AFM Missile

EDIT This was originally just a bug report, however after starting a separate thread on AFM missiles, that thread was merged with this one for whatever reason. Post 2 is the original post from the AFM missile thread.

Original Post:

I decided to look at AFM missiles (very happy to see 3000 knot AIM-120), but a few strange things happened. I pit a F-15 against a MiG-23. The F-15 eventually killed the MiG with a AIM-9, however while the AIM-9 was launching, I happened the pause the mission. The AIM-9 continued to turn while the simulation was paused. It was slow, so I don't think it effected the missile accuracy very much.

Secondly, after the planes killed each other and the pilots ejected, they continued flying. The F-15 had no engines, but would not go below 190 knots.

Trying to upload track, but my internet is painful at the moment.

Extreme Glitch.trk

AFM Missile

So I've started messing around with them, there is some good and some bad. So far, I've only seen AIM-120C, R-23, AIM-9, and R-60. And only I've only really seen the first two fly for any real amount of time.

Good news is launch speed and altitude make huge changes in missile performance. AIM-120 peak speed has been anywhere from 3000 to 3800 knots. Tight turns (like those seen in tracking glitches) cripple velocity. No one needs to fear active missiles making 180 degree turns. It's now nearly impossible.

There is some bad though, the AIM-120C at least does noes seem to have a very optimize loft trajectory. It will pull a maximum g turn coming out of loft, which basically bleeds it of energy completely. This might make the AIM-120B, AIM-7, R-77, and R-27 better than the 120C. I have to do more testing.

So I've started messing around with them, there is some good and some bad. So far, I've only seen AIM-120C, R-23, AIM-9, and R-60. And only I've only really seen the first two fly for any real amount of time.

 

Good news is launch speed and altitude make huge changes in missile performance. AIM-120 peak speed has been anywhere from 3000 to 3800 knots. Tight turns (like those seen in tracking glitches) cripple velocity. No one needs to fear active missiles making 180 degree turns. It's now nearly impossible.

 

There is some bad though, the AIM-120C at least does noes seem to have a very optimize loft trajectory. It will pull a maximum g turn coming out of loft, which basically bleeds it of energy completely. This might make the AIM-120B, AIM-7, R-77, and R-27 better than the 120C. I have to do more testing.

IIRC the missiles we have now in DCS 1.2.1 are only half AFM. That is, the AFM for missiles has not been completely implemented. I would hold off all judgement until after FC3 comes out.

This is AFM, not AMG (Advanced Missile Guidance). AMG will probably come later.

There is some bad though, the AIM-120C at least does noes seem to have a very optimize loft trajectory. It will pull a maximum g turn coming out of loft, which basically bleeds it of energy completely. This might make the AIM-120B, AIM-7, R-77, and R-27 better than the 120C. I have to do more testing.

Please, continue your investigation. It may uncover something that has been missed and will perhaps result in a fix when FC3 comes.

So, now I've compared ARH to SARH. All soon to be player useable ARH missiles pull max g turns coming out of loft, but thankfully the loft trajectory has a minimum range below which the missiles follow a lead pursuit path. SARH missiles did not loft in my test and kept their speed all the way to the target, but they are a lot slower than the ARH missiles.

At long ranges (15(?) to 26 nm), ARH missiles can be easily bled of energy by taking advantage of their loft behavior, and this may almost negate AFM modeling. At shorter ranges though, aircraft velocity is very important, as it can dictate whose missile will strike first, and whose missile will have the higher Pk.

I think a couple of important things to look at will be the loft minimum range and behavior of ARH missiles when fired from low to high, as in this case the loft should not hurt them. I will look at these next.

EDIT - Loft range is determined by the missile seeker. If the seeker is in range, the missile will not loft. If the target is outside of the seeker range, the missile will loft. Seeker range is about 15 nm. Missiles fired below loft speed are much deadlier, they can reach the target at velocities slightly below the hard coded speed limit in FC2.

SARH missiles did not loft in my test and kept their speed all the way to the target, but they are a lot slower than the ARH missiles.

This one is interesting... seeker type affecting speed?

seems seeker type is affecting trajectory/guidance which affects G and ultimately speed

seems seeker type is affecting trajectory/guidance which affects G and ultimately speed

Yes, that along with different rocket motor performance.

I will try to post more quantitative data at some point, can't access the sim right now.

I did manage to do a little bit of testing with low vs high, but the AI biases the test slightly because it only fires when the missile seeker is in range in that case, so I can't see if the loft will aid a missile fired at a target flying much higher than the launch platform. AIM-120C from 5000 ft launched at 600 knots reached a target at 30000 ft flying also at 600 knots at a final velocity of about 2000 knots. The range between the two at launch was 11 nm.

So, I had hoped to come in here and post a big chart with all kinds of missile data vs targets at various ranges and positions, etc. However, the AI makes things a bit tricky and I'm still trying to figure out how to make them do exactly what I want.

But in the mean time, I did see some strange things from missiles at high altitude. A missile launched at around 65000 ft that reach peak alt of maybe 70000 was able to pull a 10 g turn coming out loft. During this same test, the two AI planes got into a high altitude dogfight and lost some altitude. At 35-45 Kft, they were firing missiles at each other within 10 miles, but the missiles didn't seem to be able to exceed 10g. Actually, they couldn't even reach 10g, even with rocket thrust. I think the max was 8-9.

I found this odd given that the missile at a much higher altitude could reach 10g. I'd imagine that nothing in the flight dynamics is scripted at this point, so would the peak g at extreme altitude be the result of the bug or something?

On the plus side, it looks like R-73 thrust vectoring is simulated, it makes some crazy maneuvers, though at very high altitude, it still suffers some performance loss.

When setting up their waypoints, you can set up their behavior to 'no reaction' to threat.

So, I had hoped to come in here and post a big chart with all kinds of missile data vs targets at various ranges and positions, etc. However, the AI makes things a bit tricky and I'm still trying to figure out how to make them do exactly what I want.

When setting up their waypoints, you can set up their behavior to 'no reaction' to threat.

I tired this. I could get a MiG-25 to obey, but the F-15 refused to fire on it until 11 nm. I changed the MiG to a Su-30, the F-15 would fire at longer range, but the Flanker refused to hold altitude.

I've updated to 1.2.1.6055 this morning, so maybe the issue will be gone when I try to run these tests again tonight.

I think I'll also change the AIM-120 carrying plane from the F-15 to the F-18 since it's slower. If the F-15 doesn't fire immediately, its speed will be much greater than what I set it to initially. Unfortunately on the Russian side, all their R-77 carrying planes are "fast" so I don't know what would be the best to pick. Maybe a Flanker will full fuel tanks? Or can I force the fighters to never exceed a certain speed in combat? I'll look through the AI options tonight.

Any new find on this?

At the moment, no. I've been too busy, and with so many updates coming out at the same time, it might not be a good idea to collect data that may or may not remain consistent between patches. It's mostly because I haven't had time though.

I might be able to do more tests tonight and possibly on the weekend.

EDIT -

I managed to do a quick comparison test between AIM-120C, AIM-120B, and R-77. F/A-18C carried AIM-120, Su-30 carried R-77. The target in all cases was Yak-40, placed about 28 nm away, flying at 295 knots and 26400 ft. Patch 1.2.1.6192

Missile

launch altitude [Kft] - launch range [nm] - launch velocity [kt] - max velocity - impact velocity

AIM-120C

30 - 24 - 636 - 2900 - 892

22 - 21 - 617 - 2610* - 823

AIM-120B

30 - 25 - 628 - 2600 - 500

22 - 22 - 615 - 2400 - 442

R-77

30 - 19 - 709 - 2350** - 514

22.6 - 17 - 701 - 2300 - 475

* - Missile pulled max g turn out of loft just before the rocket cut out, resulting in loss of speed

** - Missile pulled max g turn while rocket was burning. Max speed was reached in a dive.

The R-77 definitely has the worst kinematic performance of the three missiles. What I found most interesting though was that the R-77's powered dive didn't seem to help it. The AIM-120's both lost power before or during the turn to come out of loft while the 77 fired from 30000 ft came out of loft with full power (though acceleration basically stopped during the turn). I'm pretty sure that it had a higher initial dive speed than the AIM-120B, but it only just bearing beat the B's final velocity despite having less distance to travel.

I'm also wonder if the missiles are maybe a little too draggy during maneuvering, but as I've not done the math I don't have anything to support my feeling. What I really want to see is a range increase from proper flight path modeling, or AMG as GG put it.

Thanks for the analisis.

More tests this morning, this time missiles fired within radar range

F/A-18C carried AIM-120, Su-30 carried R-77. The target in all cases was Yak-40, placed about 28 nm away, flying at 295 knots and 26400 ft. Patch 1.2.1.6192

Missile

launch altitude [Kft] - launch range [nm] - launch velocity [kt] - max velocity - impact velocity

AIM-120C

30 - 24 - 636 - 2900 - 892

22 - 21 - 617 - 2610* - 823

30 - 15 - 602 - 2970 - 1897

22 - 15 - 594 - 2750 - 1610

AIM-120B

30 - 25 - 628 - 2600 - 500

22 - 22 - 615 - 2400 - 442

30 - 15 - 602 - 2650 - 1369

22 - 15 - 594 - 2400 - 1069

R-77

30 - 19 - 709 - 2350** - 514

22.6 - 17 - 701 - 2300 - 475

30 - 15 - 606 - 2550 - 1106

22 - 15 - 604 - 2350 - 731

* - Missile pulled max g turn out of loft just before the rocket cut out, resulting in loss of speed

** - Missile pulled max g turn while rocket was burning. Max speed was reached in a dive.

The 77 is still behind. The missiles in this test showed a bit more sensitivity to altitude, as the impact velocity varied more between the two altitudes that it did in the long range test. It's also interesting to note that the missiles were slightly faster when launched from 15 nm despite not having to climb. This shows the effect on air density on acceleration, and range may be greatly improved once loft is fixed.

I think my next set of tests will involve the ranges and altitudes above, but with a beaming target. Then I will try to get the AI to fly a repeatable "evasive maneuver".

One last update, at least possibly until FC3 hits and I'll have more control.

I used the flyable aircraft mod for DCSW to become a missile target and give myself slightly more control over the testing. Lofting does not work, regardless of relative target altitudes. The missile always tries to go above the target's altitude, and in then burns all energy in a max g turn once close enough.

However, the SARH missiles showed some promise when lobbed to high altitudes, they were able to reach targets over 20 miles away (at launch) and retain their speed (~Mach 2 impacts) even when they had to climb.

Missile maneuverability and speed makes them pretty effective against targets that are closing, even when maneuvering. I think that once missile guidance is fixed, ARH vs SARH will become even more lopsided. When targets turn and run, it seems to make missiles bleed a lot of energy, so that makes me wonder if duels between F-15's and MiG-29's (or R-77 equipped Flankers) will see longer range kills or not. At this point it's hard to tell. I did have the advantage of the simple AI flight model, which doesn't bleed very much energy during turns, but even taking that into account I didn't expect to be out running medium range missiles launched from 5 or so nm.