darkman222

I have a small replay from a server I frequently play on, where the CPU frame time was really bad, right when I joined the server. Around 16 - 17 ms CPU frame time. When I replay it in VR and press the VR spyglass Zoom , for that time being zoomed the CPU frame time improves significantly. See the video with fpsVR on the left. Same behavior in singlethreading too. But with much worse frame times of course. Is that behavior worth noticing? Or is it just what will be expected using VR zoom? Can that be used to optimized something in DCS that helps me get better CPU frame times? Video and track file attached. System Specs below, using VR with steamVR and Varjo Aero. dsa mobettametas_Dogfight_Arena_v1.73.1-20230528-223449 16 ms permanent.trk

Thanks a lot. That worked perfect. Its a shame we have to trick DCS to look onto another drive than just having the "folder browser" checkbox in every file dialogue. I used that link: https://www.howtogeek.com/16226/complete-guide-to-symbolic-links-symlinks-on-windows-or-linux/ And used the Graphal Tool , Link shell exension. (Just skip to this part) https://schinagl.priv.at/nt/hardlinkshellext/linkshellextension.html Then pick the folder on drive D as source and paste it as symbolic link in the user/missions directory. ( re means right klick)

I mean I know I could that. But it'll probably move the saved game folder for other games that I dont want to have it on D. This seems like an over complicated action for just accessing replays and missions (I dont want to edit but just play) from D. If the checkbox "folder browser" was on every DCS file dialogue there would not be any issue here. Or even if I could edit that the default drive DCS has to look into is D instead of C.

I move the replays I want to keep in a synced drive on drive D : Also my missions are created on that drive (and saved via sync) The mission editor gives the option to switch to drive D. Of course I can start missions loading them into the ME. But for replays it does not work. And for just playing missions its not convenient. Is there any way to switch the mission load and the replay load menu to drive D :

Are you checking whats bottlenecking your system? Try to check CPU frame times. With the integrated performance tool from ED (RightCRL + Pause) or with an external one. Do that before you update to W11. I am looking forward to see the results. The main thing with W10 vs W11 is the rumor that W11 can use the newer generation of CPUs better. I did not get any hard proof that made me seriously switch to W11.... yet.

Dont know if that would help much. Like I said I just put the task on the Aircraft and he just does always the same EDIT: New try. ******************************************************************************************* Deleted everyhing. Then new F16 and fist I had this on the Aircraft itself, and deleted it as second try: But the waypoint action in both cases was: And this time the result was, that he was just flying high AO like this. But no 4G turn... null I attached the MIZ if someone is more creative than me. But I guess, that simple task is not as simple as it sounds. ai circle v01.miz

Today with DCS 2.8.5.40170 Open Beta - 18.05.2023 I made a comparsion between versions. Same mission, just a single player F16 over Batumi. To sum it up 2.7 : CPU 7.8 ms , GPU 8.1 ms 2.8.5 with Multithreading: CPU 4.6 ms , GPU 7.5 ms Good job ED! It was woth the wait. See the full test in the screenshot below with system specs. Running a Varjo Aero @ very high setting 37 PPD Lets call the issue resolved

3.2.1. and the thread is locked Seems like the F16 FM will be undergoing minor tweaks. But for the F18's FM a major overhaul was announced. I think about a year ago. I am wondering when and if that thing will stop wobbling around as it does since relase.

45 fps and not motion reprojection aka motion smoothing? How can you have a smooth ride then?!

Removed the Cap Task on the waypoint did not help. Deleting the waypoint and putting the aerobatic task directly on the aircraft without waypoint did not help. He does always the same. climb to 6200 ft and start the turn with 220 kts, 2,3 G

I want an AI aircraft to perform an aerobatics manover as waypoint action. Alt 2000 ft, 450 kts, 4 G turn. Looks like the screenshot in the mission editor: But was he does instead is is climb to 6200 ft and start the turn with 220 kts, 2,3 G Do I miss anything? Do I have to initialize something?! I attached the MIZ below. Any help appreciated. ai circle v01.miz

I really cant read through all that stuff any more. We dont need any pseudo scientific discussion here. Threre is given amount of time an F16 pilot has to withstand 9G. Otherwise he is not qualified to fly the F16, and is not in the cockpit any more. What is so complicated about the approach that if the DCS pilot cant stand the amount of time, a real pilot has to, then something is off. Take the DCS F16, make a g warmup, then speed up to 600 kts pull 9 Gs count the time until blackout. Take the measured time to an SME and ask if the DCS pilot would have been qualified for the F16. No need to prove that you can sustain 450 kts while the pilot is on the edge of conciousness, when the slightest tighter pull is about to knock him out.

As we had the discussion 100 times here. My 50 cents. As short as possible. I understand what EM diagrams are for and that we need numbers to model the flight model after. But can you actually read out the energy bleed rate from them ? How is this being modelled? Yes, its a question. I am just a computer graphics artist and flight sim enthusiast. What I understand is that sustainable and instantaneous turn rate come from a combination of different values from the FM. But different combination of values can lead to the same turn rate performance. But everything else might be off. Also energy bleed behavior. The mistrust people have is because there was a fix for the energy bleed behavior some time ago. And it lead to a difference of about 100 kts or more in a 180 degree turn. Which an undeniable big difference. Also that people found that Yaw and Pitch inertia values are swapped ( separate thread) does not build trust in the DCS F16 flight modelling. People will always do investigations. If the bleed rate was still of and just hypothetical, if it would be improved, that would lead to an even more obvious problem with the g modelling. Which leads to my second 50 cents. Of course you cant measure human performance like aircraft performance. But if ( I dont have a proof for the F16 but I know for the german Eurofighter) the pilot needs to be able to sustain 15 seconds of 9G to be qualified to operate it, then why cant the virtual F16 pilot do so? It can not be stressed enough that all of the real F16 flight performance is built around the 9 G limit. Its more cruicial than on any other DCS aircraft. The F18 vs F16 discussion is irrelevant, as long as the F18 in game is not 100% perfect FM wise. Whicht the FM of the F16 and the F18 never will. Then its a separate discussion for the F18 sub forums.

1: Deadzone would be needed to be huge. As depressing the cursor enable switch youre kinda sliding around with the thumb, accidentally moving it a lot. 2: I need all the extra switches on the lower side of the grip, and it feels unnatural to have the cursor enable switch sitting there. I use the buttons on the lower side to VR zoom. And its hard to hold zoom and cursor enable the same time this way. But I often use it to slave the cursor on the bandit, while using the vr zoom to identify the bandit and better see what he is doing before I fire the missile at him. 3: I tried to use it as a modifier in the DCS button mapping. So the depress disables the axis. But what it did was, when I was pressing it not perfectly without any axis input, although the axis got disabled, the minimal input was kept, and the radar cursor axis kept sending the last amount of input while it was depressed. Also, the problem is I could not bind it as a command when I assigned the modifier function to it. 4: I asked Winwing if Simapp offers the possibility to deactivate an axis via a button. Negative.

Even worse. The paddle is often used to prevent speeding over the corner speed, resulting in converting execess speed to turn rate with 9G. Wihtout the paddle, proper energery management would have been needed. I know what I say, I fly the F16 and the F18 for fun. You pick that described habit up in no time, trust me. And yeah, put me in an F18 and I become one of that paddle guys instantly, i confess. There was one folds of honor tournament with respecting G limits mandatory. It was very complicated to monitor and a lot of discussions arise afterwards. Its worth a look on youtube. In reality an overG-ed aircraft is off service for couple of months worst case. So if your tactics in an F14 would make it necessary to pull 10G to win the fight against a hostile, you just would not go into that fight and bug out instad.

You wont be damaging your stores, but if enough G applied the wings rip off. But agreed. Over-G-ing and damaging stores is a severe thing IRL and should be implemented. But not only to the F16 but to all the other airframes too. The next big updates for me, that will bring the F16 closer to realism will be: (hopefully soon) implementig a pilot that is acutally qualified to fly a 9 G jet, and the DTC, data cartrige, that already can be used, more or less comfortable with a 3rd party external program. But having it in game will be much easier to use.

Not sure if your issue is related. But here is a thread I started about performance degradation over time: And also there is a thread that identified the issue to come from craters that never disappear and add on CPU frame time. This issue is claimed to be resoved in the next update. Lets see if it fixes your and my issues too:

Before this goes too much off topic. Although the thread says the issue is reported, its two different things. First is that a pilot in flight is not already g warmed up. Thats a minor thing. Can be fixed by a couple of loaded barrel rolls. And it is reported. Second is that the current g modelling of the pilot not withstanding long enough to meet the demands to be qualified for a 9G jet like the F16, which is about to be revised, but with no timeline on the list. My opinion is that it will be about time to look into that while the DCS F16 is being pushed out of early access.

When you spawn, your pilot is not g warmed up. Thats why he blacks out instantly. Like I said. There are procedures. But you only have to be at a G higher than 4 or 5 G. So a couple of barrel rolls do the trick and you dont deviate too much from the course to the bandit if youre in a tournament or dogfight server.

Thanks for letting us know that this is on the list for future updates. I would not expect the seat position to play such a big role as it also needed due to the small cockpit measurements of the F16. (Little selling trick by General Dynamics here) I would rather expect that the DCS pilot will be adjusted to meet the real world requirements for a qualified F16 pilot. With that real F16 techniques will become possible in DCS. Like rating at the best sustained rate and pull the bandit in HUD, which is difficult at the moment because the DCS pilot blacks out in the final pull.

Not necessarily. When you spawned in you did not have a G warmup. If you do , the behavior is the same all the time. But still, you wont be able to use the full potential of a modern fighter. The pilot blacks out after a few seconds even loses vision at 7.5 G sometimes.

I think the only difference DCS makes for fighter jets is if the pilot has a G suit or not. So the F86 pilot just goes boom and blackout, whitout g suit, but the modelling for the F16, F18, F14 etc should be the same as their pilots have G suits. Whats also simulated is a g warmup. There are techniques but the easiest is to do some barrel rolls with more than 7 G. If you dont do it, you also go black out quickly. And the paddle pulling F18 is also not relevant here. Because there are no EM diagrams for the paddle pulled state as far as I know. And its not a common practice to do so. If people need the paddle its for compensating the lack of BFM skills. The F18 paddle is not being used in a dogfighting reality. But the F16 is built around and for the 9G limit, thats why its called a true 9 G fighter. And it makes no sense that the virtual pilot cant handle 9Gs long enough for the F16 to be in the regime where its built to perform. Sure the flight model is correct. But with a pilot who is unqualified for this jet it is pointless.

Agreed. The DCS F16 flight model might be declared as correct-as-is a hundred times more if the pilot thats sitting in it is not qualified to fly it, he becomes the limiting factor not the flight model. If the DCS F16 is about to be pushed out of early access this really needs to be looked at.

I know that this is done via Data Link. I guess the AI locks the ground target and sends it to DL for convenience of the human pilot. If you have a human wingman he has to do it manually.

You might need to try it with a human wingman. But if you use the targeting pod or the AG radar to lock something on the ground and then press comm switch inbound. ( Maybe long press) Not really sure if it works though. If it does let me know.