Master of Magic 1.5, temporal codename "Raid" development

[Seravy]

Oh, the strategy guide.
I've seen the Master of Orion version of it, I think that's more like a "the game works like this so you should..." type of document, I mean it was written based on how the game works, not the other way around.

You haven't read the MoM strategy guide?

How should I say...
Rulebook -> Coding the game -> Strategy guide.

The people making the Strategy Guide weren't the ones deciding what the game is supposed to be like, were they?

The Strategy Guide was written in close consultation with the developers and reflects how the game actually is, for the most part. The author is Alan Emrich, if you've ever heard of him. There are inaccuracies but I regard it as better than the manual. Why not read yourself and judge? http://megadrv.com/master.of.magic-strategy.guide.pdf

Like the one for MoO, I think I never found it to read it so...thx!

[FrancoK]

Why? Spearmen have less HP and less defense, so they should go first in the list of the units to kill.

It's a list of weights to be randomly selected. Not a list of order.
Sky Drake has 10 defense so it had a weight of 20. A Spearmen with 2 defense had a weight of 28. That means out of these two units, if there aren't more, you have a 20/48 chance to get the Sky Drake selected first.
So with 8 Spearmen and 1 Sky Drake, you would have about 20/264 or something chance to get the Drake selected first...but if it happens, it'll keep taking damage until it dies and then the Spearmen survive (or if there is more damage left, they also start getting killed one by one).
Ultimately, whichever units was selected first is dead, and that's random. Weighted, but random.
Ofc, the Drake has more HP than Spearmen, but if there is enough damage in the queue, the Drake getting killed with no Spearmen taking damage can and will happen at the above mentioned 20/264 chance.

93% of chance to target one of the spearmen is good, isn't it? The Drake is save from the attack most of the times, without counting any unit that could not fly cannot target it, no spells/spellcasters can hurt it, and so on.

In real life? Yes. In a game where you can target with a lightning bolt the units? I do not think so.

Depends. For starters, neutral encounters don't throw lightning bolts. Nor do Life wizards.
But even if they can do it, would they really target your swordsmen unit with 1 figure left, wasting a lot of mana killing it, or would they go for damaging a unit that still has a lot of health? Killing the most damaged unit is good strategy if you expect to lose the battle but if you expect to win, hitting the least damaged unit to get the best damage out of the mana you can spend is better.
Besides, there are a limited amount of spells you can cast in a battle, but there are, usually more units participating. At least some of them should be surviving, and by some, I mean more than always exactly one.

Yes, the distance has to be taken in account as well.

Strategic Combat does not have a distance at all. It's just roll some dice, compare the sum power of both armies+the rolls+spell casting capacity, higher wins the battle then takes damage equal to the lower army's power distributed semi-randomly.
How that damage is distributed is what can be used to simulate the strategies and movement the players employed in battle - the more evenly the damage is distributed, the better the winner was moving their units around to distribute damage.
While it might sound counter-intuitive from the loser's perspective that the most expensive units survive, it matches the goal of the winner, and it's more reasonable to assume that the winner of the battle was more capable at achieving their goals in the battle.
Additionally, more expensive units do come with better survivability, either in form of higher movement, more health, more defense, abilities like flight or various immunities, etc., or at least good enough offense/first strike/thrown/breath/gaze to take out anything that would approach them before they kill the unit. (albeit this sadly isn't true without balance related modding, some expensive units are completely worthless there. )

This help the AI to have a bunch of low power, wounded troops around with a maintenance cost as a fully healed unit; instead, if the weaker unit goes down, AI can replace them with more powerful unit. So I can finally fight with decent opponents and not bunch of spermen barely able to scratch my troops.

Not sure how being so much more slow paced than my mod affects this exactly, but at the very least I think a damaged stack of units will be able to regain health faster than building new units. Important battles are usually in cities, where they heal up faster, and if there are more survivors, there is no need to send more units to guard the city.
About the quality of the units, the AI does produce the better units when they are available based on random chance, and the weaker ones will get dismisses as soon as the global max unit cap is reached, to make room for more. (or if you don't install that change, they'll get disbanded once every 25 turns but I don't recommend that, the original code dismissed even heroes!)
Each unit has (cost^2)/10 chance to get selected for production by default (in the CoM mod I assigned actual chances for selecting each unit instead, but I won't add this to raid because the default stats of the unit do not match the mod, units that are good there are usually horrible in the original game).
Maintenance is meaningless for the AI, they have a HUGE discount on it, something like 50% around normal/hard and 75% on impossible? On top of earning 200/300/400% gold. Keeping the spearmen to be able to get more taxes and less rebels should be good for the AI on all but the lowest difficulty. If there is no need for them to guard anything, they'll get assigned to some attacking stack and get used up.

All correct, I agree on everything.
There are lot of issue to keep in game, as you said strategic combat, I never use, was always used by AI vs AI combat, also some enemies could be immune to some of my units attack, while other enemies immune to the rest of my units, so IMHO every unit should "roll" its target and only with valid ones every turn.

With the change I suggest the heroes goes down last, if it is mildly powerful, and AI should put item on it so even a newbie hero is protected; as all the human players do.

The AI does equip their heroes with items. If they find an item the hero can equip, that is. They can also cast Create Artifact or Enchant Item.

Exactly, so the heroes are "protected" by the "cannon fodder" that travels with them.

Edit : marked W128 as optional.

Thank you.

Oh, the strategy guide.
I've seen the Master of Orion version of it, I think that's more like a "the game works like this so you should..." type of document, I mean it was written based on how the game works, not the other way around.

You haven't read the MoM strategy guide?

Impressive how good he is, even without reading the SG, eh?



Seravy , if you want, I can give you the proof the manual in the game is outdated and the SG is reliable.
Go to page 18 of the Spellbook. You'll find the spell Guise, it was removed and substituted with Blur (you will not find B it in the book).

If you go to last page of the Tech supplement, in credits, you'll find the MoM manuals were written by Petra Schlunk, she is also one of the writer of MoM Strategic Guide. She worked closely to the developers to create the manuals and the strategic guide.

If you agree SG is a reliable source, as I suppose you will, can we check again the Wind Walker bug?

[Seravy]

The Drake is save from the attack most of the times, without counting any unit that could not fly cannot target it, no spells/spellcasters can hurt it, and so on.

None of that applies in Strategic combat.
The Drake is a 25 hp wall that absorbs damage, while Spearmen are 6 (or 12) hp walls. There are no "units" perticipating in Strategic combat, only armies. Then the total damage the army has taken is distributed between the units after the battle is done.
The whole idea of the Drake having a nonzero chance to die if the total amount of damage is exactly or only slightly above 25 is absurd. That would imply that other units (and they aren't necessarily Spearmen, could be other Sky Drakes, even) didn't do a single attack in the entire combat, to take damage from counterattacks.
But let's forget it's a drake for a moment. If I use 9 swordsman to fight in a battle, against 9 spearmen, chances are I end up with 4-9 surviving but damaged units, the rest dead and the enemy annihilated, and not end up with 5 dead and 4 full health Swordsmen. At least if in normal combat. Even if I'm the AI, I will use all 9 units for attacks, and they'll all take their share of damage. Some will take more, some less, but they won't stay unhurt (ok archers might but that's a different story, strategic combat isn't that detailed). Even if the enemy is trying to focus on killing my units, they'll only kill a limited amount while the battle lasts...and the rest of the units will still be hurt from participating.

IMHO every unit should "roll" its target and only with valid ones every turn.

Again, there are no units participating in Strategic combat, only armies with a total attack power, ranged power and defense power.
The only part where individual units do matter is where the damage taken has to be split up.
Strategic Combat is not very detailed. There are not even spells cast in it, it just spends mana and converts it to either health, damage or army power based on the realm, skill and number of books.

If you agree SG is a reliable source, as I suppose you will, can we check again the Wind Walker bug?

One last time that is a bug.
(all of the statements are for combat)
The unit does not gain flying if it has the wind walking ability.
The unit does not gain flying if it has the wind walk spell cast on it.
The unit can be attacked normally by the AI.
The unit can be attacked normally by the Autobattle.
The unit can be attacked by the player if it has the wind walking ability
contrary to all the above, the unit cannot be attacked normally by the player if it has the wind walking spell.

So out of 6 possible cases, 5 implies the unit does not have flying.
The strategy guide does not explicitly say it's "in combat" so it might mean overland, or idk. I rather not change a spell on such vague information.

Also, we are talking about a game company here, which do you think is more likely to be in an official book about it?
"This, this and that are bugs, sorry guys, this is a really cool game but don't cast X, Y or Z because they crash the game, and A,B and C have no effect at all, don't bother. But look D works way better than we wanted to, awesome, huh?"
or
"This is working like this, it's a feature, we definitely wanted it to be like that!"
Not to mention the writer might not be the same person as the one who designed the spell, nor the one who coded it. In fact he probably isn't, there are like 50 people on the staff list.

I didn't read all 400+ pages yet, but does the guide say for Evil Presence that it never works? Does it say that for New Minarals? ofc not...
I think I'll read diplomacy next, unless they actually say the features don't exist or exist but never function, they aren't telling the truth. (yes, I'm paranoid, helps in finding problems in game code )
Edit : "...will recognize that diplomacy in master of magic, while rich in detail, seems a less vital concern..."
nice way of saying "it does not work, guys"
"...This perception stems from the blurry distinctions between war and peace, with units maneuvering around the other player's seemingly "rightful" territory..."
not from the fact that the AI can't talk to the players at all due to bugs!!!??? Rich details? Yeah they WERE supposed to be in the game, I mean I found and fixed them but come on...did this guy ever play the game? Like, at least, idk, 2 or 3 hours? You think you would notice that none of that 500 lines of diplomacy text you added in never comes up, eh?
I wonder what the rest of the chapter will be about, will they explain how the features would have worked if they were actually coded in?

The guide was useful on one other thing though, I discovered I misread the code about fame at end of combat. But as a source I can trust? I don't think so...

Oh, the diplomacy table in the SG is wrong.

It's -100 to -80 Hate, -79 to -60 Troubled, -59 to -40 Tense and so on. Exactly 20 points for every text, 0-20 being neutral. Harmony is an exception, it is only for exactly +100 (as it would range from 100 to 119).

[FrancoK]

The Drake is save from the attack most of the times, without counting any unit that could not fly cannot target it, no spells/spellcasters can hurt it, and so on.

None of that applies in Strategic combat.
The Drake is a 25 hp wall that absorbs damage, while Spearmen are 6 (or 12) hp walls. There are no "units" perticipating in Strategic combat, only armies. Then the total damage the army has taken is distributed between the units after the battle is done.
The whole idea of the Drake having a nonzero chance to die if the total amount of damage is exactly or only slightly above 25 is absurd. That would imply that other units (and they aren't necessarily Spearmen, could be other Sky Drakes, even) didn't do a single attack in the entire combat, to take damage from counterattacks.
But let's forget it's a drake for a moment. If I use 9 swordsman to fight in a battle, against 9 spearmen, chances are I end up with 4-9 surviving but damaged units, the rest dead and the enemy annihilated, and not end up with 5 dead and 4 full health Swordsmen. At least if in normal combat. Even if I'm the AI, I will use all 9 units for attacks, and they'll all take their share of damage. Some will take more, some less, but they won't stay unhurt (ok archers might but that's a different story, strategic combat isn't that detailed). Even if the enemy is trying to focus on killing my units, they'll only kill a limited amount while the battle lasts...and the rest of the units will still be hurt from participating.

IMHO every unit should "roll" its target and only with valid ones every turn.

Again, there are no units participating in Strategic combat, only armies with a total attack power, ranged power and defense power.
The only part where individual units do matter is where the damage taken has to be split up.
Strategic Combat is not very detailed. There are not even spells cast in it, it just spends mana and converts it to either health, damage or army power based on the realm, skill and number of books.

I would like to speak about SC in another thread, the argument is really interesting.

The strategy guide does not explicitly say it's "in combat" so it might mean overland, or idk. I rather not change a spell on such vague information.

The spell is very clear:
"Wind Walking gives the enchanted unit the ability to fly and move 3 squares."
Then explains what happens to the other units, but the enchanted unit will fly. Every time and everywhere.

Not to mention the writer might not be the same person as the one who designed the spell, nor the one who coded it. In fact he probably isn't, there are like 50 people on the staff list.

The writer is the writer of the manuals, since they was never able to create a manual for MoM 1.31, the writer explain the changes already made in the game by the patch.

I didn't read all 400+ pages yet, but does the guide say for Evil Presence that it never works? Does it say that for New Minarals? ofc not...

Is MoM full of bugs? I do not have to tell you.

I think I'll read diplomacy next, unless they actually say the features don't exist or exist but never function, they aren't telling the truth. (yes, I'm paranoid, helps in finding problems in game code )
Edit : "...will recognize that diplomacy in master of magic, while rich in detail, seems a less vital concern..."
nice way of saying "it does not work, guys"
"...This perception stems from the blurry distinctions between war and peace, with units maneuvering around the other player's seemingly "rightful" territory..."
not from the fact that the AI can't talk to the players at all due to bugs!!!??? Rich details? Yeah they WERE supposed to be in the game, I mean I found and fixed them but come on...did this guy ever play the game? Like, at least, idk, 2 or 3 hours? You think you would notice that none of that 500 lines of diplomacy text you added in never comes up, eh?
I wonder what the rest of the chapter will be about, will they explain how the features would have worked if they were actually coded in?

Absolutely! the Diplomacy never worked and pissed me off, in other games I usually go through various diplomacy passages, in MoM all my diplomacy is: swap our spell, or I bribe you out of war, so we can swap spell. period.
That's why I'm really excited reading the fix you made in it, but, again, I will made a new thread.

[Seravy]

New file :
274 - AI ship carry capacity and movement fix 3

This might actually let the AI use ships properly now, at least in the test game I ran, no ship got stuck until the AI conquered 3 whole continents, and was able to keep 3 ships in motion at the same time each with their own destinations and purposes.

[Seravy]

About volcanoes.

I found code that looks like something that would add a random ore to the newly raised volcano tile at a 20% chance in the game code. I haven't found out why it never happens though.
Interestingly enough, Asfex's file contains no changes at all in this part of the code, so the two might be unrelated.
Currently investigating this.

[Seravy]

Found the problem.
The code was doing this :

1.Select the ore type to add
2.Add the terrain modifier already on tile to the selected ore
3.Delete the ore part of the modifier
4.Store the new terrain modifier.

When it should have been doing this :

1.Select the ore type to add
2.Delete the ore part of the modifier already on tile
3.Add the new ore type to the existing modifier
4.Store the new terrain modifier.

So 2 and 3 were in the wrong order and caused it to do nothing.

Now the question is, what do we do with this information?
This would imply that contrary to what documentation says, the code was intended to create a new mineral at a 20% chance when any volcano is raised. (and this to be honest would make Armageddon a horrible spell that helps enemies)

[FrancoK]

New file :
274 - AI ship carry capacity and movement fix 3

This might actually let the AI use ships properly now, at least in the test game I ran, no ship got stuck until the AI conquered 3 whole continents, and was able to keep 3 ships in motion at the same time each with their own destinations and purposes.

To test this fix it is enough adding the 3 patches? Or there are other related issues?
It is very good you relate the new patches their dependencies, can you do the same in the older patch (in this case AI ship carry capacity and movement fix 1 & 2)?

Found the problem.
The code was doing this :

1.Select the ore type to add
2.Add the terrain modifier already on tile to the selected ore
3.Delete the ore part of the modifier
4.Store the new terrain modifier.

When it should have been doing this :

1.Select the ore type to add
2.Delete the ore part of the modifier already on tile
3.Add the new ore type to the existing modifier
4.Store the new terrain modifier.

So 2 and 3 were in the wrong order and caused it to do nothing.

Now the question is, what do we do with this information?
This would imply that contrary to what documentation says, the code was intended to create a new mineral at a 20% chance when any volcano is raised. (and this to be honest would make Armageddon a horrible spell that helps enemies)

The documentation says there is a 5% chance to create a new vein for every volcano created, so volcano can be used to hit the enemies, destroying valuable tiles (food, crystal) at 100% chance with a lesser chance to create another valuable tile. Or it can be used to increase our cities output by using it on poor tiles like tundra.
Armageddon is the same, but do not hit our cities, rarely can be helpful to enemies, but it in the nature of chaos.

[Seravy]

The documentation says there is a 5% chance to create a new vein for every volcano created, so volcano can be used to hit the enemies, destroying valuable tiles (food, crystal) at 100% chance with a lesser chance to create another valuable tile. Or it can be used to increase our cities output by using it on poor tiles like tundra.
Armageddon is the same, but do not hit our cities, rarely can be helpful to enemies, but it in the nature of chaos.

All right, I checked spellbook.pdf and it does say "newly raised volcanoes have an enhanced chance of containing mineral deposits"
so I conclude that the 20% chance on creation of volcanoes was the intended effect (at least in 1.2 and earlier).
The 5% chance on volcanoes reverting is a separate, unimplemented effect that was supposed to be part of the 1.31 patch.
An extra 5% on top of the existing 20 is not very interesting, I won't waste hours on trying to implement that effect, but I think we can add the 20% original effect to our "include" list.

275 - Warp creature does not crash but stays random
276 - Fix : No ores from volcanoes on creation
277 - Allow raising volcanoes on hills, mountains and volcanoes
278 - Declaration of War chance increases after each warning message
279 - AI will not add (almost) infinite priority to targeting the fortress city.
280 - AI targeting Earthquake checks for Nature Ward instead of Chaos Ward and Consecration

277 will be optional, 278 fixes the "eternal warning spam" problem.

[FrancoK]

All right, I checked spellbook.pdf and it does say "newly raised volcanoes have an enhanced chance of containing mineral deposits"
so I conclude that the 20% chance on creation of volcanoes was the intended effect (at least in 1.2 and earlier).
The 5% chance on volcanoes reverting is a separate, unimplemented effect that was supposed to be part of the 1.31 patch.
An extra 5% on top of the existing 20 is not very interesting, I won't waste hours on trying to implement that effect, but I think we can add the 20% original effect to our "include" list.

You're right, 1.31 says the chance is 5%, I would stay stick to the latest patch and change the 20% to 5%.
Again I agree with you creating the special tile while creating the volcano or after it turns to a mountain is not so important, but a chance 4 time greater to obtain a special mineral could.

275 - Warp creature does not crash but stays random
276 - Fix : No ores from volcanoes on creation
277 - Allow raising volcanoes on hills, mountains and volcanoes
278 - Declaration of War chance increases after each warning message
279 - AI will not add (almost) infinite priority to targeting the fortress city.
280 - AI targeting Earthquake checks for Nature Ward instead of Chaos Ward and Consecration

Any chance the nightshade will help protect the city from harmful spells?

277 will be optional

Indeed, it was done on purpose to avoid volcanoes destroying enhanced tiles that lies on hills/mountains!