What are Heatblurs/ED’s Plans for the AIM-54?

[THE KING]

   There seems to be an awfully large amount of information, sometimes conflicting, out there regarding future plans for the Aim-54 and how it should work. What is truly the game plan for this missile moving into the future of DCS? What is currently broken now and needs to be fixed? How close are we to getting a finished product?
 

From what I can gather the two biggest problems plaguing this missile now are, 1. It’s very low tolerance of chaff and 2. Unnecessary and violent guidance corrections that bleed off tremendous amounts of energy.

Edited December 26, 2020 by THE KING

[Callsign JoNay]

3 hours ago, THE KING said:

From what I can gather the two biggest problems plaguing this missile now are, 1. It’s very low tolerance of chaff and 2. Unnecessary and violent guidance corrections that bleed off tremendous amounts of energy.

 

 

I'm seeing the same two things. My PK% versus AI is ~10% in the current build of the open beta for those two reasons you've listed.

[BreaKKer]

21 hours ago, THE KING said:

   There seems to be an awfully large amount of information, sometimes conflicting, out there regarding future plans for the Aim-54 and how it should work. What is truly the game plan for this missile moving into the future of DCS? What is currently broken now and needs to be fixed? How close are we to getting a finished product?
 

From what I can gather the two biggest problems plaguing this missile now are, 1. It’s very low tolerance of chaff and 2. Unnecessary and violent guidance corrections that bleed off tremendous amounts of energy.

 

Chaff tolerance, I’ve seen, is pretty damn good on the 54C. The guidance problems are an ED thing. I’ve seen the guidance to be getting better and better. 

[Doc3908]

I was hesitant to bring this up in this forum since I'm a relative newbie, but I'm glad I found this thread. I've been trying to work out missile defense strategies that would allow me to get close to any opponent. After figuring out how to deal with the dreaded AIM-120Cs using different airframes, I thought I'd up the challenge and try to learn to defeat the AIM-54 since it's supposed to be very difficult to defeat. To my surprise, I found that the AIM-54s are almost trying to avoid hitting me. Take a look at the Tacview in my video below (there are three engagements with Tacview after each one). Aside from the stupid AI firing an AIM-54 from 1.25km away (min range should be 2km, I believe), even the long range shots (from 25km and 8km away) are missing me when they they should easily be tracking. It's frustrating to me because I can't work on a strategy to defeat a missile of the missile is not even trying to hit me. I find the AIM-120C much more challenging and, therefore, more rewarding to defeat. There's no sense of accomplishment if the F-14 opponent fires an AIM-54 at me and the missile essentially defeats itself. I fly single player almost exclusively, so I imagine AIM-54s are much more deadly when fired by a human opponent, but still, I would love to see the missile performance improved in SP.

 

 

[Skysurfer]

Has been like this and changing for almost a year now. Maybe they'll fix it maybe they won't, I've personally given up hope at this point if such a basic task is taking so long.

[Naquaii]

22 hours ago, THE KING said:

   There seems to be an awfully large amount of information, sometimes conflicting, out there regarding future plans for the Aim-54 and how it should work. What is truly the game plan for this missile moving into the future of DCS? What is currently broken now and needs to be fixed? How close are we to getting a finished product?
 

From what I can gather the two biggest problems plaguing this missile now are, 1. It’s very low tolerance of chaff and 2. Unnecessary and violent guidance corrections that bleed off tremendous amounts of energy.

 

 

Honestly? Apart from what we say ourselves you have to regard any rumours or information as conjecture. There has been a couple of threads discussing the AIM-54 which we have followed but we seldom comment in those threads as a lot of it is speculation or just not possible to model in DCS or maybe any simulator.

 

In any case, what sets our AIM-54 apart from other DCS missiles is that we had a rather exhaustive modelling done of the missile itself which then informed our modelling in DCS. But it is still a lua defined missile just as any other DCS missile like how the AIM-7, AIM-120 (before their changes) or Super 530 are. What we can tune is the flight model parameters, motor performance and basic chaff resistance. This in effect makes our AIM-54 the same as many other DCS missiles just that we could base those parameters on our modelling.

 

The other thing that sets our AIM-54 apart is that ED helped us make it possible to control the lofting and seeker activate/seeker state by calling functions in the already existing API, it never was a completely new "missile API". This is what was finally introduced later autumn this year and made it possible for us to have the AIM-54 purely SARH in PD-STT and to control at what point it would go active in our code. Additionally it allowed us to turn on or off the lofting based on range or ACM modes to be closer to real life. This does however mean that AI F-14s might not use correct lofting or seeker functionality as we can only control this for player aircraft. So AI STT shots will still behave like ARH missiles as an example.

 

So as it currently stands our AIM-54s are in a stage where we are at a limit of what we currently can do and we're mostly doing bug fixing atm. In regards to chaff resistance this has unfortunately changed a couple of times lately and we're still trying to tweak it until we're happy with it.

 

Missile flight performance outside of pure FM stuff and guidance and seeker performance as it is is completely out of our hands (other than what has been mentioned) and that's just how it is in DCS currently.

 

That said we do know that ED are making improvements to stuff like this and we are looking at changing over to use the same new missile flight model parameters as the improved AIM-7 and AIM-120 currently do but it will take a lot of work as we will now have to retune it all and we're also not sure how much it will improve the missiles. We're also have no information as it stands regarding how this will affect missile guidance in any way or if it will even do that.

 

We will ofc continue to improve things working with ED and improve the missiles in any way we can in the future if the opportunity arises but as it stands currently it's mostly bugfixes and finetuning that's happening.

 

I hope that helps you somewhat!

 

5 minutes ago, Skysurfer said:

Has been like this and changing for almost a year now. Maybe they'll fix it maybe they won't, I've personally given up hope at this point if such a basic task is taking so long.

 

I'm sorry you feel this way but I can assure that it is anything but "a basic task".

Edited December 27, 2020 by Naquaii

[THE KING]

5 hours ago, Naquaii said:

In regards to chaff resistance this has unfortunately changed a couple of times lately and we're still trying to tweak it until we're happy with it.

 

So right now, the only thing that an aircraft needs to do that is under attack by an active aim-54, is put the missile on the beam and drop 1 piece of chaff. Do this and the missile is trashed. This has made it nearly impossible to score any kills against both AI and Human Opponents. This also makes it virtually impossible to hit larger targets that are beaming such as an AWACS, Tanker, Transport, etc. As you mentioned in another post, the Aim-54 does quite well against Its targets once it goes active. In game right now this is not the case and mostly has to do with the low chaff resistance of the missile. I could see an A version getting spoofed easily but I would think the C could handle it much better. I hope this is considered in the future. 

Edited December 28, 2020 by THE KING

[TLTeo]

Maybe people from HB can answer this, but is missile guidance and radar modelling the same for AI and player aircraft? I'm asking because for LGBs, it certainly isn't. I found that you can set an aircraft with no TGP (say, a Mirage) to attack a bunch of ground target with LGBs with nothing lasing, and all the bombs will land perfectly on each vehicle. That makes me suspect something dodgy may be happening with AI missiles as well.

[Naquaii]

11 hours ago, THE KING said:

So right now, the only thing that an aircraft needs to do that is under attack by an active aim-54, is put the missile on the beam and drop 1 piece of chaff. Do this and the missile is trashed. This has made it nearly impossible to score any kills against both AI and Human Opponents. This also makes it virtually impossible to hit larger targets that are beaming such as an AWACS, Tanker, Transport, etc. As you mentioned in another post, the Aim-54 does quite well against Its targets once it goes active. In game right now this is not the case and mostly has to do with the low chaff resistance of the missile. I could see an A version getting spoofed easily but I would think the C could handle it much better. I hope this is considered in the future. 

 

 

Well, 1 chaff does not equal a 100% trashed missile, embellishing and making up facts is not a good way to get me to listen to you. That aside our intention has always been that the AIM-54C should be close to the AIM-120B in chaff resistance and the AIM-54A about as much less from the -C as the AIM-120B is from the AIM-120C or less. This is what we have been tuning for but several times we have done this and the way the chaff resistance parameter works have changed from under our feet just as the patch was released. We are working on this but you have to understand that as 3rd party devs we do not always have the complete picture.

 

1 hour ago, TLTeo said:

Maybe people from HB can answer this, but is missile guidance and radar modelling the same for AI and player aircraft? I'm asking because for LGBs, it certainly isn't. I found that you can set an aircraft with no TGP (say, a Mirage) to attack a bunch of ground target with LGBs with nothing lasing, and all the bombs will land perfectly on each vehicle. That makes me suspect something dodgy may be happening with AI missiles as well.

 

Most likely not, apart from the seeker and lofting control that will not work for AI aircraft I wouldn't be suprised if it works differently but it's not something we 3rd parties have insight into.

[Skysurfer]

20 hours ago, Naquaii said:

 

I'm sorry you feel this way but I can assure that it is anything but "a basic task".

 

 

PN algorithms and a simple seeker logic can be found in various books. The math is all there and other games like BMS have used this to great success. What I meant by "basic task" wasn't a knock on you guys as I'm sure you could knock it out of the park if you had the full control over it - what I meant by that is that you guys have to now rely on ED's "API" which doesnt seem to take any finished state for over a year now and causing more troubles than there were before the missile API rework. The Phoenix and AMRAAM experience was way more consistent and at least in line with expected values before the rework, obviously not taking into account the different seeker guidence states and logic. Would really love to know who or how many people are working on the missile API at ED.

 

I think HB should be made the exception and allowed to develop/use their own custom code for the Phoenix since it's a pretty unique and complex missile.

Edited December 28, 2020 by Skysurfer

[Fri13]

3 hours ago, Naquaii said:

This is what we have been tuning for but several times we have done this and the way the chaff resistance parameter works have changed from under our feet just as the patch was released. We are working on this but you have to understand that as 3rd party devs we do not always have the complete picture.

 

I don't know how the co-op works between ED and 2nd party studios, but I would have thought that in the licensing agreements the ED would have opened up the API so well for the studios that they can see the code even how things really works behind them. And then when someone is going to touch the code from ED side, it is flagged as changed for everyone else so 2nd party can go and check their code against changes. And these changes are not suppose to happen couple days before release or without a memo from the corresponding maintainer so if there is a change, then it is few days time given for change modules code if required.

 

I am so custom to open source methods where such kind problems don't happen so easily or often, as everyone can stay up in speed that what happens and where. Even when the maintainer has the last word to say, it is easier to work around programming part far more easily than the human relationship problems where one just denies something without good reason.  And that is when it becomes politics, lobbying and selling new features, methods, ideas etc instead just showing a code that somethings are just better as such. And finally if maintainer doesn't still like and is alone against everyone else, there is option to make a fork and just maintain it and let the users decide what to use.

 In this case it would be that Heatblur would have made own AIM-54 code and ignore ED side and make it work...

[Fri13]

22 hours ago, Doc3908 said:

To my surprise, I found that the AIM-54s are almost trying to avoid hitting me.

 

That has been case for long time for all missiles. Even the R-27 missiles will pull in the last seconds a crazy "correction" to avoid hitting the target that has been flying at steady speed and trajectory through whole missile flight phase.

 

Watching missiles guidance logic in TrackIR, the missile doesn't have a proprtional guidance enabled. The target is either leading or getting behind, and then on the last seconds the missiles can start to pull high G maneuvers like 7-9G while approaching target from rear hemisphere, and then just 50-100 meters from the target the stop pulling G's and "give up" and start flying straight.

 

A straight flying, at constant speed being target should be very easy target for almost any missile. But in DCS missiles does crazy things by themselves. And it really do make them trying to avoid you in the last couple seconds.

It is almost scary revelation watching the track replays and seeing the missiles not able perform proper interception course and simply fail. But shot from the frontal hemisphere and the change to hit is increased dramatically.

[THE KING]

6 hours ago, Naquaii said:

 

Well, 1 chaff does not equal a 100% trashed missile, embellishing and making up facts is not a good way to get me to listen to you. That aside our intention has always been that the AIM-54C should be close to the AIM-120B in chaff resistance and the AIM-54A about as much less from the -C as the AIM-120B is from the AIM-120C or less. This is what we have been tuning for but several times we have done this and the way the chaff resistance parameter works have changed from under our feet just as the patch was released. We are working on this but you have to understand that as 3rd party devs we do not always have the complete picture.

 

 

Track 1: Aim-54A Mk-60 Defeated With One Piece Of Chaff and Notch

Track 2: Aim-54C Mk-47 Defeated With One Piece Of Chaff and Notch

Track 3: Aim-120B Intercepts Me Despite Outputting a Ridiculous Amount of Chaff While Notching

1.trk 2.trk 3.trk

[Naquaii]

It's all well and good folks, no hard feelings.

 

Like I said, we will continue to tune the chaff values and have a serious look at changing over to the same FM and new functions tha the AIM-120 and AIM-7 now uses if we can and work with ED on that.

 

1 hour ago, THE KING said:

Track 1: Aim-54A Mk-60 Defeated With One Piece Of Chaff and Notch

Track 2: Aim-54C Mk-47 Defeated With One Piece Of Chaff and Notch

Track 3: Aim-120B Intercepts Me Despite Outputting a Ridiculous Amount of Chaff While Notching

1.trk 160.11 kB · 1 download 2.trk 332.98 kB · 1 download 3.trk 59 kB · 1 download

 

You don't need to post tracks to convince me that it's possible to decoy an AIM-54 with a single chaff bundle. That's possible with all DCS missiles if you're lucky and I'm sure our AIM-54s are too sensitive atm which we will ofc have a look at. What I don't agree with is the fact that you're making it out to be that it always is decoyed by a single chaff, I can easily produce a track file were it isn't. Taking selective datapoints to prove a point while ignoring the rest that doesn't agree with you is not how we produce the data that we build our modules on.

[Skysurfer]

47 minutes ago, Naquaii said:

It's all well and good folks, no hard feelings.

 

Like I said, we will continue to tune the chaff values and have a serious look at changing over to the same FM and new functions tha the AIM-120 and AIM-7 now uses if we can and work with ED on that.

 

 

You don't need to post tracks to convince me that it's possible to decoy an AIM-54 with a single chaff bundle. That's possible with all DCS missiles if you're lucky and I'm sure our AIM-54s are too sensitive atm which we will ofc have a look at. What I don't agree with is the fact that you're making it out to be that it always is decoyed by a single chaff, I can easily produce a track file were it isn't. Taking selective datapoints to prove a point while ignoring the rest that doesn't agree with you is not how we produce the data that we build our modules on.

 

If its not 1 chaff it's 2-3 chaff bundles. Point is, 99% of the time it goes for chaff from a beam/notch. Has been the case since forever now. The 120 doesn't even do this or didnt do it this reliably last time I checked. 

[dundun92]

19 minutes ago, Skysurfer said:

The 120 doesn't even do this or didnt do it this reliably last time I checked.

It does, it just takes more than 2-3 chaff, if you drop like 10-15 chaff from an mildly accurate notch youll spoof it almost 100% of the time, and im pretty sure you can bring that down if you want:

At 4:52

I do agree though, the AIM-54 is definitely nowhere even 120B levels of CCM, and if that is the intent for the C than CCM needs to be improved

[Naquaii]

Yes, and that's the point, it isn't at the correct place relative the other missiles right now and that's what we're looking at.