Print

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

[Fleeting Frames]

Yet another moment of "biomes don't work that way (expect they do!)":

Revealing, probing and farm plotting my way towards tiny multi-biome embark, found an interesting situation:

Reveal shows different soil layers on tropical grassland and temperate grassland. Expected.

Probe shows that some of the same-looking areas have different biomes. Not unusual, but has potential.

Farm plots painting shows just different tropical grasslands next to each other. Disappointing, but understandable.

I guess that means 2x2 is sm-...

*triple-checks probe biome and region ids, directly on top of temperate grassland and tropical grassland farm plots*

So.

Apparently, you can have different biomes (temperate and tropical grassland, in this case) under same biome and region ids.

I mean, evilness, layer materials, temperature still shows them to be different.

But I thought at least one of those ids were unique.

[Evans]

werecursed invaders will not attack other invaders from their siege after transformation.

I hoped to spread the curse to my enemies but it appears it won't be that easy.

Edit:
Just learned that if I paint the floor with stone material in the area where trees grow, the bottom part of the trunks vanishes.

Upon saving and reloading all canopies that were hovering in the air collapse on the ground.

Easy tree cutting and no regrowth after all, but they seem to change ground from stone back to soil in the immediate collapse area.

[Fleeting Frames]

Adding even a single cave to even a PSV world with same seeds slightly changes river boundaries and majorly changes biome boundaries.

No kobolds for me, then.

[Larix]

I think i've found a close (or maybe even the exact) ratio the game uses to convert distances covered by minecarts on ramps: standard distance travelled *0,70707.

I worked out the baffling "lost" tiles when minecarts move on long ramped slopes (they sometimes move fewer ramp tiles than their speed suggests, e.g. spend two steps on a tile when going at 200 000+ speed, i.e. much faster than one tile/step, even on ramps): there's a "secondary checkpoint" effect where carts moving on ramps run a check every 100.000 normal distance and stop if they haven't moved to a new tile since the last check: long explanation here ; i had seen the effect earlier but hadn't understood it, so with some retroactive help from user fricy, who posted several data dumps, i could look up such an event in detail: open spoiler at own risk. Extracting the few ramp-check-braked shortened moves, we get these data:

Coordinate  speed    accel.      z    z-movement
26,72592   1,0816   0,0489   144   -1
27,43299   1,1305   0,0489   144   0
...
38,50771   1,6684   0,0489   132   -2
39,21478   1,7173   0,0489   132   0
......
55,78747   2,353   0,0489   115   -2
56,49454   2,4019   0,0489   115   0
...
59,99488   2,4997   0,0489   111   -2
61,40902   2,5486   0,0489   110   -1

(not quoting all of them, just three of the moves that went zero z-levels due to the effect, and one that only made one z-level)

Unfortunately, the dump doesn't explicitly contain the "distance travelled" field, so you'll have to subtract the co-ordinates to get the values. You'll see that the zero-z moves (stopped after 100.000 conventional distance) all have a distance travelled of 0,70707, and the one-z move (after 200.000 distance conventional) a distance travelled of 1,41414. I went and tested the value a bit on the high-speed turns in the dump that went the full distance and am reasonably sure that

on ramps, distance travelled is calculated normally, then multiplied by 0,70707.

(it could also be done by division, presumably by 1,41429, but that may be more effort for the CPU for pretty much the same result. Incidentally, a very human-friendly rendition of that divisor is 99/70, which at ~1,4143 is a reasonably good approximation of sqrt2.)

[Libash_Thunderhead]

Pets still follow the ghost of their owner.

[StruckDown]

Browsing legends and just found the most-pillaged site in my world:

In 497, early spring, The 639th Pillaging of Wavedent occurred as a result of The Assaults of Lancing in The Violent War waged by  The Kingdom of Eagles(Humans) on  The Superior Roar(Goblins).

The Violent War started in 200 and is still ongoing. It's 500.

Although the Goblins have killed nearly 3 times as many humans, they haven't won a single battle yet:

 The Kingdom of Eagles (Attacker)
◾Kills: 2416
◾Battle Victories: 2918
◾Conquerings: 2871 (2871 Pillagings, 0 Destructions, 0 Conquests)

 The Superior Roar (Defender)
◾Kills: 6796
◾Battle Victories: 0
◾Conquerings: 0 (0 Pillagings, 0 Destructions, 0 Conquests)

*upon further inspection, it seems that about 95% of wars, the defender will never win a battle. Even with the few that do win, they're never more than 2. How odd. Might be a bug with legends viewer

[mikekchar]

It might be.  Someone mentioned this a while ago in the adventurer section, so I've been scouring legends mode and I find that defenders in my worlds are edging out attackers, but still pretty even.  I run the latest vanilla, so either it is something to do with that, or potentially a bug in legends viewer as you say.

[TheFlame52]

I found a world where one goblin nation has been holding out against two human nations and an elven nation for over 200 years. The site where most of the battles take place has just over 60,000 deaths. More people have died in that one dark fortress than everywhere in the entire world combined.

There are still about a hundred goblins left. The only reason they survived so long is because they have ogres.

[Fleeting Frames]

You don't need a whole lot of herbalist skills to get mega-stacks out of trees.



All my herbalists are currently middling or so, no legendaries or masters -highest is adept, adequate or talented (not sure which of these is highest, wiki haz it tho )

Anyway.

"What if I brew that, will it split up the stack"



I guess no. Now, applications...

*considers doing a minecart system to automatically sort the biggest stacks into still*

*looks at weights*

Sadly, it appears all? fruits/spice are reeaaally light. None of the big stacks cross 1 urist - something that is accomplished by even a single sun berry, though the whole barrel (with many hundreds of spices/fruits) seems to clock out at around 30 urists (12 native weight).

Upon further looking into plant barrel, it seems at least those have wildly varying densities as well.

But if spice/fruit doesn't...Then assuming the weight of a single plant is around 0,036 urists, you could perhaps reasonably measure the weight of large stacks via featherwood minecart collisions (weight 4 urists, iirc) without overly large systems (a 10% increase in weight ~ stack[11] ~ requiring mere 10 or 11 tiles to tell the difference of falling behind at glance).

Perhaps this has also implications for coinstar training, though I'm sceptical.

[wierd]

Are fruits lighter than coins?

Unstacked ginko seeds might be useful coinstar fodder if they are.

[Fleeting Frames]

Hm, don't think so. Looking at stack of 500 gold coins, game claims it has 5 urists weight, which is way less than what that barrel had in plants (18 urists) (though I think it had more plants than 500, tbh). And adamantine coins are going to weight one hundredth of that.

Though, I suppose you could perhaps use split coins to ultra-precisely measure the weight of things being an exact match (with 0,0001 urist accuracy) pretty trivially with repeated (non-)frontal collisions. For non-practical applications, could be amusing if very finicky way to check the corpse being who they're identified to be, for instance

[TheFlame52]

Fruits are veeery light. I like elves in adventure mode because you can buy a hundred apples or something and they weigh 2U and last you for months.

[Fleeting Frames]

Was looking upon different ways to minimize injuries in most compact way in minecart stone quarry systems and came upon this:

Default display state when building Vertical Bars/Wall grate: X

Ok, but they're impassable after building....

For orthogonal walking/pushing.

A vertical bars/wall grate on downward ramp can be gone up and down freely for dwarves.

Neither is it an obstruction to minecart going up.

A minecart going down, if it would normally descend the hole, will.

But, if there's a untracked floor before the grate-hole, the minecart might decide to jump against the grate instead of descending, and thus come to a complete stop. A mono-directional railway

"Ok, F^2, but it is already mono-directional without the wall grate on hole, and without sufficient speed you could also use track turn - and both of those systems could easily use floodgates to block falling into hole as well. Does this do anything new?"

I guess two things that springs to mind:
When a minecart comes to a stop against the wall grate, it is in an ideal position to take PUSH signal from below, without being able to give one to it or ignore a track corner. This can be used for 1-wide monodirectional signal propagation, and of course retracting any of them could be used to mess up the system.

(I suppose push signals could also be given from the other side of the grate on same z-level with a little flight and then letting the cart fall down and ascend 1z back up.)

If you had something like a door beneath the wall grate with both having adjacant track/ramps/minecart rest places, then with opening 1 door you could get a cart signal from any of 8 possible carts (up from 4 with just a door). Not very useful, but could save mechanisms.

[taptap]

Was looking upon different ways to minimize injuries in most compact way in minecart stone quarry systems

Statues are your friend. Build on track, carts pass, dwarves don't. In dwarf-exposed minecart operations this is a good way to isolate the larger part of the operation.

[Fleeting Frames]

Yeah, and I do use statues and flying through the air where appropriate.

But sometimes, I desire something different than what I have already done. Also, statues block cart traversal to z beneath when hanging in mid-air iirc and are somewhat troublesome to get on up-ramps (need obsidian casting)