Modded Minecraft
Modded Minecraft is one of the more interesting (and grindy) entries in the factory builder genre. Building on the base game, it offers a tremendous amount of tasks and a fairly broad tech tree.
You’re tasked with:
- Survival: Finding food and shelter
- Combat: Fighting monsters for resources and survival
- Exploration: Finding rare resources that spawn in small bunches or specific locations.
- Mining: Extracting valuable minerals from the ground.
- Creative outlets: Building, setting up machines, etc.
These tasks are common to both vanilla and modded experiences, and they’re experiences that pull you in for dozens of hours over and over. I’m no psychologist so I’ll leave aside the reasoning behind why it’s such an addictive experience, but in this post I’ll be discussing some of the issues that crop up with both this game in particular and which are often shared with the factory-builder format.
Combat, (the lack of) dungeon crawling, food and healing is cheap, and harsh penalties.
Minecraft combat is, to put it lightly, not fantastic. It works fine for pre-arranged vanilla player-vs-player combat, but suffers elsewhere.
The main issue is that monsters just aren’t much of a threat. Early game you might die a few times, but you can easily avoid encounters and monster-proof your base. Midgame you have good enough armor and a supply of food to be able to wipe any normal mob. Late game, mobs are more of an annoyance than anything. Bosses are typically cheesable or not real threats, and, most importantly, armor progression isn’t tied to combat. You can acquire end-game gear by turtling your way through the game, preferring safe strip mines over caves, using monster farms for drops, and generally avoiding combat.
The second issue is one of food. Food is trivial to acquire and mass-produce. Till some fields, plant some seeds, go out and harvest every 20 minutes. Furthermore, saturation healing is incredibly fast, making food act more like a healing potion than natural regeneration. A stack of steak will win you a fight faster than any armor.
The final issue is one of consequences: there are none. On a typical modded server, you
have access to /home and /back, and there’s a gravestone mod preserving your drops.
Fixing your mistake is a matter of retrieving your gear, which you have infinite time
to do. The alternative of losing all your items is equally untenable: modded inventories
are a huge investment and losing everything would be a crushing blow. Some servers take
this a step further and turn on keepInventory, preventing you from losing items in the
first place.
A tertiary issue is one of accessibility and rewards: sometimes there’s just no reason
to go and fight monsters. A mob farm likely provides all the drops you need, and searching
out mobs has no further benefit. Mods like Roguelike Dungeons solve this handily by adding
adding large dungeons filled with steadily increasing loot and difficulty. They’re still
cheesable
though: you can mine around or rush in, grab items, then teleport out.
Addressing the problem: Dungeons on Demand
Proposed solution: Add a true roguelike dungeon system. Players may elect to challenge a dungeon whenever they wish. The dungeons are generated in another dimension and come with a few restrictions:
- An entrance fee of some sort (e.g. a cooldown, pay in resources or xp)
- If you die or leave the dungeon, your run is over. Your inventory is preserved.
- The dungeons are cheese-resistant: You can’t tunnel between rooms and you don’t recieve rewards until actually completing each room.
- There are a limited number of monsters in the dungeon. No dynamic spawns: a cleared region is safe (unless some monsters backtrack).
Rewards take the form of standard dungeon loot, enchanted armor, monster drops, and dungeon currency. Dungeon currency can be used to purchase from a dungeon shop, and must be used immediately after a run. Unused currency expires after leaving the shop.
This system has the advantages of tempering true risk without being excessively harsh or presenting an unattractive mechanic. At worst, you lose out on the rest of the dungeon’s rewards and whatever sunk cost you have (entrance fee, armor durability, food, etc). You can choose the difficulty (and thus quality of loot) to provide a steady progression over time.
It doesn’t solve the issue in the overworld, but it does provide a nice target to drive combat progression forwards.
Addressing the problem: three prong approach
Difficulty scaling over time is a common mechanic, one that minecraft actually does use. However, it’s a modest increase at best and doesn’t make handling monsters any different. Furthermore, difficulty scaling doesn’t make sense if you can lose your hard-earned gear and be forced to start over in a much harder scenario.
Solution part 1: Transient difficulty scaling
Difficulty scaling should be made much more dynamic. Chunks near active players will steadily increase in difficulty. Difficulty will also be tied to more features: monsters will spawn more often and in bigger packs, be more intelligent, spawn in higher light levels, and be generally more lethal. I’m envisioning something approaching Epic Siege Mod at the highest levels, where mobs become an unignorable threat capable of blowing through walls to reach you and spawning in full daylight.
However, killing monsters decreases the difficulty. This might seem counterintuitive at first, but it promotes active combat. If you don’t deal with the monsters now, they’ll get much worse later on. Ideally, ignoring the problem should render a region uninhabitable.
Additionally, if a region does become overwhelming, players can retreat elsewhere. They can grab their stuff and move, or rebuild their strength and wear down the difficulty by skirmishing the offending region.
Solution part 2: Intensive healing process
Saturation healing is significantly too fast. Replace it with a slower healing-over-time solution combined with a much bigger health pool and emphasize careful combat and use of harder to obtain equipment such as healing and regeneration potions to supplement natural healing.
For some example numbers, consider pushing the health pool up to 50 hearts (100hp) and dropping healing per day to 25 hearts. Healing should be possible irregardless of hunger so long as you’re not starving. Sleeping should heal by a flat amount based on the time skipped, twice as fast as being awake.
Healing potions should restore a fair amount of health instantly, but come with drawbacks discouraging abuse such as depleting hunger far less efficiently than natural healing and giving negative status effects such as weakness and mining fatigue. They should be reserved as last resort for escaping an encounter, or for intensive healing in a known safe location.
Solution part 3: Significant penalties
Losing your equipment is too harsh, but getting off scot-free with a free teleport back home is too easy. Partially taking away your gear isn’t really a viable solution either: some items aren’t evenly divisible and it’s still ultimately an inventory-preserver. However, a penalty of some core gameplay element such as, say, a heart container, won’t cripple you. Much. Especially if you regenerate it after a few in-game days.
Post-note issue: knockback
It’s a bit controversial, but I hate minecraft’s knockback. Any kind of damage immediately kills any momentum you have and mob attacks knock you around enough to be disorienting. Rapid-fire skeletons require your full attention to get close enough to hit them, though melee attacks aren’t too bad. Attacking a fleeing player pushes them further away.
Proposed solution: Remove knockback, replace it with a slowing effect. Getting hit with an arrow should cripple you (for a moment, at least), not send you flying backwards. This preserves making space with damage, but forces the defending player to retreat rather than hold ground with knockback.
Economic issues
Ease-creep / No failure modes
A notable issue cropping up with modern mods is that nothing ever
breaks or goes wrong:
machines stop when their output buffer is full,
pipes intelligently move items, running
low on energy only slows starved consumers, and laser drills provide an infinite supply
of resources.
At worst, everything grinds to a halt.
This is more a matter of taste, but engineering fault-tolerant systems could open up a whole new endgame endeavor. Mid-game factories could occasionally break down, requiring routine maintenance and replacement parts. Early-game automation wouldn’t be automation at all and instead be a balancing act of keeping temperamental machines afloat.
Denial of Conservation Of Energy
A related issue is the perpetual motion machine: Modded Minecraft has a lot of infinite loops. Explicit ones, such as the Laser Drill, generate ore from a theoretical sub-bedrock region. Put in power, get ore out. Seperate yellorium, stick it in your reactor, make back more power than you spent mining it.
Other, more egregious examples exist, such as the mushroom loop:
- Spore recreators quickly grow mushrooms from water and a bit of power
- The Bioreactor turns organic goods (such as mushrooms) into biofuel
- The Biofuel generator burns biofuel to create orders of magnitude more power than the spore recreators need to run.
Exponential growth
A third conjoint issue is one of infinite scaling: If machines never break down and you can generate infinite resources from them, what’s to stop you from scaling exponentially? Nothing, except for the server tickrate and the fact that you don’t really need all that many of a resource anyway. Power is universal, but there’s a limit to how many diamonds you need before you exhaust your capacity to care.
This one might be a bit of a non-issue, but many resources in the game simply
aren’t used for much. Even if it requires four stages of machines to make,
how much are you going to use it for beyond the stage five machine and
maybe some new armor? Basic resources suffer less from this since there’s
just so many mods that want to use them. One potential solution would be
to have a resource sink to bootstrap new tech tiers, similar to
how Factorio
has you mass-produce science packs.
No scarcity
An even worse problem is the infinite world and homogenous resource distribution. Every player has the same available resources, and can freely expand their base or mining operations to gather more should they exhaust their chunks. Furthermore, there’s at best only a handful of ways to produce any given item, meaning there’s essentially no concept of competitive advantage. When every player can satisfy all of their own needs, who needs to interact with others?
Addressing the problems (#1): Government Regulations
Consider a scenario: A server has a custom modpack. The pack removes as many infinite loops as possible. Laser drills are banned, soil is eventually exhausted by factory farming, and players are given only a limited amount of land to work with.
There’s no (or very few) infinite loops, and players have an absolute limit on the amount of material they may consume. This forces them to play efficiently, and can create a highly competitive resource extraction simulator when combined with a system for buying more land and a market as described in the next section.
Addressing the problems (#2): The Free Market
A long-standing mod idea of mine is to create a server-backed market for resources. Prices are determined by supply and demand. Players may freely buy and sell on this market (perhaps with a tax), including by automation with auto-buy/sell blocks. The market naturally trends towards a set stock for each item over time, so can be used even in a single-player scenario.
The initial benefit of such a system is that scarce or difficult-to-automate resources become easier to acquire, without becoming completely gameplay-breaking like with Equivalent Exchange.
The secondary benefit is that, over time, players will form natural supply chains and create a distributed business sim game. Natural shortages caused by some players buying will incentivize other players to build a factory to cover the demand, creating cashflow and eventually an economy. This could happen organically on its own, of course, but when was the last time you set up a long-term trade agreement with another player?
Interoperability
When you’re working with software from hundreds of developers with their own conventions and opinions, you tend to run into certain cross-mod compatibility roadblocks. Mods like OpenComputers provide incredible higher order logic and control, but intentionally suffer from the inability to read arbitrary data and must rely on APIs implemented by mod authors.
Integrated Dynamics is another powerful logic and control mod, but has a cumbersome programming process involving juggling variable cards. It is, however, very good at reading arbitrary data from the game world.
There exists no API to transfer data between the two mods. Luckily, vanilla mechanics are often universal and redstone can be used to build functional, if inefficient, 4-bit-per-channel communication channels:
Proposed solution: None. Getting hundreds of pieces of unique software to play nice is a hard problem. It’s a miracle they even run together at all!
Enough complaints! Let’s hear some positive stuff.
Modded Minecraft is genuinely a fun pastime.
If you enjoy undertaking large projects, there’s fun automation endeavors to undertake such as building a tower to house a rainbow generator and automating its 16 constituent generators. If you’re handy with programming, you can build things with OpenComputers like an automated shop where you can buy and sell items with in-shop credit, or something more ambitious like an arcade.
If you’re more of a low-tech type, there’s always building to be done and adventuring to do. There’s entire dimensions with their own tech trees and progression, and countless new materials to build stuff out of and biomes to build in.
Overall, it’s a pretty solid experience. Vanilla Minecraft is too, of course, but modded really brings out a depth of stuff to do.
Post-script: Request For Comments
I added a commenting system to this blog and I want you to comment! Discuss my ideas for solutions, counter my points, I’ve made the system as easy to use as possible: pick the comment you want to reply to by clicking the ↩️︎, type text into the box and press submit. No accounts, no email, nothing.
Update 2021-08-01: Added ‘three-prong approach’ solution