Performance

[MinhowMinhow]

My biggest concern about this game is performance, considering how other games are doing.

Stellaris, HoI4 have lots of performanxe issues. Stellaris late game become almost unplayable, due to lots of pops. HoI4 is laggy too, specially with the high number of units in game, mods like Millenium Dawn are problematic, because there are a lots of tags. Basically these are the problems that Victoria 3 will have, with lots of pops, units and countries.

Imperator was unplayble, at least for me, on my old laptop (ok, is a potato ), but shows how intensive are these new generation pdx games. And I:R can be pretty laggy sometimes, even on my new PC. But one of games that I played a lot on my old laptop was Victoria 2, and even on late game was cool.

So, performance wise, will this game be an upgrade on the issue, or will be another laggy game?

 

[elandread]

My biggest concern about this game is performance, considering how other games are doing.

Stellaris, HoI4 have lots of performanxe issues. Stellaris late game become almost unplayable, due to lots of pops. HoI4 is laggy too, specially with the high number of units in game, mods like Millenium Dawn are problematic, because there are a lots of tags. Basically these are the problems that Victoria 3 will have, with lots of pops, units and countries.

Imperator was unplayble, at least for me, on my old laptop (ok, is a potato ), but shows how intensive are these new generation pdx games. And I:R can be pretty laggy sometimes. But one of games that I played a lot on my old laptop was Victoria 2, and even on late game was cool.

So, performance wise, will this game be an upgrade on the issue, or will be another laggy game?

I doubt that Vic3 will have as severe performance problems as late-game Stellaris does, since the growth of the number of pops is not nearly as large (since in Stellaris most of the galaxy is unsettled at game start, which is not the case for Vic3). Whether there will be performance problems similar to the ones HoI4 has depends on how they implement warfare, and we know nothing about that at the moment.

 

[InvisibleBison]

I never understand why people who admit that they're playing on a computer that isn't powerful enough to run a game complain about the game having poor performance.

 

[MinhowMinhow]

I never understand why people who admit that they're playing on a computer that isn't powerful enough to run a game complain about the game having poor performance.

I forgot to say, I changed my computer, but even having a good CPU, some games like Hoi4 and Stellaris lags.

 

[rafsq]

Given CK3's performance, I assume Vic3 will be similar.

 

[C.D. Howe]

I forgot to say, I changed my computer, but even having a good CPU, some games like Hoi4 and Stellaris lags.

I have a very good CPU but my GPU is below recommendations and causes my games to lag as well. I've been meaning to try GeForce Now rather than buy a new computer to deal with this issue. What is weird is that CK3 runs pretty well for me, but Imperator and especially Stellaris struggle.

 

[nuarbnellaffej]

I don’t know a whole lot about video game design, but hopefully paradox is getting better at multi threading so their games can leverage the higher core counts of newer CPUs. Maybe they already have.

Ive got a 6 core, 10600k CPU, and a 1070 gpu and can’t say that I’ve had any issues with HOI4 or CK3.

 

[Leoreth]

Please don't use mods as a yardstick for performance. Modders can always make unwise decisions about how many entities they put into the game (and certain mod philosophies will lean in the direction of doing that as their main purpose). No engine can support any arbitrary number of game entities.

 

[MinhowMinhow]

Given CK3's performance, I assume Vic3 will be similar.

I never played CK3, is this a good thing?

 

[rafsq]

I never played CK3, is this a good thing?

Yes, CK3 runs remarkably well CPU-wise.

 

[MinhowMinhow]

Yes, CK3 runs remarkably well CPU-wise.

Good to know.

 

[EntropyAvatar]

Stellaris late game become almost unplayable, due to lots of pops.

I think it's a reasonable concern, since V3 will start with tens of thousands of pops, and that's what causes giant performance problems late-game in Stellaris. On top of that, there are other economic actors buying and selling, such as factories, farms, etc. If it is not designed well, the CPU load could be crushing.

I am hopeful that it is designed well, though. The code for how pops interact with the economy was retrofitted into Stellaris and frankly wastes a huge amount of computation. Moreover, the wasted computation is not obvious until you get deep into the game - things that seem reasonable on one planet with 20 pops can get very unreasonable as the galaxy fill up. In Vic3, the designers went into it knowing they could have very significant performance problems with all the entities they were trying to model, so they could design the system to be as efficient as possible.

 

[Leoreth]

I am also under the impression that the problem Stellaris runs into (never played since it introduced pops) is that every pop effectively interacts with everything and that there are no concepts that help limit the scale of interaction. It is easy to see how that exponentially grows the computational load as more pops come into existence in the late game.

What we know about Vic3 indicates that some interactions only happen on a state and market scale, so it's not like everything influences everything. Although their commentary about how markets merge into larger markets over the course of the game might still create problems later on.

Also, correct me if I'm wrong, but doesn't every Stellaris pop represent a set amount of people, and population growth results in more pops? That's not the case in Vic3 where the number of pops doesn't change even as your population grows, and pops are only created if your populace diversifies for some reason.

 

[EntropyAvatar]

Also, correct me if I'm wrong, but doesn't every Stellaris pop represent a set amount of people, and population growth results in more pops? That's not the case in Vic3 where the number of pops doesn't change even as your population grows, and pops are only created if your populace diversifies for some reason.

Yeah I think that's an important factor. Basic population growth shouldn't change the computational load much. Though it seems as though pops are divided by their workplace as well as their profession (the pops DD lists workplace in addition to profession, culture and religion). So a proliferation in types of workplaces within a state could increase the number of pops beyond population growth.

 

[monksimus]

Yeah I think that's an important factor. Basic population growth shouldn't change the computational load much. Though it seems as though pops are divided by their workplace as well as their profession (the pops DD lists workplace in addition to profession, culture and religion). So a proliferation in types of workplaces within a state could increase the number of pops beyond population growth.

This should really help constrain things. Using V2 as a starting point, if we assume that splitting pops by workplaces increases the number of different pop groups by a factor of 10x, and it sounds like pops exist at the state level rather than the province level which decreases it by 5x, there should be something on the order of 2x the pops there were in V2. Growth is probably similar throughout the game, though industrialization adding workplaces could inflate the number of pops significantly. V2 slowed down in the endgame (I never played past 1910 due to navy control being a nightmare at that point), but that wasn't too bad. And computers are significantly more powerful now, I don't expect it to be a major issue. I just wish I could install it on my workstation with 64 GB of ram and a 10-core Xeon and let it rip.