My Thoughts about the New Barbarian Clans Game Mode Part1
What started as a simple list of random thoughts I had about the new Barbarian Clans game mode has evolved into this elaborate text. Instead of just providing a list, I thought it was important to explain my whole thought process. While not everyone may not agree with my final recommendations, maybe my arguments will help spark new and difference ideas for this new game mode. In this post, I will focus on criticizing the different aspects of the current iteration of the Barbarian Clans game mode. In the next post, I will propose multiple novel and more complex ideas which I hope would greatly improve this mode.
0.
Major fix needed
First and foremost, the AIs need to interact with barbarian camps more. As I highlight
here, AIs currently only use the bribe diplomacy option with barbarian camps and very rarely (about 1-5 times observed for 7 AIs within the first 120 turns on deity where gold should be less of a limiting factor). They are completely ignoring the hire, incite and ransom options. It is imperative to fix this. Many of the ideas I will outline in the following text will depend on the AI properly using these diplomacy options.
1.
Bribe
As an experienced player, I currently find little use to bribe a clan. The only occasion where I could see it being useful is in the very early game (turn 1 to 10) when you have not built your first unit yet and a barbarian scout sees your city while your starting warrior is caught out of position in the opposite direction. However, I cannot use the bribe option in this case, simply because I do not have enough gold and I do not know where the barbarian camp is located yet.
One possible solution (though hard to implement successfully) would be to bring the barbarian diplomacy prompt when clicking any barbarian unit (without having to find the location of their camp ie a kind of “bring me to your leader” option if you will). Alternatively or additionally, another solution would be to change the cost formula for the bribe cost. Currently, it is set as 50+50n where n is the number of cities you have founded. At the moment, we can change the slope of the cost formula through modding, but not the intercept. If the developers do not care about changing the formula, they should at least give us more options to change it. I would personally change the intercept to 0 and see how it goes. (I talk about changing the whole formula/mechanic altogether in the “Bribe/Incite bidding war” section). Now, these two previously mentioned suggestions are more nitpicks really and are probably the ones I care the least about since I see getting overrun by barbarian early on once in a while simply part of the game.
No, the major problem concerning bribe is more tied to what it actually does which is quite underwhelming… It simply blocks barbarian units from entering your territory. Ranged barbarian units can still attack your units inside your territory and your city center while sitting just outside. Barbarian units outside your territory will also remain hostile toward you and plunder your trade routes. It’s hard to justify spending such money for so little “protection”. Unless you are getting overrun and about to lose an expansion (after the raiding sequence has been triggered), I see little reason to pay a bribe. If all the barbarian units became non-aggressive, that could however possibly justify the cost in some situations. Currently, if you are overrun by barbarians, you would certainly be better off by hiring units from that same tribe!
2.
Hire
At the moment, hiring units from barbarian camps is good, really good, too good actually! As it is now, you get a 40% discount (60% price) for hiring a military unit at the small disadvantage of a 15-turn cooldown (per camp). Now, why would anyone purchase any unit in their cities if they can do it for way less from barbarian camps? By scouting around, you can easily find 3+ camps (depending on map script and difficulty level) rendering the cooldown period no limitation at all. Now, you can mod the hiring cost scalar in the files, but unfortunately you can’t implement what I would do instead: the hiring cost would start higher (at say 110%), but would be reduced by 10% for each additional unit purchase (per civilization). That way, the 3rd unit purchased would be at a (10%) discount and you would need (30) turns at least before you could unlock it. The discount could also be capped, but chances are the camp would turn into a city from all the points it accumulated at that point before it becomes too imbalanced (setting dependent).
Hiring units from barbarian camps needs to be nerfed, because it provides 3 significant advantages over simply purchasing units from cities (apart from the lower cost discussed above):
- You can get access to better unique units from other civilizations that you would normally have not access to
- You can get access to units which require technologies which you have not researched yet
- You do need to posses the strategic resources normally required to get that unit
One other area that could be improved relative to hiring is the diversity of units available (in terms of unit classes). Currently, you can only get melee units (from flatland, woodland and hills clans), light cavalry units (from the rover, chariot and jungle clans) and naval melee units (from seafaring clans). But why can’t you also hire ranged units from the woodland camp? Or an anti-cav unit from a flatland clan? As a rule of thumb, I think you should be able to hire the unit class of the unit that is used as the defender in the barbarian camp. (This idea ties to another one, see the “Defender units in camps” section)
3.
Incite
“Does this actually do anything?” was my first thought after trying to “Incite” multiple clans. I mean beside the added gold coin stack icon beside the barbarian units, I didn’t notice anything: no additional barbarian units spawned from the camps, no change in movement/behavior, nothing. After doing some additional testing while also revealing the map, my thought was still: “Does this actually do anything?”. Actually, after numerous tests, I did see “something” happen on one occasion: one barbarian camp that had about 10 units under its control sent a small force consisting of 4-5 warriors in the direction of the AI I incited them against. It was obviously crushed by the deity AI. Why did something happen in that particular case? My guess is that if the barbarian clan does not have vision of any unit or city of the AI you are inciting them against, they won’t do anything. As you can incite clans against majors or minors which are located in a pretty far range from their camp, it’s possible they do not possess the knowledge of their location. After all, if their scout had found any of their city, they would have launched the “Barbarian Attack” operation already.
The are also some other “strange behaviors” surrounding the incite mechanic. For example, once you have incited one tribe against a major, you cannot incite another tribe against another major (the only exception is with the seafaring clan which opens the option of inciting against an additional major). Some times, for some unknown reason, you cannot incite again the same clan after the 15 turns have ran out. Also, once the incite period has run off, the gold stack icon never seems to fade away, etc. This is currently without doubt the most “broken” feature of the new barbarian clans game mode especially when you take into account how much it costs…
Now it’s difficult to determine what is the “right” value for inciting a clan when you don’t even know what it is supposed to do… For the benefits you are currently getting in the game, I would value it at zero. However, the cost of this option is currently valued very high. In the early game, it is the costliest option and its cost slowly increases with the passage of time/turns. Cavalry clans also require an additional small premium versus the other clans. Whether a clan has 1 or 10 units under its control the cost will be the same… The only way I could see myself paying such a steep price is if the incite diplomacy option launched the “Barbarian Attack” operation against the civilization you incited it against (as if a scout had found a city and ran back to its camp resulting in the spawning of multiple barbarian units). This could act as a good distraction if you are planning to attack an AI in the near future and could be a valid alternative to a joint war.
In the end, there needs to exist an equilibrium (advantages and disadvantages) between the 3 following options:
- Training your own army (build time, unlimited time, perfect control)
- Levying troops from a city-state (no build time, 30 turns (with risk of losing suzerainty), perfect control, gold needed to levy and upgrade units)
- Inciting a barbarian clan against a civilization (no build time, 15 turns, no control, weaker units, gold needed to incite)
4.
Ransom
If incite was the barbarian diplomacy option with the worst implementation, ransom is certainly the one with the best. One should however note that when you ransom a unit, you actually do not get your old unit back. The captured unit is deleted and a new one is spawned in the city nearest to the camp. (Note: this new unit does not contribute to the cost scale up of builders/settlers.) This distinction is important, especially for builders. If a one-charge builder gets captured by a barbarian, when you ransom it back, it will become a three-charge builder. Now, a bell may ring in the mind of some of you with a banner that reads “Exploit!”, but I assure you it is not the case here. Here’s why: the newly spawned builder is not affected by Amani in the city or policy cards such as Serfdom. It is however affected by the Pyramids. So, in the best case scenario you can get a four-charge builder from ransom. The opportunity cost is one builder charge, so you are effectively getting 3 build charges for the price of the ransom which is fixed in the game files as 75% of the purchase price of a new unit. Thus, in the best case scenario where you have the Pyramids, ransoming a one charge builder at 75% price is neutral to buying a new builder from the city panel on a gold/charge basis (without considering the extra hurdle of getting your builder kidnapped in the first place). In the case where you do not have the Pyramids, ransoming a builder actually costs more on a gold/charge basis.
Ideally, in a perfect world, the builder returned in a ransom would have exactly the same number of charges it had when it was initially captured. Yet, to make this change work, the ransom price would have to be adjusted accordingly by taking into account the current purchase price of a builder and the maximum number of charges you can get to calculate the appropriate ransom price on a gold/charge basis. However, that may be too difficult to code properly and I am perfectly fine about how things are currently handled on this front. The only problem I can foresee with the way things are done at the moment is if you have unlocked Amani and Serfdom and the builder you are getting back (3-4 charges) has less than 75% of the charges you would get by purchasing a new one in the appropriate city. In that case, paying a ransom makes no sense. It’s not a problem per se as it creates the opportunity for the player to make good/bad choices, but it devalues the option of ransom for a “Cartesian player”.
In a way, it is a shame that the currently most balanced diplomacy option with the barbarians is also the one that is the least likely to be used by the player as you also always have the option of getting your civilian units back by the use of force with your military units (at the opportunity cost of using them for another purpose). If you have idle military units at your disposable you should of course do so. And so should the AI. Now, I don’t know where to track in the logs (or if it is even available) the number of civilian units that were stolen from the AI by barbarians or what was their military score at the particular time or how much gold they had, but while the number of occasion where it would have made sense for them to ransom their civilian units may have been small, it most likely was not null. However, there never was in the logs any trace of any ransom made by an AI (see the “Major fix needed” section).
5.
Bribe/Incite bidding war
When analyzing the logs related to the barbarian clans, I found a strange occurrence that I had not even thought about or thought possible until that point: a barbarian camp that I had incited against another civilization was bribed just a few turns before by that same civilization. Of course, I was not aware of that when I decided to incite that particular clan as there is no way to tell this when you are in game. Also, since incite does not currently do anything in game, it is hard to evaluate how those two conflicting “orders” would play out in reality. But let’s assume for a second that incite does what I proposed earlier: trigger a “Barbarian Attack” operation (see the “Incite” section). The barbarian camp would begin to spawn its attack force, move to meet an AI city, but would stay outside its borders, because it accepted a “bribe” from the target civilization. As you can see, complying with both conflicting orders should not even be considered treason or backstabbing, it is just complete nonsense.
The simplest way to fix this without making a UI that displays each barbarian tribe’s particular stance (bribed/incited/neutral) versus all other civilizations (like it is currently done for civilizations) is to make sure that a barbarian tribe only follows one type of order at any point in time, either a bribe or an incite, but not both at the same time. So, how to determine which one it should follow? simply by establishing a bidding process. The initial bribe or incite ask price would be fixed by the number of units a particular barbarian tribe has under its control. Once a civilization pays the asking price for either bribe or incite, another civilization can change the barbarian clan’s stance by outbidding the initial bid by a certain increment (let’s say 50g). The initial civilization would then need to pony up 100g (ie two increments) to revert the barbarian clan’s stance to what it originally was. (Note: the minimum asking price should always be a function of the number of units a particular barbarian tribe as under its control. For example, if a civilization incited a tribe and it produced many units as a result, the ask price for a bribe, ie stance reversal, could be higher than 100g as a result of the greater number of units it now has under its control.) This would insure that a single barbarian clan can only be bribed or incited against one civilization at any single point in time. (Also, it would ironically fix the way the UI can only show 3 of the 4 diplomacy options at a time by combining both bribe and incite in a single column.) This system could also allow you to pay protection money (bribe) for another civilization/minor. If a city-state you are suzerain of is getting hammered by a barbarian tribe (pillaged luxuries, strategic resources, etc.), you could put an end to it.
Additionally, when a tribe is incited against you (or maybe even when a Barbarian Attack or Barbarian City Assault operation is launched) the hire diplomatic option should be locked from you. This is more of a flavorful suggestion. I just find it kind of ironic that you can hire some barbarian units to fight against their native tribe. Was this a strategy ever employed by a civilization in the past? Maybe Rome with some Germanic “tribes”? Maybe @Boris Gudenuf could enlighten us on this subject? If this was ever done, I doubt it yielded any good result… Alternatively, any hired unit should suffer an heavy combat penalty when fighting members of its native tribe.
6.
Distinguishing barbarian clans
When you play your first game with the barbarian clans game mode on, trying to find out what type of clan you are up against can be a non-intuitive task. For this purpose, each tribe has a unique name (42) and each tribe name is linked to a specific clan (7). However, when I first played the game mode, I found myself constantly alt-tabbing to the wiki to find out what type of clan I was up against. This continued until I came pretty familiar with all of the names (noticing the common theme between them) and finally memorizing them all. Focusing only on the “symbol” of the tribe can also be misleading as some symbols can be reused by tribes belonging to a different clan (such as the ant and the skull symbols). I believe there should be a more intuitive way to distinguish right off the bat what clan you are up against without having to remember names or refer to a wiki.
6.1
Defender units in camps
Usually, the first thing you notice when you meet a new barbarian camp is what unit is defending it. As some of you may already know, the idea to place defender units from different classes in barbarian camps and changing the mix of units they would spawn
originally came from me. Currently, 3 different unit classes (anti-cav, melee and ranged) are used as defenders in the camps spread amongst the 7 different clans. Thus, tribes belonging to completely different clans can have the same unit class used as a defender. Ideally, I believe that each clan should have their own unique unit class used as a defender. The developers seem to be allergic to the idea of using cavalry units as defenders despite the fact that cavalry units can all fortify in this iteration of the game. I’m no expert on pastoral tribes, but I find silly the idea that a tribe would have access to swords, but would tell all their swordsmen: “You stay here and defend the camp”. While I am unsure about this historical accuracy of using cavalry units as camp defenders, I know it would make sense from a gameplay perspective. When was the last time you used an anti-cav unit to clear a barbarian camp? That’s right, never. This idea combined with my idea of “Reinforcing camps” (which I will develop in a later section) would make the usage of anti-cav units a must in certain situations. I thus recommend that the mobile ranged, light cavalry and heavy cavalry unit classes also be used as defender units for the Rover, Chariot and Jungle clans. I will outline the specifics in the next section. This now leaves us with 6 different unit classes for 7 different types of clans…
The last undiscussed clan is the Seafaring one. In order, to give them their own unique defender class, a slight tweak will need to be made first: camps that can spawn in coastal water, in other words, pirate havens! Seafaring clans would indeed spawn in a coast tile with at least 3 adjacent land tiles and could thus have a naval melee unit as a defender. This would force the player to either use naval or land ranged units in order to clear it. An 8th type of clan could also be introduced, one that uses a naval ranged unit as a defender (and which could produce a different mix of naval units).
6.2
Clan & terrain types
There is currently something very wrong about how the game decides which type of barbarian clan it should spawn according to the surrounding terrain of the barbarian camp. It currently looks like the game only checks if a certain criterion is met within a range of 1 (for example, 1 tile has woods) and then says ok let’s spawn a barbarian tribe of the woodland clan here. Never mind the fact the 5 other tiles surrounding the camp have hills, since there is 1 adjacent wood, a tribe from the woodland clan can spawn here. I believe the game should check for the predominant type of terrain in a 1 tile range to decide which type of clan it should spawn. For example, if out of 7 tiles, 3 tiles have hills, 4 tiles are flat and 5 tiles have woods on them, then a tribe from the woodland clan should spawn here. This would get rid of aberrations where, for example, a tribe from the hills clan spawns in the middle of a forest, because there is only 1 adjacent hill to the camp. I mean if there are from the hills clan, that name must mean something right? The game should also keep away from the other extreme: for example, require that all tribes from the woodland clan spawn on a tile that has woods. This is because it reduces the different possibilities of combat strength differentials you can get when fighting a certain type of fortified unit in different types of terrain. For example, I still want to be able to fight a melee unit that is fortified on flat land, woods, hill, marsh or hill and woods. It will just be more likely that I fight a melee unit fortified in barbarian camp of the Hills clan on a hill, but it will not be a guarantee.
Another issue is that with the barbarian clans game mode activated you have less chance of finding a barbarian camp that fields cavalry (“a horse barbarian camp”). With the mode deactivated, any barbarian camp that is located within 3 tiles of a tile that has horses will automatically become one. However, that is not the case with the game mode activated. I have witnessed many cases where a camp had access to horses within a 3 tile range, but was not a member of the 3 types of clan that produce cavalry units (Rover, Chariot or Jungle clans). The logic conditions must thus change so that any barbarian camp that has access to horses automatically becomes one of the 3 types of clan that can spawn cavalry units. (Similarly, not all barbarian camps that spawn next to coast with automatically become tribes from the Seafaring clan if I am not mistaken.)
Not only are we seeing less “horse barbarian camps” in general, but the conditions required for the Chariot and Jungle clans to appear are way too strict. You can easily play 20+ games in a row without seeing even once a tribe from these clans. What should be the new requirements for these clans? I have no clear answer to offer. Removing the horse requirement from the Chariot clan and requiring that a majority of the tiles within a range of 1 from the camp are desert seems like a simple idea easy to implement. After all, heavy chariots and units from the heavy cavalry class do not require horses to build. However, people who like to play as civilizations that have a desert bias may not like that idea too much as camps from the Chariot clan should be considered of higher difficulty than the other more common ones. One way to balance it would be to make it easier by reducing the proportion of heavy cavalry units it spawns and maybe change the light cavalry units by mobile ranged. Trying to balance the Jungle clan would be even more of challenge. For a start, ivory is an even rarer resource than horses and it is not even a guarantee for it to appear on all maps generated. I also find ludicrous the idea that a camp in the middle of the jungle would start to spawn light cavalry units. The fact they can spawn heavy cavalry units is ok as long as they are Varus or other units that use elephants such as the Voi Chiens (but those are considered ranged units). Also, the wiki mentions that spears were frequently used by these tribes, but that clan does not spawn any anti-cav military units. If the developers really want to keep the ivory requirement and keep this clan as “a special case scenario” that you rarely see, I would customize it further by making the Varu always its defender unit and it would only spawn Varus, Voi Chiens and Domreys. This type of camp would only spawn in the medieval era. If you have not yet scouted or settled nearby ivories by then, I say you deserve to pay the price.
7.
Tech stealing
Barbarians also acquire technologies. This is done through a passive tech stealing process behind the curtains. When they acquire technologies which unlock new units, they are then able to train them. The developers also us to partially tweak the speed at which this is done through these 2 variables:
BARBARIAN_CLANS_TECH_STEAL_INTERVAL_STANDARD
BARBARIAN_CLANS_TECH_STEAL_VARIANCE_STANDARD
However, it seems that barbarians for some obscure reason are able to steal technologies which have not already been discovered by any other civilization yet. From whom did they steal these technologies unheard of before? Who knows? Indeed, I found in the log files related to the barbarian tribes that they had discovered Gunpowder on turn 108 and the Scientific Theory on T117 (deity game). However, in the AI Research log file (which is a bit hard to read), none of the AIs had “owned” these 2 technologies at that point in time. Under the current settings, where most barbarian camps are gone by turn 110 (due to early conversion to city-states), this might not be a problem. However, imagine a barbarian camp that has launched a “Barbarian Attack” operation that produces one musketman per turn without being limited by any resource requirement on turn 108: a scary scenario indeed. This is why I think there should be a way to limit the technological level of barbarians to be equal at most to the most advanced civilization (with maybe some other options to set it to the median technological level of the civilizations with some variance). It might be possible to achieve a similar end result by lowering the values of the 2 previously mentioned variables, but I would prefer if it were possible to establish a hard upper bound (ie eliminate any technology not researched by any civilization yet).
8.
Additional random observations
- AIs will use bribe for camps that are located far out (15+ tiles) which should be of very little use to them.
- Once a ranged unit used as a defender in a camp has done an attack, it will no longer fortify back in the camp even if it can no longer attack any units.
- After killing the defending unit in a barbarian camp, even if you put a military unit in the camp, an AI-controlled unit which has multiple movement points can “pass through” the camp and either disperse or raid it.
In my next post, I will elaborate on more complex ideas that would change more drastically the Barbarian Clans game mode.
