Caster of Magic II Game balance brainstorming.

[Seravy]

This thread will be used to collect possible issues that are not bugs but might need future attention. (Currently, the top priority is to get the game to a bug-free, properly playable state.)

Please keep in mind the standard procedure for these will be, and that I reserve myself the right to decide if something gets on the list or not :

I. Add the issue to the list of problems.
II. Pay attention to the issue while playtesting, report results (issue makes game balance worse, better, or doesn't really change anything.)
III. When there is enough information, a decision needs to be made whether the issue is a problem or can be safely ignored.
IV. If the issue cannot be ignored, we need to design a solution.

1. Crack's Call dealing irrecoverable damage doesn't make good sense but is required for game balance purposes. (Phase III)
See : http://www.realmsbeyond.net/forums/showt...#pid748918

2. Book rewards too easy to get? (Phase II)
Currently any lair that has no less than 800 budget might get one. Perhaps requiring at least 1000 or 1200 would work better?
Ideally, AIs should have a fairly even chance of finding these compared to human players (of the expected skill level for the played difficulty), so if the human player usually gets most books, or if the AI does, then we might have a problem.
It's also not good if they appear in trivially easy lairs but the budget of 800 should be more than enough for that. (No, 6-7 Nagas does not count as trivially easy, not counting sprites, a fairly decent force of medium tier units is needed to kill them.

3. Is it too hard or easy for the AI to take out lairs under the current automatic combat rules? (Phase II)
Ideally, we want a reasonable chance for the human player to get to at least their fair share of lairs on the map, but the AIs should also get their fair share, basically in a 5 player game, the AI should get about 3/5 to 4/5 of the lairs on the map before the human player unless the human player plays too low difficulty or specializes on treasure hunting strategies. (related to the previous issue)
This seems to be solved by the recent updates but pay attention to it anyway.

4. AI players should eliminate each other at a reasonable pace. (Phase II)
Basically, if they do not then they'll left behind compared to the human player who grows by conquering other player's territories. In addition, AIs not expanding and destroying each other means balance issues with many late game spells which assume the number of players is generally lower than 4-5 by the time the spells are available.

5. Destiny might be overpowered? (Phase II)
http://www.realmsbeyond.net/forums/showt...#pid746741

6. Monsters Gone Wild option might be too effective at destroying AIs (Phase II)
Obviously, being an "extra difficulty" option, this shouldn't make the game easier by the monsters destroying all the enemy players. If that indeed happens, we will need to consider the option to only affects stacks spawned "for" the human player.
Currently, the stack strength is only +50% higher instead of the originally intended 100%, so if an AI specific exception is added, we most likely should put it back to the original +100%.
Note : monster stack strength greatly depends on difficulty level, don't forget that while testing.

7. Trading might be too easy when there are too many AI players (Phase II)
Basically, it's a problem if the player can trade for too many spells, accelerating their research too quickly. If that proves to be the case then the AI's trading "patience" should wear down faster the higher the player count is in the game, allowing less trades per player per turn.

8. Too much ocean might make it more difficult for the AI to fight against the human player. (Phase II)

9. The Revolting Raiders option is useless (Phase III)
Basically, low tier raiders are not a big deal and without the Tree of Knowledge spell, neutrals cannot possibly build higher tier units so this option is nowhere near as relevant as Monsters Gone Wild.

[jhsidi]

7. Trading might be too easy when there are too many AI players (Phase II)
Basically, it's a problem if the player can trade for too many spells, accelerating their research too quickly. If that proves to be the case then the AI's trading "patience" should wear down faster the higher the player count is in the game, allowing less trades per player per turn.

On this point, it's really wildly variable how much mileage you get out of trading and I think there's more than one problem.

Problem 1: wizard generation is random, so you get weird RNG results. For example, in my current game, I've met 8 wizards so far. Out of those, 7 have 4 or more Life books (and I can see there are a couple more wizards I haven't met to have Just Cause up). So they can all trade and ally with each other. Out of the 8, there are 0 wizards with any books in my primary school. So I can't trade with anyone. On the flipside, if those 7 all had 4+ books in my school of magic, I'd be massively advantaged (although also bored, I don't want that).
Suggested solution: During world generation, place a safeguard ensuring that A) every school of magic is used by at least 25% of the wizards, and B) no school of magic can be present in 2+ or 3+ books in more than 50% of wizards.

Problem 2: trading has no relationship check. Why is a wizard who absolutely hates me willing to trade? Or a wizard I just got a peace treaty with this turn? On the other hand, changing this so they outright deny trades might be too large a change from MoM at this point in dev.
Suggested solution: For low relationships, let the trading wizard demand something else along with the trade, like gold, or a second spell of lesser value, or a declaration of war on their enemy. Ignore this mechanic for trades the AI initiates.

Problem 3: there's no limit on trading. Maybe this wouldn't be much of a problem if the above two suggestions were implemented, but for really frequent traders some small speedbump probably still needs to be added here.
Suggested solution: For every trade, reduce the relationship of all other contacted wizards with you by -5 or -10. This would represent that the wizards have a "crabs in a bucket" mentality where they get angry about seeing other wizards getting ahead. They could send threatening messages about the trade after it's made if the relationship is low enough, similar to what they do currently when you acquire a new creature type.

[Seravy]

there's no limit on trading. Maybe this wouldn't be much of a problem if the above two suggestions were implemented, but for really frequent traders some small speedbump probably still needs to be added here.

There are basically 2 limits :

1. Each time you trade, the trade interest with that wizard decreases, so you can only trade a limited amount per wizard, but this system is currently very forgiving, a fully maxed trade interest variable is enough for 4-5 trades at least with that wizard.
This was set up that way because at a max of 4 wizards (out of which 1 is on the other plane so you can't trade with them for most of the game) you generally don't have more than 1 trading partner, even if super lucky you only get 3, but the average is probably closer to 1.5 per game.

2. You have to be in contact with the other wizard. This sounds trivial on the small maps we're testing at the moment but on larger maps, it requires some effort. Probably nowhere near enough to be a real limiting factor past midgame but definitely in the early game. More importantly, it involves the opportunity cost of that wizard being able to attack you - except, not really anymore because we've limited the AI's ability to not declare war on far away empires.
However, this is pretty much the main limiting factor intended for AI trades, so we'll also need to see how that works out, do AIs discover each other slow enough or too quickly?

Limiting games with very high (7+) player counts to include every realm might be useful although multiple players of the same realm does increase game variety because it requires different strategy to beat so I'm not sure it's good to prevent that. (You can't use the "I attack the enemy weakest against my realm" tactic if you don't find any of those, and strategies with a major weak point shouldn't really be forced to always have enemies that can target that weak point, nor should they be denied the chance to face many enemies that do so. )

The above is especially true because the game has several late game spells that stack and completely change the flow of the game when multiples are present, Divine Order being a primary example, but Chaos magic has this mechanic as well (in Chaos Surge and Blazing Eyes), so game with many of these wizards are very different from "normal" games.

Low relation demanding extra gold is a possible solution, low relation dropping trade interest faster is another, but I agree that low relation should be detrimental for trading in some way. Maybe it could even be a modifier in trade value, like, each 10 relation below 0 the player has to offer +1 spell value higher than the default of at least equal.

[jhsidi]

6. Monsters Gone Wild option might be too effective at destroying AIs (Phase II)
Obviously, being an "extra difficulty" option, this shouldn't make the game easier by the monsters destroying all the enemy players. If that indeed happens, we will need to consider the option to only affects stacks spawned "for" the human player.
Currently, the stack strength is only +50% higher instead of the originally intended 100%, so if an AI specific exception is added, we most likely should put it back to the original +100%.
Note : monster stack strength greatly depends on difficulty level, don't forget that while testing.

After thinking about this for a while, I believe the issue goes beyond Monsters Gone Wild. In my earlier test games I just mentioned MGW because my games weren't really lasting long enough to notice other issues. But really, the problem with MGW extends to the player: it's simply too easy for me to take enemy cities. Here's the base level analysis of what's happening:


1) There are many landmasses, most of which are far from the wizard, available to colonize
2) The AI is coded to seek out and colonize these other landmasses
3) Colonization is a success, but due to distance and garrisoning mechanics, the wizard is not able to defend them (btw, I couldn't find the old thread on garrisons you mentioned)

Here's an example of what happened in my early game. This screenshot is a small portion of the map:



Ariel started on a small island and attempted to expand in three directions. To the east, a relatively close island was unoccupied. She quickly colonized it. However, her colonies were assaulted by monsters and destroyed multiple times -- but, it's pretty close. She managed to maintain the foothold and did... decently well. Not very, since her cities were not fully garrisoned and a couple got destroyed over many turns.

To the northwest and northeast, there were other wizards. Now, the distances are getting really long. She tried to colonize northeast, on the island north of me. I didn't want has as a neighbor and so got into a war with her. From there it was really easy, I just kept an eye out for incoming stacks and destroyed them before they could reach my cities. With the distance involved, she couldn't really mount any kind of challenge to me. But she kept trying, over and over, and it drained her resources.

That brings me to my next point: because all the AI knows how to do is to keep trying, Myrror has been no challenge in each of my games so far. With only 3 wizards on Myrror, it's going to take them a LONG time to colonize the entire plane. As a result, their forces are spread out and underpowered.

Example: in my Expert-level game, I just broke through to Myrror. I do have MGW on -- I think? maybe not -- but I found 4 of Jafar's cities on the tower landmass, untroubled by monsters. The problem was, he only had 4 units in most of his cities, and they were just swordsmen and bowmen. This is in 1509, and I've got buffed shadow demons and strong heroes. So, within a few turns, I'll have 4 new beastmen cities. But this is the real problem for him:



Before I arrived, he couldn't support his colonies. After I arrived, he can't strike back. The distance involved is so long that I have a huge buffer to mount a defense.


I don't think distance is the only reason wizards are under-performing. The simple fact of available space to spread into results in wizards constantly trying to expand, which means they never consolidate. On Myrror, at Normal game size, I don't think they'd finish expanding until probably 1515, given how much territory there is to take, and by then the game can be more or less over.

But it's also worth noting here that the tests I've done with Plane of Earth feel significantly harder. It's only 2-3 test games so this isn't a final analysis, but I think it's probably because the enemy AI is better able to reach both their colonies and their opponents.

Anyway, that's my analysis of the problems. I think there's too much ocean, but you already know my opinion on that. Perhaps tuning the enemy AI to better be able to deal with the oceans would solve the problem; they should be fielding navies, flying units, invasion forces, etc instead of hesitantly burping out settlers.

I also think there's just too much space for colonization. Not all players will enjoy massive expansion or being forced to rule empires of 30-50 cities, but if everything works out with the enemy AI, you'll eventually be forced to do that just to compete. And everything working out with the AI is a pretty tall order -- this is no longer Master of Magic, increased size and number of wizards has introduced new complexities that won't be trivial to iron out.

You could also say: just play a smaller map. But then there's less to explore, less ruins and nodes and interesting spots on the map, and I don't think it's really your intention that people play on a small map; this is supposed to be a grand strategy game. So, I don't know. What if there were an additional new mechanic that prevented settlement near strong nodes / ruins until they're cleared? And landmasses that are initially unoccupied always have a very strong location? Just spitballing a bit. Curious to hear your thoughts on the above issues with Ariel / Jafar.

[Seravy]

Well, for now I've added the ocean size to the list. I'll try to form an opinion after actually playing the game a few times.

I did notice people like to play large maps but don't like to have too many cities and units, but I don't have a solution for that paradox. More lands means larger empires and more stuff to take care of, that's one of the the core game mechanic(s) of 4X.
Restricting settler location is definitely not an option - it can result in players starting in areas where they can't build a city at all. Also, cities that are built too late are simply not relevant to winning the game anymore - by the time they can grow to a relevant size, the game ends. The most they can contribute is a place to spend gold on buildings but that too isn't relevant unless all of the player's cities are already maxed out on economy buildings, which would assume they haven't built any other settlements recently.
In the end, that would be no different from simply adding more tiles that cannot have cities on, like vast amounts of tundra that doesn't allow cities similarly to the north pole, but can have lairs and nodes on them - or changing city building rules to not allow using land that's below 5 max pop. However these would also affect smaller maps and make useful territory less on all map sizes. (In fact, small land is probably affected the most because it has a higher ratio of cities on shore with not-so-great max pop.)

[jhsidi]

However these would also affect smaller maps and make useful territory less on all map sizes. (In fact, small land is probably affected the most because it has a higher ratio of cities on shore with not-so-great max pop.)

You're definitely fighting an uphill battle here: you need one set of rules that applies to all map sizes. And the variance between map sizes is massive, much more than the original game. It's possible that finding one perfect set of rules actually can't be done.

For me, I'd be fine with toggles for just a small and large game, and different rules applying between those two game types. Cutting down options would allow for easier analysis of what's wrong or right in each map size. As-is, you may get people complaining about issues that primarily happen on a tiny map, or a normal one, or a maximal one... personally I'm gravitating toward small, as even normal maps are really massive, and maximal is just ridiculously oversized.

Restricting settler location is definitely not an option - it can result in players starting in areas where they can't build a city at all. Also, cities that are built too late are simply not relevant to winning the game anymore - by the time they can grow to a relevant size, the game ends. 

OK, yeah... restricting areas is not a good idea. Actually I used your editor in the past to set city size to occupy 4 or 5 tiles, and that was also not fun.

What if the neutral cities were a bit more numerous, and could actually be challenging? As is, even with Revolting Raiders on and game difficulty high, neutral cities are complete pushovers, for both me and the AI wizards. Especially for the human player, there's just no challenge at all to a garrison full of longbowmen (guardian wind, goodbye) or wolf riders (cloak of fear, sprites, etc). Same as MoM really, but there are more neutrals, territory and so forth now.

Neutrals could, conceptually, become the demesnes of minor wizards. They could not engage in diplomacy, would not progress in research, wouldn't settle new towns and would not accumulate resources or increase their skill pool, but they would present a challenge to expanding territory. So basically they get:
1) Summoned creatures in the garrison (including some uncommons or maybe even rare creatures)
2) Spells to cast during battles and a skill pool
3) Raider armies to try to take settlements from the major wizards (only differing from the current game in that the raiders would have some summoned creatures and spell support)

More impressive minor wizards could have more impressive hometowns, giving you a reason to actually try to defeat their garrisons before it becomes trivial to do so.

... eh, it has some flaws as a plan. But it's worth continuing to think about IMO, because of this:

I did notice people like to play large maps but don't like to have too many cities and units, but I don't have a solution for that paradox. 

What the people want, they must have! OK, me. I'm talking about me. But it is a very typical problem in strategy games and some have actually figured out elegant ways of dealing with it, so it's not impossible. We just need to find the right idea for this particular setup.

[Seravy]

What if the neutral cities were a bit more numerous, and could actually be challenging?

We had that in the original game and it doesn't work.

It basically results in one of these two, or both :
-You lose the game because a hard neutral city is nearby and sends too high tier raider units attacking you.
-You conquer all the neutrals early because medium and low tier units are trivially easy without a wizard casting spells for them, so basically you start the game with 10+ cities and win while the AI can't do the same.

I most definitely don't want neutrals to have summoned creatures or spells.

[jhsidi]

It basically results in one of these two, or both :

-You lose the game because a hard neutral city is nearby and sends too high tier raider units attacking you.

Then I don't understand why you implemented rampaging monsters from nodes and lairs. These, too, can make the player lose the game, even without Monsters Gone Wild. Why would neutral city stacks that can incorporate summoned creatures be balanced any differently or not use the game difficulty scaling that is already used to construct stacks?

-You conquer all the neutrals early because medium and low tier units are trivially easy without a wizard casting spells for them, so basically you start the game with 10+ cities and win while the AI can't do the same.

It sounds like you don't like neutrals as they already exist, then? On a normal size map, there are about 7 neutral cities at the start of the game. A fast explorer can reach several of these. And since they're defended by homogenous stacks of mid-tier units, at absolute best, and have no spellcasting support, they are literally just gift-wrapped prebuilt cities, defended by the military equivalent of tissue paper.

Anyway, it wasn't the topic but in my opinion you should remove the Revolting Raiders option. It currently does nothing and provides no challenge. If your belief is that stronger neutrals would be too powerful then the game option has no path to becoming viable. Personally I can't see how they could be powerful, as city units are vastly inferior to summoned units if they don't have buffs.

[Seravy]

Then I don't understand why you implemented rampaging monsters from nodes and lairs. These, too, can make the player lose the game, even without Monsters Gone Wild. Why would neutral city stacks that can incorporate summoned creatures be balanced any differently or not use the game difficulty scaling that is already used to construct stacks?

It's a different game mechanic.

Rampaging monsters are spawned using a budget. For example, sorcery node on turn X, difficulty Y spawns a total of 150 cost Sorcery creatures.

Raiders are spawned using the actual units in the city. For example, Dwarven town on turn X, difficulty Y spawns 4 units with 2 experience levels each that are exact copies of a random unit that was present in the city.

Basically, a lair with Death Knights will spawn zombies or skeletons in the early game. A neutral with Hammerhands will spawn Hammerhands, not swordsmen. So there can't a neutral that has Hammerhands in the early game.

Anyway, it wasn't the topic but in my opinion you should remove the Revolting Raiders option. It currently does nothing and provides no challenge. If your belief is that stronger neutrals would be too powerful then the game option has no path to becoming viable. Personally I can't see how they could be powerful, as city units are vastly inferior to summoned units if they don't have buffs.

One way to make it viable is to give a production rate buff to the neutral cities and change their production selection rules. Then they can build those higher tier military buildings on their own faster and start spawning stronger tier units...except, actually, they can't. Not anymore because the buildings require the Tree of Knowledge spell.

So I agree this option is kinda obsolete now. I'll add that to the list of problems.

[jhsidi]

It's a different game mechanic.

Rampaging monsters are spawned using a budget. For example, sorcery node on turn X, difficulty Y spawns a total of 150 cost Sorcery creatures.

Raiders are spawned using the actual units in the city. For example, Dwarven town on turn X, difficulty Y spawns 4 units with 2 experience levels each that are exact copies of a random unit that was present in the city.

Basically, a lair with Death Knights will spawn zombies or skeletons in the early game. A neutral with Hammerhands will spawn Hammerhands, not swordsmen. So there can't a neutral that has Hammerhands in the early game.

Summarizing this a bit, there is one game mechanic that does not work well (neutral cities) and one that does work well (lairs and nodes).

So... ditch the mechanic that doesn't work. Neutral cities can generate stacks according to a budget that scales by time. Simple, predictable and already well understood. It would also be obvious to players.

Also, if you make that change, you can start off neutral cities more fully built, with city walls and full garrisons. That would prevent the recurring problem I'm finding of finding neutrals in the first 10-20 turns with almost nothing guarding them.

and start spawning stronger tier units...except, actually, they can't. Not anymore because the buildings require the Tree of Knowledge spell.

Hm? But I found a neutral elf city with an armorer's guild. It was still garrisoned by longbowmen, but I got the armorer's from it before I cast tree of knowledge, and it generated pegasai as normal. So either the rule isn't working anyway, or the city appeared with an armorer's guild but somehow didn't get a garrison of pegasai.