Have to say I've had the same experience as jhsldi. Wizards rarely have anything beyond common or uncommon spells in their fortress, not to mention I gain nothing when they don't pick the same books as me. Their city isn't anything special either, it's just often a highly defended spot, and most of the good buildings like the Armorer's Guild get destroyed inevitably during the invasion. The heroes are often dangerous not because they have good items, but because they usually have a bunch of buffs on them and cast extra spells/attack your troops at range. Fortress invasions are really not worth the investment, which is why I usually swing around and try taking out their other cities when they are nearly defeated.
|
Caster of Magic II Game balance brainstorming.
|
|
Regardless, the typical wizard you'll be defeating won't share books with you so you won't get spells even if their research is at a normal level. Also, no guarantee on items even if the AI gets better at treasure hunting. Any heroes they had outside the fortress are dead and their items already went to the victor.
Some ideas of what you could receive from defeating a wizard:
Well, my primary motivation for defeating wizards is not the rewards but the consequences for not doing it.
If I don't eliminate the wizard then... -There will be a city on my continent that belongs to someone else -That city will keep spawning units -If I make peace then those units will build up into dangerous stacks and later might attack me and cause serious damage -If I don't accept peace then I have to keep killing every single unit they produce which super annoying, or surround their city with 9 spearmen which is something I rather not do below Master difficulty because it's definitely an abuse of game mechanics. -They might cast spells that cause trouble if I don't make peace or it expires - Volcanoes, city curses, drain power, spell blasts, etc. -They trade spells to other AIs that might make things harder. (Even if I don't share books with them, if they give that Prayer to the 10 Chaos 2 Life AI then that's very bad.) -If the wizard is eliminated by someone else, they (might) get the free spell(s) but more importantly, they will have a city on my continent, which will likely get designated as a frontier city on top of receiving a major bonus priority for unit production due to being in enemy territory, so if any wizard who is a threat gets the city, they can and will use it to invade. (unless I'm already at war with them and can take it out immediately before that happens) So yeah, the rewards are bad but leaving the city is worse, unless taking it out is unreasonably expensive/difficult. (and this list is exactly why I don't want a miracle solution in the form of a spell that banishes the wizard.) Quote:A percentage of their mana or gold reserve (looks possibly unbalanced, but if they're on their last legs they won't have much left; also, any substantial reward would help strong AI wizards to further snowball; could be balanced to both players and AI a scaling % based on difficulty level) Looking at the code, the attacker gets 50% of the defeated wizard's mana. (there is no notification about this anywhere, you just get is and that's all.) No gold is transferred beyond the usual amount on city conquest, we can change that if necessary. It's unlikely they have any gold left either way because they lose cities faster than they can disband troops to maintenance issues so the only way they can have a lot of gold left is if all their cities were taken simultaneously. Quote:Additional +10 fame (makes sense, I don't know why this isn't a thing already actually)It is a thing, its +5 though. I agree +10 will work better. (speaking of the reward, not sure it's still included, probably not, but banishing a wizard used to be a major part of a player's final score.) 
Been busy and couldn't chip in on the early discussion here. I haven't had a chance to play through the latest build but I have a different take on what the "issue" with AI not eliminating wizards is. I don't think it's a matter of it being too hard for AI to take the final fortresses.
It's just that, the way the AI war and diplomacy weightings are designed right now--and it's good IMO because it's how a player would behave--it doesn't make sense to eliminate wizards entirely. As others have discussed in these last couple of pages, the reward for taking the Fortress is unimpressive for the cost, both in terms of troops necessary to take it, and the diplomacy penalty to other AI relationships. Even more importantly, in the context of much larger games like in COM II, the incentive to stay at peace generally is even more than it already is in high difficulty games on Huge maps in COM I. But specifically for the Fortress: -They can make dangerous units? -- Just make peace then. -Cause trouble with spells? -- Make peace. -Trade spells to others? -- When you've got 10 competitors, denying spells to a couple of them is far less important than personally getting ahead by not wasting resources. -Taken by someone else? -- Awesome, now it's not a Fortress and easy to take. The larger the map and more players, the more the natural equilibrium of a competitive war game shifts to peace. Because unless there are enormous disparities in strategic power growth or intelligence, any resources spent on war against an effective opponent are resources that you are losing relative to your OTHER peaceful opponents who spend it all on research, economy, and peaceful expansion. Trade then multiplies the effect as a group of wizards can effectively share research and do it several times faster. What's better? Having 2x the sacked and half destroyed territory from conquering after spending a ton of troops and mana to take it plus diplo maluses from other wizards that know you're a conqueror, or having 2x the territory as allies who've never had to fight and are both at full strength and who were researching and building at twice the rate you were? Better yet, why wouldn't the allies just wait until the warring wizards exhaust themselves and come in for the easy pickings? If the conqueror insists on finishing off the victim at an extraordinary cost instead of fortifying themselves against full strength allied potential enemies, I think there would be something seriously wrong with them, and the AI shouldn't be doing that. It seems obvious that unless your race/build is entirely geared for conquering, making allies is always better. There is not a natural rush to conquer, but a rush to form the larger alliance web. With 5 wizards on COM I, this effect is less obvious, but when there's so many wizards, it's inevitable. And it happens because the AI in COM is actually good, and the balance between tall vs wide play is also reasonably good. It's not like other games where the bad AI is vulnerable to super fast rush strategies and let players snowball. More territory is not always better. If the game could be played in multiplayer, I daresay the competitive metagame would involve almost NO warring until the endgame because not many human players would want to fight any wars unless they are absolutely certain they have a massive advantage. If there's a problem, it is that AIs don't trade enough between each other as the player is allowed to. This gives human players a large advantage in research, and this is a common problem in other 4x games with tech trading. The advantage gained from having tripled or quadrupled research due to trading, or worse--FREE research from "brokering" spells--makes it trivial to overcome even the highest of AI bonuses. 
I understand what you mean but that only works in other games because everyone has the same tech tree and there is no mechanic to limit how much "techs" you can use called "casting skill".
In Master of Magic, a 5 player alliance can mean any of the following : 1. Everyone has 10 Life books. They trade a lot, and each of them have all the spells. Overall we get the result you mention, the 5 player alliance will function the same way as if it was 1 player, minus the damage to cities from fighting. ...except not really. Casting skill scales with empire size (both indirectly due to resources and directly due to amplifying towers but the former is not linear). For overland skill this means there a larger total casting skill (due to the non-linear part) but less skill on each individual wizard. As allies can't pool their units together in an army, this means a much longer duration for each individual player to build up one doomstack. Furthermore, global enchantments only affect the caster, so global enchantments eat up 5 times the casting skill in a 5-player alliance compared to one player. Overall, these cancel any advantage so I'd say it's fairly even but military potential suffers greatly. (5 players each having 2 Archangels is not the same as 1 player having 10.) However in combat, we can't add the skills together - only the player whose units are fighting can cast spells. Individual combat casting skill will be drastically lower than what a large empire can have so the 5 player alliance will be significantly weaker in battle even if they manage to use the same units. The alliance is also weaker against mana draining tactics because the enemy has to drain only one of the players to destroy that player. Finally, some globals are cumulative. 5 people casting it produces 5 times the effect. Not at all identical to one large empire. 2. Everyone has 10 books but in 5 different realms. No trading can take place. The 5 wizards, even though allied, all remain individually weak and behind in spell research. The entire alliance will lag behind and get destroyed by the larger empire. 3. Everyone has 5 books in 2 different realms. (so each of the 5 realms appears on exactly 2 players) In this case, assuming they got lucky on their spell picks and there is no overlap, they can get most spells in all five realms. Research will be very slow because there are 5 times as many spells to get, so this most likely results in the same thing as "2." but if the alliance manages to overcome it, we still don't get "1.", instead we get an alliance that's several times more powerful than one large empire as they have 5 times as many spells available. Finally, it's also worth noting that alliances don't last forever. It can be possible to wait until someone from the alliance researches Spell of Mastery and the rest of the alliance attacks that person then join the fight, and repeat this for each alliance member one at a time. Of course this comes with the added inconvenience of having to fight difficulty fortress battles.
Today's test game thoughts.
-Rampaging monsters. It feels a bit too much. Gargoyle in 1404, Chimera and unicorns in 1406? Sure, for Master or Lunatic, but on Advanced? I mean ok, I can defend 2-3 cities well enough to deal with that but 5-6 not so much and 5-6 is still not overexpansion. Normal bowmen are not really enough to deal with gargoyles and better units are expensive to have 3-4 in each city. (however I was playing Sorcery halflings which is about the worst at the early game.) The purpose of the game was spell trading and basically exactly what I expected happened. I was able to trade for all common spells and focus on research uncommons only - but I wasn't able to trade for uncommons because no one had any yet. Basically the trading strategy allows picking the last spell in the research book all the time and fill in the spells below by trades which is a great advantage but it's not all that different from how CoM I worked. Having more trading partners contributed of course but not significantly - even with only two trading partners I can get all common spells and even with 10 trading partners I still can't get an uncommon because they don't have any. So similarly to the 5 realm game, trading was better, and it felt powerful but didn't really give me a relevant advantage. (in fact I can skip researching my commons even without trading, and fill them in by finding common spell treasure.) I expect the same to remain true even in higher spell tiers, basically what the OP trading does is it allows you to keep up with all other AIs and IF you have good research production, have a small advantage over them by being the first to research spells in the next tiers, but you have to research those on your own. You can't trade for something no one has, and you can't trade for a higher spell tier without having it yourself because trading doesn't allow getting a higher value spell for a lower one. So there doesn't seem to be a need for any significant trading nerf but we might want to consider a minor one. I don't think reducing the amount of times you can trade with one person would work because the number of spells is too limited. It only takes 2 trades each with 5 other wizards to get all 10 spells of a tier. So if we want a relevant nerf, it has to be the spell value, for example, get 1 lower spell value for each 3 wizard in the game above 4 (rounded up). Basically, making trading spells of the equal tier harder for the human player and require offering increasingly better spells in higher player count games. (which makes sense because trading one good spell to 1 out of 13 players is less of a problem than to 1 out of 4 people.)
It's worth considering that the impact of trading isn't simply how many total spells you gain access to. It's the timing of when you get it.
There are a lot of enchantments in Life, Sorcery, and Nature that dramatically alter the growth curve if you can gain access to them quickly. Sure, you can skip common spells yourself and find in treasure, but the extra turns used prevents you from growing resource production that much sooner, and the growth is often exponential. Even if nobody else has the Uncommon yet, what about the moment that they do? Instead of researching 1 new spell yourself, you may suddenly have 3-4 Uncommons you can gain immediately. Getting Stream of Life, Altar of Peace (in the early-mid game, this can MORE than double your RP as 24 per city is way more than you get from all other buildings combined), a year or two earlier--or not at all--makes a massive difference, and it's much more significant if you ALSO get the military uncommons to defend yourself. The impact from Stream of Life is hard to measure as pop growth and unrest reduction has a huge number of effects, but it's definitely a world of difference to get it at the time of your first Uncommon as opposed to 3rd or 4th when it still takes 6+ turns to research each spell. The same issue crops up at the Rare stage with Uranus' Blessing (this can practically double your power output in a year, as each cast gives you more Power than the average Arcanus node on Max settings) or Gaia's Blessing, and to a lesser extent with the 3 Rare city enchantments from Life, even without the enormous impact of multiple Divine Orders being activated. Aether Binding too, at the time you first get it, could be equivalent to 1/3rd to 1/4th of total power production. An extra casting skill every 2 turns or so adds up to 10 casting skill over 20 turns, and from turn 50-70, that 10 skill could be upwards of 20% of your entire skill pool, depending on how much you've otherwise invested in it. The alliance isn't just researching 2 or 3 times faster. Once they all get these economy spells, they are possibly researching 6-8x faster, reaching the next tier years before everyone else. In COM I, I hesitate to trade these economy spells to other wizards because there aren't that many of them and I'd rather keep my own lead. In COM II with 13 wizards I'm going to hand out all these spells to my alliance partners for free if I have to just because I benefit tremendously if they are researching as fast as me and I have no interest in fighting them until every non-Alliance member is gone. On the other side with the offensive spells: -having a group of players spamming curses at every one of your cities/tiles is far more problematic due to their way bigger total casting skill. It's incredibly horrific if a team of Death Wizards spam Drain Power on the same wizard same turn and cause 50 turns worth of enchantments/summons to be dispelled from insufficient mana, or force them to spend all Power on Mana with nothing on RP/SP until they fall so far behind they can never catch up. I know that the AI isn't programmed to do this, but it's certainly a devastating strategy if actually used. -Getting both strong summons and combat spells at the same time while your enemies only have one or the other can make the difference between winning or losing the vast majority of battles in a war. But because everyone garrisons first, this mostly results in much stronger defenses you can't possibly overcome by using rush strategies.
Well, we'll have to see how that modifier I added today changes things.
Even before that, if you wanted to get 3-4 uncommons for researching one, you had to research one with very high trade value, do it before anyone else, and share it with other wizards. On top of that, the other wizards had to miss that spell entirely (otherwise they to will research it before the other spells, why wouldn't they if it's better?) or they had to find a good uncommon in treasure that's not this one. I'm not saying that doesn't happen, it does, but it still is a lot of limiting factors. However with the additional spell trade value modifier that changes by also removing the 1-2 spells in the same realm and tier from the possible things you can receive for it, further reducing trade chances. (we might want to discuss how much modifier we want. Currently it's a +4 at max players, so roughly 25-30% of a tier's value. For example Altar of Peace (20) would normally trade for Heavenly Light(19), Unicorns(18) and Just Cause (18) but with this modifier it does not. Or Stream of Life (23) can no longer be traded for Raise Dead (21) and Altar of Peace (20) as well the low value rares Exaltation (21) and Mass Healing (22). You can see the full table here : https://www.realmsbeyond.net/forums/show...?tid=10059 There is a -5 modifier for spells a player could research on their own however, so offering a spell the AI cannot get for one you could get can help you trade more. So maybe considering that a +4 is too little.) Anyway, if we assume trading and alliances are an issue there are limited ways to solve that. For the AI side, we already did - the AI scales down the frequency of their positive diplomacy rolls by player count. For the human player side, there is a large limiting factor on Alliances - the AI will ask to declare war on their enemies (or break your alliance with them) more often that they did in CoM I, and due to the AI doing less positive diplomacy, large alliance blocks don't form. Yes, there will be that one or two AI diplomat players who ally with everyone but their allies will still keep fighting each other, so you can't really do that same. So Alliances are probably good. That leaves trading - I really don't see how trading could be restricted other than requiring to hand over a more expensive spell than normally which is what I did today, and where we can make proper numerical adjustments as necessary by raising the required spell difference. On the other hand if we assume trading and alliances are not a problem and try to gear gameplay towards that, I think we hit a brick wall - it's way too luck based, you either get a lot of wizards you share spellbooks with, or don't. So winning or losing depends on which realm had the most spellbooks among all wizards - basically if you are the member of the "Life wizard alliance" group who have 8 wizards, you'll be winning, but if you're member of the "Chaos wizard alliance" which is only you and one more guy, then you lose and the Life guys win. I agree city curses among a lot of other spells are one thing that cause issues if there are too many players casting it, and not just through alliances, as multi-front wars can happen even without that. So that's one more reason why player counts should decrease, fortunately curses as rare spells just like all the others that have this sort of problem. PS : A different possible solution could be making the AI ask for gold in addition to the spell in the trade but I'm not sure I like that. There isn't much gold in a player's treasury during the common/uncommon phases of the game and the "fair" cost of a spell would be quite a lot. And in the end it's still gold that makes your ally stronger which is not as good as yourself but better than your enemies...
I see the biggest potential issue as being from brokering spells. If trading worked as you said, with the player having to have a unique spell to trade to everyone else which is also high value, then it wouldn't be so powerful. But actually, you only need a spell that any one wizard doesn't have, and for that wizard to have a spell that some other wizard doesn't. You can then take the newly traded for spell, and trade it again with the 3rd wizard. And again with the new spell...
The more wizards are around, the easier it is to do this, especially as you can trade different realm spells with wizards of dual or triple realms. That's why I said earlier that AI not rapidly trading between each other makes the player gain a large advantage. The player can do all of this trading in one turn with every AI and wind up with several new spells. Possible solutions: -limit the number of times you can trade in one turn, with other wizards refusing to deal with you -just make the AI trade as often as possible so the player doesn't have much opportunity to broker anything -make the AI more reluctant to trade spells of the current highest tier. Want any of the good uncommon spells? AI won't give it to you until they start researching Rares, unless you give them something they really, really want. |