Print

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

[misko27]

I must ask how the volunteers were made to fall off, for repeatability you understand. 40% death rate seems very low for 20 story fall, but they are small, so probably not the same rate as humans. The fact that you took casualties during the fall is a very interesting concept, perhaps we could make dwarves die while falling? Was there a ramp there? i am very interested in the circumstances here, the point of the experiment was not to kill before falling, but this just got very interesting. The lack of instant death pretty much proves the nerfing beyond doubt, but now this experiment can go in exciting new directions. Try getting a dwarf underneath a falling dwarf and see what happens.

[khearn]

I dug a 4x4 vertical shaft 30 z-levels tall. Retracting bridges are built at various levels. I thought I built the one used in this test at level 20, but it turned out to be level 19 instead. Mea culpa.

Test subjects are put into a military squad test group and assigned to a burrow waiting area that contains a room with 10 beds, a dining room, a booze stockpile, a prepared food stockpile, a barrel stockpile (because if a dwarf empties a booze barrel, he really wants to be able to put it away and you get a ton of cancellation spam if he can't get to a stockpile for it), and a refuse stockpile. This is to keep them from making friends with the research staff or test subjects in other test groups. It's bad to use test subjects with lots of friends in hazardous tests, unless all their friends are taking the same test at the same time.

When a test group has 10 members, I create a burrow on the bridge at the level from which I want them to begin their test, and schedule that squad test group to patrol that burrow. All other bridges are retracted, of course. I then waited until the entire squad is on the bridge, which involved deconstructing one bed when a squad member test subject decided it was time to sleep instead of test. Then I pulled the lever to retract the bridge and begin the test.

And yes, one of the dwarves did die during the fall. He's the one that didn't seem to hit another dwarf, so I'm guessing he hit the wall of the test chamber after falling 17 z-levels. There's a red smear there. But there's also a read smear up at level -6 where the two collided, so it doesn't necessarily prove that this one hit the wall. More testing will be needed.

Two of the dwarves also did land on other members of the test group. The landers just got stunned and walked away unscathed, while the landees took damage. This means that dropping troops onto invaders is a plausible tactic. Just make sure your drop troops don't miss.
 
Test group 2 is scheduled to do the full 30 z-level fall, as soon as I fill the test group.

[MarcAFK]

40% mortality rate seems ok to me, half of the survivors did so because they landed on someone, i would like to see the experements repeated with no dorf shaped airbags on the bottom, i'll assume it'll be more fatal.(though half the deaths seem to have been due to collision midair)
I'm actually happy with the rewritten collision physics, i admit it needs tweaking, but things are far more interesting now, falling is now survivable based on different variables, rather than how it was where damage was based on height and 100%  survivable if a kitten was directly below.
Having said that, yes things should be a little more lethal and kittens should always survive any fall period no exception damnit.

[Dude_Jebawe]

You need to put upright spears at the bottom, then.

[Blizzlord]

You also need to drop pickaxes and hammers from the ceiling. Maximum devastation.

[Viking]

well, Its still better then before. If you fell 2 stories IRL, would you require hospitalization? Now apply that to hardened dwarves.

Personally, one story would be very doable. At two stories the risk of a broken ankle or some sort of other injury even requiring hospitalization seems quite possible. This is especially true if I were to hit a rock floor with armour on. Just to be clear, by 2 stories I'm thinking two stories of falling space as in if you step off the third floor. I admit I don't have a lot of experience (or any) falling such heights.

All I can say is I hope this gets fixed.

[Dude_Jebawe]

A fall from 2 stories would only hurt you significantly if you tensed your muscles. Drunk people relax their muscles. Dwarves are always sauced. Do the math.

[UristMcDwarf]

I built a puppy dropper

i love DF.

[HiEv]

While damage from falling may have been lowered, damage from things falling on you has apparently increased greatly.

See bug #5945: Collision damage can be very exaggerated.

That bug details an example where rat remains fell down one z-level and caved in a dwarf's head.

When testing, damage from falling and hitting the ground should be separated from other collision damage.

In other words, I wouldn't include data from mid-air collisions, dwarves that were landed upon, or dwarves that landed upon others, since those are different situations than damage from simply from falling and hitting the ground.  If you do, you're not testing damage just from falling, but from other things as well, and jumbling all of that data together.

[Mudcrab]

That bug details an example where rat remains fell down one z-level and caved in a dwarf's head.

I love DF.

[khearn]

While damage from falling may have been lowered, damage from things falling on you has apparently increased greatly.

See bug #5945: Collision damage can be very exaggerated.

That bug details an example where rat remains fell down one z-level and caved in a dwarf's head.

When testing, damage from falling and hitting the ground should be separated from other collision damage.

In other words, I wouldn't include data from mid-air collisions, dwarves that were landed upon, or dwarves that landed upon others, since those are different situations than damage from simply from falling and hitting the ground.  If you do, you're not testing damage just from falling, but from other things as well, and jumbling all of that data together.

I agree. I'm about to drop test group 2 and this time I moved them into and out of the test chamber several times until I got everyone on different tiles, just to avoid the mid-air and landing collisions. I'm also going to keep notes on when each dwarf falls to each level to try and get some firm data on the differences I saw in falling rate.

Anyone know of an easy way to find out how heavy each dwarf is? Given that the baby fell fastest in test #1, I suspect lighter weight creatures are accelerating faster. But I need to know weights in order to test that hypothesis. It seems like dfhack ought to have something that tells me that.

[misko27]

Two of the dwarves also did land on other members of the test group. The landers just got stunned and walked away unscathed, while the landees took damage. This means that dropping troops onto invaders is a plausible tactic. Just make sure your drop troops don't miss.

Or just drop animals on them, more expendable if you miss. Or war dogs, so If they dont they can tear them apart. Possibilities are theoretically endless.

That bug details an example where rat remains fell down one z-level and caved in a dwarf's head.

Someone do testing on whather a larger animal landing does more damage, if so i'm dropping yaks. Does junk work? Can I drop war hammers on a head?

[ab9rf]

I dropped several jabberers and rutherers 15 z-levels earlier today.  Only one died immediately.  One had both legs broken but didn't die on its own and had to be killed by my military.  The rest all managed to walk (or crawled) away.  Some died on their own after crawling/walking some distance; others survived long enough to be killed.

In the past, a 15 z-level drop has been enough to cause nearly anything dropped to explode on impact, so something has definitely changed in this regard.

[Vherid]

It's the same reason a feather falls slower than an equal mass of lead

Just a little side thing I felt the need to point out:
That actually has nothing to do with mass, weight, or size, but entirely with friction. All objects fall at the same exact rates, regardless of anything, except for the friction that slows them down, which on the planet earth, is mostly air friction. That is why a feather will fall slower. It has been tested on the moon where there is no air friction, the feather fell at the same exact speed as a wrench or something else that was heavy they tested it with.

But I have to say this seems a little disappointing overall, I was really hoping to start making some death pits and such after reading a few things about them. People seem to be testing what used to kill people last time though, someone needs to test like a 100z level drop just to make sure that they will actually explode at some point, hopefully.

[imperium3]

A realistic death pit would include a nasty set of spikes/monsters at the bottom though, so it's still doable. Has anyone tested whether falling onto menacing spikes is just as deadly as it used to be?