/tg/Station 13

Preface: Summary:
When is the game hard enough? SS13 has been constantly evolving over time to incorporate many new features, but the addition or revamp of features constantly causes the complexity of the game to constantly increase. I conclude that there is a theoretical "Optimal" complexity in the game, but the definition of that is very subjective, so it is meaningful to discuss where this point may be and how it may be achieved.

Thought Process:

A: That there is an optimal complexity for the game
1. There exists a state of the game in which the game is too difficult/complex.

Spoiler:

Proof: One case of the game being too complicated is when the game requires the real-life knowledge of systems to work each job (i.e. chemists need to know real life chemistry)

2. There exists a state of the game in which the game is too simple.

Spoiler:

Proof: See Proof 1

3. The game cannot be both too simple and too difficult.

Spoiler:

Proof: WLOG, If we assign optimal complexity as a number, say 5, and <5 being too simple with >5 being too difficult, there exists no real number x such that x<5 and x>5.

4. Given 1,2, and 3, there exists a point of complexity that is neither too simple nor too difficult.

Spoiler:

Proof: If there exists ways the game can be too difficult or too simple, there must exist a point that these converge at, that is neither too difficult nor too simple.

B: That the complexity of the game is constantly increasing
1. The number of merges with the code increases with time

Spoiler:

Proof: The number of merges is strictly increasing along time, so as long as we have a codebase the game has constant code changes.
Point A: If we count reverts as -1 merges, the # of unreverted merges >> # of reverted merges.

2. Merges, on average, increase complexity of the game.

Spoiler:

Proof: This is the weakest point of the argument. It seems like merges have been more focused on increasing the number of available features to the playerbase rather than simplifying things, from what I have seen.

3. Given 1, 2, it is reasonable to conclude the game is getting more complex.

Spoiler:

Proof: Given that the number of merges increases with time, and that time is strictly increasing, and that merges on average increase complexity of the game, then on average the game is getting more complex over time.

Personal Opinion: Given what I have pointed out, what do you think is the optimal complexity of the game? Are there any features that are too simple or too complex right now? Why?

PS: I wish I could organize this better, but it doesn't seem like nested spoilers work. Anyone know a way?

Too much bullshit: Telescience

Not enough bullshit: Medical

Incredible amounts of bullshit if you look at it closely enough, but not that big of an issue: Atmos

One of these things is more fun than the others, one of these things just doesn't belong~

     .

4 does not follow from 3. If X and Y are opposites to eachother, we cannot assume that a value for X or NotX is somehow Neither "too X" or "too NotX." These are undefinable in a concrete manner due to opinions of what qualifies "too much" or "too little" X or NotX. Your logic is overbroad.

That is a problem of usability not complexity.
You want something easy to get into and hard to master.

Besides most systems are time-wasters anyway.

cedarbridge wrote:4 does not follow from 3. If X and Y are opposites to eachother, we cannot assume that a value for X or NotX is somehow Neither "too X" or "too NotX." These are undefinable in a concrete manner due to opinions of what qualifies "too much" or "too little" X or NotX. Your logic is overbroad.

Fuck I knew there was a problem with it. I'll think about it and correct it later

In a sense the game is pretty much a huge amount of mini games within one game of which some do interact partially with each other. Some dependencies are created to be a fact and apart from that its pretty much all.

We need to kinda go away from minigame type stuff to create larger things that everyone can work with together. But that is difficult and near impossible and most people dont want to do that at the moment we got a lot of issues with stuff as it is.

In time things will change and the question of hard enough will be different. Some things need to be easy others need to be difficult but what should be difficult? is the question

I think you need to define complexity. The one I use is "number of viable options the player must choose between at any given time". Thus false choices don't increase complexity and should be removed. This includes both no-brainer obviously good decisions (as an engineer, you always want to be wearing yellow gloves) and stupid choices you would never want to do (eating those jellies that have acid inside them). Having more complexity is good, but it also can't be increased very much for any given decision. If you were to keep adding more to one, you'll wind up with an optimal one, which means you actually reduced complexity. It's important that every choice is viable at least some of the time. That doesn't mean any given player should choose any option given some specific set of circumstances but instead that some player within the population should.

Contrary to what you might think, high complexity doesn't necessarily lead to player confusion or low usability or any of the other issues you commonly associate with complex games like DF or Aurora or SS13. Instead the real problems with a complex system is when the player makes a decision without knowing all their options or all the consequences of their actions. The general way players get confused is when they don't understand why something in the game happened the way it did, especially when it's something bad that happened to them.

Not knowing all your available options is always at least partially a failure of the game's UI. All options should be laid out clearly before the player at some point during the game. That doesn't mean the game should "play itself" or that new players should be flooded with unnecessary information, but it does mean the player should have all the information they need to solve any given problem. If the game literally can't lay that information out clearly, the system is too complex. A very common example with SS13 is pretty much anything in the "little things you learned that changed the game" thread. This kind of shit is very frustrating to players who lose and have no idea why or how they could have even won, but it can also simply lead to players missing out on fun they could have experienced if they only knew it was available.

Not knowing the consequences for a choice can be a failure of the game's UI, but it can also be a sign that the system is too confusing, interwoven with another system, or chaotic to get repeatable outcomes out of. An example would be if a closed-source game added an item with an unstated drawback. The player can't plan around it and may forget the item even has a drawback, which causes confusion when they notice they're experiencing it and can't figure out why. Another possibility would be if we made telescience require knowing how to do calculus. The inputs and results seem random to the player, making the whole system a waste.

EDIT: It also occurs to me a player could simply not care about the outcome of a decision they make, which basically has the same effect as not knowing those outcomes in the first place.

What person and Adrix said is completely correct. In game theory you eliminate non-optimal choices immediately and are generally considering less choices than are actually presented to you.

In design this generally results in the designers evaluating non-choices and trying to either change them, remove them, or in cases where they are rendered irrelevant due to another choice being too strong nerfing other items. For example unless I am missing something people really should spawn with their IDs in their PDAs at this point as there isn't really a reason not to store them there.

A good example of when this happened in SS13 was with stungloves. There used to be no reason not to make them, as they were essentially weaponized insulated gloves. Now they are a trade off, you get the option of a surprise panic stun that is hard to disarm vs insulation. In daily use you would probably want rubber gloves, so it seems like they are a first order optimal solution.

However it is acceptable for items to be segmented to highly specific uses or specific players, as well as by limiting access. Some players may have no reason to ever take an option where as others would see it as desirable. Traitors end up saving quite a bit on telecrystals just by making a set of stungloves. The other gloves still need a ton of love, but it is acceptable in design to deny high powered options to the player.

There are parts of the game that are problematic in complexity. Virology comes to mind, it is a bit hard to grasp for a new player for too little benefit, and genetics pretty much requires a guide to play as trial and error is not at all viable. Compare that to chemistry, which while powerful and having a ton of depth is simple enough a concept that a player can wrap their head around it. "Pour this into that and you get a new thing." R&D falls somewhere in between. The idea of taking stuff to a machine to scan them to get science is easy enough, but what you get from what you scan isn't generally intuitive.

However complexity in SS13 doesn't really matter in terms of job's system, as long as the job has simple tools as well as complex ones. What matters in SS13 is more about reacting to others. SS13 is too complex once it becomes impossible to understand what is happening by viewing the results. There needs to be clarity there in most cases, both to allow players to recognize threats in order to react to them and present a choice to the threatener, and so that people can fake threats.

As for when things are "hard enough?" Good hard is things like new-mining. You have a risk vs reward mechanic, and the job is difficult but not same-y. Bad hard is genetics and virology. Did RNG give you your buff? No? Roll again until success. Toxins is another good example of good hard. It is difficult to do right and dangerous as hell, but supremely rewarding when applied.

I'm kinda sad that I didn't learn the game by playing it, but by watching 30+ episodes of Plumphelmetpunk on youtube... so I never really got the full feeling for how hard this game might be to learn by playing it. And when I did start playing myself, I immediately went for the wiki for everything I wanted to do..

To answer the question "When is the game hard enough", I'd say it's hard enough already. I wouldn't want to see deconstruction or hacking be any more difficult than they are now, for example. Most things are complex enough. Chemistry, or medbay in general, could have more mechanics. But it shouldn't be just to make things harder, but to add layers of complexity, and viable options to explore. Defibrillators are a great addition in this sense.

Apart from putting the relevant guides to be available at the library, I don't know how most of the game's mechanics could be explained to new players in-game. I think most people expect games to explain themselves, but we don't even have a tutorial, or a controls page or anything. Just a link to a wiki. Then again it seems to be a trend in these types of games, I'm looking at Minecraft, Binding of Isaac, most of the modern fps zombie survival games, a bunch of roguelikes/-lights. "Part of the fun comes from things happening you didn't even know could happen" is an easy way to argue yourself out from making the game more approachable and more playable.

In SS13, part of the difficulty level comes from the infamous metagame, eg. what is being thrown at you by antagonists, how helpful the AI happens to be, how likely the captain is to be on top of their game, etc. For example I'm seeing nuke-ops rounds becoming stale when the ops are simply too robust in comparison to the crew. The game has become "too hard" for the crew. But then it's up to the players to shift the metagame, maybe next time the nuke ops won't go with the strategy they know will work but will try something more fun, or maybe a more robust captain will step up to defend the disk. In this sense the difficulty level of the game is in the players' hands.

Maybe the best way to make the game easier to learn would be to encourage new players to ask for help, preferably in character. There is bound to be someone who can help. Having to have to read the wiki while you play is the worst thing.

iyaerP wrote:As for when things are "hard enough?" Good hard is things like new-mining. You have a risk vs reward mechanic, and the job is difficult but not same-y.

I find it funny you should mention this, as the person who changed it, I find the current state of mining to be really, really bad. It's incredibly easy to circumvent the risk/reward scenario due to how prevalent ores are. There's no reason to ever step foot in a tunnel, because you can find all the ore you'll ever need outside of them. You do have a slight advantage in finding diamonds while taking the tunnels as it is the only rare ore left, plus you can run into basilisks and goldgrubs for more diamonds, but largely, the risk/reward factor was done away with once the ore rates were increased.

As it is right now, it's not challenging enough for my tastes, and still too challenging for the "average" spessman... It's in a very weird place. The design philosophy of risk/reward was largely thrown out the window when it was deemed too hard.

As for "When is the game hard enough?", in my opinion, when I feel challenged. That's all there is to it. If I'm not struggling somehow, what is the point?

I assumed the OP was talking about complexity, which is why I asked for clarification. Difficulty is an entirely different concept.

It's hard to quantify difficulty in this game. Obviously stealing the AI and escaping alone is harder than stealing a jetpack, but it's hard to say if chemistry is tougher than genetics for example.

If we're talking about complexity, most of the station fails fairly hard (honk) in that regard.

Jobs are generally based on a combination of two things:
1. Rote memorization of information
2. RNG

Part of this problem is because introducing UI to support stranger mechanics is difficult, but a lot of it is just lazy.

Botany, Virology, Xenobiology, and Genetics? they're all just slot machines. Player input on some of them matters a tiny bit, but in the end its up to the hand of god to decide your fate, and as a result we have very shallow, easy mechanics. There's no real barrier to entry besides reading a tiny bit on the wiki.

The entire medical system, Engineering, Cooking, Chemistry, Robotics, R&D etc? It's all just memorization, and how fast you can click buttons on the UI. None of these are deep or hard, and barely count as game mechanics at all. The only barrier to entry is memorizing a wiki page. Or just reading it while you do this activity.

And then we have Telescience. A mechanic that uses math (read: Outside assistance with a calculator) as a barrier for entry. Awful, and largely unused except by people who know how to abuse the system. At this point, "Complexity" has taken a backseat to being needless obfuscation.

Honestly, I don't think we really have many parts of the game that have the proper amounts of complexity that allows entry to new players, but has deep enough mechanics to make something round to round be novel on its own, except the gestalt of SS13 itself... but a newbie is still flooded with information and no way to process it due to almost nothing in game teaching the player how to do things. We have a few manuals here and there, and they can be useful, but it's mostly just wiki wiki wiki.

The game is "hard" (complex) enough to learn when the obfuscation of mechanics doesn't render people uninterested in playing it, but is still complex enough that the player improves at it over time, learning new ways to handle it.

I think that each job needs to be fairly easy to do the bear minimum to keep the station running with a layer of complexity that provides benefits to either the station as a whole or the individual on top.

Though telescience is some real bullshot, it still has a pretty good skill ceiling compared to other jobs. Speed makes the difference between teleporting the Captain out of a scrape before he dies and grabbing empty air. Sure it's muscle memory, but it's still something that makes me feel warm inside knowing not everyone is capable of it.

Atmos is in a similar place with optimal configs and bombs on top of less complex set ups.

This is a legit issue that we were already looking at it, the plan is, in simple words, making a better learning curve for the players.

MisterPerson wrote:I think you need to define complexity. The one I use is "number of viable options the player must choose between at any given time". Thus false choices don't increase complexity and should be removed. This includes both no-brainer obviously good decisions (as an engineer, you always want to be wearing yellow gloves) and stupid choices you would never want to do (eating those jellies that have acid inside them). Having more complexity is good, but it also can't be increased very much for any given decision. If you were to keep adding more to one, you'll wind up with an optimal one, which means you actually reduced complexity. It's important that every choice is viable at least some of the time. That doesn't mean any given player should choose any option given some specific set of circumstances but instead that some player within the population should.

Contrary to what you might think, high complexity doesn't necessarily lead to player confusion or low usability or any of the other issues you commonly associate with complex games like DF or Aurora or SS13. Instead the real problems with a complex system is when the player makes a decision without knowing all their options or all the consequences of their actions. The general way players get confused is when they don't understand why something in the game happened the way it did, especially when it's something bad that happened to them.

Not knowing all your available options is always at least partially a failure of the game's UI. All options should be laid out clearly before the player at some point during the game. That doesn't mean the game should "play itself" or that new players should be flooded with unnecessary information, but it does mean the player should have all the information they need to solve any given problem. If the game literally can't lay that information out clearly, the system is too complex. A very common example with SS13 is pretty much anything in the "little things you learned that changed the game" thread. This kind of shit is very frustrating to players who lose and have no idea why or how they could have even won, but it can also simply lead to players missing out on fun they could have experienced if they only knew it was available.

Not knowing the consequences for a choice can be a failure of the game's UI, but it can also be a sign that the system is too confusing, interwoven with another system, or chaotic to get repeatable outcomes out of. An example would be if a closed-source game added an item with an unstated drawback. The player can't plan around it and may forget the item even has a drawback, which causes confusion when they notice they're experiencing it and can't figure out why. Another possibility would be if we made telescience require knowing how to do calculus. The inputs and results seem random to the player, making the whole system a waste.

EDIT: It also occurs to me a player could simply not care about the outcome of a decision they make, which basically has the same effect as not knowing those outcomes in the first place.

My definition of complexity would be the amount of pre-existing knowledge required to perform a job up to task. This is still extremely broad, of course, but at least from this I can form some bases for comparison.
For one, a job with RNG would probably require much more pre-existing knowledge than one that focuses on memorization. As you mention as well, a job with more viable options at a time requires the player to understand the results of each action.

Difficulty is probably an entirely different subject; Something can be hard without being complex
Something can be complex without being hard; As long as you have full knowledge of the processes you should be fine (i.e. genetics)
Keep in mind that I am specifically talking about the mundane station-based jobs we have, and SPECIFICALLY NOT DIFFICULTY OF ANTAGS (Which I believe is different)

A good question that summarizes what I think complexity is:
How much do I have to learn/know to understand this?

This is probably also entirely unable to be quantified, to my dismay.

I just cannot comprehend atmos. I do telescience just fine, I think thats easy. Atmos is the only thing on the entire station I cannot figure out how to do. That and wire the singulo to the power grid directly.

deathhoof wrote:I just cannot comprehend atmos. I do telescience just fine, I think thats easy. Atmos is the only thing on the entire station I cannot figure out how to do. That and wire the singulo to the power grid directly.

Its literally just "trace the pipes", set volume pumps in high-flow areas like waste, set filters to 4500, and set pressure pumps in airflow to whatever you feel like. I tend to go for 300kpa because that keeps it just under enough pressure for me to wrench up things if I feel like making changes.

If you're talking about bomb mixes though, then we get into the whole meta of moles vs temperature efficiency and all that jazz.

Saegrimr wrote:

deathhoof wrote:I just cannot comprehend atmos. I do telescience just fine, I think thats easy. Atmos is the only thing on the entire station I cannot figure out how to do. That and wire the singulo to the power grid directly.

Its literally just "trace the pipes", set volume pumps in high-flow areas like waste, set filters to 4500, and set pressure pumps in airflow to whatever you feel like. I tend to go for 300kpa because that keeps it just under enough pressure for me to wrench up things if I feel like making changes.

If you're talking about bomb mixes though, then we get into the whole meta of moles vs temperature efficiency and all that jazz.

This is a very important distinction. Understanding the quirks and characteristics of the Atmos code, especially where it regards toxins and bombs, is like its own special challenge. Just doing the basic job of atmos is hilariously easy. Set up the pipes as per the wiki, making changes where you see fit, and once hull breaches are sealed, set the air pumps in that area to max until the kPa is back in spec.

iyaerP wrote:This is a very important distinction. Understanding the quirks and characteristics of the Atmos code, especially where it regards toxins and bombs, is like its own special challenge. Just doing the basic job of atmos is hilariously easy. Set up the pipes as per the wiki, making changes where you see fit, and once hull breaches are sealed, set the air pumps in that area to max until the kPa is back in spec.

That's where my initial quote of "Incredible amounts of bullshit if you look at it closely enough, but not that big of an issue: Atmos"

Its got a huge level of complexity if you REALLY want to get down to it, but otherwise its not that hard to get into initially. In my opinion, that's what most jobs should try to achieve.

Saegrimr wrote:

iyaerP wrote:This is a very important distinction. Understanding the quirks and characteristics of the Atmos code, especially where it regards toxins and bombs, is like its own special challenge. Just doing the basic job of atmos is hilariously easy. Set up the pipes as per the wiki, making changes where you see fit, and once hull breaches are sealed, set the air pumps in that area to max until the kPa is back in spec.

That's where my initial quote of "Incredible amounts of bullshit if you look at it closely enough, but not that big of an issue: Atmos"

Its got a huge level of complexity if you REALLY want to get down to it, but otherwise its not that hard to get into initially. In my opinion, that's what most jobs should try to achieve.

Atmos has gas reactions and stuff apparently.

Ergovisavi wrote:Botany[/spoiler]

Git gud, you don't even kill everyone with spicy hot chill.