Actual length of game night?
Actual length of game night?
On this page, the wiki says the game night (ie the time when solar panels generate no power) is 66.5 seconds long. However, I just measured it with a stopwatch and got a result of somewhere around 40 seconds, not 66.5 seconds. It's not a problem of the game running too fast, as machines idly draining 20kW took around 800kJ from accumulators during the night. I tried searching a bit but couldn't find anything useful. Does anyone know the actual duration of night in the game?
Re: Actual length of game night?
I just measured with my stopwatch, and these are the values I have measured (quite approximative, I'm aware of it) :
Couldn't find anything in a changelog about that change.
[Edit] : Test ran under 0.11.14, with a pretty small factory, and with a quite decent gaming rig, continuously at 60 fps, I'm sure it's not a performance issue.
The day cycle is about the same length (417ish), but dawn and dusk are longer, and full night shorter than the ones reported on the wiki, while full day is about the same.
[Edit 2] : I measured with an autonomous electrical network composed of 1 solar panel (60 kW),a pumpjack (90 kW needed), and a tank for the oil (to ensure the pumpjack always asks for 90 kW). Like that, 100% of the solar panel output is always used, and at the second it produces less than 60kW, I see it.
Code: Select all
phase stopwatch duration (mm:ss) duration(sec)
dusk 01:27 01:27 87
night 02:04 00:37 37
dawn 03:33 01:29 89
day 06:57 03:24 204
[Edit] : Test ran under 0.11.14, with a pretty small factory, and with a quite decent gaming rig, continuously at 60 fps, I'm sure it's not a performance issue.
The day cycle is about the same length (417ish), but dawn and dusk are longer, and full night shorter than the ones reported on the wiki, while full day is about the same.
[Edit 2] : I measured with an autonomous electrical network composed of 1 solar panel (60 kW),a pumpjack (90 kW needed), and a tank for the oil (to ensure the pumpjack always asks for 90 kW). Like that, 100% of the solar panel output is always used, and at the second it produces less than 60kW, I see it.
Last edited by Koub on Tue Feb 03, 2015 12:01 pm, edited 2 times in total.
Koub - Please consider English is not my native language.
Re: Actual length of game night?
Our results match then, and if wiki is right and it wasn't so in 0.11.8, it's either an undocumented change or a bug. Time to where devs will file the report.
Re: Actual length of game night?
IIRC the complete day/night cycle used to be approx. 10 minutes. According to your measurements it is more like 7 minutes, so it seems like the whole cycle has been shortened? Or was it always only 7 minutes long?
Re: Actual length of game night?
After some schooling from Rseding, it seems the wiki was wrong. The day-night cycle is 7 minutes long and night is 41 seconds long, and it has been this way for at least a year.
Re: Actual length of game night?
If anyone feels tempted, the old Factorio versions can still be downloaded (down to 0.6) to make timing measurements
Re: Actual length of game night?
Added the method I used to measure the day/night cycle in my initial post.
My guess is that the method used to time what's written in the wiki had a flaw : if your solar panels produce more than your factory needs, you don't start to see the lowering of power produced by solar before you underproduce energy.
Example : if your factory consumes 100 MW but you have enough solar panels to produce 300 MW, you'll see night begin when you're 66% into dusk or 33% into dawn.
My guess is that the method used to time what's written in the wiki had a flaw : if your solar panels produce more than your factory needs, you don't start to see the lowering of power produced by solar before you underproduce energy.
Example : if your factory consumes 100 MW but you have enough solar panels to produce 300 MW, you'll see night begin when you're 66% into dusk or 33% into dawn.
Koub - Please consider English is not my native language.