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

[Hakusho]

This is the thing people are talking about BTW
I read 4 pages now and dont want future human beings to be as confused as me because nobody linked anything
(at least i did not see any link to the FFF until now)

[MeduSalem]

[...]

they would still be usable for train stations as you'd only need half the amount of inserters and the rebalancing setup can be simpler. In fact you can reuse existing splitting and merging designs with just 1 extra tile of extra width needed.

You are right, the thought appeared to me after I read Klonan's post in the FF #128 thread: viewtopic.php?p=131966#p131966

I particularly like the following approach somehow:

train station

The above design doesn't have chests, but it wouldn't look much different... There would be a chest before the Loader and a single Inserter would load the chest.


Somehow I start to like the thought of being able to use Inserters directly on Loaders due to them having internal stacks. It might actually make the item really useful in some cases, like for example the train loading.

With some further thoughts I even think that the Loader-item could be 1x1 instead of 1x2 and it wouldn't cause major imbalances to the game because of how you basically would need splitters EVERYWHERE to profit from the Loaders, which in return naturally increase the overhead compared to Inserter based setups. But making it 1x1 instead of 1x2 would also decrease their throughput again because 1 Fast inserter probably won't be enough to keep the Loader busy.

[koisama]

I like the idea of a loader with small internal buffer. It should only interact with belts, and you will still need an inserter to move items. Basically, it should be a part of the belt system that applies inserter stack bonus to belts.

[ratchetfreak]

I like the idea of a loader with small internal buffer. It should only interact with belts, and you will still need an inserter to move items. Basically, it should be a part of the belt system that applies inserter stack bonus to belts.

Just this will be a massive boost to belt/inserter interaction.

The core issue with belt/inserter interaction is the lack of the stack size bonus. Every other item transport can move stacks.

[Linosaurus]

There is no doubt, that 95% of inserters would still be used the same way, as it would be very impractical to use the loader on common places, like smelting setups, assembler setups, etc. To use the loader as input for assembling machine, you need to split the belt line, and use a loader, that is 2X1, using 1-3 inserters is clearly easier.

I'm starting to realize that this is very true. I wouldn't use them at all in the early-mid game.

Thinking about how much speed gain loaders would give in different situations with blue belts.
Current situation:

• 1x speed : fast inserter without stack bonus. This is enough for early-mid game.
• 1x speed: current belt-chest(s) buffer storage.
• 5x speed: inserter moving between containers later on.

Really awesome:

• 5x speed: replacing a belt+inserter(s) with belt+loader+chest+inserter(s). This is *awesome*. The constraint of loader+chest+belt in a line, would add *fun* complexity to layout. And not more poweful than current bot solutions.

Not overpowered, and fine with me:

• ~38x speed: one-chest buffer storage with loaders. This seems *really strong*. But... not useful enough to be overpowered.
• 5x speed: throughput of a big buffer with a chain of chests with loaders at each end. (of course you can have several side by side but then you need splitters etc)
• ~38x speed: direct connection to trains. Overpowered, but after my home train station it becomes sort of repetitive busywork for outposts. So I don't mind.
• up to 5x speed: direct (un)loading moderately boosted assemblers (if allowed). Balance wise okay. A bit sad to see less moving robot arms.

Problematic:

• ~38x speed: direct connection to logistic chests. This.... is bad. One big weakness of bots today is long distances, and this would make small logistics networks with belts in between them super duper convenient.
• Up to 38x speed: Direct feeding to and from a train wagon surrounded by assemblers. Possibly too strong. Because of this, and I like building my main train station, maybe instead provide a hopper to *fill* trains only.
• Over 5x speed: direct connection to assemblers (if allowed) with lots of speed boosts. Honestly I think it would be more fun and challenging if you were forced to move items into chests first, because getting the right layout of inserters adds complexity

Main opinion:
I like how they make highly boosted assemblers without bots feasible.
I don't like how they make some bot networks easier.

(minor edit: added *fun*)

[ssilk]

I think a loader is fine, cause the usages will be limited to very high throughput. As Kovarex already stated this will not be the case for most existing inserters.

As I already stated I see the splitter mainly for media-changes of the items.

From belts to trains. From trains to belts. From belts to logistic network ... etc. That are the points, where the loader enables to keep the high throughput without building dozens of splitters, inserters, chests and completely wired setups with hundreds of belts...
The other stuff we will see then if the loader exists. Nobody can know yet, how it can be used then - there will be always some interesting usage cases I think.

[MeduSalem]

The other stuff we will see then if the loader exists. Nobody can know yet, how it can be used then - there will be always some interesting usage cases I think.

At this point I would say, yeah let's bring them on in an experimental build and if they are completely hurting the gameplay they can be removed again or changed accordingly. There is no shame in trying something and admitting that it didn't work out as well as it seemed on the drawing board.

At least I would like to fiddle around a bit with them before I condemn them any further to be unnecessary or overpowered. That said I would like to see them implemented the way Klonan suggested in the FF thread.

Yeah I am having a nice day for once. Better use it before I change my opinion again.

[ssilk]

@MeduSalem: Exactly.

And I found out for me something interesting:
Unlike many other games (especially for example round-based games), Factorio is a game, that simulates some kind of physics steadily. This means: You don't have some state and - booom - next turn it is a different state. The state changes are quite, quite small each second a 60th part of change. Within a minute you have 3600 changes, over 200,000 per hour.
I explain this so deep, cause that means, that if someone would describe you a device like the splitter in text-form, everybody would say: So what, that is boring, it's just a stupid thing.

But in the game-reality this looks quite different and quite complex. It behaves much more complex, than on the paper (Or the forum. Or a FFF.) just because of this.
Just because of this steady change and that there are so many changes over time. Because effects of one change have impact on other changes. Even after a minute nobody can foresee, what happens, if I take away only one item from a belt.

What does that mean? It means, that the complexity of such a simple looking loader will have impacts on the gameplay, that nobody here can see or imagine. The effects of it will be there - eventually as described, but in the long run: What will happen?

The loader looks simple, and it looks as it's overpowered, but I would blow into MeduSalems horn: Let's put it into the game with as much power as possible and as simple as it could be at that time (for me also without the usual nice graphics, that can be done, when the loader goes final) and let's try it out for 1-3 month: Which is useful gameplay and which not. And if it is really overpowered, then this can and will simply be changed. It's more difficult to do it the other way around.

[That is also the way I would suggest to do some (not all) things for my customer. It's also called agile development.]

[Zeblote]

Give it fancy graphics right away, then players will like it more

[kingarthur]

i feel like the loader needs added in as a developer mod or opt in update/ patch and let people test it to see what affect it has on the game. since the fff was posted about the loader ive thought of like a dozen areas where id like to use it and a dozen more where it wouldn't do any better than what is already available. i`ve even thought about if the effort was put into it the loader could be used for everything and never need an inserter. it would require a lot of redesign on most setups and would be major overkill

[Joefesok]

Maybe the loader could only function if there is a loader on each end of a given conveyor, and the conveyor stops functioning if either loader loses power or whatever material would be used to keep it running.

[icanfly342]

I think inserters are an important part of Factorio and loaders would be a bad idea.

[Joefesok]

I think inserters are an important part of Factorio and loaders would be a bad idea.

Yeah, but the point of loaders is to serve alongside inserters. With their current balance they are overpowered but still serve a somewhat distinct purpose.

[katyal]

As long as loaders cannot interact with assemblers they are perfectly alright, imho. Oh! Make them require electricity to run! (not as balancing mechanism... 'cause being able to turn them on and off using the power switch would open up so many possibilities).

[MalcolmCooks]

I like the idea of a loader with small internal buffer. It should only interact with belts, and you will still need an inserter to move items. Basically, it should be a part of the belt system that applies inserter stack bonus to belts.

Just this will be a massive boost to belt/inserter interaction.

The core issue with belt/inserter interaction is the lack of the stack size bonus. Every other item transport can move stacks.

I like this idea as well, if loaders can't interact directly with chests, only with belts and inserters, but have their own internal storage. Then they act like hoppers. They need to be able to work with people's tendency to keep one item on a conveyer belt, though. They might be frustrating if they are just "dumb" machines that drop any items evenly onto conveyer belt, so I think the ability to filter which items go onto which lane of the conveyer belt would be useful, and justify a running cost of electricity. Running cost of lubricant is a stupid idea because everything else in the game that needs lubricant has it as an ingredient.

[aRatNamedSammy]

electric loader, good for train-to-chest , chest to train..... conveyor-to-chest , ect.. not allowed to filled assembly machine...
train station and buffers will benefit a lot from it...
late game tech, well, right after logistic 3 could be nice... but ill propose after logistic 2, fluid and petrol, need TONS-o-lub to craft loaders, but not to keep it running...
size ..not a small piece of equipment, of course..
one loader can do slight better job then 4-5 inserters on a wagon, at the price of electric spikes when running, more than 4-5 inserters..exemple, it can load/unload a full wagon in 10-15 seconds..

my little 2 cents!!

[kovarex]

I don't understand, why the main concern is the loader to assembling machine?
It doesn't make sense to use the loader for assembling machine in almost all cases, as you need to split the belt and use 2X1 entity that is actually quite hard to fit there.

I personally like the idea of loader having an integrated chest, but after some thoughts, I start to think, that the limitation is actually wrong, as having loader general entity->belt and belt->entity mechanism adds more possibilities without adding annoyance as you need to place two entities (inserter + loader) to do one thing. It is very very far from replacing inserters overall.

[Zeblote]

I personally like the idea of loader having an integrated chest, but after some thoughts, I start to think, that the limitation is actually wrong, as having loader general entity->belt and belt->entity mechanism adds more possibilities without adding annoyance as you need to place two entities (inserter + loader) to do one thing. It is very very far from replacing inserters overall.

Integrating a chest limits it way too much.. the whole point is to use them in place of inserters in some places.

Using them with assemblers would be really impractical most of the time. Why would you make this



Instead of this



If you don't actually need the throughput? I really can't see what's overpowered here.

[katyal]

You know after seeing a mock up I'm convinced.

[MeduSalem]

Just fooling around a bit...

1.png (507.02 KiB) Viewed 6958 times

The above makes me wish we had Smart Splitters... because then we might control the Smart Splitters depending on the stack contents of the Loaders behind them with Circuit Logic, effectively allowing to input 4 source items into the same Loader without jamming the 2 lane belt.

... and 2 more variants:

2.png (563.02 KiB) Viewed 6958 times

... here 2 other variants, the lower including the notorious underground belt cheat:

3.png (983.58 KiB) Viewed 6953 times

... and another one using the Underground Belt Cheat:

4.png (490.97 KiB) Viewed 6930 times

If there are filtered slots in the loaders it would turn out quite interesting with multiple source items etc.

So I would say... Loaders allow quite unique contraptions without really compromising Inserter usage.

Though I am still against allowing Loaders to feed directly into Assemblers/Furnaces. I am also against feeding directly into Train Wagons and I am against directly feeding into chests. There should always be Inserter(s) between the Loader and the Item(s) it is supposed to fill/empty for balancing and consistency reasons. That way you force the player to still think about what one is doing, it's already enough advantage that the Loaders will auto-balance both sides of the lane. And with (filtered) slots inside them and maybe circuit network integration/feedback they would offer additional qualities in various usage scenarios.