Factorio Forums

Screenshots:
https://imgur.com/a/4Uh9W

Save File:
https://www.dropbox.com/s/9hwxz92ymib03ev/Bug1.zip?dl=0

Blueprint:
https://gist.github.com/KingIonTrueLove ... 463eddf78e

Issue:
I accidentally stumbled upon a way to change the way one belt's lane was acting, by changing the input to the other lane.
In the screenshots, I have a contraption set to tell if the bottom output lane is 100% compressed or not. If the belt ever loses any compression a stack inserter will be moved.

On the first screenshot, I have left this on for over 10 minutes without a single stack inserter being moved.
Then i remove the piece of belt circled in yellow. This should only affect the top lane of the output belt, not the bottom lane.

On the second screenshot, it shows that somehow the bottom lane is affected by the change in the top lane. As within a minute several stack inserters were moved. Thus proving, that one belt's lane can affect the other, even if only slightly.

Replication: The easiest way to do so is to use the save file. If you use the blueprint you have to wait a long time for the smeltries to 'spool up' and provide constant compression on the bottom output lane. It could be up to 20 minutes for the 'spooling up' phase to actually complete :/

I know that this isn't a huge bug or anything, but this could help determine the reason why sometimes side-loading isn't perfect, or some other belt-related bug, so I'm taking the time to post this.

The only mod running for this test was the creative mode mod. If requested I could later replicate the setup in pure vanilla if needed.

After removing the marked belt segment, the top lane on the output belt will have gaps it didn't before, because the items in the other lane are now out of position to sideload into them.

So that means there will be lower throughput on the no-longer-compressed top lane, so with just enough load on the output belt a bottom-lane item will be taken more often, so the lane will stall less often, leaving room for gaps the lighter
bottom-feeder-belt lane isn't in position to fill either.

quyxkh wrote:After removing the marked belt segment, the top lane on the output belt will have gaps it didn't before, because the items in the other lane are now out of position to sideload into them.

So that means there will be lower throughput on the no-longer-compressed top lane, so with just enough load on the output belt a bottom-lane item will be taken more often, so the lane will stall less often, leaving room for gaps the lighter
bottom-feeder-belt lane isn't in position to fill either.

Both lanes go directly into an item void. The 'load' on the bottom lane does not change at all.

In order to help isolate the mod from being the cause, what happens if you split the lanes and put them each into their own voids?

TruePikachu wrote:In order to help isolate the mod from being the cause, what happens if you split the lanes and put them each into their own voids?

Problem still persists when the belt is removed as before.

When loading the save in vanilla and replace what is missing, it works completely identical with or without that belt on top. I'll move it to pending until someone can show how to reproduce in vanilla.

I will try to do so in vanilla when I get the chance. It may be that the smelteries didn't properly 'spool up' when replacing the missing bits, and it would take too long to ask you to do so. The setup only *barely* makes more than a half-belt of plates on the bottom lane in the first place.

I have to sleep now and have plans for the morning but just give me a day or two and I will try to get it going in vanilla.

New save:
https://www.dropbox.com/s/elhmpyass0qtn1j/Bug4.zip?dl=0

I replicated it with the creative mod taken off so its completely vanilla now.


Again I have ran it for a long time without removing the belt circled in red, enough to fill storage chests with 20k+ plates. During that time nothing was put in the green area.
Once the belt in red was removed the green area slowly starts to fill with stack inserters. Sometimes it can happen seconds after removing the belt and other times it has taken a bit over a minute.

I have even had someone run the file on their own factorio vanilla install and they were able to replicate it as well.