-
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.
Ok, 60471 pops, Game Year: 2454:
YoY: 639 secs, 53.25 per month. That's almost 80% slower compared to the previous datapoint at 51k, but the big difference here is that the war in heaven is in full swing.
In fact some empires are fighting in 3-4 wars, and the tempest coumes out once in a while as well. The crisis hasn't fired yet.
Not sure when I'll post on 70k. The time it takes is quite long to observe up to that. Also the crisis might come out and make it impossible to reach 70k. We'll see!
YoY: 639 secs, 53.25 per month. That's almost 80% slower compared to the previous datapoint at 51k, but the big difference here is that the war in heaven is in full swing.
In fact some empires are fighting in 3-4 wars, and the tempest coumes out once in a while as well. The crisis hasn't fired yet.
Not sure when I'll post on 70k. The time it takes is quite long to observe up to that. Also the crisis might come out and make it impossible to reach 70k. We'll see!
- 1
- 1
- 1
And here's the last data point: 70483 pops, Year 2514.
With the Crisis (contingency) and War in heaven finished/defeated, there is no war, and everyone is in the LNAP.
YoY: Just an Impressive 341.37 secs, Average Monthly: 28.41 secs
Obviously, fleets moving, fighting, and AI planning takes significant resources. This result is about 40%-45% faster than the previous data point. Adding to that, I noticed the AI had many small fleets and large fleet fragmentation. Not sure if this is intendend or not. Also the fleets had huge amounts of missing reinforcements, and some had reinforcements set higher that what the command limit will allow.
The fleet manager of all empires is also plagued with hundreds of "Empty 0 fleets" that shouldn't be there. Probably a bug that saps resources.
I wont comment on the AIs war strategy, and planetary development - everything was handled in an atrocious and wasteful manner. Watching with debug_ai for a few hours reveals how bad the whole thing operates, and allows you to plan for even more exploitation on future campaign.
[edit:spelling]
With the Crisis (contingency) and War in heaven finished/defeated, there is no war, and everyone is in the LNAP.
YoY: Just an Impressive 341.37 secs, Average Monthly: 28.41 secs
Obviously, fleets moving, fighting, and AI planning takes significant resources. This result is about 40%-45% faster than the previous data point. Adding to that, I noticed the AI had many small fleets and large fleet fragmentation. Not sure if this is intendend or not. Also the fleets had huge amounts of missing reinforcements, and some had reinforcements set higher that what the command limit will allow.
The fleet manager of all empires is also plagued with hundreds of "Empty 0 fleets" that shouldn't be there. Probably a bug that saps resources.
I wont comment on the AIs war strategy, and planetary development - everything was handled in an atrocious and wasteful manner. Watching with debug_ai for a few hours reveals how bad the whole thing operates, and allows you to plan for even more exploitation on future campaign.
[edit:spelling]
Last edited:
- 2
hi, I want to build a new PC. Can I know what computer processor can handle 600-800 star system and 10-15 empire just nice? budget is a bit tight.
If anyone has i5-10400/i5-10600k, are they good? because it's cheap at my place.
If anyone has i5-10400/i5-10600k, are they good? because it's cheap at my place.
I'd still suggest you to get a Xen 3 CPU. Even the entry level 5600 & 5800 are quite good.
Zen 3 - Wikipedia
- 1
I have here a savegame for people who want to do some performance testing in a more late game. The year is 2872. The human player has 69,808 pops and there are 135,249 pops total in the galaxy. The endgame crisis has been defeated, all fallen and awakened empires have been destroyed. If you do want to test some earlier progress, I saved a savegame of every 5 years of this game.
- 1
I have the following system:
Processor: Intel Xeon E3-1240 v3
Motherboard: Gigabyte GA-H87-HD3
Cooler: Noctua NH-U12S
Graphics Card: Sapphire R9 290 4 GB GDDR5 OC Tri-X
RAM: Corsair Vengeance 16 GB DDR3 1600MT/s
SSD for game: Crucial M500 2.5" 480 GB
Case: Corsair Obsidian 650D
And if you are wondering about the temperature, here are the temperatures I got when playing my 69k empire in 2.8.1, while at the same time running Chrome with 150+ tabs, Lightroom and a bunch of other programs:
.png "2021-01-27 (1).png")
My setup is already 7 years old, but still performs good enough for me.
It takes about 5-7 seconds per day. I got this far by running Stellaris on one screen and doing something else on another monitor. Then you also got ... about a minute or so each month for calculations. I haven't measured the exact numbers though. Normally this is doable, but the reall 'fun' starts during a war of conquest, where I need to tender all those conquered planets.
Having 800+ civilian ships wasn't the brightest idea though: that screws up the overview.
You need to know that fleets are perhaps an even greater lag source. Also running the game on the galaxy screen is a lot slower that doing it in-system.
- 1
Wait, really? Do you have a breakdown of how much lag each major source causes? Or a way to figure it out?
anything the game needs to calculate adds lag. and fleets recalculate their paths all the time
- 2
Oh, I know that. I was just wondering if it was really more than (or comparable to) pop lag.
Is Star base and Mega/Gigastructure a source of lag? On my moded game, I purge 6000 out of 7200 pops, Kill all ai & fallen empires, delete almost all ship, and merge all species (using mod), but the game still isn’t as fast as mid game.
At the moment, I have about 60 Mega/Gigastructure and 400 starbases
At the moment, I have about 60 Mega/Gigastructure and 400 starbases
Last edited:
Not sure about starbases yet, I suppose if you build platforms they might affect speed, but smaller tests I did were not conclusive. You with the 600 bases save should be in a better place to tell us.
For negastruxtures, if they are colonies then it's the pop lag, otherwise they are very light on performance because they are more or less static.