Master of Magic 1.5, temporal codename "Raid" development

[Seravy]

The 'right' think to do in that situation is either not attacking in the first place,

The reduction of the Magic Immunity bonus in AI calculations will make this happen much less frequently. Completely eliminating it is impossible but it hasn't happened to me at all in my current game at least. The reverse was a lot more frequent, the AI walking units being unable to attack my flying defenders.

or flying into the city and parking there until time out,

That would be a good thing to do, yes, good idea. Not sure if it's worth coding for a situation that is supposed to be very rare, though, not to mention that destroying stuff is not necessarily beneficial as long as there is good chance to win the battle (either with direct damage spells, by buffing the units, or by adding more units to the stack before the next attack.)
In this particular scenario, the attacked played only Life and the defender was immune to magic, but in 99% of the battles, the attacker can keep nuking the units until they win.
Building units is much slower without mods, so even if the AI can only kill one per battle, it will eventually take the city. But 99% of the time they won't even attack it like that.

Under some conditions flying up and attacking while using heal spells to stay at 2 figures would be worthwhile, it depends on what the defender has (does healing work on magic immune units?).

Normally the AI would attack and stop doing so if the enemy to own unit ratio becomes too bad. If it was too bad to begin with, they don't risk the attack, there is no point, every exchange of hits would damage the enemy less than their own unit, or at least that's the more likely outcome. It's not possible to make an AI that always makes the best possible action in a game where there are this many spells and abilities, but making one that does the action that is best 95% of the time is enough.

Some new files :
267 - AI power distribution : Allocate 100% on mana when casting Time Stop
268 - AI will automatically use alchemy if gold exceeds 24000 to make mana.
269 - Spell Binding cannot fail
270 - fix : Create Artifact base cost (used by Suppress Magic) is wrong

[FrancoK]

or flying into the city and parking there until time out,

That would be a good thing to do, yes, good idea.

Or not.
When I conquest a city I try not to destroy the infrastructure, AI should do the same.
If my troops are ill-equipped to conquer it, I move better troops to get it.
I only go raiding the city if I want to raze it at the next attack (less population, less Fame loss).
If AI known before attack if they want to raze it at next attack, it could be a tactic, otherwise not.

267 - AI power distribution : Allocate 100% on mana when casting Time Stop

During Time Stop no mana or gold are produced.
Or did you mean to gather as mana as possible before the spell is in effect?

In Asfex mod the fix for volcano is applied when the volcano is created, so the new vein appears right after the spell is cast.

[Seravy]

267 - AI power distribution : Allocate 100% on mana when casting Time Stop

During Time Stop no mana or gold are produced.
Or did you mean to gather as mana as possible before the spell is in effect?

Yes, it is for the turns before the spell takes effect. It costs 1500 so there are plenty of turns to fill up the mana reserve.

In Asfex mod the fix for volcano is applied when the volcano is created, so the new vein appears right after the spell is cast.

Thought so, in that case I won't use it, as that isn't the official effect. The spell is bad enough as it is even without giving free adamantium to the opponent immediately :D

[FrancoK]

In Asfex mod the fix for volcano is applied when the volcano is created, so the new vein appears right after the spell is cast.

Thought so, in that case I won't use it, as that isn't the official effect. The spell is bad enough as it is even without giving free adamantium to the opponent immediately :D

Well, the manual states:
"Newly raised volcanoes also have an enhanced chance of containing special mineral deposits."

So it is an official effect, simply it did not work before Asfex fixed the bug.

[Seravy]

In Asfex mod the fix for volcano is applied when the volcano is created, so the new vein appears right after the spell is cast.

Thought so, in that case I won't use it, as that isn't the official effect. The spell is bad enough as it is even without giving free adamantium to the opponent immediately :D

Well, the manual states:
"Newly raised volcanoes also have an enhanced chance of containing special mineral deposits."

So it is an official effect, simply it did not work before Asfex fixed the bug.

It really does say that, interesting. All right I'll see if I can take this from Asfex's exe.
At least it can make it more useful when casting on own territory...horrible as a curse though.

Meanwhile :
271 - AI won't consider uncontancted players as valid targets for overland curses
Stuff like Drain Power or Subversion, but it also opens up the possibility for city curses (they would fail to lack of cities to target though). This included already eliminated players, the horrors...

I think the game should have had a different title - Master of Magic 1.31 : Infinite Bug Works
I spent 3 months playing and fixing bugs and I still find new bugs on a daily basis!

[FrancoK]

In Asfex mod the fix for volcano is applied when the volcano is created, so the new vein appears right after the spell is cast.

Thought so, in that case I won't use it, as that isn't the official effect. The spell is bad enough as it is even without giving free adamantium to the opponent immediately :D

Well, the manual states:
"Newly raised volcanoes also have an enhanced chance of containing special mineral deposits."

So it is an official effect, simply it did not work before Asfex fixed the bug.

It really does say that, interesting. All right I'll see if I can take this from Asfex's exe.
At least it can make it more useful when casting on own territory...horrible as a curse though.

Since you are there, there is an event tied to this that never happens, no two events:
The vein is exhausted and a new vein appeared.
Like good/bad moon event, special tiles can disappear or appear.
Never seen it happening in years of playing, not sure it can be fixed.

Meanwhile:
271 - AI won't consider uncontacted players as valid targets for overland curses
Stuff like Drain Power or Subversion, but it also opens up the possibility for city curses (they would fail to lack of cities to target though). This included already eliminated players, the horrors...

I think the game should have had a different title - Master of Magic 1.31 : Infinite Bug Works
I spent 3 months playing and fixing bugs and I still find new bugs on a daily basis!

Yep, there are so many bugs I opened a bottle of Champagne when the most visible were fixed by Kyrub and I get used to the others, like 271. Yes, a wizard I never met cast Chaos Rift on my capital, AI cheats anyway so I did not feel it is a bug, but you're right, it is a bug, and if you fixed it, thank you!

[Seravy]

In Asfex mod the fix for volcano is applied when the volcano is created, so the new vein appears right after the spell is cast.

Thought so, in that case I won't use it, as that isn't the official effect. The spell is bad enough as it is even without giving free adamantium to the opponent immediately :D

Well, the manual states:
"Newly raised volcanoes also have an enhanced chance of containing special mineral deposits."

So it is an official effect, simply it did not work before Asfex fixed the bug.

It really does say that, interesting. All right I'll see if I can take this from Asfex's exe.
At least it can make it more useful when casting on own territory...horrible as a curse though.

Since you are there, there is an event tied to this that never happens, no two events:
The vein is exhausted and a new vein appeared.
Like good/bad moon event, special tiles can disappear or appear.
Never seen it happening in years of playing, not sure it can be fixed.

I thought that's because the chance of finding a mineral vein (or suitable place for one) on first try is simply too low and the event gets replaced by another most of the time but...looking at the code, that does not seem to be the case, it actually tried finding one 100 times before giving up.
Holy god, no wonder!
This says "if found target<>0 then stop trying"...but "no target" is -1. So the process just goes
1. let's start, we don't have a target yet, so target=-1
2. Did we try 100 times yet? no, continue
3. Did we find a nonzero target? Hell yeah -1 is not zero, we are done, our target is -1.
...outside procedure "-1" is returned so no target was found, let's try another event.

Why did people say they had this event, just extremely rarely? With this code, they are impossible to get, ever!

New ores possible to get seem to be 2,4,5,6, and 7, so Coal, Gold, Gems, Mithril, Adamantium, but no Quork, Crysx, Coal or Silver...
Do we want to keep this limitation?
Oh and land type has to be 4 or 5, if I had to guess, Mountain or Hills?

...new file :
272 - Depletion and New Minerals work

[Tiltowait]

I'm finally ready to start playtesting the new patches. Is there a recommended starting position that would be more helpful than others? Starting parameters, spellbooks, retorts, etc. I really don't care what I play.

The "mineral" event is a copy from one from Master of Orion. A random event would say "mineral depletion" and change an ordinary planet to Mineral Poor or "deposit discovered" and the planet would become Mineral Rich. It didn't happen very often as it could be devastating or overpowered, depending on which planet it hit. Does it place the new mineral in a city catchment, or just somewhere on the map? I say include all the minerals, why not. However Mithril and especially Adamantium could be unbalancing. You have to go to Myrror to get Adamantium, that's one of the things that makes it worth going to. Minerals can only occur on certain types of land, and they make the otherwise worthless mountains worth having.

[Seravy]

I'm finally ready to start playtesting the new patches. Is there a recommended starting position that would be more helpful than others? Starting parameters, spellbooks, retorts, etc. I really don't care what I play.

The "mineral" event is a copy from one from Master of Orion. A random event would say "mineral depletion" and change an ordinary planet to Mineral Poor or "deposit discovered" and the planet would become Mineral Rich. It didn't happen very often as it could be devastating or overpowered, depending on which planet it hit. Does it place the new mineral in a city catchment, or just somewhere on the map? I say include all the minerals, why not. However Mithril and especially Adamantium could be unbalancing. You have to go to Myrror to get Adamantium, that's one of the things that makes it worth going to. Minerals can only occur on certain types of land, and they make the otherwise worthless mountains worth having.

As this isn't a "mod" in the sense of no major gameplay changes anything that worked for you in 1.31 or 1.40 should still work.
If you mean which files to try, I suggest using all AI and Diplomacy files together as some of those rely on each other. Anything else is up to you, the fewer you try at a time the easier it is to pinpoint a bug if there is one (although I really hope there isn't).
Probably best to go for Hard or lower difficulty, I have no idea how effective the improved AI is under these conditions but in Caster, playing Extreme or Impossible sometimes feels like masochism

The mineral has to be near the targeted player's city. Not sure if terrain 4 and 5 are the mountains but I assume they are.
The ore codes generated by the event might not even be the same as the ones used on the map but they most likely are.

Ehhh, now this is weird coding if I ever saw one. At the time the Event is queued up to happen, it looks for a valid target. Then does not save the target found, except for the player's ID, and when the event actually happens, looks for a target again. So there is a chance to get the event and THEN fail to target anything with it. But in this case the game will go ahead and generate an effect anyway to an unset memory address...
Did they seriously think a RANDOM target will be the same one if they call the procedure twice?
This thing definitely needs a check to drop the event if no target is found the second time.

[Tiltowait]

I meant starting conditions like size, spellbooks, retorts, etc. Everything that happens before you can play.