Thoughts: Endless Space 2

Warning: getting maps off other people makes the endgame chug a bit.

It took me a while to warm up to Endless Space 2. I regarded the first game as something of a qualified success, but the fantasy followup Endless Legend left me completely cold despite having some ideas and mechanics that were, objectively, very good indeed – it’s the first time ever that I’ve bounced off a game without being able to really explain why, and to start with I was afraid that the same might be true of Endless Space’s sequel. Partially this is because I made the mistake of buying it a week before it came out of Early Access, foolishly assuming (because I’d done the same thing with Battle Brothers and had a whale of a time) that it wouldn’t be too different from the finished article; instead Amplitude released a 3 gigabyte patch on launch day that papered over a lot of the obvious Early Access holes and made it significantly more coherent as an end-to-end experience.

Mostly though — and even after 27 hours playing it, I still think it is a valid criticism — it is because, just like Endless Legend, Endless Space 2 kind of sucks at building the emergent story of your empire’s growth from a single planet to a galaxy-bestriding colossus. This is surprising given the sincere effort both games have made to improve on the comparatively sterile experience of Endless Space by injecting oodles of character into every element of the game; ES 2’s races are fully voiced, have unique techs in the tech tree, narrative questlines that provide some backstory to who they are, and — very importantly — each have at least one gimmick that ensures they play very, very differently, from the nomadic Vodyani who have giant arkships instead of colonies and who must generate new population units by leeching life essence from foreign inhabited systems, to the mechanical Riftborn who build new population in their colony production queue and who can conjure up time-manipulating singularities around systems that provide either a bonus or a malus to any colonies located there. Heroes and leaders have 2D animated portraits, there’s an absolute ton of unique art for improvements, techs and events, Endless Space 2 has taken note of Stellaris’s early game exploration and provides its own stripped-down version where you send scout ships around to explore anomalies and curiosities etc. etc. — there is a lot to like about Endless Space 2, as well as some really good ideas around trade and military that I’ll go into in a second.

Less confusing than it looks, trust me.

The thing is, nearly all of that was also true of Endless Legend. I’m getting on better with Endless Space 2 now since I instinctively prefer the sci-fi setting — and not just on a thematic level, since the standard star-nodes-linked-by-warp-lanes layout that the galaxy map goes for fits the gameplay far better while allowing for plenty of interesting decision space around ship movement — but there’s no getting around the fact that the two games are very similar. In particular they share a crucial weak link: the FIDSI system that all of Amplitude’s Endless games are based around. This will be very similar to anyone who has ever played a 4X before, as it’s shorthand for the usual 4X production of Food, Industry, Dust (money), Science and Influence (culture). The Endless series puts a particular spin on it, however: it puts generation of these resources front and centre and has a ruthless, ruthless emphasis on making them go up. Which is fair enough in a way, since this is the feedback loop at the core of all 4X games: you improve your colonies by building buildings that generate more resources that let you build bigger buildings etc. etc. The thing is, something like Civ VI1 understands that building a Library should have an impact in the game beyond making a number go up, even if it’s a superficial one. This is why the Civ series is quite big on having your improvements appear on the map and provide some level of interactivity or benefit besides a +20% to colony Industry or a +3 Science per colony population. If you boil a 4X down to a pure numbers game it comes dangerously close to being exposed as the time-wasting End Turn-clickathon these things really are – but this is exactly what the Endless series does, with improvements disappearing into the ether the moment they’re built with the only tangible evidence of their existence being that you’re now producing 10% more Science or whatever.

If nothing else, Endless Space 2 has at least let me figure out what bugs me about the series. It’s not that the various empires don’t feel distinctive or characterful, because they do. They just don’t feel like my empires; whereas most cities or colonies in regular 4Xes end up having a fairly distinct story behind them the only story behind a typical system in Endless Space is that there was colonisable space there and you expanded into it to generate more FIDSI. This is compounded by Amplitude’s love of runaway exponential growth curves and some truly atrociously tuned win conditions. There are six victory types in the game: Conquest (capture all opponent homeworlds), Supremacy (occupy x% of the systems in the game), Science (research the four techs at the very end of each of the categories in the tech tree), Wonder (build X number of special wonder buildings across your colonies), Economic (generate X amount of money during the game) and Score (game ends at turn 200 if no other victory condition is fulfilled). It feels like Conquest and Supremacy are the same thing since you’ll usually end up accidentally doing one while trying to do the other, but otherwise there’s no problem with them; they feel like meaty achievements that have some good narrative moments scattered along the way. Economic, Science and Wonder victories are horribly broken right now, however, as they’re all expressions of how good your empire is at generating at FIDSI — and thanks to some extremely poor balancing it’s very, very easy to thoroughly break your FIDSI output so that you’ll achieve any one of these conditions about 10-15 turns after deciding to go for it. Economic victories in particular are really bad because you can break your Dust generation so hard with the trade system you’ll win by Economy every time unless you turn that victory type off.

(This is how broken Dust generation is in Endless Space 2 right now: there are achievements for generating 1 million, 2 million and 4 million total Dust across any number of playthroughs, the implication being that this should take some considerable time to do. I got all of them in a single game when I turned off the Economic victory condition that was capping my Dust gathering.)

Those numbers are going to get a lot bigger.

Anyway, the ease with which you can win means games of Endless Space 2 end extremely abruptly. Fighting wars is the best way to generate your own narrative here, but it’s very difficult to kill off more than one race before you win by other means. The fact that none of the non-war victory conditions are particularly interactive compounds the problem: the Economic and Science victory conditions are literally “Get X amount of Science/Dust to win”, while the Wonder victory condition has a little more nuance to it but boils down to having Y number of colonies with good Industry output, where Y is the number of wonders required to win. It’s all rather unsatisfying to interact with, and feels more like you deciding to end the game because you filled out the tech tree and built all the improvements than it does a real, earned victory.

But what about the quests, I hear you say? Surely it shouldn’t be hard to tell a story if there is an actual story written into the game? Well, that’s what you’d think, but it turns out the quests are just as poorly balanced as the rest of it. It’s the same problem Endless Legend had, where quest objectives are spawned on the other side of the map past three hostile empires and you just don’t bother with them — and when you do your reward is 50 of a resource that you’ve already stockpiled 1000 of, or 1000 Dust at a time when your empire is generating 50k per turn, or a shield module for your ships that’s somehow worse than the bog-standard tier 1 shield you researched 70 turns ago. Even within individual quest chains there’s extremely poor balance: one example for the United Empire asks you to pass 4 Industrialist laws, which requires you to tech down to tier 4 of the Empire category of technologies to get the required number of law slots. This takes about 50 turns of buildup, and then the next three stages of the quest have laughable requirements like “Have a colony that generates 200 Science per turn” when the techs required to fulfil the previous condition collectively cost a total of 100,000 Science. It’s really uneven and incoherent, and that’s the last thing any kind of narrative should be.


It’s not all bad news, though. In fact Endless Space 2 is far from it. Despite all of that bitching, and despite my fundamental feeling that this is more a game about how thoroughly you can break your economy than it is building a space empire, there’s a lot of cool systems in Endless Space 2. When I’m thinking of actual individual parts of the game such as the space battles I have nothing but good thoughts; there’s a lot of interesting mechanics to get your teeth into, while most of the boring faff that infests 4Xes has been shooed away. Take planetary invasions, which I mentioned earlier; something like Stellaris has you actively recruiting armies from your population, loading them onto troop transports, bussing them over to the target system, unloading them so that they can fight the invasion, and then loading the survivors back onto the transport after the planet has been pacified so that they can hike to the next one. That’s a lot of incredibly tedious micromanagement that Endless Space 2 has simply abstracted away with its Manpower mechanic. A certain percentage of your Food output is converted to Manpower each turn; this fills a Manpower pool up to a cap that you can raise with techs. Manpower is spent in automatically filling up planet garrisons and — this is the important part — in building all of your ships, since every ship your build now has a default complement of troops, and it’s these that are used for planetary invasions. Once your battle fleet has disposed of any opposition in space, it can then immediately drop its troop complement onto the planet surface to duke it out with the planetary garrison: it’s your fleet Manpower strength (which can be boosted by having more ships or ships with special troop modules) versus their Planetary garrison Manpower strength (which can be boosted with improvements). There is nuance to be found in invasions, but it’s found either on the macro empire level (where you can set your empire-wide troop balance between Infantry, Armour and Air units) or the broad battle tactics you and your opponent use for the invasion. Troop management itself has been removed from the equation entirely; once an invasion has been won or lost you simply pull your fleet back to a friendly system for a turn to replenish any spent Manpower strength.

Trade networks are another area where Endless Space meets with success through abstraction, although it’s a more qualified example than planetary invasions since they’re also the primary reason why Dust generation is so broken right now. At some point in the early midgame you get a tech that lets you build a Trading Company HQ in one of your systems. Once the HQ is built, you can then build a Trading Company subsidiary in another one of your systems. When both the HQ and the subsidiary are built, trade will automatically start between the two with no further input required from you; this brings in both Dust and a smaller amount of Science, but it also does the following:

  • Starts levelling up the trading company whose HQ you just built. The more levels it has, the better it is at making money.
  • Starts filling up an unlock bar for your second trading company. Once it’s full you can build another HQ, and another subsidiary. However, once both of these have been built Trading Company HQ 1 will be able to trade with both subsidiary 1 and subsidiary 2, and the same is true with HQ 2.

You can have up to 5 Trading Company HQs in your empire, and the implications of this should be clear when taken with the second point above: the base number of trade routes you can support is the square of the total HQ + subsidiary pairs that you have, making it a system with a literally exponential increase in effectiveness. There are additional system improvements you can build that increase a system’s Trade Value, and you can also sign Trade Agreements with other empires that give you access to their Trading Subsidiaries (and vice versa), and if you stack all of this stuff you can make your trade network an incredibly broken thing; it’s not that it generates more Dust than you can spend as there’s a hidden inflation mechanic that increases Dust buyout prices along with your income, but the problem with it is that an endgame trade network will be generating nearly twice the amount of Science as all of your colonies put together. It’ll need a couple of balancing passes to fix it up so that it’s a bonus worth going for rather than the driving force behind an endgame empire’s research, but I really like the base mechanic since at no point does Endless Space 2 have me manually faffing around with individual trade route destinations (Firaxis should really be taking notes here); it just wants me to set my hub systems and my spoke systems and then does the rest for me.


Wherever you look in Endless Space 2 it’s pretty much the same story. Very few of the mechanics present are new or innovative, but they’ve been built in a way that makes them far more pleasant to deal with than previous examples I’ve encountered. An interesting thing about Endless Space 2 is that Amplitude have clearly looked at Stellaris, swiped a couple of the mechanics (exploration and population politics) and implemented them in a much more 4X-y way – and they work a lot better for it. Exploration is based around sending scout ships to explore curiosity sites with probes; it’s a one-click process that’ll yield anything from special ship modules to planet anomalies to triggering new questlines, and while it’s less involved than Stellaris’s elaborate event chains it’s a much nicer system mechanically. Politics are a little more impenetrable — I’m not sure where the actual decision space around the political system lies beyond “Push button to influence the election to get the party you want elected”, but then I couldn’t see any depth in Stellaris’s far more detailed system either and here it’s at least a nice bit of flavour with a moderate impact on the game, as each party has its own set of laws that you can enact to provide empire-wide bonuses. The interactivity has been stripped out of space battles entirely but the depth of the mechanics has nevertheless been widened slightly – the outcome is now 100% dependent on your ship configuration and the tactic card you choose going in, which influences the range your ships will engage at. This makes ship design somewhat more thoughtful as it’s here that your battles are now won or lost, as you build ships that can make use of particular tactics cards to counter a known enemy ship design. Diplomacy has some weaknesses — I can never tell exactly what is causing an empire to hate me — but they at least give you plenty of warning before declaring war, and there’s a pseudo-warscore system in place that’ll let you — or them — dictate the terms of any peace. It certainly works a damn sight better than certain recent examples to be found in games whose names rhyme with “Spivilization”.

There’s a lot to like in Endless Space 2. Notably it’s the first 4X I’ve played in a while that I’ve had genuine fun playing — not qualified fun, not fun despite bugs or poor balance or a psychotic, broken AI, but actual moments of unalloyed delight. However, this is because I’m of a certain mindset; I enjoyed pulling apart Endless Space 2’s changes and additions and seeing what made it tick as there’s an above-average quantity of excellent design in here. The flipside is that that design is paired with some absolutely awful narrative and balance problems that cripple any long-term pull the game might have (as the current winning strategy is to build a trading network — not to optimise it, but simply to have it), and while Amplitude will probably fix them in the near future I’m getting a little tired of giving 4X games a pass on balance issues that are bloody obvious for anyone willing to spend a weekend playing the thing; since this is a genre that’s supposed to support hundreds of hours of repeated playthroughs that’s a shortcoming that I’m no longer inclined to ignore. So nice try, Amplitude; I liked this one better than your previous effort and I feel much more open to revisiting it in the future than I do Stellaris or Civ VI, but you’ve still got a long way to go before you properly nail it.

  1. And this is the only time you will hear me namedrop it positively, as it doesn’t look like Firaxis have fixed any of the broken shit I complained about eight months ago.
Tagged , ,

3 thoughts on “Thoughts: Endless Space 2

  1. ilitarist says:

    Good analysis, even though I do not quite agree with your conclusions. I like this exponential growth of Endless games. It feels right, you really have new problems and play differently: in the beginning your main problem is getting resources, by the end you are looking for a way to properly use all of this abundance to get ahead of contenders with the same problem. I’d like Civ games to do that to portray human civilizaton development trends like finding stuff to do for people after you hit a revolution in food/stuff production and suddenly don’t need most of society to work on sustaining itself. Yes, there’s something mathemctical and cheap optimization games. This satisfaction is close to playing a grindy RPG where all you do is getting stronger, but in a meaningful way: you don’t just get new levels to become able of defeating stronger enemies, but instead you encounter new challenges, understand what perfect party would be able to overcome those challenges and go grind to change your party into something able of doing that; with perfect play in theory you wouldn’t have to grind and would be able to predict all challenges creating a perfect party. Endless Space 2 feels like a strife for finding that perfect endgame configuration and getting there with what you have which is reinforced by inertia-driven political system. All the other 4X games to me feel like a bad grindy RPG: there’s really only one model of “right” endgame empire, it can do anything and it slightly adjusts its power throwing it into specific victory condition.

    Civilization feels to me like an exercise in speedrunning through the same steps to the perfect empire while ES2 has very different starting conditions and very open-ended development – you always have to strife for resources but you can do very different things with those resources.

    Trade companies may somewhat be justified by the fact they make you very dependent on integrity of your empire – so whole is more than sum of its parts, losing just one system with trade company/subsidary will make you disproportionally weaker, but it doesn’t feel like there’s enough risk there. Maybe it’d work better if you’d have some sort of galactic economic crises with trade companies not working at all for several turns every N turns so that you can not depend on them. Anyway, yes, economic victory is very broken. The could at least add some additional condition besides reaching some number, like “and generate 40% of galactic Dust production”. The sad thing Amplitude is very deceptive about their policy of listening to people. After couple of games of Endless Legend I had a list of roughly dozen obvious seemingly easy to fix issues, mostly with UI. I posted it on their forums couple of times, was ignored. Only one of those problems (anomalies no being really visible in schematic view out due to the icon being 3×3 pixel size or something) was fixed after developers were asked to fix it on some podcast.

    I only played 1.5 games and the second one is stuck due to infinite AI thinking bug (now supposedly fixed) but I don’t remember feeling such joy playing 4X since Civilization 4, I think. If you took grand strategy genre as a whole Europa Universalis 4 captured me similarly. Usually I see that game is good, rush through it and forget it. ES2 I want to play. A rare occasion!

  2. Darren says:

    I’ve been playing Endless Legend lately to stave off buying this until it goes on sale, and I think that the Endless series has a couple of big flaws.

    One is that it’s not nearly as customizable as it thinks it is. You *can* go hog wild on the tech tree, but there tend to be too many truly essential techs that you can’t ignore, which funnels you toward the same playstyles regardless of faction.

    Then there’s the fact that unique faction features don’t always have a compensating element. Endless Legends’ Forgotten need tons of heroes to act as spies, but they don’t have any Dust bonuses to pay for them.

    And it’s not like spies matter much, because Endless Legend is full of systems that the AI can’t work with well, resulting in very passive games where little happens unless you force it to, and that requires massive FIDSI output, as the only thing the AI seems good with is stonewalling you in diplomacy.

    So I like the series, but I don’t love it. But they get better game by game, and I do love their commitment to strong aesthetics and personality.

Leave a Reply