Strafing bombers [MISC_DATA] code?

[Allen]

I'm working on my B-25J and for the life of me I can't find the right [MISC_DATA] for them to strafe. Tried category=5 and category=6 but the just do nothing and fly right by.

What I'm I missing here....

[bearcat241]

Make it a fighter--unit=1, cat=1

Although these strafers often went out in small groups, their individual roles and performances were akin to fighter bombers with low-level strafing, skip-bombing and para-bombing. If they got bounced, they mixed it up with their attackers with extreme maneuvering that gave their rear-firing and side gunners the advantage.

In the sim, like in RL, these bombers won't be able to get on the tail of an attacking fighter, but they can make any trailing fighter pay heavily in a turning fight and i don't have to tell you what an attacker's head-on pass would look like...lol

[Allen]

That was my only idea too. Wasn't sure if there was another way. Now back to ed up CFS2 AI.... They dive at the target to fire guns but never open fire.

[Allen]

I have no idea why AI only flights dive at the target to fire guns but never open fire. My wingman will just fine. No amount of anything I can think of will fix this.

[bearcat241]

I suspect the air files. Make a dedicated AI model by taking the aircraft.cfg and *.air file from the stock P-38 and transferring it to the AI model. The only thing you want to change initially in the P-38 copy is the contact points--you can change engine data and other details specific to the B-25 later. If this test works then you know its a weight/MOI problem in the medium bomber flight dynamics that the sim's AI engine can't fully resolve in strafing passes.

If a big, heavy twin-engine bomber gets too fast in a diving run (unavoidable), i'm guessing that in the AI algorithm targeting would become a second priority to survival. It could be that the AI engine can't reconcile these two factors precisely as it would with much lighter fighters, fighter-bombers and single-engine Navy bombers. So perhaps it focuses on the pull-out when the diving speed is too high and skips lining up the target and "pulling the trigger".

[Captain Kurt]

As Bearcat says, it is more likely a MOI (moment of inertia) issue, possibly coupled with the aircraft weight. The player's aircraft uses rudder, elevator and ailerons for control, but some of our gurus have said that the AI only use the trim tabs on these controls rather than direct control input. Whether this is true or not, the AI controls responses are slower than the player aircraft. This means that the AI will pull out of a strafing attack without firing or dropping bombs when it senses it will not have time to pull out. And it errs on the side of caution.

You might give these suggestions a try.

The way for the AI to overcome the timid behavior is to reduce the MOI numbers in the .air file by trial and error until the AI will complete a test strafing run. Use AirEd to open the .air file and go to the 1001 Moment of Inertia Values line to reduce the values for Roll, Pitch and Yaw. I reduce each value by the same percentage to keep control more balanced. Seems to work for me. This allows the AI controls to respond more crisply. Save and fly your test. Repeat until it works.

Try also reducing the weight in the 1100 Primary Aerodynamics / *Empty Weight (lbs) =

Now, there is a consequence to doing this as it can make the player's plane controls very squirrely unless you do the following:

The AI flies completely based on the .air file data. But the player aircraft uses the data in the aircraft.cfg file first, and then uses the .air file for data that is NOT in the aircraft.cfg file. In order to not affect your player aircraft, the original MOI values need to be in the aircraft.cfg file. If the aircraft.cfg file does not already have the MOI and weight values, type in the following under the [WEIGHT AND BALANCE] section:

empty_weight =(original value)

; moments of inertia
empty_weight_pitch_MOI =(original value)
empty_weight_roll_MOI =(original value)
empty_weight_yaw_MOI =(original value)
empty_weight_coupled_MOI=0.000

Now the player aircraft will continue to fly properly while the AI uses the reduced MOI and weight values.

The other thing to keep in mind is that the higher your AI have to attack from the more their speed will build up and the sharper they will have to pull out. Set your strafing attacks from low levels and see if it helps the AI to fire or bomb.

Give it a try and see if it solves the problem and let us know.

[Allen]

Been sick all week and done vary little.

In the past I have tested the MOI idea that the AI only uses the MOI in the .air and found it to be BS. Years ago I gave the AI Oscar huge MOIs in the aircraft.cfg and it caused the AI to be unbale to move. I could see the control surfaces where moved but the play kept moving straight like it was a level bomber. I also nuke out the .air MOI numbers if found.

May be a weight prob as the J is 19,530 lbs dry. ALT is good as they are at 500 feet to attack ships and they have over 2 Nautical mile before the reach the target ship. Watching the AI with Chase View before they attack the climb up what looks to be 1000 feet or so before diving down.

[Captain Kurt]

Hi Allen

I hope you are feeling better.

I have to disagree with you though, sorry. I stand by my statements as correct. I have used this method for years with complete success.

Just to be sure though, because I am far from infallible, I repeated your experiment giving the AI a huge MOI in the .cfg file. They took off and flew just as if nothing was changed. To double check I added another 100,000 points to each value which should have immobilized them completely and it had no effect. Pretty conclusive they use the .air file.

Removing the .air MOI numbers should destroy any aircraft specific flight characteristics. I wonder if there is some default value in the program somewhere like there is a default panel and dp. I don't see anything obvious but if your planes will still fly and be controllable, there must be some generic default flight model somewhere. If you are doing this with the B-25, then I am not surprised they won't attack as the MOI definitely affects whether they will break off too soon or not.

[bearcat241]

Allen,

Just drop the weight to P-38 standard and set higher aggression and skill levels. Did you try subbing the P-38 flight dynamics for short-cut testing?

[Allen]

Captain Kurt
This was years ago. Maybe I did something wrong back than. I really don't know.

bearcat241
Not yet. The bug that got me hasn't really let go.

[Allen]

Setting MOI and weight in the .air had no effect sadly. Set the .cfg at 10,000 pounds and P-38 MoI numbers with guns set at 1500 meters is the best I have gotten. It seems okay. I'll attach the B-25J and mission if any one wants to try there luck with the AI...

[bearcat241]

Set the .cfg at 10,000 pounds and P-38 MoI numbers with guns set at 1500 meters is the best I have gotten.

Not surprising...the P-38 is the heaviest fighter in the stock hangar and it gets the ground attack job done in the AI role, even at an empty weight of 12264 pounds.

[Allen]

I'll attach the B-25. Not sure if things are working right but maybe some one will give it a test. Includes my test mission. Set up as strafing aircraft but has level bomber DP panel and model as well.

Attachment 51834