Heroes 2 - bug fixes and mod

The mod consists of several DLL files, some of which are bug fixes of the original game, and some of which modifies the original game in order to improve the gameplay and balance.
Technically, the user needs to download the archive (link), unpack it into the game folder. In addition, you need to replace the MSS32.dll file with the file of the same name from Heroes 3, or use GL Wrapper version 1.48 and newer (in this case, there is no need to replace the MSS32.dll file).
Download MSS file (if you do not use GL Wrapper) link
The folder Plugins contains DLL files with the required functionality. They work independently of each other and, if desired, some of them can be removed by the user. However, in my opinion, there is little point in using DLLs to improve gameplay without using bug fixes or AI enhancements.
Supported versions of the game (those versions for which I had an EXE file):
- English version Succession Wars 1.3 (1.3 Eng)
- English version Price of Loyalty 2.1 (2.1 Eng)
- English version Price of Loyalty 2.0 (2.0 Eng)
- Russian version Price of Loyalty 2.1 (2.1 Rus)
- Polish version Price of Loyalty 2.1 (2.1 Pl)
- Czech version Price of Loyalty 2.1 (2.1 Cz)

A set of DLL files:

Main modules:
--------------------------------------------------------------------------
1) H2_48_SpellsBugFixes – bug fixes related to spells:
- for the Bloodlust spell (duration is determined by the power of magic, not equal to 3)
- for massive spells (sometimes they did not work even for creatures without immunity, if there were creatures with immunity on the battlefield)
- for Earth Elemental creatures - invulnerability to Meteor Shower spell was removed (additional damage was already in the original code), added invulnerability to Elemental storm (according to NWC guidelines).
- for spellcasting by AI Dragon Slayer(there was a bug)
- for spellcasting by AI Shield/Mass shield(there was a bug)
- for spellcasting by AI Slow/Mass slow(there was a bug)
- for spellcasting by AI Haste/Mass haste(there was a bug)
- for spellcasting by AI Curse/Mass curse(there was a bug)
- in versions 2.0 Eng, 2.1 Rus, 2.1 Pl, 2.1 Cz – for spellcasting by AI Dispel/Cure/Antimagic(there was a bug)

2) H2_49_AI_NecromancyFix –correction of the number of skeletons raised after the battle with neutral monsters by AI heroes (in the original, only 1 skeleton was raised). This plugin should probably not be used if you are playing without a DLL to improve game balance (where the number of skeletons is reduced from 10/20/30% to 5/10/20%), as AI heroes often accumulate very large amounts skeletons.

3)H2_70_AI_Imrovements – AI improvements. The most significant is the increase in “AI thinking radius” for the choice of actions on the adventure map. In the original, this number was very small, and the AI heroes were "blind"because of this. After increasing this number, the AI heroes added great aggression (they became more similar in behavior to the AI heroes from Heroes 3), and the gameplay became more interesting. Also quite significant is the improvement of the algorithm for choosing a spell in battle - the algorithm has become more adequate and adaptive to changing conditions.

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- Well gives +1 growth (instead of +2) for Green Dragons, Red Dragons, Black Dragons, Giants, Titans.
- secondary skill Necromancy gives 5/10/20% of skeletons (instead of 10/20/30%)
- secondary skill Eagle Eye gives a 40/50/60% chance to learn a spell (instead of 20/30/40%)
- secondary skill Archery does not allow you to avoid the penalty when shooting through the fortress walls
- changed the cost of buying Red Dragons (4000 gold instead of 3500), Black Dragons (4500 gold instead of 4000), Ogres (350 gold instead of 300), Crusaders (900 gold instead of 1000), Liches (700 gold instead of 750), Power Liches (800 gold instead of 900).
- changed Ogres (speed 2->3, health 40->45)
- changed Archmages (probability of spec.ability 20%->30%)
- changed the cost of upgrading buildings Armory (1500 gold instead of 2000), Jousting Arena (2500 gold instead of 3000), Cathedral (3000 gold, 5 crystals, 5 wood instead of 5000 gold, 10 crystals, 10 wood), Archery Range (1000 gold instead of 1500), Stonehenge (1000 gold, 5 stones instead of 1500 gold, 5 mercury), Mausoleum (2500 gold instead of 3000), Ivory Tower (3000 gold instead of 4000), Cloud Castle (10000 gold instead of 12500), Adobe (2000 gold instead of 3000)
- changed the cost of building Cloud Castle (8000 gold instead of 12500), Laboratory (12500 gold instead of 10000), Adobe (3000 gold instead of 2000), buildings to increase the growth of the 1st level (750 gold instead of 1000), special building of the Knight (1000 gold instead of 1500), special building of the Sorceress (1000 gold instead of 1500), special building Barbarian (1000 gold) instead of 2000), special building of the Warlock (4000 gold instead of 3000), building of the Left Turret (1000 gold instead of 1500), building of the Right Tower (1000 gold instead of 1500), Moat building (500 gold instead of 750)
- Hypnotize spell was moved from 5th level to 3, the spell does not end after an attack (as in Heroes 3)
- damage of Cold Ring, Fireball, Fireblast spells increased by one and a half times
- changed the chances of appearing in the Mage Guild of the spells Dragon Slayer (+), Cold Ring (-), Cold Ray (-), Fireball (-), Lightning Bolt (-), Magic Arrow (+), Anti magic (+)
- Shield spell was moved from level 1 to level 2 (this is done in order to “dilute” the 2 level of the Mage Guild, oversaturated with very useful spells)
- changed the cost of the spells Hypnotize (10 instead of 15), Dimension Door (20 instead of 10), Shield (5 instead of 3), Mass Shield (10 instead of 7), Slow (6 instead of 3), Haste (5 instead of 3), Mass Haste (12 instead of 10), Blind (7 instead of 6), Animate Dead (12 instead of 10), Town Gate (12 instead of 10), Town Portal (30 instead of 20), Set Guardian... (12 instead of 15)
- spell Haunt is disabled in Shrines on map
- shooting towers attack parameter was increased by 2, also damage was increased (from 2-3 to 2-4)
- the starting armies of heroes have been changed; the Knight has 20-40 Peasants and 3-5 Archers, the Barbarian has 10-15 Goblins and 3-5 Orcs, the Sorceress has 8-12 Sprites and 2-3 Dwarves, the Warlock has 4-6 Centaurs and 2-3 Gargoyles, and the Wizard 4-6 Halfling and 2-3 Boars, Necromancer 4-6 Skeleton and 2-3 Zombies
- the restoration of movement points for heroes fleeing the battlefield and redeemed on the same day was removed (now as in Heroes 3)
- the starting number of creatures that join free of charge at map objects has been changed; Halflings 15-30 instead of 20-40, Archers 10-20 instead of 10-25, Goblins 15-30 instead of 15-40, Sprites 10-20 instead of 15-25, Dwarves 8-15 instead of 10-20
- the starting number of creatures that join for charge at map objects has been changed; Trolls (Troll bridge) 4-8 instead of 4-6, Power Liches (Dead City)  4-8 instead of 4-6
- the week growth of creatures that join for charge at map objects has been changed; Trolls (Troll bridge) 2-4 instead of 1-3, Power Liches (Dead City)  2-4 instead of 1-3, Nomads (Desert Tent)  2-4 instead of 1-3, Rogues (Wagon Camp)  4-8 instead of 3-6.
- artifact Statesman's Quill moved from level 1 to level 3
- reduced prices for rare resources for the construction of magic guilds 2-5 (each by 1)

Auxiliary modules:
------------------------------------------------------------------
H2_42_ProperRandomArts - plugin for PoL 2.1 version - denied random appearance of so-called artifacts level 3+
(like PoL 2.0 game mechanics)
The author of the patch is AlexSpl

H2_71_AI_TeleportSpell - plugin that allows the AI ​​to use the Teleport spell in combat. The spell prob. in magic guilds has been increased compared to the original game.

H2_30_BugFixes_Misc
It includes various small fixes (module will be updated as soon as new fixes will be done):
- updating max. movement points of heroes when they go from shore to boat or vice versa (to better know where hero can go tomorrow)
- bugfix of AI interactions with monsters when AI have both Diplomacy SS and Necromancy (thanks to Arstahd)
- fix of Vision spell when hero have Diplomacy (proper number of joined monsters) (thanks to Arstahd)
- some bugfixes of Quick Combat

New modules with gameplay changes:
------------------------------------------------------------------
H2_83_Anti_RedRush.
- The weekly increase in the players' castles occurs only after the corresponding player has been given a turn. That is, if a player on the 1st day of the week captures castle of some player who play later, then he will not get newborn creatures.


It is possible that not all maps from the original game will be compatible with these changes (in the sense of an interesting game and the possibility of winning), although the vast majority will be compatible. Personally, I tested many maps and I can say that the game has become more thoughtful and interesting.

Very good work. Is it possible to have it as txt file to easy modding?!
And can be added other possibilities?
Eg. horde growth rate. Its the sane for peasantas and for centaurs even thought that centaurs are few times more powerfull?!
Well growth rate can be another for all creature.
And dwellings building order can be changed?! I need another dependencies.

Quote :

In addition, you need to replace the MSS32.dll file with the file of the same name from Heroes 3, or use GL Wrapper version 1.48 and newer (in this case, there is no need to replace the MSS32.dll file).

Wait, this is of great interest to us.

What does using the newer version of MSS do?

Is this something we should add to the Ironfist installer?

MSS loads ASI files located in game root folder (you can make ASI file by renaming extension of DLL file).
Then my Loader.asi (author - AlexSpl) loads DLLs from Plugins folder.

Original MSS from Heroes 2 does not make ASI-loading, but DDRAW.dll from GL Wrapper does it.

BTW, in this project I used game structures from Ironfist project, I mean your raw_decompiled folder (thanks !).

Heroes III has ASI-loading built-in? Wut?

I'm having a closer look now. This seems like really nice work. Would you be willing to share the code with us? We can port some of the same changes into Ironfist.

Darmani wrote:

Would you be willing to share the code with us?

Yes, there is no problem to share code. Today I will provide link to source code.

***

I sent you PM.

I have used API of patcher_x86 (patcher_x86.dll (free distributed) should present in game root folder). Author is baratorch (thanks to him !)

Ben80 wrote:

1) H2_48_SpellsBugFixes – bug fixes related to spells:
- for the Bloodlust spell (duration is determined by the power of magic, not equal to 3)

After thinking about the duration of this spell for a long time, the manual also states "Duration: 1 rnd / Power", I think that this spell is "dedicated" to the Knight and Barbarian heroes who can cast this spell for three turns with only 1 in Spell Power.
For my part, and for this reason, I would leave it with a duration of three turns.

Ben80 wrote:

1) H2_48_SpellsBugFixes – bug fixes related to spells:
- for massive spells (sometimes they did not work even for creatures without immunity, if there were creatures with immunity on the battlefield)
- for Earth Elemental creatures - invulnerability to Meteor Shower spell was removed (additional damage was already in the original code), added invulnerability to Elemental storm (according to NWC guidelines).
- for spellcasting by AI Dragon Slayer(there was a bug)
- for spellcasting by AI Shield/Mass shield(there was a bug)
- for spellcasting by AI Slow/Mass slow(there was a bug)
- for spellcasting by AI Haste/Mass haste(there was a bug)
- for spellcasting by AI Curse/Mass curse(there was a bug)
- in versions 2.0 Eng, 2.1 Rus, 2.1 Pl, 2.1 Cz – for spellcasting by AI Dispel/Cure/Antimagic(there was a bug)

All these bugs fixed, nice.

Ben80 wrote:

2) H2_49_AI_NecromancyFix –correction of the number of skeletons raised after the battle with neutral monsters by AI heroes (in the original, only 1 skeleton was raised). This plugin should probably not be used if you are playing without a DLL to improve game balance (where the number of skeletons is reduced from 10/20/30% to 5/10/15%), as AI heroes often accumulate very large amounts skeletons.

The fix that makes the AI accumulate more Skeletons is great.
I want the AI to be powerful with it.

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- secondary skill Necromancy gives 5/10/15% of skeletons (instead of 10/20/30%)

Regarding the change of the Necromancy percentage since the AI seems to take full advantage of it, I would leave the original percentages.
To be tested.

I also have options regarding Necromancy right here:
https://github.com/ihhub/fheroes2/discussions/3235
To be tested as well.

Ben80 wrote:

3)H2_70_AI_Imrovements –AI improvements. The most significant is the increase in “AI thinking radius” for the choice of actions on the adventure map. In the original, this number was very small, and the AI heroes were "blind"because of this. After increasing this number, the AI heroes added great aggression (they became more similar in behavior to the AI heroes from Heroes 3), and the gameplay became more interesting. Also quite significant is the improvement of the algorithm for choosing a spell in battle - the algorithm has become more adequate and adaptive to changing conditions.

To be tested.

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- Well gives +1 growth (instead of +2) for Green Dragons, Red Dragons, Black Dragons, Giants, Titans.

True that the power of these creatures should be reduced, but not wanting to reduce their growth of the original game, I have chosen another path, with in addition, an increase in growth of the other creatures.
https://github.com/ihhub/fheroes2/discussions/3001

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- secondary skill Eagle Eye gives a 40/50/60% chance to learn a spell (instead of 20/30/40%)

I'm afraid that's not enough to attract people to choose this secondary skill.
Here is my vision for this skill (among other skills):
https://github.com/ihhub/fheroes2/discussions/2990

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- secondary skill Archery does not allow you to avoid the penalty when shooting through the fortress walls

Ouch, a bit harsh.
I chose to give more possibilities to the defenders:
https://github.com/ihhub/fheroes2/discussions/2968
https://github.com/ihhub/fheroes2/discussions/2975
https://github.com/ihhub/fheroes2/discussions/2980
https://github.com/ihhub/fheroes2/discussions/2973
https://github.com/ihhub/fheroes2/discussions/2981

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- changed the cost of buying Red Dragons (4000 gold instead of 3500), Black Dragons (4500 gold instead of 4000), Ogres (350 gold instead of 300), Crusaders (900 gold instead of 1000), Liches (700 gold instead of 750), Power Liches (800 gold instead of 900).
- changed Ogres (speed 2->3, health 40->45)
- changed the cost of upgrading buildings Armory (1500 gold instead of 2000), Jousting Arena (2500 gold instead of 3000), Cathedral (3000 gold, 5 crystals, 5 wood instead of 5000 gold, 10 crystals, 10 wood), Archery Range (1000 gold instead of 1500), Stonehenge (1000 gold, 5 stones instead of 1500 gold, 5 mercury), Mausoleum (2500 gold instead of 3000), Ivory Tower (3000 gold instead of 4000), Cloud Castle (10000 gold instead of 12500), Adobe (2000 gold instead of 3000)
- changed the cost of building Cloud Castle (8000 gold instead of 12500), Laboratory (12500 gold instead of 10000), Adobe (3000 gold instead of 2000), buildings to increase the growth of the 1st level (750 gold instead of 1000), special building of the Knight (1000 gold instead of 1500), special building of the Sorceress (1000 gold instead of 1500), special building Barbarian (1000 gold) instead of 2000), special building of the Warlock (4000 gold instead of 3000), building of the Left Turret (1000 gold instead of 1500), building of the Right Tower (1000 gold instead of 1500), Moat building (500 gold instead of 750)

I think that changing the cost of buildings and creatures is not necessarily desirable, players are used to these values.
Also, adjusting the cost for some and not others seems a bit biased to keep the balance of the original game.
Personally, I like the slightly unbalanced "balance" of the original game, which allows for different choices and strategies, rather than a game where everything is absolutely equal, in which case there might as well be only one castle.
I do have a planned option for creature costs for players who would like a little more difficulty but I haven't posted it yet.

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- Hypnotize spell was moved from 5th level to 3, the spell does not end after an attack (as in Heroes 3)

True, that this spell is completely useless in Heroes 2, and makes the player who has painfully managed to build a level 5 Mage Guild after countless efforts cringe.
Here is my vision for this spell:
https://github.com/ihhub/fheroes2/discussions/2517
https://github.com/ihhub/fheroes2/discussions/2785

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- damage of Cold Ring, Fireball, Fireblast spells increased by one and a half times

I thought of the same changes for these, and I have other changes for other spells as well:
https://github.com/ihhub/fheroes2/discussions/2994

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- changed the chances of appearing in the Mage Guild of the spells Dragon Slayer (+), Cold Ring (-), Cold Ray (-), Fireball (-), Lightning Bolt (-), Magic Arrow (+), Anti magic (+)

Are the chances of spells appearing in the Mage Guild for Heroes 2 documented somewhere?
I've never been able to find them and I'd like to know what exactly they are for all spells.
Is there also something similar for the Shrines and the Pyramid?

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- Shield spell was moved from level 1 to level 2 (this is done in order to “dilute” the 2 level of the Mage Guild, oversaturated with very useful spells)

I think in the original game, this is done so that the AI heroes always get damage spells easily in its Mage Guilds, plus the AI is less likely to get adventure spells in its Mage Guilds if I am not mistaken.

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- changed the cost of the spells Hypnotize (10 instead of 15), Dimension Door (20 instead of 10), Shield (5 instead of 3), Mass Shield (10 instead of 7), Slow (6 instead of 3), Haste (5 instead of 3), Mass Haste (12 instead of 10), Blind (7 instead of 6), Animate Dead (12 instead of 10)

Ouch, moreover I think it will handicap the AI who always casts a spell when it's its turn even if it's not necessary consuming even more magic points.

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- the bonus to the attack of shooting towers has been changed (+2 to attack for each level of the Mage Guild instead of +1)

To be tested.

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- the starting armies of heroes have been changed; the Knight has 20-40 Peasants and 3-5 Archers, the Barbarian has 10-15 Goblins and 3-5 Orcs, the Sorceress has 8-12 Sprites and 2-3 Dwarves, the Warlock has 4-6 Centaurs and 2-3 Gargoyles, and the Wizard 4-6 Halfling and 2-3 Boars, Necromancer 4-6 Skeleton and 2-3 Zombies

For what purpose this has been changed?

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- the restoration of movement points for heroes fleeing the battlefield and redeemed on the same day was removed (now as in Heroes 3)

Okay.

Ben80 wrote:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- the starting number of creatures that join free of charge at map objects has been changed; Halflings 10-20 instead of 20-40, Archers 10-15 instead of 20-25, Goblins 15-25 instead of 15-40, Sprites 10-20 instead of 15-25, Dwarves 7-10 instead of 1-21, Skeletons 10-20 instead of 10-25

Is there also a documentation somewhere for this?
Is it normal for Dwarves: "Dwarves 7-10 instead of 1-21"?
What about Peasants, Centaurs and Orcs?

In any case, a lot of changes for the original game program, kudos.
To be tested when I have some time.

Unknown_Hero wrote:

Regarding the change of the Necromancy percentage since the AI seems to take full advantage of it, I would leave the original percentages.
To be tested.

I was never talented in skeletons accumulation (neither in Heroes 2, nor in Heroes 3), however after observing the AI,  I decided to lower the percentage.
Besides, I've heard that skilled players also easily collect a lot of skeletons.
The fact is that in Heroes 2 there are quite a lot of monsters on the map with the number of "Lots". By the way, in Heroes 3, the standard number of monsters for author's maps is less, so there 10/20/30% is normal. In Heroes 3, problems with Necromancy began after the introduction of RMG,  in which monsters with a large number.

Unknown_Hero wrote:

I'm afraid that's not enough to attract people to choose this secondary skill.
Here is my vision for this skill (among other skills):
https://github.com/ihhub/fheroes2/discussions/2990

I regard this skill as secondary anyway (like Luck SS, for example).
The problem with this skill is that it doesn't fit well with the actual gameplay.
It's pretty easy to capture someone else's castle and learn spells there.
Therefore, the main efforts were made precisely to improve the gameplay.
AI now began to use Slow, Haste and other more often, which it almost did not use in the original game. Capturing castles has become more difficult (and there will still be some work in this direction).  Also, my AI doesn't rebuild villages into castles that often.
(Usually rebuilds when there are matching double builds, so maps like Pyramid, Forsaken Lands will work fine).

Unknown_Hero wrote:

Ouch, a bit harsh.

Quite classic - like in Heroes 3. I think it was an improvement on the part of the NWC.

Unknown_Hero wrote:

I think that changing the cost of buildings and creatures is not necessarily desirable, players are used to these values.
Also, adjusting the cost for some and not others seems a bit biased to keep the balance of the original game.
Personally, I like the slightly unbalanced "balance" of the original game, which allows for different choices and strategies, rather than a game where everything is absolutely equal, in which case there might as well be only one castle.

I also like this approach (a little unbalanced). Besides, as we remember Dark side should be stronger ))
Therefore, I did not pursue the goal of creating some kind of perfect balance.
However, the decision with a decrease in the growth of Dragons and Titans is simply obvious.
I try to make decisions based on game analitycs. For example,  I compared the construction costs of the main buildings and their upgrades and compared what the weekly increase of fight_value it gives.
It turned out that the main buildings are much more profitable in this regard. It is normal, it is by game design. However, in some cases, the upgrades were very disadvantageous (for example, for the Upg. Ivory Tower).  In such cases, I used adjustments.

Unknown_Hero wrote:

True, that this spell is completely useless in Heroes 2, and makes the player who has painfully managed to build a level 5 Mage Guild after countless efforts cringe.

This spell, as it appears in Heroes 3, is usually used to remove retaliation from the opponent's strong stack. I also use this spell a lot for combats with shooters (neutral monsters).

Unknown_Hero wrote:

Are the chances of spells appearing in the Mage Guild for Heroes 2 documented somewhere?
I've never been able to find them and I'd like to know what exactly they are for all spells.
Is there also something similar for the Shrines and the Pyramid?

Yes. https://handbookhmm.ru/kakim-obrazom-zaklinaniya-popadayut-v-magicheskuyu-gildiyu.html#more-4210
(there is English version of this site)
For Shrines and Pyramids process very different - there all spells are equally probable.

Unknown_Hero wrote:

I think in the original game, this is done so that the AI heroes always get damage spells easily in its Mage Guilds, plus the AI is less likely to get adventure spells in its Mage Guilds if I am not mistaken.

I think that the 2nd level of the mages guild is very problematic in the game.
There are quite a few combat spells out there, and very little variety.
The AI (and human also !) ​​often easily gains access to the Blind spell, etc.

Unknown_Hero wrote:

For what purpose this has been changed?

I've noticed that often building upgrades are not competitive compared to buying a new hero. To increase the selection space and make the game more interesting, that's what I did (reduced the size of the army that purchased heroes have). Besides, as you can see, the Knight suffered the least.

Unknown_Hero wrote:

Is there also a documentation somewhere for this?

Yes. https://handbookhmm.ru/2-objects

Unknown_Hero wrote:

What about Peasants, Centaurs and Orcs?

They are ok in original game

Ben80 wrote:

Besides, I've heard that skilled players also easily collect a lot of skeletons.

Yes, it's quite easy to get thousands of them.
But beyond a certain number it doesn't really matter anymore that the Skeletons troop can do one or four times the number of damage points to decimate an opponent troop.
But the player can always split them into several troops to optimize the damage, and also leave them in defense in his castles or divide them between different heroes to speed up the end of the scenario.

Ben80 wrote:

Also, my AI doesn't rebuild villages into castles that often.
(Usually rebuilds when there are matching double builds, so maps like Pyramid, Forsaken Lands will work fine).

However, building castles is a good way for the AI to get more troops, especially in the higher difficulty levels where the AI has bonus in gold and resources.

Ben80 wrote:

I also like this approach (a little unbalanced). Besides, as we remember Dark side should be stronger ))

Yes, absolutely.

Ben80 wrote:

However, the decision with a decrease in the growth of Dragons and Titans is simply obvious.
I try to make decisions based on game analitycs. For example,  I compared the construction costs of the main buildings and their upgrades and compared what the weekly increase of fight_value it gives.
It turned out that the main buildings are much more profitable in this regard. It is normal, it is by game design. However, in some cases, the upgrades were very disadvantageous (for example, for the Upg. Ivory Tower).  In such cases, I used adjustments.

I see your point.

Ben80 wrote:

Yes. https://handbookhmm.ru/kakim-obrazom-zaklinaniya-popadayut-v-magicheskuyu-gildiyu.html#more-4210
(there is English version of this site)
For Shrines and Pyramids process very different - there all spells are equally probable.

Thanks for the link and the clarification, I will be able to take a look into it.

Ben80 wrote:

I think that the 2nd level of the mages guild is very problematic in the game.
There are quite a few combat spells out there, and very little variety.
The AI (and human also !) ​​often easily gains access to the Blind spell, etc.

Also, I think that the Shield spell, as a level 1 spell, can be useful at the very beginning of the game against ranged wandering creatures, and is a bit less useful afterwards as it is more situational in combats between heroes (and therefore even less useful as a level 2 spell).
But, I understand your reason to decrease the chance of appearance of level 2 spells.

Ben80 wrote:

I've noticed that often building upgrades are not competitive compared to buying a new hero. To increase the selection space and make the game more interesting, that's what I did (reduced the size of the army that purchased heroes have). Besides, as you can see, the Knight suffered the least.

I see.

Ben80 wrote:

Yes. https://handbookhmm.ru/2-objects

Thanks for this link too.

Ben80 wrote:

They are ok in original game

Okay.

If we assume that a player raises 120 skeletons in a week, then this is equivalent to about 2 bone dragons.
Even if we accept that he lifts 60 skeletons, then this is a good increase. I am sure that even after my changes in Necromancy (and by the way in Evil shrine and in the necromancer artifact) Necromancers are still very strong in the late game (exactly as the NWC manual says).

Unknown_Hero wrote:

However, building castles is a good way for the AI to get more troops, especially in the higher difficulty levels where the AI has bonus in gold and resources.

This is of course my crutch decision (as in the case of the captain of the guard from the NWC ) that simply reflects the fact that the AI ​​is not smart enough.
But you can try to play and you will see that it is much more difficult to play now

Unknown_Hero wrote:

Also, I think that the Shield spell, as a level 1 spell, can be useful at the very beginning of the game against ranged wandering creatures, and is a bit less useful afterwards as it is more situational in combats between heroes (and therefore even less useful as a level 2 spell).
But, I understand your reason to decrease the chance of appearance of level 2 spells.

In fact, after moving the Shield to level 2, I started to get it and use it more often than before. I also like this spell.
(at 1st level, it rarely appear, as there are many other combat spells)

Ben80 wrote:

This is of course my crutch decision (as in the case of the captain of the guard from the NWC ) that simply reflects the fact that the AI ​​is not smart enough.
But you can try to play and you will see that it is much more difficult to play now

I started the scenario Broken Alliance, Difficulty: Normal, Player: Blue, random castle, with your full mod.
I got Barbarian, Mage Guild with Haste and Bloodlust.  
With Bloodlust, you have to be very careful to cast it at the right time (only one turn, and current turn only, with a hero who has 1 Spell Power).
The Haste spell can be cast twice only (five spell points instead of three, with a hero who has 1 Spell Power), I'll see how it goes.
I've only done a few turns so far, and I won't finish the scenario before several days.

Thank you for attention to the mod. BTW, I'm impressed about your work for bug finding in original game. I hope someday I will fix those bugs.

Ben80 wrote:

Thank you for attention to the mod.

Thanks to you.
Also, if mss32.dll (or/and GL Wrapper needed file) is free to use (I don't know), why not include it directly in your package?
I had first tried with a more recent version of mss32.dll and got an error.
With the version corresponding to Heroes 3, everything works directly.

Ben80 wrote:

BTW, I'm impressed about your work for bug finding in original game.

Thanks, it was a lot of work.
I had accumulated saves with bugs over the years of playing the game.
It was also a way for me to free my mind from those bugs now that they are reported somewhere.

Ben80 wrote:

I hope someday I will fix those bugs.

It would be nice, good courage if you decide to do it.

Unknown_Hero wrote:

Also, if mss32.dll (or/and GL Wrapper needed file) is free to use (I don't know), why not include it directly in your package?

I have added link to MSS32.dll from Heroes 3.

So I finished the Broken Alliance scenario.

Some comments on the game in progress (already reported in PM).

One of the Purple AI player's heroes surprised me with a large army on normal difficulty Month 2 Week 1.
After capturing the central island town (Peacekeeper), where he certainly learned the Dimension Door spell (the Mage Guild of this town contains this spell)(and while one of my heroes was heading to that same town to capture it), the purple hero used that spell to capture two of my castles in three turns (including my starting Blue player castle).
I think if the spell had cost 10 spell points instead of the modified 20 spell points, he could have done it in two turns.

The AI also uses the Shield spell more when needed against shooters (but not always in an optimal way, for example when the shooters do not have a high damage potential).
The three-point Slow spell is sometimes wanted by the hero in combat for small spellcasters, as well as the Haste spell.
And the Bloodlust spell is much less effective for the Barbarian hero I play.

I also noticed a problem with the Cure spell that the AI cast on a troop of Peasants (with no active spell).

More comments on the game.

As expected, AI Purple player has the Dimension Door spell in his starting castle in the south east of the map.
That's why his starting hero can leave the island because he has Expert Wisdom.
Heroes that AI Purple player recruits later cannot leave the island because there is no way to gain experience on the starting island to learn Wisdom at Expert level.

On his island the AI Purple player has recruited several heroes, including one with a large army who guards the castle.
The other heroes are a bit superfluous as they cannot leave the island or gain experience and they have creatures that would have been better used on the defending hero to have an even bigger army.

I found the Ultimate Artifact in one try, on a partially discovered map, the Ultimate Wang of Magic, +12 Spell Power.
The AI Purple player's hero didn't stand a chance with his 12 Titans against my hero with 9 Black Dragons, Armageddon spell, 25 Spell Power and 260 spell points.
Even though only one Armageddon was needed for each of his four heroes on the island.

AI also uses Dragon Slayer spell against Dragons.

Plugin that allows the AI ​​to use the Teleport spell in combat.
The spell probability in magic guilds has been increased compared to the original game. link

Ben80 wrote:

Plugin that allows the AI ​​to use the Teleport spell in combat.
The spell probability in magic guilds has been increased compared to the original game. link

Nice.
Can this plugin be added using a save game from an already started scenario, or do you have to start a scenario from the beginning for the improvement to be taken into account?

Unknown_Hero wrote:

Can this plugin be added using a save game from an already started scenario, or do you have to start a scenario from the beginning for the improvement to be taken into account?

One can play saved game and AI will use Teleport, but only when AI have it. We know that in original game prob. of Teleport in GM is quite small. Spells go in GM at very beginning of the game and plugin do not change already generated spells in GM.
I would not advice to play only to see new AI spell. May AI Teleport will be good surprise during some usual game

Ben80 wrote:

One can play saved game and AI will use Teleport, but only when AI have it. We know that in original game prob. of Teleport in GM is quite small. Spells go in GM at very beginning of the game and plugin do not change already generated spells in GM.
I would not advice to play only to see new AI spell. May AI Teleport will be good surprise during some usual game

I started the Dominion scenario (PoL), impossible difficulty, Blue player as Knight, with your full mod and I will continue this game by adding the Teleport spell plugin.
I will see if any of the AI heroes will use the spell during a combat.

New small module is available - H2_30_BugFixes_Misc
It includes various small fixes (module will be updated as soon as new fixes will be done):

- updating max. movement points of heroes when they go from shore to boat or vice versa (to better know where hero can go tomorrow)
- bugfix of AI interactions with monsters when AI have both Diplomacy SS and Necromancy (thanks to Arstahd)
- fix of Vision spell when hero have Diplomacy (proper number of joined monsters) (thanks to Arstahd)

Download: link

Hello @Ben80,

A note about hero movement with your plugin:

H2_30_BugFixes_Misc
- updating max. movement points of heroes when they go from shore to boat or vice versa (to better know where hero can go tomorrow)

If a hero with a sea move bonus, due to Lighthouses (or due to artifacts or due to the secondary skill Navigation), flees from a sea combat, he can be recruited with all the extra movement points due to the bonus of these Lighthouses (or artifacts, or secondary skill Navigation).
The hero can then travel an impressive distance on land.
Then if he fights on land, he does not get this sea bonus when he is recruited again.
I don't know if this is the desired behavior, but I think it's a little weird and goes against this change:

4)H2_80_GameplayImprovements –gameplay and balance improvements.
- the restoration of movement points for heroes fleeing the battlefield and redeemed on the same day was removed (now as in Heroes 3)

**Edit.**

In fact it's the same with or without the H2_30_BugFixes_Misc plugin.

Hello @Unknown_Hero,

Max. movement points is different from remaining movement points. Max. movement points is used only to see how far the hero can go tomorrow, the day after tomorrow, etc. (brown arrows on the map).
In the original game, when, for example, the hero gets into the boat, this value is not recalculated, so usually the next day the hero can swim further than the brown arrows show. This small bug has been fixed in this plugin (for better planning of further game etc).

Hi @Ben80,

Ben80 wrote:

Hello @Unknown_Hero,

Max. movement points is different from remaining movement points. Max. movement points is used only to see how far the hero can go tomorrow, the day after tomorrow, etc. (brown arrows on the map).
In the original game, when, for example, the hero gets into the boat, this value is not recalculated, so usually the next day the hero can swim further than the brown arrows show. This small bug has been fixed in this plugin (for better planning of further game etc).

Yes, but here is what happens now with all the plugins of the mod in the plugins folder (English version Gold 2.1 Windows 7 64 bits):

https://youtu.be/dwpiOhFrPSQ

Movement bonuses related to the sea should not be converted into movement bonuses on land and vice versa (so at most 1500 movement points can be converted).

When the hero moves on the sea, the movement points given by the bonuses (Navigation, Artifacts, Lighthouses) are consumed first.
For the remaining points, when recruited, the hero should only be able to move as much as the slowest troop in his army.

Example:
The hero is in a boat, his slowest troop allows 1000 points of movement on land.
The slowest troop is not taken into account on the sea:
1500 points of movement on sea.
The player controls five Lighthouses:
2500 movement points on sea.
The hero has Expert Navigation:
1500 movement points on sea.
The hero has an artifact:
1000 points of movement on sea.
The hero can move a total of:
1500 + 2500 + 1500 + 1000 = 6500 points of movement on sea.

First case:
The hero moves on the sea by 4000 movement points.
The hero participates in a sea combat and flees during the battle.
The hero has 2500 movement points left on the sea.

The hero is recruited, the number of movement points (2500) is higher than the maximum movement on land with his slowest troop (1000 points on land), the hero should have 1000 movement points on land (not 2500).

Second case:
The hero moves on the sea by 6000 movement points.
The hero participates in a sea combat and flees during the battle.
The hero has 500 movement points left on the sea.

The hero is recruited, the number of movement points (500) is not higher than the maximum movement on land with his slowest troop (1000 points on land), the hero should have 500 movement points on land.

Hi @Unknown_Hero,

Unknown_Hero wrote:

Yes, but here is what happens now with all the plugins of the mod in the plugins folder (English version Gold 2.1 Windows 7 64 bits):

https://youtu.be/dwpiOhFrPSQ

Movement bonuses related to the sea should not be converted into movement bonuses on land and vice versa (so at most 1500 movement points can be converted).

Plugin does not do any conversion of MP sea to ground (or vice versa), it does not change movement points. It only does exactly what I have described in my posts.

I do not know why all this happens on your video - it does not have any relation to my mod.

I do not know whether all this can happen in original game, but if one would trust your video - then one should think that code of original game do all this.

You wrote "is what happens now with all the plugins of the mod in the plugins folder", but probably you forgot to add "also it happens without any of the plugins of the mod in the plugins folder".
Which fundamentally changes the essence of the matter, right?

@Ben80.

Ben80 wrote:

Hi @Unknown_Hero,

Unknown_Hero wrote:

Yes, but here is what happens now with all the plugins of the mod in the plugins folder (English version Gold 2.1 Windows 7 64 bits):

https://youtu.be/dwpiOhFrPSQ

Movement bonuses related to the sea should not be converted into movement bonuses on land and vice versa (so at most 1500 movement points can be converted).

Plugin does not do any conversion of MP sea to ground (or vice versa), it does not change movement points. It only does exactly what I have described in my posts.

I do not know why all this happens on your video - it does not have any relation to my mod.

I do not know whether all this can happen in original game, but if one would trust your video - then one should think that code of original game do all this.

You wrote "is what happens now with all the plugins of the mod in the plugins folder", but probably you forgot to add "also it happens without any of the plugins of the mod in the plugins folder".
Which fundamentally changes the essence of the matter, right?

If the plugins folder is removed, the behavior explained above does not occur (the hero with a very large number of movement points on land).
I made tests, if the plugins folder is put back without the plugin "H2_80_GameplayImprovements.dll", the behavior explained above does not occur (the hero with a very large number of movement points on land).
Hopefully this helps.