I’ve started working, if that is the right word to call it, on what would be called The Great Material Continuum (TGMC), a long time ago. It might surprise those that don’t come from the game industry (and hardly surprise those that are) just how much preparation work can go into game design. In term of actual ‘work’ there is often little to show, especially if this is a volunteer project where you are the only designer and don’t need to keep any detailed design doc. You just keep footnotes and little pieces of documents and numbers and excel sheets that, somehow, as if by the miracle of science, come together in the last minute.
Of course, they didn’t QUITE come together that rapidly this time – you might have noticed those ‘short’ little months where we tried to catch up and release TGMC about Paradox released 2.2.
Still, the foundation work started all the way back last Summer, once Paradox started talking about their new economic changes. Two things were clear very quickly – one, this is going to be a far bigger change than people expect. Secondly, the potential performance pitfalls the new 2.2 approach might have.
When 2.2 was under work, the hope was that the change to pops, the switch from tiles to jobs, will result in a magical performance boost. While 2.2.6 performance is now satisfactory and reaches about the level of speed the game had before 2.2, we knew that pops will remain the major performance impact of the game (and they still are. According to our tests, they are responsible to about 60-70% of the calculation cycles towards the end-game).
For that reason, we’ve made the decision very early on that we’re going to cut the number of pops in the game, and especially reduce pop caps and pop growths. This has proven doubly true once PDX started talking about some of Le Guin and Megacorp features, such as the ability to build an ecumenopolis.
Of course, we couldn’t just go and arbitrary divide the number of pops in the game by half, or something. It just won’t work properly. You have too many sensitive mechanics such as immigration, bombardments, factions, purges, rebellions and slavery. So our approach was to slightly reduce pop counts, but to mainly reduce pop growth and make the pop caps be far harsher and stricter.
For the most part, this didn’t just boost our end-game performance tremendously compared to vanilla. But it also works thematically for Star Trek in general, and New Horizons in particular.
Star Trek has never been about giant city worlds. For the most part, even the busiest planets in the galaxy are not described as having multitudes upon multitudes, and everywhere, even more fascist and autocratic states, seem to be living in rather high standards of welfare.
Not only that, it was always quite clear about the significant difference about the almost paradise like qualities of Earth compared to the fringes of the Federation and its various colonies. No colony has ever been described as having a major population. Empires in Star Trek tend to be one large home world, where the absolute majority of the population lives, surrounded by smaller colonies. That’s why the loss of Remus and Romulus, despite the huge scale of the Romulan Star Empire that spanned thousands of solar systems, was so devastating.
This was also why any addition of a member world to the Federation is so meaningful, and why the conquest of home worlds such as Betazed during the Dominion War was so impactful to the Federation.
So the slower pop growth also complemented the Star Trek theme and our desired narrative, putting more emphasis on the home worlds – that become very juicy targets for conquest, assimilation of absorption. It also complements the slower and more methodological gameplay of STNH – the more intricate and careful tech tree and ship design, the smaller fleets and slower fleet movement, forcing players to make more strategic than tactical decisions.
Reducing pop growth also reduced the capacity of the game to support playing a ‘tall strategy’. While nothing if wrong per say about a ‘tall strategy’, and in fact its often my preferred gameplay style in many games, it does goes against the modus operandi of most Star Trek empires, where the strongest empires always tend to be quite sprawling in size as well. Not to mention, the logical path of ‘playing tall’ in Stellaris 2.2 would mean we would need to add something similar to an ecumenopolis as its natural end-game accomplishment, which, again, is quite contrary to the vision and direction of Star Trek.
Of course, the slower pop growth also introduced a range of new problems to the game…. but that would wait for the next post.
The line must be drawn here! This far, and no further! – Captain Jean-Luc Picard