• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

Tinto Talks #3 - March 13th, 2024

Welcome to the third week of Tinto Talks, where we talk about our upcoming game, which has the codename “Project Caesar.” Today we are going to delve into something that some may view as controversial. If we go back to one of the pillars we mentioned in the first development diary, “Believable World,” it has 4 sub pillars, where two of them are important to bring forward to today.

Population
The simulation of the population will be what everything is based upon, economy, politics, and warfare.

Simulation, not Board Game.
Mechanics should feel like they fit together, so that you feel you play in a world, and not abstracted away to give the impression of being a board game.

So what does that mean for Project Caesar?
D4RGBO3N1xr8MhsfaTGT5DNNERZhnjijvnx4KgvFi0c2ZFBuMEvrfiht3yyayH6EloTJWJNKEh1VSCH_LsaJWUASqg1j0thITZivoIM3jtOzKM-IGlJFubDx6UZP-iMTRXmnCWAVsm5uKdmQD5F77i8


Every location that can be settled on the maps can have “pops,” or as we often refer to them in Project Caesar; People. Most of the locations have people already from the start of the game. Today we talk about how people are represented in our game, and hint at a few things they will impact in the game.

A single unit of people in a single location can be any size from one to a billion as long as they share the same three attributes, culture, religion, and social class. This unit of people we tend to refer to as a pop.
  • Culture, ie, if they are Catalan, Andalusi, Swedish, or something else.
  • Religion, ie, Catholic, Lutheran, Sunni etc. Nothing new.
  • Social Class. In Project Caesar we have 5 different social classes.
    • Nobles - These are the people at the top of the pyramid.
    • Clergy - These represent priests, monks, etc.
    • Burghers - These come from the towns and cities of a country.
    • Peasants - This is the bulk of the people.
    • Slaves - Only present in countries where it is legal.

TX1paNgsYnH4SO0ZWP2NOrbtNa8O20QO9w-Ps-VwjSN8uhMZca-pxt0P2kND5gOnejQfklB6AQpb_C3XH2cB9hF_6sd6GSxbsgygmOmvnUbPCfgWS_BvIq7fPQzBYgy0mYwAccRxR-vFvYfL5jptBMs



There are a few other statistics related to a Pop, where we first have their literacy, which impacts the technological advancement of the country they belong to, and it also impacts the Pop’s understanding of their position in life.

Another one is their current satisfaction, which if it becomes too low, will cause problems for someone. Satisfaction is currently affected by the country’s religious tolerance of their religion, their cultural view of the primary culture, the status of their culture, general instability in the country, <several things we can’t talk about just yet>, and of course specially scripted circumstances.

There are also indirect values and impacts from a Pop on the military, economical and political part of the game as well, which we will go into detail in future development diaries.

Populations can grow or decline over time, assimilate to other cultures, convert to religions, or even migrate.

Most importantly here though, while population is the foundation of the game, it is a system that is in the background, and you will only have indirect control over.

What about performance then?

One of the most important aspects of this has been to design this system and code it in a way that it scales nicely over time in the game, and also has no performance impact. Of course now that we talked about how detailed our map is with currently 27,518 unique locations on the map, and with many of them having pops, you may get worried.

14 years ago, we released a game called Victoria 2, that had 1/10th of the amount of locations, but we also had far more social classes (or pop-types) as we called them there. That game also had a deep political system where each pop cared about multiple issues, and much more that we don’t do here. All in a game that for all practical purposes was basically not multi-threaded in the gamelogic, and was still running fast enough at release.

Now we are building a game based on decades of experience, and so far the performance impact of having pops is not even noticeable.


Next week, we will talk about how governments work a bit, but here is a screenshot that some may like:

1710317019801.png
 
  • 432Love
  • 170Like
  • 17
  • 13
  • 11
Reactions:
I like the granularity of having just individual pops, rather than something like Victoria, where the number of people in each pop varies too much. Is there some justification for performance, or gameplay?
Oh definitely performance. Having clones of pops that are equal in all things just adds more data entities to crunch through in loops. It's part of why Stellaris performance tanks in lategame when there's lots of pops that are just about identical.

My discussion with Nicolas above highlights the sorts of considerations devs make to ensure performance. I regarded Vic3 pops in different jobs (but otherwise identical in culture/religion/etc) to be a key detail warranting the extra quantity of data entities to track. Nicolas' approach reduces memory footprint and is essentially stating some details (like whether only a specific job type would want to migrate) aren't worth the performance (time) cost to track.

Now, that was on how pops are represented in the code. If what you meant was "can you display this number with discrete sprites so I can quickly grasp quantities instead of reading numbers", that's a separate (UI) matter.
 
  • 3
  • 1Like
Reactions:
This looks great! Thanks Johan. This looks like a real improvement to EU4’s abstracted ‘development’ which always turned me away from the game (and towards the MEIOU & Taxes mod).

Are there no plans to incorporate tribal pops? It seems like one segment of the population which really can’t be captured by the existing classes, particularly since tribal/nomadic populations are often defined by their less hierarchical / more fluid social classes (or even lack of any distinct social classes for Amerindians).
 
  • 6
  • 1Like
Reactions:
Not also that, but something like a proper 1576 would be great too (Full Spanish, Netherlands, Ottomans, Mughals, Persia etc). This makes it much more interesting to play France, England or the Netherlands, in my opinion. It is a shame that EU4 has not really expanded the later starting dates.
That was a reaction to the player market: some 95% of EU4 games started at the earliest possible date. Same with HoI 4. Maintaining many starting dates means divided attention. Sometimes it's best to pick just one thing and do it well.
 
  • 8
Reactions:
I would prefer 1411 start date.

Right before the conquest of Ceuta and the beginning of the Age of Discovery.

But if it starts mid 14th century, it means it will also finish earlier and I'm OK with that. Late 14th century. Maybe 1370
More poke poke less pew pew in the game overall.
 
  • 6
  • 2Like
Reactions:
Unless snowballing and the player loss of interest in continuing a campaign is solved, any start date is too early.
 
  • 5
  • 3
  • 1Like
  • 1Haha
  • 1
Reactions:
I love the time period of the renaissance or the League wars and would love to be immersed in the colonization of America, the 30 years war, the Franco-Habsburg rivalry, the Union of Lublin, etc. But they need to actually happen, or at least some of them, which isn't even a thing in EU4 all that often.

Thinking about that, serfdom in Poland was really enforced only during the reign of Kazimierz Jagielończyk, so if we start 100 years before, we are in a completely different socioeconomic environment. The whole "noble republic" concept was forged during his reign in the second half of the XV century, would all this be neglected?
 
  • 1
Reactions:
Most importantly here though, while population is the foundation of the game, it is a system that is in the background, and you will only have indirect control over.
Sounds frustrating.


What about performance then?

One of the most important aspects of this has been to design this system and code it in a way that it scales nicely over time in the game, and also has no performance impact.
None ... eh? Surprising.


Now we are building a game based on decades of experience, and so far the performance impact of having pops is not even noticeable.
Hmmm ... I would think the CPU consumption resulting from pops is directly proportional to how noticeable they are.
 
  • 6
  • 1
Reactions:
Dai Viet in 1350 had about 3-4 million people, why is there less than 1 million people on the map?
That's impressive. Byzantine Empire had ~2 million (yeah at that point they lost lot's of territories, but still).

Incorrect numbers can be explained by early development. Once they open "Suggestions" subforum for Project Caesar/EU5, you should post this info there together with some sources confirming these numbers so they can correct any inaccuracies.
 
  • 2
  • 1Like
Reactions:
How different are the concepts of "Believable" vs "Realistic"?

Native Americans modernizing in the face of an invasion isn't realistic, but it's believable. A Teutonic Horde isn't realistic, but it's believable. Same with norse revival. I worry of course that the first pillar is not "Fun" nor any real definable pillar with SMART definition to it. How do we quantify it, and how does it affect fun?

To compare a totally unrelated game, EU4, the best recent DLC was Lions of the North which didn't have a high degree of realism. Overlord has more realisim, but is significantly less fun.
 
  • 2
  • 1
Reactions:
My main gripe with a significantly earlier start date is mainly that the history of the era the EU-franchise plays in gets more and more watered down as important events are less likely to happen, unless heavily scripted.
And why exactly are "those" events that important in the grand scheme of things, compared to events closer to 1356?

In any case, I hope the 27th of March DD is the real controversial one, where they axe the mission trees for good.
 
  • 6
  • 1Like
  • 1
Reactions:
I'm happy that they are making official EU: Meiou and Taxes Edition, but I'm seriously worried about the performance.
I think something like high end AMD 5800X3D/7900X3D CPU will be essential. Meiou and Taxes is great content wise but it's one of the most demanding and unstable mods ever.

Also, even with 5800X3D late game (1900+) in Victoria 3 can still get very slow and it's basically unplayable on average CPUs (past 1900).
 
  • 2
  • 1
Reactions:
@MrMcQue

Sorry, my previous post is a bit on the face...

What I really meant was that, had EUIV been for years starting in 1356, you would think that the events surrounding this date would be very important, and things later would be more fuzzy.

And regarding any historical events happening after, say 1356, I'm very much in favor of them to happen... organically, based on the mechanics of the game, and not from a surimposed guideline, be it mission trees or DHE.
 
  • 5
  • 2
Reactions:
I find events boring as hell, mission trees as well, their all fun the "first" game, but otherwise they just bog down the game into specific historical outcomes or events. Basically events are dumb because by 50 years in, chances are it looks nothing like IRL, let alone 200 years in.
 
  • 7
  • 1
Reactions:
I got no clue about the code in either v3 or stellaris, nor about their performance costs of pops.

I coded the pop systems in v1, v2, eu:rome and imperator before this, and I know how those work.

This dev diary reads a lot like how I'd like to see the Stellaris pop system fixed, like if Stellaris planets amalgamated pops with same species and ethics instead of its current remnant of the old tile system.
 
  • 3
Reactions:
To add to this discussion, Johan already mentionned he'd talk about how they do flavor later... meaning there ARE plans about it. And my hunch is, having game rules, such as historical or open, the former having events meant to help the world move toward the one we know
 
  • 1
  • 1
Reactions:
And my previous point in the post you're quoting remains the same.
I think the issue is that it is then not an early modern game but a late medieval game. That could be fine, but if the purpose is to design a early modern game you want to focus on mechanics and assumptions designed for that period.

A game focused on nation states (as Project Ceasar seems to be) is not one well suited to the 1300s in which the concept of a state does not really exist. Instead the 1300s might be better modeled by a focus on dynasties and characters. But that is already very well done by CK3...
 
  • 9
  • 3
Reactions: