Inactivity System to prevent Stack Inserters from getting stuck

[teejayhi5]

TL;DR
Add an inactivity system or something to stack inserters so that they don't completely clog up a system if they dont pick up a full stack.
What?
Stack inserters wait until they pick up a full stack which can be very limiting in a variety of situations. Some examples where I have had issues is Fulgora scrap trains, I wanted it to unload onto a belt as a stack but when it comes to quality scrap it stops, waiting to receive a full stack of rare scrap or something. Another example is fruit processing on Gleba, the inserters begin to pick up the fruit, but before it receives a full stack the biochamber comes to a halt because there are 5 seeds in the machine and they must be removed before the machine can keep producing - but then the inserter is stuck with the fruit.
Why?
I shouldn't have to completely rework designs because these stack inserters (that are meant to make life easier) can get stuck so easily. I just want an inactivity system so that if it doesnt pick up any more items in a while and therefore is stuck in place it ends up starting up again, dropping what it is holding and therefore unclogging the factory.


Edit: For people that seem to be confused -
Why would I use a bulk inserter if the whole point of using stack inserters is that it outputs items in stacks? Using a bulk inserter would completely obliterate the purpose.

[Khazul]

Use a circuit to change the filter on it after inactivity.

[DefGie]

Use a circuit to change the filter on it after inactivity.

Just add a couple combinators per inserter to your layout, no big deal?

One can also look at either using multiple output inserters - a stack filtered to the highest-volume output and a different type filtered to rest, for instance - or parallel production lines that load belts unstacked, to be sorted by splitters then consolidated by stack inserters onto fewer belts.

I'm inclined to think it's a good thing that there's more reason than cost to use bulk inserters over stack inserters in some setups. There seem to be adequate tools for us to rise to this challenge, regardless that I'm skeptical circuit is a good one in this case.

[Khazul]

Use a circuit to change the filter on it after inactivity.

Just add a couple combinators per inserter to your layout, no big deal?

One can also look at either using multiple output inserters - a stack filtered to the highest-volume output and a different type filtered to rest, for instance - or parallel production lines that load belts unstacked, to be sorted by splitters then consolidated by stack inserters onto fewer belts.

I'm inclined to think it's a good thing that there's more reason than cost to use bulk inserters over stack inserters in some setups. There seem to be adequate tools for us to rise to this challenge, regardless that I'm skeptical circuit is a good one in this case.

I must admit I only tend to use filter clearing to empty a stack inserter when I already have a load of circuit control going on and it really matters - for eg biter egg handling etc.

[Muche]

A stack inserter that doesn't get stuck when its hand is not full and there is no more of that item at the source location is called bulk inserter.

[Gaagaagiins]

The above point about how what you're describing is just a bulk inserter is totally accurate. And, really, if a stack inserter is stalling because of an issue like that, doesn't that imply you don't even need the greater overall item moving speed that they provide anyway? Even if you do, wouldn't the answer then be to make bulk inserters of a higher quality, or to just change the design, like with a longer section of belt, or inserters on both sides of a belt?

I won't speak for the developers but I feel that this is a intended functionality limitation for stack inserters. They're not meant to be used everywhere, they're meant for specific applications, or at least getting the most out of them requires deliberate and careful usage of them so that you can design around the quirks in their functionality.

I'm not in favor of a fundamental change like this that just turns stack inserters into super bulk inserters. That just makes the game less interesting.

[teejayhi5]

A stack inserter that doesn't get stuck when its hand is not full and there is no more of that item at the source location is called bulk inserter.

The whole point is that stack inserters output in stacks?

[teejayhi5]

The above point about how what you're describing is just a bulk inserter is totally accurate. And, really, if a stack inserter is stalling because of an issue like that, doesn't that imply you don't even need the greater overall item moving speed that they provide anyway? Even if you do, wouldn't the answer then be to make bulk inserters of a higher quality, or to just change the design, like with a longer section of belt, or inserters on both sides of a belt?

I won't speak for the developers but I feel that this is a intended functionality limitation for stack inserters. They're not meant to be used everywhere, they're meant for specific applications, or at least getting the most out of them requires deliberate and careful usage of them so that you can design around the quirks in their functionality.

I'm not in favor of a fundamental change like this that just turns stack inserters into super bulk inserters. That just makes the game less interesting.

I dont see how it is accurate. A bulk inserter doesn’t output items in stacks, therefore using one completely obliterates the point. and I don't understand how this implies in any sort of way I have such a production that stacks aren't necessary.

[teejayhi5]

Use a circuit to change the filter on it after inactivity.

Just add a couple combinators per inserter to your layout, no big deal?

One can also look at either using multiple output inserters - a stack filtered to the highest-volume output and a different type filtered to rest, for instance - or parallel production lines that load belts unstacked, to be sorted by splitters then consolidated by stack inserters onto fewer belts.

I'm inclined to think it's a good thing that there's more reason than cost to use bulk inserters over stack inserters in some setups. There seem to be adequate tools for us to rise to this challenge, regardless that I'm skeptical circuit is a good one in this case.

The point is that I shouldn't have to.

[teejayhi5]

Use a circuit to change the filter on it after inactivity.

I really shouldnt have to do that though, its such a simple thing that could be implemented

[CyberCider]

Stack inserters were designed to not be a straight upgrade to bulk inserters. You can’t replace every bulk inserter with a stack one and expect your base to work the same. That’s just part of their design. This is why a blank upgrade planner won’t automatically upgrade inserters to stack. Because there are situations where stack inserters aren’t necessarily the way to go.

[Khazul]

Use a circuit to change the filter on it after inactivity.

I really shouldnt have to do that though, its such a simple thing that could be implemented

And the same argument applies to so many things in Factorio. So many thing I curse the devs for almost daily, but they are what they are and I doubt they will ever change and the best I can hope for is a modding API gets improved/added to allows us to complete/fix or whatever. Stack inserters as they are are very good when used as intended. I can understand that maybe they work as they do for balance reasons or to force a choice that the devs want us to have to make or fall back to circuits if you want the best of both worlds. They dont come under my heading of needlessly primitive or needlessly broken. They seem reasonable to me.
And yes - I do often end up doing little conversion setups from non-stacked to stacked, or using circuits to force them to drop their contents in specific cases. Not really a fan of having to do so mainly because of how circuits work (their horrible tick latency which I also get the reason for even if I utterly hate it), but I understand and accept it.

[Muche]

A stack inserter that doesn't get stuck when its hand is not full and there is no more of that item at the source location is called bulk inserter.

The whole point is that stack inserters output in stacks?

Yes.
And if the stack inserter doesn't have its hand full, it is not able to drop a whole stack.
By wanting the inserter to not get stuck, i.e. drop the items in its hand despite their amount is not a full stack, you want it to violate the promise of dropping only full stacks.

[teejayhi5]

A stack inserter that doesn't get stuck when its hand is not full and there is no more of that item at the source location is called bulk inserter.

The whole point is that stack inserters output in stacks?

Yes.
And if the stack inserter doesn't have its hand full, it is not able to drop a whole stack.
By wanting the inserter to not get stuck, i.e. drop the items in its hand despite their amount is not a full stack, you want it to violate the promise of dropping only full stacks.

All this would do is in a worst case scenario provide a stack SLIGHTLY less than maximum. Also most cases 4 items can be picked up, it just waits till it picks up 16.

[teejayhi5]

Stack inserters were designed to not be a straight upgrade to bulk inserters. You can’t replace every bulk inserter with a stack one and expect your base to work the same. That’s just part of their design. This is why a blank upgrade planner won’t automatically upgrade inserters to stack. Because there are situations where stack inserters aren’t necessarily the way to go.

Stack inserters are certainly the way to go in situations where I need 240i/s on one belt.

[robot256]

Stack inserters were designed to not be a straight upgrade to bulk inserters. You can’t replace every bulk inserter with a stack one and expect your base to work the same. That’s just part of their design. This is why a blank upgrade planner won’t automatically upgrade inserters to stack. Because there are situations where stack inserters aren’t necessarily the way to go.

Stack inserters are certainly the way to go in situations where I need 240i/s on one belt.

That's true. What kind of application do you have that needs 240 items/second on one belt and can't fit extra filtered stack inserters to avoid deadlocks? Usually only one or two outputs of a recipe are actually high volume, and you can use a bulk inserter for the rest of them.

[DefGie]

The point is that I shouldn't have to.

Why not?

It's a game about building logistics solutions around constraints, yes? This is one of the constraints the game presents, and all you've conveyed is, "That constraint is annoying, I don't feel like dealing with it."

That's not to say that WUBE are gods of balance don't question their judgement, or that my opinion on balance is better than yours. But the essence of your request is that you disagree with what appears to be an intentional design choice, and you're not giving a lot of reason for someone who does agree with it (as the developers presumably do) to consider changing their mind.

I favor different constraints kicking in with different recipes - so that more diverse designs are needed within an optimal factory - and this mechanic seems to be a case of that. Bluntly I think you're asking for the game to be dumbed down, in one small area.