Air-to-Air Missile Discussion

[GGTharos]

Guidance is part of it, but it's really nowhere near as simple as you may believe it is :)

 

You're also answering the wrong question I think :)

 

 

PS: You realize you're actually showing that there's a problem with the guidance in F4 right? IMHO anyway. Look at those flight trajectories - do they remind you of anything from DCS?

Edited February 4, 2015 by GGTharos

[lunaticfringe]

In my opinion, the missile is not adequately modeled.

 

Boy, nice job taking a statement regarding the interaction of decoys versus sensing and turning it into a bad remark regarding loft programming.

 

PG is still in effect during a lofted profile; the weapon's logic is processing range and aspect information as required to turn a known energy constant into enough kinetic-plus-potential energy to render a valid intercept. It is not two separate stages, but one long calculation- the missile maintains the constant lateral LOS required. This is reinforced by performance at short range: an AIM-120 doesn't need to loft with a shot taken at 5 miles. Instead, the equation is performed, energy is noted as sufficient to get downrange, and the logic is followed.

 

Loft, in relation to missile performance and guidance, is a known issue. It is not simply known issue, it's also an excessively difficult one to solve based on the grade of data that is available on the "white" side of the fence, not the least of which are the constants that have to go in. Although, if you'd like to provide exact engineering data as to the ARs of the fins of various weapons, the specific energy total of their solid fuel boost and sustainer stages, along with their full drag profiles, they might be able to get back to you far more quickly than they currently are.

 

Instead, they're having to work backwards, from oftentimes wildly inaccurate sales data, and flight data for singular regimes. And honestly- that doesn't tell you a quarter of what you need to know as to how the weapon got there from the physics side, because those inputs can be fiddled with any number of ways to get the range up.

 

One has to have that grade of intimate data to be able to effectively feed a lofted PG equation. Otherwise, you're dealing with even worse performance than we already have.

 

So frankly- as the man said, it *is*, as currently constructed, adequate. They are working on it in a near-constant fashion, but it is a painful process.

 

But, like I said- if you've got a hard line on state secrets that you're willing to spend a couple decades in a maximum security facility for releasing to the developers of a consumer-grade flight simulator, be our guest. Otherwise, don't shoot the messenger.

[flankerted]

Guidance is part of it, but it's really nowhere near as simple as you may believe it is :)

 

You're also answering the wrong question I think :)

 

 

PS: You realize you're actually showing that there's a problem with the guidance in F4 right? IMHO anyway. Look at those flight trajectories - do they remind you of anything from DCS?

 

I'm sorry about answering the wrong question, but given the truth that the BMS has already worked out an adequately accurate model of these missiles both in the terms of aerodynamics and guidance, it dose not seem that it is hard for you to accomplish such a task. If they can be modeled like that in the BMS, I believe almost all of FC3 players will give you pretty high prise of them. :)

Edited February 4, 2015 by flankerted

[Frostie]

If they can be modeled like that in the BMS, I believe almost all of FC3 players will give you pretty high prise of them. :)

Absolutely not, for a start DCS missiles have progressed to AFM which is a step in the right direction, but more importantly BMS missiles are just as useless against human opponents.

 

Yes they kill AI with ease because BMS AI once in the parameters are weak but against human opponents where so much of the opinion of DCS missiles comes from they are just as avoidable if not more so. DCS issue with the stupid AI is that countermeasures are too effective

[GGTharos]

I'm sorry about answering the wrong question, but given the truth that the BMS has already worked out an adequately accurate model of these missiles both in the terms of aerodynamics and guidance,

 

Their model sucks. It achieves what it needs to do, but as a model it in fact sucks. It's effectively the equivalent of missile SFM AND the guidance is ... I don't know what, but it ain't right, looks like they're just going pure to me which is er ... wrong. Wrongwrongwroooooong.

As for missile range, it's like that because they're not letting those missiles turn much. Which results in intercept issues and the stuff you saw on those ACMIs. DCS missiles are 'front-loaded', they pay for FM accuracy straight at the beginning because the guidance does not do anything to maintain energy.

The BMS missiles are 'end loaded' ... the guidance and FM are so bad that these missiles keep a lot of energy, but aren't able to use it for an intercept against a fighter performing a robust dive. :)

Edited February 4, 2015 by GGTharos

[flankerted]

Their model sucks. It achieves what it needs to do, but as a model it in fact sucks. It's effectively the equivalent of missile SFM AND the guidance is ... I don't know what, but it ain't right, looks like they're just going pure to me which is er ... wrong. Wrongwrongwroooooong.

As for missile range, it's like that because they're not letting those missiles turn much. Which results in intercept issues and the stuff you saw on those ACMIs. DCS missiles are 'front-loaded', they pay for FM accuracy straight at the beginning because the guidance does not do anything to maintain energy.

The BMS missiles are 'end loaded' ... the guidance and FM are so bad that these missiles keep a lot of energy, but aren't able to use it for an intercept against a fighter performing a robust dive. :)

 

Actually what I mean is not to ask you to make the missile's AFM 100% same as that in BMS, because from the video we can find the FM is absolutely wrong in the terminal guidance stage. As mentioned above, the guidance is not two separated stages, but I think these missiles should be close to pure pursuit at long distance from the target, and at short distance from the target, they should obey proportioal guidance law. The key to simulate the progress could be making the guidance more than one stage , such as three or more stages, to instead the complex function of realistic one. In a word, the AFM of these missiles is adequately accurate in aerodynamics, and the model in BMS is too simple but has something for reference to use to enhance the guidance of AFM missiles in DCS.:)

[Jamovich]

Interesting article on this whole AIM120c being "nerfed". According to this article it probably is pretty well modeled in DCS (personal conclusion). I don't claim to be an expert however probably neither are those who claim that the aim120 has been nerfed "in game" to appease the red team

 

http://www.ausairpower.net/APA-Rus-BVR-AAM.html

 

There is certainly a difference between the missiles however the aircraft are sufficiently equipped to fit in the performance doctrine of the said designers.

 

I guess you can't follow the same BVR tactics in an F15 as you do in a Sukhoi

Edited February 6, 2015 by Jamovich

[wilky510]

Interesting article on this whole AIM120c being "nerfed". According to this article it probably is pretty well modeled in DCS (personal conclusion). I don't claim to be an expert however probably neither are those who claim that the aim120 has been nerfed "in game" to appease the red team

 

http://www.ausairpower.net/APA-Rus-BVR-AAM.html

 

There is certainly a difference between the missiles however the aircraft are sufficiently equipped to fit in the performance doctrine of the said designers.

 

I guess you can't follow the same BVR tactics in an F15 as you do in a Sukhoi

I think people aren't really concerned about the "PK" of the AMRAAM, but they are concerned with it's inflight use of bleeding speed when it doesn't need to be. It's already been explained by GG, the missile seekers aren't working well with the AFM (is what i got from his posts).

 

AMRAAM real life PK data is irrelevant anyways.

Edited February 6, 2015 by wilky510

[OnlyforDCS]

Interesting article on this whole AIM120c being "nerfed". According to this article it probably is pretty well modeled in DCS (personal conclusion). I don't claim to be an expert however probably neither are those who claim that the aim120 has been nerfed "in game" to appease the red team

 

http://www.ausairpower.net/APA-Rus-BVR-AAM.html

 

There is certainly a difference between the missiles however the aircraft are sufficiently equipped to fit in the performance doctrine of the said designers.

 

I guess you can't follow the same BVR tactics in an F15 as you do in a Sukhoi

 

An interesting read, but there are a lot of assumptions about missile technology which would be very hard if not impossible to check or verify. Also some of the ranges set for the BVR missiles stated in the article are two to three times to what we have in DCS.

Edited February 6, 2015 by OnlyforDCS

[GGTharos]

APA is not a useful source of information. Yes, the 120 is 'nerfed'. It's not the only thing that's nerfed, the nerfing's pretty fair all around.

 

Interesting article on this whole AIM120c being "nerfed". According to this article it probably is pretty well modeled in DCS (personal conclusion). I don't claim to be an expert however probably neither are those who claim that the aim120 has been nerfed "in game" to appease the red team

[HiJack]

APA is not a useful source of information. Yes, the 120 is 'nerfed'. It's not the only thing that's nerfed, the nerfing's pretty fair all around.

OK so it's maybe fair in a multiplayer server with loads of different skilled players but have you played any singleplayer mission lately? There is almost no way 4 * F-15C with full load off 120's can win over 2 * MiG-23's. Take the "FC3-F-15C-Clean Sweep.miz" for a run will you? And please post the finnished track where you beat all aircrafts and arrive safely at homebase :thumbup:

[Exorcet]

OK so it's maybe fair in a multiplayer server with loads of different skilled players but have you played any singleplayer mission lately? There is almost no way 4 * F-15C with full load off 120's can win over 2 * MiG-23's. Take the "FC3-F-15C-Clean Sweep.miz" for a run will you? And please post the finnished track where you beat all aircrafts and arrive safely at homebase :thumbup:

I play a bit of single player and I think the missiles are perfectly workable there. The AI is another issue though. They are not very good at using missiles and horrendous at listening to orders. They will not lead their shots and they will not account for future actions on behalf of the target.

 

For fun I decided to try the Clean Sweep mission. It's my first time doing any DCS mission that came with the sim. I've also got a track I use for training, 2 F-15 vs 4 MiG-29. I don't know if Clean Sweep recorded properly, but I had 100% AMRAAM accuracy. I was killed by the MiG-31/MiG-29 groups though as my wingman weren't very effective. I know my 2v4 track is correct until the landing.

Clean_Sweep_1st try.trk

2 F-15C vs 4 MiG-29S 1.2.16 with good landing.trk

[HiJack]

Yes the AI does not help the situation. I have tried to just command the wingmen on the first attack resulting in them all dying against 2 MiG-23's. Have you tried that? To complete all 6 objectives in the mission you are reliant on the wingmen and they have no chance against the superior MiG-23 :P

Edited March 23, 2015 by HiJack

[Exorcet]

Yes the AI does not help the situation. I have tried to just command the wingmen on the first attack resulting in them all dying against 2 MiG-23's. Have you tried that? To complete all 6 objectives in the mission you are reliant on the wingmen and they have no chance against the superior MiG-23 :P

This was the only time I've flown the mission, but in my own missions I've seen strange behavior. Many times, the plane with the range/first shot advantage loses.

 

Su-25T beats MiG-29 in 2v2 (who would win between MiG-23 and Su-25? Battle of titans)

F-15C loses to SARH fighters

F/A-18C's on defensive defeat R-77 MiG's

 

Generally, the first salvo by the AI is too early to kill fighters. After that it was usually a mess because of the missile launch bug, but even with that fixed I've seen weird stuff.

 

Also interesting is that the AI ignoring enemies can be more effective than defending against them. I have a mission where 4 AI F-15's start attacking 4 MiG-29's. The F-15's have a larger aggression radius, but they end up shooting the MiG's as they turn to follow their waypoints, resulting in AMRAAM's being trashed. Then the F-15's get close enough for the MiG's to engage. You'd think the F-15's would have an advantage but I've seen them get utterly destroyed in this situation.

 

While I think boosted missile performance would be good for SP, I think AI missile tactics are the bigger problem right now.

[*Rage*]

Since there is renewed interest in testing missiles and understanding why they are so poor. I'm re-posting my ER testing.

 

Old testing methodology and results for posterity:

 

http://forums.eagle.ru/showpost.php?p=1 ... tcount=243

 

Overview of performance in 1.2.4 to 1.2.5 Ptrack improves .22 to .42

 

http://forums.eagle.ru/showpost.php?p=1 ... tcount=266

 

Unfortunately I havent mentioned the patch number but its likely an early one since Ptrack is .25

 

http://forums.eagle.ru/showpost.php?p=1 ... tcount=381

 

Same again. Likely early version of the game.

 

http://forums.eagle.ru/showpost.php?p=1 ... tcount=412

 

Same again

 

http://forums.eagle.ru/showpost.php?p=1 ... tcount=522

 

Patch 1.2.5 Ptrack still abysmally low at .22

 

http://forums.eagle.ru/showpost.php?p=1 ... tcount=560

 

Wags concedes that there has been a chaff bug. It is formally addressed in the next patch. Update 1.2.5.15865 increases Ptrack to 0.42 in my testing. Was previously .22

 

http://forums.eagle.ru/showpost.php?p=1 ... tcount=720

 

Kuky testing. Changing the RCS of the planes improves tracking.

 

http://forums.eagle.ru/showpost.php?p=1 ... count=1137

 

Patch 1.2.7 Ptrack improves to .53. I haven't tested it formally since then.

 

http://forums.eagle.ru/showpost.php?p=1992715&postcount=1138

 

Exorcet testing ranges.

 

Note that in all of the above testing I am testing tracking of the missiles only. Not kills. Kinetic range is not the issue. As long as it tracked well I considered that a success. Also In all situations this was a head on engagement with the shooter below the target and the target is not notching.

 

I now realize that that ground clutter and look down shooting also has a disastrous effect on ER tracking. I just haven't quantified it yet. This will be my next avenue of testing.

[Falcon_S]

...ground clutter

We have also a problem if the target is taller and run down quickly. ER go ahead ballistic and not try to go for it. Looks like target escaped from visual field of missiles and the only explanation for that is poor mobility of missile. In addition to bad guidance the problem is and speed too. If the missile was fast enough target should be invest much more effort in avoidance maneuver. Currently small target movements make missiles to miss.

[*Rage*]

Id love if ED would say whether they intend to work on seeker logic at some point.

[LowLee]

Id love if ED would say whether they intend to work on seeker logic at some point.

+1

We need a petition on change.org for this!

[Rangi]

Id love if ED would say whether they intend to work on seeker logic at some point.

 

I'm sure they intend to work on it, it's just a question of when.

[Pilotasso]

I leave again an appeal that missile sensor and flight models (aircraft ones as well) are in urgent need of fixing/tuning. It's been a looooong wait without flying. :(

[NFI_841bird-29]

inconceivable AIM120B

 

These screenshots were ACMI at the 104th server yesterday, the player launched AIM120B, shot down the first plane immediately out of the 50 nm away, then the same way to shoot down the another plane.

[GGTharos]

Sure, problems happen. Not the first time, can easily be caused by network issues.

[AtaliaA1]

Can U say "Beta"?

 

I don't see how anyone can have an issue with this.

 

I still have and use 1.2.7 the one I have the least objections to. Allowing me to use the "Beta" at my discretion. This was and is the intent of releasing a Beta and a allowing other ver. to be used. Well I guess it is?

[NFI_841bird-29]

Sure, problems happen. Not the first time, can easily be caused by network issues.

 

all caused by network?

[GGTharos]

Yes, easily caused by network issues. Packet loss to the server means important messages sent from that client do not make it to you, and missile trajectory and other processing is done on the computer that 'launched' the missile, ie. your opponent's computer.

 

It's perfectly possible for a network issue to make things look strange.