is there a way to gauge how much a performance impact a mod has?

  • 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.
Open up task manager and look at your system resources whilst the game is running.
  • Is your GPU maxed out? (I doubt it the game isn't graphically heavy)
  • Or are any of your CPU threads maxed out? (Don't look at the overall CPU measure as Stellaris struggles with good multi-threading)
Thatll tell you what's bottlenecking your game - processor or graphics.
  • If it's the latter, graphics, list mods by filesize and disable the largest ones [steam tells you the filesize in workshop - some texture mods run to over a GB each, also adding to initial boot times, as they need to be decompressed by the game]
    • (I have noticed a marginal increase in speeds in past patches by disabling large texture mods, despite my GPU not being under stress, so this is likely to do with some engine interaction with RAM)
  • If it's the former, you have to basically look at each mod, the ones that add more events with periodic (monthly/biannual) checks will be the most CPU intensive. Or mods that add more traits (or planet types - E.g. Planetary Diversity) will also add to processing times as each check will have to cycle through more types of traits, habitability types etc, than if you just had vanilla ones enabled (though this example would probably be a minor-to-negligible performance hit).
    • If you've ever used complex macros in excel and seen it chug for a bit on calculation updates, this is basically what the game is doing, it's a big fat spreadsheet running in memory. If you add more crap to the list for it to search, it'll take longer.
  • There might also be a bugged event in one of your mods slowing the game down, as you only have 13, you could check each recent mod-comment section on steam to see if anyone else has had performance issues recently.
    • But things like extra ship types or more ship components shouldn't be an issue to performance, for example, as they all build off hard-coded, existing ship modifiers.
Ultimately the largest hit to performance is still the size of your galaxy*number of hyperplane connections - scaling with the number of fleets in-play that are not docked somewhere (I.e. are actively moving about).
  • Because pathing for fleets is essentially a constrained graph theory problem, wiki example, so the larger the hyperplane network is, the more time the CPU needs to dedicate to finding a path from A-B and the more fleets there are that need to move about, the more time it spends bogged down in pathing maths, too.
  • This is also why you'll see fat lags at the start of big wars - all those fleets are processing their initial move orders.
 
Last edited:
  • 1
Reactions: