Dangerzone

Hi All, Just wondering if Olympus is still active and if it should work on the new Germany CW Server? I see there's no update since 2024, so not sure if CW Germany is supported? Thanks DZ

That's the exact issue I've been having. Not sure if you've had a chance to go through the checklist of mine posted above your post, but the solution for me (which apparently isn't a good solution, but did work) was to use RivaTuner to set the FPS at exactly 72fps. (Not above). This changed the messages from flashing orange/red to a green GPU Bound - giving an theoritical FPS limit too, which is much higher than the 72FPS that I'm limiting at. People far more knowlegable than I say that there should be no need to use RivaTuner, so take this advice with a grain of salt, but if it helps for now... it's worth giving a try with the caviet of keeping an eye on the 'chat' about stuttering because you may need to disable it once there's further updates. (Because all this though seems to have only occurred since 2.16. I don't think ED have acknowledged that there even is an issue - but I suspect (hope) that they do know of this and have found an issue, and will hopefully be addressed with the next update, because numerous people are noticing issues since the latest update). As per the thread above, if this works for you, it's an indication of a deeper issue that using Turbo mode may fix - however as mentioned in my case, while Turbomode made things significantly better, I still have times where using RivaTuner works better (albeit even though it shouldn't). IDK why and am still investigating. I have the same issues as yours though in VR, yet I run Pimax. I was wondering if it was a Pimax related issue, but given it's introduced with DCS 2.16, plus it's happening to non-Pimax headsets too - all fingers appear to be pointing to DCS.

Hmm - thanks. I'll check this out. Normally it will pop up with an alert if it blocks something, but I'll check it out and then will try again. Thanks!

Not sure if this checklist might be helpful for you: https://forum.dcs.world/topic/374746-june-2025-dcs-vr-optimization-stutterjitter-generic-checklist/#comment-5657256

hi Qcumber, I tried downloading and running the XRFrameTools installer, but was unable to get it to install. It did nothing. (There was no popup, no setup, anything). Just wondering if I'm doing this right, or if it installs silently behind the scenes, etc? Thanks actually_fred. I'll try changing mine and will see if I notice any difference. Also - very strange, but I was playing DCS on the weekend and stumbled across the stuttering again (with Turbo mode on). I task swtiched out and loaded up Rivatuner and task switched back in, and it locked straight to 72 fps, and went from flashing between Orange & Red CPU bound, to a green GPU Bound - and a theoretical limit in the high 90's. (I was running at 72fps). I don't doubt that the FPS limiter in Rivatuner is not ideal, and may cause microstutters, but it's far better than the frequent jittering that I get without it. (This changes depending on the mission/conditions. It's as though I must be close to the limit of CPU binding with whatever the spikes are, and once it goes over, it starts to fall apart). Very confusing to see how many different settings work so differently for different users. You've almost got to be an IT Technician just to use DCS in VR these days it seems.

LCTRL-F11 (I actually thought it was LALT-F11 - but I think I'm wrong ) does work in VR for sure. It doesn't move the pilot - it's a 'free camera', but moving the pilot doesn't really do much anyway. It only moves it for the player. For all other players observing -the pilot is static where they landed. About the only benefit the moving pilot really is for, is maybe some video recording work, but you wouldn't be in VR for that anyway.

Indeed. Where I've been running no shadows, no smoke, no heatblur, etc - doing everything on low to try and figure out what darn thing is causing these stutters and how to get rid of them. For now I've been fine with DCS running at 50% of what it could be - just to get a smooth experience, and then once I've got that stable and proven, I can start working my way but up to the best visuals possible. This is a major reason why I wrote the checklist. I feel like I've been 'groping in the dark' with DCS experiences trying to stumble across a solution. In some cases it seemed there were combinations, not just one thing that helped. By how much, I don't know. Part of a combination of changes too was due to the amount of hours invested, and then trying 2 or 3 things at once because the load times to get into DCS and close down again in the last couple of versions have been extreme, so I also tried utilising my time more efficiently. Dealing with the stutter issues hasn't been a smooth experience for me (excuse the pun). The FPS limiter was a solution I stumbled across, but then as pointed out by someone else - bad idea - here's a far better idea. The idea of making this list is to encourage discussions on things that may be wrong, or work better. Most of what I've found is lists, but no deeper information on when it's applicable, why it works, whether it's still applicable today, etc. Diagnosing these issues is at best a dogs breakfast to try and work through. (At least it was for me). In that spirit, are you saying that Game mode improves the experience for you? Same with Full Screen, HAGS, and HPET? That now that I have a solution, I should try undoing these settings one at a time because they actually give better performance? The list is more for me to learn from and be corrected than it is just giving other people ideas of what to try. Oh - 100%. This definitely isn't a "Do this". It's a "Check list" of things to try, as well as a list of things that need more discussion. Some of these things were recommendations, but I don't even know the why or how they are supposed to make a difference, or whether they're suggestions that are so old, that things have improved and they're no longer relevant. (HAGS, Full Screen, etc)?

I see your point, but I don't think CA with infantry gives the same experience. I've played FPS games where the idea is stealth. Run, squat hide, etc while enemy are closing in on your position. Fast paced - behind enemy lines kind of thing. It's smooth flow and high action. Using CA with an infantry guy doesn't really have the same experience as other FPS games. Don't get me wrong - I think CA is a great idea (and wish it was given move love too, TBH) - but the idea of merging FPS with DCS would be the ultimate IMO. Real guys running around on the ground, calling in strikes, etc in their own little world for a full combat simulation could have a lot of fun indeed. But I also understand that it would only serve a smaller % of the population (DCS is already a niche market, and there's far less that do MP already - and this feature would be far more nicher... (no that's not a word but I'm using it anyway ) that it wouldn't be a viable option. But who knows... maybe one day there will be an option for an API link between DCS world and some wild FPS game out there that allows the best of both worlds. If/when DCS actually goes 'full globe' - would it be much of a stretch to imagine some startup company could include basic aircraft in their game, and use API between the two to merge two different game environments together? DCS streams out where what aircraft and munitions are. The game streams back where ground troops and vehicles are running around on. I imagine taking off in an AC-130, getting on station, helping some guys out in their skirmish and RTB while hey continue on with their mission. I know it's the realm of dreams, but still fun to consider what could be one day. I know there's holes in this too that would need to be addressed, and I'm dreaming. The AC-130 isn't really happening (at least yet) - but neither is this option - just talking some day in the FUUUUUTURE. (As Andre would say). However... the 45 minute RTB helicopter RTB ride... yeah - I can see your point there. Might be time for me to come out of fantasy land.

Unfortunately I think the walking pilot is a bit of a “I wonder if we can get this to work” side quest for one of the devs rather than something to take seriously. But man, what I’d give to be able to eject and then forest person shooter my way hiding and evading the enemy and trying to get to a evac point and have someone fly a CSAR helo in, in DCS. That would be epic!! but I think not high up on the ED priority list. But one can dream.

I’m fairly sure I heard ED announce that DCE will be released as SP only to start with. MP will come later… how much later who knows. I think DCS does well without modules and maps for beginners. Free modules and maps, and a 2 week trial on most others is fairly generous. One of the biggest helps I’ve seen for beginners is co-op MP PVE environments where others are willing to help them out. When I was first starting out (and even occasionally now after having a break from a particular airframe), to comment in discord saying “why can’t I get this to work” and someone chime in and say “did you remember to set the MFD active”, etc has been invaluable. DCS is primarily SP, and I think with that often gets overlooked how much more valuable the right types of MP servers/communities/environments can be for beginners. How many RL pilots are left to their own to learn? Not many. . For this type of stuff the less popular PVE or PVPVE servers with a tight group of enthusiasts probably fair far better than the big player base more known servers. One of the larger hurdle for newcomers IMO, as well as retention is stability and smooth onboarding. Not everyone has the stamina to push through all the setting up/binding then diagnosing issues to get going and then have an update create more issues that need diagnosing. Even basic things like no default bindings - so people don’t have to know to search for another peripheral that might have auto-bound an axis would be good. Yes, basic stuff for us DCS veterans, but not newbys, some who may have fought with controls not behaving and giving up before even getting a chance to fly. It would be good to see some of these things change just to help users have a smoother onboarding experience.

Hi QCumber. Sure - I hope to be home this weekend, so will try and give this a test then. I've only been flying on CWG lately, so not sure how this goes with other maps. I actually thought that CWG was already optimized (given that it runs so well with the incredible amount of detail) - so are you saying that there's more optimization that's required on it. (Besides the bit up north of course where people seem to be getting some frames tanking - but I would take that as more of a bug than just optimization?) Either way - leave this with me and I'll test on Syria when I can and see if it is any different, and if not - get you some logs. Cheers DZ

Thanks Mordants - I wasn't aware of this one. I'll be sure to check it out! Thanks. I've posted a link to your post on the first post of mine. I think we're trying to achieve something slightly different (yours is more about getting the most performance out of your system, mine is focused more on troubleshooting jitters/stutters, but the two definitely overlap in some areas). As for some working the opposite way - could you tell me which ones, so I can tag them in the original message. I'm not saying that my list is anywhere near right - it's more for discussion (and I'll update my list as people correct me). The reason I've posted this thread is to get feedback on what's I have wrong. (After being corrected about using FPS limiters and that it's an indication of needing 'turbo mode on' - that made me realise some of the information I have been getting (and even giving too) has been wrong, and while it may help in some areas - it's covering up a far better solution. Thanks for this. What I found was that having preload radius up high (on a 4090) created major stuttering when I was taking off in a helicopter on CW Germany and rotating to face the opposite direction. Reducing the preload radius reduced the stutters. I thought having so much VRAM upping preload should do the opposite, so I'm very confused about this. However I also wonder/question if there's different types of 'stutters' - one that deals with initial loading/preloading/etc - and another one that deals with IRQ interrupts / ticks / or whatever turbo mode fixes, etc. Thanks for also adding the link to your post as well. At the moment I'm finding VR in DCS extremely confusing when it comes to optimizations, what works, what kinda works but doesn't really, etc. Thanks. I think you're right - you have to turn off core isolation memory integrity completely. I'll update my original post.

dcs.log is just a text file - you could try right-clicking and choosing 'send to zip' and then uploading the .zip file maybe?

After being corrected / shown a better options with my VR optimization the other day by @actually_fred (Thanks again!) - I just wanted to put down a checklist/summary that I currently have for VR optimization for both critique, and also for potential help - as optimizing VR in DCS seems to have a lot of different nuances, and to say I have spent over 20 hours in the last 2 months trying to fix issues would not be an exaggeration. So, here's a summary I wrote up - current as at June 2025. Some of these may be wrong (as I've recently been corrected about the FPS limiting), so please feel free to comment / correct / etc, but hopefully they will be helpful to others who are having performance issues and/or doing diagnostics. Please note: Some of these may have the opposite effect depending on your system. It's best to get a baseline check - and then test. FPS Limiting / Turbo Mode If you find using FPS limiting (such as Rivatuner) creates better frames with fewer frame drops - this points to a bigger problem that may be betterr overcome by removing FPS limiting and using OpenXR Toolkit and setting Turbo Mode to ON. (Alternatively Quadviews FR also has turbomode and this could be used too). (Thanks Fred!) In short, you should not need to or be using FPS limiters in VR. (And if you are - try removing the limiter and use TurboMode instead.) As with everything in this list, test though, as YMMV. DCS Autoexec.CFG Add the following to DCS's Autoexec.cfg file in the saved games\config folder: no_device_hotplug = true -- Disables searching for new devices after starting - may reduce performance/stuttering disable_write_track = true -- Disables track saving - may help reduce performance issues options.graphics.stereo_mode_use_shared_parser = true -- Disables rendering of unseen parts in VR ?? Don't do if it causes artifacts bkg_load_signed_plugins = false -- Stops all modules loading at start by default. May reduce load and close times? DLSS_Preset = 'K' -- No performance improvement but worth having for visuals if using DLSS HUD_MFD_after_DLSS = true --Not so much for performance, but clarity of MDF's - only works if using DLSS ... is there any others I'm missing? Disable CPU Parking CPU Parking can cause stuttering in DCS. Windows updates (and other functions) can re-enable this without the user being aware. Note: Always worth double-checking this when experiencing stutters as Windows has a habit of undoing your changes. Park Control is a very useful tool to monitor and keep this disabled. DCS Settings Turn off "Full Screen" for best results in VR Consider reduce Preload radius if having jitters when first playing DCS Start with Sharpening on 0. (Sharpening in DCS can be a performance hit - YMMV. Increase only after getting stable results) TACView Can have CPU performance impacts. Best to have this disabled while investigating Keyboard & Mouse Polling Rates Both gaming keyboards and more so gaming mice can have polling rates at 1000. This has been reported to create issues in DCS (especially when using the keyboard or mouse). Reduce the polling rates to 500 or below. Process Lasso Controversial. Some people report significant improvements limiting DCS to P-Cores only. Others (including ED's Log file AI Analyzer I believe) state to remove all affinity and let DCS do it's own thing. Yet others report denying access to core #8 after DCS starts fixing issues. YYMV. If using - for Pimax users - make sure that Pimax software has access to P-Cores as well for the VR processing. BIOS Enable Rebar Enable XMP BIOS Profiles (for RAM overclocking - note - some users have issues / crashes with this, so stress test!) Disable Speedshift, power management/saving & C-States (force CPU to run at max speed the whole time) On some BIOS's: Disable automatic software downloading. (Stops the BIOS from telling windows to download and install services). NVIDIA CPanel Settings Set performance mode to Prefer Maximum Performance. VR Pre-rendered frames. Some say to set to 1, others to 2 or 3. YMMV - each needs to test their own. Make sure Rebar is Enabled (NVIDIA Inspector). Set RBAR Size limit to 0x0000000400000000 (NVIDIA Inspector) Set RBAR Options to 0x00000001 LOD BIAS. Allow can help with some FPS, but CLAMP is better for better visuals. Only go ALLOW if GPU bound and need to squeeze a bit more. (Better to lower settings in other areas?) Low Latency Mode: Avoid ULTRA! (Use off or On as Ultra can cause jitters) Windows Checklist Set Power plan to Ultimate (and verify it stays there after windows updates) Disable XBOX Game Bar Disable Hardware-Accelerated GPU Scheduling (Only enable if fully GPU bound) Disable Game Mode Exclude DCS Directories (Saved games and install) from Antivirus Check. (At your discretion!) Set DCS.EXE to High performance. (Regedit - HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Image File Execution Options\DCS.exe Add or set DWORD 32bit option of "PerfOptions" to value 3 Stop timer from sleeping - Run "bcdedit /set disabledynamictick yes" in elevated prompt Disable Memory Compression (run "Disable-MMAgent -mc" in Elevated Powershell, or run "Enable-MMAgent -mc" to revert) Consider a fixed pagefile size of 32768 (or higher) Remove any 'Gaming RGB' services and applications from startup. Turn off Windows Indexing on Drives Disabled Superfetch (Sysmain Service) in Services. (Stops Windows from preloading frequently used apps into RAM / Reduces background disk/CPU usage) Search all USB Devices in Device Manager and disable power management (allow this device to turn off to save power). Miscellaneous Disable VR Hanger. (Reduces the amount of unnecessary VRAM used / can help with F10 map /. GPU RAM issues) Consider using Quadviews. (Uses more CPU resources, but free's up GPU significantly. YMMV!) Avoid USB HUBS where possible and plug in devices direct to motherboard. Experimental: Best to avoid unless needing to dig deep: Disabled Microsoft Device Association Root Enumerator (Research what this does and if needed first) Changed NVIDIA Shader cache from driver controlled to 10GB Computer\HKEY_CURRENT_USER\System\GameConfigStore Set 32bit DWord Value GameDVR_FSEBehavior to 2 (Disables fullscreen optimization) Consider disabling any unused Audio devices (Onboard Realtek, etc) as well as unused devices (wifi / bluetooth adapters, etc). Marginal gains, but can help with troubleshooting if they're causing IRQ/Interrupt issues. Consider reducing Soundcard rate down to 44000 (Marginal gains, but can help with troubleshooting) Turn off windows core isolation memory integrityy. In addition to these, if chasing general performance optimization, it's worth checking out @Ready's post:

Hey Yoyo. Thanks for providing this. Are you able to advise where you found the info for bkg_load_signed_plugins please?And do you have any more documentation on what it does or the change? I mean, I understand that it doesn't load all the official modules in at the start - but what are the downfalls. When does it load in the modules? Only when a player joins a server with them? Is this like a 'test feature' that ED has put in for testing with the plans of actually making it a core change in an upcoming change - or is it solely for debugging? You've sparked my interest - because it seems to make sense not to load modules in unless they're needed, but yet it's a hidden option with no documentation that I can find. The only google search for "bkg_load_signed_plugins" reveals this thread.

@actually_fred - there's a new one for me. I've tested your recommendation with Turbo to on and I don't need to use the FPS limiter anymore to get a good experience. And you were right - if anything this has not only matched the FPS limiter in terms of a workaround but it's an improvement on what i was having - even less of the microstutters. Thank you! I may revisit QVFR at some stage. I turned it off due to "CPU Bound" issues I was experiencing to start with. Seems I have plenty of GPU overhead, but am severely limited by CPU. If that turbo mode does the same thing, then if I go back to using QVFR I may end up using it again. Thanks fort he suggestion!

Thanks so much for that information. I'll definitely give this a shot. Also - just to confirm, when you say Turbo mode - I'm assuming that you mean Turbo Mode in OpenXRToolkit? Or is there another turbo mode setting/configuration that can be done that I'm not aware of? Thanks DZ

Thanks WipeUOut. I checked MSI. My GPU was selectd for MSI, however the interrupt priority was undefined. I've changed that to High now, so will be keen to see if this helps! Cheers DZ

My apologies. You are right. I misread that twice. Thanks so much for your patience and pointing it out, and again my sincere apologies. Thanks for the info on turbo mode too. I wasn’t aware that could help with this issue so will investigate that too. Cheers! DZ

But if you fly in a squadron - I'm guessing someone other than ED is creating missions and hosting the server, so you are relying on community contributions such as the server hosting, mission designing and maybe scripting too, etc? Sure thing. And I need to make an apology to you. I think my previous response to you was bad. My idea was more about how much people rely on other community contributors for their DCS experience. This includes any community contribution. Reflecting on this - it could include Mods (Such as QVFR) for sure. But it could also include non-commercial content creators (C130 mod, etc) - or 3rd party community missions, or using a community server, or using scripts, etc. It's a pretty broad spectrum. My understanding is that the majority of players are Single Player only - so this poll was a way of referencing just how many people here are happy with "ED only" material, vs how many actually rely on other community members also contributing something to the mix. I wasn't really focusing on QVFR, etc that are essential to people running DCS - but in all honesty reflecting back on this - if they are non commercial community contributions, then I was probably wrong with my previous response, and they probably should be considered, but it's not limited only to those. Or another way to put it. how much would DCS suffer if there was no volunteer / community contribution at all.

Can you please expand on this for the rest of us. Are you saying you had VSync on in DCS, and turned it off? (And have it nowhere now)? Are you also saying that turning off Full Screen fixed all your stutters. And lastly - can you clarify if you're running 2D or VR? Thanks.

Yeah... we've come so far. I remember with DOS games... I'm finished. Half a second later POWER BUTTON and walk away.

You're more optimistic than me. When ED can't even keep their own modules from having repeated issues (scoreboard, ground unit pathing, track replay desync, etc where they have all the source code and can work on it from all angles - it seems more plausible that sooner or later there's going to be a change they make that's going to change something that breaks the F15E in a similar way too - but without the F15's source code to be able to diagnose from all angles - well, as I said - if they struggle when they do have it, this can only make it far more difficult. (This isn't a go at ED with their bugs - but just to point out that if it's so involved when you have the source code, it's going to be far harder without) Agreed. I see only 2 ways that the F15E survives this. Either Razbam and ED come back to a mutually beneficial arrangement, or somewhere in the contract it's stipulated that ED must get a copy of the source code and the legal proceedings end up forcing Razbam to hand it over. Otherwise, without developers actively maintaining the source code - it would be a very hard ask to expect this to remain operational long term.

That's definitely not true in my case. If you are right - then there's a connection between you and I that is missing. To clarify though - and maybe this is where the miscommunication is: The problem isn't that I'm getting over 72FPS without limiting it - the VR does indeed limit the frames still to 72 as you say - so I agree there. (So if you think we're saying RTSS is needed to limit the FPS to 72 in the headset - that's not what I'm saying). What I'm saying is if I use RTSS and set 72FPS - I get a far more stable 72 fps with far less drops than compared to if I wasn't using RTSS. Now as for the part that RTSS only affects the monitor - if I change RTSS to 60fps - according to what I understand from what you've written - this should only change the mirror window to 60fps? That's not the case - what I observe is immediately after changing this setting, my VR drops to 60fps. As such - I can only conclude that RTSS does indeed force a limit on DCS's FPS in VR. This isn't placebo - I'm viewing the DCS's actual FPS counter in VR - and I also see the effects visually well. I can lower this significantly more to 30fps or less - and the effects are even more obvious. Obviously I have no intentions to lower FPS that far - but for the sake of this exercise to test if your theory is correct that VR isn't affected - and I can conclude it definitely is affected by RTSS. The headset can't go above it's refresh rate - but RTSS is definitely affecting the frames that the HMD is delivering. Not only visually - but I can see this also reported in DCS's FPS dialog as well. The main reason to limit FPS to 72 using RTSS in VR isn't to lower the FPS to 72 because it's rendering more than 72fps- but rather because doing so gives a clear change to the amount of stuttering and CPU binding that is going on. I don't know the reason for this. Maybe the 72FPS isn't being fully limited correctly with Pimax software, or DCS has some HID interrupts coming in the wrong place that causes issues that manually setting RTSS to 72FPS helps as a workaround. I suspect DCS since this was working smooth on the previous version of DCS but the latest updates cause issues. (And rolling back to previous version works fine again). The reasons I can only guess - but the effects are plainly obvious. For me - without RTSS I'm removing my headset after 10 minutes with a headache due to the very obvious and significant stuttering. With RTSS at 72 - the major stutters have gone (the odd micro stutter remains),but I can play for hours. I also just tested again too - running with no limit gives me 72fps in VR with significant stutters. (72FPS dropping to around 60FPS as reported by DCS as well as is plain to see - no placebo). I see CPU Binding flash between yellow and green in the DCS FPS window as well. Then, setting RTSS to limit to 72FPS, the DCS FPS window shows a green 'GPU Bound' the majority of the time, with the occasional stutter - and I see a more consistent 72FPS in the DCS FPS window - I definitely see far fewer stutters. Cheers DZ

I'm curious about this. In my case it's definitely not a placebo. I can literally be in my headset - see what DCS is doing - while DCS is still running set the FPS limit in RivaTuner, come back and see a significant difference in what I experience in DCS plus what the DCS FPS Window is telling me. If this is not supposed to have any impact - I'm very keen to find out why it does - as anything I can do to remedy the issues DCS is having another way - I'm all up for learning more. However if I change RTSS to 60FPS - it definitely locks my DCS Frames to 60FPS in VR. My best guess is that there's some sort of over-rendering happening with DCS behind the scenes, and limiting it to 72FPS helps DCS to not 'race ahead' - but all I know is what I've done from experimentation - I'm no expert at this. Are you saying we should be enabling VSYNC for the VR mirror / monitors sake instead? I don't believe I have any VSYNC enabled. The only limitor there at present is Rivatuner? I'd be happy not to have a mirror at all on my desktop - if it meant better experience in VR, but I don't think that's possible?