multisampling Vs Supersampling (google.com)

SYSTEM SPECS: Hardware Intel Corei7-12700KF @ 5.1/5.3p & 3.8e GHz, 64Gb RAM, 4090 FE, Dell S2716DG, Virpil T50CM3 Throttle, WinWIng Orion 2 & F-16EX + MFG Crosswinds V2, Varjo Aero
SOFTWARE: Microsoft Windows 11, VoiceAttack & VAICOM PRO

1569924735_WildcardsBadgerFAASig.jpg.dbb8c2a337e37c2bfb12855f86d70fd5.jpg

Link to comment
Share on other sites

Those blurring effects are strange, especially as they change by using RTSS. With OpenXR mod I´ve observed some blurry/ fuzzyness most pronounced on the end of text lines, like the briefing screens or main menu when moving the head. This fuzzyness/blurryness always goes vertically, not horizontally. Never had this effect with SteamVR for WMR before.

I assume it has to do with that OpenXR is still a mod and not perfectly optimized as native integration into the DCS World render pipeline. With MSFS2020, which I don´t update anymore, but have a look into from time to time, I didn´t observed this effect so far.

Yeah, there are really good guides here around, I´ve also gone through your guide for the 3090, which is really good and helpful for any GPU. There are also good guides on Youtube for DCS VR especially - I guess that´s no news for you :-).

For supersampling I just wanted to make a point for using the regular slider in the particular VR software, which does the same effect as supersampling in DCS option menu but much better integrated into the VR process... probably more efficient.

Yesterday I took a ride with the Apache over the Marianas, as DBurne was very impressed by this experience, but as I locked 45FPS throught the OXR Toolkit it was noticable less fluid ( constant small stutters! ) than with other modules on other maps. Apache over Marianas is kind of stress test for my setup, but in practice I don´t fly on the Marianas, so it´s fine for me.  

Edit: @dburne Apache over Mariana indeed is the most beautiful experience we get in DCS at the time :-).


Edited by Rosebud47
  • Like 1

F-14b Tomcat   /   AV-8B Harrier   /   F-16C Viper  /   KA-50 Black Shark   /   Mi-24 Hind   /   MiG-21bis   

Link to comment
Share on other sites

6 hours ago, zero2005 said:

So what's the consensus on Motion Reprojection vs Radeon Chill/standard frame locking?
I've tried both, Chill seems more stable at 45 and no jiggle/jello effect, but seems it has a bit of ghosting on fast movers in the merge
where as MR is just jello jello jello?
 

VR is different from flat screen.  Locking frame rates from any where other than in the VR software can cause problems.  One possible outcome might be, a significant delay between when something happens in game, and when you actually see it with your eyeballs.   

Not sure it makes a ton of difference in SVR/WMR -- that stack is such a mess, I suspect that's already happening anyway.  Personally I think Chill works OK in SVR; not great, but OK.  

On the other hand, OXR is pretty slick. Specifically engineered to be more efficient than SVR/WMR.    Now we are wondering if Chill is really such a hot idea.  Let's just say it's not "best practice" but if it works for you?  Cool!   Enjoy it.  It's not gonna light your computer on fire or anything.  

Of course the "best practice" is to turn MR on, but then ya gotta deal with the jelly worms and such.  Pick your poison.  

-----

MSAA is sort of a selective supersampling technique, only works along edges.  This made it popular because it's more efficient than generic supersampling.  In VR, we liked it also because it cleans up edges.   The early headsets needed more than a little bit of that.  

I say this in past tense because when DCS went over to deferred rendering, MSAA lost its efficiency advantage, because now it doesn't know where the edges are.    Then it became a  question of whether it's worth using at all.   And the answer for many people is "yes."  MSAA is probably the best tool we have for shimmering.   Only thing is, there's a lot less shimmering in OXR and I think we have the option to look at alternatives.  Among them, plain old vanilla supersampling.

-----

Where these two threads merge is on the subject of tuning. 

Let's say you want to run your G2 at 90hz.  If you want things to look smooth, you're gonna need to hit a target.  Ideally you want your frame time to be 11ms.  That way you'll be at 90FPS most of the time, which matches your monitor refresh rate. Won't be perfectly smooth -- if you're zipping along at 500kts/500ft things are gonna look a little blurry along the 3-9 line, and you'll be jonesing for one of those 144hz HMD's and griping cuz DCS can't keep up with that.  

If you can't hit that target (don't feel bad -- nobody can) then your next target is 20ms, which should give you 45fps.  Turns out, that'll be smooth too, albeit with even more blurriness along the 3-9 line.  But you have to be right on it.  If you're drifting up toward 60fps you're gonna get a great deal of microstutter which people will describe as ghosting, double images or stuttering.  As you get closer to 80fps, you're gonna be dropping a lot of frames and it's going to annoy you to no end.  

I think a  lot of people with high performance GPU's get into this situation, where they can't quite hold 90fps all the time, but can't quite figure out how to get rid of the microstutter.  Several ways of dealing with this, but clearly, turning settings down doesn't help.  You gotta turn em up to get closer to that 20ms frame time.   So if you look at the process @speed-of-heat followed with his 3090, he exhaustively studied the effect of each setting on his render times, and found the perfect combination to get him right where he needed to be.  Every bit as much work as @edmuss did with his 3070, just moving in a different direction.  Trying to slow things down a little rather than speed em up (which is a good problem to have, believe me).

For those who need to turn their settings up to reach a target -- who need to slow their computer down a little to dance better with VR -- I'd suggest looking at your resolution slider.  It's an easy way to get things just right, and it has its advantages.  Among them, it has an anti-aliasing effect and it does reduce shimmering.  Among them, it may make your sweet spot bigger, as one poster recently reported.  


Edited by DeltaMike
  • Like 2
  • Thanks 1

Ryzen 5600X (stock), GBX570, 32Gb RAM, AMD 6900XT (reference), G2, WInwing Orion HOTAS, T-flight rudder

Link to comment
Share on other sites

9 minutes ago, DeltaMike said:

VR is different from flat screen.  Locking frame rates from any where other than in the VR software can cause problems.  One possible outcome might be, running 45fps, 20ms after the action has already happened. So in other words, what you see on the screen happened some 20ms ago; under worst case, using frame locking, what you're seeing happened 40ms ago.

Not sure it makes a ton of difference in SVR/WMR -- that stack is such a mess, I suspect that's already happening anyway.  Personally I think Chill works OK in SVR; not great, but OK.  

On the other hand, OXR is pretty slick. Specifically engineered to be more efficient than SVR/WMR.    Now we are wondering if Chill is really such a hot idea.  Let's just say it's not "best practice" but if it works for you?  Cool!   Enjoy it.  It's not gonna light your computer on fire or anything.  

Of course the "best practice" is to turn MR on, but then ya gotta deal with the jelly worms and such.  Pick your poison.  

-----

MSAA is sort of a selective supersampling technique, only works along edges.  This made it popular because it's more efficient than generic supersampling.  In VR, we liked it also because it cleans up edges.   The early headsets needed more than a little bit of that.  

I say this in past tense because when DCS went over to deferred rendering, MSAA lost its efficiency advantage, because now it doesn't know where the edges are.    Then it became a  question of whether it's worth using at all.   And the answer for many people is "yes."  MSAA is probably the best tool we have for shimmering.   Only thing is, there's a lot less shimmering in OXR and I think we have the option to look at alternatives.  Among them, plain old vanilla supersampling.

-----

Where these two threads merge is on the subject of tuning. 

Let's say you want to run your G2 at 90hz.  If you want things to look smooth, you're gonna need to hit a target.  Ideally you want your frame time to be 11ms.  That way you'll be at 90FPS most of the time, which matches your monitor refresh rate. Won't be perfectly smooth -- if you're zipping along at 500kts/500ft things are gonna look a little blurry along the 3-9 line, and you'll be jonesing for one of those 144hz HMD's and griping cuz DCS can't keep up with that.  

If you can't hit that target (don't feel bad -- nobody can) then your next target is 20ms, which should give you 45fps.  Turns out, that'll be smooth too, albeit with even more blurriness along the 3-9 line.  But you have to be right on it.  If you're drifting up toward 60fps you're gonna get a great deal of microstutter which people will describe as ghosting, double images or stuttering.  As you get closer to 80fps, you're gonna be dropping a lot of frames and it's going to annoy you to no end.  

I think a  lot of people with high performance GPU's get into this situation, where they can't quite hold 90fps all the time, but can't quite figure out how to get rid of the microstutter.  Several ways of dealing with this, but clearly, turning settings down doesn't help.  You gotta turn em up to get closer to that 20ms frame time.   So if you look at the process @speed-of-heat followed with his 3090, he exhaustively studied the effect of each setting on his render times, and found the perfect combination to get him right where he needed to be.  Every bit as much work as @edmuss did with his 3070, just moving in a different direction.  Trying to slow things down a little rather than speed em up (which is a good problem to have, believe me).

For those who need to turn their settings up to reach a target -- who need to slow their computer down a little to dance better with VR -- I'd suggest looking at your resolution slider.  It's an easy way to get things just right, and it has its advantages.  Among them, it has an anti-aliasing effect and it does reduce shimmering.  Among them, it may make your sweet spot bigger, as one poster recently reported.  

This is a very interesting post....

  • Like 1

Specs: Win10, i5-13600KF, 32GB DDR4 RAM 3200XMP, 1 TB M2 NVMe SSD, KFA2 RTX3090, VR G2 Headset, Warthog Throttle+Saitek Pedals+MSFFB2  Joystick. 

Link to comment
Share on other sites

Hi Guys,

recently i moved to openxr from steam vr.  Running Reverb G2 - rtx 3090 - R 5800x - 64 gb ram. For me image quality is way better than in steam vr with shaders. when i lock reprojection to always on and set it to 30 fps its ok and smooth with jets but when i go with heli its not. but only on rotor blades - everything is shaking and wobling even hmd display. under rotor blade is ok - did you notice this and how to improve this - its only when i have repro always on and only in area of rotors.

Best

Lukas

Link to comment
Share on other sites

51 minutes ago, DeltaMike said:

VR is different from flat screen.  Locking frame rates from any where other than in the VR software can cause problems.  One possible outcome might be, a significant delay between when something happens in game, and when you actually see it with your eyeballs.   

Not sure it makes a ton of difference in SVR/WMR -- that stack is such a mess, I suspect that's already happening anyway.  Personally I think Chill works OK in SVR; not great, but OK.  

On the other hand, OXR is pretty slick. Specifically engineered to be more efficient than SVR/WMR.    Now we are wondering if Chill is really such a hot idea.  Let's just say it's not "best practice" but if it works for you?  Cool!   Enjoy it.  It's not gonna light your computer on fire or anything.  

Of course the "best practice" is to turn MR on, but then ya gotta deal with the jelly worms and such.  Pick your poison.  

-----

MSAA is sort of a selective supersampling technique, only works along edges.  This made it popular because it's more efficient than generic supersampling.  In VR, we liked it also because it cleans up edges.   The early headsets needed more than a little bit of that.  

I say this in past tense because when DCS went over to deferred rendering, MSAA lost its efficiency advantage, because now it doesn't know where the edges are.    Then it became a  question of whether it's worth using at all.   And the answer for many people is "yes."  MSAA is probably the best tool we have for shimmering.   Only thing is, there's a lot less shimmering in OXR and I think we have the option to look at alternatives.  Among them, plain old vanilla supersampling.

-----

Where these two threads merge is on the subject of tuning. 

Let's say you want to run your G2 at 90hz.  If you want things to look smooth, you're gonna need to hit a target.  Ideally you want your frame time to be 11ms.  That way you'll be at 90FPS most of the time, which matches your monitor refresh rate. Won't be perfectly smooth -- if you're zipping along at 500kts/500ft things are gonna look a little blurry along the 3-9 line, and you'll be jonesing for one of those 144hz HMD's and griping cuz DCS can't keep up with that.  

If you can't hit that target (don't feel bad -- nobody can) then your next target is 20ms, which should give you 45fps.  Turns out, that'll be smooth too, albeit with even more blurriness along the 3-9 line.  But you have to be right on it.  If you're drifting up toward 60fps you're gonna get a great deal of microstutter which people will describe as ghosting, double images or stuttering.  As you get closer to 80fps, you're gonna be dropping a lot of frames and it's going to annoy you to no end.  

I think a  lot of people with high performance GPU's get into this situation, where they can't quite hold 90fps all the time, but can't quite figure out how to get rid of the microstutter.  Several ways of dealing with this, but clearly, turning settings down doesn't help.  You gotta turn em up to get closer to that 20ms frame time.   So if you look at the process @speed-of-heat followed with his 3090, he exhaustively studied the effect of each setting on his render times, and found the perfect combination to get him right where he needed to be.  Every bit as much work as @edmuss did with his 3070, just moving in a different direction.  Trying to slow things down a little rather than speed em up (which is a good problem to have, believe me).

For those who need to turn their settings up to reach a target -- who need to slow their computer down a little to dance better with VR -- I'd suggest looking at your resolution slider.  It's an easy way to get things just right, and it has its advantages.  Among them, it has an anti-aliasing effect and it does reduce shimmering.  Among them, it may make your sweet spot bigger, as one poster recently reported.  

 

Hey Mike seeing as how you and I have similar hardware...

Have you tried NTTC at 60Hz, OXR tools slider at 130%, FSR 100%?

With a few minor adjustments using 3DMigoto.

It's gorgeous!!

It's the most "3D" I've ever seen DCS run.

Ryzen 7 5800X3D | 64GB DDR4 3600| MSI RTX 4080 16GB Ventus 3X OC  | Samsung 970 Evo 2TB NVME | HP Reverb G2 | DIY Head Tracker Cap | Logitech X-56 throttle | VKB NXT Premium |  Win 11

"Any sufficiently advanced technology is indistinguishable from magic."

--Arthur C Clark

Link to comment
Share on other sites

58 minutes ago, Sr. said:

Hey Mike seeing as how you and I have similar hardware...

Have you tried NTTC at 60Hz, OXR tools slider at 130%, FSR 100%?

With a few minor adjustments using 3DMigoto.

It's gorgeous!!

It's the most "3D" I've ever seen DCS run.

I may, only problem is, I've found myself afflicted by Liberation.  Even though I'm running it on a server I'll be flabbergasted if I can hold 60fps. 

Good news is, I've discovered I own the night. There are no microstutters at night.  

But, I imagine 60hz will be pretty sweet when I'm noodling around in my Christian Eagle.  That should be the shiznit.  Eager to try it.  May try it with Civ Traffic on, although what I really need is civilians. Seems like there should be pretty babies down there when I'm buzzing the pool at Caesars'.

 

Ryzen 5600X (stock), GBX570, 32Gb RAM, AMD 6900XT (reference), G2, WInwing Orion HOTAS, T-flight rudder

Link to comment
Share on other sites

2 hours ago, DeltaMike said:

Among them, it may make your sweet spot bigger, as one poster recently reported

..and I'll confirm it again. I previously struggled to crane my old neck over far enough to get a sharp focus on the rear side cosole dials in the hornet. I either had to adjust my seated position or use a hand to move the headset over and down a little on my face. At 150% SS however, I can focus on them without intervention.

  • Like 1

RYZEN 5900X | 32GB | ASUS Strix RTX3090 | 500GB NVMe OS 1000GB NVMe DCS | Warthog HOTAS | HP Reverb G2 | VA & ViacomPRO

My DCS Apps:    Radio KAOS for DCS      KB Quick - Quick and Easy Kneeboards

Link to comment
Share on other sites

9 minutes ago, ShaunX said:

..and I'll confirm it again. I previously struggled to crane my old neck over far enough to get a sharp focus on the rear side cosole dials in the hornet. I either had to adjust my seated position or use a hand to move the headset over and down a little on my face. At 150% SS however, I can focus on them without intervention.

I've noticed this as well. It no longer takes a series of awkward contortions to adjust the radios in the A-10CII or simply find the HMCS switch. 

Ryzen 7 5800X3D | 64GB DDR4 3600| MSI RTX 4080 16GB Ventus 3X OC  | Samsung 970 Evo 2TB NVME | HP Reverb G2 | DIY Head Tracker Cap | Logitech X-56 throttle | VKB NXT Premium |  Win 11

"Any sufficiently advanced technology is indistinguishable from magic."

--Arthur C Clark

Link to comment
Share on other sites

3 hours ago, DeltaMike said:

Won't be perfectly smooth -- if you're zipping along at 500kts/500ft things are gonna look a little blurry along the 3-9 line, and you'll be jonesing for one of those 144hz HMD's and griping cuz DCS can't keep up with that.  

If you can't hit that target (don't feel bad -- nobody can) then your next target is 20ms, which should give you 45fps.  Turns out, that'll be smooth too, albeit with even more blurriness along the 3-9 line.  But you have to be right on it.

For reference, if you're above refresh rate, be it 60 or 90 it will be smooth with no stutter or blur, no matter the speed or height. Drop below and it will start to desync the frames and you'll witness ghosting to a certain degree.

20ms is 50fps not 45. 1000 / fps = frametime.

If you're locked to 45 fps the you'll be relying on openxr vsync which will smooth out rotational movement but not translational. Looking 3-9 will give ghosting outside of the cockpit because what you're looking at is translating relative to your viewpoint.

Ryzen7 7800X3D / RTX3080ti / 64GB DDR5 4800 / Varjo Aero / Leap Motion / Kinect Headtracking
TM 28" Warthog Deltasim Hotas / DIY Pendular Rudders / DIY Cyclic Maglock Trimmer / DIY Abris / TM TX 599 evo wheel / TM T3PA pro / DIY 7+1+Sequential Shifter / DIY Handbrake / Cobra Clubman Seat
Shoehorned into a 43" x 43" cupboard.

Link to comment
Share on other sites

Im really struggling with it right now.

I got superb results with my old PC (i9-10900K @ 3080, WIN10).

Now i got my new PC (i9-12900K @ 3080TI, WIN11) i have lower frames (about 10-15) as with SteamVR. Strange? Anyone made the same experience?

Cause of WIN11 maybe? 

DCS was clear and fresh isntalled, 3 files put into bin...made same then before.


Edited by Eisprinzessin

Specs:
12th Gen Intel(R) Core(TM) i9-12900K   3.20 GHz,  RAM 128 GB, Win11 Home, RTX3080Ti

Link to comment
Share on other sites

31 minutes ago, edmuss said:

For reference, if you're above refresh rate, be it 60 or 90 it will be smooth with no stutter or blur, no matter the speed or height. Drop below and it will start to desync the frames and you'll witness ghosting to a certain degree.

20ms is 50fps not 45. 1000 / fps = frametime.

If you're locked to 45 fps the you'll be relying on openxr vsync which will smooth out rotational movement but not translational. Looking 3-9 will give ghosting outside of the cockpit because what you're looking at is translating relative to your viewpoint.

Yeah I go for 20 to account for frame time variability and also because I feel it's better to be a little fast than a little slow.  That said, fwiw my frame rate counter has been showing 46 pretty consistently so I'm sure I'm a bit above that most of the time... 

Does the same thing apply to the 60fps/60hz target?  Should I leave myself a little wiggle room or does it matter? 

-----

@Eisprinzessin I'd ping @speed-of-heat for Win11 settings, a couple of which he mentioned in his 3090 tuning thread.  Interested to know what you find, I've been sorely tempted (for reasons I cannot explain) to hit that "upgrade" button.  


Edited by DeltaMike

Ryzen 5600X (stock), GBX570, 32Gb RAM, AMD 6900XT (reference), G2, WInwing Orion HOTAS, T-flight rudder

Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...