Hacker News new | past | comments | ask | show | jobs | submit login
Nostalgia for Minecraft (reallifemag.com)
193 points by ohaikbai 25 days ago | hide | past | web | favorite | 140 comments

A multiplayer persistent world like Minecraft with physics would be amazing, and probably replicate Minecraft's success.

e.g. If I chop down a tree at the top of a hill it could roll down and potentially wreck your house.

Or causing cave-ins and avalanches from indiscriminate mining. Building dams and rerouting rivers. Building realistic defenses and then siege weapons. Fun

Add in critter breeding and predation for dynamic populations that interact with each other. So if you slay too many Gorgs soon there would be none left. Leave them alone and they'll eat all the Fraggles. Players would form culling expeditions every week. Pretty soon you'll have players giving each other quests. :)

I think the venerable Ultima Online had a few features like that, or was supposed to, not completely sure.

> So if you slay too many Gorgs soon there would be none left. Leave them alone and they'll eat all the Fraggles. Players would form culling expeditions every week. [...] I think the venerable Ultima Online had a few features like that, or was supposed to, not completely sure.

Ultima Online tried to have such an ecology, but the players didn't react to it as described despite the game designers giving players minimal rewards for killing herbivores so that they would focus on killing the carnivores. Instead the players indiscriminately killed both the carnivores and the herbivores, because the players were human. The ecology was removed.


Original lead designer of the game here.

Richard's recollection is flawed on this (he wasn't close enough to the tech to know why it failed).

Reason #1: the closed loop fell victim to hoarding, particularly because the advancement system encouraged the production of goods with no value, as explicated in great detail in Zach Booth Simpson's detailed paper on "The In-Game Economics of Ultima Online."

Slides: https://studylib.net/doc/9778818/the-in-game-economics-of-ul...

Paper: https://web.archive.org/web/20090207141813/https://www.mine-...

Reason #2: the cost of pathfinding for the searches was too great. The Sims later solved this by having the agents in a very similar system use simple hillclimbing, with resources broadcast into a map.

wow, this is why I love HN: An incidental thread on a subject sometimes ends up having a key figure on that subject joining the discussion. :)

I never really got to try Ultima Online during its heyday (I was averse to MMOs and saw them as a step backwards from the narrative qualities of single-player games.)

Re: the ecology, other commenters here and on YouTube have also noted that it was too easy to do things what you shouldn't be able to do, like chasing down rabbits and deer with a sword. :)

Thanks and looking forward to your next game!

>because the players were human

It's a very compelling story, but that wasn't actually the case. The real reason was the flawed design. People just had to kill everything for the materials because of the way equipment was implemented.

I think one YouTube comment named another good reason: The animals were too easy to kill.

"The problem is obvious. The animals were too easy to kill. Imagine trying to chase down a real rabbit or deer with a sword. You will never catch either one and if by some chance you corner them the deer would actually stand a chance of beating you."

Mmo designers are very constrained in their game mechanics because they always have to deal with players that will exploit every possibility to grief. This unfortunately leads to things like extremely unchallanging encounters for random matchmade groups as any group will have players trying not just to get a free ride but actively trying to get the party to fail the objective.

As best I recall (it seems it's not easy to find sources now) the biggest problem was with player-killing and griefing. The devs left player-character interactions relatively unrestricted in the expectation that social order would be (largely) maintained by the in-character actions of the PCs. Similar things had worked in text-based MUDs of the time. When they naturally got a Hobbesian bloodbath instead (classic MUDs have many more hours of active games-mastering per player than any MMO) they began the retreat that led to the fake-roleplaying, theme-park experience familiar from World of Warcraft etc.

"Similar things had worked in text-based MUDs of the time."

Not really, in my experience. In MUDs with unrestricted player killing, in-game policing did very little to stop it (unless you count intervention from immortals/wizards, I.e. admins)

Sounds like the game was perfectly realistic then.

That was in 1997, a time when nearly all players were using dialup modems and server hardware was pitifully slow compared to today's. The attempted ecology was no doubt very simple.

UO's original designer, Raph Koster, recently announced a new MMO project where he intends to revisit many of those old simulationist ideas.

You may be interested in TerraFirmaCraft (https://terrafirmacraft.com/f/info), then. It's a Minecraft mod that tries to increase the realism of the game, I don't know if there are avalanches or tree rolling mechanisms, yet the mining system is greatly changed to include cave ins and other structural changes.

"TFC has thrown out Vanilla generation and started on a fresh slate. Sea-level has been raised to twice the height to accommodate 3 separate, varying layers of stone underneath, each spawning their own ores and minerals depending on which of the 23 new stone types it is comprised of.

Caves and underground ravines can be massive, with stalagmites and stalactites scattered throughout. Cave-ins are also an added risk; mining ceiling blocks that aren’t properly supported may result in metric tons of cobblestone falling down on your head.

On the surface, there are 16 different types of trees, large boulders, and smaller surface rocks scattered about. Grass, saplings and flowers slowly grow back depending on the temperature, while foliage changes colors to match the season. TFC also adds new crops and fruit trees as alternative food sources.

Inhabiting the surface are the same standard Minecraft mobs, but with a TFC twist. Animals have distinguishing features depending on their gender and breeding is changed so that females are pregnant for a period of time before giving birth to a believable amount of offspring. Drops have been changed so that all animals (including squids!) drop some form of meat, and mammals drop raw-hide or sheepskin that needs further processing to obtain wool and leather."

Shame it's stuck on 1.7.10 with no update in sight. There is a modpack, RLCraft, that PDP has played recently that tries to do something similar, if more vanilla minecraft-y. If that one were overhauled with more hardcore TFC mechanics, I'd be very happy.

I believe there is also an aspect of food spoiling if you keep it in your inventory for too long too, along with stats for protein, carbs, etc.

Spoilage would happen even in storage, but the rate was temperature-dependent so you could make it last longer by keeping it somewhere colder. I remember there being some work on preserving food indefinitely by storing it in a barrel of vinegar.

You'd have to adjust your food production strategy to fit your climate. In a cold climate, you'd have a short growing season and need to set up cool storage/pickling barrels in order to survive the winter. In a very warm climate, you might not be able to get anything to keep very long, but you'd be able to grow crops year-round.

Keeping bad meat in your inventory should infect everything else.

A fact you could [ab]use by sneaking it into other players' inventories.

I just want to say, thanks for posting this! I was just playing Minecraft before reading your post yesterday, after a long break (months to years of barely playing at all), and this was exactly what I was longing for.

Used to a play a server called Civcraft that was persistent for several years - server-side plugins enabled preservation of structures (to an extent) through reinforcement (Citadel), culling of mobs using "Mustercull", biome-specific farming with RealisticBiomes, and an imprisonment system (PrisonPearl) which gave players the power to ban others from the server.

People naturally formed their own laws, governments, societies, and over the years the balance of power went from a fractal multipolar world to a massive "red vs. blue" cold war with minimal conflict. Quite interesting really. Not to mention the massive international infrastructure works, including rails, bridges, tunnels, massive towns, and lag-inducing automated farms.

There are still direct descendants of CivCraft around; CivEx, CivClassics, and CivRealms are currently online.

The lead developer of CivCraft went on to do a lot of founding work on Althea mesh networking (https://althea.net/)

All of those are pretty awful, honestly. There's no guarantee of permanence like the original had (first one shut down due to hacked map, second due to mass duping ruining the economy), the admins are pretty one sided, and the culture is ruined due to everyone focusing on building megavaults. Ttk did do well with Altheanet, yeah. I think he's on this forum as well.

Civcraft 1.0 was amazing. I remember hearing about people having bots to show player's locations on an online map using jukealert messages.

Anarchy minecraft has gotten super popular on youtube. I wonder if there's room for it to reboot.

>>A multiplayer persistent world like Minecraft with physics would be amazing, and probably replicate Minecraft's success.

That is something that I'm actually trying to create. Although my focus is a world where you can create machines that are simulated by a physics engine. Users will be able to create humanoid robots, a mechanical giant clock, etc. Think Minecraft for machines. I'm still in the early stages right now so don't really have much to show anybody yet but that is the plan.

I already tried once[1] without success so this is my second go at it.

This is my first attempt from several years ago:


Looking for help?

Thanks for the offer. I will need people to try out the first release if you are interested. Will announce the first usable release here in a couple of months.

Space Engineers might be interesting to you

Though the physics strains to keep up with more complex creations/scenarios, I have had great fun building big ships and then smashing them into things (or each other). It's another game where they give you a sandbox and the possibilities are up to you

They even give you a super-threadbare programming environment with which you can write C# code

The lack of PVE kills it for me, but it's a really fun creative game. Build a cool thing, smash it, repeat.

Try Empyrion. I was a Space Engineers player who has switched entirely to Empyrion. It’s SE with way way more complexity. Food and starvation, recipes, mobs, biomes, and so much more. It’s a criminally underrated game.

Is the game still being developed? I bought it in pre-release and haven't looked at it in years.

Yeah, it recently hit 1.0. They're also adding some things they said they wouldn't, namely more PvE stuff. I'm hoping they add NPCs of some kind with scriptable behavior to give us something to fight

> Minecraft with physics would be amazing

Chris Delay of Introversion (Prison Architect, Darwinia, DEFCON) mocked up something quite like that back in 2011. I always wished it had been developed further: https://www.youtube.com/watch?v=eQMBGLMtdFE

Never let realism get in the way of game play...


>Some muckety-muck architecture magazine was interviewing Will Wright about SimCity, and they asked him a question something like “which ontological urban paradigm most influenced your design of the simulator, the Exo-Hamiltonian Pattern Language Movement, or the Intra-Urban Deconstructionist Sub-Culture Hypothesis?” He replied, “I just kind of optimized for game play.”

This is a bit misleading. From what I've read, the modeling in sim city is heavily influenced by Urban Dynamics by Jay Forrester, a 1969 book which was highly regarded in urban planning circles when it came out and has since fallen out of favour [1].

The "optimise for gameplay" quote was in reference to how developers decide between two equally valid but conflicting outcomes to a given scenario.[2]

[1] https://logicmag.io/play/model-metropolis/

[2] https://www.princeton.edu/~starr/17star.html

What game play..?

I mean a sandbox is about having a world to tinker with.

> dynamic populations that interact with each other

It feels like every open-world game dreams of this, and backs up on it. The problem is you need so many things to go right for it to work and be fun.

* It needs to be resilient to player shenanigans * It needs to respond to player actions * It needs to be coherent and connect to the actual gameplay * It needs to reward player interaction while not breaking the economy

That's a pretty tall order when most MMOs struggle to even motivate the existence of simple NPC vendors.

> A multiplayer persistent world like Minecraft with physics would be amazing, and probably replicate Minecraft's success.

To a point, yes. After that it is EVE online and lacks general appeal. I was watching a video tonight of some players trying out a new mod that has some basic and limited physics - waterwheels can rotate shafts. Two different sized cogs let change up rotational speed. Belts can connect shafts, transferring rotation and also creating a converter belt.

All cool stuff, but part of the appeal is the unrealism. As soon as work starts reducing torque it goes from fun mental puzzle approachable by many to math and frustration approachable by few.

Minecraft's success is not only in the freedom to build, but in the reduced realism that makes that potential open to so many as fun.

Empyrion has block building and working structural integrity, the settings is in the future so it might not be to everyone taste.

IIRC, Markus Petersson, the creator of Minecraft, worked for a long time on Wurm Online, an MMO where you can kinda do what you're talking about - as a group or individual, you can level forests, dig down mountains, build roads, cities, etc. It's a slow and labor-intensive grind but it is a game where as a community you can change the whole game world's geography.

I saw it was on Steam the other day, I guess they tried to add some storyline to it.

When I read this for some reason I envisioned a game where you need a huge group of players to slowly accomplish changing the landscape, like you’re a literal worm digging away at your patch of dirt with millions of others.

Picture my disappointment when I realised they meant wurm like a dragon wyrm.

> If I chop down a tree at the top of a hill it could roll down and potentially wreck your house.

There's a Minecraft like game with this mechanic right now. When you chop down a tree it goes rolling down the hill and can damage you/enemies. It's as satisfying as you'd hope it would be. I'm not affiliated but I enjoyed playing the alpha.


There was a game or tech-demo called "upvoid miner" a few years ago. It had worldgeneration similar to minecraft and proper physics with rolling tree logs. It allowed arbitrarily shaped terrain and I think liquid physics were worked on. But sadly the studio had to shut down due to not getting enough sales. I think you can still download it and test out the free tech demo though.

Wow, that looks quite interesting. Seems dead, but the folks behind it open-sourced the code: https://github.com/Upvoid/UpvoidMiner

GPLv3, unfortunately, but interesting nevertheless.

A township tale has persistent world with physics, not exactly minecraft though https://www.youtube.com/watch?v=OfgjJOUW_KY.

Yes! That's somewhat better than what I had in mind! (the granularity of chopping down individual branches)

I would love to see the breath of the wild physics engine in a persistent MMO.

Yeah! People are hyped up about VR but I think MMOs with physics could a big milestone in gaming too.

Both of them combined would literally be a step towards living in a Matrix. :)

The performance would be pretty bad. Every time someone placed or broke a block, the game would have to check if there was a heavy object above it that now has to fall, if there was water flowing through that voxel that now has to be rerouted, etc etc. Combine this with many players in one world and you will not get many server ticks per second.

Just from the top of my head without thinking it through:

If it was all deterministic, servers might just send player input and the physics simulations could happen locally on players' devices.

And as the sibling comment said, multiple servers each handling a small region of the world (both gamespace and meatspace.)

User-side calculation makes it prone to hacking.

That can easily be engineered around using geosharding.

One of the neat things with multiplayer games is that you get to see quickly how players creatively leverage/abuse the game play primitives (the basic physics rules) to their advantage.

Eco is getting close to that. But the whole idea of consequences and interaction somewhat doesn't give the same carefree vibe you get from being in a virtual world.

A few commenters are responding on how they think certain features in addition to Minecraft would make a killer game. I don’t think Minecraft’s success had anything to do with its featureset. To paraphrase notch, he believed in a model of good enough vs perfect and continually released updates that added the minimal needed for the feature he wanted, responding to user feedback in real-time. This was unheard of at the time for how games were developed, in addition his target user was the lowest common denominator in terms of gamers, his mother, whom self described never played games.

“In addition to that, he released updates often, in accordance with the Swedish saying “hellre än bra” (meaning someone who prefers spontaneity over perfection). As soon as a new function or bug-fix was in place, he made it available via his site, asking players for help in testing and improving it.” https://www.wired.com/2013/11/minecraft-book/

Plenty of boring casual games with little success release early and often.

I don't think there is any 2 thing that made Minecraft a success. It's a combination of a great nu!bet of things but most importantly a combination of a great number of things done right together not just as you'd view them individually.

It's a computer version of legos that is actually playable.

Lego tried making some brick building games but they were universally awful. A couple hundred bricks and the computer slows to a crawl. Their later games don't involve building at all.

Minecraft scratches that building itch, with a bucket that never runs out.

As I recall Minecraft's snowball started with geek dads watching Notch live coding and buying the game for their kids. https://youtu.be/MhQ70O1MiXc

Was minecraft already starting to get popular by the time minicraft was released? I don't think the livestream alone lead to the game's popularity, it was probably spread a lot more via word of mouth.

I bought it for my kids long before the release. It spread through their primary school like wildfire, as each kid that saw it joined in, and in no time it was the number one game In there by miles.

People will attribute the success of minecraft to the game itself, but really the fact that it could be scripted with a fairly popular language meant mods, and mods meant a ton of users. That to me is why, if he had programmed this game in a better language or tried to over-engineer it, it would have sucked.

Ten years ago, my kinds were all under 12, and I discovered Minecraft from a message on a forum. So different, relaxing and engaging. And such blocky, bad graphics compared to the drive to 3d photo-realism that seemed to be the story in most games. All of my kids have discovered Minecraft on their own. I figured if Dad recommends it, that it is instantly cringy.

Four girls, one boy and all of them love it. Ages 10-22. Roblocks? Not so much.

One thing I think this misses is that while Minecraft's graphics are blocky, the terrain generator creates rather spectacular and varied scenery. Just wandering around has an appeal evocative of visiting national parks, and building even a basic hut in a nice setting is more appealing than it would be in a plain sandbox.

Maybe that's part of why it feels "somehow pure"?

It's a lot like an atomic simulation. All the items are quantized and can be combined to produce new items.

They follow very simple rules and their interaction with each other can produce amazingly complex machines.

It's the most realistic game we have.

I find it disturbing that more and more websites, and especially personal blogs, refuse to even show any content without loading scripts from google and cloudflare. It's a bunch of text with images. It's literally what HTML was made for.

Mandatory mention of this: http://motherfuckingwebsite.com/

Maybe someday the web will return to its humble roots.

I really think Minecraft mods are an underused method for teaching programming. I first got into programming in 7th grade developing mods. Times have changed, but ~5 years ago there were great tutorials, documentation, and good support channels (IRC) for modders.

If regular development is too complicated, there are mods that add programmable computers (Lua, iirc). You can use them to mine, build, automate, or in the spirit of Minecraft, really do whatever you want.

The most popular variety of Minecraft modding frameworks, Forge, is non-trivial. Think Android Java level of boilerplate and equally terrible homebrewed build tooling involving a ton of groovy. The developer friendly modding systems tend to be server-side only e.g. Bukkit. The classic "fun" mods like Buildcraft, Thaumcraft, and the grandparent of all Redstone mods, Redpower, are all Forge mods involving a ton of duplicated client-server code. Unless you enjoy IAbstractSingletonProxyFactoryBean Enterprise Java, coding Minecraft mods is a far cry from the UX modern engines like Unity provides.

Source of one of the most popular mods:


Other examples: https://github.com/agaricusb/ForgeMod/wiki/List-of-Open-Sour...

How people usually run mods (in packs): https://www.curseforge.com/minecraft/modpacks/ftb-official-p...

Open source alternative client: https://multimc.org/

The core tooling that helps with modding:


It's a set of decompilation scripts included with most modding frameworks. Not officially endorsed by Mojang but the developer is a Mojang employee.

Nowadays there is Minecraft education edition; a variant of the Bedrock version which has a code editor to create automations, and a classroom mode. Aimed much more at schools than at individual student-hacker-learners, but a lot more accessible IMO, and has a heap of resources/docs and communities for those teachers.

Is there any way to get this version? I would love to use this with my kids.

If you just want the code editor part, then https://minecraft.makecode.com/

Thank you! Will definitely check this out.

You can't just download that.. you need valid accounts linked to education etc.

There is also https://www.minetest.net/. It's fully open source and free, and although it is written in C++ it has a Lua modding API with many mods.

Minetest itself is just the engine with basic functionality such as terrain generation and basic physics. Mobs, and other things are added via mods. Mods can be installed manually or in-game from https://content.minetest.net/.

The problem I had with Minetest, is that the base sandbox is quite boring and I think you're not supposed to be playing it on its own, but then when it comes to mods there is so much choice and I couldn't find a place to start from (something like: "if you are a beginner, start from these 5 mods"), so I don't know where to start from. Any suggestions?

Yeah. That's a common problem and mostly because nobody has time documenting this. The minetest page itself helps a bit (under customizing, but it's not that helpful admittedly).

FWIW, here're my recommended mods:

Basic gameplay: 3d armor, mobs_redo + mobs_monster

Logic: mesecons, more mesecone, digilines

Hard monsters: NSSM (not so simple mobs), dmob (dragons, etc) (both need mobs_redo)

Building more: Homedecor

There're also some read-made games. Mineclone 2 is pretty good.

Just replying back to let you know that I'm enjoying Mineclone 2 a lot! Thanks again!

Thank you, this is very helpful!

Just go play on a server. Vannessa's servers or Pandorabox...

Thank you, I'm usually more of a single player type but I will give a try to the server!

It’s frustrating how Minecraft’s potential hasn’t been realized under Microsoft.

The game has such huge potential and tremendous goodwill, and it’s been so stagnant since the acquisition.

I don't know. They've added a ton of underwater content, new structures, mobs (bees, foxes, llamas, phantoms, drowns, etc), sunken ships and treasure maps.

If anything, they've had a better release cycle since the acquisition.

the new structures and content that reward exploration are great! the new mobs, not so much. they don't have a purpose in game that wasn't fulfilled already, and while I appreciate the diversity they give not something new to do, just more of the same. what'd be interesting are additions that open up new things to do, like that parcour level demo that show off the new wax block, but tbh wax and bees scale seems off compared with the rest of the game.

they really don't seem to want to explore too much. so what if they added a block that allowed automating things? what would be having a sky dimension accessible with the elyra wings? that would give new purpose to existing content without altering the game too much. what if furnaces and other production structures added taint to the biome, making it more and more polluted and unfriendly to live in? so many possibilities! luckily there's plenty mods out there, but a well balanced execution would make it so much better.

> what would be having a sky dimension accessible with the elytra wings?

This is something that definitely needs to be added at some point. One problem I would see is that there is already a very good mod that does that. In a sense they are between a rock and an hard place in trying to both copy and not copy it.

What the game really needs is what was promised ages ago: a mod API on the C++ edition.

Yea let's actually make the C++ edition playable first. Have you noticed how the cursor acceleration is actually like a joystick? I play on a computer for a reason, I don't want console nonsense creeping on me. Same goes for the whole UI. Also no linux version.

When it takes 30 minutes to load your mod pack and even with a RAM disk and a healthy number of fast cores on a dedicated server you're still dipping below 20 fps on some occasions, it's time for a leaner edition than java. I'm not even saying java is inherently the problem, but the java edition of minecraft is not performant and the bedrock edition is.

I understand the interest in some of the larger modpacks, but I try to stay as close to vanilla as possible and to me, if it takes more than a minute to start a server, you've got too many mods loaded.

Additionally, the server isn't multithreaded. It benefits always from having the fastest clocked CPU you can throw at it.

Client performance can be greatly improved by using Optifine.

I would love to have a more-well-built version (as I think the core of Bedrock is), but Mojang has taken some weird steps where things like redstone mechanics are severely limited compared to the Java edition... and they have no intention on fixing it. Now you've got these weird minor differences that will never merge and you'll never get those players to switch versions.

As I remember from setting up a server way, way back is you want to give that JVM as much memory as you can trow at it.

I haven't hosted modded servers in many years but a while ago the advice was to avoid (4, 8] GB (might have been 6.5 instead of 8, I don't remember, but the point is there's a dead zone) because of something weird with the garbage collection. I think it had to switch to 64-bit pointers or something. Anyway the worse garbage collection outweighed the memory advantage. I personally never hosted with more than 4gb.

IIRC more memory also increases GC time so after a certain point the GC causes such annoying lagspikes it's often better to play with less.

I’ve done testing and at some point more memory just changes how long/large the GC cycles are. More memory can cause larger hiccups when mods are leaking memory.

Better hope you pick the right arcane garbage collection strategy!

Latency minimal GC is the right choice for game servers. The most available one is G1, though others like Shenandoah and ZGC are on the way

I played with a few of them. A server I used to run had a leaky mod that would, when a base was large, leak so much that the GC would hit twice a minute with a 5 second stall. It didn’t matter what GC was used. I never found a solution.

And a linux version of the C++ version.

Bedrock sever is available for Linux. No client yet, although there is a hacky launcher which runs the android binary on Linux and works alright.

You don't get it, most people don't want to mod with c++. Java is more popular, runs everywhere quite easily, and isn't as hard to understand. Imagine an 12 year old trying to mod with this language, they'd be spending more time figuring out how it works than coding a mod.

Scripting for Bedrock is in beta!

Minecraft is for selling Xbox's and Windows 10. If they wanted modding they would have modding.

It’s more a matter of doing what they promised, which is also a highly requested thing. There’s a middle manager somewhere who is being told they need to never officially support modding so they can continue to sell valueless items to children.


You’re right I hadn’t seen this. It’s pretty fresh and I’ve been off minecraft for a few years.

It’s neat, but even by previous FAQ response standards it’s horribly weak. It appears like little more than an ability to make scripts of existing vanilla commands. You can’t make/register blocks, fluids, modify terrain gen, or even trigger individual events. As a framework it isn’t useful for content creators.

From what I understand, a lot of that is due to a combination of factors from bugs and spaghetti code from prior to purchase, to the necessity to not change too much too quickly, which might cause a community uproar, to simply just trying to recoup the cost of the IP purchase in the first place by developing projects around the IP, which the main game (java edition) doesn't see any progress from.

Much of the popularity of the game has a lot to do with the modding community as well, which is where you go when simply building in the vanilla game is no longer interesting. Any change to the vanilla game can have major ripple effects on the modding community.

Lastly, I'd say that one huge development they recently fixed in the code base was the block ID limit. Now that the ID system is name based rather than number based with a hard coded limit, the number of blocks that can be added to the game isn't limited by anything other than the direction of the creative team and new mechanics.

How so? They've been releasing updates at a pretty good clip each with significant changes.

Everyone is arguing with you but I 100% agree. Official mod support is all I want. And it isn't happening.

Hmm, there's been an awful lot done with Data Packs.

Didn't they unify everything so all the different clients (desktop, mobile etc) can log into the same worlds? That would have been a hell of a lot of work.

No, not really. There's a "Windows" version for Windows-based desktops that can co-mingle with the other C++ versions, but the main version of the game is still based on Java, with the features being ported to the various console versions shortly after. The Java version remains the only way to play Minecraft on Mac and Linux, for example, and it's also where pretty much the entire modding scene lives.

It's a little complicated from the outside looking in, since the base game is very similar on the surface, but there are major technical differences between the platforms that would make true cross-platform very difficult.

EDIT: The Minecraft Wiki has an excellent overview of all the various platforms. Check out the sidebar on the right for a breakdown. Generally, the "Bedrock" edition is what's on most consoles that are still being supported, and I ... think it has cross-platform multiplayer support? I play on PC so I am not super familiar with the details on console.


There are two main versions of Minecraft.

Minecraft (formerly known as bedrock) which is the C++ edition and runs and plays across Windows 10, Android, iOS, and Xbox.

Minecraft Java Edition (formerly known as minecraft) which as the name implies runs on Java and can play between Windows, MacOS, and Linux.

Feature sets of the game are VERY close now (and get closer each release) with the major difference between them now mainly being in how redstone works. Java has redstone bugs that have since become so relied on they are considered features, the C++ version doesn't have them. Someone picking up minecraft for the very first time wouldn't feel they were missing anything using C++ version.

Another important thing to mention about bedrock redstone is that it is non-deterministic while Java's redstone for all its bugs is 100% reliable.

> Minecraft (formerly known as bedrock) which is the C++ edition and runs and plays across Windows 10, Android, iOS, and Xbox.

Switch too. There are rumors that the Playstation 4 will receive a version based on Bedrock as well, but it's not confirmed.

If iOS now works with the java edition I'd consider that a major achievement

Does minecraft earth and the countless updates not count?

They are mostly working on Minecraft 2 which will be a massive multiplayer world.

oengb 25 days ago [flagged]

Stagnant? I don't think you play or know the game at all.

Would you mind reviewing the site guidelines and sticking to them when posting here? Personal attacks aren't ok on HN. But more interesting is: "Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something." If you know more than another commenter, the best way to express that is to share some of what you know, so we all can learn.


The article is more about Roblox (whatever that is) than Minecraft. Title kinda misleading.

Both of these games are excellent intros to game development. ROBLOX more so in my opinion - it’s a lot like the old school quake-c modding days and such.

I tried to get my kids to play around with web pages and tutorials, but it wasn’t until they wanted to create mini games in ROBLOX that they got busy.

Interesting article, the author kind of gripes at the end that open world games like Minecraft and Roblox eventually devolve into a competition among players for points, achievements, and status.

While that is true in many cases, so is society in general, and I think that occurs when people accept rules that limit what they create and discover in the world. 'Mini-games' are all about points and status, servers with economies result in play that is all about selling goods and accumulating wealth.

What the author doesn't mention is the side channel (typically discord) on which players are communicating and socializing.

I agree with the article, Roblox certainly doesn't please the Minecraft crowds - or at least, not everyone among them.

I'm curious about Hytale, an RPG game currently under development, made by Hypixel which is the developer behind one of the most successful Minecraft servers.

Another title that I was excited about was Cube World but after reading the reviews, it seems it didn't really hold up to the hype.

I think Minecraft was the first game with a procedurally generated immense world which was also 100% modifiable. Games with virtually infinite procedural worlds have existed and sparked imagination before (my first was Elite 2 on the Amiga), but you couldn't really interact with them in a way that changed them.

Correct me if you know of any earlier examples.

It annoys me when people call Roblox a game. It's an engine and a platform. It's like calling Unity or Unreal Engine a game.

I've also found that people tend to underestimate the development tools Roblox gives. There are games on Roblox that have external web APIs for managing servers, games that circumvent the standard "player model" and instead opt for an entirely different appearance, games that make use of the GUI tools given in meticulous ways...

Roblox does a fantastic job at giving creative young minds an opportunity to expand on their ideas — something very few other "games" accomplish.

This is exactly what I love about Roblox. My 11-year-old daughter does "commissions" and designs 3d objects to sell, she's on "Scripting teams" where they help design in-game elements. She's having fun, building / creating, learning and interacting with people. Something I could only dream of when I was 11 (I had BBS's, lots of TW2002).

My 11-year-old knows blender, photoshop, python and javascripting because of Roblox and some Minecraft.

Remind me to break out Tw2002 on docker.. And yes, the opportunies these kids have is amazing.

It's a nearly fully featured game engine. Maybe not quite suitable for a AAA game, but I think it'd be possible for a studio to make a real commercial game with it.

Minecraft is the game that made me feel old - and that I was genuinely missing out on something.

I seemingly signed up my €15 in 2011 and had fun. Then forgot about it.

Then I remember when MS bought it for silly money. Full on WTF? headspin.

Then I visited my former-bridesmaid, who shared with me what she'd built on her ipad. I'd no idea what she was talking about, quickly stopped randomly hitting stuff when her upset overcame her politeness and she told me to stop. She really cared about what she'd built.

I can still only wrap my head around it as "infinite lego with a community" - but that's pretty damn impressive.

> Then I remember when MS bought it for silly money. Full on WTF? headspin.

Considering Minecraft has over 176 million copies sold, not to mention the endless amount of merchandise and spin-offs I'm sure MS is getting royalties on, it looks like they got a bargain at $2.5B.

Maybe I didn't make my point clearly enough. To me - it was a silly fun game I spent an evening with - and then never touched again. The whole "becoming a thing" completely missed me. I maybe say it was on Android, or Xbox - but being an old fart completely missed the fact that millions had turned up and were playing it (and PewDiePie was a name of somebody etc etc)

I played it overnight nonstop after buying it, also in 2011 I think (infdev anyone?). Then also never touched it again. However I am curious how you missed the hugeness of it for so long, did you not follow gaming sites or anything gaming on YouTube similar at all?

Did you try any mods? You can make minecraft into all variations of games.

I currently have a persistent server I run for my kids. My son loves to explore and fight things, my daughter likes to build things, and I like to make machines with redstone. It's fun that we have something that we can play together that appeals to all of us.

This raises some interesting points about social dynamics that emerge in creation-oriented engines like Minecraft/Roblox based on how their different business models work.

Roblox sounds like a place where a sort of creative hustle underlies the whole experience, where everyone's creative experience is about trying to figure out what they can get from others. It's full of in-app purchases, by way of people's creations.

Minecraft, meanwhile, sounds like a more open-ended collaborative experience where people create for the sake of creating, and the worst negative externality is some people do it to gain views on Youtube. There's no in-app purchases, just the upfront fee paid to Microsoft.

One thrust of the piece is that playing Roblox promotes "indoctrination into entrepreneurship for children", by way of many of the 'tycoon' style games.

Author also points out how accruing possessions is a large part of Roblox: "Compared with Minecraft, most Roblox games are highly individualistic and use private property as their main incentive rather than skill-building or a sense of common good."

Author then frames Roblox as operating a kind of multilevel marketing scheme, where everyone's hustling to make game modes to get game currency from others, and it all flows up to Roblox. It may not match an MLM perfectly, but it really made me think about how games are places where a lot of these really negative ideas from the offline world (like MLMs, gambling, etc) can appear in similar form in game worlds.

The author goes into a digression about streamers and creators showing off their work on Youtube, and that being something that sort of sullies the purity of these creative mediums. I don't see that as particularly unique to these games though, as that's part of the broader trend of 'life as performance' which exists everywhere in social media (see also Instagram, etc).

The last paragraph is great: "By enabling self-defined goals and DIY accumulation of skills and resources, Minecraft once seemed to promise fun without fixed form, without ownership, without competition and hierarchy. Roblox has no such pretense. Though its premise of blurring the line between player and creator may seem democratizing, it transforms conventional gameplay into entrepreneurial striving while indoctrinating young children into capitalist society’s hierarchical scoreboards, its fantasy of being the best by having the most."

Is this all just a reflection of adding in-app purchases with a % going to creators? Feels like it.

The article seems to be saying Roblox is a successor to Minecraft. But Roblox was released under that name in 2005 (I created my Roblox account in 2008). Minecraft was released in 2009.

At a higher level of abstraction, git(hub|lab), or even the internet itself are still less structured construction projects.

I am a bit surprised at not seeing any reference to Second Life in this article. Is this because it (quickly) turned to more "mature" themes?

I ask because both the exploratory and (more importantly) the build/customize parts were a big element of that - also, it is still active (I believe) even if not exactly "thriving" today.

> Even as Minecraft experiences a nostalgic renaissance, the online open world Roblox has become its spiritual successor.

Too bad Roblox doesn't work on Linux even in Wine, and developers have no interest in doing anything about it. So it's far from being a proper successor.

The page is not loading its content for me. Maybe it was hugged to death?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact