Minetest Forums

I've sorta been out of the loop of the Minetest community, as a huge Minecraft modpack project has sucked me into oblivion (and to be honest I was kinda getting tired of the open source circlejerk). But I was recently browsing the Feed the Beast sub-reddit and a random comment made me angry but then incited a very interesting train of thought that I think would be useful to the Minetest modding community.

It all started with this reddit post inquiring about minimum specs for playing Minecraft modpacks. At first, it made me very upset, as I realized that a player was willing to pay nearly 30 bucks for a game that they had no interest in playing by itself, only for the priviledge of enjoying the labor of people who saw little or no money in return for their work. Now, I'm not an open source purist; I believe that good games are worth paying for if the developer(s) decide that is how they want to distribute the game. However, I feel strongly that if a person does spend money on something, it should be done with the knowledge that spending money is a capitalistic form of voting, and therefore should go to the people who actually put work into the thing the person is buying.

Oh boy, I guess I'm contributing to the open source circlejerk myself in a way! Oh well... I'll continue...

With my strong opinions on for-profit games in mind, I was very close to submitting a post which would explain why the person shouldn't buy a game just for the mods. I even had a list of alternative games which each satisfied some core aspect that modded Minecraft does, all of which are very good, which they should consider purchasing instead. And oh, boy, was the list long: Terraria, Factorio, FortressCraft, Subnautica, Starbound, Stardew Valley, Zachtronics games...

but... wait. This person didn't want to play a block building ARPG like Terraria! They didn't want to play a survival crafting game like Subnautica, or an automation sandbox game like Factorio, either. They wanted to play a Skyblock!

The original skyblock map was released over 8 years ago, and there STILL is not a decent video game on the market which captures that experience and expands upon the concept in a compelling way. And yet, in spite of this, skyblock modpacks and their variants are, to this day, the most popular way to play modded Minecraft. Just to clarify, a modpack in Minecraft is the equivalent of a subgame in Minetest. A skyblock is an experience where you start from nothing and build outward, creating more materials and structures as you progress, traditionally starting on an island floating in the sky.

Curseforge is currently the de-facto platform for modded Minecraft since the fall of Technic, and the Feed The Beast modpack brand, home of the previous [arguably] de-facto platform and a household name for modded Minecraft, is itself based on the first well-known modded skyblock pack of the same name. I looked today at the list of the top most popular modpacks, at minecraft.curseforge.com/modpacks, then sorted by total downloads, and of 20 modpacks on the first page, I see:

• 1 exploration/combat modpack (Roguelike Adventures and Dungeons)
• 4 technological progression modpacks (Age of Engineering, The Simple Life 2, Space Astronomy, SevTech)
• 1 tower defense modpack (Invasion)
• 1 unthemed progression modpack (FoolCraft)
• 2 "everything but the kitchen sink" modpacks (All the Mods, All the Mods 3)
• 1 Harvest Moon themed modpack (Farming Valley)
• 10 skyblock modpacks (Skyexchange, Modern Skyblock 2, Crash Landing, Regrowth, Modern Skyblock 3, Project Ozone, Forever Stranded, StoneBlock, Agrarian Skies 2, Project Ozone 2)

There is clearly a demand for skyblocks. And for good reason! There are many things that make skyblocks appealing over a traditional, procedural open-world voxel survival experience:

• Better performance due to a much simpler environment
• The beginning progression differs greatly from the traditional start (although there is usually some tree punching in the beginning)
• De-emphasis on combat for acquiring mob loot, increasing appeal for players who are not as good at combat, jaded by tired combat mechanics, and not wanting to put up with mob spawn RNG
• No large things in the way, allowing the player to build their base outward as they please
• The satisfaction of starting out with almost nothing and creating a huge amount of stuff from it

It should be noted that not all the skyblock packs I mentioned start on an island in the sky, and can vary greatly in difficulty and theme. Crash Landing puts the player on the ground of a dangerous alien planet, where staying alive is a challenge at first. Stoneblock takes place underground, and has a reputation for being overpowered to a silly degree. There is also an indie game called Raft which takes place on the ocean and has the player build an increasingly complicated floating raft base (free demo is on itch.io), although its complexity and depth can't compete with modded skyblocks.

As I mentioned before, there isn't really any other game that does a start-from-nothing experience like Minecraft does when using a flagship skyblock modpack. This means there's a great opportunity for Minetest to step into the scene and innovate. But the skyblock format also greatly lowers the barrier of entry for a Minetest modder interested in creating a compelling modded experience. To be honest, I wouldn't say that the skyblock format plays to Minetest's strengths either, because even though the engine and modding API are good, even the "best" Minetest mods aren't really that good (and even if the engine has good performance, that doesn't mean the lua mods have good performance). So, there's still a lot of work to be done. But the most important things that Minecraft excels at over Minetest, such as PvE combat, exploration, and thematic coherence, no longer apply in a modded skyblock setting. (I do think these sort of features should be improved upon in the long run, but I'm talking about low hanging fruit here)

If these skyblock formats seem compelling to you, I recommend checking out some let's-plays of the skyblock modpacks I mentioned on YouTube for inspiration. If you need a YouTuber recommendation, check out direwolf20. Overall there's a lot of untapped potential, and while it's still not easy as modded Minecraft players' expectations for modded skyblocks continue to rise, a high quality "start from nothing" subgame with unique progression will provide a way for Minetest to stand out in a way that would be much more difficult to pull off otherwise.

tl;dr: Modded skyblocks are extremely popular in Minecraft, yet there is no video game that capitalizes on it effectively despite the demand. Their simplicity makes them relatively low-hanging fruit for Minetest subgame developers, and a potential boon for Minetest's publicity, with lots of room for creative exploration.

asanetargoss wrote:tl;dr: Modded skyblocks are extremely popular in Minecraft,

This is why Minetest has such a mod, too.

https://cornernote.github.io/minetest-skyblock/

There's also a version by rnd: viewtopic.php?t=19625

I found two skyblocks with a quick minetest forum search and they both looked pretty boring and not much different from the original skyblock map. Cornernote's skyblock looks about the same as that.

That just leaves rnd's SkyBlock redo, which looks like it actually has some meat to it so I'll definitely check it out, although it looks from a first glance like it stays pretty close to the most common modded skyblock formula.

I still feel there's a lot more room for innovation, both in terms of theme, as I mentioned, as well as the early game. Overall, it feels to me the vast majority of the limelight for Minecraft skyblock modpacks is going to skyblocks based off of one mod: Ex Nihilo. Harvest some saplings, compost them, sift the resources. Rinse and repeat.

Now, obviously, Ex Nihilo is far from the only early-game for skyblocks. There's Garden of Glass, a skyblock version of a flower-based technology mod called Botania, which is arguably even more successful. There's Sky Resources, popularized by Modern Skyblock 2 and 3, which is not that good from a game design perspective but introduces a lot of unique ways for acquiring resources. Skytouched is a lesser-known modpack which uses that mod and innovates further by introducing a source of danger when acquiring some resources which is cleverly thematically tied to the fact that the player is in the sky. There is also a mod whose name escapes me which bases its resource progression entirely on killing mobs and ascending to higher tiers of mob killing (that one's not exactly low-hanging fruit, but you get the idea). There's also the original Feed the Beast map, whose early-game was sort of a puzzle which drew on admittedly arcane knowledge of vanilla Minecraft and Equivalent Exchange 2 (a "create stuff from energy" mod where every item has an energy value (in practice, minus some items excluded for balance reasons)). And while Raft isn't that far along, it does admittedly have a unique early-game where you fish junk from the ocean to make stuff, and have to defend your raft from a shark.

In my opinion, a good skyblock should feel like a puzzle (Feed The Beast minus the arcane prerequisite knowledge) while taking more advantage of the unique properties of skyblocks (Skytouched). In this area I think there is room for something new. Not... create one decent subgame and call it a skyblock and call it a day. But maybe I'm just some rambling weirdo on the internet talking to the wrong crowd about game mechanics and some numbers that add up to 50%.

Edit: Dangit, I meant Equivalent Exchange 2. 3 isn't released.

asanetargoss wrote:Oh boy, I guess I'm contributing to the open source circlejerk myself in a way!

That's kind of a rude way to say "I guess you guys have some valid points," but I'll take it anyway. ;')

Nice writeup. As I've been out-of-the loop concerning minecraft since microsoft bought it and I can't get my copy to run anymore I didn't really know what the 'skyblock' stuff was about.

Now I need to try out the minetest versions :-)

But I truly don't understand your reasoning here:

asanetargoss wrote:However, I feel strongly that if a person does spend money on something, it should be done with the knowledge that spending money is a capitalistic form of voting, and therefore should go to the people who actually put work into the thing the person is buying.

If you want to play a particular mc modpack, whose creators distribute it for free because it's a labour of love, why is it bad to buy the base game???

You *can't* give your 30 dollars to the modpack creators, because you then still can't play it because you don't own the base game.

Of course I agree with you if the topic is fair-trade coffee for example, where you hope more money will go to the people doing the hard work of actually growing the coffee. But if people create a modpack as a hobby, they'd actually *like* you to buy the base game and play the modpack. A completely different situation IMHO.

Mantar wrote:

asanetargoss wrote:Oh boy, I guess I'm contributing to the open source circlejerk myself in a way!

That's kind of a rude way to say "I guess you guys have some valid points," but I'll take it anyway. ;')

I mean it's more about me being self-aware about my contribution to the vibe and the community feel. Around the time I stopped checking the forums regularly, there were some people who were saying the same things over and over, and some people saying some rather... extreme views. That's what I meant by circlejerk. I don't want to be a member of circlejerk type A: that guy who says the same thing over and over.

Honestly I would be contributing most to the community in a positive way if I just finished a Minetest mod already. But Minecraft mods/modpacks are a great source of inspiration from a game design perspective, so it's not entirely a loss. I'd say a new type of skyblock subgame is a promising idea that I'd totally work on if I had infinite time. Aside from that, creating some "cross platform" mod that can be played on both Minecraft and Minetest. I dunno... maybe a "cross-platform" skyblock mod with really unique mechanics. One of the great things about skyblocks is they can entirely differ from reality and no one bats an eye.

wziard wrote:Nice writeup. As I've been out-of-the loop concerning minecraft since microsoft bought it and I can't get my copy to run anymore I didn't really know what the 'skyblock' stuff was about.

Now I need to try out the minetest versions :-)

Huh, that's interesting! I'm glad you found it informative. :)

wziard wrote: If you want to play a particular mc modpack, whose creators distribute it for free because it's a labour of love, why is it bad to buy the base game???

By bad, I mean unethical and not in the purchaser's long-term self-interest.

It's unethical to buy the base game solely for the mods because giving money to the company financially incentivizes the game industry to place less value on human labor, while the industry is already known for poor working conditions.

It's not in the purchaser's long-term self-interest because it financially incentivizes the game industry to produce things the purchaser actually doesn't like. The game industry is risk-averse and will use the base game as a template for financial success.

A bit far fetched in the case of mods I think.

I'd prefer the game industry to make their games *more* moddable. And I'm ok with it if that means they make more money because of mods I might produce for free. Modding is a hobby, you know.

So IMHO buying games which are easily moddable incentivizes the game industry to make their games more moddable, which is a *good* thing in my book.

There are lots of wrongs in the game industry. A company making a profit because of mods on their game is not one of them IMHO.

Good writing. I agree that a high quality, comprehensive and well-packaged skyblock game could attract some Minecrafters but there are limits to this reasoning as well:

- Minecraft’s items and mechanics is in itself an incredible strong convention that any modded game builds upon whereas Minetest Game can’t match that and thus mechanics aren’t as self-explanatory. One solution would be to build the skyblock game for MineClone 2 instead, bringing back all the conventions necessary for MC audience adoption.

- Most Minetest modders aren’t interested in bringing games to fruition but are either satisfied by designing mods around Minetest Game or incapable of designing coherent game experiences. I believe this is due to people not taking the time to coordinate cooperative game development (it’s hard) with a few exceptions (Farlands). A skyblock game development would have some directions from the get-go so perhaps cooperation in that case could be more feasable.

- Servers. For greater adoption there needs to be some simple services to spin up a skyblock server on demand, as in the MC ecosystem.

Also, check out SkyFactory built with the FromScratch modpack which is (currently) a clone of Ex Nihilo, mentioned in your post.

Making a moddable game requires additional work to be done.
For example, to make maps for Warcraft 3, you need to use special Map editor.
It is rather complex application which can be compared to the game itself.
So if you are buying moddable game, you are buying two applications instead of one.
Though, some people use only one of these apps.

texmex wrote:Minecraft’s items and mechanics is in itself an incredible strong convention that any modded game builds upon whereas Minetest Game can’t match that and thus mechanics aren’t as self-explanatory.

They're not self-explanatory, just well-known. I think originality is the better approach as it gives existing Minecraft players a reason to try Minetest, and overall it makes Minetest stand out for its uniqueness rather than its weaknesses (this is less true for a cross-platform version, but that gets the added benefit of additional feedback). Good game design and documentation are the right way to fix learning issues. A great example of using game design to teach the player is explained in Egoraptor's video on Megaman X (warning: language). Obviously doesn't work with 100% of cases, but it's a good thing to strive for immersion-wise, and would work well, I think, in the minimalist setting of starting from nothing.

texmex wrote:Most Minetest modders aren’t interested in bringing games to fruition but are either satisfied by designing mods around Minetest Game or incapable of designing coherent game experiences. I believe this is due to people not taking the time to coordinate cooperative game development (it’s hard) with a few exceptions (Farlands). A skyblock game development would have some directions from the get-go so perhaps cooperation in that case could be more feasable.

I can totally relate to that from a motivation perspective.

texmex wrote:Servers. For greater adoption there needs to be some simple services to spin up a skyblock server on demand, as in the MC ecosystem.

Sure, why not? I'm more of a singleplayer person myself, but I know there's people who prefer multiplayer. And, in fact, Minetest may have an advantage here, because the inconvenience of downloading a modpack can be a huge deterrent to jumping onto a modded server.

texmex wrote:Also, check out SkyFactory built with the FromScratch modpack which is (currently) a clone of Ex Nihilo, mentioned in your post.

Kinda dissapointed to see another Minetest creation using the same name as something that already exists, although I understand the appeal from an SEO perspective. Still, it looks like a good amount of work went into it, so I'll definitely check it out.

Xudo wrote:Making a moddable game requires additional work to be done.
For example, to make maps for Warcraft 3, you need to use special Map editor.
It is rather complex application which can be compared to the game itself.
So if you are buying moddable game, you are buying two applications instead of one.
Though, some people use only one of these apps.

Minecraft doesn't have an official modding API. The modding API is 100% community maintained and its success compared to modding in other popular games can be attributed to the coincidence that its Java program is easy to modify. So Warcraft 3's map editor might add value but that scenario doesn't really apply to Java Minecraft's modding scene. Most modding scenes are unofficial, and a map editor isn't really a modding API although that's more of a technical point on the type of content creation.

asanetargoss wrote:They're not self-explanatory, just well-known.

This is what I meant of course: "self-explanatory" because of prior knowledge from vanilla MC.

asanetargoss wrote:Good game design and documentation are the right way to fix learning issues. A great example of using game design to teach the player is explained in Egoraptor's video on Megaman X (warning: language). Obviously doesn't work with 100% of cases, but it's a good thing to strive for immersion-wise, and would work well, I think, in the minimalist setting of starting from nothing.

Agreed! For the purpose of focussing strictly on game design, game project collaboration and shipping (sub)games I've been toying with the idea of creating a separate space for Minetesters also interested in actual game design, sort of a "Designer's Lounge". But more important than a separate discussion fora (not even sure it's a good idea to fracture discussion this way) are tools enabling game design collaboration and development. Perhaps Github is enough, but perhaps it isn't. Any tool that helps articulate a game design is a good tool in this case.

asanetargoss wrote:Kinda dissapointed to see another Minetest creation using the same name as something that already exists, although I understand the appeal from an SEO perspective. Still, it looks like a good amount of work went into it, so I'll definitely check it out.

I think the creator isn't striving for simply a 1:1 clone eventually, but it's a true and tested model for a first milestone.