Althar93

The issue has already been discussed to some degree in the thread above : essentially there appears to be a problem with how the Aiming station input for moving the sight left/right & up/down works when using a Joystick axis/POV, leading to persistent velocity drift of the input. It is as though the logic for computing the velocity of the displacement does not reset when there no longer is any input from the Joystick (e.g. from a POV hat), resulting in a drift behaviour with a constant velocity, until the input (e.g. POV) is activated again. This behaviour has been observed in the guided missile training mission whereby the Hind is in a perfect hover & so any drift from the aircraft itself is a no factor. The issue above occurs when removing any bindings for the Aiming station up/down & left/right & have a sole axis/device for each of the actions. Setting a deadzone as suggested in the post above 'helps' making the issue less frequent but it does still occur & with the same persistence. This issue, which appears to be affecting a few people is making it impossible for me to be remotely proficient at being in the gunner's seat as I am unable to reliably aim at & target a stationary target even when my own aircraft is perfectly motionless.

While this may not be realistic, I really like the clicking sound you get in the Mi-8 when trimming the aircraft. This audio cue is particularly useful as the trimmer mode (which may be set to require a return to centre) may sometimes not register if the stick hasn't been brought back within the centering threshold. I personally frequently find myself pressing the trim button on my joystick then returning the stick to centre, only to realise I didn't fully depress the button to the point of activation & end up bobbing around until I can gain control agian. It would be really nice if we could have a similar audio cue for the Mi-24. If there is one already, then please can we have it louder because I don't recall hearing anything.

I tried it with the curves maxed out to 100, the drift still occurs.

I have tried using the whole stick & the drifting still occurs (I have a VKB Gunfighter MKIII / MCG).

I have tried setting the curvature to the maximum value & it does help, but the drift does still occur. To me this is definitely a bug / oversight in the implementation of the aiming sight motion/input.

Worth noting I am seeing this issue in the Guided missile training mission where the Hind is in a 'perfect' hover so there should be no movement whatsoever to compensate for.

I am having the exact same issue - I deleted all bindings in the Aiming station page, and the only two inputs mapped are Aiming Station Left/Right (Joystick) and Aiming Station Up/Down (Joystick) to a POV on my stick and the drift still happens. Setting a deadzone does absolutely nothing to solve the issue (I tried a deadzone of '90'). It looks like the motion of the aiming device is based off the velocity of the input, rather than the axis value ; this would make sense to have it working that way when using a mouse as you would calculate the speed of motion based on the displacement of the mouse cursor, but not for a POV.

The cockpit shape most certainly has an impact on how reflective the glass appears, compared to a flatter canopy. I do appreciate the scratches, grime & reflections but unlike in the real world where we have a much larger field of view & can focus past scratches which are a few millimeters in width, here we are faced with scratches which are several pixels wide on a screen that is a fraction of what a real pilot would see. There has to be a balance between realism, and having something playable for a sim/game. If anything the glass will tend to look 'milky' with age as more and more microscratches accumulate & the glass effectively becomes rougher and as a result the reflection less defined - you can see in this 360 degree view of the cockpit (not the bis, but close enough) : https://www.gawronsky.com/panorama-3-MiG-21MF.html

I thought the issue had gone away, it hasn't - my performance in DCS is much better than it ever was but under the right (wrong?) condtions, the radar on the Mig21 definitely still takes a big chunk out of the frametime. I have been experiencing rather large stalls/performance hits in multiplayer games, with the framerate struggling to be interactive, only to jump all the way back up to 60FPS+ as soon as I turn the radar off.

Since my original post I have updated my drivers to 466.27 as well as completely reviewed my graphical settings. I did a bit more testing of 2.7 with other modules and noticed my framerate was also rather inconsistent (not to the same degree but it did used to be pretty stable prior to 2.7) and came to the conclusion this was due to my vsync settings (which worked just fine prior to 2.7). Long story short, I went through my graphical settings from scratch, disabled vsync in game and enabled it from the Nvidia control panel as 'Fast' (which essentially is a non-capped vsync where frames in excess of the refresh rate are ditched which prevents any kind of tearing) and with my frame limit set to just above my refresh rate (in my case 61FPS for my 60Hz ultrawide). If I have my frame limit set to my refresh rate or below (like you would if you when running with Gsync enabled, which was also causing issues for me) then performance is all over the place for me. Ever since I have had very stable performance & can no longer measure any fluctuations in performance from using the radar in the Mig21. For whoever else is experiencing issues, perhaps try disabling or tinkering with your vsync settings. Without knowning anything about the specifics of how DCS does its rendering or the radar in the Mig for that matter all I can think of is a stall can occasionally occur as a result of having the radar enabled, long enough to throw the frame pacing off which can become a problem when 'strict' vsync is enabled.

Nevermind, this would appear to be an issue with server I have been playing on over the last few days.

You are correct, I was thinking of the category where the binding lives. Specifically, I have mapped 'Drop Countermeasures (ASO and SPS-141)', for whic you have to lift the little red flap under the throttle handle and press the button to drop the counter measures. I can either click or press my joystick binding and can see the button in the cockpit being depressed but nothing happens. Worth noting there is another 'SPRD / or ASO drop chaff flares' to complicate matters.

I had no issues with the Mig21 or using counter measures (SPRD) a few weeks ago. I stopped playing for a few weeks until 2.7 came out and now I am unable to release any countermeasures in multiplayer games. I have my loadout full with Flares, the cover for the SPRD is lifted & I can physically see the button being depressed when I press the corresponding binding but no countermeasures are being deployed. I see there is an old post from 2017 around here that covers the same issue. Is anyone else experiencing the same issue as I am, or am I alone. I must admit it does make me almost completely defenseless in multiplayer if I didn't have the upper hand to begin with.

Performance seems to be pretty good for me overall, minus the occasional stutter which is a bit strange. I get 50-60FPS (locked) most of the time but under certain circumstances the frame rate will chug ; if I pause the game, the frame rate picks up straight away back to my locked 60FPS. What I can say though is that the RAM usage is through the roof and has been causing crashes every two flights or so ; I was able to reduce the frequency of crashes by reducing my preload radius somewhat but there is definitely something odd going on there (guessing a leak, but even then RAM usage does seem quite high even considering what DCS simulates).

Joined Open range earlier - that fog/weather is brutal.

i7-4790 RTX2060 6GB with 16GB RAM here I would say that with identical settings, I am getting more or less the same framerate in 2.7 as I did previously, which is to say around 40-60FPS between the worst case scenario (multiplayer, on the ground in crowded areas) and best case scenario (single player, in the air). I wouldn't mind further optimisations as a higher framerate is always nice but overall the game / update runs pretty well given all of the visual additions & clouds.

It looks fine on other modules, I think it is just a side effect of the shape of the Mig21's canopy which results in much stronger fresnel/reflections simply down to the curvature of the glass. Let's hope ED allows 3rd-parties to adjust this and/or tone it down somewhat.

I agree that in some situations the inner reflections of the cockpit can be extremely strong, to the point where you can't actually see outside.

The fluctuation in performance can also be observed flying the 'Radar Operations' training mission, just set the SAU at the start of the flight & just watch as the frame time periodically takes a hit and the FPS drops before picking back up. On my system, I'll observe those micro-stutters/chugging every 10-20 seconds or so, dropping down to 25-40FPS (as I wrote above it is nowhere near as bad as it used to be but it is still rather distracting when it happens). On the other hand for the exact same section, if I leave the radar off my FPS is a rock solid 60FPS with no deviations in the frame time or noticeable stuttering.

Been doing some more testing, it seems to occur when I am close to the ground / forest in the Freeflight mission.

The drop isn't half as bad as it used to be but it is there (dropping from 60FPS to ~25FPS). My repro is to simply run the Freeflight instant action mission in Caucasus. I am sporting an RTX 2060 with the 465.89 Nvidia drivers (I just noticed there is a brand new driver out today so I'll try that). I initially thought the clouds could be the culprit, but I can toggle the radar on/off and the framerate picks up straight away as a result.

I have just updated to the latest DCS 2.7 and been trying out the Mig21-bis. With the improved visuals & lighting the Mig's cockpit looks amazing. One small caveat. The performance issue that was previously plaguing the Radar seems to be back, with the same behaviour or chugging for a few frames, before picking up again. Am I the only once with this issue?

Thanks a lot for taking our feedback onboard & I look forward to 2.7! After reading my post from a few weeks ago again, I realise it might have been a bit brash/unfair in places. I just wanted to clarify that my intention was not to be insulting or discouraging & want to reiterate that you have done a stellar job with making the most immersive & engaging DCS campaign that I've played so far.

Nevermind, I had a closer look at the track and.... J-P Lambert committed unintentional suicide. Essentially just after releasing the weapons, rather than fly away he decided to make a dive and actually got under his own bombs. I thought the whole point of CCRP was that you could release the weapons at a safe altitude/pitch & you didn't have to get close to your target. In case the images weren't clear, his own bomb hit him in the rear & I guess that must have caused enough damage that he was unable to recover later during the CAS run. Oh well... DCS AI I guess...

I have been trying to play through mission 06 of the campaign and realised after the arrival of the Hinds and switching back to the main frequency that my character was speaking by himself with no one replying on the other end. Suspecting the mission broke I exited and reviewed the mission in TacView only to realise that my lead (J-P Lambert) crashed in the ground when turning back from his successful CAS run on the objective. I have attached the TacView, J-P Lambert will meet his demise at +00:38:53. I am guessing this is going to be specific to the AI in DCS, but perhaps there is something that can be done about this? I will re-attempt the mission again at some point and hopefully it does not occur again after nearly 40 minutes in flight. Cheers, M2000_Campaign_Mission_06_Glitch.acmi