Print

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

[GavJ]

Why the condescension? I think I've made it pretty clear which information I consider unrealistic, so why are you comparing Legends mode ("This monster, which you probably won't ever encounter because it lives half the world away from you, once smashed the third finger, left hand of a human in some settlement that you couldn't care less about") with the data viewable with the Tab key ("Hey, this is the civilization that YOU ARE FROM")? The main thrust of one of my major suggestions is not simply hiding information, but hiding information based on distance, difficulty, and hostility.

I'm simply pointing out inconsistency.

You DON'T think people have sufficient self control to avoid hitting Tab to see information they don't find fun or realistic.
You DO think people have sufficient self control to avoid loading legends maps or choosing a nearer civ first for info before aborting and switching to a real fort.

And you treat the difference as being so extreme as to scoff at one version and yet show serious concerns requiring hardcode changes for the other version? This is unrealistic. People who have good roleplay self control can already deny themselves the unrealistic survey data easily. And people who don't have good self control would not be limited by the proposed change. I agree with most of the embark suggestions, but the obfuscation of info one simply doesn't make sense in its current form. Needs further revision to address the backdoor alternative or dropping it.

edit:
An example if you truly want to obfuscate info of a way of doing it might be forcing a player (in a certain game mode) to choose a civilization at the start of worldgen and then only ever being able to play an adventurer with them, start forts with them, and only able to view legends known to them, across the board.

[Dirst]

A handful of detail levels would probably work better than presenting potentially false information to the player.

Could you elaborate on this? How would your system work?

Toady first needs to decide what the absolute highest level of detail that he's willing to give the player, which conceivably could include things like detailed local topography and a rounded-off version of DFHack's "prospect all" data.  This would be the highest level of detail, probably only available very close to your civ (if not actually within its borders) due to direct experience with the area or adjacent areas.  A civ's underground tunnels would be considered within its borders for the purposes of site survey detail, but not necessarily surface roads.

The medium level would give you something akin to the current embark information with three exceptions: (1) it mentions sand resources, (2) it gives a smidge more detail than "multiple constructions", and (3) it mentions layer types, and only specific layer materials that are visible from the surface.  This needs an in-game guide of some sort, pulled from the raws, to inform the player what sorts of minerals could be expected.  The knowledge of layer types is a little game-y, but can come from role-playing explanations such as gods or divining rods or "sniffing the rock" or whatever.  This level of detail would be available for land that your civ could reasonably have surveyed (a certain radius around the borders, a larger radius around the civ for tiles with a road, plus caravans and large armies survey everywhere they've been).

The low level of detail would give you climate and surface features only, which is anywhere else in the world.  We might want to include aquifers as "surface features" since their presence is game-changing, with a role-playing explanation that they leave tell-tale (to a Dwarf) marks on the land.

One of the map modes on the embark screen would indicate where the high, medium and low levels of detail are for that civ.

Of course this is just a rough idea, the levels of detail could use tweaking.

[GavJ]

You're being unrealistically stingy with info, I think.

1) In real life, if any explorer or skilled party (caravans, armies, diplomatic missions, possibly dedicated explorers, road builders, settlers en route to existing settlements, etc.) from your civilization had ever moved past a region of the map, then a player should have full DF-prospect (surface-level) information about the region.

That's because DF-prospect-level info is basically just a diary entry, which any state official would keep as they trudged along, for exactly this reason. "There are a lot of plants here, and I sketched out the main species. The rock is of this type. We noticed a couple of veins of gold and hematite in the bluffs of such and such abundance," and so on.

Since armies and things now actually exist in the world, and caravans and other things should soon as well, this should be easy to implements. They just trace out thick lines as they move of civ known information. Caravans might have to get home safely to report it. Armies don't have to (they have implied scouts and supply lines that would carry info back continuously).

Road builders should probably even be represented by actual simulated surveyors. I.e. as the path algorithm runs to determine a road, pretend like the algorithm itself is a bunch of surveyors, and add detailed knowledge of covered ground on the map to the civ's knowledge.  And so on, you get the idea.

2) The current level of info on embark represents realistically something like hearsay, and should extend all the way to any regions of the world visited by any civilizations that your civ trades with. Often, you might even trade detailed info from partners, if you're on good terms, equivalent to #1.

3) Less info (like only biome, nothing else, basically) would represent far off lands only beyond the borders of anybody you interact with. On a small map, this might even be nowhere at all. On a large map, maybe 1/2-2/3 of the map qualifies @200 years.  At 5,000 years history or something, every civ should probably know every corner of even the largest maps (look at the influence ranges in legends politics maps)




Also, over time, civs would likely start sending out diplomatic and exploration missions, the equivalent of Marco Polo type things. They would meet with foreigners far away and gain hearsay or even occasionally detailed regional maps from diplomacy, and as they move, they'd add first hand knowledge of far away places. Since geology doesn't change much over history, this all adds up and as you go, more and more of the world knows everything about every site for longer run worlds.

But yeah, for the most part, huge swaths of the map, in a realism sense, would/should have DF-prospect-level detail to them, at the LEAST. Maybe more than that within a healthy distance of the civ's own settlements (as they start digging they learn even about deep undergroudn stuff) and those of close allies.

[2074red2074]

Honestly, we're talking about the same creatures who use quantum physics for levers and pressure plates. I'm sure they can figure out what kind of metals are under the ground. Maybe their technology can't analyse metals but can tell them apart if they are close together, so you can only know how many different types there are and not what those types are. Seeing as how their race relies on mining, mining tech should be fairly high priority.

[GavJ]

There are certain things it's easy to suspend disbelief on for 14th century dwarves. Hypothetical metal that's lightweight and super strong? Sure, whatever.

Ground penetrating radar? Notsomuch, IMO, not by a long shot.

Like the OP, this is definitely something that bugged me from my very first playthrough.



I could believe dwarven geologists, which would indeed give hints about deep resources, but A) it would be very different in nature than this information, and B) it would sort of require there to be geology.

[Niddhoger]

I agree completely.  Giving yourself 1 sand of each type and 1 unit of each food/milk for free barrels and bags can easily net you 300+ embark points worth of barrels and bags.  Barrels are simple enough to make, sure, but bags are a more involved process (process thread/shear animal->weave cloth->make bag, or kill->tan hide->make bag).  Either way, you are gaining hundreds of points worth of free items.  40 different food types is 400 points worth of bags/barrels, whereas the game normally just gives you 3 types of food in 3 barrels.  If the value isn't directly added, then it should be if you are "underusing" the container.  Asking for 10-20 goat eyes? Sure, we'll throw in the barrel for free! Asking for 1? We gotta give you a "convenience fee" fer that, mate.  Perhaps you only get 10 "free" barrels, and anything over that you have to pay for.  Better yet, it always seemed silly to only store one animal per barrel.  Sure, your quartermasters will pack badgers eyes and badger tripe in the same barrel, but they won't put badger eyes with your hyena eyes? In your fort, your dwarves will stuff all the plants into one barrel, and all your animal parts in another- why not do this at embark? After selecting all your food, it then calculates the minimum amount of containers needed to store everything.  You could still ask for 21 meat parts and 21 plant parts to game 2 free barrels... but it would be substantially better than gaining 20-30 free barrels.

Animal prices are also pretty wonked... a sheep is 51 while an alpaca is 101... both produce milk/wool in addition to meat, but an alpaca won't produce over twice the products (50% more).  A cow isn't even worth as much food as an alpaca, but worth 50% more than that (an can't be sheared).  Then there is a full grown turkey (6 embark points) that is wroth more food butchered than the sheep! A single cat is 11, but completely replaces the need of setting up a trapper (and does it so much more quickly), and is only worth 11.

Another idea is that the cost of resources can vary depending on the amount of sources (and possibly distance) in the empire.  If your empire has only one small, far away savannah fort, then cougar meat should be rare, exotic, and expensive.  Conversely, since all dwarven civs have mountainhomes, then mountain-based products should be a dime a dozen and cheap staples of your civilizations economy.  At the very least, anything not of a cavern/mountain biome shouldn't be as readily available as exotic imports from the very edge of the empire.  These are DWARVES, not ELVES.  They shouldn't be intimately familiar with above ground crops and animals.  Until DF2014 you couldn't even bring above ground crops with you, but could still import a host of exotic (above ground) animal products.  Goats are just right outside the gates of your mountainhome, but why would the dwarves want to trudge way out into the WOODS at the edge of your kingdom to hunt echidna (especially for your ONE unit of echidna tripe!)

I do love the idea of being given mandates by the king.  I always felt that lack of a goal... sure the outpost is to "produce wealth" but why would the kingdom want to bankroll your expedition to that evil glacier (lacking metals but infested with goblins) on the other side of the world? I usually like to think of my expeditions as procurement centers that will enrich my empire with exotic new goods... but again, you must choose a civ before you can view what animals/materials are available to them. 

[SixOfSpades]

Aaaaaand I'm back.

I'm simply pointing out inconsistency. You DON'T think people have sufficient self control to avoid hitting Tab to see information they don't find fun or realistic. You DO think people have sufficient self control to avoid loading legends maps or choosing a nearer civ first for info before aborting and switching to a real fort.

No, my issue with the Tab-key info has nothing whatsoever to do with self-control. Rather, I find the embark window faulty in that it provides both too little and too much information, simultaneously. It shows us too little because "Shallow metals", as I've said before. (Some knowledge of which stone types are sticking right out on the surface would be nice, too, but hardly critical.) And it shows too much because it gives a 100% accurate assessment of things that would be nearly impossible for even professional explorers to know with certainty. Sure, if a piece of land is right between a lake & a swamp, it's a fairly safe bet that there's a pretty shallow aquifer there. But detecting flux stone 200 feet down? Candy 500 feet down? And it's below us when we stand just here, but not right over there? Get real.
I do not find it the least bit abusive for anyone to scroll through the Tab-key info while embarking, nor is there any reason that I should. Relative elevation, slopes, neighbors, this is all pretty basic stuff, and it's presented in rather broad strokes--you know that the bend in the river is north and a little east of the volcano, but not precisely how far north or east. And that's as it should be, IMHO. This level of knowledge is something that I would not appreciate degrading with distance, because it's already appropriately vague.
As for Legends mode, I don't even see why you consider it a relevant comparison to the Embark window. Sure, it shows the whole world, but only at the most basic level: Forest, Glacier, Mountain, River, Desert, Plains, City, Sea. It doesn't reveal temperature, aquifers, water salinity, or savagery. It DAMN sure doesn't give any information on soil or minerals. I forget, but I don't think it even shows volcanism. You cannot search for elevation / drainage / etc. Now, I haven't downloaded any of the new versions of DF yet, but unless Toady has made some severe overhauls to Legends mode, giving it a high-resolution magnifying tool that reveals the kinds of info formerly seen only in the Embark view, you are damn right that I "scoff" at the very idea of using Legends mode to inform an embark--because Legends mode has almost NO relevant data to give. Planning an embark with just Legends really WOULD be like mashing the keyboard.
Now, you're right that it's very feasible for a player (in my own personal "ideal" embark scenario, described above) to choose a "dummy" civilization, picked solely for its proximity to the desired embark site, to examine the site thoroughly, before quitting and embarking with a civ nowhere near that site. I brought this possibility up myself. But you know what? I don't mind. It's pretty obvious that a player who does that is knowingly circumventing the "rules," so if they want to be cheesy, hey, it's their game. (This is the only part of my earlier statements that had anything to do with self-control at all, so I have to wonder how you got off on that weird discipline tangent.) I might care more about this if there were appreciatively greater differences between the dwarven civs (the only truly meaningful ones are being at war with other civs, and lack of access to iron & steel), but even then I rather doubt it.

Toady first needs to decide what the absolute highest level of detail that he's willing to give the player, which conceivably could include things like detailed local topography and a rounded-off version of DFHack's "prospect all" data.

Yeah, I thought about that, and I think I'd like a system where the "surveyors" gave you a percent chance (rounded to the nearest 10%) that the site would have [ Trace / Minor / Moderate / Ample / Abundant ] amounts of each type of mineral. So if they said they were 60% sure of moderate amounts of limestone, there might not be enough to build an aboveground castle, but you're pretty much guaranteed to have enough for flux. But a reported 20% chance of trace amounts of native silver? There may not be any there at all. (The obvious problem with this setup is the difference between layer stones and ores / gems. What a player might consider to be only "trace" amounts of diorite could easily be many, many times more than what they would call "abundant" emeralds.)

A civ's underground tunnels would be considered within its borders for the purposes of site survey detail, but not necessarily surface roads.

Good call. The general layout of the surrounding hills, and what sorts of vegetation grow on them, can largely be gained by just a quick look around--but that's hardly even [ahem] scratching the surface.

The knowledge of layer types is a little game-y, but can come from role-playing explanations such as gods or divining rods or "sniffing the rock" or whatever.

Yeah, I like the idea of giving the gods something to actually DO (besides curse people, that is). So when Toady implements the Religion arc, this would tie in nicely, with Priests offering guidance on where to dig, and choosing prospective Miners based on their faith.

One of the map modes on the embark screen would indicate where the high, medium and low levels of detail are for that civ.

Also good. I just feel that embarking should feel something akin to real-world 14th-15th century maps: They flat-out admitted that there were places they simply didn't know, with a big old "TERRA INCOGNITA". And they got things wrong, believing the hearsay, writing "HERE BE DRAGONS" and filling the seas with mermaids and sea serpents. (Okay, so DF actually has dragons, merfolk & sea serpents, which muddies my point, but you get the idea.) Or Columbus's globe, showing the full-size Earth with nothing but Ocean where the Americas (and Australia & Antarctica) turned out to be.
I want "Terra Incognita." It makes the game feel a lot more immersive, somehow.
I also want "Here Be Dragons," but I'm also okay with other people disagreeing.

1) In real life, if any explorer or skilled party (caravans, armies, diplomatic missions, possibly dedicated explorers, road builders, settlers en route to existing settlements, etc.) from your civilization had ever moved past a region of the map, then a player should have full DF-prospect (surface-level) information about the region.

Well, in REAL real life, events happen. So if you're going to play the "Everyone's a Prospector" card, I might as well play the "How Long Ago?" card, which reminds you that one of your armies passed through here once . . . but someone's built a town here since then. And this river has now changed its course. And they saw some iron here, too, but someone seems to have mined it all out & moved on. DF is more like medieval Europe (with goblins) than Lewis & Clark, so while you know that the Mandans aren't going to change the landscape, you cannot make that same assumption of the Dutch.
Mind you, I think knowledge of the surface (and depths, if the game is actually going to separately track subterranean traffic) based on civilization movements is actually a good idea, but isn't that almost certainly already taken into account with the map of a civilization's influence? Your list of "caravans, armies, diplomatic missions, road builders, settlers" all tend to travel in narrow lines, very inefficient for covering ground when compared to the other item on your list, "dedicated explorers". So to me, it sounds like frequently-traveled surface roads should confer "vision" of approximately 1 embark tile (the only things you could tell for certain at greater range than that are mountain / volcano peaks), and armies maybe 3-4 embark tiles (depending on the size of the army). The algorithm for calculating how well your civ knows the lay of the land could very well be better off calculating the explorers only . . . based on the distance from the Mountainhome or similarly large settlement, with a modifier for being within the "realm", and another for the harshness of the territory & its inhabitants. Just like traffic designations.

More later.

[SixOfSpades]

DF-prospect-level info is basically just a diary entry, which any state official would keep as they trudged along, for exactly this reason. "There are a lot of plants here, and I sketched out the main species. The rock is of this type. We noticed a couple of veins of gold and hematite in the bluffs of such and such abundance," and so on.

Dwarves have casual photographic memory, now? If this "state official" was part of the very first foray into this previously uncharted landscape, then yes, I might see him taking at least an hour out of every day of the journey to jot down notes & drawings like this. But in any other circumstance, no. He's got a job to do, and stopping every 5 minutes to estimate the "abundance" of the gold & hematite veins ain't it. Leave that for the professional explorers & prospectors.

Since armies and things now actually exist in the world, and caravans and other things should soon as well, this should be easy to implements. They just trace out thick lines as they move of civ known information. Caravans might have to get home safely to report it. Armies don't have to (they have implied scouts and supply lines that would carry info back continuously).

In the 18th & 19th centuries, before the advent of portable photography equipment, it was not all that uncommon for European armies on campaign to have at least a couple of cavalry officers who were trained amateur artists. If these scouts came upon an enemy position, they could quietly toss off a hasty sketch or watercolor, which, when brought back to the general, would be more informative than a purely verbal description. (Although I assume that the verbal part was given as well, with the scout giving notes like, "Yeah, that dark-gray blob is roughly 30 mortars-- 18- or 20-pounders, I think --arranged along the ridge.")
Except that that example of "war artists" is the most extreme example I can think of. When an army is being an army, it's just that--an army. They do army things, think army thoughts, and don't give a crap about anything that doesn't help the army in army ways. When they're huddled for cover to avoid goblin arrows or dragonfire, they honestly couldn't care less whether the rock they're hiding behind is selenite or satinspar. In short: Send a dwarven army to go slaughter the goblins and they'll do just that, while keeping a sharp eye out for goblins. NOT for aquifers, mineral outcroppings, what variety of trees those are, or the line where a Cold biome meets a Temperate one. That stuff all falls squarely into the category of Definitely Not My Job. So while an army (or whatever) could *explore* into uncharted territory, they would never reveal any more info than the embark window currently shows, and definitely not about anything beneath the surface, not even soil depth. And they wouldn't explore a whole "Region" tile, either, only as far as they could see . . . and we have no reason to assume that dwarven eyesight is all that keen.

2) The current level of info on embark represents realistically something like hearsay, and should extend all the way to any regions of the world visited by any civilizations that your civ trades with. Often, you might even trade detailed info from partners, if you're on good terms, equivalent to #1.

Your civ currently knows the name of every single river, all the way down to the smallest tributary, everywhere in the world. That's an awful lot of hearsay. Marco Polo only wishes he could exchange that level of detail with his civilization's trading partners.

At 5,000 years history or something, every civ should probably know every corner of even the largest maps (look at the influence ranges in legends politics maps)

That's a valid point--the age of a civilization / world should be taken into account when calculating data accumulation . . . although it also carries the risk that some of that data may have changed.

But yeah, for the most part, huge swaths of the map, in a realism sense, would/should have DF-prospect-level detail to them, at the LEAST. Maybe more than that within a healthy distance of the civ's own settlements (as they start digging they learn even about deep undergroudn stuff) and those of close allies.

I just installed DFHack, and I have to say, OMG NO. To know the exact mineral & vegetable composition of a site, before you even set foot there, and you have that knowledge for "huge swaths of the map"? And that's the LEAST amount of detail that you think should be shown? So much for painting me as the extremist, for wanting to introduce things like ambiguity & margins of error.

Honestly, we're talking about the same creatures who use quantum physics for levers and pressure plates. I'm sure they can figure out what kind of metals are under the ground. Maybe their technology can't analyse metals but can tell them apart if they are close together, so you can only know how many different types there are and not what those types are. Seeing as how their race relies on mining, mining tech should be fairly high priority.

Humans are the ones who can't identify stones, not dwarves. Dwarf kids have to do SOMETHING with their first 12 years, I expect that they do geology like we do elementary school, to the extent that a dwarf who could only tell rocks apart when they were next to each other would quickly be offered a ride in the "special minecart". With that said, however, it's very difficult to identify something that you can't see, hence the various calls to things like gods / divining rods / beard powers, to try to explain the Embark window's mysterious knowledge of "Deep metals". As for the well-known (and well-mocked) ability of dwarves to "link" a lever to a drawbridge through 500 feet of solid bedrock, magma, and completely empty space, with no physical connection whatsoever . . . trying to rhetorically reduce a game to being just as good as one of its least realistic elements is not generally considered to be a good idea.

I do love the idea of being given mandates by the king.  I always felt that lack of a goal... sure the outpost is to "produce wealth" but why would the kingdom want to bankroll your expedition to that evil glacier (lacking metals but infested with goblins) on the other side of the world? I usually like to think of my expeditions as procurement centers that will enrich my empire with exotic new goods... but again, you must choose a civ before you can view what animals/materials are available to them.

Yeah, having the outpost liaison relay actual commands from the Mountainhome has always seemed to be a good way to break up the standard routine. Embarking with a definite purpose from Day 1 opens the door to more royal imperatives like "As you are our nation's copper mine, from now on, you must send us 25 bars of copper in tribute every year. If the caravan fails to reach you one year, next year's demand will be doubled." Or, "The Queen is pleased with the number of trolls you have slain this year, and has graciously elevated you to the rank of Count." Or, "The elves have insulted us with another 'restriction' on the number of trees they will 'allow' us to cut. Here is the writ commanding you to release 5 of your prisoners into our custody--they must all be skilled wood cutters, and the more badly behaved, the better. We'll do the rest." Or even, "This year, you allowed a total of 40 goblins to reach our ally's civilian settlements. Prepare for your flogging."

[GavJ]

As for Legends mode, I don't even see why you consider it a relevant comparison to the Embark window. Sure, it shows the whole world, but only at the most basic level: Forest, Glacier, Mountain, River, Desert, Plains, City, Sea. It doesn't reveal temperature, aquifers, water salinity, or savagery. It DAMN sure doesn't give any information on soil or minerals.

http://dwarffortresswiki.org/index.php/v0.34:Legends#Export_Map.2FGen_Information.
  You can not only get this info, but can export it in convenient split-out maps for everything.
Elevation has even been there since 0.28.

Well, in REAL real life, events happen. So if you're going to play the "Everyone's a Prospector" card, I might as well play the "How Long Ago?" card, which reminds you that one of your armies passed through here once . . . but someone's built a town here since then.

Yes of course you only store data as of when it was most recently observed. Not that hard to do. It doesn't even require more data.  We already store history, so all you need is a date for each location of when it was observed and can extrapolate as needed, for essentially no additional disk space.

Dwarves have casual photographic memory, now? If this "state official" was part of the very first foray into this previously uncharted landscape, then yes, I might see him taking at least an hour out of every day of the journey to jot down notes & drawings like this. But in any other circumstance, no. He's got a job to do, and stopping every 5 minutes to estimate the "abundance" of the gold & hematite veins ain't it. Leave that for the professional explorers & prospectors.

"Diary" =/= photographic memory, so no, they don't and nobody was saying they did.
And generally, you stop to make camp an hour or two before sunset. They have 14 hours not on the move. Do you sleep for 14 hours? Neither do I. It's common sense to take notes during your significant amounts of free time during which you can't do anything else, for the significant benefit it brings to the realm. Relevant training would also be provided by any monarch or overseer with a head on their shoulders.

Depending on who it is, merchants for instance likely can just ride in the wagons most of the day and have even more time to observe and take notes.

while keeping a sharp eye out for goblins. NOT for aquifers, mineral outcroppings, what variety of trees those are, or the line where a Cold biome meets a Temperate one. That stuff all falls squarely into the category of Definitely Not My Job.

I've personally read several geological journals of army officers for my home region in real life, while I was attempting to locate a source of fire clay for personal art projects. This is simply wrong, and wasteful thinking. The mountainhomes (and countries in real life) need constant information AND goblin stabbing. If you're sending out hundreds of people to stab goblins, you send a few who know how to take down data about other things you need too. It would be frankly stupid not to. Especially when half the geographical stuff is tactical anyway (plants, watercourses, elevation, climate)

This is not conjecture. It happens/happened all the time.

I just installed DFHack, and I have to say, OMG NO. To know the exact mineral & vegetable composition of a site, before you even set foot there, and you have that knowledge for "huge swaths of the map"? And that's the LEAST amount of detail that you think should be shown? So much for painting me as the extremist, for wanting to introduce things like ambiguity & margins of error.

I think you might be confusing /prospect with /prospect all.
"Prospect all" is ground penetrating magic radar. "Prospect" merely gives you the composition visible from the surface, which, like I said "within a healthy distance of active settlements" is entirely reasonable. Shepherds and patrols and clothes washers and blah blah everybody and their mother is trampling past areas right next to settlements all the time. They'd know it like the back of their hand. People in the 14th century aren't sitting around playing computer games. They're outdoors memorizing their lands every day in the course of daily business. This doesn't even require diaries, this would just be common local knowledge.

More specifically, maybe 1-2 world tile radius from any current settlements.

[SixOfSpades]

As for Legends mode, I don't even see why you consider it a relevant comparison to the Embark window. Sure, it shows the whole world, but only at the most basic level: Forest, Glacier, Mountain, River, Desert, Plains, City, Sea. It doesn't reveal temperature, aquifers, water salinity, or savagery.

http://dwarffortresswiki.org/index.php/v0.34:Legends#Export_Map.2FGen_Information.
  You can not only get this info, but can export it in convenient split-out maps for everything.

I honestly did not know that. So I'm wrong about Legends mode containing that information, but I'm not wrong about it being effectively useless for picking an embark location. Export all the XML maps you want, find some way to overlay them together, you'll still find it nearly incomprehensible. Only the Biome & Site map is any real help at all, and it's no better than the world map inside Legends mode itself.

DF-prospect-level info is basically just a diary entry, which any state official would keep as they trudged along

"Diary" =/= photographic memory, so no, they don't and nobody was saying they did.

I think you might be confusing /prospect with /prospect all.

Nope, I mean /prospect. Let's consider a merchant or state official who has literally nothing to do but ride in the wagon and look at the landscape all day, except for when they're stopped for the evening and he can stretch his legs & poke around a bit more. Let's also say this guy is a top-notch dwarven geologist, and with eyesight as sharp as any human's. In that case, a realistic diary entry really would be something like "There are a lot of plants here, and I sketched out the main species. The rock is of this type. We noticed a couple of veins of gold and hematite in the bluffs of such and such abundance."

It would NOT be something like this:

Code: [Select]

Liquids:
WATER: 389 Z: 148
MAGMA: 49593 Z: 128..188

Layer materials:
ANDESITE: 11501 Z: 161..218
SCHIST: 8256 Z: 149..214
SILT_LOAM: 3682 Z: 148..174
GRANITE: 1466 Z:151..201
OBSIDIAN: 1443 Z: 151..196
DIORITE: 202 Z: 149..189
>>> TOTAL = 26550

Ores:
TETRAHEDRITE: 478 Z: 152..213
SPHALERITE: 432 Z: 154..214
CASSITERITE: 140 Z: 149..201
>>> TOTAL = 1050

Gems:
SMOKY_QUARTZ: 79 Z:157..214
PINEAPPLE OPAL: 73 Z: 163..218
Etc., etc., you get the idea. Prospect (not even prospect all) looks a lot more like photographic memory (plus clairvoyance) than any diary entry that I've ever seen.

I've personally read several geological journals of army officers for my home region in real life . . . If you're sending out hundreds of people to stab goblins, you send a few who know how to take down data about other things you need too.

Okay, alongside the "war artists", let's have "war geologists", too, and since these are Dwarves we're talking about, the number of war geologists would rise considerably. But most of these would either be too busy fighting/drinking to care about rocks, and/or be dead before being able to pass their information on. So let's just consider those dwarves in non-combat roles, especially those who don't need thrills or risks in life. These dwarves would have the opportunity to jot down notes on sources of fire clay.
When Toady adds more detail to the movements of armies, especially during actual battles, it would be nice to see the Dwarven Corps of Engineers whip out their entrenching tools and dig bunkers just outside of enemy ballista range. That would provide some real firsthand knowledge of the local strata (not to mention the freedom to tunnel straight into the enemy's base), and some security in which to write it down & send it homeward.

Especially when half the geographical stuff is tactical anyway (plants, watercourses, elevation, climate)

I'll buy dwarves (particularly soldiers on an active campaign) caring about elevation, but the rest of that stuff is for elves. Dwarves have always struck me as being far more about grrrahh stab blood beer than giving a flying fart about watercourses. Climate too cold for you? Start a forest fire!

"Prospect" merely gives you the composition visible from the surface, which, like I said "within a healthy distance of active settlements" is entirely reasonable.

If you "prospect" on a perfectly flat embark site, then yes. But pick a hillside or mountain, and you get magical ground-penetrating radar, at least in the x-y plane. Sure, in my example above, I picked a volcano to exaggerate the effect, but then volcano embarks are quite popular, and a lot of those would (I expect) include candy in the prospect screen as well, if it happened to be visible at the bottom of the magma tube. Spoiler metal isn't spoilery if you know about it before you even arrive onsite.

Shepherds and patrols and clothes washers and blah blah everybody and their mother is trampling past areas right next to settlements all the time. They'd know it like the back of their hand.

True, villagers would indeed have perfect knowledge of the local terrain. But how often does that knowledge make it all the way back to the capital, losing no detail in the process? Still, you're right: The Mountainhome's topographical maps aren't perfect, but they're pretty damn good, and when the embark party nears the site, they can just hire a local guide or two--a combination which would most feasibly be translated as having the Embark window show 100% correct detail, very near your or allied settlements. But what about a sandy desert, with its dunes being constantly shifted by the winds? Or a flowing glacier, different year by year? Or a trackless bog, or featureless moor? No settlement could reliably know these lands like the back of their hand, no matter how close they are. An Arab guide can take you between the oases, sure, and that's literally lifesaving knowledge, but apart from that, the majority of the landscape is just a blank, whether in the locals' minds, on a paper map, or in an Embark window.

[Dirst]

"Prospect" merely gives you the composition visible from the surface, which, like I said "within a healthy distance of active settlements" is entirely reasonable.

If you "prospect" on a perfectly flat embark site, then yes. But pick a hillside or mountain, and you get magical ground-penetrating radar, at least in the x-y plane. Sure, in my example above, I picked a volcano to exaggerate the effect, but then volcano embarks are quite popular, and a lot of those would (I expect) include candy in the prospect screen as well, if it happened to be visible at the bottom of the magma tube. Spoiler metal isn't spoilery if you know about it before you even arrive onsite.

Remember that prospect is a utility someone else made.  Toady could easily decide that the embark screen gives info on all "light" tiles only, or better yet reasonable forecasts based on knowledge of the actual engine used to generate the tactical tiles from embark tiles.

Given the importance of geology to Dwarves, some limited knowledge of the underground seems reasonable at least for "medium-level detail" surveys.  My first inclination was to list the layers, with what-you-can-expect-to-find summaries generated from the raws (easier than it sounds because the embark site really is generated from those raws already).  Maybe the actual layer stones if we want to be generous, again with the summaries.

It's important to note that these summaries are generated from the "potential site" not the "actual site."  For example it tells you the likelihood of emeralds based on the likelihood of specific layers times the frequency of emeralds in those layers.

[GavJ]

I forgot that DF prospect mentions layers. Ditch that part, definitely. But the rough amounts of each thing at surface are fine, minus layer info. I see nothing wrong with "80 schist, 30% andesite" etc. 5-10% precision within each category is fine. Probably would be easier to read and more useful at a glance anyway than prospect individual tile counts.

Rocks don't really need absolute numbers anyway, since you know how big your embark site is (%s tell you just as much). Plants do need absolute magnitude of some sort. But you can do "About 3,000 plants, 10% whipvine, 15% rope reed, blah blah"

Dwarves have always struck me as being far more about grrrahh stab blood beer than giving a flying fart about watercourses. Climate too cold for you? Start a forest fire!

That is a playstyle. Nothing about the hardcoding suggests this sort of thing. Just because being careful and responsible isn't as much FUN for players to do (so they don't) doesn't mean the game should ascribe irresponsibility to dwarves as a matter of course.

If you "prospect" on a perfectly flat embark site, then yes. But pick a hillside or mountain, and you get magical ground-penetrating radar

I'm not sure what you're talking about. It has been my experience that it seems to actually do surface visible tiles only. If that is incorrect, then okay, that's just a bug: the game's knowledge engine SHOULD do surface visible tiles only... in which case it being a hill doesn't matter. (and if so, dfhack should fix that as well, cause that would be dumb!)

True, villagers would indeed have perfect knowledge of the local terrain. But how often does that knowledge make it all the way back to the capital, losing no detail in the process?

The level of detail I'm talking about isn't a perfect map. It's just 75% andesite, 20% schist, etc.

Never meant to imply that they'd be giving you a detailed topo map with every little dusty defile in it and where individual village walls are built down to the tile, and so on.

Shifting dunes and glaciers don't really change the ratios of rocks or quantity of overall vegetation in an embark site.

[Antsan]

Just wanna say that I like SixOfSpades' suggestion.

[SixOfSpades]

Remember that prospect is a utility someone else made.

Oh, definitely. I'm fully aware that DFHack is 3rd-party software, otherwise I'd have downloaded it ages ago.

Toady could easily decide that the embark screen gives info on all "light" tiles only

It has been my experience that it seems to actually do surface visible tiles only.

Hold the phone! You guys are right. I took a closer look at one of the sites I /prospect-ed on, and actually counted the gems that were exposed to open air--they matched prospect's total exactly. The wealth of precise information it spat out fooled me into thinking that it gave the full composition of every z-level that was exposed in any way, but now I see it actually just counts the tiles that are considered potentially "visible" to dwarven sight--whether those tiles are right next to a road, or at the bottom of a volcano.

. . . or better yet reasonable forecasts based on knowledge of the actual engine used to generate the tactical tiles from embark tiles. . . . with what-you-can-expect-to-find summaries generated from the raws (easier than it sounds because the embark site really is generated from those raws already). . . . It's important to note that these summaries are generated from the "potential site" not the "actual site."  For example it tells you the likelihood of emeralds based on the likelihood of specific layers times the frequency of emeralds in those layers.

This. This is pretty much exactly what I wanted. But one question: You're saying that until the embark is actually finalized, the program only knows the % probability of, for instance, more than one shallow ore? That seems at odds with the embark window itself, which flat-out guarantees multiple shallow ores.

Given the importance of geology to Dwarves, some limited knowledge of the underground seems reasonable at least for "medium-level detail" surveys.  My first inclination was to list the layers . . . Maybe the actual layer stones if we want to be generous

But the rough amounts of each thing at surface are fine, minus layer info. I see nothing wrong with "80 schist, 30% andesite" etc. 5-10% precision within each category is fine.

Okay, so, how about something like the following:

Full detail = The "heartland" of your civilization, within 3 or 4 days' walking distance of major & medium-sized settlements.
High detail = All of your civ's domain that is not already covered under "Full" (small towns & the like), plus the "Full" area of allied civs.
Medium detail = Everywhere within 3 days' walk outside of your civ's influence range that is not listed above, plus the "High" area of allied civs.
Low detail = All lands more than 3 days' distance from your civ's borders, as well as the "Medium" areas of allied civs, except areas remote enough to be No detail.
No detail = Every place at greater distance from your Mountainhome than 3 days' travel for every year since worldgen, unless already covered by Full / High / Medium.
These distance measurements are just estimates--I have no idea how many World Map tiles "3 days' walking distance" might be, but that number feels realistic to me. Feel free to suggest changes to the numbers as you wish. Traveling armies and merchants of your or allied civs (as well as settlers / state envoys, if the game tracks such things) generate "traffic," which fades over time. Areas with sufficient traffic get bumped up from 1 level of detail to the next. For purposes of calculating site familiarity, underground roads generate 3x the "traffic" of surface roads. Similarly, areas with sufficiently fluid or inhospitable conditions, and/or hostile inhabitants, get kicked down 1 level of detail. If you accept your monarch's mission to settle at a specific area, that area is treated as already explored (to some extent) for you. For example, the area surrounding one of the villages inside a neighboring allied civilization would normally be shown to you at Medium detail--but if your King wants you to build a fort there to defend their land as a sign of cooperation (and you accept), your level of detail when examining the preferred region goes up to High.

Surface = Just what it says, everything that's exposed to open air. Just like DF Prospect, but treats magma as opaque.
Shallow = All tiles reachable by drilling 20 z-levels straight down from the surface. Approximates taking core samples.
Deep = All tiles more than 20 z-levels below ground.

• Full detail: All Surface features, both mineral & vegetable, displayed in exact counts. Local and migrating creatures shown in percentages showing what amount of an average year they tend to be present (and, if relevant, which biome(s) they prefer). Lairs are clearly shown, with creature type & name.
  Shallow features calculated to 3 significant figures, with 90% accuracy, then displayed as a percentage breakdown of total rock composition of Shallow layers (e.g., 43% gabbro, 31% andesite, 12% schist, etc.). Presence of caverns shown if detected in Shallow ranges. Shallow cavern flora (if expected) calculated to 4 sig figs at 90% accuracy, fauna at 80%, then displayed as discrete counts (not percentages). States exact depth of clay/sand/soil, gives the range if it varies (within each biome). States if magma expected in Shallow layers (90% accuracy, disregard if the only Shallow magma also reaches surface). All Shallow aquifers are detected.
  Deep features calculated to 2 significant figures with 70% accuracy, displayed as percentage composition of Deep layers. Deep cavern plants & animals calculated to 3 sig figs & reported with 70/60% accuracy, respectively. 90& chance of detecting any Deep aquifers.
  
  
• High detail: Just like Full, except for these changes. Surface counts now rounded to 4 sig figs & at 95% accuracy, with rock composition (but not gems) broken up into percentages. Animal frequencies are listed at 90% accuracy.
  Shallow findings calculated at 2 sig figs at 80% accuracy, predicted Shallow cavern plants & animals calculated to 3 sig figs at 80 / 70% accuracy. Soil depth now back to Little / Some / Deep, etc. 90% chance of successfully predicting aquifers.
  Deep composition now 60% accurate, at 1 significant figure. Cavern flora & fauna now 60/50% accurate. Chance of correctly spotting Deep auifers 75%.
  
  
• Medium detail: All Surface rocks & gems, as well as trees & other plants, are now calculated to 2 sig figs and shown as percentages, with 80% accuracy. Animal populations are shown at 75% accuracy. Unless it is your royally-mandated objective, Lairs are not shown pre-embark, and show only the creature type, not name. The embark screen does not show the names of brooks or streams, unless they flow to / from areas of higher detail.
  Shallow rock / gem composition now at 1 sig fig & 70% accuracy, chance of detecting aquifers / caverns / magma in Shallow zones reduced to 75%. Cavern plants & animals calculated at 1 sig fig & 70/60% accuracy, expressed as a percentage.
  No longer detects precise rock/ores/gems in Deep zones: Now states nothing but what types of layers are expected, with 50% accuracy. Deep aquifer predictions also at 50% accuracy. Deep cavern plant & animal predictions now 30/25% accurate.
  
  
• Low detail: Surface details are calculated at 1 sig fig & expressed as a 60% accurate percentage, animal populations are 50% accurate. Clay/sand/soil is reduced to Yes/No. Small and medium rivers are not named, unless they flow to / from your territory, and neither are volcanoes.
  Shallow composition no longer names actual rocks, only a 60% accurate prediction of layer types. Shallow aquifer detections 50% accurate, no Shallow magma or cavern detection at all.
  The only Deep data is a 30% accurate guess at the layer types.
  
  
• No detail: The Local view in the Embark window is not shown at all, the game only reports limited data about the general region. Shows Biome type, Temperature, the general concentration of trees & shrubs, the Surroundings (Wilderness / Terrifying / etc.), and a No / Yes / Multiple on whether or not there are surface ores. If you embark here, the game will randomly pick a site in the chosen region--if it lands on a site with no ore, it will try 1 additional time and then stop.

To be sure, this setup has its problems. For example, consider a perfectly sheer 30-z cliff. Are the tiles near the base of this cliff "Surface", because they're exposed to sight, or "Deep", because they're more than 20z 'below ground'?
What about caves? Should their entire interior be treated as exposed and therefore "Surface", and can the theoretical explorer dwarves go to the very bottom and drill their core samples, thus making it "Shallow" no matter how deep they might be? Should the explorers avoid the caves entirely for fear of the nasties there, and if so, why don't they avoid regular surface nasties? There is still much to hash out.

About "Roleplay Mode" (proximity to Mountainhome strongly encouraged, you might have a goal) vs. "Fortress Mode Classic" (embark wherever you want, your civ is irrelevant):
I can see both the "build a castle here for this reason" and the exile / fugitive routes as possibly being accessible from the same menu at the same time, as in, "Lead this expedition for me, OR I'll have you and all your friends thrown out of the kingdom." But I can't justify the monarch saying, "Do this for me, OR you'll be banished, OR you can just go wherever you like." Then again, it looks that that's exactly what the monarch will say, because having the game ask 1 question that breaks the 4th wall is a whole lot easier than creating two completely separate, yet only very slightly different, flavors of Fortress mode.

[GavJ]

I like those categories just fine.

However, I would request that knowledge of geography wouldn't be likely to be forgotten within a couple years just because nobody has walked there recently. That's fairly silly. The traffic thing works fine for building up gradual detail, but it shouldn't go back down again. Or if it does, make it VERY skewed. Like... maybe 2-3 trips might build a level, whereas losing one might take 100 years or something.

That's not to say it should magically keep up to date with new buildings or mining, though. A date stamp can go with each region, and when necessary (when you select it as an embark option), the game can go extrapolate backward from history data to figure out how much your civ would know about it having last seen it at that date. Almost no extra memory requirement, but realistic feeling geographical recordkeeping. Also applies to no longer being allied with another civ. Info about them stops being updated at that point but still stays at old time stamp.