Loader Poll - Trying to get an idea of what the opinions are

[Miatog]

I'm new to factorio so this is probably a stupid question...but what are loaders? I tried to find them and I can't find anything on the wiki. I found an old FFF that brought them up as a random idea the dev was toying with. An out of date mod post...and a lot of random reddit posts about them. Are they in game? Is this about a mod? I'm so confused >_>

[Lav]

I'm new to factorio so this is probably a stupid question...but what are loaders? I tried to find them and I can't find anything on the wiki. I found an old FFF that brought them up as a random idea the dev was toying with. An out of date mod post...and a lot of random reddit posts about them. Are they in game? Is this about a mod? I'm so confused >_>

Loaders are an item that the developers implemented long ago but made a mistake to ask the community whether it's ok or needs adjustments, and then made an even bigger mistake of actually following community's opinion.

They're still in the game but disabled, with no unlocking research, and primitive graphics. You can add them via console like any other item, plus there are several mods that unlock them.

[Omarflyjoemacky]

I'm new to factorio so this is probably a stupid question...but what are loaders? I tried to find them and I can't find anything on the wiki. I found an old FFF that brought them up as a random idea the dev was toying with. An out of date mod post...and a lot of random reddit posts about them. Are they in game? Is this about a mod? I'm so confused >_>

Loaders are an item that the developers implemented long ago but made a mistake to ask the community whether it's ok or needs adjustments, and then made an even bigger mistake of actually following community's opinion.

OMG. I'm going to print this quote on a 3D printer, spray paint it gold with a nice shiny finish, and mount it on the wall.

Really the only misstep from the dev's in an otherwise perrrrrrfect game. I wonder how many of the staff still fight for loaders and how many fight against.

[silverkitty23]

I wonder how many of the staff still fight for loaders and how many fight against.

At a guess? zero. Too much work to do on new features and bugs.

[Miatog]

I'm new to factorio so this is probably a stupid question...but what are loaders? I tried to find them and I can't find anything on the wiki. I found an old FFF that brought them up as a random idea the dev was toying with. An out of date mod post...and a lot of random reddit posts about them. Are they in game? Is this about a mod? I'm so confused >_>

Loaders are an item that the developers implemented long ago but made a mistake to ask the community whether it's ok or needs adjustments, and then made an even bigger mistake of actually following community's opinion.

They're still in the game but disabled, with no unlocking research, and primitive graphics. You can add them via console like any other item, plus there are several mods that unlock them.

Thank you so much. That clarifies so much!

[Mehve]

And here's the Proof of the Loaders OverPowerdness:

...

Now, if anyone wants to refute this, they're welcome to post their designs functioning in actual factories.

Sarcasm noted, and it's absolutely a fair question, although I might suggest you went out of your play to find some dull and ridiculous examples? I will stress that my primary complaint regarding loaders isn't about "OP-ness", but simply that they bring nothing NEW to the table, OP or otherwise, yet obscure a number of existing core mechanics. That's a gameplay negative in my books. But results aside, this was a fun little distraction

The specific mod used for the loaders here was the "Vanilla Loaders" mod. In all images, each configuration is considered A, B, C, D, etc, going from top to bottom.

Lane Balancing
The ability to take a jumbled up lane and balance things out to make more room on the production side, or maximize the buffer-ability of the receiving side. Since loader distribute what they take from the chest evenly, they're effectively built-in balancers.

In the image below, rows A and B compare the balancing of a belt with only one lane. If you know the items are only in one lane, a single splitter and belt section suffice. The loader is clearly overkill in this case.
Rows C and D compare the balancing of a scattered belt. The first solution no longer works, you need a more complex solution. However the identical loader configuration still works, and saves a significant amount of sideways space...

Lane Balancing 1

...which comes into play when you're dealing with multiple parallel belts together. Here, the spacing savings quickly become obvious.

Lane Balancing 2

Buffering
The ability to store up as much (presently) unneeded product, in as little space as possible, ideally without any impedance to throughput. The last point becomes increasingly difficult as you hit higher throughput levels. Arguably the most obvious application for loaders.

Row A is the obvious implementation. 5 spaces total, no width, 100% throughput even on express belts. A technically perfect buffer. It can be chained in front of itself indefinitely, at a cost of 5 spaces/chest.

As Row B demonstrates, without loaders, there's no hope for a solution that occupies a similar footprint while managing full throughput.

Row C is pretty much the most compact in-line solution available (that I'm aware of), using 4 fully upgraded stack inserters and 6 splitters. You get the occasional hiccup, but it manages >99% throughput. The buffer can be extended at a cost of 4 stack inserters and 4 chests/step. While nowhere near as compact as Row A, it at least offers 4 times the buffer?

Row D illustrates the similar buffer size with loaders - basically the same footprint, but you can get away with yellow loaders and some red splitters. No clear advantage, really. Depends on the cost of loaders vs stack inserters.

But you may as well just have 4 lanes of throughput, as Row E demonstrates. The minimal footprint of the loaders means you can effortlessly buffer the hell out of ALL the lanes in parallel, with no extra spacing required.

Buffering

Train Unloading
The ability to get stuff on and off a train. Alongside buffering, the other blatantly obvious application for loaders.

The first wagon is using loaders, the second using inserters. No contest. 12 express lanes offloading versus 3? In a smaller footprint? Assuming you can actually manage to route all the belts somewhere, and use the product fast enough, you'll spend more time waiting for the next wagon to arrive then you will actually unloading it. In fact, it's so ridiculous, it basically negates any OP-ness - You'll clean that wagon out in less than 10 seconds, and stand around twiddling your thumbs until the next train comes along.

Train Unloading

Belt Merging
Combining two belts into one, one product in each lane. Obviously no point in trying to use a loader in a 2-into-1 situation, but maybe there's an advantage to be found in a 2-into-2 situation? Some way that a box and loaders can do better than some splitters...

...nope. Not that I could find, anyway. Existing game mechanics are more compact, and require very little in the way of special components. Loaders aren't taking anyone's job here.

Belt Merging

Belt Sorting
Sometimes, despite your best efforts (or because of your efforts), there's a bunch of extra crap on your belt that you want to go elsewhere. We already have ways of dealing with that, of course. But since most loader implementations seem to have a filter option, so I figured it was fair game to test out.

Row A and B illustrate a common enough scenario - one of your mines is touching two kinds of ore. You don't want to give up all that ore, but you don't want the extra occasional bit. Simple solution is to just throw down a filter inserter to grab the unwanted stuff when it goes by. Or, you can use a loader with a filter to only send on the good stuff. Except that you'll probably still need to use a filter inserter to get rid of the unwanted stuff, if you don't come by and do it manually. No real advantage to the loader in this case.

But sometimes the junk isn't just occasional. Maybe you want to run one of those new-fangled "sushi belts" or "universal belts" you see forum warriors talk about? Except that you want to sort stuff out later on? Rows C and D illustrate a 2 item sorter. As you can see, while filter inserters can do the job, it's considerably more messy than with loaders.

Belt Sorting 1

With three items in the input, things get even messier with inserters. They can do the job, but not cleanly or compactly, and throughput usually suffers if any one item makes up the majority of the input. Whereas the loader solution remains dead simple and compact, always allowing for 100% throughput no matter the mix. Now, think we've run out of box sides to use loaders on?

Belt Sorting 2

...think again. A simple divide and conquer approach lets you stack loaders and sort total disasters on the input side, always with 100% throughput no matter the actual mix. This can be done with little trouble up to 3 times the number of filter slots on each loader. Beyond that, you need to seriously ask yourself why the belt is that bad to begin with.

Belt Sorting Omega

Splitters
Kind of an odd application, and highly dependent on how the algorithm handles multiple loaders pulling from the same box. But in this case, I found that they seemed to pull from the same box evenly. Which led me to compare them directly to a 1-into-3 balancer and 3-into-1 balancer. Incidentally, this doesn't work on train wagons, or we might just have a brilliant balancing option for any sum of input and output lanes up to 12.

Kind of a wash, really. Similar footprint, similar results? As I just said, this application is highly dependent on the software, and I probably just as soon stick with a tried-and-true belt balancer. But if it was proven reliable, the loader method could potentially be more flexible, since it can be based on both powers of 2 AND powers of 3, unlike splitter-based balancing. An interesting thought exercise, I suppose.

Splitters

So, discuss, criticize, whatever. I had fun making these examples.

[Zeblote]

So basically, you've just shown us why loaders are great - replacing some ugly/bad/overcomplicated solutions with nice and clean ones.

[Lav]

Sarcasm noted, and it's absolutely a fair question, although I might suggest you went out of your play to find some dull and ridiculous examples?

Smelter setup is my response to repeated claims how powerful loaders are, how useful they would be in smelting/production/whatever.

I'm not using a main bus in my current game, so I had to create balancers/splitters/mergers from scratch, though in my game there's a fair few loader-based belt mergers. Yes, as I said, they don't really offer any real advantage over splitters.

I will stress that my primary complaint regarding loaders isn't about "OP-ness", but simply that they bring nothing NEW to the table, OP or otherwise, yet obscure a number of existing core mechanics.

Well, as you've found out, they bring lane balancing, storage, possibly wagon loading/unloading (if that is enabled, as vanilla loaders cannot unload trains ATM). I don't think loaders should have filtering ability (or at least that should be the ability of another line of more expensive loaders, same as we have inserters and filter inserters). With filtering, there's the possibility to filter mixed mining output, which is definitely useful. I was just (un)lucky to have my deposits separate in my current game, plus as I said I don't like loaders to be able to filter. So I missed on this application, alas.

And grats on pointing out a useful 3-way split with loaders, for missing that I have no excuse at all.

Overall though, I don't quite agree with you. Loaders do bring some new possibilities into the game, even if a few, and if train loading/unloading is allowed for them, they could truly become useful. And they're already coded, so adding them to the game shouldn't be a problem. But the game can perfectly continue without them, that's also true.

[bobingabout]

Those nice use case examples just make me want to add these into my logistics mod.

Only thing they really need (in my opinion) to not be "overpowered" is being powered, probably electricity.

[Lav]

Those nice use case examples just make me want to add these into my logistics mod.

Only thing they really need (in my opinion) to not be "overpowered" is being powered, probably electricity.

IMHO if they are capable of filtering, they definitely need a power cost.

[Engimage]

Those nice use case examples just make me want to add these into my logistics mod.

Only thing they really need (in my opinion) to not be "overpowered" is being powered, probably electricity.

I totally agree with you. Just make them consume power and its ok. You might want to use them in some cases but they generally make designs much bigger but there are some cases where you might make use of them. I don't see them replacing inserters for general use. And the cost - if you wann automate some production line like smelting, you will use A LOT more resources - a splitter and a loader combined, and especially express ones, cost tens of times more compared to inserter. So I do not see the reason not to let people do it. Especially if given some heavy power usage.

[Mehve]

Overall though, I don't quite agree with you. Loaders do bring some new possibilities into the game, even if a few, and if train loading/unloading is allowed for them, they could truly become useful. And they're already coded, so adding them to the game shouldn't be a problem. But the game can perfectly continue without them, that's also true.

That's fair. And conversely, if they were implemented tomorrow, I would keep on enjoying the game just as much

[nemostein]

After reading this posts and tried the ingame loaders I've changed my mind completely...
Loaders aren't overpowered at all, if you consider a commom factory.

The only logistics related item that I wanted implemented was a real lane spliter (I really hate that undergroung belt ugly hack).
But now I wish for 2 things, this and loaders... ^^"

[AndrewIRL]

I don't like loaders because their appearance doesn't suggest their function. Inserters, splitters and underground belts all look nice. A loader is just a rectangular prism. What it does, particularly when filtering, is anyone's guess (from the appearance).

Unless the devs can deliver some artwork that look a lot better loaders should not be implemented.

[Zeblote]

I don't like loaders because their appearance doesn't suggest their function. Inserters, splitters and underground belts all look nice. A loader is just a rectangular prism. What it does, particularly when filtering, is anyone's guess (from the appearance).

Unless the devs can deliver some artwork that look a lot better loaders should not be implemented.

It's like beyond obvious that proper loaders wouldn't use the grey placeholder graphics.

[AndrewIRL]

I don't like loaders because their appearance doesn't suggest their function. Inserters, splitters and underground belts all look nice. A loader is just a rectangular prism. What it does, particularly when filtering, is anyone's guess (from the appearance).

Unless the devs can deliver some artwork that look a lot better loaders should not be implemented.

It's like beyond obvious that proper loaders wouldn't use the grey placeholder graphics.

Grey? I thought they were yellow, red and blue?

Here is an example of the yellow ones:

[Lav]

Grey? I thought they were yellow, red and blue?

These are from a mod that adds graphics to loaders. Vanilla ones just use a grey featureless rectangle with an arrow on top.

Real vanilla loaders

[Zeblote]

I don't like loaders because their appearance doesn't suggest their function. Inserters, splitters and underground belts all look nice. A loader is just a rectangular prism. What it does, particularly when filtering, is anyone's guess (from the appearance).

Unless the devs can deliver some artwork that look a lot better loaders should not be implemented.

It's like beyond obvious that proper loaders wouldn't use the grey placeholder graphics.

Grey? I thought they were yellow, red and blue?

Here is an example of the yellow ones:

Huh. Those must be from a mod... you can always identify mod graphics by them not being as good as vanilla ones

I'd imagine real loader graphics looking similar to this thing:


But who knows, maybe they have an even better idea

[netmand]

In respect to loading chests: If you're loading chests as fast you're producing, why are you storing the item? Just produce as needed.
For loading trains: I can see value here, gets the train loaded faster. If the devs set the power requirements of the loader appropriately then some interesting divergence of decisions can be made here.
For loading assemblers: not needed. I once blueprinted of the maximum speed assembler surrounded by speed max beacons and still was able to only need one inserter to load it (and unload it) fast enough to keep it producing non-stop.
Putting it on the tech tree: If this is a moderate-cost (in science packs) tech item then I see no issue with having it pwn inserters. After all, it can't replace inserters entirely as many have outlined that inserters are better than loaders in certain applications.
Keep them lubed? This is a wierd interpretation of the fuel mechanic: The only other items that have this mechanic is fueled items like locomotives, burning smelters, reactors, etc. I don't think a loader should be "fueled" by lubricant.

[Slayn25]

For loading assemblers: not needed. I once blueprinted of the maximum speed assembler surrounded by speed max beacons and still was able to only need one inserter to load it (and unload it) fast enough to keep it producing non-stop.

You obviously weren't producing Express Underground Belts lol.