Improving Factorio's Mining Endgame: A Programmable Bucket Chain Excavator

[XNK]

I've played this game for several hundred hours and it's clear to me (and others) that Factorio's endgame lacks a mining solution as simplistic and beautiful as the other endgame features.

While by the endgame nearly everything is fully automated, establishing new resource outposts is decidedly not. Even with clever blueprints it's inevitable that each new resource patch will require some degree of tailoring, falling aggravatingly outside the world of high-level automation designed by the player up to that point.

This is not to say that something needs to change fundamentally about Factorio--rather, there is a single endgame solution which remedies all of these problems in glorious Factorio fashion. It's not a Bucket Wheel Excavator (BWE).

I'm aware of mods which add BWEs to Factorio, and I've seen that BWEs have been suggested on the forums as a strong candidate for strengthening Factorio's late-game--I have yet to see BCEs in either. I agree with the general sentiments for BWEs (enabling higher level automation, reducing UPS and monotony, pushing Factorio's theme and endgame enjoyability forward) but the point of this post is to redirect that idea to make Factorio's endgame mining as

sexy

as possible. This will be a little long since it's a rather large potential addition to the game.

First, addressing BWEs, and why they (rightfully) don't belong in the game: BWEs are awkward in Factorio from an implementation perspective. Ore patches don't lend themselves to excavation from a stationary rotating arm.

• If the BWE can move, it's more awkward--how should it move around as it excavates a patch?

• If it's not rail-mounted, automated pathing will eat into any CPU resource gains--one of the points of this proposal.

• If it is rail mounted, the digging pattern will be unnatural, trying to make circles out of Factorio's grid or else dance on the rail strangely to try to mine in straight lines.

• Or, giving up on all this, the mining area could simply be reduced to a square, which isn't a solution satisfying enough to deserve implementation.

Fear not: BCEs solve all these peculiarities and go beyond to launch Factorio's endgame into truly exciting territory, and are more than worthy of being in vanilla.

What?

A programmable, rail-mounted, Bucket Chain Excavator (BCE)--effectively, a mobile mining outpost. But this is no lazy endgame solution--it's a smart one. BCEs mine everything on the ground, from the base of the superstructure all the way out to the tip of the boom, all at once. In Factorio this means it'll mine neat rectangular strips. Mounting the BCE on rails provides the dynamic solution we need, reduces the complexity of vehicle pathing, provides a path to total automation, and meshes with and enhances Factorio's existing train dogma and aesthetic. And it's beautiful. Devastating Nauvis with gas-guzzling, thousand-ton, automated rail-riding mechanical strip-mining behemoths is so Factorio I'm at a loss for words.

For your comparison. Bucket Chain Excavator (BCE):

Yes. (Note that in Factorio it should be rail-mounted.)

Bucket Wheel Excavator (BWE):

No.

A hasty sketch of the BCE as to be used in Factorio.

BCE.png (33.07 KiB) Viewed 5554 times

Why?

• Reduce UPS and monotony. A single venture-capitalist-size BCE with a 4m by 50m bucket chain (the excavating bit) and a counterweight could easily take the place of vast fields of copy-paste electric miners of olden times. Impressive. Big. Sexy. The BCE would also be the dirtiest of polluters.
  1/s per tile (assuming it pollutes ~ the same per tile as electric miners) * 4 * 50 = 200/s = 12k/m max unmoduled.
  Nice. The gross pollution will add interesting challenge by inviting all the biters over at once. How will you protect your new assets? Spidertron legions?

• Automate. The BCE would function just like a train station as a programmable destination, permit trains to pass straight through, and allow fluid trains to be more useful by supplying sulfuric acid to the BCE. With the new station limit coming out, combinator logic should be built-in, allowing toggling the BCE's limit by current storage load, etc. Cargo trains would carry away mined ore. Logistics robots (e.g. in Spidertron units) or trains could resupply the BCE with fuel and ammo.

• AUTOMATE HARDER. The BCE needs to be programmable. Place it on rails, tell it to go forward mining one or more select ores. It's conceivable one could also direct it around to different ore patches via the map, similar to the Spidertron remote. Expanding your production with the BCE would be as simple as adding rails. The BCE should also include radar and artillery, so you may access it no matter how deep it's gone into the wilderness.

• Pushing Factorio forward. In the real world, BWEs are mining superstars, while BCEs are mostly used for resurfacing.
  But Factorio isn't real life.
  A BCE is the optimal solution for Factorio, where all ores are sitting on the surface. Plus, dragging the bucket chain over sprawling patches like a greedy
  bastard
  (while absolutely devastating the environment) is more satisfying, efficient in scale, and less complex than BWEs. A rail-mounted BCE would support Factorio's central thematic elements and reduce complexity in the endgame: simpler is better. There's no reason we need to build and manage tons of outposts. Fundamentally, it also necessitates the player has an established, high-functioning train network, requiring them to work for the reward rather than passively receive it.

A how-to, so you start imagining the wealth of possibilities opened by this singular addition:

• Set up outbound parallel tracks at most 100m apart for 100% coverage.

• Place your new BCEs on each. Set them to drive forward, mining only one ore type each unless you enjoy the sadism of mixed ore mining.

• Enable station limit by your preferred train load size. Name your BCEs as you would your outposts and set up your train network.

• Ensure you have a one-way return rail connected to all your mining lines, so outbound trains may return.

• That's it, you're done. You can sit back and watch as your factory runs and the ores flow unceasing from your strip mines, giant bucket chains mowing over fields of sparkling ore, reeling in the good stuff. If you need more ore simply continue the rails. Obviously, there will be accompanying challenges, but the tedium of establishing outposts will be gone forever.

If you aren't convinced yet that this fits and enhances Factorio's theme, tone, and style; generally furthers the game's artistic mastery; and adds something crucial to the vanilla experience, let me know.

Details?

More (entirely suggestive) technical details, for the interested.

BCE
Fuel powered
Six (6) to Eight (8) module slots
Late-game tech
Subject to mining productivity bonuses
Preliminary base (unmoduled) stats estimates:

Size: enormous. 50m boom, 50m (?) counterweight. Probably 4-6 train cars in length to accommodate parking trains.
Storage: 50,000 ore units = 1,000 slots (equals 25 cargo wagons). Yes, it's absurd. It's also a behemoth machine larger than a house, and it digs like nobody's business.

Weight: around 5,000,000. It appears that Factorio's weight is measured (unofficially) in kilograms--which produces reasonable results, such that a locomotive weighs ~2 tonnes--though it's not specified anywhere I've seen (even the wiki). If indeed measured in kilos, this figure (5,000,000) correlates with reality, where BCEs can easily weigh around 5,000 tonnes. Side note: IMO it's unrealistic that the weight doesn't change between loaded/unloaded when there are 50,000 units of ore in storage, but that's up to the devs to decide.

Acceleration power (i.e. fuel consumption while moving): 50,000 kW. This is genuinely heavy machinery--it needs a lot of power. The power:weight ratio (0.01) is far lower than a locomotive (600 kW / 2000 = 0.3), as it should be, and the moving consumption is also considerably more than the digging consumption, as it should be. The fuel consumption serves to temper how badass it is. Consider that Nuclear fuel would last for just 24.2s of moving (i.e. relocating to a new ore patch/strip). For Rocket fuel, 2s. Three full stacks net you ~1m and 3m respectively. Even better, the intense fuel consumption (esp. if using Nuclear) provides impetus for further resource collection, leading to the cycle of usage seen everywhere else in the game and potentially opening new areas of development in the late-game (particularly with uranium/rocket fuel). Since the player effectively offloads the mining duty previously borne by electric miners to gas-guzzling machines, fuel will become even more important than it is currently.

Fuel consumption while digging: 2,000 kW max. Adjusts based upon actual digging area. This is derived as follows: 10kW per tile * 4 * 50 = 2000 kW max. 10kW per tile matches power consumption per tile of electric miners. (90kW / 9 tiles = 10kW per tile.) This means you need one Nuclear fuel changed out every 605s (10m) of mining.

Excavation width: 4m
Excavation length: 50m. It really should be this long--as an endgame tech it should revolutionize mining, switching completely to the large-scale, rail-based format that already happens with outposts.
Excavation rate: 2/s per tile * 4 * 50 = 400/s max. Even when mining strips only 4m wide at a time, with a 2/s production rate per tile (fourfold as fast as unmoduled electric miners; an appreciable upgrade) we find a base production rate of 400/s. This would fill a 50k ore hopper in just over two minutes (125s), without incorporating modules.

Pollution: 1/s per tile (whether it should pollute more or less while aggressively strip-mining the ground is open to argument) * 4 * 50 = 200/s = 12k/m max unmoduled. With modules this will change drastically.

If you're curious how the BCE and another train could both run on the same rails, fear not. With rail-mounted ramps at either end of the BCE, trains could access a loading platform inside the BCE without hindering the BCE's own locomotion. For more technical details, the BCE would always align itself to the tile grid when stopping (mining), meaning the boom will remain perpendicular to the rail while mining. The BCE could not be allowed to mine on curved or angled rails--only to travel over them. The BCE would also need to zone itself by acting as two moving rail signals. The BCE would only mine while stopped. Invisible inserters could load parked trains. There's also the challenge of what happens when a BCE that has an inbound train suddenly starts moving--this situation is basically identical to the station limit mid-route-state-switch scenario, and could likely be handled in the same way.

I'm also interested in the slightly radical possibility of throwing random stone ore (say, 20% chance on each mined ore) into the BCE's mining output since it's scraping the surface so aggressively. This would add complexity in a simple way by forcing players to deal with mixed ore.

There are plenty of minutia to discuss. But first, I'd like to hear what you think--do you see this bringing an added layer of depth to Factorio's endgame? Do you see this contributing highly to the vanilla experience?

[conn11]

I like this idea. Outposting is, like setting up more smelting areas repetitive especially in the end game. The latter being more easily semi automatable.
The whole idea reminds me, though differently implemented, of the Mining Drones mod by Klonan. Breaking monotony and gaining UPS are certainly beneficial.
The entity would be fitting the general art style. Probably rather than using fuel, a to be fueled integrated nuclear generator (without the UPS costly fluid dynamics) could be used. This would put it in the late-game territory.

A point of critique, besides the practicability of the general implementation, though automating mining, the outpost defences will still have to be built the current way (if needed).
Secondly if the BCE/BWE can be balanced such, that mining yield and throughput to trains can be similar to late game mining outpost. Considering highly beaconed builds of DI mining to smelting, high tier of mining productivity and with big practically everlasting ore patches.
edit: typos

[Amarula]

This would be great, either vanilla or a mod.

[conn11]

Addendum to my first comment:
I'dont think a fixed length chain/wheel extractor is necessary. IMO the core concept of this idea (having a more complex rather repetetice late game mining solution) would be better served, if the extraction arm is dynamically lengthend or shortend depending on ore patch size. Any BWE/BCE on a train track should deploy the extractor arm if it is under a certain threshold distance to the ore patch. Lore wise this could be explained e.g. with a collapsing extractor arm.

I find OPs idea of having a stone contamination interesting, this could be done as research/ additional minig recipe with higher yield but generation of the aforementioned byproduct. Though the late game could be missing a stone sink. Maybe some sort of minined patch fill (like landfill) should be needed to use/walk the terrain of the depleted parts of the ore patch, after the BWE/BCE Juggernaut is done (though with high mining productivity this could take a while).

[XNK]

A point of critique, besides the practicability of the general implementation, though automating mining, the outpost defences will still have to be built the current way (if needed).
Secondly if the BCE/BWE can be balanced such, that mining yield and throughput to traines can be similar to late game mining outpost. Considering highly beaconed builds of DI mining to smelting, high tier of mining productivity and with big practically everlasting ore patches.

Both very good points. For the first it's possible a BCE could be armed more generally, i.e. with gun/laser turrets, so that it becomes feasible it might be able to defend itself. (This could also be accomplished by providing it an equipment grid, though I'm not certain if it makes sense for mining machines.) Then there's the possibility of assigning Spidertrons to essentially escort the BCEs (perhaps the two devices could be linked) since Spidertrons can usually easily handle any sort of biter threat. I think it would also be rather cute.
Your second point is also a very important challenge. How it gets balanced precisely in-game will require a lot of careful tweaking--mining area, base extraction rate, number of module slots, etc. And like you say, decentralized smelting outposts are highly popular for their train throughput. It's possible, though it would be strange, that the BCE could feature onboard smelting that preprocesses ore before storing it away as plates. There are plenty of other questions that that brings up, though, like that uranium would need to be ignored...

[XNK]

I'dont think a fixed length chain/wheel extractor is necessary. IMO the core concept of this idea (having a more complex rather repetetice late game mining solution) would be better served, if the extraction arm is dynamically lengthend or shortend depending on ore patch size. Any BWE/BCE on a train track should deploy the extractor arm if it is under a certain threshold distance to the ore patch.

Though the late game could be missing a stone sink. Maybe some sort of minined patch fill (like landfill) should be needed to use/walk the terrain of the depleted parts of the ore patch, after the BWE/BCE Juggernaut is done (though with high mining productivity this could take a while).

Absolutely. And your second point was actually also something I thought about--the mined area would leave behind a special 'water' texture (like what they're doing with the new home screen) made to look like an excavated pit so it's uncrossable. Imagine if you could program it to dig out specific tiles, and if water could spread to these dug out tiles when connected--you could make artificial waterways in a much more satisfying way than other mod implementations (e.g. like waterfill). This would also add trench digging to factory defense, which would be rather powerful.

[conn11]

A point of critique, besides the practicability of the general implementation, though automating mining, the outpost defences will still have to be built the current way (if needed).
Secondly if the BCE/BWE can be balanced such, that mining yield and throughput to traines can be similar to late game mining outpost. Considering highly beaconed builds of DI mining to smelting, high tier of mining productivity and with big practically everlasting ore patches.

Both very good points. For the first it's possible a BCE could be armed more generally, i.e. with gun/laser turrets, so that it becomes feasible it might be able to defend itself. (This could also be accomplished by providing it an equipment grid, though I'm not certain if it makes sense for mining machines.) Then there's the possibility of assigning Spidertrons to essentially escort the BCEs (perhaps the two devices could be linked) since Spidertrons can usually easily handle any sort of biter threat. I think it would also be rather cute.

More concerning is protection of stacked or sheduled trains, having them destroyed can become quite annoying.

Your second point is also a very important challenge. How it gets balanced precisely in-game will require a lot of careful tweaking--mining area, base extraction rate, number of module slots, etc. And like you say, decentralized smelting outposts are highly popular for their train throughput. It's possible, though it would be strange, that the BCE could feature onboard smelting that preprocesses ore before storing it away as plates. There are plenty of other questions that that brings up, though, like that uranium would need to be ignored...

Having integrated smelting seems a little bit OP. Two possible solutions I can think of: Firstly this technology might be late game, but not super late game whe decentralized mining/smelting is much more prevalent.
Secondly mods like Nuclear Furnace (credit: darkfrei) have high throughput smelting. Having this as additional placeable wagon with DI from the BCE/BWE could solve the problem. Besides it can be inline with a nuclear powerd theme.

[conn11]

I'dont think a fixed length chain/wheel extractor is necessary. IMO the core concept of this idea (having a more complex rather repetetice late game mining solution) would be better served, if the extraction arm is dynamically lengthend or shortend depending on ore patch size. Any BWE/BCE on a train track should deploy the extractor arm if it is under a certain threshold distance to the ore patch.

Though the late game could be missing a stone sink. Maybe some sort of minined patch fill (like landfill) should be needed to use/walk the terrain of the depleted parts of the ore patch, after the BWE/BCE Juggernaut is done (though with high mining productivity this could take a while).

Absolutely. And your second point was actually also something I thought about--the mined area would leave behind a special 'water' texture (like what they're doing with the new home screen) made to look like an excavated pit so it's uncrossable. Imagine if you could program it to dig out specific tiles, and if water could spread to these dug out tiles when connected--you could make artificial waterways in a much more satisfying way than other mod implementations (e.g. like waterfill). This would also add trench digging to factory defense, which would be rather powerful.

Or at least allow for shallow water forming, preventing buildings beeing placed, while not blocking biter movement. This would give this alredy implemented terrain some use.

[ssilk]

The bucket chain excavator idea is really old ( viewtopic.php?f=6&t=1980 ), but this is nicely worked out.

I’m nearly sure this can be done as mod.

[Koub]

https://mods.factorio.com/mod/BigDrills ?
Not quite OP's suggestion, but I admit I'd love to see this in vanilla Factorio .

[XNK]

The bucket chain excavator idea is really old ( viewtopic.php?f=6&t=1980 ), but this is nicely worked out.

I’m nearly sure this can be done as mod.

Yea, I've seen it around a few times. The most important difference here was to ensure that the concept of a 'mobile miner' aligns with Factorio's core gameplay as closely as possible. Other concepts like mobile mining trucks, robot/worker depots, or simply larger miners never really aligned with those principles, at least to me. I.e. encouraging continuous development/progression of infrastructure (by necessitating an advanced rail system), increasingly advanced manufacturing processes (by demanding inordinate power in the form of fuel), higher level automation (by rolling train stops and mining into one intelligent device), and of course wanton environmental destruction. The BCE wouldn't be a 'shortcut' like most mods tend toward--I doubt new players could get it to work on their first playthrough, simply because of the need to actually figure out Factorio's train system, which would make it all the more rewarding. It would be a legitimate progression element, with many accompanying, genuinely novel problems the player would have to deal with. And, of course, I would be perfectly happy to see this as a mod first, particularly so it can get playtested and optimized, but getting it into vanilla would really change stuff up for the better in the late game.

[XNK]

I'dont think a fixed length chain/wheel extractor is necessary. IMO the core concept of this idea (having a more complex rather repetetice late game mining solution) would be better served, if the extraction arm is dynamically lengthend or shortend depending on ore patch size. Any BWE/BCE on a train track should deploy the extractor arm if it is under a certain threshold distance to the ore patch.

Though the late game could be missing a stone sink. Maybe some sort of minined patch fill (like landfill) should be needed to use/walk the terrain of the depleted parts of the ore patch, after the BWE/BCE Juggernaut is done (though with high mining productivity this could take a while).

Absolutely. And your second point was actually also something I thought about--the mined area would leave behind a special 'water' texture (like what they're doing with the new home screen) made to look like an excavated pit so it's uncrossable. Imagine if you could program it to dig out specific tiles, and if water could spread to these dug out tiles when connected--you could make artificial waterways in a much more satisfying way than other mod implementations (e.g. like waterfill). This would also add trench digging to factory defense, which would be rather powerful.

Or at least allow for shallow water forming, preventing buildings beeing placed, while not blocking biter movement. This would give this alredy implemented terrain some use.

Exactly. It also adds a surprisingly large degree of aesthetic freedom to the game, in terms of making artificial lakes/rivers/canals/etc, in a very simple way. I would thoroughly enjoy being able to add decorative waterways to my factory, or even functional ones (i.e. for pumping).

More concerning is protection of stacked or sheduled trains, having them destroyed can become quite annoying.

Yes. The player will need to be (or become) competent at train networking to ensure trains don't back up along mining lines, or you'll quickly end up in a terribly awkward position wrt biters. I mentioned the upcoming station limit feature will help greatly with this (networking), but designing an optimized system will be both difficult and highly rewarding once finished--it's undeniable it would replace the use of miners as entirely as possible, once players figure it out. I see this as very much akin to the transition from burner miners (whose inputs are tedious to try to automate large-scale) to electric miners. Electric miners, whose outputs are tedious to try to automate large-scale, would be subsumed by BCEs, which may be automated in every facet, large-scale.

[NotRexButCaesar]

Yes. The player will need to be (or become) competent at train networking to ensure trains don't back up along mining lines, or you'll quickly end up in a terribly awkward position wrt biters.

Or just wall everything in everywhere

[XNK]

Yes. The player will need to be (or become) competent at train networking to ensure trains don't back up along mining lines, or you'll quickly end up in a terribly awkward position wrt biters.

Or just wall everything in everywhere

For sure. There's the question of how the BCE reacts to placed buildings, trees/rocks, etc. It starts pretty simple--basically just a start and end length for an angled, adjustable bucket chain (e.g. see the original BCE photo) would be enough to get it 'over' a protective wall and stop short of power poles/roboports. It's possible it could ignore entities entirely but that doesn't seem fun. In my envisioning it would destroy trees, rocks, or anything else (nests lol) it drives into but would skip over player-placed buildings in the way (i.e. that it can't conform around) by lifting the chain up. And of course, it doesn't need to mine anything except ore patches and whatever might lie on top of them unless the player programs it to excavate a certain area, if that feature is incorporated. So we're basically talking about a boom with two variable-pitch points and variable length. Also, if power poles realistically want to be able to fit next to the tracks the boom would have to start pretty high off the ground (again, see the original photo). There's plenty of specific details that it would bring up, and I'm not denying it's a challenge since it really brings Factorio's quasi-3D experience under complete scrutiny.