[0.12.29] [Reproducable] Inserters Jamming

[Xpyder]

If you change an assembler recipe while an inserter is moving a material to an assembler, the inserter jams and won't automatically put the wrong material back and has to be manually rotated to clear it.

While this could be argued to be a valid "feature" for basic inserters, Smart inserters should automatically put the materials back on the belt for a simple reason: The game is about large scale automation.

Once you get to circuit logic and start reprogramming your assemblers, having to manually check each "smart" inserter in a 20, 50, or larger assembler chain is impractical, especially as there's no symbol or notification to tell you an error condition has occurred. You have to zoom in and check the tiny end of each and every inserter to make sure it's not stuck. It's a fine behavior for basic inserters, and it would make a reasonable argument for switching to smart inserters once you unlock them, but needs to be fixed for smart inserters (and above if there are any).

To reproduce:

Build an assembler with a feed inserter
set it to a recipie and have it start building
change to a different recipe while the inserter is in the process of moving a material
Note that the inserter jams if the new recipe doesn't include the material it was loading

[Koub]

If you change an assembler recipe while an inserter is moving a material to an assembler, the inserter jams and won't automatically put the wrong material back and has to be manually rotated to clear it.

While this could be argued to be a valid "feature" for basic inserters, Smart inserters should automatically put the materials back on the belt for a simple reason: The game is about large scale automation.

You're complaining because something you did manually has an effect that doesn't fit the automation spirit of the game ... Isn't that a little contradictory ?

[Xpyder]

This was predicated on the assumption that your assemblers could be part of the circuit network allowing for adaptive programming of manufacturing bases.

I've gotten far enough to discover this isn't the case. I suppose the assumption is that you will set up a logistics robot network with an assembler for each item type and just let them have at it instead of dynamically adjusting your base based on your evolving needs.

[Rseding91]

If assembling machines get the ability to have the recipe set from the circuit network this particular quirk may be addressed. Until then this isn't so much a bug

[Xpyder]

Fair enough, feel free to move this to won't fix until (if) that changes.

In case it comes up, the way I was expecting assemblers to connect was through a control panel device.

Think 16 (0-f) channels, with each having the potential to have an item assigned to them, and assemblers being assigned to a particular channel, letting you set up a mass production sequence with a chain of up to 16 assemblers x however many rows you want to make

[Xpyder]

NOTE: if you this becomes available, it's probably a good idea to let Smart Inserters choose a production channel for their item type and decision filters.