Noctrach

If it ever finishes, sure. The "recent" AI updates only made guns-only against the AI (and thus OP's campaign experience) even sillier. At least they're no longer pulling endless vertical loops. The AI simply has not got the same restraints as the player. Fly against an F-5, MiG-21, F-16, Su-27... it's all the same experience. If it has chosen "rate fight" as it's tactic, it will not get out of afterburner and end up rating around at very high speeds (0.9+ mach) while using G to gain angles. As such you can regularly see an AI F-5 pulling 9G. I've seen them go as high as 11. If you so much as scrape them with a bullet, they will pull their own wings off on the next turn. They will not stop doing this even if you slow down and go 1C, so you will get endless snapshot and missile opportunities, but you'll never get on their six as they will simply always have a higher energy state. If it has chosen "radius fight" as it's tactic, it will wallow at just above stall speed with perfect nose authority. You will never get "better radius" because the player's flight model does not have the same level of control at those speeds. However, you can just speed up and take the fight vertical for endless high to low guns solutions, because they will never take a proper gunshot anyway (only head-on for some reason) and will not have the energy to climb with you. If you have managed to get behind them, they will light the afterburner and just start pulling 6-8G jinks out of plane every time you threaten a gunshot (for which they have perfect awareness). Once again, you're getting literally endless opportunities for snapshots and missile solutions once it starts doing this, as the AI will never translate a jink into a defensive break or similar way to force an overshoot. It will just keep plowing on at 450 knots or more. They are not flying anything close to proper BFM, they're just using perfect awareness and unrealistic flight models. Most fights will end with a relatively high aspect guns snapshot kill, or if you carry missiles... by a nice rear aspect shot after the 3rd turn. You can beat even Ace level 4th gens in an F-5 while never coming out of MIL. Sooner or later it will fly into your gunsights, run out of fuel... or into the ground. It's simultaneously incredibly easy to beat them, and really really frustrating. If you want to learn about the handling characteristics of your aircraft and get really good at snapshots, use the AI. If you want to practice proper BFM basics, start running butterfly sets with another player.

See bug thread and attached tracks. Interestingly, the glitch seems to disappear at certain sun settings, shown in 2nd track. It requires some attention though because this can be an extremely obstructive bug for target designation. ShkvalGlitchDaylight.trk ShkvalGlitchDisappearsAtDusk.trk

Yep, this bug is AI only from where I'm standing. In my very limited experience against players they universally go down in 1-2 hits depending on what systems get taken out. If I let the AI Ka-50 shoot at me it pretty much disintegrates my Ka-50, no matter the aspect. It will pretty much always take out my tail and at least one engine. Same story with the Mi-8. AI Ka-50s can't engage player aircraft with Vikhrs though so I am unable to test this. Ka50_destroyed.trk Ka50_destroyed2.trk Ka50_destroyed3.trk Mi8_destroyed.trk Mi8_destroyed2.trk Mi8_partiallydestroyed.trk

Hey Flappie, thanks for your response I tested some more and I'm seeing the same results, 3-4 Vikhrs frontally for most jets (F-16 and F-18 result in a pilot kill most of the time). After 2-3 Vikhrs most of them make it home safe with some damage on wings and nose-cone. But they almost always to go down with a single hit from rear aspect. Helicopters have a similar story, 1 hit to the tail usually results in tail rotor/tail boom loss. Mi-24P is an exception, which continues flying after 12 (!) Vikhr hits from tail and side aspect. Happily flying along without a tail rotor, so I'll put that one as an exception to the rule (attached track) Makes me think it's an interaction between the damage modeling in general, but also the armor penetration power of the air-burst Vikhr. It currently seems to have the same effect as small arms fire. Iglas seem quite weak overall, as they don't even take down jets reliably on rear aspect hits. Helicopters seem to shrug off their hits like its nothing (everything except for Huey, which requires 2-3 to reliably destroy the tail rotor) Mi24_ActualTank.trk

This is why emergency AP disengage is a very easy-to-reach button on the cyclic... Firstly, this is a sim so whatever works for you is great. That said, proper procedure would be to hold down trim and make slight adjustments, that way your hover hold will just settle you in your new position. Best to avoid using the Flight Director mode and learn to work with the autopilot, it will pay dividends in the long run. Flight Director was not used in real life outside of certain training scenarios. Tools like hover hold are there to reduce your already substantial workload, it makes no sense not to use them and getting comfortable with them will make the Ka-50 a much easier helicopter to employ.

As per title, Iglas and Vikhrs do unrealistically little damage to AI helicopters. For Iglas, I've scored 4 direct hits and most AI will simply return to base. For Vikhrs in AA mode, it seems you need 3-4 direct hits before they finally go down. On the other hand, the 2A42 cannon with HE rounds can down a helicopter with as little as 2 hits. Not sure if this should be in weapon bugs, AI bugs or Black Shark 3 bugs, but since it primarily affects this module I figured this would be the place. Tracks attached. VikhrVsCannon.trk IglaVsCannon.trk

This would imply less than 23% safety margin between the rated 7.33G and actual airframe destruction for a clean configuration. This would also mean an F-5 with just a centreline 5xMk-82 ripple rack should start suffering catastrophic wing failure at less than 7.5G. To me this seems rather unlikely for any airframe, let alone one that has such a reputation of mechanical reliability while being used in an aggressor role. (That said I don't think I've ever seen the wings break off my F-5 in hundreds of DCS hours, so I think something else is going on)

There was a newsletter... a year or so ago? That mentioned the wing AIM-9 drag as an improvement point. Right now if you pull >15 AoA with only a single AIM-9 on the wing, it basically flips the aircraft sideways from the drag induced. You have to put in full opposite stick and a lot of rudder to compensate.

Is it not more likely that this is drag related caused by the addition of stores? All of the youtube viper drivers seem to comment on stores drag feeling off and it has always been the part where DCS FMs have been most wonky.

Try flying the Ps = -200 FPS line. For example see if you can maintain speed by dropping ~12,000 feet per minute (it's about 15 degrees nose low). It's not an easy profile to fly, but when you hit the numbers you'll see that it also matches up. I completely agree with @gortex. You're really better off looking elsewhere because this FM is not going to reveal any magical causes for anyone's BFM woes. The only reason we're having this discussion is because the Mirage and Hornet have no charts to conclusively say they should or shouldn't out-rate the Viper in their modeled configurations.

Its still pretty much the same airframe though no? It seems to me that a turn at idle power would minimize the engine difference and be pretty close to a 1:1 comparison on drag or is that not a valid assumption? Any suggestion on how to quantify the drag performance otherwise? --- Besides, the tables of the Blk 42-220 and the Blk 50-129 are very very close, with the Blk 50 having a marginal advantage. No matter how many tests and data comparisons I do on the Viper, it pretty much always ends up matching the available documentation. No matter how hard I look, I can't see any significant discrepancies for any of the clean performance charts outside the sub-3% "this is a videogame" margins. I think if there's anything suspect on the Dogfight server match-ups, the Viper is not the jet to look at.

So let's turn this on its head, how do we demonstrate this? What are we expecting to see if we fly the Ps = -200 line at a given weight, standard day, 15,000 feet? Is there a way to generate that data and fold in into a track? There is a training document on the web containing a F-16CG Blk 42 diagram showing energy bleed rates for turns at 15,000 MSL, 22,000 lbs gross weight clean. Example: At 425 KCAS, idle power, a 6 second max G level turn should leave you between 250 and 275 KCAS, average turn rate ~16 deg/s I know this is an F100-PW-220, but would this be a valid starting point? ---- Trying this a couple times: it leaves our jet at around 245-265 KCAS with an average turn rate well above 16 deg/s (around 17.5 deg/s, slightly over 105 degrees of turn). Makes me think it's pretty well tuned tbh... I've run various scenarios in this table and it seems spot on in the sim. Bleed rates, turn rates, radii, it all matches up almost perfectly. ---- Clean BFM Viper á la dogfight servers (the topic of this thread) seems honestly extremely accurate on all metrics, including drag.

The CAT I/III switch in the F-16 does not limit G. It limits roll rate, rudder authority and max AoA. The FLCS will still do 9G in CAT III. It's purpose isn't to protect stores integrity, but to prevent the aircraft from entering flight regimes in which it is prone to departures. To OP: This behaviour is definitely modeled, the warning is not.

Tbh I think proper G tolerance modeling would go an enormous way already. Model some difference in suits, seats and stamina between the various aircraft. Model smooth G increases vs peaks. It doesn't have to be elaborate or complicated, but it would already make such a massive difference in how dogfights are played in DCS. You're just not gonna be holding G forever, no matter how reclined your seat is. Suffer a couple minutes of rating around at 7+Gs and fatigue will start setting in hard. Pull an instant 9-10G and you're not gonna stay awake for very long. Some of the other combat sims can already give you examples of how fights like that play out. It's more about energy/angles management and less about horizontal rating on the deck at the speed of stupid.

First and foremost, I'm not gonna say the F-16 can't win a dogfight against peer adversaries because it does quite alright against the F-14, 15 and Mirage 2000 and pretty much crushes anything of earlier gens. On the whole I think the only thing we can comment on for the Viper specifically is that the general sustained numbers follow available documentation very fucking closely. I will however say that there are precious few fighters in the 4th generation that can't follow an F-16 in uphill flow. This is mostly down to how much speed it bleeds when tightening the turn over the top. If there's anything wrong with the FM, it's likely found in how fast the jet shoots up in AoA. (That's for a clean jet, it gets much worse with even just outer pylons) The problem there is that the F-14, 15 and 18 have almost identical climb performance, but Don't suffer as badly from AoA-induced drag (partially because of their lower wing loading, but I am looking at the Hornet FM with moderate levels of suspicion). Don't give a damn about high AoA manoeuvring because they don't have a limiter. So in practice, going vertical against either of those three mostly ends up with 2 slow jets, one of which can just put nose on and trigger down. Pilots being equivalent, no BFM errors, the Hornet is the only jet against which I feel the Viper hasn't a single viable game-plan. Then again, none of the other jets do either. However, this is all mega theoretical and that proposed scenario doesn't factually exist. I don't think the Viper is as helpless as a lot of people seem to suggest.

Dear @BIGNEWY I was wondering if there was an easy way for us to fix this or if this is something that will be solved when full migration to the MT build goes live? The sim is unfortunately quite a stutterfest for me with only borderless as an option. Is there a line we can add to the script to also apply the original fix for the MT branch?

Hey there, I enjoyed the first 3 missions of the campaign so far but mission number 4 seems absolutely impossible with the new SAM kinematics. The SA-10 system will launch mere seconds after you pop up from behind the ridge which at that point leaves 16-18 miles between you and the target. As DCS Mavericks have a hard lock limit of 10 nmi even with perfect contrast, that leaves 6 miles of missiles closing at mach 4.5 in the best of cases. Not to mention 16 nmi is very much within the danger zone of an SA-10. Combined with chaff and jammer being pointless in today's version of the sim. This seems like a scenario set up for failure, is there anything I'm missing?

If you look at the TID readout when you're in a roll attitude on a cranking target >10 left or right you will notice the antenna elevation will actually go crazy beyond the expected azimuth-to-elevation transpose. So they might be separate issues but I definitely feel they've the same root cause. The AWG-9 has no roll gimbal and will transpose ATA onto elevation and vice versa to generate the required angles to position the dish during banking turns. I've long felt something is fundamentally scuffed with the simulation of how it does this. 40 ATA plus a 15 degree roll can have your TID elevation readout shoot up way into the 30s, which is not how that math works out. In doing this you will also see the DDD return fade out and the TID track start dropping, even though STT is maintained throughout. I feel that it is somehow doing a partial simulation of the AWG-9 going out of bounds on its gimbals, without actually doing so. Not sure why this never got fixed though since it's crazy easy to reproduce.

Yep, has been happening for >1.5 years now A combination of antenna elevation + azimuth + roll exceeding a certain value will break the track but not the lock. Seems to happen when the sum exceeds 55, even if none of the individual components exceed gimbal limits. Quite annoying during BVR cranks or flying an aggressive intercept since as a RIO it means you lose target SA in STT unless your pilot reverses the crank. was reported as investigating but never received a fix

Issue is that they also lock onto lampposts and objects that it can't see. You can easily replicate this in a night mission like the Stop Rockets Rolling IA for the F-16. I'll see if I can attach a track for this tonight, but it's been reported in this other thread before and is a known issue DCS is a sim, mavericks just lock onto game object, not on true contrast. You can clearly showcase this due to the fact that it can't lock onto most world statics like ships on the PG map, even if they show as a bright blob of light on the feed. The easy solution would just be to ignore lampposts as lockable targets, but it's quite likely this would affect the entire category of statics like powerlines, meaning it might become a real a pain to find objects to boresight on.

As per attached track, since the recent update the BFM AI routinely exceeds 10G flying full afterburner at 500+ KIAS Easiest to observe in F-16s as they seem to have a very high top-speed and acceleration, but anything from a Hornet to an F-5 will replicate this behaviour. Aircraft restrictions do not apply, there's also no G-onset (snap from 1 to 10G nearly instantaneously) or GLOC (observed AI sustaining 10G for 5+ seconds). AIBreaking10G.trk

Still seems to be a significant issue, especially on the Syria map. The F-16 Stop Rockets Rolling IA mission is unplayable because the Mavericks will go for roadside lampposts every time. Even with a brightly contrasted rolling truck against cold terrain, they will never go for the truck.

@IronMike Would it make sense to discuss the AI defensive behaviours with ED in context of the ongoing missile work? An AIM-54Cmk47 launched between 20 and 40 miles has a Pk of near 0 because the AI will always opt for a split-S defence An AIM-120C at identical parameters will have a Pk of roughly 0.5 because about half the time the AI will just crank and fly into the missile. The exceedingly rare occurrence where the AI doesn't immediately split-S and burn away the Phoenix tends to arrive with a little bit more energy than the AMRAAM on long-to-medium shots. Kinetically there's no reason why the Phoenix would perform as much worse as it does... it's all in the AI response.

I'm not sure if this is pure chaff simulation in singleplayer or the AI's ungodly ability to notch SARH missiles, but the AIM-7M has abysmal performance against chaffing AI when launched outside of 4-5 miles. Missiles launched at high mach, medium altitude, at ranges between 7 and 11 miles against MiG-29s and F-5s have a Pk that is very close to 0%. Attached are 24 such launches. Missiles were launched low to high to prevent aircraft radar being notched at any point. Disabling chaff on AI brings these same shots up to a kill probability of nearly 100%. I think a root cause of the issue here is that AI has perfect awareness of the exact angle it needs to fly to notch a SARH (non-emitting) missile. Where it should only have accurate knowledge of the launch platform radar. This negates any BVR tactics like diving down to prevent look-down notching, as the large amount of dropped chaff coupled with a guaranteed split-second missile notch leads to immediate and permanent decoying. AIM7Chaff5.trk AIM7Chaff3.trk AIM7Chaff4.trk AIM7Chaff.trk AIM7Chaff2.trk AIM7Chaff6.trk

Addendum: There's definitely a factor of the object/LOD mix that is together on-screen. E.g. removing either nearby trees or distant objects from the camera FOV will also completely eliminate the framedrops. All shadows AND terrain object shadows turned off Game in fullscreen (script removed)