Humankind Game by Amplitude

[Kwami]

The biggest problems weren't just numbers that needed rebalancing, though. The biggest problems were a really bad, uninformative UI (worse than CiV VI even!) and a great lack of depth in pretty much every core mechanic. I doubt that they were able to fix either of those problems in the last month or so.

 

[Krajzen]

The biggest problems weren't just numbers that needed rebalancing, though. The biggest problems were a really bad, uninformative UI (worse than CiV VI even!) and a great lack of depth in pretty much every core mechanic. I doubt that they were able to fix either of those problems in the last month or so.

That sounds worrying, but I don't think I remember playing a grand/4x strategy game which didn't receive "depth" complaint, including the most highly rated ones, so I shal see if it is a local subjective opinion or a symptom of a larger problem... How would you exaxtly define "depth" in a game like this?

I didn't find civ5 shallow at all, even before expansions, but I had this profound hard - to - explain feeling with civ6. Maybe that's due to the combination of "horribly low difficulty level as AI can't use neither unstacked combat nor unstacked cities" and "awful pacing and unberably static, preditable second half of the game, with boring victory conditions". As the result of those two major failures, I had no motivation to interact with game's numerous mechanics, because I have felt like "all I have to do is conquer some braindead rivals early and I am untouchable runaway economy by the renaissance era - why bother with any other mechanics".

My biggest worry regarding this game is that Amplitude is too attached to the idea of their FIDSI yields and builds too much of their games around the endless linear accumulation of yields, via the economy of a thousand small micromanaging cuts but no major macro - level strategic decisions. But I didn't play later non - combat betas so I don't know how far away are we from this problem.

 

[Kwami]

That sounds worrying, but I don't think I remember playing a grand/4x strategy game which didn't receive "depth" complaint, including the most highly rated ones, so I shal see if it is a local subjective opinion or a symptom of a larger problem... How would you exaxtly define "depth" in a game like this?

I didn't find civ5 shallow at all, even before expansions, but I had this profound hard - to - explain feeling with civ6. Maybe that's due to the combination of "horribly low difficulty level as AI can't use neither unstacked combat nor unstacked cities" and "awful pacing and unberably static, preditable second half of the game, with boring victory conditions". As the result of those two major failures, I had no motivation to interact with game's numerous mechanics, because I have felt like "all I have to do is conquer some braindead rivals early and I am untouchable runaway economy by the renaissance era - why bother with any other mechanics".

My biggest worry regarding this game is that Amplitude is too attached to the idea of their FIDSI yields and builds too much of their games around the endless linear accumulation of yields, via the economy of a thousand small micromanaging cuts but no major macro - level strategic decisions. But I didn't play later non - combat betas so I don't know how far away are we from this problem.

Humankind has some of the same problems. If you start warring your neighbors, then you'll be picking up military stars, district stars, population stars, and territory stars all at once. You'll get generating more science and influence with every city capture. It's a huge snowball, just like other 4X games.

Aside from that, though, the systems in Humankind just feel unfinished. Religion is 95% passive and just gives you bonuses. The only thing that you can really do is build your (limited) holy sites. Culture spread is entirely passive. Trading with the AI is just a matter of guessing. And a lot of the city improvements and districts just don't matter because the unique districts are so much better than everything else. So, most of the game seems pointless.

Maybe some of this would change if the AI was remotely competent, but it isn't, and history suggests that it won't ever be. Almost no 4X game has good AI and Amplitude's past games certainly do not.

Still, I hope that they work it all out and that I'm pleasantly surprised. I'm looking forward to playing a new game!

 

[FinalDoomsday]

Humankind has some of the same problems. If you start warring your neighbors, then you'll be picking up military stars, district stars, population stars, and territory stars all at once. You'll get generating more science and influence with every city capture. It's a huge snowball, just like other 4X games.

Aside from that, though, the systems in Humankind just feel unfinished. Religion is 95% passive and just gives you bonuses. The only thing that you can really do is build your (limited) holy sites. Culture spread is entirely passive. Trading with the AI is just a matter of guessing. And a lot of the city improvements and districts just don't matter because the unique districts are so much better than everything else. So, most of the game seems pointless.

Maybe some of this would change if the AI was remotely competent, but it isn't, and history suggests that it won't ever be. Almost no 4X game has good AI and Amplitude's past games certainly do not.

Still, I hope that they work it all out and that I'm pleasantly surprised. I'm looking forward to playing a new game!

I've played 4X games where I'd describe the AI as competent and those include Endless Legend and Endless Space 2. Because of the vastly assymetric factions the AI would perform much better with some factions and far worse with others but you would still face some strong mid game challenges. Old World and Gladius also have AI that can play with one unit per tile rules very well.

Civ 6 has remarkably bad AI that cant pose any kind of threat beyond the very early game and while making fun AI for games like these is very hard these other games do prove it can be done.

 

[Kwami]

I've played 4X games where I'd describe the AI as competent and those include Endless Legend and Endless Space 2. Because of the vastly assymetric factions the AI would perform much better with some factions and far worse with others but you would still face some strong mid game challenges. Old World and Gladius also have AI that can play with one unit per tile rules very well.

I haven't played enough ES2 to comment on it, but it had ridiculously bad AI at launch. Famously so. The AI in Endless Legend, nine years after release, still doesn't know how to do any quests except for faction quests. Really! And it just kills every minor faction village because it doesn't know how else to handle them. And if you turn the minor faction difficulty up, the AI just dies because it can't deal with the roaming armies. The AI in that game is horrible. If you're losing, it's because the bonuses granted to higher-level AI players are insane.

 

[DaviddesJ]

I haven't played enough ES2 to comment on it, but it had ridiculously bad AI at launch. Famously so. The AI in Endless Legend, nine years after release, still doesn't know how to do any quests except for faction quests. Really! And it just kills every minor faction village because it doesn't know how else to handle them. And if you turn the minor faction difficulty up, the AI just dies because it can't deal with the roaming armies. The AI in that game is horrible. If you're losing, it's because the bonuses granted to higher-level AI players are insane.

It's perfectly reasonable to just give the AI players bonuses, to compensate because they can't or don't do quests, etc. Comes out basically the same.

 

[Kwami]

It's perfectly reasonable to just give the AI players bonuses, to compensate because they can't or don't do quests, etc. Comes out basically the same.

Quests are a major part of the game, though. And many quests are cooperative or competitive. When the AI can't do those quests, they don't function correctly.

Besides, lots of folks here complain about the Civ VI AI getting bonuses. I'd imagine that those folks won't like the AI bonuses in Humankind, either.

 

[Elhoim]

I've found the Humankind tactical AI quite acceptable. When you fight an AI in equal combat strength terms, they can do a pretty good job and you need to be careful about it. What might complicate the AI, and combat difficulty in general, is it being a bit behind on tech or lacking resources, so battles can become very loopsided even with just being one unit upgrade higher. But that can also happen to the player as well. I'd say the AI is competent on the general game, and good on the tactical side. It gets bonuses in higher difficulties, but I wouldn't consider it bad as a baseline.

 

[SupremacyKing2]

What might complicate the AI, and combat difficulty in general, is it being a bit behind on tech or lacking resources, so battles can become very loopsided even with just being one unit upgrade higher. But that can also happen to the player as well.

There have been plenty of historical battles between two sides that were not evenly matched technologically. And the side with the better tech having a big advantage is realistic. And like you said, it can happen to the player too. So I don't think that is a big problem.

 

[Boris Gudenuf]

There have been plenty of historical battles between two sides that were not evenly matched technologically. And the side with the better tech having a big advantage is realistic. And like you said, it can happen to the player too. So I don't think that is a big problem.

Bigger Tech Advantage works unless you happen to have that advantage at Adowa, Isandhlwana, or Little Big Horn. Numbers count too.

Although, when the Dupuy Institute did one of their first statistical studies of warfare and battles, they found in analyzing battles from Ancient to Modern Eras that having superior numbers was one of the least important factors in determining victory. It seems that weapons technology, organization, leadership, morale, and terrain all had much more effect. This is interesting because in Humankind's tactical battles, terrain and weapons technology seem to be the most important determinants of victory, while Leadership, Morale and Organization have to be (since the game has no Generals) internal to the units.

It will be interesting to see what form further development of the system takes: I don't see them introducing Generals, because the game has no individual 'figures' at all, but they might add some Technological or Civic changes/choices that affect Leadership or internal Organization in battle. Something like the infamous German Auftragstaktik or French revolutionary/Napoleonic 'Mixed Order' as tactical 'soft factors'

 

[Taefin]

Maybe I just played too much Civ6 but it is surprising refreshing to hear a developer acknowledge the actual balance issues identified by the community. I’ve been playing Old World lately and their team is awesome and very receptive to our feedback, and I hope that Amplitude follows suit. Identifying tech/era pacing and the flatness of religion and civics seems like they had a strong idea of what was right (I.e. keeping era pacing so fast through each beta), but were ultimately open to feedback that went counter to their intuition.

Min-max wise, I feel the game is very nearly engaging, giving a strategic depth at least of how long to invest in each yield before cranking out an army. For this to work the AI needs to be aggressive and semi-competent enough to win when it has the superior force and put pressure on the player if they get too greedy in one direction or play too cautiously. And to me, the game feels mostly about stacking some crazy culture bonuses in different ways each game. Betas felt like playing 4 games of Civ6 in one, getting to move onto the next as soon as I got bored of the last.

If they can get Vassals to team up with potential adversaries during subsequent wars, and perhaps stack war weariness against the colonizer, I think that the massive gold boost you get for subjugating a neighbor can be balanced by the increased risk when they rebel.

I also rather enjoy the HK UI, sure it suffers for lack of middle-click tool tips but I enjoy that you can play the entire game with just the mouse. Excellent if you play while exercising. And the game is quite simple once you have the basics down. It’s really learning the ins-and-outs of what exactly each bonus affects where I feel this lacks the most. Hopefully we get that yield summary when hovering over infrastructure options at release.

 

[Krajzen]

Maybe I just played too much Civ6 but it is surprising refreshing to hear a developer acknowledge the actual balance issues identified by the community.

Yes indeed, what a curious alternative to the of course equally valid approach of "silence for months and years until single patch drops, fixing one fourth of the fundamental problem, meanwhile we add zombies to the game"


On another note, I am currently reading about Sanskrit, Vedas and Vedic India and sweet brahman, Vedic/Indo - Aryan culture would be such an amazing thing for Humankind's ancient era

Epic, impossible in other games like this, not overlaoping with neither Maurya nor Harappa, and orobably the only other legit aestethe/especially scientific ancient culture

 

[DaviddesJ]

Quests are a major part of the game, though. And many quests are cooperative or competitive. When the AI can't do those quests, they don't function correctly.

No, that just makes the game asymmetrical. Almost all computer games are asymmetrical, the AI players play differently than the humans. Most people playing these games don't actually want the computer opponents to play exactly the same as the human players, although you are not wrong that there are a few who grouse vociferously when they don't. But those people can just be unhappy with the game; that's fine with me. There is no point in trying to please everyone.

 

[Haig]

From what I have seen/read about HK and beta feedback, my biggest worry is the lack of personality.

Civ has memorable leaders and factions, Endless Legend had very unique factions(leaders were bland though) but Humankind doesnt have that at all.

I really dont warm up to the changing cultures either, like how your enemy morphs between nations. I like how in Civ your empire has its power eras andnsome bonuses that are useful for each era.

It seems that HK ai doesnt play with the same rules as human player when it comes to combat and some mechanics, it is also annoying when it happens at civ like how in Civ 5 ai doeant pay upkeep for armies.

I feel HK has some "emperor's new clothes" going on for it with some people, but we will see, I am definately curious and might buy it if it impresses.

 

[Boris Gudenuf]

From what I have seen/read about HK and beta feedback, my biggest worry is the lack of personality.

Civ has memorable leaders and factions, Endless Legend had very unique factions(leaders were bland though) but Humankind doesnt have that at all.

I really dont warm up to the changing cultures either, like how your enemy morphs between nations. I like how in Civ your empire has its power eras andnsome bonuses that are useful for each era.

It seems that HK ai doesnt play with the same rules as human player when it comes to combat and some mechanics, it is also annoying when it happens at civ like how in Civ 5 ai doeant pay upkeep for armies.

I feel HK has some "emperor's new clothes" going on for it with some people, but we will see, I am definately curious and might buy it if it impresses.

As I've posted elsewhere, Civ and HK are taking their basis from different schools of historiography: Civ the Great Man view of history, HK the Great Forces view (once upon a time called "Marxist"). I agree, that means that in HK there are no Great Men (or Women) - in fact, no named individuals at all, in any form. That will take some getting used to and will definitely not appeal to some people.

On the other hand, I think (from what I've seen) that HK models the long-term trends in history better, and gives the gamer some very interesting interactions/decision-points between Problems brought on by long-term interactions with other cultures and factions, available resources and trade opportunities, and diplomatic interactions and gives the gamer a better 'feel' for how individual decisions by the gamer can affect what happens to his faction in the game. We've yet to see, of course, how well all that will work in the game when it is Released and subjected to play by a vastly larger group than any 'test' group they've exposed it to.

And I was extremely sceptical of the 'changing culture' mode when I first heard about it: as an avowed historian, By Herodotus I want to see historical progressions: Greeks to Byzantines, Zhou to Han, Tang, Song Chinese, etc.
But, after playing with it through several OpenDev iterations, it grows on you. For one thing, each type of Faction has a different ability/attribute that carries over into the succeeding Eras, so that after 2 - 3 varying choices, your Medieval or Early Modern pick does not have quite the same attributes in every game. To some extent, then, you 'build' a unique Faction the longer you play, until your Modern Australians can have 'bits' of Khmer, Mycenean, Siamese, Ming, and Austrian in them - or any other of myriad combinations. Again, it will be interesting to see how gamers 'take' to this in the marketed version of the game.

Finally, based on comments from some of the VIP testors and other hints, it appears that the oher Major Factions do play under rules similar to the Human Player, unless you play at higher or lower than 'average' difficulties. Like most computer games (and all the Civ variations), the difficulty differences are based almost entirely on the amount of 'cheats' given to the human or AI players. On the other hand, (and this is Very different from Civ VI, at least) the AI and the human seem to be pretty evenly matched in Tactical Battles - the AI uses terrain very well, picks its attacks pretty well, and seems to know when to go on a 'hedgehog' defense to extract the maximum losses from the human's units when there is no other good tactic available. Again, it will be interesting to see how well it all works with thousands of players trying it in the (presumably) 'tweaked' release version of the game. - And at least in the Open Dec versions, the AI major Factions were paying Upkeep for their units, but the Minor Factions (think Civ VI's Barbarians and City States combined) don't seem to pay upkeep and 'spawn' units at a ridiculous rate - very similar, unfortunately, to Civ VI.

Humankind will be different. In fact, a lot of Civ VI players will doubtlessly complain, in so many words, that it "isn't Civ" and therefore familiar in its mechanics and depictions. This is short-sighted: we've had the Civ Version of 4X historical for years, the complaints about it have filled digital volumes of posts on this and other Forums, it is past time to try some new things in the genre and see what works differently, or (again, as I've posted elsewhere in this Forum, along with many others) what can be used, modified, assimilated or 'massaged' to make the next iteration of Civ better . . .

 

[Zaarin]

From what I have seen/read about HK and beta feedback, my biggest worry is the lack of personality.

Civ has memorable leaders and factions, Endless Legend had very unique factions(leaders were bland though) but Humankind doesnt have that at all.

This was a concern of mine as well, but I think that @Boris Gudenuf 's post pretty well sums up what my actual experience with the game has been. Also worth noting the AI leaders do have something of a "personality," albeit not as vibrant as a Civ leader. HK has chosen to downplay leaders and play up culture. That should make the anti-leader crowd in the Civ forums happy. For me, I find it an interesting approach in HK that I don't want to see migrate over to Civ; it's best the two franchises go their own separate ways.

 

[Taefin]

What I’m excited/dreading to see at release is whether they’ve added a retreat option during a battle, or done something so that the purpose of individual battles is no longer to annihilate your opponent’s army in one fell swoop. The siege mechanic is awesome! And I was finding in Beta that the AI’s army was dead by the time my force got to their cities, then the siege process is trivial and boring. It felt very off that at some point a few turns into a big battle, after ~20% casualties, it becomes clear who will win the battle, but you keep playing a few more turns while the winner massacres the entire enemy army. Such opportunities (per my understanding) we’re rare historically, and seem to place of the stakes of the battle on inflicting casualties rather than securing territory which (again per my understanding) is what the dynamics of warfare actually hinge on.

I hope that this can at least be a game option, since all it would require is programming the AI to judge the victory likelihood of a matchup each round, which it seems may be no different than what they already do when deciding to retreat in the first place and whether to send in reinforcements. If fighting in hedgehog, and sufficiently outnumbered they should retreat.

Endless legend had a similar issues where the game quickly becomes about picking battles and their boundaries to isolate small portions of your opponent’s army (and/or drawing the battle boundaries to create artificial cover and choke points) and then killing everything with minimal casualties yourself. HK’s reinforcement system alleviated part of this problem, their hedgehog behavior eliminates another (AI won’t march “endless”ly against your artificially fortified position), but the main dynamic still feels too similar to me. Also it seems that HK has a soft version of this option, with the capture the flag goal, but I’ve found this rarely prevents the massacre of the losing side.

And in retreat, units are still vulnerable, especially to cavalry armies that can outrun them (and maybe I’ve listened to one too many Dan Carlins, but wouldn’t that be more historical as well?).

One additional bonus of a mid-battle retreat option would be that when a sieged city was clearly lost, it could be surrendered, saving population and units. This would allow AI (and players) to lose a war and even become a vassal, but retain some of their fighting strength to pose a risk afterward.

 

[Boris Gudenuf]

What I’m excited/dreading to see at release is whether they’ve added a retreat option during a battle, or done something so that the purpose of individual battles is no longer to annihilate your opponent’s army in one fell swoop. The siege mechanic is awesome! And I was finding in Beta that the AI’s army was dead by the time my force got to their cities, then the siege process is trivial and boring. It felt very off that at some point a few turns into a big battle, after ~20% casualties, it becomes clear who will win the battle, but you keep playing a few more turns while the winner massacres the entire enemy army. Such opportunities (per my understanding) we’re rare historically, and seem to place of the stakes of the battle on inflicting casualties rather than securing territory which (again per my understanding) is what the dynamics of warfare actually hinge on..

Battle hinges on Morale. The task is to convince the enemy, and especially his leaders, that they cannot win or have already lost, and so they go away (sometimes at high speed). This can be related to suffering casualties, but not necessarily directly. On the one hand, L. Trotsky famously said that the quickest way to demoralize the enemy was to kill him in large numbers, on the other hand, there are numerous historical battles in which the victor suffered higher casualties than the loser, but could still claim victory because the enemy retreated or otherwise surrendered the battlefield to the victor.

I hope that this can at least be a game option, since all it would require is programming the AI to judge the victory likelihood of a matchup each round, which it seems may be no different than what they already do when deciding to retreat in the first place and whether to send in reinforcements. If fighting in hedgehog, and sufficiently outnumbered they should retreat..

There are many more variables, and unfortunately many of them are 'soft' factors like Leadership, Morale, Doctrine, Tradition, and other factors that don't yield themselves well to numerical (computer) calculations. Back in the 1950s the US Army put together a Maneuver Control manual for umpires to judge the outcomes of tactical exercises and full scale 'war games', but despite reams of statistics on casualties and outcomes of tactical actions from WWII and the Korean War, hey still wound up relying, more often than not, on the judgement of an umpire with combat experience to decide what would happen in a given tactical action. Having known a few professional rules-writers and written a few myself, I can tell you that In tactical miniatures rules, "Morale" and its effects are one of the stickiest things to manage because of all the Uncalculable 'soft' factors that apply.

Endless legend had a similar issues where the game quickly becomes about picking battles and their boundaries to isolate small portions of your opponent’s army (and/or drawing the battle boundaries to create artificial cover and choke points) and then killing everything with minimal casualties yourself. HK’s reinforcement system alleviated part of this problem, their hedgehog behavior eliminates another (AI won’t march “endless”ly against your artificially fortified position), but the main dynamic still feels too similar to me. Also it seems that HK has a soft version of this option, with the capture the flag goal, but I’ve found this rarely prevents the massacre of the losing side..

When I first saw it, I thought that the "capture the Flag' option would be a great way to show that one army had lost the fight in their minds and could therefore retreat, but that doesn't seem to be the way it worked in the Open Dev versions. On the other hand, especially in the late-game, if a side is not wiped out in the first 3 rounds of Battle, you progress to the next Game Turn and anybody within range can join the on-going battle. Still missing the retreat option, but the result (according to at least one Reddit poster) can be a massive stalemated battle in which more and more units get sucked in, casualties mount, but neither side can wipe out the other - shades of some of the mass battles of WWI and II. I believe in one example they had over 50 units on a side slugging it out, which on the one hand is awesome but on the other hand, compared to the general scale of the rest of the game, seems a bit OP.

And in retreat, units are still vulnerable, especially to cavalry armies that can outrun them (and maybe I’ve listened to one too many Dan Carlins, but wouldn’t that be more historical as well?)..

Cavalry pursuing running infantry is another definition of a massacre. On the other hand, troops who think they are pursuing a beaten enemy and run into a counterattack by fresh forces or an intact rearguard can get their heads handed to them before they realize the situation is not what they presumed: Jack and I have researched a number of instances of exactly those things happening in October 1941 in front of Moscow, when ''pursuing' German forces ran into enemy forces that were not only not running, but were attacking. The result was at least 4 German infantry divisions and 3 panzer divisions getting badly mangled before they could pull themselves together tactically.

Unfortunately, that means that when a strategic game tries to 'descend' to the tactical level the way Humankind does, it get complicated and the number of factors to consider multiplies quickly.

One additional bonus of a mid-battle retreat option would be that when a sieged city was clearly lost, it could be surrendered, saving population and units. This would allow AI (and players) to lose a war and even become a vassal, but retain some of their fighting strength to pose a risk afterward.

Sieges and Battles have to be handled differently. In the open field, troops that decide that there is no victory to be had can and will 'vote with their feet' and retreat with or without orders. Stuck inside a besieged city, the only option is to surrender before the enemy successfully assaults the walls - once the attackers are in the city by assault it is too late to surrender, and almost inevitably a general ransacking and massacre takes place. That's why. in the 17th and 18th centuries, a 'protocol' of sieges was developed in Europe: once certain conditions were met (the attacking cannon and engineers had blown a large hole in the defenses, for instance) the city defenders could 'honorably' surrender and in many cases, were allowed to march out and either rejoin their own forces or be exchanged for ransom or prisoners of the attacker's own army. That would be a great mechanism to have in the game, but I've seen no sign that they have it or have contemplated it. Maybe for a post-release Fix or DLC . . .

 

[Taefin]

Cool info/insights, thanks for sharing! I wonder if adding a little more space behind each flag would capture some of the retreat effect I’m otherwise hoping for.

We’ll see about the we 50 unit battles. My second Victor game ended when the game crashed during a musket war with 15-20 units on each side. Sounds like they tidied up the gunpowder unit mechanics, I could hardy line up any shots with the line of sight mechanic, and it got awkward when everyone jumped in the water.

I’m guessing, if they can keep the player and AI on roughly the same tech it will be pretty fun regardless.

 

[Krajzen]

From what I have seen/read about HK and beta feedback, my biggest worry is the lack of personality.
Civ has memorable leaders and factions, Endless Legend had very unique factions(leaders were bland though) but Humankind doesnt have that at all.
I really dont warm up to the changing cultures either, like how your enemy morphs between nations. I like how in Civ your empire has its power eras andnsome bonuses that are useful for each era.

I am not that afraid of a lack of personality, because I have played a lot of Paradox games, where you interact with:
- a myriad of faceless countries, without even a picture of their leader (EU 4), only via diplomatic proposals menu
- the same but with the bare minimum of foreign leaders having portraits (HoI4, Vic2)
- alien races randomly generated from the pool o interchangeable characteristics and leader avatars (Stellaris)
- a myriad of actual characters with personalities, who are however constantly changing and without too close relations with any single one of them (Crusader Kings)

...and in every case it sort of works, despite all those systems of opponents having a lot less personality than civ5/6 style "a constantly returning staff of fully animated leaders with dialogue, personality, quirks and stuff". Different games, different mindsets and ways to immerse the player. Personally I'm open to play in a 4X game without eternal leaders due to the novelty and lack of several issues of that system - the inability to add cultures without well documented Big Personalities, figures of singular men overshadowing the cultures they are from (India always being ultra pacifist as a whole because of Gandhi is my most hated example), constantly recurring faces, the eternal headache of female representation etc.

Regarding morphing cultures, I am actually very enthusiastic to try this concept due to its sheer revolutionary novelty. All 4X games ever had the problem of AI factions being monoliths unchanging from beginning to the emd with their core characteristics, powers, aestethics and behaviour, which is boring in fantasy 4X and downright unrealistic in historical ones. IRL the history is not "the set of cultures who remain from the ancient to the modern era, unless they are conquered from the outside". Humankind is the first step to the much more dynamic, interesting and chaotic flow of historical simulation. The second step would be new factons dynamically emerging/rebelling/splitting during the game and then being capable of threatening biggest powers - just like in real history.

In general, I respect HK devs for not being afraid of shaking 4X holy dogma of "12 factions start the game and the exact same 12 factions (minus conquered ones) end the game". It's not like in our timeline the year 2021 is a battle of dominance between Sumer, Elam and Harappa