Friday Facts #128 - Back down to earth

[malcmiller]

After seeing both Klonan and Kovarex's ideas, I think Kovarex's implementation of the loader is better. I think the loaders automatically buffering everything would make them do too many things at once, and i feel like if the player wants to have buffers they should have to actually build them. It's a chest that's filled by a belt that also functions just as a 2x1 chest which is very useful in it's own right. It seems like too much of an all in one machine that I think goes against what makes factorio good. No one machine does everything, and the best designs are when inserters, belts, splitters, chests and assemblers all come together to do something else. Kovarex's design keeps chests as a 1x1 entity and makes it so the loader (or chute I like that name!) is a bit more awkward to use and doesn't just end up adding buffers everywhere.

Also I like the idea that the loader should unload at full belt speed. Not unloading at full belt speed makes it seem less like a part of the belt system, and more like an inserter, which its not. But i think to do so it should have some cost that makes it a challenge to set up and properly harness its power. Which is why I am still on board the idea that some guys said that it should consume lube. The other alternatives to making it expensive like high research/build costs/power requirements are all things that a good factory can already handle easily, but lubrication requirements would be a totally new puzzle to work on. I like puzzles.

One idea I saw thrown around was that it should only work on raw materials or something isn't a very good idea. Every item in factorio is treated the same, from copper wire to rocket silos, and suddenly adding something that arbitrarily only works on one type of item really just seems out of place.

Most of all i want to say THANK YOU DEVELOPERS! I love this game and I really like that you give us that opportunity to express our thoughts on something before you implement it! Whatever we all say i trust you devs to make the right decisions.

[TimmPure]

I'm not that enthusiastic about the idea of a loader, but I'm well aware I probably represent just a small subset of Factorio players - let's say, the more 'harcore' ones. In the design/functionality that's in the FF post it seems like it would trivialize some content that I currently find enjoyable. But then again, optimizing throughput and achieving belt compression might be a nuisance to some, instead of a joy.

All in all I can understand if the idea does make it into the game, but I would receive it with mixed feelings.

[Degraine]

To be honest, I think people would be having the exact same reservations as if splitters were never in the game. You'd have to split using inserters! Its part of the fun of the game! Splitters just trivialise the logistic trouble of splitting a belt, which makes them overpowered, and they don't even need electricity? Splitters over powered, and they dont add anything we cant do already.

This is 100% correct. Someone asked on Reddit what the big deal was and I was tempted to reply 'because people don't like anything to be easier than it was when they first started playing'. Again, forgetting that bots utterly trash belts in convenience, throughput and even CPU usage. Bring on the loaders, I say.

[Fatmice]

The "hardcore" ones do not care about sphagetti belts nor belt balancing. We've debelt our huge factories a long time ago and are more concerned with optimizing logistic network layout, and beacons layout. This will just be another one of those cute toy to add to the arsenal of making the factory churn out even more stuff. "We" do not care about the strawmans arguments that have been raised multiple time in this thread, i.e. the loader/unloader reduce complexity of belts, inserters, and splitters. What complexity exactly? Inserters are still needed for the production aspects and people have it wrong when they think the unloader/loader can interact directly with crafting entities. What about belts and splitters? There is nothing complex inherent in them. Even the most obscene sphagetti mess of the 16x16 balancer is, which is really a waste of space and resources, quite simple in operation and utterly useless with logistic robots.

So in reality, the "hardcore" ones are the silent majority here. For the most part, we are busy making, fixing mods, and playing with them in our own factories. It is somewhat bemusing to watch the going on about nothing in here.

P.S. Nice to hear you're still around, Degraine.

[legleg]

i think a belt loader would be a great idea personally . limit it to working with chests and rail carriages and i dont think itd be OP at all. i use chests with inserters all the time as a buffering solution for when im producing more than im using and to soak up unexpected extra usage . if this isnt implemented in the game then i hope someone makes it as a mod. theres never too many toys to play with

[bobingabout]

I love the idea of the loader, I mean, the Mining machines have an un-loader built in already.

But yeah, Go tiered, so you have Yellow (standard), Red (fast) and Blue (Express) versions, that can shovel items at the same speed as the equivalent belt.

I also think using them to load machines isn't such a bad idea... Think about it, when loading a machine with a robot, a robot loads a chest, and an inserter can move 5 items from that chest to the machine at a time. therefore, to make Loaders viable in the same situation, you would need to use a loader to load a chest, then use an inserter to load the machine. Otherwise using an inserter from belt to machine would be slow. If you instead go from a belt, straight to the machine, it can potentially load as fast as the belt can move items, potentially out-performing the current robot method. If you then also allow it to move items from 1 chest to another, or from a chest to a machine, then it effectively takes the role of an inserter, and works as well for robots as it does for belts.

Basically, I see a loader as an upgraded inserter, therefore it still needs a throughput limit.

Ideas for graphics... do something similar to the underground belt. split it up into 4 parts. Firstly, you'd have an input side, and output side, each taking one tile. the input tile would then again be split in two, the first half looking like some sort of frame connected to the chest, the second basically being the belt. Do the same on the output side, so there's approximately one tile worth of belt in the middle. The input/output frame graphic may change depending on what it is connected to, so if it is connected to a belt, it would look more like a belt with a bit of a frame on it, if it's connected to a chest, or a machine, it would look more like the underground belt, with a canopy.

[Lallante]

The belt loader/unloader is too powerful alone. Some suggestions for balancing:

- Make it 3x3 (its too compact at present) and require at least 6x inserters of the relevant tier (as well as some higher tier components) as input.
- Make it require "fuel" of some sort - suggest coal for the yellow belt tier, heavy oil for the red belt and a new fuel processed from light oil and lubricant for the blue belt?
- Make it use a craptonne of power
- potentially split up loading and unloading and make the latter higher tier / more expensive.
- make it NOT completely compress the belt. Maybe fill to 80% max compression. Leave a slight advantage for inserter setups (or complex setups using both methods together)
- Do not allow them to load or unload machines (or trains?), only containers. Keep many use-cases for inserters.
- Potentially add a fourth tier after blue-belt that allows a customisable throughput

[SHiRKiT]

I really missed something. This is a 27 pages topic in 3 days.

[FlyHigh]

So in reality, the "hardcore" ones are the silent majority here. For the most part, we are busy making, fixing mods, and playing with them in our own factories. It is somewhat bemusing to watch the going on about nothing in here.

On a side note, It is rather curious that we keep asking for devises to flatline the learning curve and at the same time fill our games with mods that increase that same complexity curve.
It kinda reminds me of EA's fiasco games for the last... decade.

[Nemoricus]

I find the degree of controversy around loaders and unloaders to be interesting, given that I see a lot of differences between them and inserters that make each distinct in usage. Now, there are some assumptions that I'm making about how they'll work:

• They can only load/unload from chests and maybe trains.
• They require a belt in-line with them to operate

Now, for the differences between loaders/unloaders and inserters:

• Footprint: Loaders have a 2x1 footprint, while inserters are 1x1. This is already a significant difference, as it means that more compact designs will use inserters, while loaders need a little more space. When combined with a chest and an input/output belt, the effective footprints increase to 4x1 and 3x1 respectively, making the loader even more unwieldy in compact situations.
• Filtering: This is another significant difference, as a loader will simply move whatever it's given, while a smart inserter can take specific items, react to circuit network conditions, and so on. This is particularly pertinent for trains, since they might carrying mixed loads and you might only want to take out some items from the wagon, but not all. Unloaders can't pick and choose, while a smart inserter can.
• Feeding furnaces/assembling machines/non-train vehicles: This one's quite simple. The inserters can do it, but loaders can't.
• Speed of operation: This is the first place where a loader/unloader has a distinct advantage over inserters. Given a sufficient supply of items, an unloader can saturate a belt on its own, while several inserters are needed to do this and even then, it's a more involved design process.

The combination of these means that I believe that loaders/unloaders are most useful where high throughput is of paramount importance, while inserters are more practical in other situations.

To reinforce this distinction, I would like to see loaders/unloaders require the following:

• Electricity: This would help keep the choice between loaders and inserters from being too clear cut, as neither can operate without cost. A loader should have a higher idle power drain than an inserter, so that for lower volume operations, an inserter is more efficient on a kilowatt basis.
• Cost: Loaders should be modestly but not extravagantly more expensive on a resource basis than an inserter. This would reduce temptation to use loaders everywhere possible, while making potentially more cost effective in situations where their higher throughput can be utilized.
• Tiers: Like belts, loaders should come in yellow/red/blue tiers, so that they aren't a one-size fits all solution. If you want the ability to saturate an express belt, you're going to pay proportionately for it.
• Position in Research: Logistics 2 or 3, perhaps. Players don't have immediate need for high volume tools, and so it doesn't need to be available immediately. At the same time, it's not so game-changing as to be gated deep in the tech tree.

If given appropriate stats, I believe that loaders would be a useful element in the game, and while they streamline certain tasks, they do not remove challenge from the game. As has been mentioned elsewhere, the presence of loaders doesn't remove the need to design set-ups where their high throughput can be useful, nor does it provide the sinks for such high volumes of resources.

Additionally, figuring out how to achieve a task while using a limited set of Factorio's tools is part of the appeal of the game. Sure, sure, you can grab a loader and easily store the full output of an express belt, but can you do the same thing without using it?

Personally, I think I'd love to have these as control points for bulk resource flow. By inserting them at key points in my belt network, I could toggle them on and off to throttle production, while it might take multiple inserters to get the same degree of control. Anything that reduces the number of entities I need to wire in a circuit network is something I'm in favor of.

[osvert]

First of all, thank to the devs for an amazing game that made me happy for over 48 hours on my first week-end of play! I don't know how I could have missed the past 2 years
I am a new player, so don't take my answer as as important as an advanced player. Personally I don't like the idea of this "loaders" they feel unnatural, and I feel they would break the fun of creating the factory. For now I feel that balancing them in any way would seem artificial. I did not have time yet to play with trains and other vehicles, and maybe then, as a train station or car garage element, but that's it. Thanks. Great day to all!

[mideg]

Funny thing is, most people call the Loader/Unloader overpowered and require them to use extensive ressource to research and work with.

But then compare those to simple belts. They are incredibly powerful, extremly low-tech and cheap as anything.

They are just not overpowered because they were there first.

My point is: Yay for Loaders/Unloaders. They are technologically so simple they shouldn't be high-tiered or extremly expensive. Maybe introduce them at red-belt-level with faster version unlocking later. If you go by overpoweredness, you need to put belts far later in the tech tree.

[miturion]

I like those loaders/unloaders for train loading/unloading.

[RobertTerwilliger]

Loader vote:

NO

Because, as it is said millions times before - it replaces old solution. More of it, old working and ballanced solution.
Also gating it under some high-level techs will break factory into pre-loader and post-loader chunks, because of changing working setup is too bothering.

Also loader will have very limited real purpose - only storage (while IMO, allways when possible, resources should work rather than collect dust), train loading/unloading, transfer to logistic system. Ceratinly, considering inserters would fit assemblers more, while being more compact and not cutting the line (this loader's feature makes it even bigger, adding splitters)

Also what would we get by upgrading train station to loader-based? Really, what? All we get is reduction of train being at station. So my train which is staying, say, 20 seconds, will stay only 5, saving 15 seconds - is it worth intending new element? I don't think so.

Do we really need higly-specified basic element, while the whole Factorio spirit is making complex things using simple elements?

[orzelek]

Loader vote:

NO

Because, as it is said millions times before - it replaces old solution. More of it, old working and ballanced solution.
Also gating it under some high-level techs will break factory into pre-loader and post-loader chunks, because of changing working setup is too bothering.

Also loader will have very limited real purpose - only storage (while IMO, allways when possible, resources should work rather than collect dust), train loading/unloading, transfer to logistic system. Ceratinly, considering inserters would fit assemblers more, while being more compact and not cutting the line (this loader's feature makes it even bigger, adding splitters)

Also what would we get by upgrading train station to loader-based? Really, what? All we get is reduction of train being at station. So my train which is staying, say, 20 seconds, will stay only 5, saving 15 seconds - is it worth intending new element? I don't think so.

Do we really need higly-specified basic element, while the whole Factorio spirit is making complex things using simple elements?

Silly question but I'll ask: if you think it will be very limited and devs(and others) think that it will be useful - why vote no for something that others will find useful?

And whats so wrong with replacing the old solution - thats the topic that I see repeated here all over again. I do understand some people like to build huge inserter contraptions - noone is taking that away. For others let them use loader and focus on other things.

[nallar]

I like those loaders/unloaders for train loading/unloading.

Agreed. Train unloading looks silly, and is generally a pain.

[Art1985]

I would like to see automated distributor, something like splitter, but with option to select on what place of belts it will put items.
For example it would put iron on right side of right belt and so on (like with smart inserter)

[sahaka]

I like the idea of improving train usage but I don't think that the loader as shown is the best way forward. I think there is a strong argument for implementing a loader type device when combined with a sort of silo. Someone earlier in the thread mentioned hoppers and feeders and I think that is a good direction.

I envision that a silo can be a large storage device similar in size or slightly larger than a liquid tank and it can hold large quantities of one material. The second part to this would be a train car that works just like the silo. This would be perfect for handling raw materials. The only way to get materials in and out is with a loader type device(which only works with silos/train silos) that can transport the goods rather quickly so that trains and mines can operate without needing long rows of chests and inserters. While I see this working really well with raw materials such as iron and copper ore, coal, and stone, I am not sure how to balance that with finished or intermediate products. This concept could also be applied to oil and liquid transport on trains where a tanker would line up to a pump that fills the tank car with the fluid.

Raw materials like ore are normally stacked in piles as shown by the hopper/feeder, but finished and intermediate products like gears or circuits would not normally be just tossed in a pile. I think that there could be a warehouse type block that can be sorted and organized. It would need to be rather large both in physical size and storage size and would allow for logistics organization and sorting.

Right now, at least for my friends and I, once we get to logistic robots we start to just have arrays of chests that are a mess of materials. The use of a sortable warehouse would allow for a more organized factory. I personally dislike that once you hit the robotic endgame you stop really having to use belts, inserters, or trains. So anything that can encourage and increase the use of trains is something I would love to see.

[Venatos]

im a little on the fence about the "Loader", i realy want it for high throughput cases where you would need 8? fast inserters to max a expressbelt.
but trains can allready be full loaded and unloaded in 10seconds, bringing this further down to like 5seconds would realy bother me for some reason.... idk. just feels wrong.
if its a box to belt exclusive, im all for it.

[Koub]

Silly question but I'll ask: if you think it will be very limited and devs(and others) think that it will be useful - why vote no for something that others will find useful?

And whats so wrong with replacing the old solution - thats the topic that I see repeated here all over again. I do understand some people like to build huge inserter contraptions - noone is taking that away. For others let them use loader and focus on other things.

Human nature is such that most (not all but most) will take the easiest option when they are given the choice, even if it's detrimental to the experience of the game. As a (kind of) proof, I'll give :
- The flourishing business of goldsellers
- The ever renewed success of sites with cheats/exploitable glitches
- The regular bans for exploiting bugs in online games
- The intemporal success of sites of walkthroughs
- ...

Lots of people make the choice of easy result if they have it. You may say "well it's their choice". I'd say "game designer knows best what's the game experience he wants to provide". Give people a "win" button, and you'd be surprised how many use it, and without realizing it, ruin their experience of the game.

That's why, in my opinion, the devs and game designers have to think twice before implementing and adding something that could unbalance the game even if some people ask for it. I don't favor the hardcore extremists who only want what's the hardest, who only play with nonexistant ressources and maximum biters, and find that too easy so theu do it screen shut down, sitting on piled glass. But I don't thing intoducing too easy things would make any good to the game either.

That's why I think the kind of things I (along with many other people I won't enumerate) suggested would be appropriate, adding a legitimate functionality, without f*cking-up the game balance.