Pops - Why the opposition?

[kpaw]

Level of abstraction is a very important design choice. Once you pick how much you want abstracted in the game, you want interacting mechanics to operate at similar levels of abstraction so that the outcomes make sense (especially in a causal sense, but ideally also fitting in with a historical theme in history-based games). You don't want crazy micro interactions in one facet and then close to no control in another facet, unless the latter doesn't influence the game much.

Could EU 4 benefit from being more granular? Possibly, but I caution against it since the game's time period (1444-1821) had some pretty massive differences in how armies operated (both in direct combat and logistically), how common people think, and in a few cases the way countries were run. HOI 4's extra granularity makes sense; its unit of time is an hour rather than a day, and the time period in question is ~12 years. The result is that HOI 4's total time units/game is shorter, despite those time units being hours rather than days! Managing stuff that happens on the time scale of days-months makes a lot more sense than doing it in EU 4 as a result.

Yes that's a fair point, and it wouldn't be easy to model such complexity. I certainly don't know exactly how it could be achieved. At the moment, though, it is pretty extreme towards the 'all or nothing' end of the scale, and I do believe that bringing in a little more granularity would be a net positive.

I would say the core of EU4 is about using your resources efficiently. The question is whether the player is making meaningful choices. In many areas that is successfully achieved by distilling several disparate concepts into the mana system, leaving it up to you how best to spend that. The problem for me is that the rather binary approach to things like culture and religion mean that you aren't making interesting choices when it comes to these hugely important elements of governance.

Take an example from my recent Ethiopia run. I naturally chose religious ideas, which paired nicely with the 1.5% Coptic missionary strength and my national +1 missionary. Deus Vult is also extremely powerful. I could have gone humanist, but that would have been a comparatively terrible choice, so it didn't feel like a real decision based on pros and cons. Once I had nailed my colours to that mast, all I did was click 'send missionary' over and over again upon conquering new land. Often I hired an inquisitor, and occasionally I used the enforce religious unity edict. What should be one of, if not the major driving force in my society during this period, reduced to mindlessly clicking on a list every few minutes.

I know that sounds a bit negative, and I want to stress that I am absolutely not shitting on the game. I love it. But I maintain the belief that introducing at least some granularity in representing populations would be a good thing that allows for more meaningful interaction and decision making.

 

[alexti]

That's a fair point, if you consider the gameplay in EU4 micro. (braces for backlash) I do not. It's just the strategic choices of a GSG. *Every* strategy game has menus, and clicking, and choices, but that doesn't make it micro-management.

I consider "micro" to be managing details that are not strategic in nature.

I was agreeing with your post until this point. But what is strategic in managing carpet sieging and ferrying troops to avoid crossing month boundaries in transit? To me this is the worst kind of micro. It involves completely trivial decisions and a lot of clicking.

 

[alexti]

A "simulation" (aka game rules) is an AI -- only it has different objectives then "agent" AI.

That's a new definition of AI. Calling the scripts that control non-player nations in EU4 is already a stretch of a definition. But calling something like pop_size_next = pop_size*1.005 an AI?

Where else simulations are called "AI"?

AI might do some simulations internally (like bridge AI does), but that doesn't mean simulation is an AI.

In any case, in the context of the discussion, regardless how we call it, Paradox can implement game rules of simulation kind and they have already proven it. It's more of a real AI that would control nation with some strategy in mind that is a challenge, but it is not needed for the simulation.

 

[SolSys]

That's a new definition of AI.

I have a preference to the textbook definition.

 

[Spartakus]

That's a new definition of AI. Calling the scripts that control non-player nations in EU4 is already a stretch of a definition. But calling something like pop_size_next = pop_size*1.005 an AI?

Where else simulations are called "AI"?

AI might do some simulations internally (like bridge AI does), but that doesn't mean simulation is an AI.

In any case, in the context of the discussion, regardless how we call it, Paradox can implement game rules of simulation kind and they have already proven it. It's more of a real AI that would control nation with some strategy in mind that is a challenge, but it is not needed for the simulation.

Correct, also this challenge is what I referred to when I started this derail by pointing at PDX's track record. It's not that Pops are inherently problematic for an AI, you can write an AI that works with pops just like for any other system. It's more that pops suggest a design flexibility that they don't really have. Pops are incredibly easy to fiddle with with huge impacts on the economy, precisely because they are at the bottom of the simulation part. And that future DLCs will do exactly that is a givin IMHO, thing is, you need to adjust the AI everytime you do or - if you do huge redesigns as Stellaris did - completely start anew.

I have a preference to the textbook definition.

Definition of ARTIFICIAL INTELLIGENCE

a branch of computer science dealing with the simulation of intelligent behavior in computers; the capability of a machine to imitate intelligent human behavior… See the full definition

www.merriam-webster.com

A simulation is not an AI.

 

[zorkman]

There are arguments for both sides, & I do like POPs, particulalry the ones in V2. I also like character interaction & think there should be more in EU5, as currently even your rulers are bland & boring. At the end of the day, I think all sides are looking for more out of the game than before, particulalry strategy & gameplay, as besides war EU games are quite lacking, & whereas that wasn't a problem in the past, it is now.

 

[alexti]

I have a preference to the textbook definition.

Any specific textbook? There are different flavors of AI definitions, but I find it very unlikely to find any textbook definition that would match your earlier post.

 

[alexti]

Correct, also this challenge is what I referred to when I started this derail by pointing at PDX's track record. It's not that Pops are inherently problematic for an AI, you can write an AI that works with pops just like for any other system. It's more that pops suggest a design flexibility that they don't really have. Pops are incredibly easy to fiddle with with huge impacts on the economy, precisely because they are at the bottom of the simulation part. And that future DLCs will do exactly that is a givin IMHO, thing is, you need to adjust the AI everytime you do or - if you do huge redesigns as Stellaris did - completely start anew.

I think the problem with AI in the current generation of games is that it's emulated by scripts that have various tuning parameters that are optimized for particular version, so every time new patch changes something in the balance those parameters (and possibly scripts) need to be adjusted, so updates keep breaking AI in the game. This doesn't mean that the same approach has to be taken in the new game. A lot of research, tools and libraries have appeared since EU4 has started, so if AI is given sufficient prioritization in the game development we may see much more adaptable (and thus much less fragile) AI that can evaluate game rules and find out how to use available tools to achieve its goals.

 

[Sergeant Major Gross]

People that are drooling about Vicky's pop should (and I know they do) hope for a Vicky 3 and leave EU4 alone.

This.

 

[dessoul]

My 5 cents on this topic are: i would doubt, that paradox could do it. They have proven at stellaris, that they can only fail at such a topic. Play stellaris, and you run into some stuttering lag-hell in mid-late game. That is my main reason, why i am not playing this game anymore. (Even as it counts as one of the best game in this area, but lets take this aside.)
I just checked the population numbers in Victoria 2: they are terrible inaccuate. But only then was it possible for this game to function properly. And if i try to imagine some pop-installment in a game like EU4 (or EU5): No thanks. It would not work. The game has plenty of issues. Lets not add another game-breaking topic to it.

 

[Red Death]

I'm wondering how many of the people complaining about micro-heavy Imperator have actually played or taken a look at it recently. Really the most you need to do is for example check the demographic size and make-up of newly conquered regions to see how valuable it is/whether they're going to cause problems. Besides that you can look at the statistics in specific regions or the overall country to decide what macro-level policies are appropriate, and which buildings should be built where. There really isn't much micro involved with the exception of:

Yes. You don't *have* to manage them, but the problem with this, is that not moving slaves & not making sure to hit production # quotas for whole integer increases in trade goods is significantly sub-optimal play. Now, I totally get that there are a lot of people that don't care about ignoring available mechanics in order to avoid game aspects they don't like, but I am not one of them.

If a using a provided game system (e.g. manual pop movement) allows my empire to better succeed against my rivals, then I'm going to use that provided mechanic . . . unless it's such a pain the a** that I just end up not playing the game.

Having to do heavy micro on pops for optimal play sucks, and if the micro is available, it's *going* to be optimal, because no auto scripting (in the near to foreseeable future) ever match player cognition and strategic capabilities.

If the options are: 1) micro, or 2) sub-optimal play, then I chose *neither* and will instead go play a game that isn't an annoying mental grind/chore.

Manually moving slaves around to hit whole integer thresholds for whole integer trade good production is a serious design problem for the reasons you mentioned (I especially agree with your statement on the fact that it being the optimal solution means the game is encouraging unfun micro).

It is however, the sole exception in the system, and a problem that is unique to imperators' handling of trade goods. It's not exactly an argument against the system as a whole, especially in this context, as an EU game has no need for this kind of trade good system.

 

[Sergeant Major Gross]

I don't want increasing lag mid to late game like Stellaris.

 

[alexti]

My 5 cents on this topic are: i would doubt, that paradox could do it. They have proven at stellaris, that they can only fail at such a topic. Play stellaris, and you run into some stuttering lag-hell in mid-late game.

I am making opposite conclusion from it. Now Paradox knows how poorly that implementation of pops performs so they are unlikely to repeat the same error.

 

[Xdevo]

I am making opposite conclusion from it. Now Paradox knows how poorly that implementation of pops performs so they are unlikely to repeat the same error.

That assumes that Paradox doesn't have a pattern of making the same mistakes over and over and over and over again.

 

[JagLover]

I don't have a great deal of experience with Vic 2, but I have put a lot of hours into EU Iv and what I think it most lacks, IMO, is dynamic development (including population growth). You could call it "pops" or just organic development and changes as was seen historically. I don't think anything beyond total population, culture and religion split, is really required from a "pops" perspective in the context of the game.

Far more interesting to me would be the ability of organic population growth and decline to take place, particularly in the later game period, where it really was a thing. This would then mean that more realistic "nation building" strategies were appropriate. With the devastation inflicted by war having the potential to set back population totals by decades if not a century or more.

State controlled development still has its place and occurred, but what the game lacks IMO is the dynamism of natural growth or decline and this is something I would ike to see in EU V to set it (potentially) above EU IV.

 

[dessoul]

I am making opposite conclusion from it. Now Paradox knows how poorly that implementation of pops performs so they are unlikely to repeat the same error.

In a perfekt world this would be maybe the case. Only bad, that this does not work that way. I was assuming, that it is because there are different people working on different projects. And person B tends to make the same mistake as person A during the process.

I saw this at "Imperator: Rome", when i was still playing it and following it shortly after release. What i had to see, that they were making the same mistakes as in EU4. And i was sitting there and was asking myself, if the people working there are even aware of the same games in their company. At some point the things which annoyed me in I:R piled up so much, that i lost any fun at the game and i stopped playing it completely.

So: even as they have made mistakes, it does not mean, that they are learning from it.

 

[alexti]

In a perfekt world this would be maybe the case. Only bad, that this does not work that way. I was assuming, that it is because there are different people working on different projects. And person B tends to make the same mistake as person A during the process.

It can happen, but usually functional companies do share knowledge and despite all criticism Paradox delivers working products of a reasonable quality, so I think they have general awareness of at least big mistakes.

I saw this at "Imperator: Rome", when i was still playing it and following it shortly after release. What i had to see, that they were making the same mistakes as in EU4. And i was sitting there and was asking myself, if the people working there are even aware of the same games in their company. At some point the things which annoyed me in I:R piled up so much, that i lost any fun at the game and i stopped playing it completely.

So: even as they have made mistakes, it does not mean, that they are learning from it.

I don't think they knew those were mistakes when they've developed I:R. EU4 was a very successful game and it was probably hard to tell that it was successful despite certain mechanics rather than because of them. When this became clear after I:R release, they've started to work on changing I:R rather quickly. Now it's fairly good.

 

[Zhetone]

yeah I think it's difficult for Paradox to see what is popular and what isn't. They've said before that evaluating what works based on forum posts and steam reviews makes no sense when everything still sells, and it could always be a case of loud forumites who are a minority vs. a fairly silent bulk of the actual player base. Same goes for steam reviews. They might plummet and say the game is terrible, but then it continues to sell and have a large active group if you look at players on steam. So how do they know what is good and what is bad?

With Imperator I think it was fairly obvious because everything converged. Forums said it was bad, reviews said it was bad, player base dropped like a boulder almost immediately. I don't think it was at all a hard decision to change its direction with those circumstances. But with EU4, we always see the cycle where a DLC comes out, the forums and steam reviews complain a lot, but more people are playing than ever.

 

[love sweden]

yeah I think it's difficult for Paradox to see what is popular and what isn't. They've said before that evaluating what works based on forum posts and steam reviews makes no sense when everything still sells, and it could always be a case of loud forumites who are a minority vs. a fairly silent bulk of the actual player base. Same goes for steam reviews. They might plummet and say the game is terrible, but then it continues to sell and have a large active group if you look at players on steam. So how do they know what is good and what is bad?

With Imperator I think it was fairly obvious because everything converged. Forums said it was bad, reviews said it was bad, player base dropped like a boulder almost immediately. I don't think it was at all a hard decision to change its direction with those circumstances. But with EU4, we always see the cycle where a DLC comes out, the forums and steam reviews complain a lot, but more people are playing than ever.

The rewiev thing doesnt make sense since if nobody bought the game there wouldnt be any bad reviews since no one ever played it to find out.

 

[Vulkandrache]

yeah I think it's difficult for Paradox to see what is popular and what isn't

Would it not be one of the primary functions of the director of development to foresee such trends forming before they become a problem?

EU4 had years of staunch support despite its many, many flaws and bugs.
With enthusiastic voices en mass on each DD.
And then you had (almost over night) hundreds of downvotes with the tone on the forum taking a dive.
Suddenly critic that was subdued and largly limited to the bug forum kept being brought forth.
Surely this would have provoked some reaktion so see what is going on?

And even if such a thing is limited to only one of 4 similar games:
You'd have regular meeting between the Dev directors (say once a month), would you not?
To talk about such things.
Or am i romanticising how i thought a company with more than a handful of employees is run?