Sermon of Courage instead of Fanatacism?

[archaeolept]

as I understand it, if only a few units are suffering morale penalties, sermon of courage could cover them all, and so the AI will preferentially cast the weaker spell. Hopefully there's no difference in these cases.

[tinkthank]

ok.
but why? What is he "conserving" if the cost is identical? (If I can only drive 55 mph, why get the Ferrari when you can get a second-hand Ford LTD? However, if the Ferarri is the same price...)

[Molog]

Is there any benefit to multiple priests chain casting fanatacism?

[tinkthank]

If you are a level four (or higher) priest, is there ever a reason to cast Sermon of Courage instead of Fanatacism? The fatigue cost is the same; it seems like Fanat is just a better Version of SoC with no drawbacks or risks whatsoever. It seems like any AI would have to choose Fanatacism over SoC, especially if there are many units on the field.
So why do my level 4 priests sometimes cast SoC, even at later stages of a battle when many troops are scattered across the field?

[Karacan]

Each casting of Sermon of Courage or Fanaticism reliefs some of the negative morale modifiers of the units, like having had to watch half their squad eliminated or being subject to missile fire, etc.

So chaincasting Fanaticism makes sense, yes.

I admit I've never seen my level 4 or 5 priests casting Sermon. Are you certain it wasn't some level 3 priest included?

[Nagot Gick Fel]

I remember someone from IW once said SoC _might_ be more effective than Fanaticism wrt morale failure, although he wasn't 100% sure of this. If this is true, there would be a good reason to pick SoC over Fanaticism in special cases.

[tinkthank]

Ah right - it could relieve more moral loss albeit over a smaller area? OK.

[Kristoffer O]

The AI makes multiple (as in very many) castings of each spell (different targetings) and evaluates the relative effects. If the effect of sermon of courage yields a higher effect in one of these castings (most likely if there are few units with morale loss) it may choose to cast this spell instead of fanaticism. There are also some random elements in the choice.

It is very unlikely that the AI would choose sermon over fanaticism if there are lots of units with moral losses. If it happens you should be pretty certain your troops are faring well and doesn't need much encouragement.

[ May 14, 2004, 13:06: Message edited by: Kristoffer O ]

[PhilD]

Originally posted by Nagot Gick Fel:
I remember someone from IW once said SoC _might_ be more effective than Fanaticism wrt morale failure, although he wasn't 100% sure of this. If this is true, there would be a good reason to pick SoC over Fanaticism in special cases.

There may be a valid reason, or there may not be one, but does the spell AI know the truth?

Actually, that's a question of how the battle AI chooses actions - does it "know" the rules, and base its decisions on them, or does it have more or less fixed appreciations of each spells?

If one were to mod a spell's effects (I don't think that is possible right now), would that change the battle AI's use of it?

I believe the same question can be asked of modding spell costs/levels, and the strat AI's "liking" of research options. If, say, all Conjuration spells were to have their costs and path requirements increased, would the AI still research Conjuration as much?

(hard to test without access to the source code, since it isn't easy to decide what the AI has already researched)