already reported New Flight-physics model Amraam?

[Geraki]

In the last update the Amraam lose easy the lock from target if the target NOTCH the Amraam.

in low distance you have to release the Amraam below 5 miles with 1.5 mach in order to hit the target if the target notch aggressively.

I think it is need some refinement  the Amraam AIM 120C5  to be more aggressive and not lose lock or energy so easy.

[BIGNEWY]

We have a reported open already for the team to check with the notch at close range, if you have something specific please include a track replay example. 

thank you

[TobiasA]

Out of 6 shots, 6 were a close miss against the AI. I have not scored a single 120 hit since the last update, but I don't know what causes it. Will try to get a track at the end of the week. 

[Moonshine]

Check these threads:

 

[TobiasA]

 

Am 24.10.2021 um 23:51 schrieb BIGNEWY:

We have a reported open already for the team to check with the notch at close range, if you have something specific please include a track replay example. 

thank you

I managed to get a track with two 120's just missing the targets on two MiG-19, a medium range shot of about 15 - 18 miles in TWS, supported until pitbull. No chaff.
The missile pulls way to much lead, causing dramatic energy losses (it is about worthless at more than 15 miles out), then misses the target by a solid nautical mile.

Sorry for taking a few days.

My default view should show the miss just right away on both missiles, I attached the tacview and track. The point is that the missile pulls so much lead that it shifts the target out of the FOV which then leads to relying on datalink as it seems.
As of now, I can see two problems:
- Pulling way to much lead mid-flight
- Placing targets outside the own seeker FOV during pitbull
 

The problem is the same with the second 120:

I hope this helps and this is really considered a bug, because it kinda ruins the experience
Thanks a lot!

AIM-120_outright_fail.trk Tacview-20211027-194716-DCS.zip.acmi

P.S.: However, what I saw is actually promising. Because the datalink works. And in earlier tests, the missiles did indeed track even if the radar lock was lost and reached their target if it didn't maneuver aggressively.
I appreciate the work that did go into a better AMRAAM.

Edited October 27, 2021 by TobiasA

[TobiasA]

As suggestion or "best guess": I think and I am pretty sure that the real thing limits lead to a percentage of the seeker FOV.

[BIGNEWY]

Hi 

I dont see a miss, the AIM-120 hits its target.

 

[TAW_Impalor]

I replayed the track twice and also see the first missile hitting the target. The attached tacview file did not, however, show it.

Funny how tracks work. In mine I always crash in the trees shortly after taking off... 

 

[TobiasA]

vor 42 Minuten schrieb BIGNEWY:

Hi 

I dont see a miss, the AIM-120 hits its target.

 

Now that is really weird.
Does the tacview work for you?

I'll record a new pair

[TobiasA]

Well, second try...

Tacview-20211027-211254-DCS-AA_test_120.zip.acmi

Well at least the track works, the end is a bit weird tho since it continues to run after I exited and I get shot down what didn't happen in the match.
But you can see the weird maneuvers, I replayed the track to be sure.

Sorry for the double work.

AIM-120_fail_second_try.trk

@BIGNEWY that should work now.

Edited October 27, 2021 by TobiasA

[TobiasA]

Weird thing is that the timestamp of track and Tacview are the same on my computer... 

Strange. I owe you a beer when we meet in real life one day. 

[TAW_Impalor]

Now I see Mig19 notching the first AMRAAM for one second, and it was enough to escape! Maybe the logic is to immediately re-target whatever the missile sees in the seeker. Can't be true to life, but I am no expert. 

[Hummingbird]

Also haven't hit a thing with the amraams since update, any target that just notches is entirely safe.

[Aquorys]

I have seen so many AMRAAMs fail and do absurd things that I'm just going to say almost everything about them appears to be broken. The most recent example being an AI Su-33 at Mach 0.7 that successfully dodged 6 high-energy AMRAAMs that were fired in ~5 second intervals.

 

They frequently fail to guide right after having been fired, even just into the general direction of the target, although they should, even if the aircraft that fired them magically disappears a second later. They already have all the necessary information to guide to a point where the target will end up if its vector doesn't change.

 

They practically don't work at all if they're fired from high altitude at low altitude targets, even with guidance to active range (which is one of the most well-tested scenarios in real life, because that's how long-range tests are done).

 

They almost always fail to hit the target if the target dives, apparently because they predict an intercept course that's a mile below ground level (that's not how vertical missile guidance works).

 

They frequently fail to fuse even when they pass so close that they almost collide with the target aircraft.

 

They lose track against notching targets virtually always (although it's pretty well known from the reports of Serbian pilots who where shot down by AMRAAMs and survived during the Yugoslav wars that AMRAAMs are extremely resistant to notching).

 

They reach the target area and choose implausible targets (for example, you guide them to active range, so the missile has up-to-date information on where the target is, but instead it chooses a friendly that's 4 miles further east, going the opposite direction at twice the speed and 10k ft lower, a change in target position and vector that's physically impossible).

 

In multiplayer games, it becomes even worse, because even if the missile tracks correctly, at some point the target can just jump away hundreds of meters, probably due to latency, flight-path updates being applied too late, etc., and the missiles are not updated in the same way (could be that the target coordinates for missile guidance comes from the client, but target coordinates come from the server, and it all desyncs sometimes due to latency, at least that's what it looks like) - obviously making it impossible for the missile to hit the target that's now suddenly somewhere else.

 

The outcome reflects this too, the rate at which AMRAAMs intercept successfully is quite low, even if you only count textbook-quality shots. What's funny but irritating is that arcade-game-like shots are actually somewhat more successful (e.g. same altitude or below, level flight and head-on deployment) than more realistic ones (e.g., higher altitude, higher speed, positive V/S, head-on deployment).

 

Even first generation AMRAAMs achieved a kill probability of better than 75 percent in the less than ideal conditions of the real world. If you replicated all those shots in DCS with the way more modern C-5 that's supposed to be modeled in the game, I would not be surprised if less than 25 percent of them hit.

 

End of rant, as it is probably pointless anyway. I mean, it's been like that for how many years now? We are never going to get AMRAAMs that actually work, and the reason in this case is probably ED insisting on trying to model too much realism into them without knowing how. Virtually everything about missiles is classified. Noone, literally noone, is ever going to tell ED any specific numbers and formulas concerning the guidance principles, trajectory algorithms, fusing logic, radar characteristics, warhead effectiveness, etc., just not gonna happen. The story of realistic AMRAAMs is over before it begins.

Edited October 28, 2021 by Aquorys

[TobiasA]

Well my targets don't even really notch. 

The 120 corrects and guides but places the target outside the seeker FOV or calculates weird lead solutions.

Sadly, the 120 in DCS has been made fun of with Tacview gifs in other communities...