Choose the number of items inserted by inserter

[bruno]

Hi, and sorry about my poor english

This is an idea :

when an inserter fill in (or "put in / into") coal in boiler, it "does" 5 units.
When in put "regular magazine" in a tureet, it does 10 units.
It could be great to be able to modify these numbers. Example : fill in with 50 units.

It could be like a rule drawing with
0 5 10 15 ... with a cursor for choosing which number we want, and set to 5 by default.

It could be set for all the boilers (for example) with a check box.

Bye bye

[ssilk]

Can you also explain, why this would be useful, please? Cause I really have no idea, why that could be needed.

[Nemoricus]

I can see some situations where this would be useful. In general, the higher an inserter fills a structure to, the greater the buffer the structure has against shortages or heavy use.

Gun turrets are a nice example. They're needed for only brief periods at irregular intervals, and the more ammo they have, the longer they can fire before they need to be reloaded. During heavy assaults, the default filling limit may prove inadequate. There are some ways to increase the stack limit indirectly, like putting a chest beside the turret with an inserter to load it. But that's less space efficient and may not be able to keep up. If inserters instead filled the gun turret to its limit directly, it would have a much greater buffer.

However, the ability to choose the fill limit for inserters should be a feature of *smart* inserters only (ie, add Ammo/Fuel/Ore to target while target has less than X of the item in it). Regular inserters just move things according to their default parameters, while the smart ones can make more complex decisions.

[MeduSalem]

Can you also explain, why this would be useful, please? Cause I really have no idea, why that could be needed.

I'd know an application where this would be useful... if you still remember my multi-purpose smelter setup from some months back, ssilk...

https://forums.factorio.com/forum/vie ... 832#p47500

There it was always a problem that due to inserter stack size bonus and default stacks in furnaces and recipe requirements there's the possibility of an odd number of ore/items in the furnace, which is pretty much undesireable because of how it might block the entire furnace. xD

But for that application it would need to be more finegrained. Steps of 5 items isn't accurate enough.

[Cloner]

Can you also explain, why this would be useful, please? Cause I really have no idea, why that could be needed.

Can be done by hand - so in a game of "automate everything" should be easy automatable.

[Trepidati0n]

The stack bonus right now is driving me bonkers on producing sulfuric acid. The stack bonus + speed modules = stalling because it waits till it is empty before it moves it. Thus..no matter what..I have a full inserter delay. Probably dropping my battery production by near 25%. Just need to add more..but I should have to.

[DaveMcW]

Delay for a fast inserter is 0.25 seconds... that is not the problem unless you are producing 40 sulfuric acid per second.

[Cloner]

The stack bonus + speed modules = stalling because it waits till it is empty before it moves it.

Same for me, with speed modules the assembler buffer is not enough, but inserter is staling often.
Maybe add a late research to DOUBLE all building buffers?

[DaveMcW]

Same for me, with speed modules the assembler buffer is not enough, but inserter is staling often.
Maybe add a late research to DOUBLE all building buffers?

You need faster output. Output into a chest for stack size bonus, add more output inserters.

[Trepidati0n]

I think that is overly complicated. I look at it like this. when you go to the grocery store to buy milk for a "large family" you buy lets say can reasonably store 4 gallons. But lets say some weeks you use 2 gallons and some weeks you need 4. Do you wait until you are out of milk because you can buy "4 gallons". If you do this it causes unnecessary movement and disruption. The logic for direct insertion should be "move the most you can based upon empty space up to the research limit". If you have room for 2..move 2. If you have room for 3..move 3. But don't wait for "maximum". Isn't that what the internal buffer for the factory/assembler is for?

The point of direct insertion is to make things faster...but eventually, it makes things slower for 0.5 speed items at the top research. Top direct insertion research is great for trains but it screws up other shit too much. A research option shouldn't force you to go back to belts..that is just seems silly.

[ssilk]

@thread writers: Please don't mix the inserter stack size bonus ( https://forums.factorio.com/wiki/inde ... size_bonus ) with the insertion limits ( https://forums.factorio.com/wiki/inde ... ion_Limits ). This thread is about the insertion limits.

And my own opinion is: There needs to be more than good reason for a change of the inserter behaviour.

[Trepidati0n]

@thread writers: Please don't mix the inserter stack size bonus ( https://forums.factorio.com/wiki/inde ... size_bonus ) with the insertion limits ( https://forums.factorio.com/wiki/inde ... ion_Limits ). This thread is about the insertion limits.

And my own opinion is: There needs to be more than good reason for a change of the inserter behaviour.

IMO, they are not mutually exclusive since once impacts another at a certain point of research. Arguing otherwise is against simple logic. Fixing one item or the other solves the problem...but making a decision in vacuum when options exist isn't good design practice. I know rules are rules per the "stickied thread", but sometimes you need to know when following rules is just an excuse for not dealing with a problem.

[MeduSalem]

My opinion on the topic is that we should have a slider at each individual Inserter that specifies how many items the Inserter grabs in each turn.

That would be the most logical solution, since I don't know why we shouldn't be able to "downgrade"/"override" the Inserter stack size bonus manually for each individual Inserter if we wish to do so for some reason.

At least one can't undo the research once completed anymore and there are quite some interesting scenarios where it could come in pretty handy if we could specify the amount of items being grabbed.

The possible settings should obviously be (1,2,3,4,5) and an additional "default"-mode in which the Inserter will scale to the maximum amount of items possible whenever a Stack Research is finished so that all the inserters keep on scaling while the ones with a manual override do not. Newly placed Inserters would always work in the default mode until they are manually overriden.



If someone wants to suggest a slider for Gun-Turrets to control how many mags there are buffered it would be nice.

For stuff that is done in assemblers it would become quite ugly though especially for products that have 2 or more ingredients. 4-5 individual sliders would seem pretty dumb. ^^

On the other hand it doesn't really work for furnaces at all, because furnaces themselves don't even know in advance which ore they are going to smelt and the recipes don't match up in their values to make a general assumption.



So I'd favor the slider in the inserters because they seem like the least amount of work and are pretty self-explanatory.

[ssilk]

An inserter, that takes only one makes eventually sense. But 2,3,4? I have no good usage case for that, and if, they are so complicated to build, that it don't makes sense, cause you always need to care about the case, when the chest/whatever it takes from is nearly empty.

And how to make that clear on map? You come back to game after one day pause and forgot that this inserter is special. Isn't it easier to use then 2 inserters then, one which puts it on earth, and the other grabs it from there?

But it would be a good idea to see the current inserter stack size if you hover an inserter.

Edit: I thought about the insertion limits: makes eventually sense to have a switch "default / double". Cause for assemblies/furnace I don't see the need to have that for every item type. If I hovering this switch it shows the exact numbers that will be filled in now.

[MeduSalem]

An inserter, that takes only one makes eventually sense. But 2,3,4? I have no good usage case for that, and if, they are so complicated to build, that it don't makes sense, cause you always need to care about the case, when the chest/whatever it takes from is nearly empty.

It would open up some possibilities in advanced logic circuitry that aren't currently do-able without a lot of extra wiring or tons of other workarounds, or aren't do-able at all in the first place.

At least in most cases It always becomes quite a mess to wire up all the inserters, especially if they should work only in a special condition like picking up stuff only when there's exactly 1, 2, 3 or 4 items in the chest all while wanting to reserve the circuit network inputs for other conditions rather than something simple as that.

Even more frustrating are scenarios where you want to grab exactly 1/2/3/4 items but can't wire the inserter in a way to achieve that because robots may deliver more items than you want to because they are also limited to potentially grab 4 items on maxed out research. Imagine you want to grab only 2 items per turn but the bots may deliver anything ranging from 1 to 4 items... how do I get the Inserter to grab only 2 items out of the 3-4 the bot may have delivered? It's not easily do-able. Not in a space-saving manner. I already had a couple of scenarios where that got a huge issue and I had either to invent a ridiculously complex workaround to deal with that or discard the idea altogether.

Also on the other hand I hate wiremess. I try to avoid it if I can. I hate everything that ends up with me connecting wires manually, especially if it's not possible to blueprint without risking stupid auto-connection issues, or when Shift+Right/Left-Clicking might connect the Inserter/Chest to the wrong electric poles, thereby forcing you to redo everything from scratch.

Sometimes it's quite hard to work with the limitations of available circuit network conditions as well, so it would save up a bit on that too.

In my opinion having the slider to manually override the stack bonus is not only like giving an additional flow-control, but also something that can be used beyond the regulation of plain production throughput in complex logic setups.

And how to make that clear on map? You come back to game after one day pause and forgot that this inserter is special.

As a simple answer on how to make a clear indication: While hovering over the inserter there could be an entry stating "current stack bonus: xx" or something like that and you'd know what the inserter is set to. Or maybe a small hovering icon with a number (1-5) above each inserter that is set to a manual stack override, how about that? Inserters that are not manually overriden have no floating icon above them and only show the stack bonus when the mouse is hovered above them.

Isn't it easier to use then 2 inserters then, one which puts it on earth, and the other grabs it from there?

To have an Inserter that puts stuff on the ground just to be picked up by another Inserter again is one of the ugliest solutions I have ever seen in the game and I'd rather avoid it because it wastes space and looks rather questionable, in terms it looks like a workaround to improvise on a basic feature that the game is lacking.

[ssilk]

Ok, good arguments.

For 1 I already come into situations, where I needed it (a circuit, which worked with no stack size bonus, but not with 1). I think this is really annoying.

But can you give us one or more good use cases for 2,3,4,5? Cause I really find none.

[sillyfly]

If it is mostly for circuit network stuff, maybe:
1. Wait until the new circuit network stuff is introduced in 0.12, it may change/alleviate some problems.
2. Give only smart-inserters the option to specify exactly how many items they should grab