darkman222

Well let me put it that way. You can do it as shown. If you have the time, the fuel and if the bandit has no 9Xs or wingmen in the air.

As the tile says. I am considering upgrading from a Ryzen 9 5900x (not overclocked) to a i9-12900KF (no overclocking intentions). The problem is that my CPU frame times in Multiplayer are very high. Between 20 ms or 17 ms. Creating stutters even in 60 Hz mode for my Pimax 8KX, which is a big immersion breaker. This is how it looks in fpsVR. So if I compare Passmark tests and benchmarks between the the 5900 and the new i9-12900KF I should expect an improvement of about 18 % which is a lot. See the passmark results for single core performance here: But I also watched some in-game benchmarks users did on youtube. Nobody did a VR benchmark though. But it turns out the higher the output resolution in game, the less the difference between the processors cpu frame times. The Intel was only significantly faster in 1080 but not in 4k. Which is much closer to the Pimax resolution. Unfortunately nobody ran tests in DCS. If you can compare that directly to VR it would not make sense to purchase the Intel processor. But if not, in my case for CPU frame time 18 % more performance would result in a frame time from 20ms to 16 ms. Which would be a good improvement. Any predictions or ideas?

Most of the people who were fighting here went to the DCS dogfighters server. Just search for "dogfight" in MP menu. But beware. People are really hardcore here.

...well since this thread was moved to "weapons bugs" I assume it was not realistic that the AIM120 was still keep tracking me after I reappeard behind the mountain. Although that does not answer the question how long is long enough to break an AIM120 lock, it gives me some hint to the answer to my question.

Wow. Back to 271 ? I did not expect performance impacts on MP gaming in particular. But it seems so. I'll give it a try reverting back to such an old version. Thanks for the heads up. EDIT: Went back to 272. I feel that it improved performance quite a bit. I also tried 271. But I think that was before the 8kx was out, so it does not recognize it.

I installed the new Pitool V274 Beta as suggested. Mostly for testing the new Gemini hand tracker. Not talking about hand tracking here. But it feels like 274 is giving a worse performance than 273. I am constantly switching betwen 273 and 274. And I feel expecially on MP servers the performance drops and stutters are noticably worse using 274. Does someone experience the same issues?!

I had this encounter you can watch in the video. You can see the view of the Aim 120... An Aim 120 was fired on a MP server on me. I was terrain masking for 3 seconds. But the Aim 120 seeked just picked me up again, or it never lost me. Although I know that 3 seconds terrain masking is very short and the seeker might do some interpolations like the radar of an aircraft like the F16 does, when it loses sight to the target. But how long is sufficient to break the lock? And, more important. If I terrain mask, but reappear in the sight of the Aim 120 seeker, will it re-aquire me? Will it go back to a maddog state? Which means losing lock, going back to mad dog, and aquire the next visible target again? Or will the AIM 120 only go mad dog once and should not re aquire any targets after it lost the initial lock? In case someone is interested. Here is the (very long) track file: https://www.dropbox.com/s/2v7f7398s58yrjt/mobettameta's Dogfight Arena v1.29-20220106-000045 amraam terrain mask.trk?dl=0 I disconnected right after that. So its at the very end of the track file.

Thanks for letting me know. I really dont understand why ED just does not fix the aiming reticle. I understand CA is not made for VR. I dont need all the information about ammo etc. I just need a working reticle to have some fun in VR and shoot stuff. Thats all.

I cant play until next year. So I cant test the update. But no VR fixes have been announced anyway. The major issue was the aiming overlays if you also want to play in a vehicle. Just try to lock something up (enter key) and see if the aiming guide , the thing that shows the necessary lead actually does show lead in first person and in third person view. If the aiming cross is still glued to your heads movement, it has not been fixed (yet).

Is DTOS fully functional now? I did not see it mentioned in the updates. I cant play DCS until next year. So I am not able to test it.

Thanks for the heads up. Will I the Pi Tool already include the new Gemini driver update for hand tracking which is now supported in the newest DCS version? Or do I have to update something manually? DCS 2.7.9.17830 Open Beta - 23.12.2021 (->Leap motion gemini 5.2 drivers. Users may need to update their leap motion drivers to the latest gemini drivers.)

What impressed me pretty much in my little test here is that an almost empty (DCS) F16 has at 340 kts a TRT of 20 deg/sec. If the FM was claimed to be complete, I would do some more testing but until then its pointless just to have temporary values to work with in future dogfights.

Thanks for the detailed answer @Frederf But I needed something easier to wrap my head around. Sloppy test from a simple gamer I know. But what I did here was to pick a speed the F16 could maintain, while circling around with full, and then almost empty tanks. I should have done this closer to corner speed, but that would have been much harder to compare when the F16 cant maintain speed long enough, or if G forces cant slow the F16 down long enough, for valid results. But I guess it illustrates what I was wondering about. F16 turning circles with 340 kts: 1300 lbs fuel -> 6,3 G -> sustained TRT 20 deg / second 7100 lbs fuel -> 5,2G -> sustained TRT 16 deg / second But now the same 7100 lbs heavy aircraft and higher speeds. F16 turning circles with 400 kts: 7100 lbs fuel -> 6,3 G -> sustained TRT 17 deg / second The heavier the aircraft is, the lower the turnrate gets at the same speed. But if I fly the 7100 lbs heavy F16 at 400 kts instead of 340 kts, now I am gaining 1 deg / sec in sustained TRT This is what I was expecting. A heavier aircraft needs higher speeds than a light aircraft to get a "better" turn rates (but just a little) . But that also proves what you are saying that the speed does not matter too much. 1 degree per second can make a difference, but not a significant one. Disclaimer: Please dont start a how realistic is the DCS F16 turn rate. We have been there, this is not my interest in that thread. I just want to gain understanding how speed, weight and TRT are related to each other.

Like I said, it does not make sense being labelled "correct as is" because JHCMS might be correct as is, but its a radar issue. You can try the same without JHCMS with the HUD radar boresight cross with the same results. Both radar issues have status: "investigating" or "will be tweaked". Just wait until the mods come online to clarify.

I think the tag "correct as is" is absolutely misleading. You encounter a combination of issues ED is aware of for a long time, but did not fix them (yet). They only seem to adress only the most pressing issues, while developing the rest of F16 module further. Maybe adressing these issues will happen in a second development pass. I dont know... Its not a JHCMS issue, its more of a radar issue. The combination of the issues you are running into (that are announced to be fixed) are: -Radar look down ( will be tweaked) -Radar locking the wrong target, if mutiple are close together (Investigating) Get ready for that this thread also being merged with one of those The corresponding threads are:

Thanks Krippz, but I am not testing I am just trying to understand. And @Frederf What you are saying is, no matter if my tanks are full or almost empty (at Bingo fuel). The best turn rate will always be around 450 kts CAS? And the feeling that the F16 really starts to perform incredibly well below 2000 lbs fuel is just because it can hold 450 kts for a longer amount of time? Makes sense to me now. So for the sake of a simple answer to my question: What I need to do is stick with 450 kts ( at 5000 ft) for max turn rate and just dont care about fuel state, but just care about how long I can hold 450 kts due to fuel state. Correct?

A question which sounds simple but I am not really sure about it. There are a lot of diagrams out, but when I am fighting on a dogfight server with a clean F16 I start with 5000 lbs of fuel, burn down to bingo, and that F16 starts to become a turn rate monster below 2000 lbs of fuel. Are there diagrams related to aircraft weight, or if its not allowed to post, what is the advice for best turning speeds starting off with almost full tanks: 5000 lbs -> 450 kts ? And when approaching Bingo below 2000 lbs -> 420 kts? Is that in the ball park?! All speeds close to sea level 5000 ft altitude, where most of the dogfights end up at. Talking about CAS speeds I think, right? Any advice?

I know. Let me come up with a better testing solution. I cant see your system specs in the signature. Can you give the processor, GPU and headset?

Hmm no idea. You need the F16 and thats it. As I am sure high CPU time has nothing to do with MP in my case, let me see and find and, or build a SP mission. Just one minute in lenght needed, looking forward with the headset to eliminate every factor that might lead to false results.

7.5 ms, no spikes. Emtpy mission in Caucasus. @dureikenlike I said, your track starts on taxiway with horrible 40ms for me. You can try my track instead. Its much shorter, starts in mid air and you can watch the video I posted for easier reference, when the CPU time goes up.

Wow. In the future I will save tons of money. I'll spend my holidays on a yacht in DCS This is just awesome.

I tried your file. CPU time on the ground does not make too much sense to compare in my opinion. It its always very high. I did not even play through your file as I already had like 40 ms on the taxiway and I got some script errors right on the start. You can try this file here. https://www.dropbox.com/s/rzarmvnuz6ld4j4/mobettameta's Dogfight Arena v1.26-20211215-230047CPUframetime.trk?dl=0 Its from the dogfight MP server. Very short replay. CPU time goes up before I shoot the F18 which is very close. That is the live CPU frame time in the fpsVR graph. The second one after the short black screen is from the track replay when a distant ground of 3 bandits appear. Also like 18ms CPU time. Again Ryzen9 5900x and RTX 3090 on my end. See the video:

I tried a lot... Reinstalling, CPU overclock GPU overclock. Reinstall again etc. VR shader mods. Installed 1000W PSU. Never used a Wifi connection. I have the fastest connection available where I live. Rember, I am on a Pimax 8KX which is a not really common headset. And my investigations found that even if I lower the SS resolution to 0.5 I still have high CPU frame times. It seems that the fact that as soon as DCS 2.7 has to render (or the CPU has to prepare images) for both eyes CPU frame time drops A LOT. I can run DCS MP in 2D 3440×1440 resolution with ridiculus low CPU times like 5 ms, often less. Also depends on how the MP mission is setup. There is a non updating MP dogfightserver called "just dogfight" without clouds, maybe old scripts which are not so sophisticated but CPU friendlier. That server often is playable until more than 3 players show up simultaneously. There is a newer server called "DCS dogfighters" with clouds, newer scripts for sure. No matter how little players are online its just a stutter show for me. CPU frame times go up to 16 ms - 20 ms ....

I gave up. In fact its a known thing in the game industry. As hardware gets more capable the software and games get more hardware demanding to deliver a better experience and visuals. Same for DCS. My hardware was on the edge for the Pimax 8KX. Development goes further to 2.7 and my hardware cant keep up any more. I hope that EDs multiprocessing optimisations and Vulkan API will do the trick at some time for me. If not, well then I need to wait for better hardware and upgrade the rig whenever hardware development makes the next huge leap and it makes sense to purchase better hardware.

The download for the setup file does not exist any more? "SimShaker software family SimShaker for Aviators by f4l0 " This item might not exist or is no longer available The One Drive is empty?! Is that correct?