What's Minetest still missing over Minecraft?

[gunvolt]

Since Minetest has shaders, I think it needs fancy lighting as well.
Otherwise, as a game engine, it misses little, except for: built-in attacking-enemy support; built-in hunger/stamina; interface re-theming; block breaking creates dropped items; automatic pickup of dropped items; and extra ease of use (I.E. don't show a full block of JSON for an entity when hovering over it, just say the name of it)

[burli]

If you like all that, create a subgame, Name it Minecraft and add all the mods you are missing. Well, except oft the fancy light shader

[rubenwardy]

Please keep your dropping items well away.

Better ui and teeming would be good.

(I.E. don't show a full block of JSON for an entity when hovering over it, just say the name of it)

Sounds like you have debug mode on (f5). In 0.4.14, a dropped entity's name will be shown when debug is off. Other entities can pick the message which is shown. It's like infotext, but for entities - infotext is the text that appears when hovering on a furnace, sign or locked chest.

[orwell]

One thing that Minetest is definitely NOT missing over MC: Infotexts.
We need infotexts for players! Would add nice possibilities.

[benrob0329]

Players are entities, so technically they could have sign like infotexts (latest git entities can have sign like infotexts)

[Don]

Please keep your dropping items well away.

I don't like dropping items either. What I would like is to have items be picked up when you walk over them. Would be great for saplings and fruit after cutting down a tree. Would be better in Minetest Game then having to use a mod.

[MineYoshi]

Please keep your dropping items well away.

I don't like dropping items either. What I would like is to have items be picked up when you walk over them. Would be great for saplings and fruit after cutting down a tree. Would be better in Minetest Game then having to use a mod.

+100!

I totally agree with Don.

[cy]

What I would like is to have items be picked up when you walk over them.

This is the one I like. Works good, not too heavy load on the server.

Now all I need is to figure out how to get it working with hoppers.

Would be better in Minetest Game then having to use a mod.

I don't really agree it should go in the main game... but there is a mod for it, at least. Just wanted you to know.

[cy]

Oh, and to add my 2c in, how to make minetest fun.

1. On-demand content loading. Nobody uses Ambience because it takes frikkin forever to download all the media, because someone had the brilliant idea to delete the downloader after the game's started, instead of figuring out how to download more after starting.
2. Client-side logic. Everyone writes mods and stuff for servers, since they're control freaks and don't want to let clients decide what accelerates where. But seriously, the game would be a lot better if clients were trusted more, and less load went on the servers. For instance...
   • Client-side entity motion logic, at least of the passive stuff like pipeworks. All the server should need to do is say "here's path A" then "entity B goes down path A, arrive in 10s" and then just wait 10s.
   • No anti-cheat. It's nice that nobody can use a hacked client to teleport, but it's not worth causing jerky backtracking when a laggy server throws a fit over players moving at normal speeds. It'd be better to just report speedhax in the logs, so you can ban if you care. Jerking players around just helps the speedhackers figure out how fast they can get away with.
   • The anti-speed-dig might be more important, but make it heuristic for goodness' sakes. I've used a good pick to dig out an area, only to have myself entombed in stone, and sitting there watching the laggy server replay my actions block, by block, one block a second. At least it should only check dig rate over the nearest 10 blocks, instead of each and every single one.
   • Better client-side animations. Let the clients perform the FSM, instead of just passively doing the next animation the server sends. All the server needs to know is the mob's proximity, and it's logical state. What part of its animation is running (the "walk" part or the "swing" part) and how fast shouldn't be transmitted every time.
3. Animations paired with sound effects. Currently minetest has set_animation, but then you have to manually figure out when to play the "oof" "ugh" "clonk" or "boing" sound. There should be a way to group those together better. And have the client do it, so the server doesn't have to send two messages "get hit" and "say ugh" and pray they both arrive at the same time.
4. Higher saturation. Seriously, minetest suffers from "reality is brown" syndrome.
5. Less ugly, useless junk. Mostly this is a server admin choice thing, but there are a lot of useless, ugly mods, some enabled by default. Poison ivy. Dryplants. Ferns. Molehills. Woodsoils. Young trees. (gee is this all in the plantlife modpack?) Some are downright infuriating, like "trunks" that make running past a tree impossible, and how "bamboo" can't be easily cut away.
6. Broader biome variation. Perlin is terrible for picking out biomes, because you end up with one normal biome, and every other biome is concentric rings around it (I'm looking at you, Ethereal.) We need an algorithm to divide a map up into irregular splotches, and divide those splotches up into splotches, and do biomes that way. As-is, adding mods just makes your surroundings look busier, but when you run around on the surface, it's just more of the same.
7. Technic. Minetest lost a ton of appeal for me when technic crashed and burned. We need more automated stuff, more chain reactions and time savers. People say if you don't want a challenge then go on a creative server, but there's something amazing about simulating that stuff and seeing your work proceed logically, instead of picking out the thing you want on a random menu.

Vanessa once said that what's screwing up Pipeworks (and therefore technic) is not having a drawtype for pipes, so needing a ton of different pipe objects/wires/etc ends up bogging down the server... or something. I think it's the server-side entity movement that's doing it, but either way, something needs to be done that gives us some form of technic back.

[Ferk]

what's screwing up Pipeworks (and therefore technic) is not having a drawtype for pipes

Actually, I think the connected nodeboxes from latest release could help there. You'd only need one node and configure it to connect itself to nodes of the given groups. The connection logic is just done client-side. You'll still need extra nodes for maybe some special shapes, though

But I'm not so sure about that being the biggest resource hog. I also think it's more about the server-side entity control, and in general all the game logic that is exclusively done server-side.

I know that relying in the clients is very complicated for things in the game world that need to be in sync. But at the very least the formspec logic and all GUI elements should be completely client side. At least that would be a start. It makes little sense that every little interaction in a formspec results in a back and forth with the server. It should be more like in a web browser, where the javascript decides when something needs to be sent. Maybe the only exception being editable shared inventories that actually need to be synced.

[orwell]

Oh, and to add my 2c in, how to make minetest fun.

1. Client-side logic. Everyone writes mods and stuff for servers, since they're control freaks and don't want to let clients decide what accelerates where. But seriously, the game would be a lot better if clients were trusted more, and less load went on the servers. For instance...
   • Client-side entity motion logic, at least of the passive stuff like pipeworks. All the server should need to do is say "here's path A" then "entity B goes down path A, arrive in 10s" and then just wait 10s.
   • Better client-side animations. Let the clients perform the FSM, instead of just passively doing the next animation the server sends. All the server needs to know is the mob's proximity, and it's logical state. What part of its animation is running (the "walk" part or the "swing" part) and how fast shouldn't be transmitted every time.
2. Animations paired with sound effects. Currently minetest has set_animation, but then you have to manually figure out when to play the "oof" "ugh" "clonk" or "boing" sound. There should be a way to group those together better. And have the client do it, so the server doesn't have to send two messages "get hit" and "say ugh" and pray they both arrive at the same time.

We really, REALLY need client-side lua...
Does someone have too much time?

[Fixer]

ambience, survival (hunger, sprint, mobs), cornerstairs, weather, boost carts, mesecons (those exist in mods and can be incorporated into mtg to have near minecraft experience)

[benrob0329]

Why do we need to have a "near Minecraft experience"?

I don't think we should try to be MC, nor do I think we should try not to be MC. I think we need to be our own thing, and that when we stop worrying about how we're better/worse than the 10,000+ MC & clones and just focus on how do we make Minetest better, that we'll be recognized as our own thing (*couph*Wikipedia*cough*)

[TumeniNodes]

Why do we need to have a "near Minecraft experience"?

I don't think we should try to be MC, nor do I think we should try not to be MC. I think we need to be our own thing, and that when we stop worrying about how we're better/worse than the 10,000+ MC & clones and just focus on how do we make Minetest better, that we'll be recognized as our own thing (*couph*Wikipedia*cough*)

I agree 100%

[mrmister]

Why do we need to have a "near Minecraft experience"?

I don't think we should try to be MC, nor do I think we should try not to be MC. I think we need to be our own thing, and that when we stop worrying about how we're better/worse than the 10,000+ MC & clones and just focus on how do we make Minetest better, that we'll be recognized as our own thing (*couph*Wikipedia*cough*)

THIS

We should really focus on making a better minetest instead of being a minecraft clone. Its already different to minecraft at its core since its features is mostly made up of lua mods.

However, it isnt wrong to add a bunch of features that minecraft has as far as tools, weapons, blocks, mobs, etc.

Wanna repair items with an anvil? Sure, make a mod for it. Hunger and experience points? Sure. Even when it comes to mimicking popular mods in minecraft like Smart Moving, Mine and Blade, and others. Just remember to keep the minetest way, which is to have the freedom to customize minetest easily, extending its features with mods.

[burli]

Why do we need to have a "near Minecraft experience"?

I don't think we should try to be MC, nor do I think we should try not to be MC. I think we need to be our own thing, and that when we stop worrying about how we're better/worse than the 10,000+ MC & clones and just focus on how do we make Minetest better, that we'll be recognized as our own thing (*couph*Wikipedia*cough*)

I don't need a "near Minecraft experience". I don't want to be better than MC. But it shouldn't be worse just to be "different" from MC.

Example: eating. I really hate the MT solution because I really often eat accidentally and waste food. In survival a bad thing if you run out of food because you eat when it's not needed.

Another example: drop and pick. I never really played MC. I just tryed the demo for a few minutes after I played MT for a while. But I never liked the MT behaviour.

So as long as we don't find different solutions we should copy them from MC, if they are better than the current once from MT

[TumeniNodes]

What's missing in comparison to minecrosoft?
The ability to pay $30.00 for a single install license. (one device per copy)
The ability to pay an additional $30.00 for a "learn how to mod" service... (same single license applies)
Did I miss any additional fees?

Minetest is opensource, community driven, has developers who work hard voluntarily, it's easy to mod and, if not so easy for some..., there are other users and devs who are always willing to help and answer questions, and even to go as far as generously offering their time to contribute to your work.
In my own opinion, Minecrosoft should be far more impressive than it is, when one considers the insanely, enormous amount of $$$ which has been made from it.
When cost and finances generated from Minetest are compared to the other..., I would say Minetest blows them out of the water in several areas.
This entire statement is merely my own opinions and not related to any opinions or feelings of anyone else in the community or the developers of Minetest.

[burli]

In my own opinion, Minecrosoft should be far more impressive than it is

Well, you are right if you talk about the vanilla game. I would expect more than that. And it is still not stable and has performance issues, even on a high end PC.

But modpacks like Life in the Woods are really impressive. It looks great and there is so many to discover

[shaalazin]

Minetest is missing a game. This is a great engine. I can't comment on if the game is great or not; I can't find one. I found this 'game style' called 'Minetest Game.' But, upon 'playing' it, there doesn't seem to be a game to be found there. There just seems to be a lot of testing resources for the engine designers to test their engine with.

I tried playing the game online. But, there doesn't seem to be a game to be found online either. Just a bunch of servers running this engine test called 'Minetest Game' with varying grab bags of mods thrown in. All the mods made by different people with no cohesive art direction, quality control, or purpose coordinated in their designs. It's like everyone was just doing their own thing, and no one was focusing on actually making a game.

Minetest does not come with a vanilla game. The mods are fun, but slapping a bunch of mods (made by a bunch of people not cooperating together for any one specific art design, purpose, or quality control) together doesn't feel like a cohesive living world that I want to be a part of; like Minecraft does. Instead, it feels like a mess. A fun mess. But a hot mess. You don't have a game here with Minetest. You have a fun, hot mess here with Minetest. Minecraft has a fun, cohesive game.

Your engine is amazing. It's the most complete Free Software voxel game engine I've ever found. The engine is made superb, made to be extendible, from the ground up. I do not fault one single thing you've done with this engine. This engine is amazing.

But, a game an engine does not make. It's very clear that the team here is focusing on making an engine, not a game. Because there is no game here. Maybe they plan to make a game once the engine is done? Maybe you should have another team focus on making a game with your engine. Isn't that basically what Voxelands is doing?

That's what this engine needs. It needs a separate team that focuses on actually building a game with this engine. That's an entirely different thin than a bunch of modders. An engine alone doesn't make a game. Nor does an engine with a bunch of mods added on make a game. A true game comes from a team of coders and artists all working together with an engine to create one unified vision of a playable game that creates a living world that players want to be a part of. You just won't ever get than from an engine standing alone or with varying mods from varying people.

You guys keep up the great work with the engine. It's amazing, and going places. I'm going to check out Voxelands some more; and I hope they make a great game out of their fork of your engine. Maybe that's the relationship between you and them. Maybe they'll be the people to turn your engine into a game. Or, maybe some other team will come along and do that. Or, maybe your team will make a game out of this engine at the end. But, right now, this is an engine; and nowhere near close to a game; with or without the many mods.

Whereas, by comparison, Minecraft is a complete game, modded or not modded. Minetest is, well, a test; an engine building test. It's not a game. Even the game is called 'Minetest Game.' The test in the name showing that it's not a complete thing yet either.

You can't look at an engine, compare it with a game, and then ask people what's missing. What's missing in Minetest is the game. That's what's missing. It's not a fair comparison at all. Because these people have made a great engine; and some modders have made great mods. But Minetest, in its current state, can't stand up or hold a candle to Minecraft in any way shape or form at all. Because Minecraft is a game, Minetest is an engine test; not a game.

That's my view on the issue. But, it's the best Free Software Voxel engine out there. Why do you think Voxelands forked from you? Because you're engine was worth forking from. Nobody forks from an engine that isn't any good. But, from reading what they had to say, they're goal is to actually make a game. Something I can truly say Minetest isn't, at least not yet.

[burli]

I read your post, buy I still don't know what you mean. What do you expect from "a game" ?

[shaalazin]

I read your post, buy I still don't know what you mean. What do you expect from "a game" ?

I don't know, kind of hard to tell. I guess there are three things I feel are missing.

1: Unified design (artistic and code.)

2: Vastness of Content

3: Popular Online Community

In detail.

1: Game should have graphics and sounds all made by a dedicated team of people. That way they all share the same style. Or, at the very least, there should be some sort of content guidelines imposed on mods; if you're going to depend on modders to build the game aspect for you.

The terrain looks blocky, almost Minecraft blocky. That's good. Animals look super realistic, that's very, very bad. They should also look blocky like the terrain. Seeing animal mobs done as realistic shapes and not blocky like Minetest and Minecraft is a really bad design choice because it pulls me out of the game entirely. Someone should have talked with the animal mob mod designer and explained to him/her that his/her art style is clashing with the rest of the game and that he/she should design that animals the correct, blocky way; not the realistic, wrong way. If someone is going to make animal mobs that are so obviously not the right art style for this game, they defiantly shouldn't be featured as the main animal mod pack on the game's customize, mod dropdown menu page. I was looking for animal mobs that fit the design of Minetest/Minecraft; not things that look like they belong in an entirely different game.

2: Then the mob pack for creatures. It looks great. Too bad I can't figure out how to open it. Is it only for older versions of Minetest? It keeps erroring out when I try to apply it to the newest version. When I open up Minecraft, I have horses, sheep, chickens, farming, crafting, a magick table, so on, so on, so on. There is so much content in Minecraft, I've been playing it for about a year in both single player and online and I don't think I've scratched 10% of the types of things that I can mine, craft, build, and ride, combine, and enchant.

Honestly, I felt bored in Minetest in about 5 minutes.

Minecraft = Played for more than 1 Year and discovered what feels like less the 10% of total game.
Minetest = Discovered total game in 5 min. Watched a youtube video showing my block types, dug dirt. There's the game.

There are no mobs. There are no animals. There are no creatures. There is no hunger. There is no anything. It's not a game. Everything Minecraft has, or at least the same amount of stuff, should be in this game by now. It's been in the making for years; and it still lacks so much in the way of content.

I want to enter into Minetest and feel like I could play the game for 72 hours in a row and not feel like I'm anywhere close to finding everything I can do with it. That's what Minecraft still feels like, a year later. I'm still finding new surprises in Minecraft. Minetest, I can dig stuff in the ground. Everything that can be made or found can be shown in a 5 min youtube video; and has been. Cool rainbow cat. I shouldn't feel like a 5 min video can encapsulate everything in your game. But it does. It really, really, does. You can't squish the whole of Minecraft into a 5 minute review.

Minetest Wiki says Minetest started in 2010. Wikipedia says Minecraft was first released to the public in 2009. That means Minecraft only has a 1 year head start on you. Just 1 year.

So, in all reality, Minetest should almost mirror Minecraft by now. It's been in development for almost the same amount of time. Therefor, it should have almost the same amount of features. Yet, and this is going to sound mean, Minetest still looks like a year 1 project. It doesn't look like a year 6 project. I have to wonder what you guys have been doing for 6 years. Making an engine, clearly. Making a game, not at all.

There is absolutely no reason why you shouldn't have just as many or more mobs, blocks, block types, horses, biomes, and everything else as Minecraft does by now. Yet, I think Minecraft had more to do in it by year 1 or 2 than Minetest has now at year 6. What's the problem with this project? Why is it so woefully behind Minecraf? It's really sad, honestly.

Look, I'm not a programmer. So, I can't speak to that. But why wasn't this game pushed on kickstarter as the Free Software, GPL, mod-friendly alternative to Minecraft back in the years where it could have pushed Minecraft off of its throne and outdid Minecraft at it's own game?

Everyone loves Minecraft. Everyone loves that you can mod Minecraft. Everyone hates that Minecraft is proprietary. Everyone wishes Minecraft was GPL.

BOOM, kickstarter for money to fund more programmers and graphic artists to get to work on Minetest. With a reasonable kickstarter goal and a reasonable timeframe so that Minetest would be more content filled than Minecraft, and be GPL compatible. I think a lot of people would spend $5-$10 to put a Minecraft close with GPL ahead of Minecraft in functionality.

Yet, I have combed the Internet for a game that was Minecraft but not propriarary. The only two real options I find are Minetest and it's offshoot Voxelands. Voxelands has more content, but you're both pretty far behind of Minecraft.

Why wasn't this funded? Why wasn't this pushed? This could have Firefoxed Minecraft. Instead, 6 years later, it kind of just puts onward slowly with a whimper.

I don't know, I'm not a programmer. I have no right to talk. But, I wanted something GPL friendly that I could promote to my friends as being a decent alternative to Minecraft; rather than shoving proprietary software at them. But Minetest doesn't Firefox Minecraft. It doesn't even libre/open office Minecraft. Mintest is a 6 year old project that looks like it's somewhere between years 1 and 2 of development. It's not where I was hoping something like Mintest would be at this point. I see why Voxelands split off and tried to make a game with the engine that was there, rather than working endlessly on revamping an engine that no one is actually building a game for. What's the point to that? Coding practice?

3: Because the game is so far behind, it's severs are pretty empty. Honestly, there's more people here than I think are on Voxelands. But, it's still pretty sad. I get about 15 people if I'm lucky. The vanilla sever I go to in Minecraft gets 2 to 3 times that amount daily. Other servers get far more. People aren't playing this game, because there is no game to play. Don't you think if this game was better than Minecraft, that the modding community would ditch Minecraft for this game in a heartbeat? Just like the users who ditched IE for Firefox, because Firefox was better and respected user freedom.

Again, I don't know how to program. It looks like the work that was done here is very well done. It also looks like it's the best GPL friendly engine out there. But, there is just not a lot to actually do in the game. It's a very empty world. Minecraft feels inviting. Like it's a world filled with secrets to discover. Minetest feels... like I need to add about 5 mods to the game to have something to do, two of the five mods are for older versions and won't run, and even with the mod packs I feel like someone should design about 15 more to get the same level of content Minecraft has. It's just so empty. It looks like it's still in a testing stage; 6 years in...

Hedgewars has been at it for 12 years and basically almost finished replacing Worms Armageddon. Considering it's only 1 year behind Mincraft in development time; how much longer do we have before Minetest is content competitive against Minecraft? (Not counting the mods; the actual game itself.)

[burli]

Well, first of all, Minetest is developed by voluntary in their spare time. They don't get payed. I could say: for all that Money Minecraft should be a lot further than that.

Second: Minetest is a game engine and Minetest Game is only the minimum to have something to play. This is by intention. To make a real game is part of the community and there are a lot of so called subgames out there

But to get a team to make a game is difficult because everyone has different ideas how this game should look like. Most of the developers don't want to make just a clone of Minecraft

The idea behind Minetest is different to Minecraft. It is not to have just one game. It is about to have totally different games. Some people just want to build. They don't want mobs, they don't want to dig. They just want hundreds of different nodes to built. Other like to dig, fight against mobs, play in an ancient or futuristic world. The idea of Minetest is to build your own game by installing the mods you need for "your" game

So, a clone of Minecraft would be just one of many possible games. Which one should be "the game"?

But I agree that lots of the mods need to be polished or completely rewritten.

[Martno]

I think this is promising:
viewtopic.php?f=50&t=13677

[burli]

I think this is promising:
viewtopic.php?f=50&t=13677

Oh, first time I see this. Good to bring that up again

[rubenwardy]

I disagree that there are no online games, capturetheflag is definitely a game. It has a clear goal and different types of gameplay. Admittedly it's not a deep game, but still a game

Yes, VoxelLands has more vanilla content but Minetest has more content because it is moddable

I agree that good mobs are missing, and that MTG lacks good game play