[2.0.14] Inserter not inserting into recycler that processed red chips

[MSgtGunny]

I created a small setup to delete items on Fulgora above a certain amount, but the logic when its recycling red chips breaks when given red circuits due to what looks like an inserter bug in relation to recyclers. The recycler has no items in its input, has red circuit ingredients in its output/result boxes, and a few items on a belt that lead to an inserter to put them back into the recycler. The initial input inserter that inserts the red chips is set to only insert if there are no items on the belt so that it doesn't get blocked.

Reproduction Blueprint:

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

Copy blueprint

Inserter saying waiting for source items when there are items on the belt beneath its hand

Bugged inserter state

Screenshot 2024-11-05 221115.png (2.36 MiB) Viewed 991 times

Recycler showing no input ingredients so its available to accept any input ingredient and it has space in the result for byproducts

Empty recycler input/ingredients

Screenshot 2024-11-05 221815.png (469.53 KiB) Viewed 991 times

It does not immediately lock up, but once the output belt is full and there are still items in the recycler result, the inserter gets locked and is preventd from putting an item into the recycler even when its input is empty. To quickly reproduce using my blueprint, turn the right inserter so it no longer is pointed to the recycler and temporarily disable the circuit On/Off condition on the left inserter, once the output belt is full and there are a few items in the recycler's result box, reset the inserters back to their original state.

[Rseding91]

Can you please post a save file that shows the issue you're describing? I'm having a hard time understanding how to reproduce what you're describing.

[MSgtGunny]

Can you please post a save file that shows the issue you're describing? I'm having a hard time understanding how to reproduce what you're describing.

Sure thing, heres a slimmed down save with just the bugged setup on fulgora

[Rseding91]

Looking at the save it looks like two things interacting here to make it confusing:

1. The current recipe produces plastic when recycled but is not overloaded on the output (yet) and would accept more red circuits
2. The next item to insert is plastic but there *is* an overload of plastic so it won't insert it

With the two combined and because you've blocked loading more red circuits they both fail to insert. You need to wait for the recycler to have nothing in the output before blocking the red circuits from being inserted.

[MSgtGunny]

Looking at the save it looks like two things interacting here to make it confusing:

1. The current recipe produces plastic when recycled but is not overloaded on the output (yet) and would accept more red circuits
2. The next item to insert is plastic but there *is* an overload of plastic so it won't insert it

With the two combined and because you've blocked loading more red circuits they both fail to insert. You need to wait for the recycler to have nothing in the output before blocking the red circuits from being inserted.

Could you explain 2.? From what I know, if you recycle plastic you have a chance of outputting plastic, and since the output slot on the recycler has room for more plastic to be stored, I don't see why plastic isn't a valid insertion recipe target with the recycler in its current state.

[Rseding91]

A recipe is considered to be “output overloaded” if the output contains 2x the results. Since advanced circuit recycling produces multiple plastic bars at once 2x the result is larger than the single plastic bar recycling recipes 2x results.

The advanced circuit recycling recipe is not output overloaded.

The plastic bar recycling is output overloaded.

[MSgtGunny]

A recipe is considered to be “output overloaded” if the output contains 2x the results. Since advanced circuit recycling produces multiple plastic bars at once 2x the result is larger than the single plastic bar recycling recipes 2x results.

The advanced circuit recycling recipe is not output overloaded.

The plastic bar recycling is output overloaded.

I see the logic thats occurring here, but I think the text state of both entities is wrong then at the very least. The recycler shows "No ingredients" and the inserter shows "Waiting for source items".

Screenshot 2024-11-19 174315.png (1.95 MiB) Viewed 702 times

Given what you are saying I would expect the Recycler and Inserter to both say "Target Full" as shown below, although in this case this seems like a related but different bug since there is only a single plastic in the Result box, but it still won't insert more advanced circuits:

Screenshot 2024-11-19 174415.png (1.18 MiB) Viewed 702 times

[MSgtGunny]

A recipe is considered to be “output overloaded” if the output contains 2x the results. Since advanced circuit recycling produces multiple plastic bars at once 2x the result is larger than the single plastic bar recycling recipes 2x results.

The advanced circuit recycling recipe is not output overloaded.

The plastic bar recycling is output overloaded.

Actually, after experimenting with this a bit more, I think I found a pattern. I think the recipe is only considered overloaded if an inserter is trying to change the assigned recipe of a recycler by inserting a different item into it than the one it last processed (in this case plastic, wire, or green circuit is trying to be inserted into a recycler that last processed advanced circuits) and the result box of the recycler has 2x the results of the ingredient trying to be inserted. And so the inserter sees it can't change the recipe of the recycler and so it doesn't pick up an item causing it to show the state "Waiting for source items" and the Recycler shows "No Ingredients" because its input is empty.

If the inserter is not trying to change the recipe it will happily keep inserting more of the same item that the recycler previously processed, well above the 2x result limit. This only happens on recipe changes, you can have a recycler assigned the plastic recipe, with 40 plastic in the result, and you can still insert plastic into the input both manually with your hand as well as with an inserter.

[Seppl1]

As far as my tests go: Recycling products are always placed into the recycler's output slots in a specific order. Recycling Low-Density Structure (LDS) will always produce steel in the first slot, copper plate in the second slot and plastic in the third slot.
If you empty out a recycler and recycle one LDS, and it happens to not produce steel, then the first slot will stay empty (and the other products are placed in slots 2 and 3, if produced). This is very apparent when recycling scap, which is very likely to leave gaps.

If any of the slots that could be filled by recycling a given item are already occupied by a different item (i.e. the potential product can't stack onto it), then the recycler consumes the input item and fills the progress bar, but none of the products will be placed into any of the output slots and the recycler stops there. This is the case, even when the recycling did not actually produce the conflicting product. For example, if (after LDS) you recycle copper wire, there's a 75% chance of producing nothing at all, but it might produce copper plates in the first slot, so if the first slot is still occupied by steel from the LDS, the recycler will stop. And if you then remove the steel from the first slot, the recycler immediately continues, which 75% of the time will not actually place a copper plate into the slot you just cleared.

But you can recycle steel chests after LDS, because those only produce steel in the first slot. And if you move the steel to a different slot and recycle more steel chests, the new steel is still placed in slot 1, and the recycler stops once that slot is full. It won't add to the existing half-empty stack in another slot.

So the recycler does not treat the output slots as "just inventory", which is quite unintuitive.

This interacts with insertion in the following way:
If the recycler is stuck at 100% and empty input slot because the last consumed item is a copper cable while the first output slot is blocked, then:

• you can insert copper cables into the input slot by shift-clicking them from your inventory, but you cannot do so with other items.
• you can insert anything you want into the input slot by picking it up by left-clicking the items in your inventory and placing them into the input slot by left-clicking there.
• you cannot insert any items at all with inserters. The inserter won't pick up any new items. But interestingly, it says "Target is full" for me, and not "Waiting for source item", so there might be even more hidden and confusing behavior going on.

This is extremely confusing because it doesn't look like the recycler is full. It can have plenty of room in the output slots, no item in the input slot, and yet, you can't insert new items to process.
With this behavior, it's difficult to use one recycler for multiple input items.

I hope this will be fixed (or at least clarified in the recycler UI), and I hope that this text helps other people who are confused.