So how many people does this 'game' have playing it currently? Also, has there been significant fallout regarding the release of the game vs. current state?
-
Welcome to rpgcodex.net, a site dedicated to discussing computer based role-playing games in a free and open fashion. We're less strict than other forums, but please refer to the rules.
"This message is awaiting moderator approval": All new users must pass through our moderation queue before they will be able to post normally. Until your account has "passed" your posts will only be visible to yourself (and moderators) until they are approved. Give us a week to get around to approving / deleting / ignoring your mundane opinion on crap before hassling us about it. Once you have passed the moderation period (think of it as a test), you will be able to post normally, just like all the other retards.
No Man's Sky
- Thread starter Blaine
- Start date
Perkel
Arcane
- Joined
- Mar 28, 2014
- Messages
- 15,858
It is from game design perspective.
Only way it can work is when you focus on some city building or something because creative aspect of game fuels gameplay and it is actually tried method that works (like DF). Because you treat pg as tool for making your environment and not as main feature of gameplay.
In this case, "creative" refers to actual creativity. I suppose it's important to distinguish between "create" and "generate." It doesn't matter if you have an entire cluster of supercomputers at your disposal. Computers aren't creative (yet); only human beings are.
Relying on the computer to generate content dynamically can work in certain contexts and to a certain extent, but certainly not for the purpose of generating more than a handful of planets in a game centered around planetary exploration.
Relying on the computer to generate content dynamically can work in certain contexts and to a certain extent, but certainly not for the purpose of generating more than a handful of planets in a game centered around planetary exploration.
But we don't want tolerable.
We want excellent.
And for that we need talented people designing shit deliberately.
procedurally generated content has become a buzzword. it doesn't mean anything more than automated mixing of available assets.
if you create enough assets for the mixer to have more than 2 mixing possibilities, it will mix more. if you just want to use the buzzword for marketing and only create a handful of assets, the mixer will not have anything to mix.
if the mixer also has to keep in mind a few things it's called an algorithm. an example for good algorithms is in most roguelikes. you can recognize its patterns pretty well if you look at terraria world maps. it doesn't mix all available materials completely randomly like a retard but sprinkles them around (procedurally) at different depths in varying amounts, keeping in mind certain biom rules with their specific characteristics.
if you put too little thought into the algorithm it's gonna produce crap. if you give it too little material to work with it will again produce crap. it's not a magic wand for free content. you still have to put a lot of work into it. i can imagine after a certain point of depth it's even easier to do everything manually instead.
i rated your post retarded since it's got nothing to do with computing power. the computer will run an algorithm as long as it has to. so if an algorithm is overly complex it will just take longer. it won't just quit its job and retire.
I don't claim to be well informed about the breakdown of the terminology, but that's the issue: All of the assets and "procedures" (however you categorize them) must be designed by a creative human being. They may be well-designed assets and well-designed procedures, but after a certain point, the assets will repeat and so will the patterns created by the procedures.
I suppose procedural generation works best when its behavior is interconnected with meaningful, active choice-making on the part of the player, or if you prefer, player participation/intervention in the procedural generation. This is usually the case in roguelikes et al.
If the computer is simply spitting out a near-infinite succession of randomly-assembled planets, then the player can only interact with the static results of procedural generation, and not with the actual generation process.
I suppose procedural generation works best when its behavior is interconnected with meaningful, active choice-making on the part of the player, or if you prefer, player participation/intervention in the procedural generation. This is usually the case in roguelikes et al.
If the computer is simply spitting out a near-infinite succession of randomly-assembled planets, then the player can only interact with the static results of procedural generation, and not with the actual generation process.
thought experiment:
2 spaces, 2 characters,
posibilities: oo ox xo xx
rules:
no xx: oo ox xo
no oo: xx ox xo
both: xo ox
already we can see there's a strong limitation on rules based on available assets and space. filling the void with space will allow for much more rules and possibilities but the repetition of assets will be vomit inducing apparent:
4 spaces, 2 characters,
possibilities: xxxx xxxo xxox xxoo xoxx xoxo xoox xooo oxxx oxxo oxox oxoo ooxx ooxo ooox oooo
rules:
no xx: xoxo oxox ooox ooxo oxoo xooo
no oo: xoxo oxox xxxo xxox xoxx oxxx
both: xoxo oxox
no xxx: xxox xxoo xoxx xoxo xoox xooo oxxx oxxo oxox oxoo ooxx ooxo ooox oooo
no ooo: xxxx xxxo xxox xxoo xoxx xoxo xoox oxxx oxxo oxox oxoo ooxx ooxo
both: xxox xxoo xoxx xoxo xoox oxxx oxxo oxox oxoo ooxx ooxo
and these were just simple filter rules. applying pattern filter rules with only 2 assets/characters would not even be distinguishable from simple filter rules. so let's make things interesting:
5 spaces, 5 characters: x,o,f,r,s
possibilities 5x5x5x5x5=3125
rules:
after every f comes an s if the f will not be followed by an o:
xofsx xsfsx xsffo ... there's a lot more remaining possible combinations even after applying a pattern filter rule.
the complexity of the pattern filter rule itself is more interesting than most combinations. with 5 characters and 5 spaces, i think it's pretty obvious for everyone that stuff like xxxxx or ooooo or even fofof or xofsr is pretty boring.
so what we want is as many and as complex as possible pattern rules and as many characters/assets as possible with just enough space that they don't become too repetitive. only being able to walk two spaces to the left or the right is not satisfactory as a dungeon level. neither is a 50000x50000 spaces square filled with only rats, cats, dogs, spiders and bees.
so what is procedural generation good for if you still have to create just as many assets as you would without it and then even keep track of complex pattern rules to sieve the chaos of randomness without the rules conflicting with each other?
well, there's a reason it's mostly used for roguelikes.
a game that kills you before you get bored will have a huge replayability value if and only if it doesn't play the same way when you start over. so random procedural generation is not a cool feature, it's a core element without which perma death would also kill the game for the player. it's a necessity, not a "smart way" to make games.
why perma death? for maximum excitement. the possibility of losing all progress, all gear, all the time you have invested makes for the most exciting games you can ever hope to make. but without randomization of almost every element, perma death devolves into the frustration of having to play the same thing all over again the same way.
this is core ideology of roguelikes. it's also the reason why roguelites fail to be as replayable. they both add a feeling of grinding in the case of saved power progression as well as a much weaker excitement factor due to deaths that don't wipe all progress.
how does this wall of text apply to no man's sky?
simple answer: it's not a good roguelike.
what's random in the game? planets, animals, that's pretty much all i remember from trying it
crafting and power progression based on crafting makes sure that you will always start the game with no hope of finding a random significant power boost. no randomly significant power boost means less randomization. yes, even character progression must be random for excitement and replayability. while perma death makes you fear losing, random character progression makes it exciting to start over. dcss will sometimes let you find a +12 accuracy / +12 damage ring on the first level making you play a warrior for a few levels even if you started as a mage.
the game features several difficulties that don't have perma death. cutting away the only other core element of a roguelike and excitement.
now, without it being a roguelike, what is the random generation supposed to add to the experience? it seems to me it's all about empty exploration. just site seeing of mixed assets. and few assets at that, some say.
thank you, no.
2 spaces, 2 characters,
posibilities: oo ox xo xx
rules:
no xx: oo ox xo
no oo: xx ox xo
both: xo ox
already we can see there's a strong limitation on rules based on available assets and space. filling the void with space will allow for much more rules and possibilities but the repetition of assets will be vomit inducing apparent:
4 spaces, 2 characters,
possibilities: xxxx xxxo xxox xxoo xoxx xoxo xoox xooo oxxx oxxo oxox oxoo ooxx ooxo ooox oooo
rules:
no xx: xoxo oxox ooox ooxo oxoo xooo
no oo: xoxo oxox xxxo xxox xoxx oxxx
both: xoxo oxox
no xxx: xxox xxoo xoxx xoxo xoox xooo oxxx oxxo oxox oxoo ooxx ooxo ooox oooo
no ooo: xxxx xxxo xxox xxoo xoxx xoxo xoox oxxx oxxo oxox oxoo ooxx ooxo
both: xxox xxoo xoxx xoxo xoox oxxx oxxo oxox oxoo ooxx ooxo
and these were just simple filter rules. applying pattern filter rules with only 2 assets/characters would not even be distinguishable from simple filter rules. so let's make things interesting:
5 spaces, 5 characters: x,o,f,r,s
possibilities 5x5x5x5x5=3125
rules:
after every f comes an s if the f will not be followed by an o:
xofsx xsfsx xsffo ... there's a lot more remaining possible combinations even after applying a pattern filter rule.
the complexity of the pattern filter rule itself is more interesting than most combinations. with 5 characters and 5 spaces, i think it's pretty obvious for everyone that stuff like xxxxx or ooooo or even fofof or xofsr is pretty boring.
so what we want is as many and as complex as possible pattern rules and as many characters/assets as possible with just enough space that they don't become too repetitive. only being able to walk two spaces to the left or the right is not satisfactory as a dungeon level. neither is a 50000x50000 spaces square filled with only rats, cats, dogs, spiders and bees.
so what is procedural generation good for if you still have to create just as many assets as you would without it and then even keep track of complex pattern rules to sieve the chaos of randomness without the rules conflicting with each other?
well, there's a reason it's mostly used for roguelikes.
a game that kills you before you get bored will have a huge replayability value if and only if it doesn't play the same way when you start over. so random procedural generation is not a cool feature, it's a core element without which perma death would also kill the game for the player. it's a necessity, not a "smart way" to make games.
why perma death? for maximum excitement. the possibility of losing all progress, all gear, all the time you have invested makes for the most exciting games you can ever hope to make. but without randomization of almost every element, perma death devolves into the frustration of having to play the same thing all over again the same way.
this is core ideology of roguelikes. it's also the reason why roguelites fail to be as replayable. they both add a feeling of grinding in the case of saved power progression as well as a much weaker excitement factor due to deaths that don't wipe all progress.
how does this wall of text apply to no man's sky?
simple answer: it's not a good roguelike.
what's random in the game? planets, animals, that's pretty much all i remember from trying it
crafting and power progression based on crafting makes sure that you will always start the game with no hope of finding a random significant power boost. no randomly significant power boost means less randomization. yes, even character progression must be random for excitement and replayability. while perma death makes you fear losing, random character progression makes it exciting to start over. dcss will sometimes let you find a +12 accuracy / +12 damage ring on the first level making you play a warrior for a few levels even if you started as a mage.
the game features several difficulties that don't have perma death. cutting away the only other core element of a roguelike and excitement.
now, without it being a roguelike, what is the random generation supposed to add to the experience? it seems to me it's all about empty exploration. just site seeing of mixed assets. and few assets at that, some say.
thank you, no.
Last edited:
why the fuck would any game have to populate any fucking thing in fucking real time, you fucking imbecile?
and who the fuck looks at game trees to check if they are really as different from each other as mother nature would approve of?
adrix89
Cipher
No Man's Sky could have worked. It just should have been User Generated instead of Procedurally Generated.
If it started with Minecraft like building and Sandbox Engineering like From The Depths with a faction and economy like X3 Terran/Albion then you would have a good framework in generating a lot of user generated content.
Next is upload the player's bases and shit on the cloud to a shared universe. Now you have a universe to explore that you can dynamically download and populate shit with.
You can have enemy player empires and ancient player ruins with unique technology which might be outside of your current randomized technology tree(like in 4X games).
This is very cheap for the developers. Like in sandbox engineering games developers can focus on creating new parts and technology with new mechanics and it will slowly trickle down into the user design.
Of course you have to do the initial hard work of sandbox engineering and X3 factions and economy.
But there are already many sandbox engineering games(From the Depths, Factorio, Terratech, Space Engineers, Emyrion) that you can analyses the mechanics and see what works and what doesn't. Design wise they already solved even if they are hard on code and technical details. There are code mankeys that can achieve it.
X3 faction and economy also has other examples(Starsector, SPAZ 2, The Guild 2, Patrician 3) and there is a whole 4X and Grand Strategy genre you can borrow elements from.
For creatures you would need something like Spore creature editor and players actually have to play it in a kind of roguelike where they need the ability to script their species behavior. There are games where you scrip AI bots and it would be similar, you then upload the scripts and the shared world populate it on planets with some requirements of resources and parts for the creatures.
If it started with Minecraft like building and Sandbox Engineering like From The Depths with a faction and economy like X3 Terran/Albion then you would have a good framework in generating a lot of user generated content.
Next is upload the player's bases and shit on the cloud to a shared universe. Now you have a universe to explore that you can dynamically download and populate shit with.
You can have enemy player empires and ancient player ruins with unique technology which might be outside of your current randomized technology tree(like in 4X games).
This is very cheap for the developers. Like in sandbox engineering games developers can focus on creating new parts and technology with new mechanics and it will slowly trickle down into the user design.
Of course you have to do the initial hard work of sandbox engineering and X3 factions and economy.
But there are already many sandbox engineering games(From the Depths, Factorio, Terratech, Space Engineers, Emyrion) that you can analyses the mechanics and see what works and what doesn't. Design wise they already solved even if they are hard on code and technical details. There are code mankeys that can achieve it.
X3 faction and economy also has other examples(Starsector, SPAZ 2, The Guild 2, Patrician 3) and there is a whole 4X and Grand Strategy genre you can borrow elements from.
For creatures you would need something like Spore creature editor and players actually have to play it in a kind of roguelike where they need the ability to script their species behavior. There are games where you scrip AI bots and it would be similar, you then upload the scripts and the shared world populate it on planets with some requirements of resources and parts for the creatures.
Last edited:
adrix89
Cipher
Are they adding in the game yet?
circ
Arcane
#remind me when it isn't shit
Morkar Left
Guest
Oh, Scientology is behind NMS. Finally, a lot of it makes sense.
why do they even bother updating it? no one gives a shit now
Siveon
Bot
- Joined
- Jul 13, 2013
- Messages
- 4,509
Good on this guy for finally cutting ties from this game. Clearly, nothing they're gonna do is gonna make this better as a game. Although ARG's are always fun to read about, it doesn't actually affect the game itself.
Makabb
Arcane
- Joined
- Sep 19, 2014
- Messages
- 11,753
Probably Sony pressured them to release it early
Perkel
Arcane
- Joined
- Mar 28, 2014
- Messages
- 15,858
Imo they earn a lot of goodwill.
They made a game and they banked like 1mln+ coppies they could just walk away and do nothing really.
Their progression system got reworked a lot but still lacks proper goal.
Worst part is that for a space game it is pretty shit space game. There are no gas giants, non atmosphere moons, ocean planets and so on and so forth where NMS could really shine considering they don't want to go strictly toward science.
They made a game and they banked like 1mln+ coppies they could just walk away and do nothing really.
Their progression system got reworked a lot but still lacks proper goal.
Worst part is that for a space game it is pretty shit space game. There are no gas giants, non atmosphere moons, ocean planets and so on and so forth where NMS could really shine considering they don't want to go strictly toward science.
Last edited:
MadMaxHellfire
Arcane
but there are sand planets with huge worms inhabiting them! there are! i swear! it's just that you haven't found them yet.
fuck off, no amount of forced-by-contract work can redeem them.
fuck off, no amount of forced-by-contract work can redeem them.
The worse thing about it is that if it somehow turned into a pretty good game by intense patching, no one would know.
If a patch makes a game good, but no one plays it, then has a patch really made the game good ?
If a patch makes a game good, but no one plays it, then has a patch really made the game good ?
As an Amazon Associate, rpgcodex.net earns from qualifying purchases.