Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 2 [3] 4 5 6

Author Topic: Dwarf fortress: Margionally complex, but mostly just detailed  (Read 14096 times)

Hammurabi

  • Bay Watcher
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #30 on: August 07, 2010, 11:30:59 pm »

...clip...
Is any of this an "Interesting Decision"?  Well, I can't speak for anyone but myself, but I certainly am fascinated by it.  But "Interesting" is ultimately a subjective term.  Personally, I'm not interested in seiges.  I'd much rather DF be focused upon the societal modeling aspect of the game, as well as its engineering aspects.  But it doesn't hurt me much to let the mechanics reset the cage traps and unjam the weapon traps every few months if that means other people enjoy the game more because of it.

I'm not seeing any "Decisions" in your ideas.  Each plant type will have a preferred pH level.  Players will look up this value in the Wiki and apply the appropriate amount of fertilizer.    Different crops will need different amount of water.  Players will water them according to the Wiki values.  Crops will need to be rotated.  Players will rotate them according to the Wiki.  Maybe this level of detail will be interesting to some people.  But I don't see any real decisions here, other than doing it the optimal way or purposely doing it the suboptimal way.  Don't add complexity if it doesn't add choices for the player to make.

Here's a quick idea of Interesting Decisions.  Have the different types of plants/food have a real effect on the dwarfs. 
Plump helmet - baseline food, no modifiers
Pig tail - Makes the dwarfs work harder, more productive, but creates some unhappy thoughts
Sweet pod - Makes the dwarfs sleep more (lazy), and more contented (happiness goes up)
Quarry bush - Adds small boost to combat skills, with corresponding loss to non-combat skills
Dimple cup - Adds small boost to non-combat skills, with corresponding loss to combat skills
Muck Root - Not a Dwarf favorite, but the Humans value it highly in trades

Now there are interesting decisions to make.   The player will have to plan ahead, deciding what to plant now base on the type of food he wants 6 months from now.  Give the player several plausible choices, with the choice possibly having a real effect on the game. 
Logged
Back in 1971, Nolan Bushnell of Atari said, "All the best games are easy to learn, and difficult to master," a design philosophy now treated as instinctual by nearly every designer in the industry.

Normandy

  • Bay Watcher
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #31 on: August 07, 2010, 11:52:17 pm »

I think the SimCity analogy is actually a very good example of how Dwarf Fortress could progress forward.

The SimCity 4 interface is surprisingly like DF's interface. It feels like it was designed in 1994 and has a way of transmitting absolutely no information whatsoever about what you're placing. But in a way, it's very intuitive - I did not know for quite some time that school radius was more of a rough guide than an absolute circle, and I think I was better off for not knowing, because it vastly simplified the placement of schools for me, at least until I understood enough about the mechanics of the game to know better. By the time I got past this obtuse interface for school coverage, I could appreciate the complexity behind it. It was subtle enough for me to not notice it before, but pronounced enough that I changed my school placement strategy.

The system of unlocks in SimCity 4 is brilliant. Though some may criticize it as a feature that does not fit in a sandbox type game, I would propose that those people haven't actually played that much SimCity 4. At what population should I place a convention center? What about a disease control center, is that necessary for a city my size? The unlocks felt like goals, but really they were guidelines. They are ridiculously easy to get (for the most part) for the experienced player and were not a hindrance to city-building. A modestly skilled player would see the unlocks and say, "hey, these are goals that I can achieve" and provides structure and framework, without hiding anything from the player. A completely unskilled player would understand that these unlocks were for more developed cities. You can spot an inexperienced player by how they place certain optional features, such as schools, health centers, and public transport, before their cities are ready to profitably accommodate for them.

I think that Dwarf Fortress would benefit from "gamey" features. Things like workshops or reactions requiring certain nobles or conditions to be met - these may seem contrived at first, but really do add a lot of structure to the game. Adding easily unlocked rewards is just a way of helping players navigate complexity, without removing any of it; thus, you can add so much more to the game.

How does this tie into the discussion at hand? I think the complexity of DF can be as high as it wants to be, as long as there are some barriers to discovering this complexity. The cavern system is absolutely ingenious for introducing bigger and badder beasties gradually, same goes for megabeasts and the thief-ambush-siege kind of growth. If you had plants that required a certain fertility of soil in order to be grown, then you're ensuring that players with the experience to grow these plants would be able to handle the added complexity of them. No scripted tutorials required. Or if building a jeweler's workshop requires a certain jeweler noble that arrives only after a certain amount of gems have been discovered - you're teaching new players that the jeweler's workshop is not something to build at first, at least until they've got their mining operations up.

Also, 'interesting decisions' varies from player to player. I absolutely love small detail that seems trivial and almost hindering to other players. It's really fun to experiment with different combinations of crops and rotations - after all, people are still coming up with room designs, aren't they?
Logged

NW_Kohaku

  • Bay Watcher
  • [ETHIC:SCIENCE_FOR_FUN: REQUIRED]
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #32 on: August 08, 2010, 09:28:26 am »

I'm not seeing any "Decisions" in your ideas.  Each plant type will have a preferred pH level.  Players will look up this value in the Wiki and apply the appropriate amount of fertilizer.    Different crops will need different amount of water.  Players will water them according to the Wiki values.  Crops will need to be rotated.  Players will rotate them according to the Wiki.  Maybe this level of detail will be interesting to some people.  But I don't see any real decisions here, other than doing it the optimal way or purposely doing it the suboptimal way.  Don't add complexity if it doesn't add choices for the player to make.

Here's a quick idea of Interesting Decisions.  Have the different types of plants/food have a real effect on the dwarfs. 
Plump helmet - baseline food, no modifiers
Pig tail - Makes the dwarfs work harder, more productive, but creates some unhappy thoughts
Sweet pod - Makes the dwarfs sleep more (lazy), and more contented (happiness goes up)
Quarry bush - Adds small boost to combat skills, with corresponding loss to non-combat skills
Dimple cup - Adds small boost to non-combat skills, with corresponding loss to combat skills
Muck Root - Not a Dwarf favorite, but the Humans value it highly in trades

Now there are interesting decisions to make.   The player will have to plan ahead, deciding what to plant now base on the type of food he wants 6 months from now.  Give the player several plausible choices, with the choice possibly having a real effect on the game.

So it's only interesting when crops do different things, but not when the means to get them are different?  Doesn't how you get them, things are only interesting for what they do?  Seems like a silly differentiation.  Whatever, having crops that do different things is, of course, also part of my attempt to create diversity.

Of course, much of what YOU are asking for is something that won't be available without the medicine/drugs corrolary with syndromes, so it's not possible until Toady completes that arc, as well.

Still, one of the "crops" I have for underground is the "Sapphiric Rock Mold", a mold that leeches the aluminum out of highly alkaline soils (and only grows in extremely alkaline soils) that, if left to grow for years, has a chance of producing sapphires as well as some more common quartz or crystal.  The mold, however, acidifies the soil powerfully, so you can't simply keep producing these rock molds, you'd be best off looking for new places to plant them unless you work on making the soil far more basic with other crops or actions.
« Last Edit: August 08, 2010, 09:40:50 am by NW_Kohaku »
Logged
Personally, I like [DF] because after climbing the damned learning cliff, I'm too elitist to consider not liking it.
"And no Frankenstein-esque body part stitching?"
"Not yet"

Improved Farming
Class Warfare

Grimlocke

  • Bay Watcher
  • *kobold noises*
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #33 on: August 08, 2010, 12:01:16 pm »

I would actualy like to see the current farming system replaced with something more complicated. Right now it just comes down to slapping a patch of farm somewhere and setting it to churn out plants, while there is a lot more interesting things you could do with it. Basic crops should only require so much, but more specialised crops could require specific humidities, acidities, etc. There could even be plants that only grow in 1/7 to 3/7 water, giving us an actual use for all those fancy pumping mechanisms.

Of course I would still build rediculous magma pump stacks that blast out fire and death on my enemies even though a simple corridor of traps would suffice, I just think it would be neat if we could make these extensively complicated contraptions and gain something we couldnt get by just randomly slapping down a row of workshops and setting them to repeat orders, only occasionaly having to repeat the orders because Urist McPotashmaker went bonkers.

Im not realy a fan of the whole unlock system though. Even the 40d dungeon master had me rather irritated whenever I ran a fort I wanted to keep below the 50 dwarves. Each time I had to let a whole bunch of dwarves migrate, wait for the dungeon master to show, and then kill off the lot of them. Often this went wrong and the fortress became larger then I realy wanted it to.
A 'line of gameplay' can be nice, but it becomes restrictive very quickly and its hard to do when the game doesnt have any set goals or gameplay styles. Its not quite the same thing as, for example, adding a more complicated farming system. A hermit dwarf can set up a drainage system for some watery plant, but if steel making only becomes possible after Urist McMastersmithy migrates to your fortress then the hermit just cant make steel.

tl;dr version!: Its nice to have complex game mechanics to work on and have an actual reason to work on them. A reward you couldnt gain in some much easier way, which is something the game lacks right now.
Logged
I make Grimlocke's History & Realism Mods. Its got poleaxes, sturdy joints and bloomeries. Now compatible with DF Revised!

Normandy

  • Bay Watcher
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #34 on: August 08, 2010, 12:48:24 pm »

I understand your worries, Grimlocke, but a simple init switch that could turn on/off rewards might solve all of our problems.
Logged

zwei

  • Bay Watcher
  • [ECHO][MENDING]
    • View Profile
    • Fate of Heroes
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #35 on: August 08, 2010, 01:42:04 pm »

OP:

Simplified game would actually help with my pet issue: most selection menus are huge (hundred or more items) and finding blocks of certain rock amongts all those woods, metals, bars and blocks is pretty iritating (sometimes you miss it and have to try again ...)

So mod with only one type of wood, one type of metal, etc... would be interesting.

---

As far as tutorial goes, it is also a bit about fact that your embark should be more connected to world. Ruler of your civilization should have good idea what he wants of you: Are you supposed to create military presence in area? To be hub for local villages and export processed food and other similar products to mountainhome? Or are you there to exploit mineral resource and provide armors and weapons for kings army?

From this, "tutorial goals" could be very well defined: build housing for ten soldiers that will come in next migrant wave, export 50 food items, etc ...

NW_Kohaku

  • Bay Watcher
  • [ETHIC:SCIENCE_FOR_FUN: REQUIRED]
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #36 on: August 08, 2010, 10:58:21 pm »

Actually, a couple things I should have put in my last post:

First, because crops are going to be so much more differentiated that we get many, many more of them, it should hopefully be the case that you can't ever rely on getting ALL of them in one fort, even through trading.  Simply copying setups from the wiki aren't going to help if you wind up with different sets of crops to work with every time. 

However, someone had mentioned before having "Procedural Crops", where different cultures could domesticate crops over time, producing very different plants, especially if all crops (or at least all but a couple ol' standbys, like Plump Helmets) entirely were procedural, and worked off of a formula that balanced soil needs with crop yield and value.  You would have to learn how best to work crop cycles out for every fortress.

Second, keep in mind that many of the things Hammurabi suggested are just as, if not more easily gamed than what I have been setting up.  Quarry Bushes boost military effectiveness, but down everything else?  Use burrows to ensure that only military dwarves eat them!  No choices necessary! 

Also, I hope you know that you don't eat Dimple Cups at all...

I really don't think the way to give players "Interesting Choices" is through "Curtain A or Curtain B" decisions - those can generally be very easily broken down into finding a single best answer, or gaming the effects of both.  Rather, I think that the farming we have now is actually a decent example of what happens when you are given an open-ended choice:  To farm, you must flood a field.  Probably in your first season or two, unless you are abusing the current lack of livestock food requirements, and just butchering animals.  That means that you have the problem of muddying a field, but it is entirely up to you how water gets from point A to point B, and it perfectly highlights the abilities of the game engine to make you find a way to move water to wherever you want to farm in a controlled fashion.

This is what the game needs more of - problems where the solution is a complex, context-sensitive solution that changes with every fortress you play, rather than simply allowing every fortress to be cookie-cuttered out with the same layouts and same decisions on how to train your military every time, and all the potential complexities of the game simply receding into "boring, unimportant detail".

Logged
Personally, I like [DF] because after climbing the damned learning cliff, I'm too elitist to consider not liking it.
"And no Frankenstein-esque body part stitching?"
"Not yet"

Improved Farming
Class Warfare

Hammurabi

  • Bay Watcher
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #37 on: August 09, 2010, 10:25:48 am »

I really don't think the way to give players "Interesting Choices" is through "Curtain A or Curtain B" decisions - those can generally be very easily broken down into finding a single best answer, or gaming the effects of both.

Maybe I didn't explain it well.  But your explanation of "Interesting Choices" is not at all what I had in mind, not even close.

Sid Meier said that gameplay is a "series of interesting choices."  His game Civilization exemplifies this perfectly.  Right at the start, you choose between The Wheel and Agriculture, both which can have big influences on how the game plays out.  Later you discover other civs.  Do you fight them or friend them?  There's no "single best answer" nor can you "game the effects of both", as you called it.  My trivial example of food effects was just a quick example of how food could have more "Interesting Choices".  I didn't suggest it to be implemented as is.

DF has many interesting choices.  The player choses how many dwarfs to put towards the military versus how many work.  Production is allocated between necessities, trade goods, military gear, etc.  Fort layout choices consist of where to place rooms, where to mine, how to layout the defenses, etc.  Players choose the starting location and the skills and items for the initial seven dwarfs.  There is no single best answer for any of these.  These are not "Curtain A or Curtain B" decisions.

NW_Kohaku, your comment on irrigation is an example of an interesting choice.  Irrigating the farm can be done various ways, with buckets or piping water to the field, with advantages and disadvantages to both.  But adjusting pH balance, rotating crops, fertilizing, leaving fields fallow, all are non-choices.  You choose to adjust the pH balance or the crops grow poorly.  You choose to fertilize the fields or the harvest is low.  Maybe these are "choices", but not interesting choices.  Solving these problems is akin to my child solving her long-division problems on her homework.

This is what the game needs more of - problems where the solution is a complex, context-sensitive solution that changes with every fortress you play, rather than simply allowing every fortress to be cookie-cuttered out with the same layouts and same decisions on how to train your military every time, and all the potential complexities of the game simply receding into "boring, unimportant detail".

We agree on this completely.  Just that if a complex problem only has one solution, it becomes more work than fun.

Logged
Back in 1971, Nolan Bushnell of Atari said, "All the best games are easy to learn, and difficult to master," a design philosophy now treated as instinctual by nearly every designer in the industry.

NW_Kohaku

  • Bay Watcher
  • [ETHIC:SCIENCE_FOR_FUN: REQUIRED]
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #38 on: August 09, 2010, 02:13:49 pm »

Honestly, I wouldn't use Civilization as an example for making games that are fun and not just a bunch of micromanagement...

The problem with those "Interesting Choices" is that there really was only one choice - what sort of strategy you were going for from the beginning - and then everything else is just a matter of whatever gives you the most bonuses towards furthering that strategy.  Players, especially in a single-player environment, will typically just make the same choice time and time again.

It's like there being an "Interesting Choice" in all these games where you can either be Good or EEEEEVIL, which are so roundly mocked nowadays - you get powerups based on consistantly making the same choice, and the decisions all wind up giving you functionally the same power-ups so that there is "game balance", so it was never more than the choice you made at the start about whether you'd be playing a good guy or a bad guy, and once you've determined that, there is no further choice - you just have to figure out which is the "right" choice (which is often easy, as these "choices" are typically "give an orphan some food" or "rip out his heart and eat it for no particular reason other than that it's EEEEEEVIL")

Frankly, if you want to set up the game as the sort of "Interesting Choices" that you are talking about (or at least, based upon what you are excluding, although some of what you dismiss so easily leads me to believe you just don't understand)... you really aren't going to get much of a game.  Or, in other words, Sid Meier's wrong (at least as you interpret him).

Even if we throw the fact that you would need to build a FAR more intricate irrigation system for a system where plants need water constantly rather than just once out the window, throw the fact that your own "Interesting Choices" were even less choices than what I'd built up (which implies you don't really know what you want), it still comes down to the fact that it doesn't have to meet your totally arbitrary and subjective "Interesting Choices" criteria.

It's purpose is to create a system, not a choice.  It purpose is to create something that appears to the player as though the game is actually simulating something similar to real life, even if that life happens to be growing giant mushrooms underground.  This system may force you to do a few tasks, yes, like spreading manure on the fields or planting Nitrogen-fixating crops.  Does this mean that, in order to be a good system, which precise nitrogen-fixing plant you choose should be a life-or-death fate-of-the-fortress-hangs-in-the-balance decision on the part of the player?  Not at all.  Does this mean it's nothing but routine homework if it isn't a life-or-death decision?  Not at all.

First of all, there's the entire matter of verisimilartude - even if complexity doesn't force particularly DIFFICULT choices, like having your character eat when they're hungry, it still has a purpose in both helping you recognize that the character is meant to be "alive" and a character you are meant to empathize with, not just a series of stats, as well as preventing players from just ignoring civilization entirely, and letting them march infinitely without resting or needing to resupply. 

The reason why Farming Improvements is such an overwhelmingly popular suggestion is that people are unsatisfied with a 5x5 farm plot and a single farmer completely nullifying all the food needs of their fortress, which makes the game feel shallow and gamey.

So, then, what does this Improved Farming system do?  It differentiates crops in a real way.  Some crops simply cannot be planted time after time, even with fertilization, because they deplete the soil too fast.  You need to let the soil lay fallow between heavy feeders, or better yet, plant cover "crops" or nitrogen-fixating crops to restore the soil. 

In a game, you really CAN'T make every choice equal, and sweep away all other choices.  A sandbox is only interesting when your decisions have consequences, and it's just absurd to have the consequences be "potentially equally good in the right circumstances" every single time. Sometimes, you have to give the player the "choice" of jumping off a cliff, because if you don't, they'll run and hop and wrestle right along the edge of the cliff, secure in the knowledge that the "invisible wall" will protect them from making "the wrong choice regarding the cliff".

With that said, the purpose of crop rotations is not to make ONE TRUE SET OF CROPS (especially since pests push you to diversify, and you won't always have the same crops to plant), but to give players access to a system that they can either run well or poorly, depending upon how well they want to analyze the situation and react to it. 

You, in fact, agreed that making a player design at least a simple irrigation system is a good way to make the player show a little mechanical skill and use water's in-game mechanics to achieve the objective of the player - an open-ended solution to a problem.  (Where a "Stupid Choice" is flooding your fortress because you didn't respect or understand the fluid system.)  How do you not see the relation between this and finding a working crop rotation system (Where a stupid choice is not understanding the system, and having crop failure)?  The objective is to use the crops, and their mineral depleting and replinishing properties, in a cycle so as to create a self-sustaining cycle.  It's also something that you only have to solve once, but which, thanks to differing crops, will take subtly different methods of solving the problem every time.  (Especially if we involve procedural crops...)

Even better still, you can change the exact crops you use to better suit your needs - cover "crops" that you turn over into the soil restore soil nutrients better than nitrogen fixation crops that you actually pick, such as legumes, but legumes are an additional, if sub-optimal, food source even while you restore your soil's quality for the heavy feeders.  Speaking of heavy feeders, you can use light feeders that can be grown far more frequently, since they don't deplete the soil as fast, or medium or heavy feeders, which may have more attractive fruits, but which require more intensive upkeep.  Beyond that, we're also getting "fruit orchard" trees (and mushrooms) that provide fruit once a year for fairly low upkeep, but which may not be as productive as well-tended seasonal crops will be, plus the ability to grow trees for the lumber industry in the same soil, as well as growing flowers (or decorative mushrooms) for decoration. 

Please, tell me how any of this is somehow inferior to the "Interesting Choices" you came up with for making Dimple Cups arbitrarily and nonsensically up your weaving skill and drop your dodging skill...

And you know what?  This system, even if it can be put in the background when a player has worked out a self-sustaining system, is exactly what this thread was all about: It's complex, not merely a buch of meaningless details.  The problem can perhaps be "solved", but the numbers in the equation DO have meaning to the player.  That's the whole purpose of this thread.
« Last Edit: August 09, 2010, 02:47:56 pm by NW_Kohaku »
Logged
Personally, I like [DF] because after climbing the damned learning cliff, I'm too elitist to consider not liking it.
"And no Frankenstein-esque body part stitching?"
"Not yet"

Improved Farming
Class Warfare

Normandy

  • Bay Watcher
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #39 on: August 09, 2010, 03:31:57 pm »

Although most of your points are fair, NW_Kohaku, I don't think you should dismiss Sid Meier's Civilization so easily. We're talking about Civilization, not Galactic Civilizations, which has the sort of good/neutral/evil "choices" you are talking about. On the contrary, Civilization has none of this pretext (pretext which isn't necessarily such a bad thing in the first place, but that's a discussion for another day). In case you have not played Civilization, we'll talk about the free alternative, FreeCiv.

In FreeCiv, you can choose your government type - each has its benefits and its drawbacks. Very early in the game, you must choose between Monarchy and Republic; both of these governments are unlocked by very different tech branches (relatively speaking), and thus you must choose one over the other. A very simplified way of putting it is that Monarchy allows you to garrison military units for free while Republic allows you to spend more money towards research. This does not deal with any hidden factors, which is essentially what is being proposed here. Here, I'm defining a hidden factor as anything that does not directly translate into gameplay metric - i.e. soil pH is first put into some formula dealing with plant fertility, which then determines the crop yield of that square. By my definition, soil pH is a hidden factor. Soil pH isn't necessarily hidden from the player, but I call it hidden because it introduces new information to the game without adding a new gameplay metric or interface for it - hence, it is "hidden", like encapsulated variables in a class.

The interesting thing about Sid Meier is that he has no hidden factors in his games. Everything is plainly visible to the player (except for maybe combat balance). Yet, I'd get just as many people to say "I wouldn't use Dwarf Fortress as an example for making games that are fun and not just a bunch of micromanagement..." Personally, I find this comment to be very hypocritical. I may not disagree with specific suggestions, but I agree with Hammurabi in that sometimes, the systems you are proposing amount to little more than homework. Though I admit that I like number-crunchy systems like the one you propose, I am afraid that it is hardly an 'interesting' decision, by most player's definitions. It's a non-trivial one that takes a little thinking to get through - which is interesting by a few people's definitions, including my own, but only because it takes a little thinking to get through.

Contrary to your assertions, there would be a 'best' method for balancing soil fertility, pH, 'feeder' crops, etc... I have done quite enough mathematics to know that such a system can be easily decomposed and turned into a simple model that will output me an answer as to what crops I will plant and when. I reiterate - this is interesting to me, and to some people. But this is not an 'interesting decision'. It's not a decision at all, after a certain point.

The key to making a fun game is not piling on hidden factors under the pretense of introducing 'interesting gameplay'. Sometimes it can help; like in the case of the material system. But to naively throw it everywhere, to bring unnecessary variables where they might not belong, is a very dangerous game.
Logged

Makaze2048

  • Bay Watcher
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #40 on: August 09, 2010, 04:19:59 pm »

The system contains interesting choices so long as you are reacting to it. Planting X or Y is an interesting choice if you're weighing the pros and cons of each and come to a decision. At the point you have solved the system and have perfect knowledge of which is better then there is no longer an interesting choice. Making things more complicated tends to stretch the time it takes someone to solve the system (and don't be fooled no matter how complicated you make it someone will solve it and post it on the wiki) and transition from interesting decision to rote following. But making things too complicated can prevent early decisions from being interesting as a novice player will not know how to even fathom the various pros and cons. Thus simply making it a guess which is not interesting. Additional complication also tends to water down the importance of individual decisions and after a certain point choices begin to loose interest value due to their lack of noticeable effect ie. planting X vs. Y will give me a 0.5% boost in total crop output... woohoo party...
Logged

HebaruSan

  • Bay Watcher
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #41 on: August 09, 2010, 05:14:54 pm »

Please, tell me how any of this is somehow inferior to the "Interesting Choices" you came up with for making Dimple Cups arbitrarily and nonsensically up your weaving skill and drop your dodging skill...

While I wouldn't want to call it something as inflammatory as "inferior", I do see two ways to strengthen this proposal. (My specific examples here are pretty boring, since they're only meant to illustrate the point; a good game designer ought to be able to come up with better ones.)
  • Different advantages to different approaches. I think this is what is meant by "interesting choices". So far we're talking about a difficult optimization problem with one unique solution, which would mostly recede into the background once solved (with some replayability with population growth, and in the next fort if crop availability varies). But if I could optimize the system towards different goals, then I have a worthwhile reason to come back to it later as my goals change. This is where "either-or" trade-offs tend to come in. Maybe one crop is best for taste, another for health, another for labor-intensity, and another for safe storage. Maybe some combinations give dwarves indigestion and unhappy thoughts when mixed in the gut. Maybe one irrigation method gives a higher average yield with erratic variations, while another is more consistent but less bountiful. There can still be good and bad choices within such a system; my design might fail to perform as I intended. But since the player has chosen the goal towards which he is trying to optimize, this has the potential to be more engaging.
  • Complex interaction with other subsystems. The proposed farming system sounds mostly self-contained so far. Insert water, seeds, ground, labor, and time, and in return you get some number of food units that prevent your dwarves from starving and supply some industries. The time you spend thinking about agriculture would be spent thinking only about agriculture, and if you're successful, all you get is a farm system that works. But if I have an incentive to use a large outdoor area for a particularly space-inefficient crop, then suddenly I have to worry about defense of all that space. If some crops required tilling the soil with (expensive?) metal plows, that crop might be out of reach until you start a metalworking industry. If the elves start selling terminator seeds to grow lembas wafer plants, then I start caring about diplomacy. Maybe some plants give off strong tell-tale scents that goblins can use to find you more easily. And so on.
These considerations would make or break a new farming system for me. If all it does is increase the amount of attention I have to pay to the process of making food, I'm going to wish I could go back to what we have now (unsatisfactory as it is). But if I get new strategic choices and new wrinkles in the overall operation of the fort in exchange for that attention, then that's all for the better.
Logged

Normandy

  • Bay Watcher
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #42 on: August 09, 2010, 05:55:17 pm »

Complex interaction with other subsystems.

This, ever so much. It's also precisely the difficulty in implementing a magic system, as Toady has alluded to that in the past in the DF Talks, so this is not something new to him. Try and shy away from isolated systems.
Logged

NW_Kohaku

  • Bay Watcher
  • [ETHIC:SCIENCE_FOR_FUN: REQUIRED]
    • View Profile
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #43 on: August 09, 2010, 06:27:48 pm »

Although most of your points are fair, NW_Kohaku, I don't think you should dismiss Sid Meier's Civilization so easily. We're talking about Civilization, not Galactic Civilizations, which has the sort of good/neutral/evil "choices" you are talking about. On the contrary, Civilization has none of this pretext (pretext which isn't necessarily such a bad thing in the first place, but that's a discussion for another day). In case you have not played Civilization, we'll talk about the free alternative, FreeCiv.

I wasn't talking about Civilization when I was talking about good and evil.  I was talking about how "Interesting Choices" often wind up not being very interesting and really wind up only being one choice that you stick with throughout the entire game.  I could go with something like focusing on rapid population growth and expanding to many small cities as fast as possible in a 4X game, but that would have been fairly complicated and required more explaining than the easier metaphor of games that have Good/Evil side mechanics, like, say, pretty much everything Bioware makes.

I didn't say this was a property of Civilization, but of "all those games that have Good/Evil mechanics", which does not include Civilization.

In FreeCiv, you can choose your government type - each has its benefits and its drawbacks. Very early in the game, you must choose between Monarchy and Republic; both of these governments are unlocked by very different tech branches (relatively speaking), and thus you must choose one over the other. A very simplified way of putting it is that Monarchy allows you to garrison military units for free while Republic allows you to spend more money towards research. This does not deal with any hidden factors, which is essentially what is being proposed here. Here, I'm defining a hidden factor as anything that does not directly translate into gameplay metric - i.e. soil pH is first put into some formula dealing with plant fertility, which then determines the crop yield of that square. By my definition, soil pH is a hidden factor. Soil pH isn't necessarily hidden from the player, but I call it hidden because it introduces new information to the game without adding a new gameplay metric or interface for it - hence, it is "hidden", like encapsulated variables in a class.

Contrary to your assertions, there would be a 'best' method for balancing soil fertility, pH, 'feeder' crops, etc... I have done quite enough mathematics to know that such a system can be easily decomposed and turned into a simple model that will output me an answer as to what crops I will plant and when. I reiterate - this is interesting to me, and to some people. But this is not an 'interesting decision'. It's not a decision at all, after a certain point.

Even if there is a "best" method of dealing with soil pH, I return again to the metaphor of the cliff.  DF is a game that isn't very hard in the normal sense, you can pretty easily build almost anything you want fairly quickly.  What the game lacks is any "guard rails".  Or in other words, when you find a cliff (or anything Fun, for that matter), you can be assured that the game is going to do absolutely nothing to stop you from walking off that cliff.  The cliff is there, and if you don't pay attention, you'll walk right off it.  This applies to the way that water or magma works - leave a gap, and you'll flood your fort, potentially dooming the whole thing.  This applies to forgetting to set up a farm or leaving a hole in your defenses and letting goblin ambushers get in or plenty of other ways you can have Fun. 

Basically, DF is easy when you know what you are doing, know where the cliff is, and respect it.  It's not an "Interesting Decision" not to walk off a cliff, there aren't any "different advantages" to the "different approach" of walking off a cliff, but having the cliff there, and knowing you can walk off of it is not only perfectly valid, but give the game real meaning.

Think about a real-life sandbox.  You can push that sand around however you want.  You have plenty of "Interesting Choices" in whether you push sand to the left or to the right.  But your actions never have any real consequences.  The game is given meaning by the fact that it restricts your actions, but still allows for very great and wonderful things to be accomplished, so long as you respect the properties of the fluid system.  Magma mishandled can easily destroy your entire fortress, but properly handled is the solution to a huge number of problems.

Or, to sum it up, it doesn't matter if you don't think it's an "Interesting Choice" or not, especially on paper, where you are throwing some fairly false assumptions onto it.  What matters is that it's a system you have to put some thought into.


I also don't see what you mean by it being "hidden" because it doesn't allow any "interface for it"... The player will be able to alter soil pH through his/her actions, especially through the addition of fertilizer (nitrogen fertilizers tend to be highly acidifying), so it's not like the properties of steel or iron where you can't really change them, you can only pick the best one, it's a variable that slides up and down as the player does things to the soil.

The interesting thing about Sid Meier is that he has no hidden factors in his games. Everything is plainly visible to the player (except for maybe combat balance). Yet, I'd get just as many people to say "I wouldn't use Dwarf Fortress as an example for making games that are fun and not just a bunch of micromanagement..." Personally, I find this comment to be very hypocritical. I may not disagree with specific suggestions, but I agree with Hammurabi in that sometimes, the systems you are proposing amount to little more than homework. Though I admit that I like number-crunchy systems like the one you propose, I am afraid that it is hardly an 'interesting' decision, by most player's definitions. It's a non-trivial one that takes a little thinking to get through - which is interesting by a few people's definitions, including my own, but only because it takes a little thinking to get through.

It's not really hypocritical, I wouldn't hold up DF to be something that frees you from micromanagement, either.

I am, however, interested in reducing this micromanagement even as I increase the complexity through a system that provides you with the tools to automate as much as possible - specifically, when you "solve" the problem of how to rotate your crops, the interface, as I have suggested through scheduling of tasks, would potentially allow you to have an agricultural system that is automated for as long as no disease suddenly breaks out, or farmers die for some reason. 

This is why I compared it to muddying the soil of a cave just once - you have to solve the problem once, which is its own problem with right and wrong solutions, but once you have solved it, you have solved it for good, without having to keep coming back to micromanage it.

The point is that you have to build that system in the first place.

(It's also only number-crunchy if you want to approach the whole thing from a "Plan out your entire agricultural system from embark" route.  Ultimately, if you just look at your soil, see it's high in nitrogen, moderate in phosphorous, and high in potassium, and has mild acidity, and that you have your watering mechanisms set up, then you can just plant pretty much any depleter that doesn't have high phosphorus requirements, and when things get low, plant replinishing crops.  Eventually, you'll be able to work out a stable system without really having to get your hands dirty, so to speak.)



1.Different advantages to different approaches. I think this is what is meant by "interesting choices". So far we're talking about a difficult optimization problem with one unique solution, which would mostly recede into the background once solved (with some replayability with population growth, and in the next fort if crop availability varies). But if I could optimize the system towards different goals, then I have a worthwhile reason to come back to it later as my goals change. This is where "either-or" trade-offs tend to come in. Maybe one crop is best for taste, another for health, another for labor-intensity, and another for safe storage. Maybe some combinations give dwarves indigestion and unhappy thoughts when mixed in the gut. Maybe one irrigation method gives a higher average yield with erratic variations, while another is more consistent but less bountiful. There can still be good and bad choices within such a system; my design might fail to perform as I intended. But since the player has chosen the goal towards which he is trying to optimize, this has the potential to be more engaging.

I think I went over this pretty thoroughly above, but, again, something isn't necessarily a bad idea just because some answers are obviously "wrong".  If every choice was just as good, but somewhat different from another choice, then there'd be no way to lose, now would there?  In that case, (as with the .5% increase reference Makaze makes), NONE of the choices become interesting, because you don't have any real negative effects.

In any event, again, I've described a system where there are many different kinds of crops you can use to start a repeating cycle.  Different crops do different things, but all you really have to do is recognize that some crops will cause changes to the soil that you have to balance out with changes caused by other actions in order to balance them out.  Particuluarly valuable crops will have very high and finicky requirements that will require several years of specific resetting of the soil to be able to plant them again (such as, again, the Sapphiric Rock Mold, that grows precious gems out of highly alkaline soil), while others are very easily put in cycles.  Different crops also have more uses beyond just simply feeding your dwarves, as well, as some crops make great alcohol, while others cannot make alcohol at all.  Some crops are used for making cloth or dies or potentially the components of future reactions.  When we gain the ability to start applying chemicals to weapons, you can bet that most poisons are going to be grown in your gardens.  Toady is also putting in farmable trees so you can get your lumber from farms if you want to set it up that way.  You can use orchard trees for low-maintainance, low-yield crops.  You can even make farms into gardens or parks for your dwarves to enjoy.  All of this by managing the balance of a few factors in the soil, but where you can choose to have completely different results for your work, depending on what you want to gain from it, and whether or not you are capable of balancing out the checkbook by finding a way to put back everything you took.

In other words, what you are arguing that I put into my suggestion is already there.

I keep listing these things, but nobody ever seems to want to acknowledge this...

  • Complex interaction with other subsystems. The proposed farming system sounds mostly self-contained so far. Insert water, seeds, ground, labor, and time, and in return you get some number of food units that prevent your dwarves from starving and supply some industries. The time you spend thinking about agriculture would be spent thinking only about agriculture, and if you're successful, all you get is a farm system that works. But if I have an incentive to use a large outdoor area for a particularly space-inefficient crop, then suddenly I have to worry about defense of all that space. If some crops required tilling the soil with (expensive?) metal plows, that crop might be out of reach until you start a metalworking industry. If the elves start selling terminator seeds to grow lembas wafer plants, then I start caring about diplomacy. Maybe some plants give off strong tell-tale scents that goblins can use to find you more easily. And so on.
Goblins already find you through cheating and not having any Fog of War.

Anyway, again, this is something that's already rolled into the system, as best it can be done.

Tree farming is already a part of this, so you have the lumber industry in it.  Your highest-grade irrigation system is to use metal pipes to create a suspended set of "sprinklers" using the water pressure you can get from standard hydrostatic pressure, which isn't that far off from requiring metal plows. 

You are also forgetting just how you procure fertilizer, or what it is...  "Fertilizer" is made of either dead stuff (plant or animal), manure, or possibly a few rocks that can be mined (especially if all you want to do is raise soil pH, you add Liming agents, which basically means powdered limestone or chalk).  Animals become a big part of farming, especially when they finally start having to be fed, as well.  Bonemeal is probably the best source you'll get in-game for restoring potassium.  Potassium actually gets its name from Potash, which means burning trees (which could be funny if you were trying to grow trees in a farm just to burn them into potash to spread back on a farm). 

While you probably won't need to actually consume much metal continuously, or use precious gems or glass at all (unless we want a "real greenhouse"), pretty much the rest of the resources in DF will be involved... Especially as these farms necessarily become much larger and more complex, and require constant inflows of water meaning that you need continuous access to a replinishable source, since that means that you'll build your farms where they need to be built so your irrigation pipes can be supported and fed, and the rest of your fort has to be built to accomidate that, the way that magma forges demand a reshaping of your fortress plans currently.
Logged
Personally, I like [DF] because after climbing the damned learning cliff, I'm too elitist to consider not liking it.
"And no Frankenstein-esque body part stitching?"
"Not yet"

Improved Farming
Class Warfare

zwei

  • Bay Watcher
  • [ECHO][MENDING]
    • View Profile
    • Fate of Heroes
Re: Dwarf fortress: Margionally complex, but mostly just detailed
« Reply #44 on: August 10, 2010, 02:18:51 am »

In FreeCiv, you can choose your government type - each has its benefits and its drawbacks. Very early in the game, you must choose between Monarchy and Republic; both of these governments are unlocked by very different tech branches (relatively speaking), and thus you must choose one over the other. A very simplified way of putting it is that Monarchy allows you to garrison military units for free while Republic allows you to spend more money towards research. This does not deal with any hidden factors, which is essentially what is being proposed here. Here, I'm defining a hidden factor as anything that does not directly translate into gameplay metric - i.e. soil pH is first put into some formula dealing with plant fertility, which then determines the crop yield of that square. By my definition, soil pH is a hidden factor. Soil pH isn't necessarily hidden from the player, but I call it hidden because it introduces new information to the game without adding a new gameplay metric or interface for it - hence, it is "hidden", like encapsulated variables in a class.

That example is actually not that good:

Say, my "strategy" is to be #1 in science.

Then, there are no interesting choices: Build barracks or library? Library wins. Choose Republic or Monarchy? You choose whatever increases your output in science.

And I do not see improved farming to be really all the different: Crops differentiation based on ph and similar factor, rotation, etc will mean that you will adapt to local features and that will be it: I(stead of interesting choice, you will have yet another insignificant detail (this fortress has best conditions to grow longland grass and fisher berries, guess this means it will be know for two types of alcohol and fisher berry cake.)
Pages: 1 2 [3] 4 5 6