Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
When you see the casualty report at the end of combat and not all the ships destroyed are registered as combat losses, are the hits against them included in the tally?
If you study the results carefully, there are lots of clear reasons not to trust their accuracy and completeness.
Now, the way to control for that is to set up a controlled environment and test various fleet builds in combat against each other. You can start out small and simple - a minimum size galaxy with two empires, and then just give them individual ships to throw into the killing stars against each other. You could then experimentally verify whether the fleet combat reports are accurate or not.
At some level of complexity that is going to break down.
So another way you could do it is through induction. You can calculate theoretical combats between fleets. There's a spreadsheet containing detailed weapon stats in the stellaris/common/component_templates folder. One is a .csv you can import into a spreadsheet program, and one is a .ods which can be viewed through an open source spreadsheet program like libre office. Frankly that alone should be sufficient for fleet design purposes, as you can tell quite clearly every stat off every weapon. Calculating the levels of armor and shielding would be a bit harder because the stats are buried in the codebase. But it's a fairly easy operation to achieve through the fleet designer - calculate the number of hull and armor points in, say, a titan. Then throw it into combat against a fleet of corvettes and see if the damage inflicted matches the various hit points of the craft.
If you determine the end results are more reliable than they seem to be, I'd love to hear your report.
If you want them to be reliable by fiat, I don't see how that's going to happen. They either are reliable or they are not. You could also simply take them at face value, and choose to trust them as reliable despite knowing their accuracy is doubted by many. Bearded guys in the sky aren't the only receptacles we can pour our faith into.
EDIT: Fixed a minor error in the path to the weapon_components spreadsheet (it's in "component_templates" not "component_sets", my bad
Thanks for writing all that. Your hypothetical testing sounds like an accurate method. Yet my regular questions are "this" fleet in "this" game against "that" empire, real situations in mid-to late-game, and I can't imagine duplicating that particular situation, much less in a timely manner.
Under /stellaris I don't have a "common" folder, and I did a PC search for the component_ file and nothing came up. Maybe they updated their folder system with later updates? (I happily play 3.7.4).
From your first paragraph, it seems that the casualty report might not even state accurately how many enemy ships were destroyed. Hmph. So is it best just to ignore that screen? It's a screen that pops up so often, and it could be so useful, and it sounds like it could be easily fixed; instead of "amazing" new DLC's, if the designers spent a bit of time addressing these kinds of things, that would be great.
II suspect you're looking in your saves folder instead of the game folder. Often in "Program Files" there is a folder called steam.... so the final path will often look like: C:\Program Files (x86)\Steam\steamapps\common
Then from there to Stellaris/Common/weapon_templates
What surprises me about your concern over this issue is that I thought you play pretty frequently and observantly. Have you never noticed this before? You say you're playing especially close attention to the after-action combat reports. You've never detected any discrepancies? That's not unnatural or unusual, I suppose. It just surprises me. It seems as though, by now, you'd have a clear grasp on its reliability and if anything should have observations that explain the inconsistencies others have observed.
There are lots of things the developers could do to improve the base game that they won't. The economic imperative of DLC-based development doesn't allow for it. They need to drop a splashy novelty bomb on the game once per quarter. That's it. That's what they do. The idea that is somehow "supporting" or "advancing" the game is a delusion of corporate simps.
This game got pushed and pulled in a lot of unfortunate directions, I think. They catered too hard to fascist fantasies of genocide in its early years, and now they've rebounded into development of a game for teenagers. The hard sci-fi title this game originally aspired to be recedes further and further from view with each release. Which is why I've picked the path for engagement that I'm on - I froze the game on a version that works relatively well for my needs, and am stripping it down to get at the simulator, which is kind of a marvel of ingenuity buried under the layers of poor game design choices.
I get this. I don't see how accurate after-action reports are essential to tailoring your fleet, but I don't want to judge your process or your concerns. It does sound, though, like I've shaken your confidence in a feature of the game that was essential to your enjoyment of it. And I do feel a little bad over that. I didn't mean to strip you of your joy. And I hope there's a path for you to hold onto or reacquire it.
An accurate AAR would incorporate the, so far as I can tell, relatively working breakdown that shows how effective weaponry and defences are on either side. That way, you can see clearly what is working better against your opponent - and by extension, against you by your opponent - so that you can tailor your fleets to counter them if it's going to be a protracted war.
In a lot of games where AARs of this kind are displayed after each engagement, they tend to be quite useful sources of information for the attentive player to examine and collate the data for. Not so much in Stellaris. Whether that's due to Paradox not being able to program them properly at the time or because the very old UI and code is no longer up to snuff with all the changes over the years or some other reason, that's currently the way it is.
That's what led me to the OP. And as time goes on, the discrepancies seemed to grow, but without hard numbers, it's not so easy to pin down exactly what I think I have been seeing.
It's easy to complain about details, but it's still and incredible game.
Well, I wouldn't say it's exactly essential to it, and hakuna matata: no joy has been stripped! Yet the idea is a good one - a post-battle report detailing what happened - and it would simply be nice if it did what it pretended to do.
RCMidas' entry above shows how a good report would be helpful (even stating war exhaustion: great idea, I never thought of that)
You're a smart and enjoyable contributor, thanks for your posts!
So, I presume the ship classes designed by a central planner safe on Stronghold Prime, wherever that may be. And so I design them for thematic coherence more than out of any specific need. So like, every ship size tends to get a "bruiser" that is heavily armored and loaded with short range weapons. There's usually an energy blaster and a kinetic one. Artillery, of course. A carrier if the hull type allows it, and so on and so forth. I construct mixed fleets with mixed capabilities.
Then when war strikes, I let attrition winnow out the losers in the fleet. Usually the bruisers are the first to go, and in large numbers. But they're not replaced 1:1, they're replaced in the same balance as the original production. So let's say a hypotehtical fleet has 25 ships of 5 designs, 5 of each design. Then war comes, it loses 4 brawlers and 1 point-defense screener. The reinforcements come out one of each of the 5 classes, meaning it now has only 2 brawlers, 5 point defense screeners, and now 6 each of the 3 vessel designs that took no casualties. In that way, my enemy basically carves the raw material of my fleet into something best adapted to kill them.
That's not even "meta", that's just common sense. Especially when the AI gets to upgrade their ships when new technologies emerge too. And even more so when you take into account that you're hoping your fleets will actually win engagements as opposed to being wiped.
The survivors won't necessarily paint an accurate picture of what happened either. Maybe you lost lots of brawlers because they charged head-in and took the brunt of the damage. However, they also DEALT the majority of the damage because their weapons were better suited to bypassing the enemy defences - meanwhile, all those artillery and carrier vessels that survived by warping out did not pull their weight, but you'll see only that they survived and therefore should prioritise building more of them.
You can absolutely build for thematic coherence, but you can still do so with a hint of actual design ability.
I understand a lot of folks need to work pretty hard to keep up with the AI. I've never really had that problem. Ninety-nine percent of combat planning in Stellaris is keeping your forces grouped and bringing more firepower to the fight than the other guy. Beyond that, I find it's just screwing around to keep yourself entertained.
What's fun about my method for certain kinds of players is it gives you responsive fleets while introducing a fairly simple selection rule whose results may not be clear in advance but nevertheless expose the contours of comparative advantage clearly.