AIM-7MH Issues

[near_blind]

It's me, again.

 

I've been trying to test Sparrows since the synchronization of the code, and I've noticed some odd behavior with the AIM-7MH that seems to be unique to the F-14.

 

Taking shots at ~20 miles against a maneuvering target with no CM, this missile will consistently miss despite ostensibly not being notched and having the speed and battery life to make the cut off for intercept.

 

Generally speaking, the engagement would proceed such that the missile begins it's terminal steering at ~6NM from the target. At ~2NM the AI target will begin a last ditch evasive maneuver of varying intensity, but the missile will maneuver aggressively to achieve cut off despite having a speed of roughly Mach 1. Once the missile has passed the target, it will select a 6G turn in the direction of the target before going dumb. In testing this occurred just about every time.

 

This behavior doesn't seem to be present in MHs fired by F/A-18s (I haven't had the time to test F-15s). For completeness' sake I also tried shooting ED MHs from the F-14 I'm not 100% convinced they're immune, but when they weren't notching themselves they seemed less prone to this behavior.

 

Neither does the HB AIM-7M seem to be as seriously effected by this issue. Launches from a similar situation will see the -7M maneuver aggressively over the last mile to connect >66% of the time. In the remaining 33% it would behave like the MH and fail to maneuver to aggressively enough to hit the target.

 

I've got no idea what's causing this, with the exception of the autoloft setting on the -7M and -7MH, y'alls code is identical to the ED Sparrows.

HB MH Miss 1.trk

HB MH Miss 2.trk

HB MH Miss 3.trk

HB M Miss 1.trk

HB M Hit 1.trk

HB M Hit 2.trk

[captain_dalan]

The launches were made in PD or P STT? I want to try this as well. Cause i actually tested both missiles at shorter ranges (10-15NM) and they seam to work the same

[captain_dalan]

Just did a few quickies. I can't seam to find any significant difference.

AIM-7M P-STT.zip

AIM-7MH PD-STT.zip

AIM-7MH P-STT.zip

[KlarSnow]

Go repeat it in a hornet, in our tests, same scenario, no countermeasures, and accounting for kinematic defeats and dropped locks, the hornet with an AIM-7MH will connect 7/8 times. Repeat that with the tomcat and we only had 1 connect.

 

Of note these are all longer shots 20-15 miles, the missile has the kinematics and ability to make that happen. Up close where the motor is still burning there doesnt seem to be any issue, shots sub 10 miles seem to work just fine.

 

The issue is that in the hornet the AIM-7MH is a reliable BVR missile. In the tomcat, in perfect conditions it is not.

[near_blind]

The launches were made in PD or P STT? I want to try this as well. Cause i actually tested both missiles at shorter ranges (10-15NM) and they seam to work the same

 

Perhaps the extra 5-10 miles makes the difference? I attached a collage of the tacviews from my track generation. The common behavior between them seems to be making a 6G last ditch turn towards the target about 2NM after it could have done anything, and the falling helplessly outside the targets turn circle. Shots were taken in both Pulse and Doppler STT.

 

We eliminated chaff as explanation. I don't believe the target notched the missile seeker as they don't unload to one G and glide. The missiles have not exceeded their battery life, and they are still locked by the launching platform.

[Sonoda Umi]

The issue is that in the hornet the AIM-7MH is a reliable BVR missile. In the tomcat, in perfect conditions it is not.

 

Actually, both AIM-7M and MH are not reliable BVR missile for F-14. This issue exists a long time.

[KlarSnow]

We are controlling for the loft/non loft differences. Saw the same behavior differences for the same range shots unlofted between the hornet and the tomcat with the AIM-7MH. IE we uncaged and removed the LOFT cue on the hornet and took a bunch of shots at the same ranges we were in the tomcat.

 

Same disparity in hit results, with the hornet missiles connecting greater than 3/4 the time and tomcat missiles connecting less than 1/4 the time, and yes accounting for target maneuvers and dropped locks in there as well.

 

We only counted as a "miss" what had a chance to connect with the target, radar held lock all the way to terminal intercept, bandit didn't maneuver to cold/ terminal intercept occured with sufficient (~1mach or greater) energy left on the missile for it to aggressively maneuver to fuze/impact. Anything that fell outside of these parameters we are not counting.

We also removed Chaff from all targets to remove that variable.

 

We were testing against Veteran AI, medium altitudes (20,000 feet for targets and shooters) in MP, both with Jester and a human RIO, and in SP and saw all of these results mirrored.

Edited October 4, 2020 by KlarSnow

[Sonoda Umi]

We are controlling for the loft/non loft differences. Saw the same behavior differences for the same range shots unlofted between the hornet and the tomcat with the AIM-7MH. IE we uncaged and removed the LOFT cue on the hornet and took a bunch of shots at the same ranges we were in the tomcat.

 

Same disparity in hit results, with the hornet missiles connecting greater than 3/4 the time and tomcat missiles connecting less than 1/4 the time, and yes accounting for target maneuvers and dropped locks in there as well.

 

We only counted as a "miss" what had a chance to connect with the target, radar held lock all the way to terminal intercept, bandit didn't maneuver to cold/ terminal intercept occured with sufficient (~1mach or greater) energy left on the missile for it to aggressively maneuver to fuze/impact. Anything that fell outside of these parameters we are not counting.

We also removed Chaff from all targets to remove that variable.

 

We were testing against Veteran AI, medium altitudes (20,000 feet for targets and shooters) in both MP, both with Jester and with a human RIO, and in SP and saw all of these results mirrored.

Thanks for the experience.

 

This is a critical subject which need to be solved or F-14 will lose most of long-range attack capabilities under the circumstances without AIM-54 - that is unacceptable and not matches the reality.

[captain_dalan]

Go repeat it in a hornet, in our tests, same scenario, no countermeasures, and accounting for kinematic defeats and dropped locks, the hornet with an AIM-7MH will connect 7/8 times. Repeat that with the tomcat and we only had 1 connect.

 

.....

 

I'm not saying the missiles don't perform better in the Hornet, just that there doesn't seam to be a noticeable difference between the M and MH variants in the F-14. I find them to perform practically identically. This was what i tested, i didn't get to try them out in the F-18.

 

......We were testing against Veteran AI, medium altitudes (20,000 feet for targets and shooters) in MP, both with Jester and a human RIO, and in SP and saw all of these results mirrored.

Ah ok, now we can standardize the tests. I was using the Ace AI and that maybe skewering my results as well. I'll try and reproduce with the new parameters in mind in both planes. :thumbup:

[captain_dalan]

Sorry for such a late response, but it was only last night that i finally got to play with DCS a bit. Unfortunately, tacview doesn't want to cooperate with DCS after the last update, so no tracks and no bandit monitoring this time :(

 

 

Just like in the previous patches, there doesn't seam to be a noticeable difference in performance of the AIM-7's when fired from either plane. The bandit is a veteran F-15C with no CM loaded. 20NM shots are in range, but will never connect to a maneuvering target. Sub 15NM shots are possible in both planes. Only sub 10NM shots (at this altitude 15-20000ft) can be considered kill shots with any degree of certainty. Tested both AIM-7M and AIM-7MH. No considerable difference here either. Lofting the missile doesn't really help it in this case, but i tested even that. Are you sure you are on OB?

 

Here is a previous video i made, with the default Constant Peg mission. Just to confirm the observations from a few patches away.

 

[KlarSnow]

K take a look at these.

 

These are the cleanest examples of the disparity I can find in my tacviews

 

Hornet AIM-7MH Loft ON

 

HOrnet AIM-7MH Loft OFF

 

The first two are hornets with AIM-7MH's, first Loft on shooting at ~20 miles, greater than 50% hit, accounting for dragging bandits, more than 75% hit. This is how Sparrows were before they got borked across the board around the supercarrier release. We found that bug which still appears to be present in the tomcat sparrows, reported it to ED and they fixed it. It is still present but only on the tomcat as exemplified by the Tomcat tacview below.

 

Tomcat AIM-7MH Miss

 

Note they are shot at 12 miles, both bandits while maneuvering continue hot and don't do anything more violent than what the hornets targets did, the bandit in the west just does a lazy 2G roll while pointing straight at the missile, and both sparrows which should have the bandits dead to rights barrel roll around targets endgame. That's the issue. This is what we reported to ED ~4 months ago and got fixed for all sparrows, and it is still present for tomcat AIM-7MH's.

Edited October 10, 2020 by KlarSnow

[captain_dalan]

K take a look at these.

 

These are the cleanest examples of the disparity I can find in my tacviews

 

Hornet AIM-7MH Loft ON

 

HOrnet AIM-7MH Loft OFF

 

The first two are hornets with AIM-7MH's, first Loft on shooting at ~20 miles, greater than 50% hit, accounting for dragging bandits, more than 75% hit. This is how Sparrows were before they got borked across the board around the supercarrier release. We found that bug which still appears to be present in the tomcat sparrows, reported it to ED and they fixed it. It is still present but only on the tomcat as exemplified by the Tomcat tacview below.

 

Tomcat AIM-7MH Miss

 

Note they are shot at 12 miles, both bandits while maneuvering continue hot and don't do anything more violent than what the hornets targets did, the bandit in the west just does a lazy 2G roll while pointing straight at the missile, and both sparrows which should have the bandits dead to rights barrel roll around targets endgame. That's the issue. This is what we reported to ED ~4 months ago and got fixed for all sparrows, and it is still present for tomcat AIM-7MH's.

 

I just did :(

Man, that was totally awkward, i can't believe we are flying the same sim. I routinely get 10NM kills, even against chaffing targets against the AI. Even at lower altitudes. Something is very weird here. I haven't tried it in a MP environment though. I usually fly only dogfights when i go online.

[captain_dalan]

Edit: for some reason i get an invalid file report when i try to upload the Tacview files (i finally got it to work again). Any way around this?

Edited October 11, 2020 by captain_dalan

[NeedzWD40]

As of the latest patch, this issue plagues not only the AIM-7MH, but also the AIM-7F. The AIM-7M exhibits some similar issues but in testing seems to hit more consistently.

 

In essence, it seems like the 7F and 7MH are aiming at some point behind the targeted aircraft and are either maneuvering too late or not enough. I tested against IL-76s set to no reaction, no chaff or flare, and ECM off. First shots at 25nmi, second at 19nmi, and third at 10nmi head on. All three shots with the F and MH fall behind the target, as if they're lagging behind somehow. Ms will have one of the three hit. Switching to boresight mode and shooting tail-on results in missiles shooting up and behind the target. Unfortunately my tracks go off the rails after the turn so they're only marginally good at showing the head-on results.

 

F-14A AIM-7MH Failure.trkF-14A AIM-7M Success.trkF-14A AIM-7M Success 2.trkF-14A AIM-7F Failure.trk

[Reflected]

AIM-7Ms seem to go stupid pretty often.

[Skysurfer]

It's kind of a bummer since Sparrow's used to work perfectly fine and according to RL charts at one point. Then ED decided to mess with missiles again. Having broken missiles for almost a year now in a combat simulator really blows, not gonna lie.

Edited December 18, 2020 by Skysurfer

[Sonoda Umi]

2 hours ago, Skysurfer said:

It's kind of a bummer since Sparrow's used to work perfectly fine and according to RL charts at one point. Then ED decided to mess with missiles again. Having broken missiles for almost a year now in a combat simulator really blows, not gonna lie.

 

I agree.

I have abnormal BVR/long-range A2A Attack experience almost a year. Saddly, in this year, I nearly withdraw the radar-guided missiles from my birds, just conducting Carrier Qualifications, Aerial Refueling and BFM gun only training...

Edited December 18, 2020 by Sonoda Umi

[near_blind]

Just to add more evidence to the pile:

 

The MH is doing the late stern conversion almost every shot at medium range. The M seems to be doing it more than it did, but is still fairly reliable. I haven't done any tests with the F. 

 

M Miss.trk

 

 

M Multi Miss.trk

 

MH Miss Close Loft.trk

 

MH Multi Miss 2.trk

 

MH Multi Miss.trk

 

MH Quatro Miss.trk

 

On a side note, @IronMike, are SP tracks alright or are MP tracks preferred for this sort of thing due to the desync?

[IronMike]

59 minutes ago, near_blind said:

On a side note, @IronMike, are SP tracks alright or are MP tracks preferred for this sort of thing due to the desync?

 

I dont think either work well unfortunately, to visualize it, likely tacview shows most. This all however points to ED changing the aim7s and our code being out of date, although we recently checked, weired. We will investigate again, thanks.

[near_blind]

5 hours ago, IronMike said:

I dont think either work well unfortunately, to visualize it, likely tacview shows most.

 

Can do chief, Here are the tacviews from the tracks I posted today:

M Miss

M Multi Miss

MH Miss Close Loft

MH Multi Miss 2

MH Multi Miss

MH Quatro Miss

 

Here are another three from this weeks patch I didn't take the time to save the tracks for

https://drive.google.com/file/d/1aHhO8_1OoUkUf_vNO77_niEQ8UpwLLL-/view?usp=sharing

https://drive.google.com/file/d/1Mfxoky8VDnvhax8cA0y4ErnIRHovhuTV/view?usp=sharing

https://drive.google.com/file/d/1xfthlvQdbstAG2QEToWQlsXBLv-46Anb/view?usp=sharing

 

5 hours ago, IronMike said:

This all however points to ED changing the aim7s and our code being out of date, although we recently checked, weired. We will investigate again, thanks.

 

I appreciate everything the team has done with regards to Sparrows since release, you all have been responsive when issues were raised. I am also aware that right now the AIM-7s are going to be in flux for a while as ED is changing things with the missile API. I would just like to try and clarify that since the APN and classname issue was "fixed" in May, this issue has affected the HB AIM-7s and has never been remedied. You all have been extremely kind taking the time to refactor the Sparrows defined in the F-14s weapons.lua to match EDs changes, repeatedly, but even when the missile definitions were carbon copies the issue persisted in the HB AIM-7, not the ED one. In no way am I trying to make any accusations that you guys have done anything wrong, just that the steps to date have been ineffective and it seems to me (admittedly a layperson and an outside observer) that the issue dwells somewhere deeper than the lua layer of the simulation. 

 

Thanks again for taking the time to look at these threads, and happy holidays. 

[captain_dalan]

On 12/18/2020 at 5:19 PM, Skysurfer said:

It's kind of a bummer since Sparrow's used to work perfectly fine and according to RL charts at one point. Then ED decided to mess with missiles again. Having broken missiles for almost a year now in a combat simulator really blows, not gonna lie.

 

Indeed, tried the GENIII Constant Peg mission this afternoon. Not a single Sparrow made it. And with the 54's having the issues of their own, we are restricted to guns and Winders only. Dogfighting is still hampered with performance issues though, possible store drag related.

[Skysurfer]

8 minutes ago, captain_dalan said:

Indeed, tried the GENIII Constant Peg mission this afternoon. Not a single Sparrow made it. And with the 54's having the issues of their own, we are restricted to guns and Winders only. Dogfighting is still hampered with performance issues though, possible store drag related.

 

And the fact I can barely break Mach 1 in a clean Tomcat at angels 30 

[captain_dalan]

8 hours ago, Skysurfer said:

 

And the fact I can barely break Mach 1 in a clean Tomcat at angels 30 

I haven't gone that high up yet....
Actually, scratch that. I did, but it was in the modified Protect the Vicksburg mission, but  i wasn't really burning up there, just cruising and loitering. I did notice the B struggling to break mach 1 with 4 Sparrows and 4 Winders down low this afternoon though. A 2g bank was enough to make her subsonic. Things are long way from finished it would seam.

But i haven't tested a clean birds yet (decided to stop flying back in November until major reported issues were resolved), only laden one. So maybe my observations steam from over modeled stores drag?

[Golo]

Same here, its still same as in previous update, when I wrote first feedback. I still can barely get A above mach in more or less level flight with about 13k lbs fuel, 2x9s, 2x7s (st. 4,5), 2xXT(empty).

Its either a bug or 14s are really shait getting past big m. 

Edited December 20, 2020 by Golo

[MBot]

*Forum just ate all the text I wrote and only posted pictures, will try again*

Edited December 20, 2020 by MBot