[RESOLVED] more air combat issues

[Jmsimer]

Ah, here's the one I was looking for. This is right as the engagement is starting. The incident is occurring SE of Tromso. I've used AWACs to bring in groups 4261, 4262, and 4266 in alongside to slightly behind the MiGs, F16s' radars off to avoid detection. The F16s started at low altitude. When this autosave was made, I'd already started climbing them to high altitude to see if that was preventing them from firing.

[Jmsimer]

Here's another save. Group 4352, with radars on, is going to continue flying right at these approaching MiG-29s and never fire a round. They will both get shot down without even attempting to engage. I realize it can get depressing if you're stationed up near the Arctic Circle in winter and haven't seen the sun for three months, but if they're suicidal I would have expected them to just fly into the ground!

[mikmykWS]

Try checking off investigates contacts outside the patrol zone or at least expanding the area to cover the area the aircraft are currently flying in and see if the results are better. Right now the zone you've set up is pretty small and the F-16's aren't really going engaged offensive because the targets you want them to shoot are well outside of it. I think there is a lot of push and pull going on between orders and responses which is to be expected during a complex engagement and small zone is probably the culprit.

Just an FYI. If you select a mission in the mission editor and move the window you can see the zones you selected. The game also gives warnings when you sent up strange zones (z's instead of squares etc) when you load. You have a number of those situations. When you set up missions try the cntrl right click menu define area option. You can click and drag areas and it will lay out 4 points correctly for you.

Let me know how you do.

Thanks!

Mike

[Jmsimer]

No luck. Did all of the above, F16s still fly around aimlessly and don't engage the MiGs.

[mikmykWS]

Odd worked fine when I did it.

Mike

[Jmsimer]

I got better results when I backed up 20 seconds from the autosave and tried again. Still a lot of odd behavior like one aircraft in a group flying toward the enemy without engaging while the other turns away (both have full weapons and plenty of fuel), but I was at least able to get a roughly even exchange with the MiGs.

[HaughtKarl]

I fired up the game after being away from it for several months and I've noticed the same thing with regards to air to air combat. I'll setup CAP zones with 'investigate contacts outside patrol area' checked and more often than not my F4's will get casually gunned down by a J-6 that mozied on over while the other CAP planes ignored it. It defiantly feels like something is off because I can't recall my units acting in this manner back when I regularly played the game.

[Jmsimer]

Been much better for me with mikmyk's recommendations. It just didn't work in that first game... for some reason I had to back up to an earlier autosave and then make the changes, and things were better. I still get goofy behavior occasionally but it's much more manageable, and if I restore from an autosave the goofiness usually won't repeat.

[Dimitris]

ORIGINAL: HaughtKarl
I fired up the game after being away from it for several months and I've noticed the same thing with regards to air to air combat. I'll setup CAP zones with 'investigate contacts outside patrol area' checked and more often than not my F4's will get casually gunned down by a J-6 that mozied on over while the other CAP planes ignored it. It defiantly feels like something is off because I can't recall my units acting in this manner back when I regularly played the game.

Can you provide a save file? We cannot investigate anecdotes.

[Dimitris]

ORIGINAL: Jmsimer
Ah, here's the one I was looking for. This is right as the engagement is starting. The incident is occurring SE of Tromso. I've used AWACs to bring in groups 4261, 4262, and 4266 in alongside to slightly behind the MiGs, F16s' radars off to avoid detection. The F16s started at low altitude. When this autosave was made, I'd already started climbing them to high altitude to see if that was preventing them from firing.

Just tested this.

* First attempt to order Group 4266 to engage the Flankers: The Falcons had still a few seconds on their OODA loop before they could begin attempting to engage. (The absolute best way of quickly learning why a platform will not fire: Try a manual weapon allocation and see the feedback. We state this in the manual too IIRC.)

* Second attempt, after the OODA countdown finished (you can watch the number if you enable targeting vectors): The AIM-9s would not detect the Flankers. Quick check at relative altitudes and weather: Falcons at ~10K feet and climbing, Flankers at 36K feet, light clouds between 20-23K feet. The clouds are blocking the AIM-9 seekers from acquiring.

* Once the Falcons cleared the cloud layer, I didn't even have to order them to attack: They acquired and began firing immediately.

Looks like WAD.

[Dimitris]

ORIGINAL: Jmsimer
Here's another save. Group 4352, with radars on, is going to continue flying right at these approaching MiG-29s and never fire a round. They will both get shot down without even attempting to engage. I realize it can get depressing if you're stationed up near the Arctic Circle in winter and haven't seen the sun for three months, but if they're suicidal I would have expected them to just fly into the ground!

Just tested this. Starting from the provided save, the F-16s have a 15-sec OODA delay which prevents them from engaging immediately (you can see this if you enable targeting vectors). Once the countdown reaches zero they do begin firing their weapons.

I would suggest playing a few smaller scenarios and getting familiar with the detect-track-prosecute killchain.

[Dimitris]

Is any of the OP reports still an issue ?

[Ancalagon451]

Number 3 is very much alive and well. In fact not only shots are taken against a contact "teleporting" within range due to uncertainty, but weapons fired against uncertain ground contacts change trayectorie to follow the uncertain contact even when the target jumps a hundred miles or more.

This generates ridiculous results when you lose the weapons quality sensor track while the missile is flying and the target starts "dancing" due to uncertainty. The missile tries to follow the new position and goes totally off target. This often happens when you fire a ARM missile against a radiating SAM, if it stops radiating during the flytime and you only have a bad surface search radar tracking it the missile goes totally off target.

It's in fact better have no sensors than bad sensor in this case since if you merely lose the contact a memory equipped ARM (or any inertial-GPS weapon) keeps attacking the last known position and can still score a hit.

Ancalagon

[Scorpion86]

ORIGINAL: Ancalagon451
This generates ridiculous results when you lose the weapons quality sensor track while the missile is flying and the target starts "dancing" due to uncertainty. The missile tries to follow the new position and goes totally off target. This often happens when you fire a ARM missile against a radiating SAM, if it stops radiating during the flytime and you only have a bad surface search radar tracking it the missile goes totally off target.

I can confirm this. Playing Crimea River, I lost many a good HARM this way.

[Dimitris]

ORIGINAL: Ancalagon451

Number 3 is very much alive and well. In fact not only shots are taken against a contact "teleporting" within range due to uncertainty, but weapons fired against uncertain ground contacts change trayectorie to follow the uncertain contact even when the target jumps a hundred miles or more.

This generates ridiculous results when you lose the weapons quality sensor track while the missile is flying and the target starts "dancing" due to uncertainty. The missile tries to follow the new position and goes totally off target. This often happens when you fire a ARM missile against a radiating SAM, if it stops radiating during the flytime and you only have a bad surface search radar tracking it the missile goes totally off target.

It's in fact better have no sensors than bad sensor in this case since if you merely lose the contact a memory equipped ARM (or any inertial-GPS weapon) keeps attacking the last known position and can still score a hit.

Ancalagon

Is there a suitable pre-fire save file we can investigate? Thanks!

[Dimitris]

ORIGINAL: Scorpion86

ORIGINAL: Ancalagon451
This generates ridiculous results when you lose the weapons quality sensor track while the missile is flying and the target starts "dancing" due to uncertainty. The missile tries to follow the new position and goes totally off target. This often happens when you fire a ARM missile against a radiating SAM, if it stops radiating during the flytime and you only have a bad surface search radar tracking it the missile goes totally off target.

I can confirm this. Playing Crimea River, I lost many a good HARM this way.

Is there a suitable pre-fire save file we can investigate? Thanks!

[Ancalagon451]

ORIGINAL: Dimitris

Is there a suitable pre-fire save file we can investigate? Thanks!

Well, I'll be damned.

I've been trying for an hour to make you a testbed and I haven't been able to reproduce the conduct. Which annoys me to hell and back because I'm SURE I've seen the issue in action but I can't generate hard proof so, unless Scorpion86 manages to do it, I'll have to retract my words here.

If I ever see the bug again I'll make a propert report. Until then sorry for the bother.

Ancalagon

[ARCNA442]

ORIGINAL: Dimitris

Is any of the OP reports still an issue ?

I believe that 1, 2, and 3 are still issues (though perhaps not as bad as they seem to have been when the OP posted 4 years ago) but I don't currently have any saved games to prove it.

With point 1, aircraft in turning fights appear to fire on the first enemy to get within boresight, which can result in them prioritizing distant targets rather than more dangerous close targets.

For point 2, it seems that with both aircraft and ships, if multiple weapon types are available the AI defaults to the longest-ranged weapon rather than the closest shooter. In a recent scenario I played Hornets with AMRAAM were deferring to Tomcats with Phoenix even when the Hornets were much closer to the enemy. In the past I also have seen ships with short range SAM's hold their fire against incoming missiles to let ships with long range SAM's engage even though the long range SAM's cannot possibly intercept the missiles before they reach the ship with short range SAM's. However, in both cases the shorter ranged weapons will eventually be employed at point blank range.

(another, possibly related point is that planes with combinations of Phoenix/AMRAAM/Sidewinder will often fire their longest-ranged weapons even when they are within range for their shortest-ranged weapons - this can be infuriating because it wastes the more valuable weapons and causes the planes to RTB to reload instead of staying on station).

Point 3 definitely still happens with antiradiation weapons on ESM contacts.


I'll see if I can create a test scenario to prove some of these.

[Der Zeitgeist]

ORIGINAL: Dimitris

ORIGINAL: Scorpion86

ORIGINAL: Ancalagon451
This generates ridiculous results when you lose the weapons quality sensor track while the missile is flying and the target starts "dancing" due to uncertainty. The missile tries to follow the new position and goes totally off target. This often happens when you fire a ARM missile against a radiating SAM, if it stops radiating during the flytime and you only have a bad surface search radar tracking it the missile goes totally off target.

I can confirm this. Playing Crimea River, I lost many a good HARM this way.

Is there a suitable pre-fire save file we can investigate? Thanks!

You can maybe use the save file from my bug report on the same issue back in 2014:

tm.asp?m=3564688&mpage=1&key=&#3564688

[Dimitris]

Hi,

ORIGINAL: ARCNA442
With point 1, aircraft in turning fights appear to fire on the first enemy to get within boresight, which can result in them prioritizing distant targets rather than more dangerous close targets.

I can see how that can be a problem, but I'm sure you can also easily imagine the backlash if we change that behavior: "My turning aircraft had A BLOODY CLEAR SHOT at this more distant but incoming threat, and instead it just kept turning with that nearby MiG that was not an immediate threat to him. In air combat if you have a shot against any of the bad guys YOU BLOODY TAKE IT. Just how dumb are the pilots in this game?"

So there is no clear-cut optimum algorithm here, it's all very case-specific.

For point 2, it seems that with both aircraft and ships, if multiple weapon types are available the AI defaults to the longest-ranged weapon rather than the closest shooter. In a recent scenario I played Hornets with AMRAAM were deferring to Tomcats with Phoenix even when the Hornets were much closer to the enemy. In the past I also have seen ships with short range SAM's hold their fire against incoming missiles to let ships with long range SAM's engage even though the long range SAM's cannot possibly intercept the missiles before they reach the ship with short range SAM's. However, in both cases the shorter ranged weapons will eventually be employed at point blank range.

I think this is WAD. You need to keep in mind what was happening before we enforced this behavior. Long-range shooters would first fire against the incoming, then short-ranged AAM fighters would close in and also fire regardless of the LR missiles already being in the air, so there was a large waste of weapons. We changed the behavior so that the short-armed fighters are aware of other shots incoming and hold their fire unless the target crosses the self-defence threshold, in which case they fire regardless. It's not a 100% solution (then again what is), but it works better than before.

(another, possibly related point is that planes with combinations of Phoenix/AMRAAM/Sidewinder will often fire their longest-ranged weapons even when they are within range for their shortest-ranged weapons - this can be infuriating because it wastes the more valuable weapons and causes the planes to RTB to reload instead of staying on station).

This is hard to examine without a save at hand. Perhaps at the time of firing the target is within _nominal_ range of the SR-AAM but outside its _effective DLZ_ at this moment? A save would help a lot here.

Point 3 definitely still happens with antiradiation weapons on ESM contacts.

I'll see if I can create a test scenario to prove some of these.

That would help a lot, thanks!