On the other hand AI attacks with great numbers of aircraft are programmed attacks. It's in the hands of the scenario designer to design such attacks.

Only in part,

firstly the writer can only program attacks on targets known at game start such as bases and secondly can only specify the make up of the attacks and the time. These attacks therefore are always straight line and fairly predictable eg I used to start a scenario by looking at the 'expected' air assets of all the basis to guess where strikes would emanate, these days I try not to make use of that sort of thing - ditto I like to play without the range rings on for the enemy as it gives me information I shouldn't really have at least initially.

Good scenarios will have some enemy patrols set up to spots where they are likely to detect ships etc and once detected they will usually attack. These are less predictable especially first time you play the game or if well thought out variable start positions have been used.

Also the AI may actually put together its own strike, so slightly smart, in fact if you ever get a large group attacking you in one of my scenarios it was the AI that did it

which brings up an interesting (to me) question, how many scenario writers program large group strikes?

it also brings up a question I've recently pondered (with the new abilities provided by Tony voa the exportDLL functions) how can we level the playing field a bit at least for poor old AI? Any ideas?

how can we level the playing field a bit at least for poor old AI? Any ideas?

Please start a new thread for this very involved topic.

I always assumed that large airstrikes against surface units by the AI are actually programmed patrols, meaning that for example 20 Backfires were sent into the middle of the Atlantic by the scenario writer and if by chance during their journey down there a sub or a Bear made contact with a convoy, the Backfires were vectored in to attack that. I did not realize that it was the AI launching a group of 20 Backfires to make the strike.

 

Smaller groups of attack aircraft against ships are, as far I undestand, often split units from the formation editor.

So pre-programmed ASuW patrols from the formation editor are split from the base to attack a ship being discovered by some other asset.

I always assumed that large airstrikes against surface units by the AI are actually programmed patrols, meaning that for example 20 Backfires were sent into the middle of the Atlantic by the scenario writer and if by chance during their journey down there a sub or a Bear made contact with a convoy, the Backfires were vectored in to attack that. I did not realize that it was the AI launching a group of 20 Backfires to make the strike. Smaller groups of attack aircraft against ships are, as far I undestand, often split units from the formation editor. So pre-programmed ASuW patrols from the formation editor are split from the base to attack a ship being discovered by some other asset.

 

The AI will continue to make scripted (programmed) attacks as long as there are aircraft remaining to perform the mission.

 

It will also launch (unscripted) attacks of its own against newly discovered targets (often ships) and typically these are large scale attacks.

This story gives you some idea of the complexities involved.

 

Syrian Mig-29 Fulcrums escorted the 28 Russian jets that deployed to Latakia hiding under cargo planes (Aviationist)