[Idea] Loader recipe and research suggestions

[Gergely]

It might be a good idea to discuss and playtest (through modding) recipe and research ideas for loaders. If you think there's a fair recipe and research cost for them in their current state, feel free to respond to this topic.

You can already obtain loaders in-game with commands.
Normal: /c game.player.insert({name="loader"})
Fast: /c game.player.insert({name="fast-loader"})
Express: /c game.player.insert({name="express-loader"})

Here's a quick template for modding:

Code: Select all

data:extend({
{
    type = "technology",
    name = "loader",
    icon = "__base__/graphics/item-group/logistics.png",
    effects =
    {
      {
        type = "unlock-recipe",
        recipe = "loader"
      }
    },
    prerequisites = {<prerequisites>},
    unit =
    {
      count = <count>,
      ingredients =
      {
        {<science pack>, <amout>}
      },
      time = <time>
    },
    order = "a"--order
  },
{
    type = "recipe",
    name = "loader",
    enabled = false,
    ingredients =
    {
      {<name>, <amout>}
    },
    result = "loader"
  },
{
    type = "technology",
    name = "fast-loader",
    icon = "__base__/graphics/item-group/logistics.png",
    effects =
    {
      {
        type = "unlock-recipe",
        recipe = "fast-loader"
      }
    },
    prerequisites = {"loader",<prerequisites>},
    unit =
    {
      count = <count>,
      ingredients =
      {
        {<science pack>, <amout>}
      },
      time = <time>
    },
    order = "a"--order
  },
{
    type = "recipe",
    name = "fast-loader",
    enabled = false,
    ingredients =
    {
      {<name>, <amout>}
    },
    result = "loader"
  },
{
    type = "technology",
    name = "express-loader",
    icon = "__base__/graphics/item-group/logistics.png",
    effects =
    {
      {
        type = "unlock-recipe",
        recipe = "express-loader"
      }
    },
    prerequisites = {"fast-loader",<prerequisites>},
    unit =
    {
      count = <count>,
      ingredients =
      {
        {<science pack>, <amout>}
      },
      time = <time>
    },
    order = "a"--order
  },
{
    type = "recipe",
    name = "express-loader",
    enabled = false,
    ingredients =
    {
      {<name>, <amout>}
    },
    result = "loader"
  },
})

[Nexela]

There are already multiple loader adding mods on the portal

[Tym]

is there an actual definition of a "Loader" somewhere? I'm assuming it's some sort of higher-volume inserter, albeit with a different graphic.

[daniel34]

is there an actual definition of a "Loader" somewhere? I'm assuming it's some sort of higher-volume inserter, albeit with a different graphic.

See here: https://www.factorio.com/blog/post/fff-128 (scroll down to Loader)

The devs later implemented the stack inserter instead of the loader, but the code for the loader is still in the game. You can get it using the console commands that Gergely posted or by using one of the many mods on the mod portal that unlocks the loaders.

[Lav]

As far as I can tell, there's only a few use cases for loaders:

1. Loading/unloading chests (including bus buffers) at belt speed. With loaders you can dump entire belt into a single chest. With inserters it's a nightmare and tons of wasted space.
2. Loading/unloading trains (completely replace inserters for bulk goods, used in combination with inserters when some filtered loading setup is used, probably useless for filtered unloading setups).
3. Lane balancer (belt - loader - belt) and buffered lane balancer (belt - loader - chest - loader - belt).
4. Belt splitter/merger, especially when dealing with multi-color belts. Merging two red belts into one blue and one yellow can be easily done.

Frankly they look pretty tame for me. I'd estimate their worth as somewhere between underground belt and a splitter of the same color.

UPDATE. Added some loaders via the console. Apparently they don't load into chests or wagons. That makes them completely useless as far as I'm concerned.

[darkfrei]

UPDATE. Added some loaders via the console. Apparently they don't load into chests or wagons. That makes them completely useless as far as I'm concerned.

Why not? You must use R for reversing unloader to loader.

[Lav]

Why not? You must use R for reversing unloader to loader.

Oh, so that's how it works? That's, like, totally counter-intuitive.

I actually tried clicking R, but I thought it merely switches loader's direction, not changes it's function.

Thanks, will try again today once I get home.

One other thing I noticed is that loaders actually have filtering capability. Personally, I would rather completely remove it, that's a job for inserters which actually have the mechanics to pick a specific item from a pile. Loaders should work in bulk IMHO, not caring about item type or lane.

Ideally, when unloading from a container, they should pick two random items (instead of two items from the last stack in container's inventory), with chances being proportional to item's amount in the container.

[Kryzeth]

Yeahh, loaders are pretty OP. Honestly, it's not even the technology cost or the resource cost that makes them OP, but just the way they were implemented. I added loaders into a vanilla-plus playthrough, and at first I only used them for in-line buffer chests. Circuits, gears, steel; that sort of thing. The resource cost for the mod I was using was pretty expensive for the early game (5 inserters and some other junk), but later became trifling. They are SO much more powerful than inserters.

Soon, I started using them for trains, realizing that instead of requiring a fancy setup for inserters to unload and compress their output to a belt, I could just use a few loaders. I don't even have to route power to them.

And THEN I realized you could input directly into an assembling machine with loaders. My landfill processing increased exponentially (I started on an island with Very Big water). Now I use them everywhere I possibly can.

Add to this, the fact that loaders have filtering capabilities, and now almost ANY problem that required inserters can now be solved with loaders, even the cheapest kind. The only thing they lack is circuit and logistic network conditions.

I think the only way to possibly balance them, is by having a (HIGH) idle and active power drain, removing their ability to load and unload from machines, AND removing their filtering ability. Because honestly, they're powerless filter stack inserters and lane balancers all in a single 2x1 tile space. WAY too OP as they are now.

[Lav]

Hmm. On the other hand, I don't see a lot of use for loaders.

They're useful for in-line buffering (nice, but not critical), train loading/unloading (where they replace a number of overcomplicated designs), and supplying ingredients for a few recipes with excessive resource requirements.

I was originally quite enthusiastic about merging/splitting lines, but loader-based setup often takes practically the same amount of space as splitter-based, if not more, depending on whether your belts are parallel or orthogonal.

Loaders don't randomize their output, so they're useless for sushi belt building (though they would make a very poor sushi belt setup anyway).

They seem to balance lanes, but are quite unreliable in doing that as far as my experiments show. If there's space, splitter-based balancer is still superior, but loaders could still play this role in space-constrained setups.

So the only place where loaders really shine is train loading/unloading, and let's face it, the game direly needs a specialized solution for that.

Oh, and there's a nice niche for filtered loaders - splitting output of mixed resource mines. But as I mentioned, I'd rather have the filtering functionality removed from loaders entirely.

P. S. It's always hilarious to read about OP loaders when a regular splitter is performing an immensely more complicated job at no power cost.

So, a quick checklist:

1. Belt-to-chest and chest-to-belt. Loaders rule.
2. Belt-to-factory. Inserters rule unless you need a huge volume, in which case loaders rule at the cost of more used space. Only relevant for a few recipes though.
3. Factory-to-belt. Inserters rule - there's no recipe IIRC that would need loaders.
4. Factory-to-factory. Inserters rule supreme as loaders cannot do that at all.
5. Train loading/unloading. Loaders rule.
6. Belt splitting/merging. No clear winner between splitters and loaders, depends on specific belt configuration.
7. Lane balancing. Loaders use less space than splitters, but take more resources (2 loaders + chest vs a single splitter).
8. Dense multiple belt setups - it would be very funny to see someone try to use loaders to get resources into a factory/lab from 3 parallel belts - something that inserters do easily.
9. Sushi belts or two lanes of different resources per belt - forget about loaders, they will mess up everything. Inserters are the only choice as soon as you're dealing with lane tricks.

In short, I don't see anything OP here. Loaders have their niche role in bulk loading/unloading setups - a niche that's currently empty and is being filled by overcomplicated inserter-based solutions. Sure, those solutions will become useless with loaders. But there will be new solutions and new tricks, so who cares?

Strangely enough, nobody complains about pumps being able to fill fluid wagons - even though they're essentially the same loaders in disguise. Shouldn't loader-haters object against pump-based loading and promote loading barrels into cargo wagons instead? It's the same principle.

[darkfrei]

And THEN I realized you could input directly into an assembling machine with loaders.

Yes, it can be useful for instant smelting.
https://mods.factorio.com/mods/darkfrei/Loader-Furnace

[Lav]

Erm - how is this useful? Unless your smelter is over-beaconized to the point of ridiculousness (and it's not, considering I don't even see production modules on the screenshot), inserters would be much more convenient in this setup.

UPDATE. Now this is a real loader-based smelting setup. Notice how it's so much more efficient than those inserters... not.

[darkfrei]

Loaders will be overloaded.

[bobingabout]

Erm - how is this useful? Unless your smelter is over-beaconized to the point of ridiculousness (and it's not, considering I don't even see production modules on the screenshot), inserters would be much more convenient in this setup.

UPDATE. Now this is a real loader-based smelting setup. Notice how it's so much more efficient than those inserters... not.

And that's not even how I do smelting!
I keep coal on it's own belt on an inside lane, for when I eventually "upgrade" to electric power.
but yes, much bigger, lots of splitters.

[Kryzeth]

Strangely enough, nobody complains about pumps being able to fill fluid wagons - even though they're essentially the same loaders in disguise. Shouldn't loader-haters object against pump-based loading and promote loading barrels into cargo wagons instead? It's the same principle.

I'm pretty sure this and that are two very different matters. Pumps have a completely different role in other areas.

I suppose if we speak of purely vanilla, then loaders are probably not overly useful. Although I think it does replace any 1-1 lane balancer by virtue of chest buffering. Instead of having to do a silly little splitter squirrelly-do.

For me, I was discovering the usefulness of loaders in two separate but similar modded playthroughs; both cases involving warehouses and ore sorting. One mod was Ore Chaos, which places an amount of random ores inside of patches, forcing players to figure out a sorting system. So I'd load them into a train, fly them back home, then load them out of the trains, load them into a warehouse, and then load them out of the warehouse wherever they need to go, using the power of filters and powerless voodoo. You encounter a similar problem using angel's-bob's, where the output of various ore sorting recipes results in massive amounts of side products. So the easiest solution is to loader into it all into a warehouse and then unloader it all into separate processing belts.

The last modded situation I used loaders in involves Factorissimo 2, specifically where you can use boxes to transfer products from inside and outside the factory building. So instead of having to lane balance any of the output, you can just haphazardly shove it all into a box, and the output will always be a perfect 50/50 balance.

I suppose if they couldn't filter, then they might (MIGHT) be on the same level as splitters, in terms of usefulness, in a purely vanilla setting.

[Lav]

For me, I was discovering the usefulness of loaders in two separate but similar modded playthroughs; both cases involving warehouses and ore sorting. One mod was Ore Chaos, which places an amount of random ores inside of patches, forcing players to figure out a sorting system. So I'd load them into a train, fly them back home, then load them out of the trains, load them into a warehouse, and then load them out of the warehouse wherever they need to go, using the power of filters and powerless voodoo. You encounter a similar problem using angel's-bob's, where the output of various ore sorting recipes results in massive amounts of side products. So the easiest solution is to loader into it all into a warehouse and then unloader it all into separate processing belts.

The last modded situation I used loaders in involves Factorissimo 2, specifically where you can use boxes to transfer products from inside and outside the factory building. So instead of having to lane balance any of the output, you can just haphazardly shove it all into a box, and the output will always be a perfect 50/50 balance.

I suppose if they couldn't filter, then they might (MIGHT) be on the same level as splitters, in terms of usefulness, in a purely vanilla setting.

From my experience, loaders are much less useful than splitters in vanilla gameplay, though I didn't have to sort ores anywhere (very low frequency big patches do have this effect) and thus never had to use the filtering function. In fact there was only one place where I used loaders in bulk - my hub rail station, where I stored all major resources before belting them to science and production facilities. Pretty sure I could replace all other uses with splitters without losing in performance and, depending on loader production cost, actually win in invested resources.