rossmum

Would be nice to see this looked at and it's good to see a fairly detailed and grounded look into it instead of having 1-2 relevant posts lost in a sea of unrelated complaints or complaints around other aspects of the FM (incorrect or not). It's very easy to build bad habits with things as they are and I'd be curious to see what the effect is on some other things where the numbers don't quite line up (takeoff run distances for example).

I have yet to see a single case of someone complaining about the 21's landing characteristics, where they were finally persuaded to provide a trackfile, only for it not to become immediately apparent that the problem was shockingly bad technique (usually because they were "inspired" by another player's genius "lifehack"). Of all the problems the FM has, landing characteristics are not one of them, and takeoff issues mainly come down to how much the jet rocks on its suspension combined with too lengthy of a takeoff run compared to the real jet. Almost all bad takeoffs I've seen (as well as my own messy ones) are from holding the stick back and lifting too early, versus holding the stick back until the nose attains the correct attitude and then holding the nose there. You can make comparisons to the Spitfire's cooling all you want, it doesn't change the facts of the matter: the only reason people complain about landing the 21 in DCS comes down to terrible technique every single time.

...But it's not unstable at 10 degrees indicated AoA? What weight are you landing at? Loadout? Are you cutting the throttle below 80% N1 before the main wheels are on the ground?

Thanks so much for the performance fixes, it feels absolutely great now!

Judging by the rest of your post, this isn't directed at you, but at everyone who's complaining about the changed flight characteristics, or firing delay (which has long applied to many fixed wing modules in the game, and should probably apply to even more of them for many weapon types). As for those people - if you can't cope with the helicopter being made to behave more like the real aircraft, I don't know why you're here. There are plenty of games where you can fly helicopters that magically disobey the laws of physics in favour of being easier to fly or more "capable". Personally I enjoy the quirks and individual character that's added when things are revised to behave more like real life, it's the entire appeal of DCS in the first place and what makes a particular module fun to learn and fly.

Was hoping to get back into DCS after about 2 months away, but not sure where I'll go now. There was (and is) nowhere else that played the same, had the same level of challenge for the Cold War jets, or attracted the same level of play. I'm going to really miss the Combined Arms cat-and-mouse, even with the CA bugs, and the cheeky SAM sites. Thanks for running the server for so long through thick and thin and sorry I couldn't be around more lately - would've been nice to see it out with a bang instead. It looks like DCS Cold War is going to be really something in a couple years, and the little one might be a little less labour intensive, so if you decide to kick things back off you know I'll be there.

The aircraft has a fly-by-wire system whose entire purpose is to interpret your inputs to figure out what you want it to do, then manipulate the controls accordingly. There is no direct linkage (except in mech reversion mode which is an emergency backup), there is always a computer between your stick input and the control surfaces. You are not flying the plane, you are telling the computer what you want the plane to do and it is flying the plane for you. You will also find that even many older aircraft without fly-by-wire will not necessarily give 100% stabiliser deflection with 100% stick deflection, because many of them have pitch-gearing systems which serve to stop you breaking the plane at high speed and do so by adjusting the ratio between stick and tail movement. MiG-19, MiG-21, Mirage F1, and from memory the Viggen are all examples of this.

I am nitpicky as hell and it's still my most-flown and favourite module, do with that info what you will. It's fun despite its shortcomings.

The MiG-21bis, if we assume that the US either never got a sniff of the war reserve mode, or they did and programmed their RWRs to process it as a launch. Other than that (manually initiated by the training/combat regime switch) frequency hop, which has to be performed before launch, as far as I know there is no state change for the RP-22 either.

The MiG-21's RSBN uses the radio beacons to correct a dead-reckoning system, which in turn tells you your location (as far as it knows) with regards to the selected beacon. So for example, when you lose line of sight to an RSBN station, the aircraft should still know within reason where it is and how far away you are from it, with drift building over time and then being corrected once signal is regained. Theoretically you could undertake an entire flight with little to no actual RSBN beacon capture and still have a fairly accurate idea of where you are. This system is already implemented in the L-39 as far as I know, but the 21's nav system in DCS is piggybacked off the old FC3 nav system which uses airfields as waypoints and is quite limited. The way it acts as a dollar store TACAN knockoff ingame really belies what navigation capability the jet actually has, around this era it would be very uncommon for a non-export (ie F-5) type to lack some kind of dead-reckoning, be it radio, visual fix, or doppler corrected (for example the Viggen's is a mix of the latter two - not an INS as commonly believed), or a genuine early INS. The auto-approach feature also has some issues at the moment and so is usually more harm than good, and there are also some inaccuracies with how ARK works, though I don't usually use that system as much so I don't remember the specifics. I think it's something like the sector setup is really simplified and the channel buttons don't do what they're supposed to. As for the autopilot - stabilisation mode currently puts an awful damper filter on all control inputs so it makes the jet handle like trash. What it should actually do is act as an always-on attitude hold. The control stick itself has a little play between it and the extension, and there are 8 (4 sets of 2) microswitches that press against the inside cup of the stick extension when the pilot exerts pressure on the stick. These are there to disconnect the autopilot when the stick is moved deliberately, and then reengage it when pressure is released. You're flying along in attitude hold, you move the stick, the jet responds how you would expect it to, you release the stick, the jet now stays where you've pointed it (though the system loses accuracy with extreme pitch or bank angles). It will also roll the wings level if you are within a couple of degrees of the horizon when you release the stick. From the verbal description in the manuals, it's about as close as you're getting to autotrim in that era, though I don't know if it has a means of coping with things like asymmetrical loadouts or not as the way the AP is currently modelled does not even in recovery mode. Currently the aircraft does have an attitude hold modelled, but you have to engage stab mode and then press an additional bind (this does not exist on the real aircraft) to engage it. It then needs to be disconnected manually, it can't be done by stick movement nor regain control after it like the similar systems on Viggen or Mirage 2000.

ED modules have had the same issue in the past, I wouldn't be surprised if others did/do too. In this case at least it's 80km/h or so, not 300, and the 21 has some hard limits that stop you achieving warp speed 9. Needs fixing at any rate, so hopefully that can be done without too much delay. I'd recommend detailing your findings on the M3 bugtracker to make sure it gets seen, as often forum threads don't. e/ APU-60-II was mostly reserved for MiG-23 and occasionally 25, IIRC, as there were never enough in supply to meet demand. Photos of them in use on 21s are relatively uncommon and usually foreign (I've seen them on Finnish and Indian 21s). It's hard to say what limitations there were to prevent loading 8 missiles but it's possible that the reason wasn't just wiring, but perhaps negative impacts on stability particularly at low speeds, as the 21's manual is full of such limitations for pretty much every store except the existing air-to-air missiles and rails prior to R-60 being brought across to the 21. R-60 itself didn't pass state trials until after the bis was in production and so it's unlikely its launch rails would have, either. Early prints of the aircraft manual make no mention of the weapon at all.

I would prefer to see all DCS aircraft brought to the same standard (no state change = no launch warning, or lock to be considered the same as a launch warning as is apparently US practice for the S-200's FCR) rather than a module which does things correctly be gamified to give warnings when it shouldn't, to be quite blunt.

Currently, yes.

Make sure you have not switched on the isolation valves above the throttle.

...I'm going to go out on a limb and say we've probably isolated the problem already. You've achieved a lock in the third image, but you've also switched your radar scope into low altitude mode, which turns off the search antenna and leaves only the tracking antenna active. It also means you won't see the tracking symbology on the AR-18 sight glass (I don't remember if you should or not in low alt mode - either way, ingame, you don't). Leave this switch alone unless you're at low altitude (~4km, clutter beginning to fill search radar scope). Note that the switch positions are currently the opposite of what the manual (and, presumably, the GR guide you followed) states - it's a known issue and will hopefully be sorted soon.

25's top speed at low level is limited by severe flutter @ 1200 IAS. It could go faster (albeit not by much) if you really hate having your ailerons still attached. It is however also a very big aircraft that has a surprisingly poor TWR when fully fuelled. Regardless, good luck finding any aircraft that can exceed M 1.2 on the deck in standard conditions, while in combat configuration (ie full onboard equipment, painted up, external antennas present, etc.) - especially from that era. You won't, because they either physically don't have enough thrust to keep them accelerating, or because something happens at that speed which is mutually exclusive with continued flight. Any claim the F1 could achieve north of 800 knots at sea level is naive at best and intellectually dishonest at worst, which fits with the moving of goalposts and failure to address any argument which isn't easily deflected or dismissed as some sort of logical fallacy, as if fluid dynamics care about high school debating technique.

The RWR has worked fine in multiplayer for a very long time for me, more than a year at this point - locks are directional unless someone is right up your backside. In MP, a lot of people often are before they lock you, which is why you might be getting the impression that the new behaviour isn't working. Powerful radars like the AWG-9 will also overwhelm all 4 sensors at quite long distances, too. I'm not sure the "experimental features" flag actually does anything currently - it just hasn't been removed yet. Since the topic of the module's age and state of its code has come up: remember, this was a mod before third parties were a thing that existed, and it very nearly didn't happen. It is every bit as spaghetti as DCS itself, because (in an issue it shares with the Viggen) a lot of systems the module needed weren't officially supported, or didn't have proper APIs to keep things consistent, or there was a different perception of DCS as a game versus DCS as a high fidelity simulator. There are a lot of very old bugs/inaccuracies which have their root in this origin - the way the nav system piggybacks off of FC3 nav code and the 21's RSBN/ARK are fully self-contained (and not functionally correct), the way the ASP has perfect, always-on CCIP including for weapons/modes it shouldn't (like the Su-25T's air-to-ground mode), and probably most obviously, the fact the radar sets off the Ka-50's laser warning receiver, but does not set off lock warnings on any RWR in the game, and operates suspiciously like the Su-25T's Shkval, especially where the Grom missile is concerned. There is a lot of work to do, probably most of the module needs to be recoded from scratch. It's still a lot of fun to fly, even knowing what's wrong or having to manually work around the systems just to avoid training more bad habits that are specific to the way the module works/doesn't work. There are still periodic improvements or fixes even while M3 is trying to finish other projects and we're still waiting for a big rework. I reported a fairly small but significant issue with the hydraulic system's magical ability to stay pressurised regardless of engine RPM until the aircraft rolled to a stop on the ground, within a week or so it was fixed and now you have to be very careful any time the engine isn't windmilling fast enough, because you'll find your controls locked up and the AB nozzle actuators frozen. It brought a pretty sneaky and unpleasant bug with it, that was then fixed within another week or so. It makes sense that small but relatively simple fixes are a better cause to push for now than the full rework the devs know is needed anyway, but which would be a serious problem to take on now, before the Corsair is out. My point is, we can start reporting some of those easy fix issues for now, rather than stating the obvious and doomposting about it every 4-6 weeks.

They are and always have been. I don't know where this myth came from, but it's pretty evident if you spend much time observing them. The situations where missiles appear to fly through proximity fuzing range or even directly through an aircraft are due to differences between server and client interpolation in MP. I have died to missiles my teammates decided to lob over my shoulder at the guy I was engaging and I've watched my own missiles just barely reach proximity fuzing range before running out of energy, so they detonated in an exhaust plume or just barely within reach of a wingtip. They're modelled.

This would be a great fix for Cold War communities, at the moment the missile is virtually useless against anything but a straight-and-level target.

There are some perculiarities to the 21 to bring up here: according to the flight manual, below 5000m and speeds of M 0.4, the aircraft does not develop any prestall buffet at all and will progress straight to wing rocking. Conversely it will enter buffet far before any risk of stalling at higher speeds. You end up with a situation where you get no real warning where you really need one, and then a premature warning that can prevent less experienced pilots from attaining maximum performance for fear of a stall that won't happen. The former is probably exactly why the SUA-1 warning lights were added where they are, to make sure you're aware of how close you're getting.

This would be a small but really helpful fix for the Cold War community, between the inability to uncage the missile and then this issue, it makes it quite hard for Viggens to hit anything other than completely unreactive targets at the moment.

It's worth mentioning as well that the aircraft handles quite differently based on whether you use the stabilise mode or not. Reading the real aircraft manual will give the impression you should just fly with it on at all times except combat, which would be true if it actually worked, but in DCS all it does is slap a gigantic damper on your inputs so it makes the aircraft feel completely different to how it feels with the system off. Anecdotally it seems quite a few people haven't realised this yet (though I doubt you're one of them, it still bears mentioning) and so are experiencing a completely different aircraft to the rest of us. Also to back up Matt's post - yes, this. I finally started playing DCS in late 2018 and at that time and for most of the following year, the 21's FM would simply switch off and it'd start dropping towards the ground while still retaining whatever AoA you commanded. When the wing rock finally appeared (or reappeared? I've been led to believe the module had it at or near launch but it 'broke' at some point) it made the jet feel much more alive and forced me to improve my flying a bit. There was also that one patch in early 2020 or so which broke the FM so the aircraft wouldn't exhibit any departures at all, I had to stop flying for a month and a half so I wouldn't pick up bad habits from the 21 suddenly flying like a Hornet. In general I'd say the FM could probably use some tweaks to out-of-envelope behaviour including extreme low speed handling, but it's better now than any other point since I started playing and it still exhibits more believable post-stall behaviour than some other modules do. e/ I think the overspeed flameout is just a DCSism to prevent people accelerating to mach stupid, but I'm not sure. 1300 IAS is a dynamic pressure limit IIRC but I'm pretty sure the M 2.05 limit is due to directional stability, like most double sonic MiGs. The fuel starvation flameout is pretty similar to how other modules model it (and makes perfect sense, you're not going to have fluctuating RPM if combustion just stops), and the R-25 is pretty hard to choke otherwise. You'll get comp stall bangs and fluctuating RPM if your antisurge doors fail or the cone is in the wrong position, but that's about it, to kill it by excessive AoA in DCS requires load factors beyond what the aircraft can structurally tolerate.

There are a few others around the cockpit as well. I'm not sure if M3 would rather wait until the rework to address typos, but I've been slooooowly working on my own cockpit retexture project that corrects typos/incorrect labels where I find them as well as adjusting the font. It's here if you're interested: https://www.digitalcombatsimulator.com/en/files/3325380/

Good request, this would help a lot of people learn the jet more quickly and painlessly, I think. It'd also be great for setting up custom detents/adjusting curves to catch them.

No. Rearm will obviously rearm you, will refill the nitrogen tanks for the brakes and chute, will repack the chute, and in general do absolutely everything except top off the radar juice. You'll need to repair to do that. I usually run a repair every second or third sortie, depending how heavily I've been using the radar - and I'll only turn it on near the combat area. I usually try to go back to standby between engagements, but I've become a bit lazy about it recently. On the way home I switch it off and don't put it back to standby until after the next takeoff and climb out.