Print

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

[GreatWyrmGold]

In my opinion, we should always err on the side on not caving in. If it makes you feel better, imagine that the dwarves are shaving off some stone here and there to make it lighter.

Mir: I agree. Soil happens to currently have the same values as stone, though.

[Jeoshua]

I agree thirdly.

But yeah go look at materials_template_default.txt

SOIL = STONE

[NW_Kohaku]

Well, that's presumably because Toady has never really cared, and neither have any of the players who look these sorts of things up.

If we start seriously rating things in terms of the shear stress of each type of stone as a building material, then it starts getting pretty complicated.

Stone can take compaction pretty well, but it doesn't handle shear stress very well at all.  Stones like chalk or sandstone, especially, would be liable to snap, and the concept of making a bridge out of sand or silt is just laughable.   

[Jeoshua]

I don't know if "didn't care" is really the reason.  There's just never been any reason for soil to be anything but a Stone.  If you mine it, it doesn't produce a boulder, so you can't build anything out of it.

But if cave-ins go in, and their logic used the material properties of soil... it's really not too much work to modify those variables to make soil "caverns" less likely.

The same goes for more specific values for different building materials.  Little by little the actual material properties would be put into the game, either by modders, by Toady, or first by modders and then incorporated officially into the game by Toady.

And "complicated" is kind of DF's middle name.  Just about says so in your siggy NWK

[King Mir]

If soil is made to be as bad for digging rooms in as it really is, then one of the first additions needed is a support Constructions, so that you can build workshops with supports going through them.

The ability to reinforce dirt walls with wood or stone would be useful too, by essentially building a stronger wall in place of or as a decoration to a dirt wall.

And of course it would make aquifers that much more trouble.

[Jeoshua]

Supports exist already (Link), they're just pretty useless unless you're making a F*** The World lever.  I don't think I understand why you assert that you absolutely would need "support workshops".  Wouldn't it be good enough to just put supports and/or walls at the four corners?

And if you're creating anything really mission-critical in the dirt above your fortress... maybe you should invest in some stone walls or maybe a tower to cover it up, instead of just digging a hole in the dirt.

ARE WE ANTS? OR ARE WE DWARVES?!


All joking aside, support walls could be made now, without any additional construction types, by simultaneously mining out a strip of wall and replacing it with constructed walls of stone.  If the hallway is already stone, it could be done with walls of metal (but really wouldn't need to be done at all, since Stone should be a good solid material by the looks of it's properties)

[King Mir]

Supports are buildings, not constructions. They should really be both, so that you can have lever triggered ones, and supports under workshops. Depending on the area of a tile, it may not be realistic to have unsupported 3x3 or 5x5 squares of dirt ceiling. Dirt is much much worse than stone for digging large rooms in. A large dugout should have a good amount wooden or stone columns and possibly bracing on the walls.

[Sadrice]

I'm not a programmer or an engineer, so maybe I'm missing something, but why would it need to recalculate the forces acting on all tiles above and below it?  Wouldn't it just need to move outwards until it found tiles unaffected by the change, then stop?  That way most mining operations would only check a few tiles, and there would only be lag if you were removing support from a large, complex, poorly supported structure.

[Jeoshua]

That would go under "optimizations".  My example was a worst case, brute force scenario, using more calculations and being altogether more pessimistic about the ramifications of those calculations than needed to be.  And it still only came out to a fraction of a second to check most of the map in that style.  It's not something you'd want to happen every time you mined a tile or moved anything around, but it's also not a definite FPS killer.

Optimizations like keeping the last mechanical stress number and comparing it to the new would increase the memory footprint of the system, but would seriously speed up its execution.  That's a good point too, and something that should be really considered.

If storing one number per tile and comparing it to the last valid result, as that would, could potentially save a lot of calculation, as that would, then it should be considered.

However, it doesn't really change the amount of calculations needed to be performed on a Cascade-style system.  If the weight support calculations "flow outward" from a spot, any time a tile's support environment is changed those changes in force will need to flow outward.  If the system is coupled with the reductions in transferered weight that I outlined, at a certain point that "flow" will stop naturally.  This basically models the same situation you're talking about, but would be a bit more abstracted and a touch less computationally intensive.

[NW_Kohaku]

I'm not a programmer or an engineer, so maybe I'm missing something, but why would it need to recalculate the forces acting on all tiles above and below it?  Wouldn't it just need to move outwards until it found tiles unaffected by the change, then stop?  That way most mining operations would only check a few tiles, and there would only be lag if you were removing support from a large, complex, poorly supported structure.

You need to check all tiles to the side, and then draw a downward-pointing cone from that point upwards to check all the tiles that could be impacted.  From there, you start at the top, and then check weights going downward if we are looking for "crushing force" in a "cascade" support check.  (Which is one where all tiles have to have a load-bearing wall capable of bearing the weight of all the tiles above them, or face collapse.)

If we only check for shear forces and making ceilings or ledges collapse, then we can check the much narrower area of the ceiling of the tile we just excavated.

[GreatWyrmGold]

Alright, does anyone mind if I take a moment to have you guys ask yourselves a vital question:

Where should the threshold of caving in be placed? In other words, what kind of structures should cave in and which should be stable, and how long would it take the unstable ones to collapse? Say: a winding entrance dug in soil, three tiles wide with the path frequently crossing under earlier parts of itself; or a wall of soap; or a big tower with a barracks hanging off one side near the bottom; or a huge statue of stone on a platform of wood, with space to live in underneath.

[kaenneth]

I build pillars in my open areas for 'flavor' anyway, and I don't see the harm is adding a betting cave in system as long as "[CAVEINS:NO]" is an option.

And while real material properties would be interesting, I think a simple system would be best for both players and code.

[Silverionmox]

- Checks don't need to happen often (monthly, at a random date or when triggered by eg. an earthquake miracle by a god). Cave-ins are a long-term concern.
- This gives us a little more computational room to arrive at sensible results. Let's take into account some of the material properties of materials as they are defined in the raws anyway, the measures, and maybe even construction skills and culture influence.
- To make it sensible for the player, first interface improvements like the "Depth by darkness" thread are needed to visualize depth and spaces properly.
- We'll also need to display the pressures on a given map when the (d)ig command or similar is selected. This would just make diggable squares flash with a color corresponding to the pressures on it, eg. purple = almost cracking, green = can handle more. (At least for dwarves, if the correct noble/official is present. The other races don't have one and therefore will have more accidents). This will allow the player to see what he's doing, and to see whether he's doing well.

[Sadrice]

If checks don't happen often, there might be the possibility of something remaining totally unsupported hovering in the air for a month until the game realizes what's going on. 

[King Mir]

Alright, does anyone mind if I take a moment to have you guys ask yourselves a vital question:

Where should the threshold of caving in be placed? In other words, what kind of structures should cave in and which should be stable, and how long would it take the unstable ones to collapse? Say: a winding entrance dug in soil, three tiles wide with the path frequently crossing under earlier parts of itself; or a wall of soap; or a big tower with a barracks hanging off one side near the bottom; or a huge statue of stone on a platform of wood, with space to live in underneath.

In terms of how long, I think for gameplay it would be best for all unstable structures to collapse instantly, or be triggered by an earthquake or weather event. Everything should be either stable or unstable, though you may have several stable stages that approach unstable.

As to how wide and unsupported a structure can be, that should depend on the material properties. I don't know how high you can build with rock or how large a balcony. But it's all a combination of compressive and sheer forces, and in principle you could have both acting on a given tile.

Unfortunately for detailed suggestions, no one has actually proposed a workable solution. The original suggestion was proven deficient within twelve posts. There's been a lot of call to make it realistic, but not much detail on how to do this.