Jump to content

Biggus

Members
  • Posts

    839
  • Joined

  • Last visited

Everything posted by Biggus

  1. Some never seem to render at all. Those tend to be the ones that cause the freeze. Flickering is just one of the symptoms with a strong correlation to the unrendered units and client freezes.
  2. Suspiciously like the one I reported awhile ago. Often leads to a client freeze.
  3. Yup, I still have the 900ft AGL frame reduction. Out of curiousity, has anyone who has this problem tested Kola to see if the problem is also evident on that map? I'm desperate for a map that isn't mostly sand and rock, but I am not really willing to risk having this bug on two maps.
  4. In my testing after today's patch, the bug remains and neither changing LOD nor clearing fxo and metashaders fixes the issue.
  5. I'll test it. I do this every patch as a habit though. But at this point I'll do anything for a fix in the short term.
  6. I've just come across this myself trying to put it on my Moza and Virpil bases. It only further reinforces my intense dislike of this company.
  7. I've asked Moza repeatedly about the fabled Winwing adapter and I've been met with zero responses. It's apparently available in at least one market but I've not yet found it. I've become frustrated enough that instead, I'm selling my Winwing stick and I'm giving money to Thrustmaster for their Hornet stick and offset adapter. I'd rather have bought the Moza adapter, but it's not an option.
  8. Two nights (~4-5 hours of uptime in total) with the TTI mission for PG, there have been zero client crashes. I haven't had so few issues with DCS either as client or server in months. No evident weapons flicker, no evident aircraft flicker, no unrendered aircraft, no weirdly flying aircraft. Interesting as TTI includes MOOSE, AIRBOSS and the super carrier. I'll continue testing for the a few more days and possibly roll the mission over to another large one like Operation Snowfox. I suspect it's a DCS bug and not a Liberation/Retribution bug, although those generators might be really efficient at building missions that exhibit the issue.
  9. 95% of the missions my server runs are Retribution, but not every one of those missions exhibits the flickering as far as I'm aware. I'll load up a TTI mission this week and test it for a few days. One additional thing I've noticed is that client crashes tend to be after more than 30 minutes, most often in the 40-50 minute range following mission start. I wonder if late activated units spawning has something to do with it?
  10. @SilentOtto I'm going to echo @MAXsenna, it's incredibly cool. Thank you very much for sharing!
  11. Had another instance of this bug appear last night. Weapons flicker was evident, aircraft rendering issues on display in this clip: First part of the vid is not long after spawning in. Two AI F-4s are on the ramp awaiting their takeoff time which was roughly half an hour into the mission run time. After a short sortie, I returned to rearm and refuel and noticed that the AI that was glitching had stopped rendering entirely. It later took off with it's wingman on it's mission. I ended the mission shortly after observing that. It is possible that the glitching aircraft bug is separate to the weapons flicker, but in my experience if you have one of these issues appear in the mission, the other will also appear too at some point. Client log reported the following after checking with the discord log analyzer: I'm pretty sure I've seen the bug appear without this error appearing in the log, but it seems to crop up from time to time even after multiple repairs. Only mods on client are SRS and Tacview. Server side has those plus Olympus. Bug predates installation of both Tacview and Olympus, and is not evident in every mission. There is an error in the server log related to Olympus, however as mentioned this bug predates the Olympus install. I'll see if I can re-run the mission without Olympus later on to confirm, but as it's a Retribution mission and the state saving function can mess with things when replaying missions, I expect to see other errors there. Server track. Client track. Server and client have had their XMP profiles disabled. Server dcs.log dcs client with flicker.log retribution_nextturn.miz Edit: Re-ran the mission. Tacview was still enabled unfortunately, as I'm not quite sure how to disable it without uninstalling on a dedicated server. Same thing occurred with that same AI phantom, at which point I ended the mission so that I'd have a more manageable trackfile. You may not see the weapon flickering in this track, however if the mission had continued to run it would have appeared eventually. I just wanted to have a usable trackfile for the bug report. Attaching server log as 2nd dcs server log.log, client log as 2nd client dcs.log and client trackfile as retribution_nextturn-20241212-105016.trk. Server log throws an error relating to Retribution as predicted. I also noticed that the server pagefile was quite low in the original log and increased it to 32GB in the second round of logs. Didn't seem to help. 2nd dcs server log.log retribution_nextturn-20241212-105016.trk 2nd client dcs.log
  12. That's an interesting idea. Would you mind sharing your workflow to develop the profile?
  13. Looks like what I've reported here. Occasionally an aircraft will stop being rendered entirely, and it is likely to crash clients when that unrendered aircraft is engaged by enemy units.
  14. Pain. You could try a hardware work around on the arduinos by doing an RS485 network and then you'd cut down on your port numbers, but that's a fair bit of work if you've already got everything built. I'd still go through and look at your USB topography. Count your used endpoints for each controller and see whether that's part of the issue. Other than that, I'm stumped.
  15. That's annoying. Hopefully it's a strong one.
  16. Check the number of endpoints per controller, each device can use up quite a few and you've got something like 127 possible endpoints per USB2.0 controller, from memory. Just rereading your problem, @JohnnyChicago is it your ESP32 device ports that are giving you 'permission denied'? I ran into that a few weeks ago with mine and I initially got around it by running socat in a powershell.
  17. I have a vague recollection that the Uno and the Mega don't have that diode. I believe most of the pro micros do have it, though. @JohnnyChicago Could you be running out of USB endpoints? Might be worth having a look with USB Device Tree Viewer to see how many endpoints you've used.
  18. How's the wall thickness? My first one had walls around 4mm thick and I didn't trust it. Once I bulked up the walls to around 7mm thickness, my level of confidence rose pretty significantly.
  19. Haven't noticed because I'm running a dedicated server. Somehow tied into all of this is that sometimes aircraft appear to fly really weirdly. Like they're being slewed at odd attitudes. I'd assumed it was AI only, until this happened: I was on the server by myself. Had noticed some weapons flicker occurring but hadn't yet seen any aircraft disappearing or failing to render. I've attached the server track and log as well as my client log, and I can supply a client track if required. Mods had not yet been removed at the time of this happening, but I figured that what happened here was unusual enough that it might prove to be useful. dcs - client.log Server - bug track.trk dcs - server bug.log
  20. I'm running a 3D printed extension of roughly 200mm, not exactly a gooseneck like those alloy ones but offset rearward with a forward tilt. I took the design by Terry Dactyl in this thread, added 100mm to the bottom and made the internal walls quite thick. PETG, something like 60% infill with six walls. It's ridiculously strong and coping with a Warthog on top of it perfectly fine. Absolutely no concern with snapping. I think I'm going to replace it with a shorter one and raise the base a bit though, as I'm not a fan of the long pitch axis travel with my current setup.
  21. @BIGNEWYThanks for doing that so quickly. I'll see about grabbing a modless set of server and client logs for you over the next few days.
  22. Haven't noticed the flicker so far with today's patch, but two clients crashed tonight clicking on the same H-6J. @NineLine, I wonder if this thread might be best moved to the multiplayer bugs forum as it sounds like an MP-only issue? I suspect it's a dedicated server bug.
  23. As I've attempted to tweak the Phantom profile over the last week, I've become less certain about things like whether the Moza software is measuring AoA in degrees (it says it is, but I suspect it's arbitrary units) and what combination of spring, damper, friction and inertia will let me retain correct AFCS engagement behavior with a heavy stick on a 200mm extension, or how to limit travel in pitch without the software creating weird curves in the background. I think rather than having SMEs sharing profiles, we'd be better off with them sharing anecdotes about how they remember it and cross-checking that with data from studies on the control systems of the relevant aircraft. There are too many hardware combinations to be able to reliably cut and paste from one person to another. One thing that would be good would be better documentation on all the functions and how the sim responds to those functions.
  24. I've been seeing crashes related to unrendered aircraft in the time since I reported this. I've had unrendered Foxbats and Fitters, and when they are engaged, it generally leads to clients within a certain proximity to crash. Players all seeing the same thing though - unrendered aircraft. Crashes occur without generating any log entries to help. Game freezes, clicking mouse brings up the 'do you want to wait for the game to respond?' and I either click to end the process or I let it sit hung forever.
  25. My feelings about my AB9 are a bit complicated and probably sound more negative than I intend. I love the hardware. The software situation lets it down in terms of both minor bugs and the failure to comply with the terms of the license for what they've taken from Walmis. The activity of the software devs on the Moza discord went from constant discussion before the discovery of the appropriated code to zero immediately after. If I had the time available to wait or to build, sure, I'd have grabbed one of those other options. The Moza was available immediately, didn't require me to put it together and was a reasonable price. If the Winwing FFB had been available when I ordered, I probably would have had a harder time deciding. With all that said though, I have a pretty big smile on my face every time I fly now. It's been quite transformative.
×
×
  • Create New...