MMH5.5 - Heroes of Might and Magic 5.5

2dredknight

2% skill: by level 21 the overall chance to catch the skill is about 45% (0.65/0.67 ^ 20 = 0.55; 0.67 means that, on average, we already have several skills filled with a total weight of 33%, i.e. the 2% is selected from the remaining 67%)
3% skill: about 60%
4% skill: about 71%
5% skill: about 79%
6% skill: about 85%
8% skill: about 92%
10% skill: about 96%
11% skill: about 97%
12% skill: about 98%
16% skill: about 100%
18% skill: about 100%

In other words, starting from 5%, the chance to catch the needed skill is basically quite high (over 80%), but with a significant difference in odds, the difference in how quickly you catch the skill (at early levels) will still be felt.

In theory all the math is correct but in practice things are a little different.
Lets say you play a game and get up to level 2 where you get "War Machines".
The skill by itself is completely rubbish you need to level it up and get some of the perks to start doing something.
It is better to level up "War machines" up to the desired point on the next few levels then to pick 3 new skills.
Why is it better?
Heroes V level up secondary skills mechanics works the following way. When a hero levels up he gets 4 choices to pick from.
- 2 of them are perks
- 2 of them are skills where one is upgrade of an known secondary skill and the other is a secondary skill the hero does not yet know.

So on level up you get proposed precisely one new secondary skill. The math on the example above works for the new skills exactly as you say.
But what is bothersome is how the second upgrade skill offering works. The one for the already picked skills.

The proposal you get for upgrade of the already existing skills does not care about the % distribution. The game always offers to upgrade one of the 2 latest picked new skills.
So lets assume you play "Engineer" class fortress and want to max out his starting skill "War Machines" early so the hero can defeat creatures on the map more efficiently.
Hero levels up to level 2 and gets proposed "Offense" as a new skill. Thats a good skill lets pick it.
Hero levels up to level 3 and gets proposed "Defense" as a new skill.Thats a good skill lets pick it.
The problem you have now is that on the 4th level up you will almost never get "war machines" to upgrade.

The same is valid for "Perks". Heroes get proposed perks from the latest learnt or upgraded secondary skill (if available).

So math does not work in the long run. Same rules go for skills like "Summoning" and "Destruction" in the mage department. They have to be maxed out fast for maximum mana/damage efficiency.
The reason why the % are so near is that even if you play the same hero class each game it is very likely the hero will get more than 1 or 2 choices for new skills on level 2. This is good because not all heroes in a class are fit to one "best" strategy for the class.

1) All the listed classes have about 45% Spellpower and about 30% Knowledge; it would be possible, at the stat level, to give less Spellpower somewhere (30%) but more Knowledge (45%)
2) If Sorcery is mandatory for all magic classes, Enlightenment is not (it can be moved into rare skills somewhere); the Summoning + Destructive pairing can also be split up with other schools of magic to avoid so much repetition

1) True. The easiest way to explain this is with the 3 classes - Might are the weakest in magic, Balanced is better in magic, Magic utilize the strongest magic.
Magic strength depends on Spellpower so the more Mage is the class the more spellpower.

But lets elaborate a bit more.
Might class has 2 main skills Attack and Defense and they are almost equally balanced at some point in the future when the big stacks come becouse damage caused and survivability gained is proportional.
You can have hero with 40 Attack and 0 Defense or the other way around and he will still do good enough.

Magic heroes does not care about the army though. Unlike army damage and army endurance they have Magic damage and Magic endurance.
Magic damage is the average damage/effect strength per cast they do.
Magic endurance is the average time they can cast magic through the battle.

Throughout the game if Magic heroes and Might heroes carry their whole army in PvE battles Magic heroes suffer more casualties because due to low protective skills their army take hard hits if they miscalculate.
This is why as a Magic play you carry only what army you need for the battle and rely on your spells to get the opponent.
Whenever you go out of mana your are done. You may still win but casualty number will be great.
Some builds and some battles (vs creatures) are a bit of gamble for the Magic heroes.

Getting back on the skills now. Knowledge and spellpower are not equally distributable as Attack and Defense. There are a few rules that are required to be kept in order Magic to be good enough.
Magic heroes of a certain level have to make more damage/impact per cast then their Might alternatives of the same level hitting with a sword.
This is important because Might heroes damage comes for free over time with the level. If the "hero melee" hit makes more damage than a spell the Mage will be forced to hit instead of cast which is not good however you look at it.
This is why we are forced to keep spellpower % gain high.

Let me explain this another way.
1 attack/defense stat gives 3.33% bonus for the whole game. But even if that stat does not increase for months it still gets better over time because the hero army stack increase with time.
On the other hand spellpower efficiency drops every week because all stacks globally become stronger (Players and map creatures).

So comparing plain numbers and % stats in here is very subjective because there are so many in game factors.
As a matter of fact I can say that heroes franchise may never be balanced, ever. Because the core hero and classes balancing anywhere (might and magic 3,4,5,6,7 etc) is very dependable on the map you play.
For example if you make a 2 player map where opponent castles are a week distance from each other without border guards Magic heroes would own 100% of the time.
Early on Might stats does not count for nothing because there is no army. Also hero melee hits kill like 2-3 tier 1 units which is nothing.
On the other hand Magic heroes will blast all their mana and after that their strength will be as competitive as that of the might hero because they will do the same damage per melee hit and primary stat difference will be negligible for the week 2-3 army size.
How hero development interacts with the maps and the zones inside is something very seriously taken in MMH55. Credit goes to Magnomagus.

I believe the ARMG is about 30 000 lines of code.

P.S. I want to add here that there is one Magic hero class with more knowledge than spellpower gain. This is the "Artificier" class.
Academy hero classes go as this "Balanced" "Magic", "Magic". This is how their primary skill distribution go as well - 50/50,25/75,20/80
They compensate for the Might skill loss by transforming the knowledge in Attack and Defense stats through the creature artifacts.

(About Might heroes with Shatters) It is not so much blasphemy as the fact that these skills are very specific/situational; it is hard to make them attractive enough at the level of ordinary skills.

That was we thought as well. Shatters were actually picked around the mid game when you start preparing for the PvP attack. It was really unpleasant that those skills offer no bonuses before the PvP.
Another think we thought about is that if you go shatters and do not pick magic schools the Magic Guild building becomes unusable.

This is how the idea about summoning prime elements was born. If a hero has an advanced (or higher) shatter he gets free elementals in all his castles at the begining of every week.
The number of the elementals is multiplied by the level of the mage guild. Also multiple shatters can be combined for a greater effect.

Some people were a bit sceptical about the usefulness of the new ability in PvP games but after trying it out it was actually overpowered!
We head Stronghold heroes with 4 shatters getting elementals and selling them on the marketplace for huge amounts of gold (I think one elemental costs around 800?) and their economy exploded.
Even if they don't sell the elementals they just stomped you over with them.
Now the current setting is still ok for PvP but not that empowering. There are also some common artifacts that further enforce the strength of that summoning so it is a viable strategy to use an elemental instead one of the six faction creatures.

Here is a good reply from Magnomagus as well.

Shatters are added to all factions to replace skills that were in TOE completely useless and didn't fit the class from a lore perspective. For example a knight cannot use summoning and destructive properly because it doesn't develop the required spellpower. It also just looks weird. H4 and H3 had generic spellschool defenses as spells, so it is just a minor step to make it skills, nothing new or blasphemy. A knight with destructive magic is blasphemy. A paladin with both light and shatter dark is perfect harmony.

That is not so:
- there is Magic Resistance (15% chance to reflect a negative spell);
- there is Magic Proof (15% reduction to magic damage);
- there are bonuses to health (+X% to health)
- there are specialized things like: Seal of Light, Seal of Darkness, Weakness of Darkness, Elemental Balance, Banish

If there is a problem with early adepts of the Destructive and Summoning schools, it can be addressed differently: significantly raise the costs of top-tier spells of levels 3-5, so that they are oriented toward the late game (and require a lot of Knowledge). Low-level spells deal less damage.

Lets say we exclude shatters and take the proposals above one by one.
- Magic resistance % and Magic proof - I am not going to discuss how they work but they are not enough. Occultism penetrates both plus they do not cover most of the Summoning or Light spells.
Also lets imagine they were good enough. It is not fair against the Magic heroes as the Might will spend 2-3 secondary skill points to protect against all magic.
Shatters make you invest in something. Also default shatter perks are not that useless anymore. The new battle scheme allows the battle to last longer thus Mage heroes cast spells (which are now more expensive too) more times during the battle.
Shatter perks make them either burn more mana, cut spellpower efficiency or make them cast less often. If you get one of those things correct in PvP you already make the opponent mage suffer.

High level spell costs are significantly increased. Also some high level spell efficiencies were modified appropriately.

What is bad here is that this ability effectively requires leveling War Machines, otherwise it performs many times weaker. On the other hand, when leveling the Ballista, it is very important to level Attack with Archery/Flaming Arrows, since the damage ramps up many times over. In other words, there is too rigid a link between Attack and War Machines (the Ballista). That said, it was like this in the original too, so this is just my personal opinion.

It will simply be too powerful if they were in the same branch. Tent and Catapult will never be picked. Basically here I can answer the same way I did for "Retribution" the synergy is too good.

+6 is the theoretical maximum (the old maximum was above +10). Real bonuses will be around +1-2 on average. This is an unattractive ability; the information has been checked and proven in practice (there is a special Random Tactical Arena format where players compete in leveling, and nobody wanted to level such an ability because it performed weakly).

the aim is to benefit the hero with around +2. This is how much all perks should give on an average.

Have you tested this ability? In the original game it worked only for Orcs and did not protect any other creatures from magic.

This was reported a long time ago. May be it is fixed in 3.1 I don't know. But I have tested it before and it works.
I will do another test when I have time the next few days and will get back to you.
Magnomagus for sure have tested it. He double, triple checks everything before it is included inside.

Luck is stronger, because Expert Offense+Archery gives 20% damage (4 abilities), while Expert Luck gives the same 20% with only 3 abilities.

True but luck is a chaotic skill. You can get luck for a hit with your weakest unit or a hit with your strongest damage dealer, you never know.

At +3 luck - 20% to 30% (not 35%). At +2 luck (Advanced) - 13% to 20%. The prerequisite is situational Soldier's Luck, which not everyone needs.

Those are different things, 10% luck strike damage modifier is not 10% average damage increase.
Luck strike damage modifier is x1.66 . Now multiply it by 1.1 and you get ~1.83.
Now if a monster hits 10 times with 3 luck and he gets 3 lucky hits the damage formula with critical strike is 7*X + 3*1.83*X = 12.49X compared to no critical strike 7*X + 3*1.66*X = 11.98
So the hero gets roughly ~5% damage for that perk.

You already weakened Retribution from 5% to 2.5%; it can be weakened even further - to 2%, then the bonus will be between 4 and 10% (from +2 to +5 morale).

True. A lot of the changes Magnomagus has taken from the ToE Tournament edition version. This one is from there too.
From what he said a lot of the changes in the Tournament edition were made after a lot of testing so he really keeps most of them.
Anyway I will bring the topic up to see if he has something to say.

If it stays in Leadership, there is an interesting idea: give +1 initiative rather than speed. In any case this is a "final" perk, i.e. ultimately strong (both with +1 speed and with +1 initiative). Although you wanted to avoid "final" perks.
What is bad about giving +1 speed is that it greatly strengthens rush tactics. Run up and kill on the first turn. +1 initiative strengthens the army overall well, but it is not as useful for rushing as +1 speed. Finally, initiative is what the Leadership skill works with (triggered morale increases turn frequency, just like initiative).

True but just because Leadership synergies with initiative we do not want to give extra +1. More initiative means Leadership will work even better. And if you get this perk this means you have +5 Leadership so that 1 initiative is even stronger that it is.
Speed is OK as there are very few in-game speeds boosts overall (boots of speed for +1 is the other boost I think). Also the aura of speed require 2 not so great perks before that so its strength is compensated but the weakness of the required perks.

Also the all in strategy is not justified because damage boosters are nerfed throughout the game and this strategy just does not work anymore like it did in ToTe.
When I tested the game we did a lot iterations of every PvP battle where different strategies were tested just to be sure if they work and how good they work.
Probably the only faction that may actually do it is the Stronghold because blood rage protects them to some extend. But that is what they are made for.

I know that in Random Tactical Arena, in 100% of cases magic builds take Sorcery at 10/20/30%. In other words, this is a mandatory skill and does not need strengthening. Even when mana may run out (100-200), the skill still performs brilliantly.

There is sorcery + occultism + a magic school. This is already 18 levels of skill points. Besides those Enlightenment is very good skill of magic heroes. Those are 24 level points. Yes the combination is strong but the opponent Might hero will be strong enough to go through your army like butter.
Plus what if he has the shatter for that magic school? You got to pick second there is no escape.

Added after 50 minutes
[/U]
(regarding Combat - retaliation bonus) This skill is very, very weak. Its bonuses are effectively the same as Offense, but it works 4 times less often (since only 25% of melee damage comes from retaliations).
You can additionally give +1/2/3 to Defense, then it will be fine.
Or there is an alternative option: before each battle, additionally give +3/6/9 to the hero level (numbers as examples) for the duration of the battle (WarpHeroExp script). This will affect: damage from the hero attack; specialties; the power of some spells depending on level, including warcries. And it will not affect hero stats (even under Enlightenment) or abilities.

What you say only refers to half of the skill. Combat main purpose is to make the hero hit.
In MMH55 hero damage is increased significantly(I don't remember how much check the default stats). The purpose of the skill is the hero to hit. Creature reatliation is just bonus. It is like the hero is on the battlefield and leads the formation.
all hero classes that start with combat always have the middle perk "Martial arts" which makes half hero attacks do double damage.
I have seen heroes kill 100+ Dwarves with a single strike or 10 Archdevils.
The other perks in this skill tree provide great battle tactics variation.

Chain shot can be combined with some of your high init creatures that will make the hero punish the enemy harder while "Avenging strike" will prevent the opponent use one of his 7 units or suffer penalty damage.
For example it is quite often that "Avenging strike" is cast on the most important enemy stack or the archer stack. I have played against players who do not attack with those units to prevent damage.

The retaliation bonus and "preparation perk" are meant to give the skill alternative usage with combination of Defense. Where the hero and his army become a moving stronghold where the hero prefers to be more defensive and cast spells instead of doing a melee hit.

+1/+1 to Attack and Defense is noticeably weaker than +2 to Attack or +2 to Defense.
A bonus distributed between several stats is weaker than a bonus focused on one stat, even if they are stats from one group (like Attack and Defense, Spellpower and Knowledge).
For example, if we have a Might hero against a Mage, then the Defense bonus is not especially needed. And if a Mage is against a Might hero, then the Attack bonus is not especially needed.

Correct, this is why the perk is in the end of the branch. Take it if need it. Also the perk give +4 primary stats which is double compared to the average perk bonus.

Have you used Preparation often enough? Are you satisfied with how it works in practice? In the original game it was taken very rarely even without penalties (though it had the inconvenient Tactics requirement there).

Before it was nerfed there was a Haven, Paladin build as follows - Expert defense + "Stand your Ground", Combat + "Preparation", Expert Light + all mass perks or the "Guardian Angel" tree.
Haven units in combination with so much defense were nearly unbeatable. Hero cast blessing and has available 2 resurrects - his own and the Angel one. Even "Puppet Master" was not doing good job because the Paladin units had high defense but low damage so hitting each other did not helped a lot.
Same strategy is valid for Fortress Might which also combine this with Runes and high magic resistance.

That was a lot of versions ago though and a lot of things change since then. Preparation is good as it is now.

In practice, Distract works effectively only against the enemy hero, but not especially against neutrals. Also, if there is no Sorcery or mass spells (0.1), it does not work at all. If there are mass spells (0.1), it works as 10%. If there is Sorcery, then 12-15%. By the way, what is the reason for the knockback after mass spells being 0.1 rather than 0?
Update: I remembered that Distract throws the hero below 0, but only until the next unit's turn comes (at that moment the negative position on the scale is reset). So a 5-10% knockback happens in any case, even without Sorcery / mass spells.

"Distract" alternatives are "Dark Ritual" and "Consume Corpse" they are nearly on the same level of usefulness.
0.1 ATB for the mass spells is because of the hardcodded issue. There is no way you can put 0 value. It crashes the game and 0.1 is the lowest possible.

1.25 - with the basic skill leveled, without perks? Ok, I misread the numbers (thought it was 1.75).
Fine, then what about the maximum percentage: why not make it 25% at all skill levels, rather than 15/20/25%? After all, expanding gating from tiers 1-2 to 1-4 and 1-6 is already a sufficient strengthening of the skill; the 5% bonuses look out of place here. Especially since Master of Gating gives a 5% bonus, and in the current implementation it looks much weaker than simply leveling the base skill.

I agree with you about the 15/20/25% bonus. I will pass that to Magnomagus. I believe the reason behind this is because it was like this in ToE so it was just suited for the mod the same way.

This is a design mistake. If you push the game into long time frames (3 months or more), then you need to make sure Knowledge is just as full-fledged a stat as it is at ordinary time frames (1-2 months). This is done by increasing the costs of top-tier spells and/or increasing the mana regeneration rate (which depends on Knowledge).

The problem with knowledge is considered a little less stronger than the other skills is that its beneficial effects cannot transform in direct gain. Especially in the early game high knowledge does nothing.
So heroes with Knowledge has slower development path.

Knowledge effects are accumulated over time through artifacts or the ability of excessive spell casting. Unlike other skills mana does not helps that much 100 and 500 mana does not make difference. What it matters is do you have enough or not?
Besides the +2 knowledge bonus for a perk there are some small gains so the hero can catch up with the rest.

Also the level up bonus is not a big deal later on because unlike ToE the level gain curve is modified to be more smoother.
Now it is possible to get up to level 40 on one big map.

No, I do not really understand graphics - so if some portraits repel even me, they may put off many other players as well.
All I wanted to say is: add only those heroes whose portraits clearly win public approval from 90% of people.
Heroes 4 or 6 has a different visual style, so simply taking graphics from there and assuming everything is ok is not the best idea.

I don't want to sound rude but this is the first complaint regarding graphics (that I know). I may be wrong however. I am not a graphical guy too ;).

Regarding spellbook this is a bug in Quantomas Exe file. I will explain it below where you ask about versions.

So the modification is based on 3.0? And for what reason? There are problematic things, for example, in 3.0 "Flaming Arrows" ignore 100% of the target's defense (imbalance), and some settings for 3.0 are missing compared with 3.1.

Modification is installed the following way. Install heroes V, patch it to 3.1 and put Quantomas Exe on top.
Quantomas exe is based on a personal source code with personal bug fixes with huge amount of improvement. You can consider it as 3.06 version. it has some bugs from 3.0, but also some bugs fixed from 3.1.
It doesn't recognize most of the script commands from 3.1, It has 95% widescreen implementation (only the spellbook doesn't work).

As you can see "Flaming Arrows" description says it ignores defense so it is ok. It is not that strong because in PvP if they see you with a kick ass pumped balista it will die after the first or second round.
Balista is good for an early game rush on a smaller maps but it can still be countered. After a few games you get to know what each class is capable of so you get prepared for some of the strategy approach.

P.S.

Response to the second @Nargott post about Balista

Balista is indeed very strong but it dies in 1-2 hits in PvP. It is very vulerable against Destruction, Dark + puppet can be used to kill it too.
War machines "Balista" perk is T1 for a reason. This way the hero can pick "fist aid tent" + "Balista" + "Engineering" or all three "War Machines" perks - "tent" , "Balista", "catapult".
You are correct that balista is the best with fire arrows but sometimes you don't want to invest so much points in balista. Also it is very important to know that unlike other war machines Balista damage depends on hero attack and defense skills.
So yes triple balista is strong but not with 5-10 damage. Especially in late game I have sacrificed Defense and spellpower just to have a kick ass balista for PvE and PvP.

Probably one of the best alternatives to balista is the Fist aide tent because along the way it makes your army snowball in size due to reduced losses. It is especially important early on when some might heroes have issues with gaining a momentum.

"Brimstone rain" is a stub yes. We really wanted to include it so by making it gives +4 bonus level to hero in the battle is good as it make the Might hero more offensive in sieges.
I admit I have never used it so far but I have two builds I want to try with it and seem pretty valid.

I get what you mean about the middle perk (battle frenzy). Actually middle perks has a little higher chance to appear than the outer one (Magnomagus did it. I dont know how).
there are a lot of stubs one can take to fill in spots. Lorewise it feels ok it give and it gives some bonus so why break it if we don't have anything better?

I forgot to mention that archer tower damage will be changed (if not already but I cant find it in any of the release posts so it is yet to be) following this formula
Main Tower Dmg = 40 + (TL^2)/8
Side Tower Dmg = 20 + (TL^2)/10

TL - Town level
So at level 31 the side towers will do - 116, the main tower will do - 160. This is 400 pure damage every 10 initiative time. This is more than twice increase in damage and due to the battles being longer than before will accumulate over time.

Nargott, Knowledge now affects skills in Runes, Leadership, and Artificats

Apparently this is not enough, since the person writes that Knowledge is so weak that it still needs additional bonuses like levelup. Although levelup by itself is already self-sufficient as an ability, without bonuses, if taken late.

Level up is not that of a big deal anymore. Please take a look at this page for more information.

Thank you for the answers! Good luck with further development of the modification!

Thank you :). There were a few nice points I have already forwarded to Magnomagus and it will be considered for the next release.