dlder
-
Posts
251 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by dlder
-
-
On 8/1/2023 at 9:58 AM, Hotdognz said:
I've just now catched that: why is your resolution so low? (11.8 Megapixel)
PimaxXR for me shows "4.312 x 5.102" (20.98 MP), which is also weird and not the native resolution of "5.760 x 2.880" (15.82 MP).
-> this means I'm currently rendering double the amount of pixels... why?
-
So, did some more testing:
Oh, first things first: whenever I change something in the Pimax-Client, or when OpenXR-TK tells me to, I restart DCS. So pretty much every test was a new game start.
-> graphical glitches (like lightning flashes) with Pimax "SmartSmoothing"; at least in DCS (I've created a video showing that:
-> Pimax FFR (fixed foveated rendering) doesn't really change anything either? Neither visually, nor performance wise
What is Minimum and Maximum anyway? "min" is about 0.5 fps slower
-> "Hidden Area Mask" area mask isn't doing anything either; it seems to me, with the exception of FFR, that the Pimax-Client settings are only for SteamVR?
Spoiler04-08-2023, 09:40:51 DCS.exe benchmark completed, 4504 frames rendered in 108.187 s
"2d settings"
+pimax SmartSmoothing + Hidden Area Mask
Average framerate : 41.6 FPS
Minimum framerate : 9.9 FPS
Maximum framerate : 46.9 FPS
1% low framerate : 22.9 FPS
0.1% low framerate : 1.2 FPS
04-08-2023, 09:49:10 DCS.exe benchmark completed, 4501 frames rendered in 107.500 s
"2d settings"
+pimax SmartSmoothing
Average framerate : 41.8 FPS
Minimum framerate : 30.9 FPS
Maximum framerate : 47.0 FPS
1% low framerate : 23.5 FPS
0.1% low framerate : 22.9 FPS)04-08-2023, 09:54:41 DCS.exe benchmark completed, 4666 frames rendered in 107.453 s
"2d settings"
Average framerate : 43.4 FPS
Minimum framerate : 31.3 FPS
Maximum framerate : 50.4 FPS
1% low framerate : 29.9 FPS
0.1% low framerate : 29.8 FPS
04-08-2023, 10:01:20 DCS.exe benchmark completed, 4658 frames rendered in 108.000 s
"2d settings"
+pimax Hidden Area Mask
Average framerate : 43.1 FPS
Minimum framerate : 31.9 FPS
Maximum framerate : 49.4 FPS
1% low framerate : 29.9 FPS
0.1% low framerate : 29.8 FPS
04-08-2023, 10:08:25 DCS.exe benchmark completed, 4617 frames rendered in 106.875 s
"2d settings"
+pimax Hidden Area Mask + FFR MAX
Average framerate : 43.2 FPS
Minimum framerate : 31.6 FPS
Maximum framerate : 49.4 FPS
1% low framerate : 29.9 FPS
0.1% low framerate : 29.8 FPS
04-08-2023, 10:13:06 DCS.exe benchmark completed, 4597 frames rendered in 107.110 s
"2d settings"
+pimax Hidden Area Mask + FFR MIN
Average framerate : 42.9 FPS
Minimum framerate : 30.9 FPS
Maximum framerate : 49.4 FPS
1% low framerate : 29.9 FPS
0.1% low framerate : 29.8 FPS-> OpenXR TurboMode doesn't do anything
-> NIS 75% is a huge performance gain of course; DLSS would be much preferred though. But when enabling NIS, why is it saying 100% resolution is "4312x5102"? The HMD has "5.760 x 2.880"!
-> FFR works; but the performance gain isn't all that great
Spoiler04-08-2023, 10:26:38 DCS.exe benchmark completed, 4654 frames rendered in 106.796 s
"2d settings"
+pimax Hidden Area Mask
+openxr turbo mode
Average framerate : 43.5 FPS
Minimum framerate : 30.9 FPS
Maximum framerate : 50.4 FPS
1% low framerate : 29.9 FPS
0.1% low framerate : 29.8 FPS
04-08-2023, 10:33:57 DCS.exe benchmark completed, 6618 frames rendered in 105.859 s
"2d settings"
+pimax Hidden Area Mask
+openxr NIS 75%, mip-map conservative
Average framerate : 62.5 FPS
Minimum framerate : 45.4 FPS
Maximum framerate : 73.2 FPS
1% low framerate : 44.6 FPS
0.1% low framerate : 43.5 FPS04-08-2023, 11:00:47 DCS.exe benchmark completed, 4646 frames rendered in 102.516 s
"2d settings"
+pimax Hidden Area Mask
+openxr FFR, quality, balanced
Average framerate : 45.3 FPS
Minimum framerate : 33.6 FPS
Maximum framerate : 51.8 FPS
1% low framerate : 29.9 FPS
0.1% low framerate : 29.7 FPS -
Hm... OpenXR Tool Kit can capture performance data? I guess I've overlooked that bit; gonna try that next.
Hz / Fps: the only correlation between those two are, if you are using VSync (or FPS limiter in accordance to GSync best-practice). I do (that's probably why my max FPS are 91), but I have a frame limiter at 77fps, but that doesn't take effect in VR, so I didn't even think about deactivating Gsync, as that is also very much monitor dependent, BUT: because of that I have Vsync turned on "globally". Now, I don't know if the nVidia DCS profile is actually being used in every instance - this being beta-multithreading exe - so I can't be sure.
Long story short: the "max" fps might not be correct, but that really is besides the point anyway, because I don't really care about getting more then 90 fps if all I'm getting is 27 at Marianas^^-> I now know, that this is really an unusual case and most seem to get their performance from Caucasus or some lighter maps like Nevada / Syria (sand maps).
That's exactly why I used your track file; to be on the same playing field.
Pixel Density: I use 1.0 in DCS (and everywhere, as already stated), because this is effectively supersampling and much more power-hungry then MSAA with MultiSampling Transparency AntiAliasing (which is what I use). To get about the same level of AntiAliasing you'd have to use something like 125% SS (at least), which takes more FPS then MSAA.
Tacview: is imho a crucial program to review ones performance; but it's running only on my Stable install anyway which I don't use for this threads benchmarking (to again be comparable to you and most every other people). This install is also clean, so no mods. Also, after comparing the performance with your trackfile and the exact same settings, both Stable and Beta performance are about the same (1-2 fps difference).
I still have to watch your video, but with the big things out of the way - like VBS (virtualization based security, which basically activates Hyperthreading, which means your Windows installation will be running virtualized on a hypervisor) - I don't think any "Windows Hacks" which have been around since Win 98 will be of any measurable help^^
Thanks for the video anyway; maybe I actually will learn something new^^So, what does that leave me with:
- I could try a fresh Windows install, but from all the things I know, that probably won't help (I might do it on a test-SSD, just to make sure)
-
to not play Marianas, as the performance in Caucasus is "good enough"
- wait for dynamic foveated rendering
- wait for DLSS (or use NIS)
I'll continue my testing in a more "practical" matter (doing missions) and will report back on my findings and hopefully my then-final-settings.
-
1 hour ago, WipeUout said:
Do you have your GPU frametime results?
Sorry, can't find any option in Afterburner
-
-
I'll do another run including the frame times (gotta check Afterburner on how to do that^^). -> please note if you use mirrors or not; I didn't check it and default DCS settings seem to disable them. I've since enabled the mirrors again for VR High (which is everything maxed)
@BIGNEWY Quick question: are we sure that "msaaMaskSize" is doing anything? I've run it with 4xMSAA (just to max out the GPU usage and thus the potential performance gain) and it doesn't do a lot; 0.5 fps seems more like a benchmark tolerance
02-08-2023, 14:14:54 DCS.exe benchmark completed, 3378 frames rendered in 107.797 s
dcs beta MT
"vr high", 4xMSAA, 1.0 MsaaMask
Average framerate : 31.3 FPS
Minimum framerate : 22.7 FPS
Maximum framerate : 36.2 FPS
1% low framerate : 22.4 FPS
0.1% low framerate : 22.3 FPS02-08-2023, 14:19:00 DCS.exe benchmark completed, 3349 frames rendered in 108.937 s
dcs beta MT
"vr high", 4xMSAA, 0.1 MsaaMask
Average framerate : 30.7 FPS
Minimum framerate : 4.9 FPS
Maximum framerate : 35.8 FPS
1% low framerate : 22.4 FPS
0.1% low framerate : 1.2 FPS -
1 hour ago, BIGNEWY said:
NVIDIA control panel, try
low latency - ultra
shader cache - unlimited
threaded optimisations - On
power management - maximum performancedidn't really change much:
02-08-2023, 13:20:11 DCS.exe benchmark completed, 4746 frames rendered in 106.937 s
dcs beta MT
"vr high"
Average framerate : 44.3 FPS
Minimum framerate : 32.6 FPS
Maximum framerate : 50.4 FPS
1% low framerate : 29.9 FPS
0.1% low framerate : 29.8 FPS
02-08-2023, 13:27:19 DCS.exe benchmark completed, 4904 frames rendered in 109.454 s
dcs beta MT
"vr high"+optimized nvidia settings
Average framerate : 44.8 FPS
Minimum framerate : 0.8 FPS
Maximum framerate : 52.1 FPS
1% low framerate : 0.8 FPS
0.1% low framerate : 0.8 FPS -
Will try!
So, I've now tested with a "clean" Beta install (non-steam) and comparing 1:1 to my steam-stable version. Doesn't seem much difference: (all without Mirrors)
02-08-2023, 11:38:27 DCS.exe benchmark completed, 9535 frames rendered in 109.438 s
dcs beta MT
"default dcs vr settings" (no mirrors)
Average framerate : 87.1 FPS
Minimum framerate : 3.9 FPS
Maximum framerate : 91.2 FPS
1% low framerate : 44.2 FPS
0.1% low framerate : 1.0 FPS02-08-2023, 12:06:04 DCS.exe benchmark completed, 4855 frames rendered in 108.641 s
dcs beta MT
"all maximum w/o motion blur" (no mirrors) -> does Beta currently have a problem with MotionBlur? I figured that out through trial&error, that if this is activated, the screen is blank and only blurry MFD-green-text is visible...
Average framerate : 44.6 FPS
Minimum framerate : 9.9 FPS
Maximum framerate : 52.4 FPS
1% low framerate : 29.8 FPS
0.1% low framerate : 1.2 FPS
02-08-2023, 12:22:08 DCS.exe benchmark completed, 4675 frames rendered in 107.313 s
dcs stable mt
"2d w/o motion blur" (no mirrors)
Average framerate : 43.5 FPS
Minimum framerate : 32.6 FPS
Maximum framerate : 50.4 FPS
1% low framerate : 29.8 FPS
0.1% low framerate : 4.6 FPS
02-08-2023, 13:05:30 DCS.exe benchmark completed, 9311 frames rendered in 109.672 s
dcs stable mt
"default dcs vr settings" (no mirrors)
Average framerate : 84.8 FPS
Minimum framerate : 0.4 FPS
Maximum framerate : 90.5 FPS
1% low framerate : 0.4 FPS
0.1% low framerate : 0.4 FPSI've also tested my Marianas flight and it too is the same FPS. So it doesn't seem to be a problem with Tacview or Texture-Mods.
Just for comparison's sake, I'll keep using your Track file in DCS beta.
-
Thanks all!
I'll do some more testing; I'm thinking about doing a clean Windows install on a different SSD (don't wanna ruin my current installation^^), but I doubt it would help (only if something critical is out of tune with Windows - Settings).
I'll try Caucassus next; I wanted to use Marianas because if it runs here (those trees are very taxing for whatever reason), then it will run Caucasus and all the other maps too.
Just for completeness sake, here are my settings:
nVidia
Pimax (needs to be open for the headset to start?)
OpenXR (not much to set up)null
-
Thanks for
15 hours ago, WipeUout said:Since the Crystal and the 8kX share the same number of pixels,
True; the resolution is different, but the amount of pixels is apparently the same. Really weird.
1 hour ago, Hotdognz said:In OpenXR toolkit are you still running the native resolution, I have set mine to override it and set it too this.
Yes, as I've stated in my post, I have it set to 100% everywhere. Maybe some Upsampling can work, but if it's not dynamic foveated rendering, then it's a pretty big clarity loss.
Can I ask: does your Crystal support eye-tracking & auto-ipd? I don't see those options in the Pimax app anywhere... (as can be seen here: https://github.com/mbucchia/Quad-Views-Foveated/wiki/Pimax-Crystal)
The biggest problem I have: even if I fine-tune the graphics and nVidia driver settings, nothing get's me to that level:
15 hours ago, WipeUout said:56-67 FPS in 3D using my Pimax 8kX at 4k per eye
Sure, I don't know his Pixel Density setting in every middle-wear, but lowering that is like a last-ditch effort. And the 4090 can handle the amount of pixels; in 2D. But that shouldn't make that much different then in VR, one would think.
-
Hi,
I've literally tried every setting in DCS (Stable, MT) so far and checked it in a 1500ft flight over Marianas. I'm running OpenXR + PimaxXR and DCS forced to OpenXR. I've set pixel density in PimaxPlay, OpenXR and DCS all to 1.0.
Here are my specs:
SpoilerWin11
i7-13700KF
64 GB DDR5
RTX 4090
PCIe4 M.2
Here are my findings:Spoiler2D, everything max, 2xMSAA, 0xSSAA, same resolution as the VR HMD - 66 fps
VR, same settings - 20 fpsnVidia driver settings don't do much: +0-2 fps (that's mainly from setting Transparency AA from Super- to MultiSampling)
-> setting "VR PreRendered frames" to a higher setting then 1 doesn't matter either!Shadows = Flat only: ~2 fps
Visibility Range:
Low: 8-9 fps
Medium: 6-7 fps
High: 4-5 fps
Ultra: 2-3 fps
MSAA has a large impact: 5 fps for 2x / 12 fps for 4x
DoF has quite a large impact too; but only for external view: 5 fps
Motion Blur has also quite an impact: 6 fps
Clouds to Low: 1-2 fps
SSLR: ~1 fps
SSAO: 2-3 fps
Forest Visibility to 30%: 1-2 fps
Forest Details Factor to 0.1: 1-2 fps
Anisotropic Filtering to 0x: ~2 fps
Terrain Object Shadows:
Off: 2-3 fps
Flat: 1-2 fps
Rain Droplets: ~1 fps (only when raining and droplets are visible on the canopy)
Mirrors: 1 fps=> the lowest playable MSAA setting imho is 2x, while Visibility Range is High or Ultra. That means I can get those 20fps up to 41 to 47; although I'm gonna check that again tomorrow, as in a last test, I got only about 35 fps with the above settings
So, with that out of the way: how is that possible? From 66 to 20 FPS? There are dozens of posts of people running an older i5 with an 1060ti playing in VR... I just don't get how!?
Do you have any tips on what else I could check?
"Windows Game Mode" is on (should help with Win-AntiVirus which isn't power hungry in the first place)
"Windows Graphics preference for DCS" is set to High performance
"Hardware-accelerated GPU scheduling" = onThanks for "hearing me out"^^ I'm truly baffled tbh. I've only tried "Serious Sam VR" (SteamVR, not OpenXR) and it ran flawlessly with max settings. Sure, that's an older title (to be fair: so is DCS), but it shows, that there doesn't seem to be anything "generally wrong" with drivers, APIs and such.
-
The fuselage tank can hold 85ga of fuel and the gauge goes up to 85
What I am not sure about: are the right/left wing fuel gauges non-linear? Because the 75ga is pretty much at the maximum on those gauges, even though it holds 92 gallons:
null
So ist the very last, 100% mark actually 92 gallons? Or is 75 not gallons, but the percentage (what I doubt)? I just can't find that answer in either the DCS manual, nor Army FlightOps, nor Dr. Google...
Cheers!
-
Ok... small update: I'm mildly optimistic that it works, just the way I've configured in the VPC Config.
Seems I just had to restart DCS:
null
I'm gonna continue to configure this Hotas, but I might just finally have figured out how to do it... (fingers crossed)
Thanks anyway
-
I've been playing around with those stupid 4-way HATs that Virpil uses for literally hours now.
If used as default, you only have 4 physical and thus 4 logical buttons. Diagonal are just 2 directional keys combined; that doesn't work in DCS.
There is NO WAY to set them to 8-way; you can use "Toggles Decoding" to create 4 virtual buttons for the diagonal, which does work somewhat (you get a new logical button which can be assigned in DCS), but because both the directional keys + the diagonal key get activated, it can happen that DCS doesn't activate the diagonal, but one of the directional keys. Very frustrating that.So, the only other way I thought might be possible: create a POV; it doesn't use buttons and thus there is no... "cross talk" between the directional and diagonal keys.
The problem here though is: DCS doesn't see POVs? Windows does:
But I can find now way to bind the POV to anything...
null
Any idea what could be wrong?
Or does anyone know how to use a Virpil 4-physical-button-HAT as a 8-button-HAT?Thanks,
dlderPS: for completeness sakes, here's the Virpil forum post: https://forum.virpil.com/index.php?/topic/16977-hatspovs-with-8-buttons/
-
just to add my 2c: I've always been playing at native resolution (3440x1440) and the highest graphical setting. I NEED to use barebones labels and or finding targets is impossible!
Not all pixels are the same; clearly. PPI of displays just changes things and having a higher density / smaller pixels is visually better. But not in DCS, where you get punished for going beyond 19", 1280x1024 it seems.
That may not have been a problem in 2008, but it is now!
-
3
-
-
What you're doing is an awesome idea!
Especially the liveries; nobody needs dozens of GB just for that. The F-14 alone has 11.5GB of liveries...; the F-1 has 7.1GB. AH-64 are 5.1GB, etc, etc.
I hope you'll downsize all of the default ones.
Cheers!
-
2
-
-
Thanks; I did do that:
-
Here's the related topic where I asked about this problem:
The gist of it is this:
-
the manual states
Quote
Launch the missile when the target enters the allowable launch range and the targeting cross starts to flash.
- so, to tell if you are in range to launch your AGM-65, you watch the seeker's crosshair in the TV and if it starts to blink, you can fire the missile
But that's not what happens.
The crosshair starts to blink as soon as you lock onto a target! Doesn't matter if it's in range or not!Here's a video showcasing the problem:
I've also included the track file.
Thanks!
-
the manual states
-
So, I'm pretty sure I cannot post a Trk for that, as it's input related, so let me explain what I mean:
- when using the MAV, I can move the targeting circle around with my mouse. Up, down, left right, just as expected.
- but it doesn't work for bombs for some reason
- I can only move it left and right with the mouse
- to move it up/down, I have to use the keyboard
Here's my mouse view/tdc setup:
Cheers!
-
On 4/4/2023 at 11:30 PM, Flappie said:
Try tweaking your NVIDIA settings
Thanks for that hint; I did go through pretty much all nVidia settings and now I know: it's not Reshade, or Auto-HDR. Nor is it GSync and Ambient Occlusion.
- it's MFAA
I never expedted this to be a problem; for me, it was just a newer, better way to do old-school MultiSampling AntiAliasing.
Seems it can actually cause artifacts in some games (DCS is the first tbh though)...
-
So, I read a bit more in the manual and on page 95 it really does state, that the flashing crosshair indicates "in range":
QuoteLaunch the missile when the target enters the allowable launch range and the targeting cross starts to flash.
-
9 hours ago, Flappie said:
Hi @dlder. I tried your track in current OB but could not replicate the issue shown in the video.
Try tweaking your NVIDIA settings:
See if G-Sync is responsible for your issue:
Will do and report back asap; ty!
-
This is not a new, DCS 2.8 issue (I'm on latest stable), as this has been an issue in 2.7 too; I hoped it would've been resolved by now, but I guess not a lot of people are experiencing that, so it never came up?
It's a kind of shimmering, which isn't always there; I guess it depends on the time of day? Its apparent in the first two missions in the A-10A campaign; it's gone by the 3rd though... but it is at least reproduce able, which is always good to find bugs^^!
Makes me wonder, what special settings I use for that to happen... I'm on max graphical settings in DCS. I had that problem with my previous computer (Ryzen 9 3900X + 1080Ti) and with my new PC (i7-13700KF + 4900), so it's not the hardware.
Would be awesome if other people could try those two missions (you start right next to the water, so it only takes a few minutes to get into the air and the bug is visible)!
But here is a track file nonetheless; hope it helps!
Thanks
-
Just to add: I have these issues too, especially visible with clouds
Help with performance with Crystal + 4090
in Virtual Reality
Posted
Thanks for confirming that this is a "normal" issue^^
Thanks, good to know!
Nope; these numbers - Render Target / Pixel Density - do not correlate with the resolution of the displays. I still have to find out, how:
correlate with:
I've tried a Render Target of 0.75, which would only be 12,37 MP -> the FPS were of course fantastic. But the picture quality isn't what I like about this HMD: "Crystal" clear picture^^
-> If you have a lower one, then you are probably using "balanced" in the Pimax-Client too... not "maximum", which would be Render Target of "1"
Well, of course one can lower the number of calculated Pixel; I just don't see the point. Why buy a high resolution HMD if you are not rendering 100% of those pixels; UpSampling like NIS (which OpenXR can do) is possible and not as bad as just lowering the resolution, but it's still not DLSS, which imho might be up to par quality wise to 100% resolution.
AWESOME, thank you!
I hope that will explain what's happening
I just now from testing that lower then 100% isn't an option. It's like in 2D, but even more apparent, because the displays are right infront of your eyes. So 100% resolution + 2xMSAA is the least one can use.
One can only hope, but testing with FFR makes me believe differently. You just can't set the quality so low that you'll gain like 20% FPS or the visuals will be greatly degraded!
-> so I doubt it will be a game changer. Yes, everyone hypes about the newest and greatest tech / feature, and Hope Dies Last, but I'm not too optimistic.
Good thing I bought the 4090 which is way, way too expansive (still), but with these many pixels to be rendered, I doubt that any type of optimization will substantially improve performance.
I've tried a lot.
I will try more (from the videos that were suggested).
I'm now playing with "High" graphics settings and have 40 (ground) to 90fps 4000ft and up (on Caucasus though^^)
Custom2.lua