What's Minetest still missing over Minecraft?

[thomasthespacefox]

one of the main advantages Minetest has is the Built in mod support. so the mods don't all break with every update like with a certain game i know...

also Mesecons has more features than redstone
ex you can easily convert redsone schematics to Mese schematics with a few tweaks. but the opposite can be almost impossible with all the microcontrollers, digilines, luacontrollers, and logic gates

not to mention how much space insulated mese saves.

i agree withtrev on the fact that minecrat is a commercialized mess of code that has been cash cowed to near oblivion.
unified inventory is a great help when im trying to find blocks i also like the apps that work with it like the armor mod.
the reason we Minetest folk have such nice things is because we are an open community. minecraft, well you need to pay for it for starters, and the code is now hidden behind Mr Big "company that named a operating system after a glass opening in a wall to let light in" corporation.
minecraft has a mid-evil age feel to it with a corporate mindset to match.

Minetest has a open, modern feel to it, with a community that cares.
Minetest seems to reflect the real world a little more. know what i mean?

[Linuxdirk]

On a technical side … it misses a proper lighting engine. (No, just altering texture colors is NOT lighting.)

[thomasthespacefox]

true... but the great thing about free open scource software is that it's alot easier to get trivial things like lighting done!

[ExeterDad]

true... but the great thing about free open scource software is that it's alot easier to get trivial things like lighting done!

Provided one of the handful of devs is interested and or capable of doing the task.
They say "Many hands make light work." This is true and is one thing MC has. But with Minetest... few hands are doing a huge amount of work. And that's pretty impressive when you look at it from that angle.

[jojoa1997]

true... but the great thing about free open scource software is that it's alot easier to get trivial things like lighting done!

Yeah but I have also seen people get pushed away because their code which fixed the problem was not added due to a dev wanting to do it their own way later. I still dont see lighting in the game

[thomasthespacefox]

true... but the great thing about free open scource software is that it's alot easier to get trivial things like lighting done!

Yeah but I have also seen people get pushed away because their code which fixed the problem was not added due to a dev wanting to do it their own way later. I still dont see lighting in the game

that can happen but Most of the people here just want to have fun and make a great game!

[Calinou]

On a technical side … it misses a proper lighting engine. (No, just altering texture colors is NOT lighting.)

I don't think Minecraft has real hardware lighting either.

[4aiman]

Minecraft is always different, always evolving. That is what Minetest is missing.

To begin evolving Minetest need to have a good default game.
A game is what users see in the first place.
But minetest_game would be of interest only as a "creative" game. (Which is quite nice for touch-screen devices!).
On PC creative becomes a "lego".
Not bad, true.
But in what did LEGO evolved?
Action.
Be it console, PC or "tangible" games to play in real world.

So there should be a game of that quality.
Here I'm trying to make one:

It's very MC-like algorithm-wise but still MT-like surface-wise.
I believe *that* is a kind of a game MT supposed to have.
And, since it's useless to make anyone to feel the same way, I'm just doing what I can. Those who would like my game will eventually join.
Right now I'm enjoying the process.




In the case Minetest to be treated as the engine, "evolving" would mean removal of "broken-by-design" implementation and making a good alternative (get_player_control, player physics, mesh animation etc etc).

The concept-decisions are hard to make, true. That is why Freeminer exists.
But it turned out that FM is not an engine I wanted - The direction it took IS interesting, but eventually it turned into an engine that won't meet *my* expectations.
License-wise and direction-wise.

MT still tries to be an engine, on the other hand.
That's why it still can meet my expectations as well of all those people out there.
The only thing hindering its "evolution" is the system of approval.
It's not a FM-like "majority wins" but "all must agree".

Note, I'm *not* judging the system.
I'm only hinting that many questions are being discussed over and over and over and over...
Those are "hanging in the middle of a room" - just like a semi-empty balloon with helium.

We must wait for ideas to be perfected. Minetest development is not a paying job for modders like it is for equivalent development of Minecraft. With more time, we will see many things turn Minetest into an amazing engine that powers many servers. I hope for rare biomes to be added, mobs with more complex AIs, and hunger to be added to most survival games. With each release, the sense of adventure that one feels playing the game grows. Before too long, Minetest will catch up and be the world's best free and open source sandbox!

Hunger and mobs won't be added to default MT game easily.
There are dozens of hunger mods.
Very simple (rouge-like) ones, middle-ware (Blockman's HUD, "farming" mod expansions) and MC-copying ones (4hunger, part of Magichet).
None of those was used.
Why?
I was told it's due to the fact minetest_game is not supposed to be a bloated platform for modding.

Quite justified opinion, if you ask me.

But then... where those "additional" games to be shipped with minetest releases?
I'll tell you where.
There are almost no such games.
All but several WIP projects are building upon default minetest_game by throwing in some mods.

Server-owners are kind enough to code things occasionally, though.
VanessaE, TenPlus1 and many others *do* create new mods or improve existing.

The problem is, those mods are fully compatible with every game.
So, we have a good choice of universal mods.
But we also have countless "clones".

It's hard to be unique, but It really pays off.

Magichet is a thing that is*not* compatible with most mods out there.
But that enabled me to create "connections" between mods. I'm not forced to support minetest_game or any mod made for it and/or add hacky support (like external_modifier.lua in Specialities mod) for it.

There's no way to get rid of any "clones" if minetest_game won't become a set of APIs to use (not a game).
But that approach was forfeit. For me it seemed only logical, though. Demands for API are much more higher than for a subgame. "Common" subgame, in turn, didn't want to become a set of APIs.

Maybe it all was due to a complete different set of reasons, but who cares?


Clones or not, but everyone want a good game to play.
Sounds simple - just throw in a set of well-tested mods for everything and make ppl disable those (MC modders do it this way).
Preferences do differ, though...

So... what will be the platform to bulid upon?

While anyone can make yet another "clone" it's difficult to decide.
But certainly *not* the minetest_game.
Minetest_game has no feature.
It loses to every game built upon it with mods.
Either in "lightweightness" or feature-wise.
Moreover, games like Carbon wouldn't have appeared if minetest_game had had a wider set of mods.
On the other hand, "bloated" default game doesn't seem like something devs would agree to have.


I think everyone interested should pick subgames he/she like and support development of those.
I.e. Instead of improving mods for *every* subgame modders should make mods for a *certain* subgame.

Your phone or window isn't wide enough to display the code box. If it's a phone, try rotating it to landscape mode.

Code: Select all

It is good to have multiple games support, but functionality should prevail over universality.
To the extent where no other subgame would be able to use the newest changes.
If a mod is "universal" it only means that there's no unique game in that aspect at all.
The more "universal" mods are in the "game" - the more it becomes a "clone".

you can easily convert redsone schematics to Mese schematics with a few tweaks

Yep. There are always just a "few tweaks"...
It will take more than that to create a copy of MC Redstone.
Some "features" of Mesecons are just a simplified version of Redstone mechanics.
It was logical and convenient to code the Mesecons mod the way it is.
But a *true* copy of Redstone will require either tons of additional checks in the signal processing or completely re-writing a good part of it.

I don't think Minecraft has real hardware lighting either.

MT won't have even shaders, let alone HW lighting...
OGL, OGLES1, OGLES2, DX, SW, BV - it's impossible to have one and the same shader for all those renderers.
MC uses only one renderer.

Complexity of MT sometimes is a disadvantage.


But what bugs me a lot is why some features (like OGL-only shaders) were added where other half-made features doesn't (or got removed).

[TheEpicJames]

this game really needs a way to use skins,
something like a skin system similar to Minecraft PE's skin system would be nice.

[thomasthespacefox]

this game really needs a way to use skins,
something like a skin system similar to Minecraft PE's skin system would be nice.

currently all the existing skin mods are on the server end meaning if you want a custom skin on a server you would need to send it to the admins manually.
server>client where > is the direction the skin is sent

what you are discussing i presume, would mean the client sending the skin server-side I'e like this:
server<client where < is the direction the skin is sent

someone correct me if I'm wrong but this would require a change to the engine wouldn't it.

minetest does have some mods to use skins but no way to send them to a server from a client.

[AnxiousInfusion]

To quote a core developer:

It's literally just sending a few hundred bytes to the server, nothing tricky about that.

Of course, he is only speaking about sending skin files from client to server, not the entire implementation.

[thomasthespacefox]

To quote a core developer:

It's literally just sending a few hundred bytes to the server, nothing tricky about that.

Of course, he is only speaking about sending skin files from client to server, not the entire implementation.

the tricky part is making a minetest-wide skin system in minetest so minetest can actually send those bytes. and of course it would need to have a server-side fallback skin (that the server admins can set) in case someone is not running version (insert version this would be implemented here) or later.
it would also need to send the skin to other players already on the server.
How that would work? I will leave that up to the devs.

[Sokomine]

I think everyone interested should pick subgames he/she like and support development of those.
I.e. Instead of improving mods for *every* subgame modders should make mods for a *certain* subgame.

I do have a diffrent opinion in that regard. Mods that can be more or less universal ought to. And if a mod can be used in combination with various subgames, it leaves the player freedom to choose which subgame and mods the player likes best. MT is about everyone beeing able to create his/her own set of mods and distinct subgame. But with all freedom, there's the drawback of having to choose.

Players sometimes ask for more polish of subgames, more fitting textures and so on. I'm sure any help in that regard will be welcomed by the subgame developers. Making a subgame look like it's something connected (and not just a collection of mods) is an important task as well, but it's the task of creating/maintaining/extending a subgame - and not a mod. If mods support such tweaking by subgames, that might be best.

[4aiman]

@Sokomine:
I agree that universal mods give freedom.
I also agree with you while mods are concerned.
But so far there are over 100 games with only 5(?) of them being not a bunch of mods.
In terms of evolving MT needs a great subgame. This is where "mod development" becomes "subgame development".

In other words, if there would be unique incompatible games, it would be still possible to write a universal mod, but MT will have several improved "platforms" for modding.

I hope that a situation like that will help even clones to be more polished. ;)

Regards!

[Linuxdirk]

I don't think Minecraft has real hardware lighting either.

They don’t but however their lighting engine is still better. Just compare sunset and sunraise in both games.

https://youtu.be/rhdTgGhGMmQ?t=48s

[4aiman]

While I wasn't mentioned, I'd like to comment:

I can't see any real difference aside from textures and speed of the sunrise itself (seems to be a little bit slowed down).
Where's the catch? :)

[SAMIAMNOT]

I agree that another subgame should be added, but both Minetest_game and Minimal should stay, because they might run better on weaker computers. However, we're likely months away from them (the devs) deciding to add a game and the process of choosing which game to add would likely be long.
That thing about Minetest being bloated is rediculous though. Look at the Mac OS builds, they come with 4, 5 games, including Overcraft, which is quite large.

EDIT:Also, Minetest shouldnt try to be like Minecraft. Minetest is best when it comes to uniqueness. For example, I like foghting Dirt Monsters and Sand Monsters and Tree Monsters from Mobs Redo better than I like fighting Zombies from Simple Mobs.

[benrob0329]

Polish, thats about it...with the skinsdb mod people can upload there skins to the database and it will be on the server.

[programmingchicken]

We need to part from MC completely, b/c MC *might* be going downhill and we need to become our own game.
Who friggin cares if minetest doesn't feel enough like MC. MT AIN'T MC

I think that RBA has been stalling the lighting overhaul for a LOOOOONG time now.
We need it. And we need it BAD.

[benrob0329]

YES!!!!!

[SAMIAMNOT]

...we need to become our own game.
Who friggin cares if minetest doesn't feel enough like MC. MT AIN'T MC...

Amen!

...I think that RBA has been stalling the lighting overhaul for a LOOOOONG time now.
We need it. And we need it BAD.

Yeah, finally a meaningful update! Unfortunately, this would likely push the release back and I really can't wait for the new /spawnentity command.

[programmingchicken]

...we need to become our own game.
Who friggin cares if minetest doesn't feel enough like MC. MT AIN'T MC...

Amen!

...I think that RBA has been stalling the lighting overhaul for a LOOOOONG time now.
We need it. And we need it BAD.

Yeah, finally a meaningful update! Unfortunately, this would likely push the release back and I really can't wait for the new /spawnentity command.

lighting overhaul which may add cool shader mods and ambient light versus a builtin /spawnentity command that you could probably install a mod for.
Honestly I want DM's and rats back into the game with 3D textures.

[rubenwardy]

There is absolutely zero chance that a lighting overhaul will make it in into 0.4.13, given that we're currently in feature freeze (only bug fixes are allowed, no new features or big changes)

There are more important things than lighting to do at the moment, after 0.4.13 is over, such as fixing the networking code and adding client side lua. I'd currently prefer RBA to work on improving the minimap.

[benrob0329]

Speaking of RBA, I rebuilt my MT and the new ore textures didn't pop out like they did! I have bumpmapping enabled, but something changed in the normalmap...

[4aiman]

Minetest is best when it comes to uniqueness.

You're so wrong here, SAMIAMNOT...

@2 everyone
Minetest as a sanbox engine (quite good, if you ask me) IS unique. Nothing can change that.

But what IS ridiculous it's to compare an engine and a game.

That being said, in terms of *game* uniqueness, Minetest is a large number of clones.
Or, to satisfy more "precise" view of some community members, Minetest is the *family* of subgames.
Those are not "more different" than a son and his father.

It's not Minecraft and never will be.
But the point of this conversation is what MT missing from MC.
So, why not follow the thread's title and discuss precisely that?
After all, in the case some MC feature would be considered "lame", then, I guess, MT does NOT miss it from MC at all.
End of story, bring in next MC feature and discuss it :)

So instead of posting "no MC, please" type of messages let us focus on what can and *should* be "ported".
And please, don't tell me that you don't want *any* MC-like feature.
'Cause if you do, you should get rid of all tools, farming, crafting grid, "creative" mode and much more.
I'm not saying those are MC inventions, but it wasn't me who don't want MC-like features added even as an completely optional mod.

Regards!

[benrob0329]

Polish,
head movement (I don't think that that's very reasonable with Irrlicht)
Lighting
Aaaand...um....I can't think of anything....

[4aiman]

> Polish
It's is a bit too vague.

I'm going in the direction of polish ~= "make smarter, more consistent, add features, let players do MORE than just playing with digital LEGO".


>Head movement
Well, that is entirely possible w/o any changes to Irrlicht.
But while it would be nice to have moving head, the game (and I mean *any* subgame) is not yet ready for such changes. It would be as weird as having Ultra HD textures for mobs and 16x textures for the environment.

>Lighting
+ 100500
Too bad it will be a major overhaul.
Knowing that there were some snippets done, I guess that may take a year or so.

[benrob0329]

By polish I ment things like head movement, lighting and such. I would help with lighting but my C++ skills are fairly non existant right now...

[4aiman]

Then, if I were the one to decide, this kind of "polish" would be a no.
Only because those are not essential to make a fine subgame.
I'm all for new lighting system, but visual prettiness should be improved when everything else is just "cool".
Currently some pretty textures + 3d armour + visual wielded item will suffice.

I would help with lighting but my C++ skills are fairly non existant right now...

Heh... Once upon a time I was disliked to the bottom of the reputation well for this kind of a message. )))

But in all seriousness, knowing C/C++ syntax and some compiler tricks won't add much to your ability to code lighting for Minetest. You'll need to learn Irrlicht as well...

[AnxiousInfusion]

Heh... Once upon a time I was disliked to the bottom of the reputation well for this kind of a message.

Story time!