Bay 12 Games Forum

I've been giving some thought to the notion of how to evaluate the quality of rooms in the fortress and I'd like to share my current thoughts on the matter.  It seems like a more processor-intensive solution isn't a horrible idea since the value of a room changes very infrequently.

There are some basic ideas that I'm trying to incorporate into my method:
1) Big rooms tend to be higher-valued.  Or, at least, small claustrophobic rooms are lower-valued.
2) High-quality materials that go into a room reflect the overall quality of the room.
3) Interesting architectural designs should make for higher-quality rooms.
4) Emphasis should be placed on the effective and dramatic placement of furniture.

Then I got to thinking that maybe (1) is not really about the overall dimensions of the room.  A common theme in architectural design is that a space does not have to be big to feel big.  Peoples' perceptions of the space really change the way they appreciate it.  As such, I think value of a space should depend on the way it's actually perceived.  Essentially, when a room is created or re-sized, the game looks at the walkable squares in the room and considers what's visible from each location.  In doing so, the game can see what the big corridors of vision are in a room.

These corridors of vision are essentially line segments within the room itself.  The important part of those segments are the endpoints.  After the game identifies several critical corridors of vision (the longest segments would be preferentially more important), it labels the endpoints of these axes as critical locations.  Furniture placed in a critical location is scrutinized much more by the valuation process than furniture placed elsewhere.  The process would then look at each piece of critical furniture.  Starting with the actual monetary value of the item, this value can be enhanced by considering the visibility of the item.  Open spaces around the item are good.  Materials that contrast with those of other things in surrounding spaces would be a boost.  In this case, special cases can be defined.  A statue would greatly benefit from its surrounding tiles being water in a reflecting pool.

I think every item should detract from the value of items in close proximity, with some exceptions.  Two statues placed side-by-side is design-wise a dangerous thing.  If the quality isn't closely matched and the materials complementary then it doesn't work.  The worse of them brings down the value of the better.  In DF, I think an item should suffer a penalty of a slight proportion of the difference in value between itself and its surrounding items.  This would include engravings.  However, some items are designed to be placed together, like tables and chairs.  These would not decrease value.

The ultimate effect here would be to make rooms with longer lanes that have interesting things at either end more valuable.  This method would be ideal for modeling the display garden of your artifacts, and having the artifacts stand alone in their own exhibit down the hallway where you can see it as you approach.  Rooms like these will start to become relatively more valuable than simply cramming a current room with engravings and hoping for the best.

  Well, my first possible creature doesn't want to appear in worlds, and my second has caused massive brushfires and asphyxiated.  Does either fire or heat damage fireimmune creatures?  Also, it would seem that strands of silk shot as webbing don't have the same properties as silk left as webbing when the map starts.

Ability to assign armor/weapons to dwarves, who then keep said armor and weapons in the appropriate furniture that they own.

The descriptive statues one is also really good.

Steam power could be added simply by having steam be pressurized or not.  Steam already exists in the game, it just needs that one addition.  Don't do turbines, but make waterwheels movable by steam flow also.  Then, you have steam as a source of power and/or a source of aesthetics.

Misters, when you get right down to it, are a grate plus pressurized water.  We have these now (they just have the slight engineering challenge of killing everyone).

Also, I'm all for the drive belts.  It's a nice, simple thing to add.

On the .999r=1 thing.  I always thought the confusion arose because 1/3 of one equals .333r.  So .333r+.333r+.333r should equal 1, but it actually equals .999r.  I'm not a mathematician, so I don't know if that means .999r=1 or what.

You have the gist of it.  1 = 1/3 + 1/3 + 1/3 = .333... + .333... + .333... = .999...

More precisely, you have .999... = the sum as n goes from 1 to infinity of 9*10^-n = 9* the sum as n goes from 1 to infinity of 10^-n.  That sum is a power series with a ratio of 1/10 and a starting point of 1/10.  It converges because 1/10 < 1, and so it converges to a/(1-r) = (1/10)/(1-1/10) = (1/10)/(9/10) = 1/9.
Returning to the original statement, .999... = 9 * (1/9) = 9/9 = 1.
It's kind of weird, but rational numbers have two ways of being written.  If it helps, you could think of writing 1 instead as writing 1.000...0001 with an infinite number of zeroes between the 1. and the 1.  So, rational numbers are essentially defined to be the limit from above and the limit from below of those two decimal expansions.

Back to on-topic, why not just have atheist dwarfs not have a Worships entry at all?

Can you specify your question please?

Are you looking to find the minimum temperature required to give you glaciers?  Or, do you know that temperature and you cannot implement it because it gives you infinite rejects?

If your problem is a rejection one, then set the game to log map rejects (it creates a file for these in the base game directory) then see what the actual error is that's preventing the map from working.  Otherwise, I believe there's a manual override that can be done in the actual world-generation process to bypass rejects of a certain type.

I'd recommend playing around with your world creation parameters, however, until you get a template that works.  It's nice to have these templates because then later you can go back to them to get a similar world.

Keep us updated on the problem.

I think we'll be up to spore 3: "hey you need to buy this expansion pack to get the real experience", before Df is 'done'

Of course, by that time, DF will be 'done' because it will be fully sentient and new versions will be made by the program itself, which it will refer to as its 'children'.

That would indeed be excellent, but unless things have changed, you cannot retire in a goblin town, friendly or not.  And, I've discovered that you CAN retire in an elven town, so I found a world where dwarves are at war with elves and am filling up the elven side with mercenaries.