Print

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

[OcelotTango]

Ok, so I was in Chemistry and we were talking about various forms of chromatography, when I realized that it might be possible to do something similar in DF.
While not entirely the same my goal was to separate objects with different weight in DF by how far flowing water would push the objects.
Here is what my set up looked like http://i.imgur.com/YIjMEqG.png

While not entirely successful it is worth mentioning how my tests went.
Before the test  http://i.imgur.com/28BvmMW.png
After a long time  http://i.imgur.com/GrQVOg6.png

I've concluded that the only movement generated by the item was from the initial flow of water, similar to my proof of concept tests http://i.imgur.com/jhh12kM.png

Water does move items, and it does move them depending on the weight, but as of now it's too hard to track them, which makes testing and developing for this dwarf science not worth the effort. As of now, unless anybody has any compelling solutions to the muddy tile item bug, or a way to generate a level steam of moving water, I'm going to abandon this ≈SCIENCE≈.

Edit1:(Fixed the title)
Edit2:(Added data)

[Button]

1) Forbid the objects before you start the separation (obviously this will throw a wrench in the eventual automation of separation, but it's good enough for initial ~~science~~!)

2) One of DFHack's tweaks will fix items that go missing when pushed by water. It has to be used after the fact, however - it doesn't fix the bug itself, just brings back items that went missing.

Watching with interest. There's a longstanding question of whether metal studding on weapons adds to their weight, and if you can fine-tune the Dwarven Centrif- er... Linearfuge? this would be a perfect way of testing this.

[Crashmaster]

How do you figure item weight is taken into consideration during flow-pushing? To me it seems more like the distance an object moves is more related to the number of times water-height changes over the tile the item is in. Of course I have done no !!SCIENCE!! on this.

[Em3rgency]

Sorry to rain on your parade, but water flow to push items is notoriously unreliable. Just repeat the experiment with the same items a few dozen times and you'll see how much the results vary. And as Crashmaster said, you can't assume weight is even taken into consideration.

[Urist_McGamer]

I love how much actual science DF encourages. Seeing if weight matters when items are pushed by water will be interesting. If that doesn't work, maybe a minecart shotgun would. You'd be able to put so many items in at once that you could fire it once and obtain all the results you'd ever need.

[Hurkyl]

Sorry to rain on your parade, but water flow to push items is notoriously unreliable. Just repeat the experiment with the same items a few dozen times and you'll see how much the results vary. And as Crashmaster said, you can't assume weight is even taken into consideration.

Thus, the desire to do Science to test the hypothesis....

[chevil]

If this works then I can go all !!science!! on my drowning traps.
In order to get goblinite to point X I need Y amount of water to drain in Z tics.

[slothen]

OP's setup is probably not adequate.  You'll need a setup that ensures a steady flow, or even better, uses only a fixed quantity of water so that successive trials are as similar as possible.  To test if weight is even a factor you would need to have two items where one is known to be much heavier "race" each other multiple times, such that after say 20 trials you can be sure that the winner isn't a completely random variable.  Then you'd need to look at the raws to isolate other variables like item size and density.  The object testing arena is probably the best place for this.  From my own experience, water flow, particularly as it pushes creatures and items, is so random/unpredictable that its a terrible method of measuring anything.  Verifying that it can be a useful metric would require a working knowledge of statistics, and everything would be apparatus-specific.  I suppose the holy grail would be some kind of ultimate-slip-n-slide that is long enough and accurate enough that one could guess the actual weight of an item within 10% with say a 80% confidence interval based on only one trial.

[Trev_lite]

for a first test i would recommend that you test 2 items of the same size (like 2 metal weapons of different metals) and repeat the experiment many times so that you can average out the randomness. the items being the same size so is important since size might make a difference and you want to test mass. two metal items of differing metals is what i recommend. try the lightest metal (excluding candy, its too light and might float) and the heaviest. gold and aluminum are my recommendations. (lead isn't all the dense, its only cheap)

a back up idea is using to woods of different densities.

hope to see the results.

also does the item being less dense than water make a difference too?

[Centigrade]

Clicked the thread hoping to see some fire-related chromatography as advertised.

Turns out it was just someone misusing or misunderstanding what is implied by surrounding a word with !!.

[Noodz]

Sphalerite's experiment where he shot whales with ballistas showed larger creatures are harder to push with water than smaller creatures. I don't know how that applies to object and/or if weight is an issue.

[i2amroy]

Clicked the thread hoping to see some fire-related chromatography as advertised.

Turns out it was just someone misusing or misunderstanding what is implied by surrounding a word with !!.

!!SCIENCE!! refers to all true dwarven science, whether it includes fire or not (though as is the way of things, most dwarven science does). It's a generalization to differentiate normal old SCIENCE from true dwarven SCIENCE.

[Namfuak]

I love how much actual science DF encourages. Seeing if weight matters when items are pushed by water will be interesting. If that doesn't work, maybe a minecart shotgun would. You'd be able to put so many items in at once that you could fire it once and obtain all the results you'd ever need.

Dwarf Fortress !!SCIENCE!!  We do what we must because we can.

[OcelotTango]

Ok, so I'm very busy with exams at the moment and didn't expect such a prompt reply. I agree that the object testing area is indeed the best place for this. My hope is that over a very long distance the randomness of water pushing will average out, and objects of similar weight will tend to stick together. If anybody knows a way to generate a continuous stream of moving 4/7 through out a long hall way, that would be awesome. Otherwise ill do some initial tests with the inferior set up as a proof of concept.

[Centigrade]

I love how much actual science DF encourages. Seeing if weight matters when items are pushed by water will be interesting. If that doesn't work, maybe a minecart shotgun would. You'd be able to put so many items in at once that you could fire it once and obtain all the results you'd ever need.

Dwarf Fortress !!SCIENCE!!  We do what we must because we can.

For the good of all of us, except the dwarfs who are dead (all of them).