Jump to content

Ready

Members
 View Profile  See their activity

  • Posts

    677

  • Joined

  • Last visited

 Content Type 

Everything posted by Ready

  1. Ready
    Why not run my dcs.log through ChatGPT. Here is what it said (after I disabled a bunch of mods to be able to get onto a new server).. Looks like a bunch of old mess. Will go through them one by one and see what the outcome will be. Will share results in next post. -- Got it—here’s a focused, do-this-now plan based on your log. I’ll group fixes by what the log actually shows, starting with the biggest noise makers and stability risks. 1) Remove/disable the broken/legacy mods Your log is flooded by errors from community mods and asset packs. These create conflicts, physics issues (negative drag/weight), and “replace not allowed” spam. Move them out of Saved Games\DCS.openbeta\Mods\ (use OVGME for profiles), then repair DCS (step 4). Mods to remove first (they’re explicitly erroring): VAICOMPRO – not supported and throwing a Lua parse error. Remove Saved Games\DCS.openbeta\Mods\tech\VAICOMPRO. dcs “unexpected symbol” & “not supported”. J-11_Au2th – hard Lua error: attempt to index global 'warheads' (a nil value). Remove Saved Games\DCS.openbeta\Mods\aircraft\J-11_Au2th. dcs HighDigitSAMs / AM2 / SAMSitesAssetPack / LLH_Counter_Insurgency / Insurgent-SF-4x4 – “unit replace … not allowed” + duplicate mounts. These conflict with current core assets. Remove their Saved Games\…\Mods\tech\… folders. dcs Civil Aircraft Mod / Military Aircraft Mod – multiple payload physics errors (negative drag). If you keep them, expect log spam; otherwise remove. dcs Hawk / MB-339 community versions – geometry link errors (“Cell … link loop/self”). These are obsolete versus today’s modules. Remove any community Hawk/MB-339. dcs Result: this alone will eliminate most red ERROR/WARNING lines and prevent mission load weirdness. 2) Fix corrupted/stray livery files F-15E livery .zip fails to open – corrupted archive; a DCS repair will re-download. dcs F-14 livery .gitkeep being mounted as a zip – delete those .gitkeep files in CoreMods\aircraft\F14\liveries\… if you manually added them, or just run a full repair (next section). dcs 3) Clean up texture/font misses Missing texture KevinWakePattern812x1024.dds – likely from a custom livery/ship/water mod. Remove the offending livery or re-install it cleanly. dcs UI font cannot load – indicates a missing/altered dxgui\skins\fonts path; the repair in step 4 will restore it. dcs 4) Run a thorough DCS cleanup/repair (restores core files) After moving/removing the mods above: Close DCS. Open an elevated PowerShell in D:\Games\Eagle Dynamics\DCS World OpenBeta clean\bin and run: .\DCS_updater.exe cleanup (say yes to remove leftovers) .\DCS_updater.exe repair slow (verifies & re-downloads corrupt core files) This will fix the F-15E/F-14 livery issues, UI font path, and any stock file drift. dcs 5) Purge shader caches Delete: C:\Users\Edwin\Saved Games\DCS.openbeta\fxo\ C:\Users\Edwin\Saved Games\DCS.openbeta\metashaders2\ Your log shows the shader caches loading; after large mod churn, rebuild them to avoid stutters. dcs 6) Tidy your audio devices (optional, reduces device spam) You’ve got 9 devices (VB-Cable A–D, ButtKicker virtual, Topping DAC, etc.). Consider disabling unused ones in Windows Sound settings so DCS doesn’t enumerate that many every boot (minor startup savings). dcs 7) Fix graphics settings (you’re stacking AA incorrectly) From your options.graphics dump: you’re running Upscaling = DLSS and AA = DLAA and MSAA = 2 with Scaling = 0.66. That’s a performance & clarity conflict. Pick one path: Path A (best clarity, powerful GPU like 4090): Upscaling: OFF AA: DLAA MSAA: OFF Scaling: 1.0 (remove the 0.66 downscale) Anisotropy: 16x (cheap on 4090) Textures: High; Terrain Textures: Low/Std if VRAM tight; otherwise High This uses native res + DLAA’s high quality edges. Path B (more performance headroom): Upscaling: DLSS = Quality AA: OFF (don’t stack MSAA/DLAA on DLSS) Sharpening: ~0.2–0.3 (you’re at 0.3 already) Scaling: 1.0 (DLSS handles scaling) Cap maxFPS only if you need frametime stability (RTSS at 90 or 60 can help); otherwise set maxFPS = 0. Your log shows DLSS initialized 1280×720 → 2560×1440 while you also had Scaling = 0.66—that double-scales and harms clarity. Fixing this will both clean the image and stabilize GPU frametimes. dcs Other small wins in your current config: Anisotropy 4 → 16 (free on 4090). Forest details 0.1 & clutter 0 are already performance-friendly. Clouds 0 saves GPU; if you raise clouds, offset with MSAA=OFF and DLSS Quality. Windows/driver/system hygiene Your log shows: Hypervisor is active and Process Lasso is running. Both can impact scheduling/timers. Do this: Disable Hyper-V / Virtualization-based Security (Core Isolation/Memory Integrity). This removes the “hypervisor is active” warning and often improves input latency/frametimes. dcs Process Lasso: you’ve constrained DCS’s affinity (process affinity mask: 11111000, unavailable cores {0,1,2}). On a Ryzen 7 9800X3D (single-CCD X3D) you generally want all 8 cores available to MT DCS. Let Windows/Game Mode handle scheduling; in Process Lasso, clear any forced affinities/priorities for DCS.exe. dcs Timer resolution shows 15.6 ms at start; DCS/driver usually pull it down when rendering, but if you run apps that fight timer resolution, you’ll get choppier frametimes. Keep background apps minimal (RGB suites, overlays, browser tabs, Discord hardware accel off, etc.). dcs NVIDIA driver: you’re on 571.96; that’s fine for DLSS. In NVCP set: Power management: Prefer maximum performance (per-app for DCS) Low-Latency Mode: Off (let DCS manage) G-Sync/Freesync: as your monitor supports; then uncap maxFPS or cap via RTSS. 9) Reduce services you don’t use You’re loading LeapMotion, CaptoGlove, SensoryxVRFree, VoiceChat services every boot. If you don’t use them, disable those services via module manager/options to shave startup time and a bit of memory. dcs 10) After the cleanup, re-test and re-enable selectively Re-add mods only via OVGME, one pack at a time, checking the log after each. Prefer actively maintained packs that don’t “replace” ED units. Keep a “vanilla” profile for multiplayer/“pure” servers. TL;DR order of operations Move out the listed mods (VAICOMPRO, J-11_Au2th, HighDigitSAMs, AM2, SAM packs, old Hawk/MB-339, etc.). dcs cleanup + repair slow. Delete fxo + metashaders2. Fix graphics: either DLAA native or DLSS Quality, not both; MSAA = OFF in both cases. Set Scaling = 1.0. dcs Disable Hyper-V/VBS; remove Process Lasso affinities so DCS can use all cores. dcs Optional: prune audio devices & unused services. dcs If you want, upload a fresh dcs.log after these changes and I’ll spot-check what’s left and fine-tune graphics for your exact modules/maps and whether you fly 2D or VR.
  2. Ready
    Good you ask. tbh the results are from what I remember. I just checked my change log in my thread on this forum for the numbers (you can find the link in my sig), I found the MSI-util tests I ran back in december before I did a complete reinstall, but... I see the results themselves were not written down in the log. I think I also ran some tests with the MSI change after my reinstall a couple of months later, but those tests are no mentioned in the screenshots of my log. I see some files dissapeared from the thread, only filenames remain. When I am back at my rig I will see if I can pull them straight from my original excel. With regards to how I measured; I have used various tools and methods for the testing, mostly focusing on frametimes, latency and reducing those irritating microstutters that sometimes tend to appear out of nowhere. In my thread you can read about most of the ups and downs with the various tools. iirc I was focusing on input latency towards the end of the thread.
  3. Ready
    Yup I have had good results with it too.
  4. Ready
    Interesting! Might give it a spin after returning from vacation.
  5. Ready
    Yep that's weird, never saw that before. Was this only after the latest update?
  6. Ready
    Nice!
  7. Ready
    For DCS in-game settings I found it to work the best by just turning everything down and one by one check what each setting does for your performance and your personal visual experience when you play around with it. Imho some things are just too taxing for very limited gain and some I can even hardly spot. When I am in the middle of the action I also don't really look at how nice reflections are etc. It is still immersive when everything is at low because I get used to it. But I do get irritated by trees popping up, unclear instruments, lags or microstutters and things like that, those things for me break the immersion. Some settings can be played with while in the game. For example the sliders for trees etc. I turn them down when I know we will fly low in green fjords. Best to try each change for yourself. Just run a test before/after and also see what your experience is within the headset. I sometimes had good numbers but strange things happening visually. I sometimes tried a couple of settings at the same time, trying to save some time, but to be honest it does happen that I still have to go back and check them one by one because the effect was counterproductive and I was not sure which one was the cause. So I would recommend to just take them one at a time. The current experience is pretty stable and I still have quite a backlog of things to check and further work on. But I am not in a hurry and sometimes just like to fly as well
  8. Ready
    Yes I think our objectives have quite some overlap. I aim at getting the best performance out of my own hardware, which to me translates in a stable/stutter-free experience, which for me began when I first got my G2 and was fighting to get a decent picture in my headset. So I try to get as close as possible to 90fps without microstutters (1% lows), which I try to realize by stripping and silencing Windows (which is a pretty big chunk of it) and optimizing all the hardware, DCS and the additional software my rig is running. Your checklist is fine; it points to items for people to look at. Some of the things that work differently for me (and are sometimes contrary to popular belief or recommendations that are given here and there); game mode, full screen, HAGS, HPET, overclock on my CPU and GPU, limiting FPS. So my main point would be to check the recommendations, but not blindly follow everything and test things for yourself (if you have the time and interest to do so). Look at the numbers and the experience in the headset (sometimes for me they did not match). Write down the changes and their impact in a log to track and compare and use various tools to do so.
  9. Ready
    Nice helpful list! Some of the items work the other way around for me though. Recommend running baseline tests before and after each change and keep a log of all the changes with their impact on your performance. I have also spend months aiming to get the best out of my system. Did you take a look at my list of changes? There might be some items in there you might like to add.
  10. Ready
    Nice! Have you considered/looked into CapFrameX? I used to perform the Excel calculations before as well using various tools to record, export and analyze. I found that https://www.capframex.com/ is way easier to quickly get to useable visual data and to quickly compare settings. No more exporting, importing, manual stuff, no more Excel wizardry needed. Just run the game and use hotkeys to start/end the log recordings, add titles/descriptions to the recording tracks and voila. I think it will save you a lot of time going forward and I assume it will have most of what you are looking for. I have used it 500+ times in my quest to silence Windows, stabilize my rig and get the most performance out of the hardware, testing all setttings and scenario's I could imagine and stuff I found online, I still use it to test the effects of updates or setting changes. You can find some decent examples of this tool in the postings of my thread (don't want to self plug). I also use XRFrameTools, but mostly to get realtime visualization of stability and get direct feedback of things I change on the fly. I now only use Excel to keep track of all the changes I make (either by for example DCS updates, driver updates or me tweaking something); with each change I create a log entry and log the frametimes etc. This ends up in one chart to show historically how my frametimes are improving or degrading with changes. It also is great to lookup stuff I have tested. Hope it is of any help!
  11. Ready
    Thanks for sharing!
  12. Ready
    Walk back the update? I frequently also have updates messing things up. If it works don't touch it if you don't have to
  13. Ready
    Hi I try to keep the first posting in this thread updated, I also added graphs where you can see the items that gave my system the biggest improvements. Of course your miles may vary. I did not use Atlas or any other preconfigured/debloated install because I wanted to decide for myself which functionality I disable and I wanted to stay in control. The end-result may be similar, I am not sure. I have seen some videos on comparing installs like Atlas and tbh was not regretting I went the long route. Whenever there are new thoughts or things to try I will add them to this thread and once in a while I will update the original posting with a list of my most recent changes.
  14. Ready
    Thanks I will check it out. Maybe it will be the solution to a weird issue I see in latencymon.
  15. Ready
    Nice one! Looking forward to the pilot body for a long time! Would be great to get additional visual improvements from within the helmet like moveable visor, the overlay of the visor and maybe some additional symbology.
  16. Ready
    I killed the process and the speed stayed the same, but haven't restarted yet. Not sure yet how it behaves.
  17. Ready
    Mouse Polling Rate Checked the polling rate of my mouse and thought it was still set to the lowest when I configured it last December. But somehow it was restored to its original 1.000Hz. Readjusted it to the lowest value. Logitech G Hub needs to be removed again, but will have to see what that does to the polling rate. https://www.clickspeedtester.com/mouse-polling-rate-checker/ Thanks @GPatricks for sharing the link!
  18. Ready
    Good it helped! With regards to what should be realistic; I once saw/heard a comment somewhere that IRL you look on top of the dash/UFC and should be able to see some screws/bolts or something like that. Can look it up but would take a bit of digging. I did not bother since I just move the chair to where I like it. I don't like bending forward and look up to peek into the hud, I much rather just have it lined up when I am sitting straight. Call me lazy
  19. Ready
    Set a Hotas binding to your seat elevator. I lift the seat up at startup and before take off just high enough so I can see the compass bar above the metal frame of the front dash. During flight I sometimes just adjust it if needed, but rarely have to adjust. This way I have the best possible view of the HUD. I VR recenter when needed.
  20. Ready
    Hi all is well Busy with some work stuff. Moving to a 9800X3D is a good move, you can never have enough cpu power in the current period. If you have to update your system board as well make sure you have good compatibility so you can do some little overclocking to get the most out of your system.
  21. Ready
    I don't see my rig getting the same improvements as in this video, will hold out for more data.
  22. Ready
    This is how I did it with (with W10). It really took me a long time to get the new install fully working again with all the various apps and configurations to my liking. If needed I could step back into the old install, which I did a bunch of times. With the dual boot you can take your time.
  23. Ready
    I have installed a fresh W10 on another SSD and copied my DCS also to another SSD. The new setup is working fine and if needed I can still change back to the old setup by booting into it. All still works. One I am absolutley certain I have everything I need in my new install I will scrap the old, or maybe I will just keep it around in case of emergency.
  24. Ready
    OBS Adjustments Changed a bunch of OBS settings to further calm down its CPU/GPU consumption. Not sure if all of them have any effect, but the combined results are positive. Updated OBS to latest version Added OBS to Windows Graphics Settings | High Performance Excluded OBS from Windows Defender Changed OBS settings in Nvidia Control Panel Anti-Aliasing: Off Low Latency Mode: Ultra Power Management Mode: Prefer Maximum Performance Changed some settings for OBS in Process Lasso CPU Affinity: Core 0,1,2 (away from DCS) I/O Priority: Normal (might change it to High in the future) CPU Priority: Normal (might increase it in the future) OBS Encoding (have to check the visual quality after a couple of recordings) Video Encoder: changed from NVENC HEVC(H.265) > NVENC H.264 (less CPU consumption) Rate Control: Changed Bitrate from Variable > Constant Preset: P5 Slow (Good Quality) Tuning: Low Latency Multipass Mode: Single Pass Profile: High Look-Ahead: Disabled Adaptive Quantization: Enabled B-Frames: 2 Bitrate: reduced to 25.000Kbps (for 2k, in accordance with Youtube Guidelines) Preview when flying: Disabled Replay Buffer: Disabled Aligned Audio Bitrate of all audio tracks (320) Improved Microphone audio quality and volume by adding and adjusting some filters Gain Noise suppression 3-Band Equalizer Expander Compressor Limiter Improved Microphone quality Added a couple more VB-Audio Cables and added one as Monitoring Device from within OBS. This way all the Filters that have been applied to my microphone in OBS also apply to SRS and Discord, hopefully improving quality and volume of the mic. Winhance (new tool to Keep Windows debloated and optimized) Thanks @Reed for sharing. There are quite some interesting things in there. I am too scared to let this tool do all the registry editing for my W10 since it is mostly made for W11 and I don't know what it is doing exactly. The creator mentions that in W10 it has some issues. I like the idea of adding the scheduler to keep things out. It seems to be a great app, but is a bit rough still when you can only switch on a long list of settings or switch them all off. Would be great if inside the categories you can just select the ones you want, so you can test their impact in a more controlled manner. Now you have to enable all or none, which worries me a bit. So I put some of the things on my list to check out.
  25. Ready
    Thanks for pointing it out and explaining. Changing authorizations for the handful of DCS VR related things seem to have improved my performance. I don't yet know exactly what caused the improvement and I check and reduce the rights that have no effect. I was not aware of running risks when joining MP servers. I normally disable Defender when playing and enable it again after disconnecting. But if I understand correctly this could expose my pc to additional risk. Will look into it further.
×
×
  • Create New...