Improving response time of AA guns?

[Flashy]

Hi all,

 

Does anyone know of a way to get AA guns (either the vehicles or fixed guns, machine gunners etc) to react more quickly to an enemy aircraft in their range? I am using a force complete low MCU as soon as the gun is spawned, and am not using an AttackArea MCU, but it still takes about 8 seconds from when the unit is spawned to when it actually starts firing.. by that time the plane has already overflown the gun and is almost out of range (especially machine guns).

 

Does anyone know of any tricks to make them respond faster?

[Stonehouse]

Possibly ensure that the gun is facing in the direction the aircraft are expected from. Particularly heavy AAA guns have a reasonably slow rotation rate. Also make sure the trigger zone that activates the gun is at least 10 kms for heavy AAA and 5 km for light AAA. This latter point is quite important in my experience as it gives the gun time to rotate and aim at the incoming aircraft. 

 

An example to illustrate. A Flak 37 has a rotational speed of 9 deg per sec. It has a max elevation of 82 degrees so once the aircraft is within 8 degrees of vertical the gun must rotate likely 180 degrees as the aircraft overflies the gun before it can fire again as there is a no fire inverted "cone" above the gun.

 

Also, the AI gunners have a built-in search time - in this case 12 secs in the stock game and also a target recognition delay based on the skill of the gunner team which is 10 (high), 20 (normal) or 30 (low) secs in the stock game.

 

So even if you leave out the search time and target recognition delays, if the gun is facing away from the enemy aircraft, it will take 20secs to rotate 180 degrees to face it. In 20 secs an aircraft moving at 300mph/ 260 knots or 134 m/s will move 2680 meters. So if your gun is facing away from the enemy and you activate it based on say a 5000m trigger the gun will have very little time to fire (maybe 1 or 2 shots considering the rate of fire is 1 shot every 3 secs in stock for a Flak 37) before the aircraft overflies the gun and forces it to rotate 180 degrees again (20 secs) before it can fire on the receding enemy aircraft. Light AAA obviously has a higher rotation rate although likely still cannot fire directly overhead and the AI has an 8 sec search time with the same target recognition delays.

 

This is why I suggest having the heavy guns face the expected direction of the enemy attack (which is what they did in real life if they could) and have at least a 10 km radius on the trigger to activate the gun and at least an 11km radius on the deactivation trigger. Light AAA usually does ok with a 5000m activation and 6000m deactivation. MG style weapons can be less for activation and deactivation, but they often can fire out to 1000-1500m so should get at least 3000m and 4000m unless the gun is very short ranged.

 

A lot of stock missions and the career still use something like a 5000m radius and don't always set up the guns with a useful orientation.

 

In real life the heavy AAA was setup such that there were overlapping zones of fire where each fire zone was shaped kind of like a bunt cake and light AAA was positioned around the heavy AAA to stop low flying aircraft attacking the heavy AAA. In this way as one gun became unable to fire another had the target entering its field of fire.

 

A typical German heavy flak battery was 4 to 6 (later in the war) 88mm guns with 2 20mm guns for close in protection. A light battery was 12 20mm guns and 4 60cm searchlights or possibly 9 37mm guns and 4 searchlights. A typical mixed battalion was 3 heavy batteries, 2 light batteries plus the headquarters. Usually, the smallest deployed unit was a battery as particularly for the 88s this was the lowest level for automated fire control from the battery commander (88s were able to receive elevation and bearing and target height data etc from a central fire control). Allied flak batteries were much the same and usually were also 3-4 guns with light guns supporting.

 

Your attack area MCU radius should also approximately match the max range of the gun unless you have a specific reason otherwise as the gun will not fire further than the MCU radius. EG the Flak37 has a range of around 10000 metres, so you handicap it badly if you give it a 5000m radius attack area.

 

See here possibly as well if you wish to consider mods

 

Edited October 20 by Stonehouse

[Flashy]

Thanks for the comprehensive reply @Stonehouse, lots of very useful information in there! I should have mentioned that I am creating AA zones for WW1 planes, so the speeds are much lower and I am only using the MG-08 and Hotchkiss AA machine guns in this case. I have been using a 750m checkzone because I wanted to keep them small to make sure they were only triggered when they "should" be, but the slow response speed of the AA means I might need to go for bigger (but fewer) checkzones. I will test and report the results back here.. I will also try your AA mod and see if that improves the existing smaller checkzone logic..

[Flashy]

Tested a bigger checkzone array last night (1500m instead of 750m). Results are a bit better in terms of AI ground unit response, so might stick with that size.. The only thing I dont like is, since I am using a sphere as the "plane in" checkzone, it can still be triggered while flying over it at 1500m. This is too high for the machine guns so they just sit there and do nothing as the plane flies over.. not ideal. It seems the sphere always sits on the ground, and there is no way to move it down into the ground so it doesnt trigger so high..

 

Wish they could give us a cylinder checkzone which has a height limit (i.e. wont trigger above a certain height)..

[Jaegermeister]

Also check "Spawn at Me" in the advanced properties box of the spawner MCU and it will face towards you when it spawns, cutting down the reaction time.

[IckyATLAS]

If you create two Check Zones one of Cylindrical format and one of Spherical format centered on the same point or not. Then you can test that the plane is in both. This means that you will have created a common volume of detection that is the intersection of both geometries as long as they have an intersecting volume. Playing on Sphere and cylinder size and how they are positioned you may get interesting possibilities. You can even go further with more than two. You can the have logical operation of AND and OR. Possibilities are limitless. This may help solve your problem.

[Flashy]

18 hours ago, IckyATLAS said:

If you create two Check Zones one of Cylindrical format and one of Spherical format centered on the same point or not. Then you can test that the plane is in both. This means that you will have created a common volume of detection that is the intersection of both geometries as long as they have an intersecting volume. Playing on Sphere and cylinder size and how they are positioned you may get interesting possibilities. You can even go further with more than two. You can the have logical operation of AND and OR. Possibilities are limitless. This may help solve your problem.

Hmm that is a very interesting idea which I hadnt considered - nice! I am not sure when I would use that though, but I will think about it a a bit more - thanks for the idea!