Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Immortal]

Forgive me if I missed it, I read several pages now.
I would like to know peoples stance on mud? I liked it, the whole install a grate(foot cleaner) and the mud goes no farther, in real life mud would be tracked all over requiring shoes to be removed to walk in a home or in the dwarven case a cleaner to allow them into the fort from the farms, also it required actual cleaning of the fort.. Any ideas?

[Draco18s]

No, what Jiri Petru posted was bad design. Actually it wasn't design at all.

Go read Jiri's post.  He already said basically what I said.  http://www.bay12forums.com/smf/index.php?topic=60554.msg1478119#msg1478119

And I agree with the post just after it:

Sorry, but that sounds like no fun at all. You want to dumb it down even compared to the current system. A little micromanagement is good, and I like to choose my plants and set up crop rotation myself. You haven't described a simple and intuitive interface, BTW, you have described automation, i.e. AI doing things for you, which is a completely different thing.

[Atanamis]

The long term goal is for fortresses to have surrounding villages which would supply them with food. Toady's recent screenshots demonstrate the kind of villages that would support a town or fortress. Farming in most forts should be an interim solution, used until you get a noble who will be able to generate villages for extra food. In a late game with a king present, the only local agriculture I would expect to see would be specialty agriculture that cannot be obtained from the surrounding villages. These villages would send in food automatically as tribute, and in exchange you would be expected to house them during an attack by enemy forces. In fact, sieges might be telegraphed by refugees pouring into your fortress. Lock them out and let them die, and you don't have food wagons coming in anymore. This would create an incentive to destroy sieges as well, since growing enough food to handle your 200 residents plus 400 plus refugees in a besieged fortress would be challenging. (Obviously performance improvements would be needed.)

Short term, this can be dealt with by expecting most fortresses to need to buy a lot of food from traders. To ease in new players I'd be happy with an "easy crop" that becomes unacceptable as the fortress grows and dwarves become more demanding. Noble mandates should be used to inform the player when new crops should be added or rotated, making them part of the tutorial process. More desirable crops should require increasing complexity. Having dynamically generated crops would be great, so long as their usage could be determined by the player without too much difficulty. Or maybe that would be part of the fun?

[NW_Kohaku]

I'd really hate to see this much work go into a system that we don't even use, it's just silly.  Especially if seiges will have all your farmers from the fields running into your fortress, wouldn't it be better to have large farming layers to help ease the amount of food that you need to stockpile?  Especially if we go with making food rot like in this thread, we can prevent people from just stockpiling their way through seiges, and you would have to feed and house a sudden 300-dwarf influx of refugees in those cases.

This is the closest I'd be willing to go to an "Easy crop"... the infamous "Lumper".  You get a few years of easy harvests before the Blight destroys it:

NAME: Waterlump Tuber (The Lumper potato)

notes: This is based upon the Lumper potato strain.  For those who aren't that big on American/English/Irish history, the Lumper was the potato strain that fell victim to the Potato Blight that caused the Irish Potato Famine that led to millions of Irishpeople fleeing their native land to places like the United States. (See: http://www1.american.edu/ted/potato.htm) Unlike most modern potato strains, the lumper was actually not a particularly tasty potato, instead being rather watery.  It also grew somewhat slower than modern potatos because, as a early strain of potato that shares more in common with wild potatos than domesticated potatos, and as such, was leafier and spent more time blooming flowers.  The key to understanding why this was the crop that caused the famine is that these potatos were much easier to grow than other breeds (soil fertility-wise), and that they had essentially all the nutrients you needed to survive, except for Vitamin A and Calcium, which you could get from milk or cheese, which most poor people relied on for their needs.  This is why it became virtually the only crop grown in Ireland, making the lumper such an unprecedentedly easy target for the potato blight, which in turn made a single disease's rapid spread (thanks to the incredible density of the lumper population for that disease to feed upon) capable of starving a million people, and forcing even more people to flee the nation. 

PREFSTRING: lumps and knobs

EDIBLE: pest, cooked, grazer
TILE: 232
COLOR: 6 0 0

NUTRIENT_N: 130: 80: -22 (--)
NUTRIENT_P: 90: 50: -5 (-)
NUTRIENT_K: 90: 50: -2 (-)
SOIL_PH: 107: 127: 87: 147: 0
WATER_REQUIREMENT: -80 (--)
BIOMASS: 60: 130: 20: 190: 0

FULL_HARVEST: 33
VALUE: 1  (Actually, since this is a "for sustainance only" crop, I do think perhaps reducing the advantage for having a crappy crop (1 value basically means you get twice as many crops as a 2 value crop), so that we still count a 1 value crop as a 1.5 value crop for the purposes of balancing the equations.  This would make this plant only produce a FULL_HARVEST of 22.)

PEST:  Waterlump Blight, tuber weevil, wireworm

BIOME: Cold to warm, wet, marginal lands
SEASON: All year, aside from freezing
GROWDUR: 660

SEED TYPE: Destroys crop

Otherwise, why are we bothering?  People are trying everything they can to get out of learning a new system, but they're inevitably going to love it when they actually have it.

[Draco18s]

notes: This is based upon the Lumper potato strain.  For those who aren't that big on American/English/Irish history, the Lumper was the potato strain that fell victim to the Potato Blight that caused the Irish Potato Famine that led to millions of Irishpeople fleeing their native land to places like the United States. (See: http://www1.american.edu/ted/potato.htm) Unlike most modern potato strains, the lumper was actually not a particularly tasty potato, instead being rather watery.  It also grew somewhat slower than modern potatos because, as a early strain of potato that shares more in common with wild potatos than domesticated potatos, and as such, was leafier and spent more time blooming flowers.  The key to understanding why this was the crop that caused the famine is that these potatos were much easier to grow than other breeds (soil fertility-wise), and that they had essentially all the nutrients you needed to survive, except for Vitamin A and Calcium, which you could get from milk or cheese, which most poor people relied on for their needs.  This is why it became virtually the only crop grown in Ireland, making the lumper such an unprecedentedly easy target for the potato blight, which in turn made a single disease's rapid spread (thanks to the incredible density of the lumper population for that disease to feed upon) capable of starving a million people, and forcing even more people to flee the nation. 

Actually, it was the only potato (or really, the only crop) being grown in Ireland because the British weren't letting them grow anything else, which is why the blight was so effective.  And rational farmer would grow more than one crop so if disease comes through it only kills part of your food supply vanishes.

[NW_Kohaku]

Technically, they were also growing corn in the warmer months (for export), but potatos were what they used as their storage crop to get them through the Winter.

But yes, this is basically what naturally results from some outside ruling entity taking an overly simplistic view of farming, and wanting to produce only one "easy crop"... which is exactly why I bring it up.

[Atanamis]

NW_Kohaku, the potatoes sound great as a starter crop. All the essentials for survival, but it can't really be relied on exclusively long term unless you want to risk a famine. I still suspect that long term most fortresses will be depending more on their villages for food based on Toady's past comments, but I think that the ability to model complex farming is desirable regardless. Like you said, during a siege it may be necessary to supply large numbers of refugees, which would be excuse enough to plant some emergency farms. Even when farming villages are implemented, perhaps we will have enough direct control to mandate crop rotation and selection.

[nil]

a method for making farming easy at first but then hard later

I think soil-depletion could fulfill this need without the requirement of an specific crop.  For the first few years, a monoculture could be productive without much extra work.  After that, however, the soil becomes barren and one must either find new fields, fertilize the existing ones, or find the equivalent of a nitrogen fixer to mix into a rotation of crops.  Jiri Petru is right in that the interface and amount of auto-automation will make or break such a plan, but if done right I think it could scale up the difficulty level in a very organic and realistic way.

By the way, I love your procedurally generated crops idea.

edit: ditto the stuff you said about pests in the FotF thread

edit2: I'm gonna actually read this thread 30+ pages in the suggestion forum is daunting but a lot of what I'm hearing is interesting

[Jiri Petru]

Hello everybody. Kohaku invited me over so here I am 

First, let me say I'm both impressed and alarmed by your suggestions here. Impressed because there's a great amount of amazing material than can be used to make the game much better - great job with browsing sources, Kohaku. Alarmed, because is seems you are only adding stuff without much thought about how to handle it in terms of interface, user-friendliness and gameplay in general. A lot of people already can't grasp the current military (not because the system is bad, bud because the controls are confusing and user unfriendly). If another such fiasco gets added to the game - and in a system that is so important as farming - there's a huge danger DF will lose many players. Not because they're stupid and can't figure it out, but because the effort/entertainment ratio will get too skewed.

There's also a question of how much attention

should farming get. I guess most DF players are interested in wars, conflict, politics, trade, architecture, not in another Harvest Moon. And while many people voted for this thread in the eternal suggestions, let me point out that was before you came up with this complicated system. Most of them probably meant: "farming is too easy now, make it a bit harder, but you don't have to make it the central game element".  If the suggestion weren't "Farming Improvement" but "Complex NH+kPh system, soil quality, fertilizers and irrigation" there would probably be much less votes. What I'm trying to say: don't fall into the false assumption that more stuff makes a better game.

So from the gameplay point of view, this is how I think it needs to be done:

The golden rule: farming should be extremely easy to setup. In ideal conditions (temperate grasslands), setting up a farm should be as easy as designating a farm plot and enabling farming labours. Anything more (even crop selection like in the current system) is just a needless hassle. Note: this doesn't mean the system should be primitive. Even a complex system can be made easy by clever defaults and automation.

• Defaults: upon building a farm, the AI would automatically select appropriate crops and setup crop rotation. The crops would probably be food only.
• Automation: irrigation, fertilization, harvest and other maintenance issues would be handled automatically. The field would simply queue jobs as needed. Crop rotation should be automatic too, as above. Let me also stress that farms in temperate grasslands shouldn't require any irrigation whatsoever, the rain is enough.

Ideally no (or almost no) micromanagement and regular player-tasks. Again, in ideal conditions the player would only setup the farm and then forget about it completely. The AI should be clever enough to maintain the field by itself (see automation above). What we're trying to avoid is the player starving because he forgot to click the "Plant!" command this spring.

I know you now want to protest that this is "dumbing down" the game, but bear with me. The above requirements do not prevent complex systems of soil quality, etc., they only need such system to be bundled with a clever AI. That's because the challenge should never stem from controls and interface. Controlling the game should be as simple and intuitive as possible. Complexity and customization are fine, but only as long as they don't impede the game. This is what happened to the military in DF2010: the new system allows unprecedented levels of customistion, which is great, but it also forces the player to customise while not allowing any simple solution, which is dreadful.

Having point-click-and-forget farms is a good thing because it allows the player to spend less time doing menial tasks and focus more on Fun. There would still be Fun in farming, but would be derived from special conditions, events, etc., not from controls.

• Challenge 1: Effectivity. While the defaults and automation would maintain the farm indefinitely without much player's intervention, it wouldn't be very effective. Not deliberately, but simply because the AI can never be so good. It would produce food, right, but it would take a lot of time and dwarfpower. Designing clever irrigation canals or reaping machines or whatever is fun. Carefully selection crops for specific soils, weather and seasons to maximise efficiency and save dwarfpower is fun but should always remain optional.
• Challenge 2: Less than perfect conditions. The above simple system works for good lands where farming is realistically easy. Worse conditions could require more effort and forward planing. Farming in underground, irrigating deserts, drying marshes, fighting against salty soils, etc. etc. These can be very difficult and complicated. The point here is that the player chooses the embark spot, and as long as he has a chance to embark upon easy farming land, all is good.
• Challenge 3: Events and scenarios. Grasshoppers, diseases, floods, fires and other natural events... Sieges that destroy crops, evil snails who poison your wheat (that was a nice idea, Kilo!)... all of these require immediate player attention and can also require more complex countermeasures (like manually changing the crops you grow). This is Fun! Micromanagement in times of crisis is a neat gameplay element, we just want to avoid micromanagement in times of boring.
• Challenge 4: Special crops. Some crops like those needed for cloth or dyes could require more attention.

So yeah, that's about it. Perhaps I've forgotten about some challenges but I think you get my meaning. The design goal is to implement a farming system that is both easy to setup and maintain AND allows for interesting challenges. We don't want a system that allows for challenges but is also user-unfriendly, hard to learn and pain in the ass to micromanage. Farming is such a key element of a successful fort and economy that every player needs to be able to control it easily.

Note that my points are not in direct contradiction with the complex system you've come up with here. It all depends on implementation and if Toady can implement your system while satisfying my wishes here, I say go for it. But unless these points can be met, I say don't bother. Anyone else than the most "elite" players would simply stop farming altogether (and import all their food instead) or say "Screw it, I'm done with DF!". It's already happening now with the new military.

I like your NH+kPH (or whatever it is) system for it would allow for more interesting biomes and varied farming culture. But I admit I'm still not convinced it can be implemented in a user-friendly way, and I'm in some doubt about whether it's worth the trouble. Communicating the NH+kPH stuff to the player would be a very difficult task. And this:

"Plant stalks look healthy"/"Plant stalks seem slightly thin"/"Plant stalks seem slightly thin, with thin purple veins appearing"/"Plant stalks are thin, with purple veins, and a slight blueish leaf tint"/"Plant growth is stunted, with spindly thin stalks and purplish-blue veins and splotches on the leaves"

is not a good solution (please don't be insulted). The player would still need to consult a table or wiki to know what the hell the "blueish leaf tint" means. If the interface said "Soil quality: 3" instead it would have the same (bad) effect. So yeah, I'm still not convinced this system can be handled in a user-friendly way.

And I also doubt farmers in medieval had all these information! It sounds way too modern in scientific for my taste.[/list]

[nil]

Jiri--is your ideal farming setup one where dwarves will make choices such as what crops to use on a rotation schedule, whether or not to use fertilizer, etc themselves, or one in which the player sets up a system (hopefully through a straightforward, easy-to-use interface) which the dwarves will then carry out without further supervision?  Personally, I prefer the latter so long a player isn't confronted with the challenges until the fortress is established and simple but less effective alternative options exist.

I've read a pretty good chunk of the thread, so let me pontificate--I am a moderate on the farming question.  My ideal system:

• All soil and muddied rock tiles have an abstracted 'fertility' value.  I'd use no more than ten nor fewer than five levels of fertility.  A tile with a fertility level of 0 would be barren, and one with the maximum value would be comparable to a farmable tile in the current game.
• Muddied caverns and soil layers have varied default values, but generally (with a few exceptions like sand) will start with high fertility levels.
• Farms slowly drain fertility when used.  Non-stop farming of a monoculture should be effective in most soils for several years--4 or 5 imo--before production starts to wane, and a field should be able to take at least a couple more years of abuse after that before crops start failing completely.  Some crops should be more draining of fertility than others, and certain plants (which for the sake of balance should otherwise be inferior) should have a positive effect instead.
• Fertility is regained through fertilizers (created from refuse, potash, certain mineral deposits) and the aforementioned crops.  Fields left fallow or unfarmed slowly return to their default soil fertility; for most soils this will be a good thing.  The player uses a simple toggle in the farm menu to activate the use of fertilizers (much as we already do).
• Fertility is displayed in the farm menu, graphically by the shade of the farm tile. and perhaps through 'k.'  Crop failures (i.e., a soil tile with a farm built on it reaching zero fertility) should have a red announcement.
• Farms should take up 4-9 times more space than they do right now.  The simplest way to do this is probably to change what is now a single tile of farmland into a 2x2 or 3x3 one--one seed goes in, 1-4 plants spout up scattered on the field
• A bucket of water should be required to plant some--but not necessarily all--seeds
• Implement something like NW_Kohaku's vermin ideas
• Procedually generated plants, hell yeah
• pie in the sky: we get a raw file for 'tile characteristics' so modders can go crazy making super-realistic farming soil simulations

So, in the sort of system I'm looking for player can afford to totally ignore farming for several years, planting plump helmets season after season on a single set of plots.  After that time, he or she is forced to choose between a very simple solution (just build another farm), a pretty simple solution (obtain and grow fertility-promoting crops) and a relatively complex but potentially very effective solution (create and apply fertilizer).  Between a single abstracted fertility system, increased farm size, increased attrition from vermin and possibly spoilage, and water requirements, we might finally get farming balanced--and if it is overkill or underkill at first we'll have a number of parameters to work with.

And a lot of the irrigation ideas in here are really cool and I hate to snub 'em but ultimately it's not worth the trouble (in both player overload and programmer time) which you can do something as simple as requiring a bucket of water during the planting job.

[Jiri Petru]

Jiri--is your ideal farming setup one where dwarves will make choices such as what crops to use on a rotation schedule, whether or not to use fertilizer, etc themselves, or one in which the player sets up a system (hopefully through a straightforward, easy-to-use interface) which the dwarves will then carry out without further supervision?  Personally, I prefer the latter so long a player isn't confronted with the challenges until the fortress is established and simple but less effective alternative options exist.

For me personally, the former is the way to go.

I'm a huge proponent of good default settings in the game. When you build a farm, it should already come with good default settings (either fixed or determined by AI on the spot) that the player may leave as they are. Customisation is good, but should remain optional. For farming this means that yes, the fields would come with crop rotation/fertilisation/irrigation schedules already set somehow. Not necessarily in a brilliant way, just in a functional way. The player would only change them when he has a specific goal in mind. (Now apply the same thinking to military and other aspects of the game).

Note this would not dumb the game down, nor would it make it any less challenging. It would only save the player a few minutes of initial frustration/confusion when he is setting the farm for the first time. Great for newbies, and the experienced players can ignore it and overwrite the defaults right from the beginning.

EDIT: Apart from the initial setup, some things should be automatic and some not. Automatise the boring tasks that have to be done all over (planting, harvest, watering, fertilisation) and are not "fun" to play. But whenever an interesting event or challenge comes (say, the goblins salt your soil), the player should handle it manually. That's what I consider a good design: automate the boring parts, let player focus on the interesting events.

Also: I like your system more than the more complex one 

---

EDIT2: I think a nice compromise would be a "plant anything" option in the crop selection, enabled by default. With this setup, AI would choose crops, fallows and their rotation itself. It could be clever enough that when I run out of specific seeds, yet have half a field empty, it could fill it with whatever I have left. This default settings wouldn't probably be very effective, and would be heavily focused on producing cheap, easy-to-grow, safe but not very tasty crops. You could overwrite it by your own settings - for luxury foods, textile industry, alchemical ingredients, etc.

Fertilising and irrigation should always be automatic (the AI would figure out how much is needed), unless you manually turn it off and setup your own schedule.

[Zalminen]

I think a nice compromise would be a "plant anything" option in the crop selection, enabled by default. With this setup, AI would choose crops, fallows and their rotation itself. It could be clever enough that when I run out of specific seeds, yet have half a field empty, it could fill it with whatever I have left.

This solution I like.

The possibilities provided by NW_Kohaku's system are great but I also think they should be coupled with 'dumb but working' defaults.
Anyone new to the game won't starve while figuring how everything works, but it's easy to start adjusting it to something better when they're ready. 

[NW_Kohaku]

Jiri, what you're overlooking is that Toady was the one who wanted a NPK+pH system, and said that he simply was looking for a way to present it to the player in a way that conveyed meaning without conveying simple soil fertility numbers. 

Regarding the "farming improvements" goals, how much detail is currently planned for tracking soil quality?

A completely accurate model would probably be a lot of effort/information with little to gain from it, though enough detail to properly encourage crop rotation seems like something that should make it in eventually.

We haven't made any final decisions.  I think a NPK+pH model does give you something back, because you'd get some really great varied local landscapes and it would take care of crop rotation, composting, naturally poor soil, or whatever else, but it introduces a farming interface problem to dwarf mode in terms of conveying the information in wholesome terms and allowing you to solve problems that come up.

Which is exactly why I went and looked up what a lack of Nitrogen in a plant would look like, and how real farmers can tell when they need more nitrogen.  This is not me thrusting "an overly complex system" onto the thread (I actually argued something simpler against Silverionmox, who wanted this complex version), I am trying to weave what Toady wants into the best system I can in the suggestion.

As for saying that nitrogen is wrong because "someone would have to look up what it meant"... well, that just brings up the same argument over "having to look up what an aquifer means" all over again - so what?  We can't have anything nice in this game because then we might force someone to learn something?  Sorry, but once again, that's just not a valid argument.  Making a good system comes first.  Making it easy to understand comes second, if that.

This also goes for Nil's "fertility" only system.  NPK is being used specifically because of how it takes a serious crop rotation system to satisfy it, and, presumably, because Toady's a really big fan of using realism whenever he can because he enjoys adding as much verisimlartude as he can to the game, and more power to him for it.

As such, I have spent this time trying to come up with the best way to actually impliment the complex system Toady wants, and how to best display it to the player.  There's a reason the "this is what the system should have" parts of the argument died down...

[nil]

Jiri--is your ideal farming setup one where dwarves will make choices such as what crops to use on a rotation schedule, whether or not to use fertilizer, etc themselves, or one in which the player sets up a system (hopefully through a straightforward, easy-to-use interface) which the dwarves will then carry out without further supervision?  Personally, I prefer the latter so long a player isn't confronted with the challenges until the fortress is established and simple but less effective alternative options exist.

For me personally, the former is the way to go.

I'm a huge proponent of good default settings in the game. When you build a farm, it should already come with good default settings (either fixed or determined by AI on the spot) that the player may leave as they are. Customisation is good, but should remain optional. For farming this means that yes, the fields would come with crop rotation/fertilisation/irrigation schedules already set somehow. Not necessarily in a brilliant way, just in a functional way. The player would only change them when he has a specific goal in mind. (Now apply the same thinking to military and other aspects of the game).

Note this would not dumb the game down, nor would it make it any less challenging. It would only save the player a few minutes of initial frustration/confusion when he is setting the farm for the first time. Great for newbies, and the experienced players can ignore it and overwrite the defaults right from the beginning.

I am really wary of this level of automation.  Not so much because of a fear of dumbing things down (although I do think it's reasonable to expect even newbies to do something as simple as selecting which crop to grow, which is for example a far less complicated decision that deciding which ones to embark with in the first place) but because unless its done flawlessly automation it becomes confusing and troublesome in its own right.  Why are my dwarves wandering down into the caverns and getting killed by giant cave spiders?  Because my dumb, automated loom is initiating jobs I didn't order to create a thread I don't want.  Why aren't my adamantine wafers getting created?  Because my dumb, automated hospital is using the most valuable material in the game to suture wounds.  Why are my farms spamming me with planting cancellation notices for some nitrogen-fixing tuber and/or bird guano when I don't have any of the required seeds and never intended to plant the required crops?  Because my dumb, automated farm thinks its doing me a favor by running a rudimentary crop rotation and fertilization program.

I understand your concerns about newbies being overwhelmed, but as I said I think this could be avoided by having initial fertility values high.  If soil depletion doesn't significantly affect your yields for the first several years, then setting up farms wouldn't really be any more complicated than it is now.  I've read every post in the "Things that turned your off dwarf fortress" thread and I don't think I've seen anyone complain that the current system is too complex.

Just to be clear, I am not in favor of micromanagement around the lines of "loosing crops because the player forgot to order watering" or requiring any more input into fetilization beyond making the fertilizer and telling your farms whether or not to use it.  On the other hand, I'm not sure anyone is in favor of that much micro.

Jiri, what you're overlooking is that Toady was the one who wanted a NPK+pH system, and said that he simply was looking for a way to present it to the player in a way that conveyed meaning without conveying simple soil fertility numbers. 

Doesn't sound like Toady is fully endorsing a BPK+pH system, just interested in it and exploring the options.  Which is great, and if Toady can smoothly implement it without turning farming into the next military system I'll be thrilled--I've always been interested in soil management and the like.  But three to four independent soil qualities are a lot, and if realistic values are used (i.e., only a small number of soils have optimal levels to begin with, most require at least some fertilizer or crop rotation right off the bat) you definitely run a risk of throwing newbies off the deep end.

[Draco18s]

Jiri, what you're overlooking is that Toady was the one who wanted a NPK+pH system, and said that he simply was looking for a way to present it to the player in a way that conveyed meaning without conveying simple soil fertility numbers. 

Doesn't sound like Toady is fully endorsing a BPK+pH system, just interested in it and exploring the options.  Which is great, and if Toady can smoothly implement it without turning farming into the next military system I'll be thrilled--I've always been interested in soil management and the like.  But three to four independent soil qualities are a lot, and if realistic values are used (i.e., only a small number of soils have optimal levels to begin with, most require at least some fertilizer or crop rotation right off the bat) you definitely run a risk of throwing newbies off the deep end.

Its why my original suggestion was just an NP system (Nitrogen and Phosphorous) because we had five main underground plants:

Plump Helmet: Grows anywhere, has no specific soil requirements (but does have an optimal soil, being middling N and middling P).
Cave Wheat, Sweet Pods, Pig Tail, and Quarry Bushes would all exist in one of the four quadrants of a 2 axis system: requiring one high (N or P) and the other low, and depleting the high one and fixing the low one.  In soils that weren't ideal, but were near-ideal, they'd grow, but produce less.  In soils that were diametrically opposed to their requirements, they'd either grow extremely poorly or not at all.  This is why plump helmets grew anywhere: there was no diametrically oppose soil quality: only "best" and "good enough."

Not entirely realistic, but playable without needing fertilization (eg. crop rotaion) or without crop rotation (eg. fertilization) depending on what you wanted to grow (hell, you could even grow two crops and do partial fertilization!).

And having read up more on the NPK+Ph system I'd actually swap out my P for the K.