Friday Facts #128 - Back down to earth

[Odhan]

What if the concept of the loader was altered slightly so it wasn't something that sat between a belt and a container but was a container of sorts in itself? It would be able to accept content from a belt and store it like a silo or something and also spit the contents out onto a belt. Pickers could still interact with it.

[kaze47]

the first time I played the game I thought that the belt can insert items into the chest, but no

I do not see the difference between this:
1. belt->loader/unloader->chest->loader/unloader->belt
2. belt->belt->belt->belt->belt
where is OP? speed is the same,
but loader/unloader it should only work with chest

cool solution in multiplayer when sending items to friend, if you have two factories (you can create post office )

[Klonan]

What if the concept of the loader was altered slightly so it wasn't something that sat between a belt and a container but was a container of sorts in itself? It would be able to accept content from a belt and store it like a silo or something and also spit the contents out onto a belt. Pickers could still interact with it.

Like this?
viewtopic.php?f=38&t=20847&start=230#p131966

[Marsbar]

What if the concept of the loader was altered slightly so it wasn't something that sat between a belt and a container but was a container of sorts in itself? It would be able to accept content from a belt and store it like a silo or something and also spit the contents out onto a belt. Pickers could still interact with it.

Like this?
viewtopic.php?f=38&t=20847&start=230#p131966

This seems like a good balance. You don't get the super crazy fast chest loading, and the loader helpfulness would scale with the inserter stack bonus.

Even the speed thing wouldn't be that big of a deal, especially if there were tiers (yellow, red, blue). It would allow for a red loader to fully saturate a yellow belt, and a blue loader to come close to full saturation on a red belt. This doesn't seem OP to me. I know my main use for this would be to dump items into logistics chests anyways.

It seems like the loader is basically providing a compressed chunk of belt. It has the ability to store the same amount of material as X tiles of belt, and the only advantage it has is the inserter stack bonus. Seems fair to me, and (in my opinion) would greatly add to the game.

[ratchetfreak]

the first time I played the game I thought that the belt can insert items into the chest, but no

I do not see the difference between this:
1. belt->loader/unloader->chest->loader/unloader->belt
2. belt->belt->belt->belt->belt
where is OP? speed is the same,
but loader/unloader it should only work with chest

cool solution in multiplayer when sending items to friend, if you have two factories (you can create post office )

the chest can hold a lot of items (16-48 stacks depending on which chest you use) while a belt can only hold a dozen individual items. having a chest inline like that is a very easy buffer while the current buffers need a lot of inserters or a bypass point to be able to get full throughput.

[Zeblote]

the chest can hold a lot of items (16-48 stacks depending on which chest you use) while a belt can only hold a dozen individual items. having a chest inline like that is a very easy buffer while the current buffers need a lot of inserters or a bypass point to be able to get full throughput.

And how is adding new components that allow building better buffers without needing a lot of space a problem?

"adding x directly is much faster than building a huge workaround" is not an argument against adding them - you're explaining why we should have them.

[Glubo]

"I think people would be having the exact same reservations as if splitters were never in the game." EDIT: (I think this nails it on the head)

As a person who actually remembers Factorio before splitters I can't hold my reaction. I exactly remember my mixed feeling from splitters because it made obsolete many hours of developing my own workarounds how to split/balance belts without splitters, but after few days (maybe weeks) I accepted them and my feeling of their OPesness (they really felt OP at that time for me) perished under pressure of fact that they made possible much more sane builds.

Same volatile feeling of OP was there for me with removal of 180 degree bent penalization on belts, removal of evil central position on belts, even introduction of underground belts. So I say go for loaders, at least some sort of hooper feeders for trains. Volatile feeling of their OP power will fade away and after few years there would be someone arguing with "I think people would be having the exact same reservations as if loaders were never in the game."

[sillyfly]

Posts: 1
Joined: Sat May 25, 2013 11:15 pm

You, sir, have been waiting a long time for this one first post. Congrats

[Prootje]

The loader idea is cool and only changes some gimmicky gameplay. As it's working in the graphic it's a bit overpower though. I would make it 3x2 so you can't place them perfect inline with conveyor belts. This would make it a powerfull item to load/unload chests and trains, but it would make it less suited for mass storage purposes.

[Ext3h]

Loaders including the chest itself feels like spot on.

Question would then just be, how many stacks the internal buffer should have. Not too many, that's sure.

2 slots? So that, even if two item types are mixed, one can unroll to the belt while the other item type is getting refilled. This avoids a buffer underun caused by switching item stacks.

More feel unneeded, you can still chain chests if you like.

Peak unload speed from/to a 2x1 loader/unloader ist 5 blue inserters, each topping out at 10-12 items/second, for a total of 50-60 items per second. More realistic is 4 inserters, resulting in 40-48 items per second. However, you can put any number of chests in between, to boost the buffer capacity.

Hrm, that's 2.5k / minute for each pair. That's a bit too much. A full blue belt with a single loader, and with 2 item stacks, that's still two seconds of buffering for free.




No, I don't think 2x1 is the correct size, it's too easy to tune for it USING the size as an advantage. 1x1 actually sounds more reasonable under that aspect.

This is how I expect 1x1-loaders to be commonly used:

Code: Select all

...=L-C     C-C-C-L=...
    | |     | | | |
    C-L=...=L-C-C-C
    | |     | | | |
...=L-C     C-C-C-L=...
    | |     | | | |
    C-L=...=L-C-C-C

Scaling for storage looks quite nicely, it can scale both in width and height, and the motivation to make use of all 3 sides is decent.

In contrast, with a 2x1, the buffer would scale differently:

Code: Select all

...=LL
    ||
    CC
    ||
    LL=...
    ||
    CC
    ||
...=LL

Code: Select all

...=LL
    ||
    CC
    ||
    CC
    ||
    LL=...
    ||
    CC
    ||
    CC
    ||
...=LL

So the 1x1 loader actually promotes a more mesh like layout, while the 2x1 loader, due to its broad side, primarily promotes a wide spread layout. Since 4 inserters are sufficient for blue belt saturation, there is no motivation to grow in the other direction.

Using a dedicated loader without integrated pickup buffer of its own leads to a similar storage pattern as the fully integrated 1x1. Even though I see issues with factory layout due to the effective 1x3 space requirement, so you can't just build a compact checkerboard factory. Even direct loading to the factory (if allowed) isn't working properly.

That is a problem since it makes the belt solution still unattractive compared to the robot or inserter+chest transport method which allows for a 66% space efficiency. With the 1x1 loader, I can still think of a competitive 50% space efficiency solution.

For train stations, the integrated 1x1 and the dedicated 1x2+1 don't differ much, apart from two additional spaces. Whereby the dedicated one even gives proper buffering "for free" on top, which the 1x1 can achieve at the same space requirement optionally.


As for the cost: An electric engine unit should definitely be a component. Maybe even more than one, given that it easily replaces a dozen robots or so, in a tight layout. Not too much though, as it should become affordable as soon as the player COULD switch to bots, or the player will inevitably switch the production line over and give up on the belt infrastructure way too soon.

And as for naming?
"Loader" doesn't sound right. It's actually more of a "belt compressor", if you get what I mean.
Because that's all it does, it's not actually capable of "loading" anything, it's just compressing item stacks, nothing more, nothing less.

[Zeblote]

Including a chest ruins the whole purpose - quickly getting stuff into/out of whatever container. That could be a chest, a train, an assembler making copper wires, etc.

Obviously, you can also connect both a loader and an unloader to the same container.

[R3vo]

Instead of the loader, I'd rather see some upgrades to inserters, maybe new ones, or upgrades via research.

The research could for example improve the speed and add the ability to pick up multiple items from a belt at once

[seronis]

Would love this. Would need to keep the 1x2 footprint or its worthless to me. For balance reasons it would be fair to require one side to be a belt. Also it should be 10-25% less efficient electrically PER ITEM MOVED. So if its moving 4x as many items per second as a fast inserter it needs to consume 4.4 - 5x as many watts.

After grudgingly reading 25 of the 32 pages in this discussion I also noticed that I did not see one single reason AGAINST them that I consider valid. Just a lot of prejudice that would be solved by "dont build them if you dont like them". But to appeal to these people I think it would be acceptable to add the loader/unloader in a micro mod that is disabled by default so the dissenters can ignore its existence.

[vaderciya]

I don't know about the loaders, I see both good and bad things. Perhaps I'd feel differently if they weren't being brought up so late in development?

[Zeblote]

I don't know about the loaders, I see both good and bad things. Perhaps I'd feel differently if they weren't being brought up so late in development?

What are you talking about? The game is still in alpha. Everything can change.

[Aikonn]

Loaders no. Too OP.

I would rather suggest:
1. remove stack bonus for bots
2. some kind of "highway/monster belt"
- 2x2 pieces ... clunky belts
- 8 lanes (items could be scaled down or overlapped)
- inserters would use stack bonus with these belts

bots will be scalpel. belts will be hammer. Both will be usefull.

[MeduSalem]

The game is still in alpha. Everything can change.

Probably not everything. They invested a lot of time and resources to develop the game into a certain direction so they won't fundamentally change some of the things anymore but try to add features that don't exist yet or compliment/balance existing ones.

[Darthlawsuit]

I don't think it should work with items but work with raw materials where they are abstractly represented, picking up a single piece of ore(lol). Like quickly unloading an ore train into a hopper then unloading the hopper as needed. I don't think they should use chests or crates but a hopper that can only be accessed loaders/unloaders. Should also take up a lot of space due to elevating material and storing a lot of material.

That way it increases throughput but also increases the footprint of the buildings. So you can get more compact with inserters or take up more space but have more throughput with loaders/unloaders.

[Ext3h]

That way it increases throughput but also increases the footprint of the buildings. So you can get more compact with inserters or take up more space but have more throughput with loaders/unloaders.

Except that the footprint for a transport belt solution is already bad enough as it is, part of the reason why casual players prefer robot + inserter based production fields, rather than belt based production lines. Plus you don't want to force players to rebuild their entire production line once they research stack size bonus, which would be required if the loader had the wrong scale.

Belts still have the added disadvantage of complex routing, especially for recipes with 3+ components, so the end points might as well be compact - that advantage is still annihilated by the requirement of having feeding lines for multiple materials, plus you still need inserters to actually load and unload materials into and from factories.

Besides, making it too big on its own, means that you will never need to scale in number - whereby the compact solution might as well require to feed a single factory with 2+ lines and the corresponding number of belt compressors, in order to get past the 12 items/s limit a regular blue inserter still has. There is no need to artificially turn the belt compressor into a rate limited choke point, if the size is sufficiently small, that happens automatically be the number of accessible sides.

[jaworeq]

Plus you don't want to force players to rebuild their entire production line once they research stack size bonus, which would be required if the loader had the wrong scale.

You already force players to rebuilt once they hit electric furnaces. And when they get robots. As much as you want to avoid it, game already includes a lot of rebuilding.