Print

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

[Tenth Speed Writer]

Change water into booze in the raws.

You will never curse an aquifer again.

Oh, to hell with the adamantine!

[Greep]

Take some limestone, melt it in magma, make a fountain out of it.  Voila! The White Rock Beverages soda fountain!

[derekiv]

Molten Mercury Oceans in every tomb

[DarthCloakedDwarf]

Once magma has flow like water does, it may be possible to build magma waterwheels using nether-cap wood... from what I hear, the material has fixed_temp, and may be magma-safe, even though it is wood.

[Danjen]

Also, molten pitchblende generators. The closest DF will ever get to a nuclear reactor.

For now.

*ponders*

Oh god.

[PTTG??]

This has been suggested for almost as long as firearms, if not indeed longer.
Luckily, Toady One does intend to add arbitrary fluids, but they have a number of challenges. I wonder if he could make it be entirely contaminant-based so that new fluids can be added without restructuring the whole CA system.

[Tenth Speed Writer]

I would think it possible simply to treat each cell as having 7 of any given fluid of significant quantity. Any less than, say, 20 (arbitrary) units of uristian volume would be a contaminant; at 20, that contaminant becomes 1/7th of a greater uristian volume unit of liquid.

Liquids could have simple properties for solubility (completely soluble in x, completely insoluble in y, all unmarked would separate by density.)

It would involve a slight decrease in performance (each cell containing fluid also being associated with as many as 7 subentities), but it isn't too terribly difficult to imagine, and it would allow for such fun (perhaps Fun) models as convection currents and (semi-)realistic pressure.

[Tenth Speed Writer]

Hm...


The main entities would be the individual sevenths of fluid. The checks on each subcell would be dissolution/separation (is contacting soluble substance? has been an emulsion for [arbitrary] amount of time?), separation (if denser/lighter than z-1/7 or z+1/7, switch with z-1/7 or z+1/7), and flow, which could be modeled on a cell-by-cell basis as it is now, perhaps with a randomization of vertical order for subcells flowing sideways. If we're getting fancy, convection of temperature (perhaps using a simplified thermodynamic model) could be considered.

Toady knows his code better than any of us, though. It'd be his call whether this is all feasible or not.