I didn't see a correction in this thread, so I'll post it, even though the thread is a year old.

For each combat you get your full spell skill. If you have casting power 20, you get 20 in each fight in a turn, regardless of whether you fight 1 battle or 100 battles. The mana in your mana reserve does not replenish between battles, but your spell skill does.

No. It's per turn, not per combat. If you exhaust your casting power in the first fight, you get no more in subsequent fights until the next turn.

Heroes use their own separate casting power and mana reserves, however.

Bookwyrm627 is right, and UniversalWolf is, surprisingly for him, wrong here. Test it yourself, it's very easy (I just did it in 1 minute).:
1. Make sure you have lots of mana. Alchemy all your gold into mana, break a few magic items, whatever.
2. Have a couple of units in range of a node or enemy/neutral city, send in one unit - cast your full casting skill (what UniversalWolf is calling casting power) in the battle, then retreat or lose.
3. Send in unit #2, you'll have your full skill to use again (though as Bookwyrm627 notes, your mana reserve does deplete, which is why I've included step 1)
4. Repeat as many times as it takes to convince yourself.

This is true for the vanilla game, for the Insecticide patch, for every version of MoM ever, AFAIK.

Yeah, you're right. It's bugged in all versions.

I don't believe it's bugged but just the way it's supposed to be. It's not a bug if it's working as intended.

After all is there a good reason you should not be allowed all your spell skill in each battle as long as you have enough mana ?

According to the manual, the Casting Skill is specifically, "the amount of mana you can cast during one battle." So you're right. It's not a bug and the description in the Plight mod is wrong.

FWIW, I hate that and think it should work the way I thought it worked.

I prefer it. Age of Wonders does what you described and I find it annoyingly constricting.

If you consider a wizard with a meagre CS of 60 or so depending on book selection, that's enough to cast Flamestrike, so he can do so in as many battles as he has to fight during a single turn as long as he has the crystals. If the rule worked the way it should work, he would have to be judicious about casting Flamestrike, because he could only use it in one of the battles. That's reason enough right there for it to work the way it should instead of the way it does.

The way it works now makes the most overpowered spells even more overpowered, and decreases the need to think strategically about how to use all combat spells in turns with more than one battle (and you never know for sure when you might be surprised by a battle you didn't expect).

Why should your CS replenish after a battle, anyway? Does your wizard chug a bunch of mana potions or something? Then why can't he do that for artifact creation?

The alternative would be just as easy to exploit. Consider this scenario: first you attack an outlying town you don't really care about, letting the enemy wizard cast all the spells, and maybe defeat you.

Then in the same turn you attack his fortress. Checkmate.

I'm with TwoHandedSword on preferring that each battle gets your full spell skill.

On the technical/rules side: He's already mentioned the AI abuse (the AI already can't handle of lot of things well, and the level of planning required for anticipating what battles will be fought in a turn AND planning what spells you might need in those battles is crazy high), and getting jumped by unexpected battles in a turn where I have other battles planned (or see the risk for them) is exceedingly rare. There are just too many ways to see things coming (like spearman on patrol, if it were necessary), with the main exception being units (like new spawns) using enchanted roads to move a long distance. You are still limited by your stock pile of mana crystals; I don't know about you, but I'm usually running a fine line of "just enough mana crystals to get me to next turn without losing anything while I cast my full spell skill overland". I tend to run 0 mana income (putting it all into spell skill), and so gold/mana management is a big issue in my early/mid game, especially when I have overland spells I need to maintain. There is also the casting range multiplier eating into your mana crystal supply as well.

On the flavor side: one explanation for why you get your full spell skill each battle comes from the complexity of overland spells versus combat spells. Generally, combat spells will be enormously easier to cast than overland spells because of the scope of effect. Setting up a wind of death that will sweep across both worlds just sounds much harder than putting a unit into a temporary coma or throwing a blast of [element] at it. Creating a holy armor spell that will pull power from a battery (your mana supply) and last as long as it has power would be harder than throwing enough power into the same spell to make it last a day (or even an hour). The amount of mental involvement is much greater in creating a permanent magical item than in summoning a magic being that only needs to last for one battle.

All that said, maybe you just need a few minutes or an hour to clear your mind of the battle you just fought, and put your thoughts in order (or, on the flip side, you take a few minutes to prepare yourself for a series of rapidly cast spells just before each battle starts). Your spell skill in general reflects your ability to manipulate the fabric of magic in the worlds, so a wizard with more spell skill is just better at putting spells together, both overland and in combat.

Except you're not exploiting anything with the rules as they are now, because it's WAD. Illogical, exploitative, and WAD.

As for combat spells being easier to cast, keep in mind your wizard is casting combat spells across vast distances from inside his tower. If he's enchanting a suit of armor or a sword, presumably it's sitting on a table right in front of him. Even if he's casting a massive overland spell, that spell originates from his tower and doesn't have a specific target that may be on a distant continent or another plane.

Quick note, Wolf: At this point, I'm replying in the spirit of having a friendly argument. I'd be plenty happy to continue if you would enjoy continuing. I'm a little unsure about the tone in your latest post, so I can't tell if you are enjoying this debate or if you are more on the irked side. If you'd prefer to stop, then you don't need to read the rest of this post (I'm just arguing back).

-------
Sword called the alternative just as easy to exploit because you indicated that you consider the current method for combat skill to be exploitative.

Casting combat spells across vast distances is represented by the multiplying the actual mana cost of the spell when you cast it. For example, a fireball would be the wizard going "make a ball of fire and then hit that guy way over there with it", and then s/he just wraps the spell in a big enough magic shell to make the trip (or uses more mana to weave the spell on site from such a long distance away). A longer distance means that more shell is needed. A channeler is adept at this distance casting, so they don't pay the extra. A wizard might get a similar effect to a fireball from making a molotov cocktail and then throwing it really hard. Hand crafting weapons and armor, especially GOOD weapons and armor, is a much more laborious process. If the fireball dissipates right after it hits the target, who cares? The victim has already been burned. If the armor enchantment falls apart 30 seconds after the hero puts it on, then the entire point of making the armor has been defeated. Likewise, making a molotov is vastly simpler than setting an entire city (or planet!) on fire all at once.

Heh, no I'm not irked at all. For future reference you can always assume I'm not the least bit irked, because it takes a lot to get me irked and if I was it would be obvious.

I guess that's what these dumb things are for.

:)

If I might insert a point, I'd add that casting at a distance isn't all that impressive... combat spells can only be cast where you have units in battle, and presumably all the wizard's troops carry some totem that allows the wizard to communicate with them, give orders, receive sighting reports, and cast spells through. So yes, I'd suggest temporary, localized combat spells can be considered so trivial compared to the permanent or world-affecting overland spells that they simply don't put a dent in one's overland casting progress.
Though like bookwyrm, I find myself stymied by mana shortage more often than skill shortage.

One more argument against skill being limited within a turn - the exploit mentioned by 2HS and bookwyrm, aside from being a programming nightmare, also causes an artificial dilemma in choosing where/when to cast. Because of the turn-based system, one has to make a decision (early in the turn) without knowing whether an attack (later in the turn) will occur or not (and indeed, an attacker may choose to mount the second attack, or not, based on how much skill one reserves). However, all action during a turn should be considered to be, more-or-less, simultaneous - the game isn't supposed to be modelling a world where only one stack moves/acts at a time. The fact we run battles one at a time is an artificial constraint, necessitated by our human (and game!) limitations.
If favouring a "casting skill is limited per turn" model for combat casting then, I'd suggest the only fair solution would be for attackers to declare all attacks in advance, and then both sides can choose where to allocate that limited casting skill. Cumbersome, and less fun. We already have this problem with mana, but that can, potentially, be maintained at a high enough level not to matter.