POPs: development as RAM size increases

This is for, what the discord lovingly calls, 'top hat beaver' content and ideas. Mechanics that will be used to flesh out the game over the absolute necessary mechanics of the core concepts. Eventually, as the development of the SOTE project continues, things in this forum may move to the 'Core Concepts' forums as things are completed and what were once future ideas become pressing and more important to develop
Post Reply
User avatar
David Buunk
Posts: 75
Joined: Thu Jun 22, 2017 4:46 pm
Location: Netherlands

POPs: development as RAM size increases

Post by David Buunk » Thu Aug 03, 2017 7:55 pm

The current idea about POPs is as follows. Information on the number of people in a certain POP type (farmers, craftsmen, and the like) would be stored at the tile level, while the bulk of data would be stored at the province level. (Think of cultural/technological data, goods/trade, and the like.) Call this POPs generation 1.

As RAM size increases, we move, piece by piece the province-level data to the tiles. In the end, provinces influence the calculations far less, and exist mostly for helping the human visualise things. The simulation code will be a bit simpler, quite a bit more accurate, but at the cost of more memory and execution time. Call this generation 2.

I assume that apart from the collective POPs, SotE will also come to represent various individuals, rulers, heirs, ministers, generals, governors, and leading nobles and merchants. Their numbers, I assume, will increase throughout the development, to the point that a larger part of the upper class is also individually represented.

Now, in what I assume will be 12-15 years from now, we drop the concept of POP with the upper classes, and represent all of their members as individuals. As time goes on, and more memory and processing power becomes available, we will convert more and more POP types to an individual-based model. In the end, we will reach generation 3, where POPs are gone, and the simulation will deal with individuals only. This is again simpler to make a model for, much more accurate, but will take mindbogglingly large amounts of RAM.

Now, this all assumes that we use the memory/processing power increases on improving POP simulation. There are other uses as well, we might introduce a variety of animals into the sim, or we might shrink the tile size. Whatever is most desirable will become clear in due time.
Programming SotE.

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest