Core Keeper University

[trim kestrel]

Question: do kills using soul powers (only) give skill points? Presumably vitality. But melee for thunder beam, or ranged for tornado?

[long dagger]

Would be nice to test it on a fresh lvl 1 character but you'd need to beat the bosses. U could join someone else for it and not fight to stay lvl 1

I doubt it'll give xp for range or melee, as for vit I'm guessing no

[gusty epoch]

I doubt it gives xp at all

[inland compass]

Do bombs give vitality?

[trim kestrel]

Hmm, I assume the AFK spot for Azeos still works (if you can get there quickly enough). Then just wait for the spike traps to do their job...

Speaking, of, didn't they used to have a 2 second activation rate? I measured both, today, at 2.5 seconds. Maybe part of the nerf that broke insta-farming Glurch and Ghorm with them, maybe?

[trim kestrel]

Probably not, but I wondered...

I had this passing notion of AFK farming mob spawns from behind an obsidian wall, proc'ing beams with an NPC for all the damage. 🤔

.
Hive spikes not far off drills' 12 DPS. (Is that once per second? Will have to double check.)

Do hive spikes have increased damage in hard mode...? Given they can hurt player, too.

[inland compass]

I don't think so, i think they were still hitting me for around 30

[long dagger]

Pretty sure the patch for Insta killing ghorm and glurch wasn't from a spike change

Well...

Ghorm breaks the spikes now but before u can 2 spike hit it without spikes being destroyed if the player is atleast a few blocks away

As for the slime boss idk what they changed, I don't think I tried killing it for a long time now

[inland compass]

Maybe it is a 0.5 second trigger with a 2.0 second cooldown

Which would allow bosses to smash them before triggering

[lilac gale]

There's no damage increase that I noticed for hard vs normal.

[long dagger]

Any solutions for destroying larva sacs in a mob farm without player interaction? So far I'm just thinking of having them spawn at the very end of the kill track just to make sure they're removed

[long dagger]

Pretty sure nothing agros onto them and mobs won't fight if they're far enough from the player

[inland compass]

that would be a good application for the crab bubble firing squad, but would limit the size of the larva spawn field

[floral tendon]

the best thing i've thought of so far is putting crab islands throughout the room

but at that point you might as well leave the cocoons cause you're losing a lot of space anyway

if your mobs are fighting each other already it's probably best to just have them pass through the room and hope the hit some

[inland compass]

pink on the left is crabs, pink in the middle is larva, blue on the right is mold for bait mobs

you keep the bait mobs out of range of the crabs, but the crabs still attack and blanket the area with bubbles

[floral tendon]

that's a p narrow strip innit

[inland compass]

yup

probably like 8-10 tiles

[floral tendon]

dang

[inland compass]

i can't think of any other way to automate attacking passive objects

[floral tendon]

yeah me neither

well, specifically cocoons anyway

big larva will smack mushrooms at least

[long dagger]

I'm wondering which would be faster or if it wouldn't affect rates if it's quick enough. Have strips in the mob farm for mob types then either have it spaced with walls inbetween so they won't agro and stay in the room so it'd empty the spawn cap in that area or have them fight in that room and be pushed along the way

Currently I'm thinking of just having them spaced with walls

Atleast until they leave the spawn rooms then I could connect them

[trim kestrel]

Nah, they hit just as instantly, I think. With a 2.5s cycle I timed over 10 hits (25s total).

But checking my archived footage, this is actually half of what it was in e.g. 0.3.7! Spike traps used to hit every 1.25 seconds!

I knew they felt slower, but didn't realise the devs nerfed them that hard!

So, now, you really need to get mobs to run across a bunch of them for them to be effective. Which makes auto-loot collection that much harder.

But so far I've only found gardeners that can't be kill by running around on multiple spike traps (without poison).

Has anyone managed to get mobs to slide on slippery (blue) slime?

I was disappointed that it seems to act effectively like regular slime on them - immediately kills their speed.

[long dagger]

Damn, a big nerf that nobody noticed 🤯

I prob won't bother using spikes for my mob farm

[trim kestrel]

I think mob-vs-mob is definitely where it's at.

[halcyon shell]

Cant slide when its under conveyor or spikes it seems

the mobs that went on blue slime under conveyors and spikes didnt slide around for me

I tested it using my mold farm

[trim kestrel]

The [slippery] debuff doesn't work through conveyors at all.

But I think I tested cavelings [directly] on just blue slime, no spikes, and this is what I'm talking about.

[halcyon shell]

Ah yeah

it sux

we really need some stronger slime. Im having to put some metropolis mages in a 2x2 to shoot at the mobs ontop of having spikes + poison

[inland compass]

i can't recall if blue slime worked in my mob arena

[trim kestrel]

Player does slip on it, even through spikes. And the debuff particles animate on mobs.

[halcyon shell]

yeah idk

it doesnt work in my mob farms like anywhere

[trim kestrel]

It's just that mobs don't have momentum, I guess.

Wanted to have them slip through a line of spikes.

[halcyon shell]

I just think we need like scarlet spike traps or something

spike traps that do only like 60 damage

is way too little

especially in hard mode

[trim kestrel]

But the slide distance for player is kinda disappointing too; I wanted to have a fun slippery slide as a mode of transport. 😢

[trim kestrel]

It's what everyone's been asking for, so I expect that's what we'll get... 😐

But I'd hope for something with a different weapon mechanic, more creative or, or at least opening up new build options.

[inland compass]

how about wall spears

that way we can keep conveyors down lol

[halcyon shell]

I was thinking more so ballistas

that pierce mobs

but wall spears work too, although id want poison dipped wall spears

:)

[inland compass]

i wonder what their original intention was for spike traps

[halcyon shell]

probably cheesing bosses and thats it XD thats about all I used them for until I started making pointless mobs farms that are INSANELY slow

[trim kestrel]

Anything that has more reach would be good, yeah.

"Original intention."...?

[halcyon shell]

Everything being added to a game has an intended use

Such as a higher tier pickaxe being needed to mine the next tier

[inland compass]

right, when they developed first added it, how did they intend the players to use it

[halcyon shell]

same thing but for the spike traps

[inland compass]

like why nerf it?

[halcyon shell]

think of it like the ancient pick

It was intended as a devtool

to mine everything

[trim kestrel]

You think they didn't mean for them to be used against bosses at all? Just for base defence, say?

I think the devs are very cognizant of the game's structure evolving over time as they add content. So 'nerfs' may often be more about opening up space to add in the new stuff.

But yeah, I think they also throw spaghetti at the wall, to see what sticks, when it comes to the actual player base. Can't beat real world testing.

[trim kestrel]

Oh, so not even cavelings will aggro on (full) cocoons? Or trigger them?

[inland compass]

yeah like maybe just base defense, not mob farms or boss kills

or maybe just ghorm

[floral tendon]

afaik nothing

[trim kestrel]

I think they expect mob farming, for sure. At least at a low level. It's done to death in Terraria, etc.

[trim kestrel]

You could have waves of crabs and infected released periodically up parallel conveyors, instead of continuously in situ. Say, hourly, or less.

Kinda complex to implement, but much greater coverage per amount of mob-collision-based fps-loss.

[inland compass]

would be hard to sync the bait mobs with the crabs

[trim kestrel]

Automatic doors. Near player's AFK position.

On long period timer.

But, I'm not sure if cocoons even slow the spawn rate of Larva...?

Like, different mob types (e.g. slime colours) don't slow other types. I've not tried to test if specific mobs slow spawning of others.

Like Mold Tentacles and Infected. It kinda seems like the tentacles aren't limited by a big build-up of infected. Only by their own presence, maybe.

So it could be only an issue of blocking that one specific tile. Which isn't a big deal (compared with mushrooms).

[trim kestrel]

Think this alliance diagram is correct?

Does anyone fancy testing if shoomen and slime take any damage from Omoroth?

[halcyon shell]

BIRD

[trim kestrel]

Loney bird... Makes me wonder if we'll see more bird-people in the dessert or crystal biome...

[trim kestrel]

Geez... Even Shroomen have >12 self heal per second... 😵‍💫

Even with the comically weakest enemy, you gotta poison them to get a kill with acid burn alone.

[pale merlin]

is there a good guide for building mob farms? im not so great at building AFK farms, mostly decor/farming

[trim kestrel]

I'm working on it... 😜

[pale merlin]

👍

[trim kestrel]

Depends what you wanna farm, really.

[pale merlin]

ah, cavelings, ig
slimes and shroomish still do well under spikes, but the sentry and esp with the gardeners escaping water+wall has me confused.

drills went a little slow. wondered if there were better ways, or if mixing the mobs to let them kill each other was better

[trim kestrel]

I think mob vs mob is definitely the way to go, late game, yeah. There's been a bit of discussion, above, with different ways of doing that. E.g. bubble crabs with infected as decoys, for cavelings. But I don't know of a simple, 100% reliable design, yet.

Although, spikes plus poison slime will kill everything, in normal mode, I think. If the mobs are able to move around enough to get hit faster than that base 2.5 hits per second.

KanSer used a combination of the above; just dumped all the mobs in a fair sized poison spike area, together. Let them fight it out to speed things up.

But that means having to manually collect the loot. Preferable with the help of the magnet rings.

Did you have a caveling farm working with (just) drills? (Got a pic, if so?)

[pale merlin]

ah, not working. belts go down, drills point up. but ig it'd make sense it doesn't do enough on hard mode

haven't worked on any farms since pre-sunken update, but trying to see what goes well now and just. not workin out

[trim kestrel]

Oh, hard mode farms is another thing entirely. @inland compass has done the most (I've seen) on this. And yeah, that's gotta be mob-vs-mob.

When even the weakest enemy in normal mode has over 12 self heal per second, and I'm assuming that's doubled with hard mode health scaling, too... Yeah.

Health doubled, spike damage output halved. Factor of 4 harder to kill mobs in a simple way (than pre-Sunken Sea).

Oh, and there's some random mob teleporting, that may be unavoidable, currently.

But you can prevent mass pile-ups, through walls and spilling into the void, by making sure that there's always an empty tile of ground at the end of any conveyor line that carries mobs. (When you're over 205 tiles away, mobs effectively turn into items and get stacked off the end of the line, otherwise.)

[inland compass]

the 2 methods i've seen are crabs, even without the bait, they seem a bit more tanky than mold

and then lots of mold tentacles with infected in front of them as meat shields

[trim kestrel]

I guess those are the only two viable options... If one's including Stone Moss Cavelings, and so, Brutes. Which have to be dealt with at range.

(Sea tentacles too flimsey and slow to respawn, on their own. But maybe ok as a backup to compliment Crabs.)

[inland compass]

So one option, might be super janky

Hive mother larva, however if you did stone moss you'd still have the brute issu

[trim kestrel]

I'd just been thinking about bosses, and yeah, hive mother's the only viable one. And as you say.

[inland compass]

And malz for non cavelings

Feed a bunch of infected and slime to him lol

Might be hard to collect loot though

Too bad ghorm doesn't drop their loot

Maybe morpha for cavelings too

[trim kestrel]

My thought was that none of those bosses would work in practice... Maybe I'm wrong?

In each case, you'd have to be in their AFK spot, right? So they don't aggro on you... Then even small amounts of damage will chip them to death? Like farming them with spike traps.

.
Does core keeper keep a crash log anywhere?

(Game crashed out while I was AFk for 3 hours running a spawn rate test, so I need to know exactly how long it was actually up for, to be able to use the mob counts, with any confidence.)

.
Oh, and a cheeky farm idea for killing non-cavelings: sit Brutes behind an obsidian (maze) wall so they can repeatedly pummel mobs on the other side of it with impunity. 😅

[trim kestrel]

Drills got nerfed from 1 second to 2 seconds now, too.

But you can at least farm shroomen with 3 of them. 😅 Not slimes, cos they aggro and break the drills.

[floral tendon]

was it ever stated whether the wood not growing on wilderness ground was a bug or a feature

[long dagger]

It wasn't stated in the patch and I don't think the devs even mention unlisted changes

[long dagger]

it has been done

ill prob chip part of the dirt and start working on the layout for the farm another day

[floral tendon]

here's a riddle... why does nothing spawn in the circled area

i can't imagine that local mob density is maxed out by just the tentacles in the lower room

and the upper platform spawns fine

[long dagger]

Try adding a platform of mold floor to the left and right of it and check if anything spawns

[floral tendon]

hmm i see the train of thought

will have to see

[long dagger]

The tentacles do take up the mob cap, as for the range of it idk. Had a bunch in my farm before removing it. Prob had 5-7 or so on my screen at all times

Since ghorm can destroy obsidian and replaces/adds dirt ground and the slime bosses can add ground, u think the slime bosses can destroy obsidian? Also would it be possible to bait them further from their spawn (bait from spot of agro for ghorm)? Wether it'd be relogging or something

[primal rapids]

I think they can destroy obsidian, but you can only kite them a short distance from their spawn

[trim kestrel]

Frost welcome the void into his world. 😅

[trim kestrel]

Did things ever spawn in that bit?

How far away is the other side, there? Like 8 tiles? I think even if it were totally packed full of infected, it'd only block mold spawns out to 8 or 9 tiles away...

So maybe you've found a mob-spawning dead zone, like I got in my rate test setup, near my core. 🤔

[proper coral]

On the right side is the typical design using conveyors. On the left you can see you can just use chests on both sides

it is heavier on drills but it is significantly faster since it doesnt have to waste time on conveyors

can do the same to smelteries, of course by inverting the timing signal on the "refined metal" arms

[trim kestrel]

Ah right, thanks. I think this is similar to the design @long dagger came up with back here...: #📷-ck-media message

It makes sense, space wise. And is still just 1 drill per chest, right?

[proper coral]

yep 1 drill per chest

actually so yeah same cost not heavier

it makes more sense for when you want to quickly sort a system though

[proper coral]

nice, yeah. i was looking for someone who had done this

[sharp nimbus]

Is there a practical use for this? Or is it just for fun/testing the automation systems in the game?

[trim kestrel]

If you're producing a lot of different items, say in a big mob farm, then it's handy to sort them away automatically. Rather than manually sorting them every time.

[sharp nimbus]

Ahhh gotcha

[proper coral]

it becomes a massive pain to sort everything out when you try to save everything

[trim kestrel]

Otherwise, I'd say, it's a bit of fun.

[proper coral]

and yes, its quite fun

[trim kestrel]

Your (and Frost's) design(s) uses half the number of arms as a one-sided design too, of course, right?

Although, I'm not sure I'd like the two sided access. In terms of player work flow. But I guess it depends just how much storage you've got and how you use it.

[proper coral]

yep it is more arm efficient. i find you can fit 15 double chests in a single column, accessibility wise

can easily power 2 columns of 14 chests with 1 power source

I do wish this game had more in terms of logic and industry, it's pretty fun but there's only so much you can do

[trim kestrel]

Needs more input output devices for wiring to interact with the environment, for sure.

[sharp nimbus]

I imagine at some point there will be water manipulation? (I hope)

And maybe some advanced tech from the metropolis style ruins

[long dagger]

u can extend it with just this

[proper coral]

oh nice, i thought i tried that but i must not have

[fossil topaz]

Has anyone successfully created an Acid Larva farm? I haven't been able to get the suckers to spawn so far.

[trim kestrel]

They can't be spawned aside from during terrain generation (in exploring a new world).

[fossil topaz]

Oh Damn, I need the statue for my collection. Off I go exploring new worlds then.

[serene kraken]

why have the mage 369 life?

[trim kestrel]

The shaman has 356. You mean, why less than cavling miner...? Cos they are a delicate ranged unit with a lot of scary fire power...

[trim kestrel]

Urgh... The order of precedence for stacking floor layers doesn't tie in with what blocks spawns, a lot of the time...

Like, stone tiles appear to be underneath slime, but still block slimes spawning. Unlike lit floor tiles, which permit slimes but block caveling moss, etc, as expected from appearances. 🙄 I've not finished with testing all of these yet:

[hoary osprey]

Is moss will block the spawn from mold floor?

[long dagger]

I think zero was talking about something like that that i skipped over, he may know

I'm having slow spawn rates for my crabs, I made a nice fat chunk for them and bridged off a couple tiles yet they've been spawning slowly. I doubt they would've made a seperate type of mob cap for the new biome but I think it may be different

[floral tendon]

i had a struggle trying to spawn crabs that way

not sure if it was the wall or the conveyers, i ended up just spreading conveyers spaced out by 5 across a big area

mostly used pre-generated terrain instead

[long dagger]

Idk if zero mentioned crab spawns, I skim through too much. I'm getting lazy, I don't wanna rework my crab platform too much

Digging out the stone biome was too much

[trim kestrel]

Yes, Caveling mosses should take precedence and block mold floor.

Only the top spawn surface layer should be able to spawn anything. (Of course I've not tried every combination to be certain of this.)

[trim kestrel]

How big is that area of beach sand? And what do you call slow? What did you expect?

With 1k tiles in a square of mostly naturally occurring beach sand (all in the sea biome) I got a spawn rate of ~1.1 crabs per 100 tiles per hour. Which felt fast to me, compared to Ruins cavelings on stone, only, for example.

This is my full experimental spawn rate matrix, as of this morning. I'm starting to feel that a lot of the variations have been down to spawn rate anomalies in patches of the map. I.e. dead patches and (adjacent) bright patches. That are consistently high or low in spawn count.

[long dagger]

i didnt consider that, natural spawns just feel insanely slow

mines 52 x 60

was just hoping for consistent fast spawns

[gusty epoch]

Crafting 1 of WoodBridge, got 1 extra, variation: 0
Crafting 1 of WoodBridge, got 1 extra, variation: 0
Crafting 1 of WoodBridge, got 1 extra, variation: 0
Crafting 1 of WoodBridge, got 1 extra, variation: 0
Crafting 1 of WoodBridge, got 2 extra, variation: 0
Crafting 1 of WoodBridge, got 1 extra, variation: 0
Crafting 1 of WoodBridge, got 1 extra, variation: 0
``` Looks like this skill can proc twice on one craft. I never noticed.

[trim kestrel]

Just over 3k tiles, So my expectation would be for ~30 crabs per hour, from that. Plenty for stocking up a shooting gallery. But not going to get a huge amount of drops, of course. Was there any particular drop you were after?

[long dagger]

not rly, was just disappointed at the spawn rates at the time, extra figs would be nice tho

[trim kestrel]

Oh, so the bonus item can spawn a bonus item...? I'd not noticed either. It should happen quite a lot, though, at 0.3 * 0.3 = 0.09. I.e. 9% of crafts.

[gusty epoch]

Yea I was just setting up a drill rig and noticed when I glanced at my console window lol

I lazily used that to log crafts and make Polished Scarlet Chunk Necklace when crafting that was broken

[trim kestrel]

I have a suspicion that it may be possible to get much high spawn rate per tile of sand, with crabs, by using strips. Say 1 row of sand to 2 tile of something else.

This would use as much, or a little more, space per spawn-rate. But a lot less sand. And you could mix in strips of the 3 different slimes, together with the crabs, as they are all friends. Or non-allied mobs, if you just want the drops (not stone cavelings, of course).

[long dagger]

I'm a lil burned out, going to sleep in a bit but if I do test it then it'll prob be 20+ hours from now

[bitter turret]

fr

[inland compass]

Natural spawn rates are crazy low, that's why they added moss

[hoary osprey]

If natural spawn rates are high as moss, it is too hard to explore then

[inland compass]

Re-explore you mean.

[hoary osprey]

Just too much tentacle and crab on the sea and both of them are ranged

[inland compass]

Right, but that isn't from respawn

When the world gets generated, it gets populated. That rate wouldn't need to change.

[trim kestrel]

It's not populated up to the [spawn limit] though, is it? 🤔 So, it could get significantly more crowded, if you're not a very fast, decisive explorer.

Like, the mold dungeons already kinda get outa hand, if you set up base within range of any.

[trim kestrel]

@inland compass - would you say I was being too picky (or wrong) if I said the map tool's chunk grid looks like it's one tile (pixel) too low? 😅

And with lines on an even symmetry (0 in between tiles), when the world appears to be odd (0 in the middle of tiles).

Did you scrutinise this pixel alignment carefully wrt to the world generation chunks we saw in the maps with -nographics on?

[inland compass]

0,0 is the 2nd tile down from the core

here's a -x at 0 y tile

and the core tiles

[floral tendon]

while we're nitpicking might i suggest reorganizing the breakable crate checkboxes

most of them are under furniture but jellyfish is under sunken sea and large jellyfish is under boulders

[lilac gale]

+1 to this

[inland compass]

i haven't seen a data dump like we had on the previous update, so I really don't know what tilesets they belong to in order to organize the new tiles like that.

[floral tendon]

oh is it based off something like that

they put large jellyfish under boulders??

[inland compass]

no, all the sea biome crap i added manually by looking at pixels i had on my map

[floral tendon]

oh gotcha i misunderstood

[inland compass]

which is why i don't have any of the furniture or anything for the seabiome

originally someone data-dumped the tileset list into csv format and that was the basis for filtering

[floral tendon]

i see

[inland compass]

i thought it was amusing that you could see the jellyfish from the map

and that the large jellyfish turned out like boulders

so i added them 😄

[trim kestrel]

Heh. I knew you'd be on top of it, thanks for the clarification! 😅 👍

It's possibly relevant to what I'm currently boggling my brain over, with the patterns of spawn-gaps...

[trim kestrel]

Did you see Cysidus's 0.4.3 parsed data dump in the Wikizet server? Any of that helpful?

[inland compass]

nope

[trim kestrel]

As in, not helpful?

[inland compass]

their discord?

i didn't see it

lol yesterday it appears

doesn't look like it has the tileset info, Mz had pointed me to a spot where I should be able to grab it but i forgot

[long dagger]

2 clips put together so there's a slight cut in the middle. I'm still thinking raid parties just select already spawned mobs and have them run towards u

[floral tendon]

not just run, i'm very sure the game can just yoink them from anywhere nearby

[trim kestrel]

That is beautiful visual comedy! 😆

[trim kestrel]

When you have interleaved strips of spawn surface, like that, each spawn event commonly gives a mob (or two) from each biome type.

So, what you showed would be expected, if newly spawned mobs get set as a raiding party, at time of spawning.

[bitter turret]

what is a raiding party

[trim kestrel]

Occasionally (usually after killing some mobs), you will get chased down by a small group of mobs, making a bee-line straight for you, from well off screen.

They'll sniff you out by running along the path that you just took.

[bitter turret]

ohhhhh

so that's why i felt like several cavelings were ganging up on me at times

[trim kestrel]

Mob spawning mechanics investigation:

A couple days ago I watched @tidal dock's video on mushroom/shroomen farming, where he observed big gaps and bright spots in mob farms, too: https://youtu.be/UxEt-d9fyWY?t=356

I said I'd tag him here when I posted some findings from laying out big mats of fungal soil (and spikes) around my spawn-rate anomolies, on a new world, too, so here I go...:

Map view of my old main world (left) and new test world (right) where I laid out fungal soil and AFK'd for 2 hours and 4 hour respecively:

(New) World 2's main area had unexpected line pattern of spawns (I thought it would look more organic, with diffuse bright and dark areas):

The horizontal lines of high spawn rate, at 16 tile intervals, extended down to the core, too:

And I saw very similar in my old world, at the same 16 tile division lines. So I think everyone should see the same in their worlds, with similar experiments.

It seems like the spawn algorithm starts at a multiple of 16 and looks up for a valid spawn surface. Then looks across to the right. Hence the accumulation of spawns on the left edge of the areas, but not the right and top edges.

But if the spawn area is restricted, it kind of overflows up to the next row of valid tiles, extending the spawning area up.

At least, this is what happens to the x=0 co-ordinate. I think it might be less pronounced, out away from the world axes.

Any thoughts?

I know at least one person was previously posting (possibly in here?) about gaps in theri shroom farm. And another YouTuber was seeing dead bands in their (just one small part) of their mob mega-farm.

So it's a wide-spread phenomenon, it seems. But harder to observe outside of killing every spawn on the spot.

I needed to dupe up more fungal soil for these tests (getting only 300 or so from the new world's dirt biome).

And I'm tempted put aside my aversion to mods, to install the placement mod, to make a much bigger test viable to set up. Unless @gusty epoch or someone has a better suggestion..?

[gusty epoch]

I can give you however much fungus you want lol. Using placement plus would make putting it down a cakewalk

[trim kestrel]

Thanks. I'm good for making more materials. And yeah, I may need to go with placement plus... Unless someone else fancies 'tiling the plain' for me, so to speak. 😜

In terms of exploiting what I've observed, above, I think you may be able to get bonkers per-tile spawn rates by positioning tiny spawn areas at the intersection of these n*16 grid lines.

E.g.: Above and to the right. Say, 5 spawn surface tiles in a horizontal line. Maybe a row below that too, if the spawns from the grid below will get pulled up that far (not sure). it's probably more complex and mess than this, when moving around the map.

[lilac gale]

Simply have them spawn better, that's my suggestion 😁

[inland compass]

try it without spikes

i don't have spikes in my farm and i have never seen any type of line behavior

off to the right, not very big but it would fill with mushrooms

it might attempt to spawn them in 16x16 squares, and if that fails it pushes them to the bottom row

[trim kestrel]

That is quite narrow, and that seems to cause the lines to wrap up to the next row. Also, I guess having a tile blocked by an object (e.g. mushroom) will cause it to search for the next tile...

Of course, my objective isn't to rid myself of lines, but to see what we can figure out from them, about the algorythm. And is they tie in with bright and dark spots, directly.

Next level of test ready for some afk...

[unborn summit]

Now thats a lot of mushroom ground you have there

[trim kestrel]

Yeah, I cheated. 😢

[unborn summit]

😭

[long dagger]

My msg didn't go through cause I slipped a word in and I can't copy paste...

Just checked some of the pics and counted the tiles for one, looks like a 9 tile gap

[silk crescent]

I recommend you don't just resend a photo of what you typed with the word still in the message.

[long dagger]

Tbf it wasn't targeted at anyone, I'll edit it out, sec

[silk crescent]

Thank you

[long dagger]

I'm a bit more confused now, I can't think with all this first thing in the morning

[trim kestrel]

To maximise spawn-rate per tile, in this (maybe unusual) section of map I've looked at, it looks like maybe having a single row (or short strip) of spawn surface, every multiple of 16 tiles upwards (starting at map origin zero), might be very efficient.

But alternating stripes of spawn surface is useful in general. Because there are specifically caps on how many mobs, of each type, can be spawned together in a single event. But they all operate independently, it seems. So you can get one or two of each type, simultaneously.

I don't know what kind of algorythm the game uses to decide which tiles each of these come from. I previously imagined a little circle, of randomised x-y offsets. But maybe it more kinda crawls about linearly...?

Btw, I saw, repeatedly, a pattern of 2 mushrooms and a shrooman appearing together, just off screen, while I was setting up this big strip (and clearing it ready for launch).

[trim kestrel]

Dynobot clobbered me for "🇳 🇮 🇬 🇬 🇱 🇪 🇸" the other day. As in, minor qualms. Too close to the bad n-word.🤷

[long dagger]

Canceled by a bot

[long dagger]

Still a bit confused with exact positioning but I'm willing to wait for results, I'd prob screw up my own testing by doing it incorrectly

[trim kestrel]

I usually do. But sometimes you learn something unexpected in the process. At least, that's how I console myself.😅

[long dagger]

Tryna convince me to do some tests? 😅

I'm a bit lost of where to start for it besides testing some forms of spawning strips but if u can explain it in a way I can understand then I could use some time for testing it

[trim kestrel]

So, a couple things I'd like to be sure of: that these lines do indeed appear in this spot I've already tested, in everyone's worlds.

Then, to see if they occur all over the world, far from the axes (x or y =0).

For that, I'm planning to set up just a big spawn box of fungal soil with spikes. See what it catches.

Orange slime could be used instead, for a slightly less clear picture, as those mobs die in 2-3 hits. And, infact, I'd like to try that on the exact same spot I already did, in my world 2. To see if the pattern is the same for other surfaces.

I've not made a clear representation of this yet, but these spawn lines should lie along the chunk grid lines of the map tool and sub-divide those further into 16 squares, of 16x16 each. (Heh, numbers!)

[long dagger]

Knowing this will help out with the testing

I'll prob test it away from my core tho since it's kinda occupied

[floral tendon]

something I've been curious about actually

i know it's been said for a while that stripes/checkers are "better", but better in what metric exactly?

like would 20x20 striped with two spawn types be better than two separate 10x20 sections? what about 20x20 striped compared to 20x20 of a single type? would spawning multiple types at once necessarily be more than spawning more of one type, and does this depend on the specific mob?

i'm sure you probably covered some of these in your testing already but searching is a bit of a pain lol

[trim kestrel]

With slime, a 10x10, with 50 tiles of orange and 50 of purple (striped), produced both types of mobs at about the same rate as two separate 10x10s filled with 100 of each.

And I think that worked with a 3 way split of caveling mosses. But my testing was done near this core spawn anomaly, so I'm not sure about quoting numbers.

And yeah, I struggle to find my own stuff in here! Doesn't help that you can't narrow down Discord searches to within threads only.

I need to finish bringing it all together into the fabled guide post. But the details are hanging me up and the final shape of it all is still amorphous, in my mind.

[floral tendon]

hmm that's interesting

i wonder if that's intentional design or not

feels like not but hard to say

[trim kestrel]

If not for Brutes, an optimal ultimate farm might be to just stripe everything together on conveyors. Blended in such a way that one ends up with only an easy to kill mob piled up the end, maybe.

[floral tendon]

well if it weren't for brutes you would just flood everything with mold cavelings and have them picked off at the end with scholars or crabs

but alas

the fact that shamans and brutes spawn together is a huge headache lol

[trim kestrel]

I peeked at some assets, and it seems that the terrain object and mob (re)spawning code is all combined. So it may be quite thrown together, currently, and up for major refactoring... (Just when I have it all figured out, probably.)

[floral tendon]

yeah

the pain of early access research

[lilac gale]

The pain of research in general tbh

[trim kestrel]

I still think there's a food chance that the right ground pattern in the right places of the world may avoid ever spawning brutes. I might try to parse their spawn requirements later. But there's definitely aspects I can't see, in what I've looked at.

[floral tendon]

who knows 🙃

though i'd prefer figuring out a way to farm everything if possible anyway

[trim kestrel]

The horizontal lines do lie along the chunk grids, exactly. And divide each chunk into 4 vertically. I marked them with notches on the side of the test area.

And video of me walking around looking at where all the spawns landed (even more uniform than before): https://youtu.be/jbelyMCCLcQ

I AFK'd in this spot for 3 hours. Looks like nothing at all spawned within 5-7 tiles of me. As expected, pretty much. Drops a tile or two closer, due to Shroomen trying to make a dash for me, no doubt. Heh.

NPC didn't put them off, but none spawned in with him.

[trim kestrel]

Note that blocking the spawn algorithm from moving horizontally is what appears to have caused it to expand up and fill in the space either side of my AFK spot.

Also, the only place where a vertical line was created, was in the one tile wide strip that overhung the x=0 axis, at the top of my test strip.

I think means that it can look only up to 16 tiles to the right, as a limit. In a 16x16 spawn grid. Once it hits that limit, it then looks upwards.

What I'm wondering, is if a spawn in a tile gives only that specific tile a cooldown from spawning. So next event, that counts like a blocked tile and it skips over, extending the line to the right. Or if it's more stochastic than that. I should be able to test this by recording a (long) video of spawns in action.

What blows my mind, though, is the 2 tile notch 24 tiles away at the right edge, that appears to correspond to the indent on the far left side... Like how is that affecting it so far away, if that's what it is...?! (No missing drop spots anywhere else down the right hand side.)

[bitter turret]

how'd you do it

[trim kestrel]

Duplicated a stack of 300 fungal ground tiles i harvested from this world. By repeatedly restoring a world from backup, after taking the stack from it. (Don't do it at home kids.)

I'm keeping it all confined to this test world though. It's bad enough that I caved and made a new world to harvest, after farming over 100 fungal tiles from drops in my old world.

[floral tendon]

hey man, gotta do what you gotta do, i ain't got time to farm a gorillion fungal tiles either

[trim kestrel]

That's a fun mix. Bit of a rough shuffle, though... How'd the conveyors collect them like that?

[long dagger]

i wasnt nearby when they piled up

They get clumped when its not loaded and they expand and spread out a bit when u get close enough but in this case there was too many to fully expand

[trim kestrel]

Oh! Now that is fun! Kinda like centrifuging your cavelings by width. After stacking them all on the same tile at the end of the conveyor... 🤔

[tidal dock]

Independent sounds about right. Since I got a dead zone for slime but then replacing the slime tiles with lush I started getting spawns of azeos mobs. Still requires a bit more testing to see variations

[shadow grove]

hey guys, wasnt really active here to follow up all interesting farming videos... does anyone has a good and most effectif method how to build mob farms? i think there is a lot of university staff whats the best size of farms to and in which area to build - to get the most loot/mobs per tiles

[shadow grove]

you just copied the map file? or the character file?

[trim kestrel]

I don't want to give a tutorial on how to cheat the game. But I'll say that I acoid messing with the character file at all. It'd been said that it's encrypted and might be more prone to weirdness.

If you mess about with game saves, always back up the whole directory to a zip file archive, first, to be safe.

[trim kestrel]

It's interesting why that might happen...

Do the different surfaces get treated separately in the spawn algorithm?

Or did you move around some spawn surface near by that let the algorithm come in for a different angle, or something...?

I think I should try to test slime, next. Going to take a lot of hoeing if I decide to clear out all this fungal soil, though! 😅

[trim kestrel]

I think I'm getting close to being able to say with some amount of certainty what's best. If there's not any more too major surprises.

That spreadsheet pic that was linked gives an idea of a couple of instances where spawn rate should be a lot higher for e.g. larva on clay caves or urban moss in sunken sea.

But the smaller variations may just be noise from the effects of the algorithm at different world grid (relative) positions. And I can't rule out that causing some of the big variations too, at this stage.

But using alternating strips of spawn surface is definitely good for increasing mob output per tile.

[bitter turret]

guys

what are the highest and smallest signal strengths

in core keeper

[inland compass]

I'm not aware of any items in CK that emit electromagnetic fields

[bitter turret]

...

[inland compass]

What signals are you talking about

[bitter turret]

just like in minecraft

[inland compass]

Never played it

And that's the wrong term for it

[bitter turret]

well that's what all the MC redstone engineers use

[inland compass]

So if i recall correctly it is 20 tiles from the generator before there isn't enough power to power devices

[bitter turret]

basically the longer a circuit is the less "strength" that redstone or part of the circuit transmits to other blocks

[inland compass]

Well lets not inherit the wrong terms from Minecraft

The longer the circuit, the more resistance, and less current to apply to a device

[bitter turret]

it's 24 lanterns

but yeah the last 4 lights are not very powerful

[inland compass]

Try with drills

I've only seen it referenced with drills, perhaps they need more power than lights

Signal strength... Ugh the cellphone generation

[bitter turret]

still 24

the 25th placed on any of the 3 sides of the 24th will be inactive

i mean

not inactive

not powered*

[inland compass]

K

Would be nice if they made the next generation metal super conducting. And added current draw, such that say 1 generator could power the equivalent of 2-4 drills

[bitter turret]

idk

so basically

a generator can power 24 electric circuits in any direction

and the components of the electric circuits can power 24 - the current index (?) of the element in the circuit

in the other directions

[inland compass]

Ugh no don't think of it like that

The generator powers anything in a 24 tile radius. Power goes through components in the circuit. Imagine it just has additional pass through wires

[bitter turret]

what i meant to say is

you have 16 electrical components going in a straight line from a generator

now you can out another 8 upwards, downwards and to the left/right of the 16th element

and so on and so forth

[floral tendon]

i mean ... you aren't wrong but that's needlessly complicating it lol

is it within 24 tiles, y/n that's it

lights start dropping off around 15

[trim kestrel]

And this is why a self-feedback delay circuit loop will tail off in power level (brightness) in the last couple of seconds before it turns off. Each of the tiles distance around the loop get added to the effective path length.

Which I guess is an interesting hidden value. Not sure if useful.

[trim kestrel]

I mean, I had a mobile phone in my teens (technically). I didn't realise you were such an old grumpy, Craigins. 😜

I think we're onto the "wifi" = internet connect generation now. 🙄

[inland compass]

Lol. I had the family nokia for road trips when i was 18-19 does that count

[trim kestrel]

Almost.

[inland compass]

By 21 is when the richer students started carrying cellphones.

[trim kestrel]

Oh right. So you're a (late) gen-x, then?

[inland compass]

1981, so i believe that makes me a very old millennial

[trim kestrel]

And I thought I was stretching "elder millennial" at '83. 😅

[inland compass]

I guess I'm an ancient mellennial then lol

But i do a lot with electronics at work, signal strength is talking about data communication. Signal to noise ratio. Stretching that could be like the wave form required to power a device

[trim kestrel]

Oh right. I studied cybernetics with a large portion of electronic engineering. So yeah. But. 🤷

.
Anyway. It looks like this my be the reason some of you were struggling to get any mold tentacles, or such, spawning on some of your farm kill-areas. @floral tendon?

Spawn events seem to start looking from the bottom left tile of each 16x16 sub-chunk. They look right first. And if there's enough spawn tiles there, all the spawns will land in that area. So none will then loop up to the next spawn area, above (where I'm standing, or the bigger one above that).

[floral tendon]

that's pretty weird lol

yeah it sounds right though

[trim kestrel]

And I had a (delayed) thought about a finding of this, earlier. Assuming it was valid and not a spawn algorithm artefact...

If caveling floor tiles block slime from spawning any mobs. But do allow the debuff to take effect. Then they are ideal for putting under poison ground in mob kill areas, in instances where you want to avoid spawning purple slimes.

E.g. when you're using drills on cavelings, who don't aggro on them, but slimes would break the drills. Or they may rarely cause brutes to attack them and break the setup.

I'd be grateful if someone could check this, though. 🙂

[floral tendon]

well

i have a feeling drills will break them even through slime

because that's what happens with logic circuits

[trim kestrel]

Oh yeah... hmm.

[floral tendon]

and caveling floor tiles behave similarly

for some reason they get priority when hitting the tile

[trim kestrel]

Still. To avoid mob-vs-mob aggro with rare slime spawns, could be handy.

[floral tendon]

it's worth considering yeah

man that means i should reroute my brute route

pain

i don't wanna move my crab platform 😞

[trim kestrel]

There may be easier work-arounds. Like extending your upper spawn area out to the left, to pick up spawns from the next cell over (but only on the tip).

And, to be honest, I've not finished checking this applies to other mobs, fully (running orange slime now).

What's that map area of yours look like, with chunk grid lines on?

Maybe we'll have to ask @inland compass to add 16 tile sub-divisions, next, once I'm more condident. 😅

[lilac gale]

Man is dust at this point, so sad 😞

[inland compass]

Lol

[floral tendon]

not at home to get the screenshot, but I'll check the subdivisions when i get back

[tidal dock]

The floor tile I changed out. Slime tiles gave number dead patches complete dead. I only replaced 3 rows with lush . Left three rows below showing the dead. Total of 6 rows I believe were dead.

[trim kestrel]

Have you got a map-tool screenshot of that, with chunk lines? And for the area you showed in the mushroom video, if it's a different place? 🙂

[tidal dock]

Yea. It's another screen over from my video

I mentioned it on stream at one point. Probably have a time jump to it somewhere. I'll check or screen shot it here when I get a chance

[shadow grove]

does the chunks has any effekt of spawn rate or farms?

[inland compass]

the only conclusive evidence of chunks is that the game generates the world around you in the 64x64 chunks

[trim kestrel]

OK, so I've just sat and watched this for half an hour, at the end of a 4 hour AFK session. And I can tell you all that:

(1) Orange slime does indeed use the same spawn algorithm as mushrooms. That is based on a 16x16 grid of cells (or sub-chunks).

(2) The primary spawn always lands on the y=16xN lines. If there are enough spawn tiles in width to grantee an open tile (not on cooldown, or randomise chance of skipping over, not sure yet).

(3) If a secondary spawn gets rolled, in this case a red slime, it can spawn far away from the primary. Furthest I saw was 14 tiles up. But I think , still within that same spawn cell. Always simultaneous with the primary. (Same as what I saw, in passing with 2 mushrooms + 1 shrooman.)

(4) Each spawn cell has it's own 15 minute timer. Very little variation from this (seconds) across 3 spawn events in these two adjacent cells I was watching.

(5) This pattern appears to happen everywhere. It looked just the same with these additional spawn areas top middle and mid-right. So not at all only on origin axis.

Also, I may have to re-visit my assertion of a 105 tile loading range for spike traps. Because it seems like mobs still got killed on the spot without being able to roam. Despite being well over 105 tilse from my AFK spot...

Maybe it was mob AI that stopped interacting the same with traps, in my previous test... 🤔

I don't understand what happened in this bottom left corner, though. I expected the opposite, that the orange slime would take all the spawn events and block that corner from making shrooms... 😕 I'll need to think on that.

[trim kestrel]

But it seems to me, that a very spawn tile efficient way to deploy a farm is in 2 tile tall strips (or maybe 4, need to test directly). 1 tile about the y=16xN lines, 1 tile below. Then a 30 tile vertical gap until your next strip.

If you are farming only one type of mob. For sure if that's mushrooms. I need to test how exactly striping different spawn surface works, here.

Questions, comments, confusion, general disbelief and doubt? 🇽

[inland compass]

yes

[trim kestrel]

Any bits in particular?

[inland compass]

test it out, see if it aligns

[trim kestrel]

Oooh! Sweet! 🤩

But, I mean, all the stuff I've done does align. Because I already marked it out in tile at the side of all the areas. E.g. the notches you can see up the side of the biggest column, here: #977128133142659102 message

[inland compass]

yeah i just wanted to make sure the offsets are right

and that they didn't do anything weird with rounding or something

(like start at 1 )

[hoary osprey]

Thank you very much for the 16*16grid

[trim kestrel]

Oic. check your working. Heh.

[inland compass]

yeah

[hoary osprey]

But how to approach to the new version of map tool?

[inland compass]

ctrl+f5

do a hard refresh to force the browser to download the latest files

[hoary osprey]

It works. Thank you

[trim kestrel]

That looks right to me. Given that it seems the devs have defined whole numbers to be on the transition between tiles. you mob grid lines have highlighted the tile above and below each. And to the right and left of each. 👍

My notches alight with the tiles the spawns land on, which is the first tile up/right from the 16 lines.

[inland compass]

ok looks good

[trim kestrel]

This will very be handy for me to look quickly at my previous spawn test areas, and try to figure out what was happening. And maybe set up new ones far away from the origin....

[inland compass]

bet your spawn distance is really 208

🙂

13*16

[trim kestrel]

I refined that to rule out 206 tiles, though, specifically.

I feel like it might have been defined as 5 tiles exclusion around player and then +200 for mob spawning.

[inland compass]

ahh

so would that make optimal mob farm rooms 16x16?

and same for spawn rate testing, if you had a 10x10 room that spanned multiple 16x16 cells, that could in theory spawn more than a 10x10 in a single cell

[trim kestrel]

Optimally standardised mob spawn rate testing rooms, at 16x16, yes.

Optimal spawn rate per tile, as I say, probably thin (horizontal) stips with a lot of air between.

.
It depends on the observations I don't fully understand, yet. And how far striping different surface types can be stretched, in terms up multiplying up the number of spawns.

.
These two chambers I'd already identified as having below expected spawn rates. About half.

And it looks like that's because they fall entirely within a single spawning-cell. So they only get 1 timer. While the majority of the rest are bisected by at least 1 grid line. So have at least 2 timers.

[trim kestrel]

Hmm, that was unreasonably fluky... My (several months old) spawn-cap test chambers all fell within single spawn cells....

Except the one in the wilderness. Which topped out at ~12 slimes vs 8 in the most populated of the rest.

So that might be down to spawn cell mechanics instead of biome (or distance from core) specific mob cap variations.

[trim kestrel]

Oh, snap... Probably not a coincidence that spawns just happened to be (mostly) blocked in the closest 4 columns to the full chamber...

They were almost entirely in the same cell as most of those mobs. (And 2 tiles at the bottom of each slot, outside of that cell, may account for those couple stragglers.)

(6) So I'm gonna call probable spawn cell based per-mob type spawn caps.

Which is why that cell managed to get so full! Because they were being pulled to the left (by conveyors), freeing up the sliver at the right (and dot at the bottom right) to keep spawning more. 😅

[inland compass]

so i wonder if you could get a small 2x20ish strip with conveyors pulling the mobs away to overdrive spawn

get them covering 6 spawn cells lol

[trim kestrel]

And this was the first 'spawn-rate anomaly' I zeroed in on. The two little rooms (with only a strip of spawn surface at top and bottom, each) spawned twice as many mobs as the T-shaped room in between.

They each got first shot at the spawns from two different cells...

And that also explains how the T-room snagged all the darn sentries, but the little room none! 😮

The secondary mobs have to be spawned away from the primaries. So it was throwing them all to the other side of the wall!

I think that should work with brutes, too... 🤔

[inland compass]

i upped the alpha to 0.5 from 0.3, should make the lines a little easier to see

[trim kestrel]

Yeah. I mean. The name of the game is to maximise use of each cell's 15 minute spawn timer.

I think strips might be the easier way to get others to organise that. With just a 1 wide conveyor every 16 tiles. (Thinking of one spawn type here to start with.)

But, come to think of it, the shape of the spawn surface area used shouldn't really matter...

Like, it could be a small square in the middle of each cell. And the algorithm will still find it every time.

As long as it has enough tiles to use up all the spawn-chance. Which seemed to be less than 16 tiles. Maybe 10, or 8 even. At which point, you'd be reducing down from 256 spawn tiles per cell (e.g. @solemn saffron's mega farm), down to, say 8. A factor of 32 efficiency gain! (Or 16 with 16 tiles.) Pretty huge.

[floral tendon]

it's only efficient in the sense of spawning items needed right

still takes up the same amount of space

[trim kestrel]

Yeah. Like an atom is mostly empty space. You could have something else in between bits, like a wood farm...

A lot quicker to set up and less laggy, too. So you could plausibly make your farm bigger than 200 tiles radius...

... That is, if there is some kind of 1-timer duration catch-up (?!) that makes that somewhat worth while...

Like with roots, where they all harvest when you come back...

[trim kestrel]

Aww, pants... 😶 My spawn range test may be an unsafe conclusion too. At least I'll need to re-run it again. The cut-off happens to aligns with a cell border.

Might we be about to see a rare win for team chunk loading...?! 😬😅

This was my initial thinking. With the curve accounting for the partial number of mobs in the last slot.

[inland compass]

Time to redo everything

Lol

[trim kestrel]

Literally! 😐 Like, this much higher rate for larva in clay caves (which sounds logical)... Look at the near perfect 4-way split on that chamber! 🤦‍♂️

But I might be able to just straight up calculate the theoretical rates. After establishing validity for doing that with a couple tests for each.

[crisp goblet]

What about your 'stripe' tests? Are they perhaps related to these grids?

[trim kestrel]

Striping different types of spawn surfaces alternately? I think that makes use of the second part of the mob spawning algorithm. Which is the Next Thing (TM) to test and try to optimise.

[inland compass]

wonder how well this will work

So far it is churning quite well lol

[rich belfry]

how big is each farm there?

[inland compass]

10x10

Urban moss pumped to shore with like 6 crabs on shore

[inland compass]

not a bad haul for an hour

[inland compass]

only problem with doing this in the sea is tentacles can steal the spawn

slightly upgraded for a better kill box. all the crabs get funneled towards the opening, and there isn't a direct line of sight down the conveyor to allow the scholars to pick off the crabs

[trim kestrel]

Yrah, in theory you should get about as many spawns from that as if you covered those 8 spawn cells entirely with 10 times more moss. Quite an efficiency gain. 🙂

[trim kestrel]

Hmm, yeah, I'd guess that is what will happen. Unless the tentacle spawn lands so close to the chamber as to also trigger a spawn in there...

I was wondering, last night, what would happen with just one tile row of beach sand, in each Sea cell. (With conveyors for collection.) 🤔

Like, does the algorithm make a very low probability roll for every single tile of sand, in a 16x16 cell. And still spawn no crab sometimes?

Or does it cap out after a certain number of tile rolls, for efficiency...?

I've had a few crabs spawn on 1 tile wide boxes fir my tentacle soawn-rate measurement. But it's not guaranteed, even after an hour or two.

[hoary osprey]

I do not understand the spawn grid.
Is that I should only put the spawn tile on the grid and place killing zone / connecting device at other place?
If I place other type of spawn tile not on the grid too, will the other type of mob spawn or not?

[trim kestrel]

Each square has it's own (15 minute) spawn timer. If your spawn area is only in one square, it gets only one spawn event per 15 minutes.

Whereas, if it overlaps enough of 2 squares, it can get twice as many spawns. Or 4 squares, on a corner, it can get 4 times as many spawns.

Other spawn surfaces in the same square can take a spawn event, though.

[trim kestrel]

Of course, one could also catch all the spawn events on a shelf beneath your kill area, to prevent any purple slimes spawning (rarely) on the poison, and breaking automation (that cavelings don't).

[hoary osprey]

Thank you for explaining.

[inland compass]

moved my farm to a fresh world. i couldn't even afk in my old world, i would crash after afking an hour or so.

[trim kestrel]

Ha. So maybe your minecart/mining AFK crashes, weren't related to the activity?

[inland compass]

well the mining i started in a new world and still crashed

so far so good in this new world

wonder if bubbles cause issue, i had like 10-12 crabs when i restarted this morning lol

[inland compass]

still going. Soooo many moldy boys

a 4 corner mold spawn, and 2 4-corner urban moss spawns.

[crisp goblet]

I'm not sure that I follow your reasoning re: the secondary mob spawning, have you been able to replicate this?

[trim kestrel]

I've thought about making an overrun core decorative build, packed with larva, but infected would do. Maybe for halloween...

[trim kestrel]

Yes. Further up, I talked about watching orange slimes always spawning on the 16 tile lines. And, half the time, a red slime spawning simultaneously with them. But always a good distance away: #977128133142659102 message

Seen with mushrooms as primary and shroomen as secondary. And inferred with scholars + sentries (there) and caveling miners + brutes, from weeks ago confusing findings.

One side experiment, today, I was trying to figure out how far away the secondaries need to land. Didn't get it yet, will need more focused testing.

[trim kestrel]

4 corner mold kinda scary at up to 16 infected per 15 minutes. Assuming that doesn't need too many mold tiles to ensure max of 4 each time. (Again, more testing.)

Btw, I tried alternating tiles of mold, fungal, slime and stone moss. I saw the first three sometimes spawn together, in a little room. So the mobs don't even have to be allied, to share a spawn event. (Cavelings might have too, but left no drops.)

[inland compass]

i mostly see only 1 guy spawn at a time

i wonder if my squares are too small

should be about a 5x5 in each cell

[inland compass]

what would be interesting, if i expand the room out, and then just put 4 dots of spawn surface

in each extended part, that way it would check to see if it looks at all possible spawn points in the cell to generate

[trim kestrel]

This could be a limitation on optimising the spawn-rate, like this, yeah. We need to figure out if

(a) The secondary spawns are forces to be over a certain distance away (but still inside the same spawn cell, I think).

(b) ... Or do they require just a certain quantity of additional spawn surface? That can't be cheese as much.

[trim kestrel]

And here it is; can I get an 🇫 for my 205 tile radius theory, please...? 😳 😢

Looks like it might actually be more of a 12 spawn cell radius. Or some intersection between absolute distance and the cells. That would take me a long time to pin down exactly. Given that each measurement takes ~2h of afk after setting up the build.

I don't know how that one scholar got in the middle slot, there. Maybe in the seconds while I was leaving...

Or the one that teleported into the closed area above. Mabe both teleports. Which muddies things.

But the mushroom chambers above and below line up perfectly with the main 3 columns and the cell borders. When 205 tile radius would have been 4-5 tiles to the right..

[trim kestrel]

And I'm more unsure what's happening with spike traps range; when I TP'd back to check (from 200 tiles away AFK for 2 hours) they activated to kill (at least) 3 accumulated Shroomen. But it seemed like all the mushrooms had already, previously been broken.

So maybe there's a 105 tile range but also, certain events can a spike trap. Or the whole cell gets intermittently reloaded...? 🤷

[inland compass]

seems to be working well. lush on the bottom, urban on top

i noticed lush mobs seem to spawn in pairs, gardener and hunter

the strip seems to be a better way to spawn them so far

less wandering for them to do, gets them to the kill box quicker

[glass night]

Is there a way to get a robot arm to only pick up one specific item?

[inland compass]

not really

there are sorting builds out there for stackable items

[long dagger]

our best option wouldve been pedestals for fewer items but they removed the mechanical arm interaction with it, so far it doesnt seem like we can filter items in other ways without using the arm, chest, and drill. would be nice to find a way to take a stack out of the chest but then filter back a stack of 1 of the same item that was removed to keep the filter in place

simple equipment sorter but u cant fully separate equipment from other items in a way to store it in chests instead of the repair stations

havent played much for the past 3 days, i cant get motivated to work on any big builds atm

[bitter turret]

someone needs to make a turing machine in core keeper fr

[long dagger]

turing?

[bitter turret]

yes

alan turing

turing machine

[long dagger]

havent heard of that before

[bitter turret]

https://en.m.wikipedia.org/wiki/Turing_machine

[bitter turret]

not a CS student i see

there is also the term of "turing completeness" https://en.m.wikipedia.org/wiki/Turing_completeness

[bitter turret]

is there a way to implement state with what we currently have in the electronics table?

[floral tendon]

yeah

but not... in any practical way lol

a feedback loop using a logic circuit can serve as a latch to hold a high value

but you'd need an extra piece to break the loop to bring it back to low

and whether or not you can actually use the stored value in any way... no idea

[long dagger]

i find it confusing, cant find a vid that explains it well enough for me. based off what i saw and if im not mistaken, it could be possible with just 1s and 0s to some extent but then there was some other stuff there that i couldnt understand with letters getting involved.

https://tenor.com/view/idiot-elmo-im-stupid-shrug-im-an-idiot-gif-16140879

[shadow grove]

will the enemies spawn on any ground in sunken sea? or do you have used a specific ground?

[long dagger]

I haven't seen anyone showing off crab spawns in the cities. Dunno if he mentioned about the ground type needed for crabs

[trim kestrel]

Has to be beach sand, for crab spawns in sunken sea.

Not clay, even though it looks very silimar in a storage chest.... 😳

[trim kestrel]

Always have done, yeah. 😉 Except for sometimes in my Wilderness grass native spawn-rate 1k chamber, where I now think some of the pairs got split by the bottom wall, one half outside.

[trim kestrel]

Is that horizontal stripes, 1 tile thick?

[inland compass]

Yup,. 1 tile width horizontal

[trim kestrel]

Have you seen it produce any sentries?

[inland compass]

the strips are lush

urban is square, and when i enlarged the room and peppered a few squares away from the main square, i didn't see any real increase (wasn't measuring)

scholars still seemed to mostly spawn as single mobs

[trim kestrel]

I don't think it's currently possible in the current wiring system, to make a general purpose computer. Because wires can't cross at all.

I mean, I made a binary counter. Which is a register/accumulator. With a reset function. https://youtu.be/k6DIZvPfRys

I guess if you were happy to reset by counting up to 0 (or flipping each bit manually in an expanded layout) then you could add in numbers from a second register.

But I'm not sure that you could multiply/divide, like I showed in my Terraria "Digi-Comp 2" (hoiktronic) build: https://youtu.be/MpZBsTTdXUc

And I've not found an alternative, like in that, like moving player or mobs through rail/conveyor/corridors to switch states in the environment.

But anyway, to be a Turing machine, you would need the memory to be extendable without obvious limit.

So if there is a design to fit that bill, in Core Keeper currently, it would probably look more like a 1 dimensional cellular automata. Like from Stephen Wolframs "A New Kind of Science". That evolves states as the input propagates down adjacent the rows, into an enormous expanse.

But I don't think even that will be possible, currently. (My brain too foggy to think it through clearly, these days.)

[trim kestrel]

Turing machines are a fundamentally unintuitive concept.

It's saying that once you reach a certain, very minimal, level of computational complexity, that basic computer can do anything that any other computer can do. However much more complex you make any future computer.

The catch is that it may take exponentially more time and memory for a simple machine to do those more complex calculations. Such that it becomes impractically slow almost immediately.

The letters you mention are probably the "states" of the machine, behind the read/right head.

It's not too important what those are, exactly. They are just a limited memory for the machine to look-up and remember what action it is currently taking. At a basic level, if it's moving the tape a certain distance left or right, reading a bit or writing a 0 or 1.

This Minecraft build kinda shows an example of this, with a (blue) array of look-up wiring that he turns around to show. (Criss-crossing wires, like what we can't have in CK.) https://youtu.be/1X21HQphy6I

In principle, if a human brain operates deterministically, at some level, then a magical demon with a calculator, and an unlimited amount of paper for writing down their working, could compute any human thought.

All the actual 'thinking' happens in the change of state of the program data on the tape, though. The states in the Turing machine, itself, are dumb. As dumb at the physics operating in the synapses of your brain. It's just rules for how to turn the handle to evolve one pattern of information into the next pattern.

[crisp goblet]

Does this hold true if neurons use, in part, probabilistic quantum effects?

[trim kestrel]

My feeling is that quantum computation will found to be, at some level, deterministic. That quantum computers can take shortcuts, but aren't somehow magical.

And quantum processes not truly random, from the right frame of reference. Physics has always been a series of less wrong approximations.

[crisp goblet]

I am by no means any kind of quantum scientist, but haven't some experiments shown that, even if quantum processes are at some level deterministic, that they are fundamentally uncomputable?

[trim kestrel]

I mean, quantum uncertainty means that an observer can't know a particle's exact position and velocity, because any process of pinning one of those down changes the other in an unpredictable way.

But the interpretation of quantum phenomena is very abstract and nuanced. I don't think it's been definitely proven that anything is fundamentally incomputable. To me, that concept, itself, sounds like nonsense. 🤷

[crisp goblet]

One must certainly be skeptical of Quantum Woo

[sly sable]

End-user-level question: how many spawn tiles would make an entertaining farm of mutually-hostile mobs in which there is almost always at least one of each type alive slugging it out, or wandering around waiting to find each other and start slugging it out?

I would like to stroll past fields of ongoing gladiatorial carnage

[trim kestrel]

Well, if you have, say, one mob spawned per 15 minutes, per spawn cell (16x16 tiles). You could start working out how many you'd need for a fight every 10 seconds, say...

And there are 12 cells within your the spawn radius. Which gives an upper limit.

[crisp goblet]

Any tests on the spawn radius shape? Circular/square/taxicab?

[sly sable]

How are the spawn cells checked? Because I built a silly "prison" of 3x3 spawn tiles, 4 types all adjacent to each other with walls between them, and sure enough there are cavelings inside. Now I don't know how long they took to fully populate but I did watch two different types pop into existence simultaneously.

[trim kestrel]

Circular. Roughly. But as I was saying yesterday, it's not a simple 205 tile radius, as I previously told everyone. It's actually something like 12 (16x16) spawns cells radius.

[crisp goblet]

So a jagged circle of spawn cells?

[trim kestrel]

Probably. Would take a lot more long tests to be sure.

[crisp goblet]

Have you tested whether it's indeed 12 spawn cells, or 3 chunks? That would dramatically change the shape of the 'circle'

[trim kestrel]

I was figuring this out and explaining what I'd found in detail a couple nights ago.

A spawn event starts looking through all the tiles in its cell from the bottom left. Going right first, then looping back up to the next row.

If it finds a spawn surface tile I think it rolls a number to see if it spawns something, with weight based on that spawn tile type.

Then, it may spawn a secondary mob (or groups of mobs) depending on what and where there are other spawn surface tiles further on in its search.

I don't know the criteria for this, yet. But you can certainly spawn at least 3 different biomes of mobs from one event (simultaneously).

[trim kestrel]

It's not 64x64 chunks. Definitely more fine-grained than that.

[trim kestrel]

Did you see KanSer's mega-farm, above? And the gif of the mob kill area in action? for an idea of (approaching) maximal carnage rate: #977128133142659102 message

I think we could do 2-3 times more than this, using striped spawn surfaces. Probably with like 1/4 of the spawn surface, too.

Maybe less.

[sly sable]

Oh my god, that's glorious.

[crisp goblet]

Honed in on any minimum spawn surface requirements?

[trim kestrel]

Well, with fungal soil, all the mushroom tend to land within 16 tiles, so all on the first row. Hence the lines of drops I was observing, above.

But the shrooman may soometimes land a lot higher up. And I'm not certain of the mechanics behind this.

With stone moss cavelings, 90% landed on my first strip of 8 tiles. But wasn't getting any secondary spawns on the two rows about that, as I had hoped.

.
It seemed like horizontal stripes gave me only cavelings, before. While vertical stipes gave me a higher propotion of shaman and brutes, but fewer mobs overall.

And I need to reexplore this, with the new spawn cell theory in mind.

[sly sable]

I see now the merits of fussing about with conveyor belts, even without a spike trap kill plan. It delivers larger numbers of mobs to each other than would spawn wild in a visible space.

[trim kestrel]

Conveyors take things places. 😜

[sly sable]

Are brutes slap-happy even without being aggroed? Or will they ride a conveyor belt quietly to an arena full of crabs and tentacles?

[trim kestrel]

No random attacks or trying to swat anything over 3 tiles from them.

They'll even let you drill them, unlike slimes.

[sly sable]

I guess if they spawn and ride in peace I can still use the moss so they don't need to be a marine offshore farm

[trim kestrel]

Hmm, I finally retested my old Ghrom one-tick farm. And, indeed, it's very useless now...

Even maximising spike surface area, he still slides out of there before the second spike activation (2.5 seconds later).

It's funny watching him mow down slimes on his slime ring... Like, the circle of life, so beautiful... 😵‍💫 😅

Do players always find a whole stack of slime on the track, when they first get there? Seems weird and I don't remember that.

[bitter turret]

turns out i briefly converted Core Keeper University to Computer Science University 😅

[trim kestrel]

I just made a simple observation, but with somewhat significant implications for mega-farm limits...

After being away from this area for well over 15 minutes (over 200 tiles away, just about) there were no immediate spawns in any of the 4 spawn cells around me. They continued over the next 10 minutes or so, at staggered timings.

This means that their timers don't increment while you're far away. So there's no catch-up when you return.

Which means that you couldn't, for example, alternate between two mob farms locations (over 200 tiles apart), to trigger all their spawns every 15 minutes. One farm of (approx.) 200 tiles radius is all you get. (Or 12 spawn cells, whatever it is.)

[long dagger]

Just woke up, I'll look at the vid in a bit

[long dagger]

R u willing to try this once more but by teleporting away quickly to c if the result differs

[sly sable]

Oh that makes sense
And explains why my first passive farms produced so little slime and larva meat

[trim kestrel]

Why would that maybe help? Or hinder? Can one complete a teleport before boss spawns in?

[long dagger]

When it was possible to kill it in 2 spike hits without it destroying the spikes, u had to be a few tiles away from it to make sure it wouldn't destroy anything. I'm thinking the range of that may have been increased to prevent builds like that from working up close. It could be possible they just changed it to destroy them no matter how far the player is like the wires having no limit

I don't have much hope on the result but it'd be interesting if it works and doesn't destroy the spikes

[trim kestrel]

There's no way anything's going to help unless it allows the spikes activate faster, or causes Ghorm to take longer to recover from enraging and start moving. Or the spikes magically do more damage per hit.

If there is time to TP far away, this could be a method to test spike load-in distance, though.

[long dagger]

I was just hoping it'd prevent it from destroying the spikes by being further away if they just increased the range that the player had to be unless if they reworked that part which would make the test worthless

[trim kestrel]

It's just down to timing. If he moves slower, then maybe.

[long dagger]

This video explains it better for me, as for how it'd be possible based off my dumb guesses. The belt would have to be stationary and small but as for the belt reader I got no clue how thatd go back and forth while skipping tiles and being able to come back. Idk if u could use an item as the bit but it's still be extremely limited and prob wouldn't be considered as a Turing machine. As for the changing between 2 bits, it may go off as an endless loop if a small design is even made to work. That's just what I think

Terraria may be 2d but atleast it's got multiple tools and wires that let it overlap which should make it possible, as for Mc I'm surprised someone made it with just that 10 years ago, I was expecting a design using command blocks to cheese it

[trim kestrel]

Here's 4 different biome mobs all spawning with the same event in a single cell. (Like with the 3 caveling biomes, above.)

It's kinda like watching a cloud chamber and seeing the wake of a high energy charged particle having zipped through... 🤔

Anyway, it makes me think you you might be able to get all mob types spawning simultaneously of a single event. By patterning their surfaces together. 🙂

Even if just one of each is spawned, that could still be pretty efficient, if it doesn't take too many tiles of space, in total.

[sly sable]

Yes, yes! That's the sort of thing I was thinking of! Brutes complicate things but they could be imported.

[trim kestrel]

I saw my first secondary moss caveling spawn on these 3 separated rows of 7 stone moss, too (after about 12 spawn events):

I think the algorithm does a primary spawn with high chance, then the secondary with much lower chance, per tile, scanning across then up.

I don't know if a secondary or one type precludes the chance of a secondary of another (biome) type. Or if that ends the spawn event and prevents any more primaries, even.

[crisp goblet]

So, you figure it may be possible.. though unlikely.. for a secondary mob to spawn low within the cell?

[inland compass]

I wonder if it is a mold cluster spawn that just picks the other biome tiles around it

[trim kestrel]

Going in tile search order, that was a stone moss spawn that got picked first.

But yeah, I did initially think that one mob getting spawned triggers the rest, too. But that was my thinking before spawn cells.

[trim kestrel]

A test for cross-biome cluster spawning might be trying to force a crab spawn, off the back of an mold tile (or something else).

[inland compass]

ooo

[trim kestrel]

Or just getting more mob types to spawn simultaneously, as a counter indicator. I mean, i had 5 entities spawn there, but you might count the 2 mushrooms as one, because they always spawn together.

[rich belfry]

I finally finished the test farm. the mold tentacles are still building along the smaller kill zones, but the main kill zone is up and running. Im going to see what I get after a couple hours.

[inland compass]

Too close together:(

[rich belfry]

each farm is in between 4 spawn chunks. anc the kill zones are all in at least 1 or 2 spawn zones of their own

[inland compass]

If you look, purple and fungus share 2 cells

[crisp goblet]

I don't think that should matter though....

[inland compass]

Depends what the goal is

[rich belfry]

I see what you mean Craigins.

[inland compass]

Each cell spawns once every 15 min

[rich belfry]

the stone and urban moss share some cells too. as do lush moss and eggs

[inland compass]

Yup

[crisp goblet]

Each cell gets one event, but couldn't that event be split between the spawn surfaces?

[inland compass]

So really building at the 4 corners is mainly to reduce the required building materials

[rich belfry]

I'll probably keep this design for now because I just dont have it in me to rebuild. but I can move some things around to ensure they're all in their own 4 cells each

I already have some of it built. I just need to move the rooms

I figured the brutes, large larva and sentries could benefit from a death line of tencticles before they even reached the main kill room

try to mitigate destruction

[inland compass]

Yeah i need to set up stone still in my new world

[rich belfry]

I had to compact this a bit because directly north of the eggs and lush moss is Glurch. and directly south of the bottom 2 zones is my main rail line...which I can move, I just dont want to

I dont think it matters, but I made my walls out of city walls to try to minimize larva destruction there too

[inland compass]

Brutes can pop chrysalis too

[rich belfry]

really? maybe I run the brutes through the crysalis room before they run through the tentacle gauntlet

I put in my own entrance to that room and the mushroom room so I could go in there and harvest them

[bitter turret]

@trim kestrel my wood farm (following your steam guide) has like 2 holes in it in terms of wood

should i be worried

i just messed up badly when initially growing it out

just those 2 wood pieces missing

wait wot

why didn't it embed

here

[inland compass]

yeah they won't grow together

what's the question

[bitter turret]

will it impact the growth/efficiency of the farm + is there a way to fill in those gaps without nuking the whole farm

i spent way too much time on my coral wood farm and i somehow got it to be complete

[inland compass]

don't have to nuke the entire farm, just the whole right side

[bitter turret]

like

the right-most wood column

?

[inland compass]

delete the red, plant at green

let it grow out how you want it, so you'll need to dig up the root areas on the right

[bitter turret]

i am never extending either of these farms

i spent too much time getting them to grow

[inland compass]

i mean they grow pretty quick

the initial first row might take a bit

but as soon as you start adding additional roots it speeds up

[trim kestrel]

Technically, you don't even have to replant the starter root.

[bitter turret]

it took me at least an hour to grow my coral farm into its final form

well

i am stupid

[inland compass]

true, just not sure how far it can grow without it

[bitter turret]

but

[inland compass]

if there is a distance

so look at this, as soon as the blue grows, plant the green

don't wait for the whole thing to fill out

[trim kestrel]

And if you want to baby-sit the regrowth, you don't have to dig up the existing roots, either. Just break any wood that forms not attached to the main branch.

[inland compass]

ah wood will grow next to an adjacent root?

[trim kestrel]

Yes. A root stump.

[bitter turret]

forget it

i have 5 stacks and going of wood

i really don't think wasting more time on it would be wise considering how much trouble i've had with it in the past

[inland compass]

i really don't understand the point of them

it is to save drills, but drills are cheap

[trim kestrel]

I'd say shouldn't take more than 10 to 15 min. But not a massive deal for overall output rate. Just looks silly.

[trim kestrel]

Drills and space. Maybe fps... ,🤷

And to feel accomplished.

[inland compass]

i had like a 12 root farm, it was producing so slow, i just made 8 individual roots with 4 drills per

and it produced like 4x as much

[trim kestrel]

Lies. 😜

Depends on design of the shared root farm, I guess. But 1 stump can only utilise 2 drills, max. 3 and 4 are definitely a waste.

[inland compass]

yeah 4 is just convenience

[bitter turret]

should i just make my wood farm feed the wood into a saw

free money

maybe i should set another farm but for money

or i could just put wood into a saw every now and then

[trim kestrel]

Which might be ok if the spawn event continues to search to any distance, after triggering a spawn. As I'm suspecting.

In fact, the bigger waste of spawn rate will probably be sticking to one surface type per chamber, instead of stripping all the ones that combine.

Although, still gotta figure out what's needed to get those secondary spawning mobs. If one is farming for trophies or their unique items.

[rich belfry]

Ive gone under the assumption that ground+moss has an effect on that but have done no testing. Though anecdotally, I didnt notice any brutes in my stone moss farm until I put stone ground under it.

[trim kestrel]

The mold kill zones look too close the the through conveyor to avoid brutes attacking and breaking it. And too short in length to kill things for sure, if not...

[rich belfry]

don't need to kill them. the conveyers are also coated in purple slime so their hp should be low enough that by the time they get to the main kill zone they die fast and easy

I hope

we'll find out

😛

my last farm was a total mess. brutes smashed everything.

[trim kestrel]

Nah. What's under the moss is always irrelevant. Brutes are just rare enough to confuse things and lead to false assumptions.

[rich belfry]

thats reasonable

[rich belfry]

though I am concerned about brutes smashing things. I have many notes already for changes to make in version 2. and maybe making things wider is part of it

[trim kestrel]

If you are in a normal mode world (?) just feed the cavelings to a line of spike traps with poison (and arms at the side for loot).

I have a slightly more complex design in mind for trying to deal with when the arm's hit box stops existing, while the mobs still do...

[floral tendon]

do spikes not make them angry

[trim kestrel]

No. Nor drills. So you can kill and collect from cavelings easier than slimes, ironically. (In normal mode.)

[floral tendon]

huh didn't know that, maybe i'll just use spikes then

i always thought it was by %hp or something

[trim kestrel]

Ok, so, got all the loot drops pasted in, but a whole load of the middle section needs experimental discovery, still... 😵‍💫

[trim kestrel]

I've just posted all 3 pocket size (7x9) rainbow mob battles I recorded from one round of spawn events, here: #📷-ck-media message

Here's one taster.

In all three, each of the three types of slime spawned, and each of the 3 cavelings (including gardener & hunter) and infected, plus two muchrooms. Larva missed in one case and shrooman in one too. Still that about 9 mobs per throw. Pretty productive.

And the interval of spawn events in this case was closer to 20 minutes, I think. So that's going to take a bunch more observations. That would still be about 27 mobs per hour. So a rate of ~43 mobs per hour per 100 tiles. WAY faster than anything I'd measured before.

Not secondary mobs seen at all though (big larva, cocoons, mold tentacles, shaman, brutes, sentries). I think that will require a bit more of each spawn surface for (at least double this 7 tiles). But more observations needed for that.

Oh yeah, maptool view... And that little mushroom farm seems to work ok. Picked up most of the first round of spawn events in 4 cells.

[inland compass]

So you got 9 mobs in one spawn tick?

[rich belfry]

today I learned that conveyer belts cover up purple slime. you dont get the debuff. I somehow never noticed that before

[rich belfry]

whats the arm contraption do?

[inland compass]

Mushroom picker

Spawn on the spikes, spikes pop them, arms place them on conveyors to store

[rich belfry]

nice

[bitter turret]

what do you do when the chest runs out of space

do you just place another wire and another arm and another chest

and so on and so forth

robot arms are very much like minecraft hoppers

[pulsar anvil]

ever improving guides!

[crisp goblet]

So no noticeable difference between the patterns?

[trim kestrel]

Yuppers. Pretty sure that inter-biome cluster spawns aren't a thing. RIP idea of forcing crabs.

[trim kestrel]

Not really. Well spotted.

I think the spawn algorithm treats each spawn surface independently. All that matters is the number of spawn surface tiles in a spawn-cell. And even then, only because that gives more rolls for triggering a spawn.

Also, the number of mobs of that type in the cell, which block more of them spawning at a certain number. Or, might just make the odds longer, to start with.

[inland compass]

so mixing even seems to increase the original spawn chance

i was getting 1 scholar per spawn with just urban moss, rarely getting 2 or 1+sentry

i dropped lush and orange slime and i'm regularly getting 2-3 scholars as well as 2-3 slimes and a gardner+hunter

[trim kestrel]

Can you screenshot your spawn chamber for clarity? 🙂

[trim kestrel]

I think this is accurate now...:

And BTW, the tiles that always block mobs spawning also block merchants from moving into a hour on their tile. 😶

[trim kestrel]

I recorded 3.5 hours of these rainbow chambers and noted down the times of all the spawn events to calculate their intervals: they were all 21 min 57 seconds (+- 2 seconds)!

But the timings in my new test world were very close to 15 minutes, also near the core. So the question is why those were faster?

Throttling due to processing burden? (FPS was at ~60 as normal.)

Scaled by number of spawn cells within range? (I hadn't spawned terrain all the way out to 200 tiles in every direction of my new world.)

Throttled by number of total mobs in spawn range, or world, like a mob cap equivalent?

[trim kestrel]

Actually. It was a minor thing, so anecdotal, but the mushrooms didn't seem to like the vertical spawn strips as much. Never got 2 simultaneously on that. Horizontal and diagonal got more overall, too.

But did get one or two shroomen on the vertical. And my goodness, they're amusing when they try to participate in fights! 😍 😅

Other observations:

(1) I never saw a second mob of the same type spawn into a chamber where there was already one still alive (from the last round). Except once with infected.

(2) One time there was a Caveling (miner) spawned (in the horizontal striped chamber) right in front of Shaman, too (a secondary).

But I'm thinking that we could bias, maybe even force spawn types onto the secondary mob types, by having some of the base type already in the spawn cell... [Edited]

(3) Never got a big larva or cocoon, probably because the little larva always died.

(4) I built bigger stone moss spawners above the top two chambers, to see if they would pick up only secondary spawns.

I messed up, forgetting to put any stone moss in the diagonal grid and had the collector conveyor one tile over into the right hand spawn cell.

However, the accumulation of mobs still blocked the spawns, so I think that mobs might register as more than one tile wide, being counted in both cells simultaneously.

(5) No cavelings (stayed) spawned in the top right cell (but were soon blocked by accumulated) spawns.

However, I thought I'd seen 1 frame of a caveling appearing in it, then getting pulled away to the bigger area at the top, where a shaman also spawned. Video was only 30fps, so could have missed this. (Or I was mistaken, only half watching.)

The last caveling to spawn was a brute, after there were already 4 miners and 3 shaman. Also, hence thinking about forcing spawn types.

[hoary osprey]

How you prevent the brute to spawn?

[trim kestrel]

I didn't, on purpose, but I think they may require another caveling to exist in the cell, in order to spawn. As per (3), above.

Either from the same spawn event (making the brute a secondary spawn), or already existing from a previous time.

.
So, if there's only a slim bit of (horizontal) stone moss, and you always clear it out, you may not get any brutes.

[hoary osprey]

But having 1/3 spawn as caveling should make the caveling stay in the arena?

[trim kestrel]

Or, maybe if you have a brute already in that spawn cell, that might block further brutes spawning in it.

It would take a long duration experiment to give this statistical credibility (unless it just disproved it).

[trim kestrel]

Not sure what you mean...?

[inland compass]

here you go!
conveyors were already down so i just randomly placed ground covering where i could

[trim kestrel]

Grey square, gottit... Thanks... 🙄😅

[trim kestrel]

I found the actual (terrain generation) spawn chances for boulders, in a game asset file, so here's a definite edition of where one can find each boulders, and with what (relative) probability:

[bitter turret]

is just covering this place in poisonous slime good enough

[unborn summit]

yes

and add some spike if you want

[rich belfry]

ya thats enough to make spawns happen. if you want to kill the mobs automatically and its normal, add spikes. I like to add conveyer belts to force all the mobs to a single spot where the spikes kill them

[rich belfry]

mob spawn rate is 1 per type square every 20 minutes?

in general

[trim kestrel]

It'll spawn some purple slimes, of course. But if you've been reading some of my above findings, you'll know that's the most inefficient way to spawn mobs.

Better to check the "mob grid" in map tool and have like just 10 ground slime per grid cell.

Then to increase output, spread your farm out as much as possible, repeating this in differetn cells. Adding each new spawn surface in adjacent strips in all the cells. As each spawns independently of the other types.

[trim kestrel]

I've seen spawn events every 15 minutes exactly, in a new world and every 22 minutes, exactly in an old world. Both near the core.

[rich belfry]

cool

Im collecting a lot of the info you've been figuring out and want to make sure Im accurate

[trim kestrel]

Needs more research to be sure of generalisations... 😮‍💨

[rich belfry]

sure

[rich belfry]

if I said mobs dont spawn closer than 25 tiles from the Player character, and more than ~200 tiles away from the PC, would that be accurate?

[trim kestrel]

The exclusion zone in only 6 tiles away.

And the max spawn range is only approximately 200 tiles, because it looks to be based on a number of 16x16 spawn cells. Possibly 12. But not sure where the transition happens with respect to where the player is within a cell.

[rich belfry]

cool. I had seen some of what you said before on the end tiles like 205 at first, then you tested and found it was a bit less, but varies

[inland compass]

so 6-7 to 192-208 seems to be the range

closest I've seen is 7 tiles personally, it was a 5 down and 5 over spawn on moss

[trim kestrel]

I'm pretty sure I've seen closer, like 5 tiles. But depends how one counts it. Peace showed some pretty close spawns a while back, too.

But even if less, 6 (or 7) would be a better rule of thumb for players to be sure they're not accidentally blocking spawns, I guess.

[inland compass]

right, mine was 5 tiles diagonally, which is a distance of 7

really sucks that we count in tiles but the game is distance based lol

[trim kestrel]

Pretty easy to test this range quite directly, now. Given that only 6 tiles of spawn surface needed to get almost guaranteed spawn(s) and can do several types simultaneously...

Be interesting to see if it just bins the mob or squeezes it out to elsewhere, if there's more spaces it could land.

[long dagger]

Have u tested patterns yet with the 6 tiles? Wether u can have 2 columns of 3 with a wide gap or something

[trim kestrel]

So, I checked the "MinimumTilesRequired" I found in game asests. That's 6 for everything except Infected and mushrooms, which are just 4.

So, bottom right, with 1 less than requirements, spawned nothing for many spawn cycles. While top right, with exactly the tiles required, turned on and spawn one of each thing every time, pretty much. Because the chance per tile is pretty high for the base level mobs.

The different stripe patterns had virtually no apparent effect, previously. Including splitting the minimums tile requirement either side of the chamber, for the diagonal stripes.

Interestingly, that left chamber had 1 too few stone moss, but 20 tile in an are just about (also within the same 16x16 spawn cell). And the first 2 cavelings still spawned in the chamber. So I guess it counts up all the tiles of that type before rolling spawn probabilities for each of them.

After those two, another miner and a shaman came from that top section. As if the first two miners may have been blocking more close by. But I'm about to try a experiment on whether that is proximity based, or count per spawn cell.

[trim kestrel]

Hmm, I just got a 15-16 minute spawn cycle in the outer reaches of my main world, where get 22 minute cycles at the core.

So, I think it may well get scaled slower when more of the world around you exists. Which is a bit awkward.

Also, a entire 16x16 spawn cell filled with (over 250 tiles of Chrysalis) still only produced 1 of each mob type. As expected from the limits I've seen in game assets.

Next to find out what the minimum is to roughly grantee each mob type, each cycle. And see if that can all be crammed into a single cell, for every type together, to absolutely optimise spawn rates.

[trim kestrel]

Yeah, 15 minutes 25 seconds every time, over 5 cycles. I think 15 minutes will be the quickest you can get. But it will take a lot more experiments to see how much map you can make use of while still getting that rate; trade-offs.

It looks bad for getting guaranteed spawns of every type. Even with half a cell of chrysalis I tend to get only big larva or cocoon. Never both. And half a cell of stone moss has given me no brutes and a shaman half the time.

I suppose I should graph the odds for each, by number of tiles...

[trim kestrel]

I'm now pretty confident that mobs block/limit spawns only in the cell they are in. Not by radial proximity. (The cell border preventing spawns from the left chamber limiting further Larva spawns.)

And, I think, only of their specific mob. E.g. the cocoon spawned on the right after there were already some larva in the right hand cell.

Oh, and I happened to notice that Brutes can't break Stone Moss ground tiles. Which I guess is sensible. (They do definitely break e.g. Chrysalis, though.)

[bitter turret]

i want to dupe one on a test character to see how big character HP could get

[shadow grove]

will all these kind of spawn information saved in a place? sometimes its really difficult to follow al conversation and what kind of farm has the most effectiv spawn rates at end of the day.

i noted that a new wiki exist with some nice knowledge

[bitter turret]

what's that wiki called

[shadow grove]

https://corekeeper.wikizet.com/en/Core_Keeper_Wiki

[bitter turret]

ohhhh

yeah j read some articles from it

it seems more

technical

than the fandom one

anyways i am currently eating duped mushrooms

just curious if there's a limit to HP

[shadow grove]

there is no limit i think so

did you do the dupe via charakter clone?

[bitter turret]

nah

world clone

lmao

[inland compass]

takes longer to eat them than it does to dupe them

[bitter turret]

unfortunately

[lilac gale]

This wiki has been around forever.

[trim kestrel]

Zoomer "forever"? About 5 months for the main page. 😜

The [wikizet's] creator/owner/admin/editor, Cysidus, says the Fandom got a headstart (main page created 8 months back). Hence why the wikizet's less known, lower ranked in Google.

Where'd you just come across it, @shadow grove? Out of curiousity.

[lilac gale]

I mean, in all aspects of the game being released on Steam it has been around. Hence the term used, forever.

It may be a couple days after the actual release, but it's been out basically since the game was launched on the 8th

The fandom site has existed longer likely because the actual creator didn't get into the earlier testing periods

or just had no thought to make one until later

either way, that's true that the fandom has been about longer

[trim kestrel]

I'm kinda dumping my research notes in here as I go. For my own clarity and so others can follow along (or participate) if they want to experience every twist and turn with me.

But I'm getting close to having the overall shape of the spawn system sufficiently nailed down that I'll be able to make a first draft guide soon(TM). 🤞

[hoary osprey]

The fandom is just not updated, but the wikizet is also not contain all the information of the game...

[trim kestrel]

There's a difference between information and knowledge, though. And looking at their respective cooking pages is a very good example of this:

Fandom attempted to list every single procedurally generated dish. With a picture for each! But doesn't really tell you anything about how cooking works.

While the wikizet lists the resultant buffs for each ingredient and the relevant skills. In a much more compact fashion. 🙂

[floral tendon]

so in other words, neither is good

i think so far the only useful info I've managed to get from either wiki is what drops what

[inland compass]

i prefer wikizet because i'm mostly looking up numbers. Fandom seems to have more qualitative information.

[trim kestrel]

Well, it depends what level of game experience one is coming in at, how much hand-holding you want, etc, as to what's going to be good, I suppose.

But, flipping it around, what have you tried to look up and failed to find (on either)?

[floral tendon]

it was a while ago, but mostly stuff related to food

i was trying to find all the different kinds of sushi

actually i did use fandom for stats, that was useful

[trim kestrel]

Hmm... I suppose if one want's to do a reverse look-up on a dish you previously made, or were shopping for some specific aesthetic, then the full food list could be useful... It's so sooo unwieldy to deal with, though. (Hence my guide, etc.)

[hoary osprey]

Most people do not need a full combination table of the cooking effect, but wikizet not even have a combined list of all cooking effect or combined list of effect of all armor / accessory like fandom.

Make it hard to try to reach a certain special stay, like 90%dodge or maximizing the mining damage with bare hand.

[trim kestrel]

wikizet not even have a combined list of all cooking effect

What would this look, like, I'm not sure I understand...

This table shows you everything you need to know about what buffs any food will give:

[trim kestrel]

And aren't these "accessories" pages almost exactly equivalent? Is there something I'm missing? 😵‍💫
https://corekeeper.wikizet.com/en/Accessories
https://core-keeper.fandom.com/wiki/Accessories

[serene kraken]

Does anyone know if poison slime/acid slime affects Ghorm?

If so.. it makes me think just maybe it could be possible with traps for a one-tick farm

Probably still not but maybeeee

@trim kestrel Ur the man to know this sorta thing hahaha

[serene kraken]

I think I actually prefer wikizet honestly, darkmode for one always a plus and just feels a bit easier to navigate in my personal opinion.

[trim kestrel]

Ghorm's a very big Larva and they're all immune to acid. Sorry.

[hoary osprey]

Sorry for not finding the page when it is actually exist

[serene kraken]

Yeah I thought so just happened to not use any slime on him ever and thought maybe..

That sucks man

My hopes and dreams of your one tick ghorm farm are far gone

[hoary osprey]

And I have wasted so much effort on recording the cooking effect of the epic food...

[serene kraken]

Lmao..

All recorded on the internet for free use already for you

The internet can be a great place at times

[trim kestrel]

No worries; I am the king of failing to find things (even stuff I know exists).

And wasting time in-game... Yeah, couple weeks, just now, recording nonesense counts for mob spawn rates. 🤦‍♂️

[long dagger]

U be doing the majority of the work here

[hoary osprey]

Would you please add the Prefix and the Product for the ingredient of cookings?

Sorry I do not know the prefix generated from giant mushroom

[long dagger]

Barely opened core for a few hours last week, I just want something to work towards and have more wiring tools

[trim kestrel]

You could make a 1 second, AFK (Ghorm) boss farm, with enough e.g. urban moss cavelings spawned in around him. 200 ranged damage per scholar.

[serene kraken]

That sounds ridiculous hahaha

[hoary osprey]

How to put the idol down under 200 scholar?

[serene kraken]

Honestly that would be a site to see.

[serene kraken]

Quickly

Hahahaha

And pray a little bit.

[serene kraken]

The world's your oyster, build a huge beautiful farm or try to think of a design to map out and build.

[trim kestrel]

Heh, yeah. Nip out with a carefully arranged wall setup to control the firing lines. You can bait them to fire too.

[long dagger]

I dug out a large chunk and worked a bit on the farm but stopped since I saw zero finding out more about spawning mechanics

[trim kestrel]

Heh, sorry.

[long dagger]

Didn't wanna rebuild my farmed a 2nd time

[serene kraken]

I'm on about a beautiful farm

Filled with crops

Not a mob farm silly!

[long dagger]

It's not ur fault at all, I want efficiency and if anything u helped me save time

[hoary osprey]

should it be better to make small room of each spawn grid with mixed spawn than make a big room of spawn farm?

[long dagger]

I'm not big on a building a crop farm rn since I maxed farming and I'd want better ways to farm like auto harvesting with drills but it'll be chunky. I'd kinda rather wait for more tools to be added

[hoary osprey]

Auto harvesting with drill should be possible with current version but there is no auto planting

[long dagger]

I ended up wasting a lot of time on this
@serene kraken

[trim kestrel]

For what it's worth, I think it's going to depend upon what you want to farm, as to how one lays it out.

E.g. If you're after brute figurines, there's no substitute for very large areas of stone moss. Because the spawn chance per tile is so low.

But if you just want the ancient gems and cogs, then you want just 6 tiles of each moss type in as many spawn cells as possible. To make use of all the events.

[long dagger]

Ye, I'm hoping for auto planting but it'd be op

[long dagger]

I'd want a mixed farm so I'd be prepared for another Easter like event

[serene kraken]

I think a 3x3 planting tool like the scarlet hoe would be great or

A perk for gardening to replant

Now that

Would be fantastic

[trim kestrel]

I want a seed gun. 😆 Planting go brrrrRRR...!

[serene kraken]

Hahahaha that reminds me of terraria

With the special biome guns

[trim kestrel]

Oh, I was thinking coin gun...

[long dagger]

Or atleast a way to farm them while blocking specific spawns with a single switch

[serene kraken]

I remember the gun that like stops corruption

Just sprays everywhere, it's beautiful

[trim kestrel]

You could request that yourself, in the wikizet, by dropping in the Discord (if you don't fancy giving editing a go, yourself): https://discord.gg/F7PqGqXq

[serene kraken]

Do you actually edit in information for the wiki @trim kestrel

I've always wondered how like they authorise people to do that.. like finding someone and then trusting them not to just put like a big inappropriate picture on the website.

[trim kestrel]

I've done a couple little bits. But I'm trying to focus on getting this mob farming thing done. And, ideally, making more videos. If I miraculously get the cognitive function for that again.

[serene kraken]

You should take a break honestly, you're going to burn out hahaha

The work you do is insane.

It's like an actual job!

[trim kestrel]

Anyone can edit. Admins have to deal with vandalism occasionally and ban IPs. Wikis are made easy to revert.

[serene kraken]

Ah I see yeah

[long dagger]

If I'm not mistaken was it @primal rapids doing a pacifist run? There's another immunity glitch that could be used but annoying to pull off, I doubt it's patched yet, I've experienced the bug part but not the immunity. As for a friend of mine he experienced both and killed the sea titan with melee without taking dmg

[trim kestrel]

Respawning in bed was tweaked a bit, by the devs, so that if you do anything at all now, it wakes you up and ends invulnerability. (Not just moving.)

Not sure what happens if the bed just breaks and you don't move, if you can then melee...

[long dagger]

There's a way to bypass that rn but it'll take a while, not by the amount of attempts to pull it off but waiting for the bug

It'll prob trigger easily on lower end pcs

I'm willing to let mz know or whoever was doing the pacifist run so they can get it done before it gets patched

[serene kraken]

Set up mob farms on every boss, technically ur not doing anything.. haha

Total pacifist

[trim kestrel]

Right. That would count. I think the mosses make pascifist more manageable. But would be tricky to get enough stone cavelings (Shaman specifically) lined up to do in Ghorm, without him running them over when he angers.

I feel Mz's taking a break. but then seems like almost everyone is, eh?

► Place your bets!: which mobs, if any, will spawn here?

Better be quick, I've already got a result... 😅

[silk crescent]

It's always a poison slime

[long dagger]

Mold and both slime

[trim kestrel]

After 3 spawn events, it's pretty clear that anything closer than 6 tiles is blocked. But 6 tiles is OK.

(I killed the first miner and another spawned.)

It's arguable that the centres of two ofthose tiles (of the 1x3) are a little over 6 tiles away.

[bitter turret]

you know

i'm kind of discouraged to try hitting the integer limit using duped mushrooms anymore

eating 85 million mushrooms would take an eternity

[primal rapids]

Currently the biggest issue is ghorm. If someone figures out a consistent way to stop ghorm (seen the glitch a few times in the discord) then that would be great 😄

Or any other way to beat ghorm with no spikes, no melee, no ranged, no bombs (I think those were the rules, it's been a while)

[long dagger]

That is an issue... you'd just have to manually bait shamans into a room and that'd be a pain

[primal rapids]

You can now spawn them so it's less of a pain, but they don't aggro on ghorm, so you'd have to bait the fireballs too, and if ghorm hits the shamans they die in 1 hit

I've been meaning to just try it to see what happens, if it would be even close to being possible

But without some shroom/grub grind it would be nigh impossible even if it worked somehow

Get hit with one fireball and you're pretty much dead

It might be doable if ghorm didn't enrage. You could just run in front and bait fireballs from the sides

[trim kestrel]

Do you think the player base is ready for log scale graphs...? 😅

[trim kestrel]

The linear scale version is more intuitive, but very cramped at the bottom end...

Incidentally, these (theoretical) numbers (that don't take into account a variable called spawnChanceDecay) indicate that spawning every mob type in every cell is a no-go. 😢

Even with a cell full of e.g. stone moss, there's only a 72% chance of brutes. Same as a crab on beach sand, incidentally (for those who thought that was too low spawn rate). And for sentries or mold tentacles.

[inland compass]

should have done larva

only 22 million

[bitter turret]

yeah give me a larva and i'll do it

@inland compass also

how much does amber larva give

i forgor 💀

someone suggested a giant mushroom salad since it gives 50 HP as opposed to 25

@inland compass

[trim kestrel]

Fair. Cooking amaber larva with another one would give 100, right?

Dupe a stack and then leave button pressed with AFK. Is that your current strat?

[serene kraken]

I'm a fan of this honestly.

[bitter turret]

well yeah

wait

you can cook food with itself?

[serene kraken]

Yeah..

[bitter turret]

aka giant shroom and giant shroom

wtf

[serene kraken]

Any food can be cooked with itself

[bitter turret]

why

why did i not know this

[serene kraken]

Why not

[hoary osprey]

Just split it and put in the both slot

[serene kraken]

Hahahaha

[bitter turret]

and you get <food> Salad by doing so?

[hoary osprey]

Problem is the effect won't be double except perm hp

[bitter turret]

i don't care

the point was to see player HP reaching integer limit

[trim kestrel]

it's generally a wasteful thing to do. Unless you just have an enormous surplus of one particular ingredient item. Only the permanent health food buffs stack.

[bitter turret]

unfortunately duping all of the food would take a lot less than actually eating

😢

[serene kraken]

Yeah honestly I don't see the point in it.

The hp max is probably like a billion

You're going to be sat there a long while to reach that

[bitter turret]

having Goku as a playabke character in game 😄

[serene kraken]

And then what haha

[trim kestrel]

just cheat engine it or learn to mod. Be quicker, heh.

[serene kraken]

Just use a mod or hack to let you be invincible

[bitter turret]

what's the highest HP boss

😱

[trim kestrel]

merchant npcs at 100k....? Wait, I should know this...

[bitter turret]

wait what

i thought merchant NPCs had unlimited HP or something

that's why i gave up hitting them after like 20 hits

[serene kraken]

People used to farm them once upon a time

[bitter turret]

you can respawn the NPCs by destroying and re-placing their respective item thougj right

[trim kestrel]

They just come back on their own.

[hoary osprey]

Not necessary be salad

[bitter turret]

do they..... drop anything?

💔

[hoary osprey]

Perm Max HP is stacked if you cook them together

[bitter turret]

cool

did you guys create multiple worlds just to find more giant mushrooms and eat them on your main character?

[trim kestrel]

Not any more.

[bitter turret]

what did they drop previously

there's no reason to kill them

why would you want to kill them

[trim kestrel]

No. Stuck to one world as a rule.

Stacking health is kinda broken.

[trim kestrel]

Their whole inventory.

[bitter turret]

yo that's awesome

imagine cooking amber larvas or giant mushrooms together and also getting double food

😳

[serene kraken]

I mean if you're playing legitimately it's a lot more exciting

[trim kestrel]

It can happen.

[bitter turret]

yeah when i was a newbie i stumbled across a giant mushroom but i was so far away i was too lazy to try cooking it in hopes of getting double food

i did get double food when i cooked my amber larva

[bitter turret]

yeah i'll keep my 9 million HP character there just for the lolz but i doubt i'll actually use it for anything other than goofing around

[serene kraken]

I mean fair you do you, it's your game haha.

[bitter turret]

yeah plus i've got plenty of things to still do

i need to place drills another 12 or so boulders, make a railway to Ivy, Omoroth and Morpha, find some Octarine boulders

[manic shore]

What the heck is that

[primal rapids]

probabilities and log scale graphs 😄 the two things everyone is intuitively good at

[trim kestrel]

It's so pretty... 😍 But yeah, I'm gonna stick to the linear graph, I think. 😅 Maybe add a zoomed in version.

[trim kestrel]

"Probability for at least 1 mob to spawn simultaneously in a [16x16 spawn] cell, with X tiles of the relevant spawn surface".

Simples, right? 🙂 I'm honestly interested to hear how much people understand of this. To know what basics to emphasise in the explanation.

[serene kraken]

I like a guide that is dumb to the point of just like, if you make this square and put slime on it in this size - it's good.

You will have an efficient farm

Enjoy

The technical science behind it all is mumble jumble smart people stuff.

[trim kestrel]

Yeah. I'm gonna have to give example builds. (And watch in despair as many copy irrelevant details of them.)

But there's going to be quite a few eventualities, of what design to use. Depending on what one wants. And that's half the reason to go into more detail. So, in theory, players can figure out alternative options.

And to see the justification for my designs. Or at least, see that there is justification.

To be honest, coming up with definitive 'simple' designs is the harder part of this, in a way. Because they have to work against a lot of different situations, and levels of player (mis)understanding.

[serene kraken]

I guess that's what a guide is for

To explain a big hard thing

In smaller steps to understand

[trim kestrel]

What would you consider is an overall spawn chance that's pretty much as good as certain? 90%? 95%?

[crisp goblet]

Statistically speaking, 95% is often used as the threshold of significance

[trim kestrel]

Yeah. 🙂 That would be my inclination.

But for players making farms, the extra tiles to go up from 90% to 95% could probably be better used for more spawn cells, because of diminishing returns. It' puts the number of tiles needed for big larva, shamans, etc, over the 256 maximum, too.

And those following my guide are likely to go with whatever tile number I highlight... 😅

[crisp goblet]

Fair points

[trim kestrel]

I'm thinking of adding the 95% (or 90%) numbers to each mob in in this table (that I completely remode from showing my spawn-rate measurements).

Not sure if that will tip into TMI overload. (Probably did already. 🙄 😅 )

[crisp goblet]

What does 'min tiles' = 0 mean?

[trim kestrel]

It means there is no minimum. That's how it is in the game files. But good point, I should probably convey that differently, here.

I guess I say "1" instead, as that's more accurate, in a way.

[inland compass]

Well technically it is 0 right?

0 tiles at 30% chance per tile is 0% spawn

[trim kestrel]

I mean, because you can never have anything spawn without at least 1 tile, the minimum number of spawn surface tiles (to get any chance at a spawn) is really 1.

[serene kraken]

90/95% either really

[trim kestrel]

The last couple days I've also been running a couple experimental setups to confirm the spawn chance per tile numbers.

I had an increasing number of tiles of slime all set u inside one spawn-cell, to collect and count the number of spawns. Starting with 1 tile per collection bin, as there's very high odds initially.

I ran a baseline reading for a couple hours yesterday, then today, for 6 hours, one with a single orange slime sat in the middle of the cell to inhibit the spawn chance...

I compared the cumulative counts with theoretical numbers calculated from the data-mined chance values.... And got an extremely close fit! Which always makes a refreshing change.

That is, if I divided the theoretical chance per tile by 6, it gave the exact same curve I saw for orange slimes when 1 orange slime was inhibiting spawns.

Red slime spawns were unaffected by this blocking slime.

The curves for scholars and sentries (off urban moss) fit about as closely as could reasonably be expected, too.

Which all means that I can't see any use for the "respawnChanceDecay" values that are declared. At a value of 0.8, for all of the above mobs. (Any ideas...?)

[bitter turret]

those slimes look adorable

[serene kraken]

So good news and progress from what I'm understanding? Hahaha

[trim kestrel]

Yeah. It's a nice little thrill when experiment matches theory.

I just need to repeat this to check how much 2 slimes inhibit spawn chance. (Compound or additive factor of 6, presumably.)

Then do the slightly trickier maths to calculate the chances for getting all 4 possible infected, with X tiles of mold ground... And for 2 mushrooms.

Then, I should be ready to finish up the infographics and write up the guide theory. So I can actually design some farms! 😅

[crisp goblet]

Makes sense, though I think maybe you mixed up some numbers? For cavelings you have 6 ground minimum, and 0 moss minimum

[trim kestrel]

Very well spotted, thanks! 👍 Yeah, I switched the ground images around, to a more logical order (descending rate). Forgot to switch those number around with it.

[crisp goblet]

That's what threw me off, was curious how you could have a '0 moss minimum', for ground it makes more sense but as you mentioned above perhaps '1' would be more clear.

[long dagger]

It'd be sad if the devs tweaked spawning mechanics after seeing all this work

[crisp goblet]

lol, I was just thinking... watch the devs get back from vacay and start changing up spawn mechanics

[serene kraken]

Bruh it's quite likely as well.

I can definitley see spawn mechanics altering

and all this work @trim kestrel will be made mostly redundant

[rich belfry]

thus is the nature of early access. its still a fun way to kill time between patches

[shadow grove]

is it possible to show the chunks(16x16)? Is it better to fit a farm field into that chunk? - and that means if a map is maximum covered the spawn rate will increase more than 22 min?

[shadow grove]

you have the slime under the belts?

[inland compass]

Map tool has the mob grid button

And I'm not sure what you mean by map is maximum covered

My guess is the game allocates X amount of time per tick for spawn calculations. As you uncover more of the map, there is more to calculate. Thus the calculations start to slip

The slowest would probably be having every type of spawn surface with at least the min spawn chance in a spawn cell within a 12 spawn cell radius.

[long dagger]

For open world sandbox games that im aware of, they usually change spawn mechanics atleast once. First off there's mobs ending up in water, void, and occasionally in walls but dunno if that's even related to spawning, just mentioning it if it is by chance. I wonder if any of the devs looked at his work and considered to not change the spawning for the future. If anything I do think spawning will be changed but it'll prob be minimal

[serene kraken]

Honestly it could help the Devs with making changes and be a positive thing exactly.

[trim kestrel]

My life is sad, lul. 🥲 😅

As much as I put it out of my mind to work on all this, I do think it may well be tweaked/overhauled before long. Learned that much from wood farm and bolder guides. And I'm asking for change to the cooking system...

I think it would be sadder, overall, if there were no changes and my (hopefully to be released) guide stayed relevant. Like my 4 year of TerraTech crafting guide, that covered so many bugs and confusing idiosyncrasies. But nothing ever got fixed (cos they were too scared to break it worst, I think).

I think, and hope, Pugstorm are bold enough to set my efforts aside and make the game as good as possible. 🙂

[trim kestrel]

Yeah, just in the places shown in the map pic. So their number adds up like this, in cell search order (anti-clockwise):

[trim kestrel]

I don't know if 22 min is the max (slowest). That's what I got at my core, where all world chunks within range have definitely been generated.

[shadow grove]

does exist to shown the chung grid?

chunk

[trim kestrel]

https://www.craiginsdev.com/maptool.html#

[shadow grove]

ah no way to show in the game?

[trim kestrel]

You'd have to measure and mark it out by hand. As I've done.

Of course the devs don't want you to see the mechanics going on behind the scenes. That's why it's taken so long to figure it out.🙂

[trim kestrel]

Are you saying that you think more spawn surface would scale back the spawn event rates? (I don't see this, but not directly tried to test that...🤔 )

The cycle rates are extremely stable, to within +-2s, which I think can be put down to mini-hangs, auto-save, whatever.

I think the cycle scaling is probably all pre-calculated. And first guess is that it's by world generation. But could be by measuring the amount of ground (and water?) tiles, more specifically.

Probably not map reveal, as dark map can have a lot of mobs spawning (e.g. Ghorm ring). But that'd be funny, if you could just delete your map file to boost the rate.😅

[inland compass]

so why 15 to 22 minutes?

[trim kestrel]

You got any ideas? 🙂 I'd say maybe 15 minutes is just a dev arbitrary upper limit to spawn rate. 🤷

A 12 cell circular radius would mean up to ~450 cells in range at any time.

15 minutes is 900 seconds. But 22 minutes (when all those cells seem to be in use) is 1350. Pretty close to 3 seconds per cell. Quite a leisurely rate.

[inland compass]

So really it is 6300 checks, 14 different spawn types

But they are probably not keeping track of tiles, so it could mean up to 1.6 million checks

Not super significant unless they try to do it all in a single update cycle

Which they could be doing. Toss all the spawn cells in a list and go over x number per update. They may check radius at that time, so the list would have all spawn cells in the world

[trim kestrel]

Sorry, where's your 6300 from?

I imagine there's a lot of ways one could approach coding this kind of thing to be efficient. But, from observing the game behaviour, from an algorithmic perspective, I think it might be quite simple and wasteful.

I think it may literally look up the type of every single tile, for all 256 tiles, in each cell, when it has a spawn event.

I don't feel it going the other way around, and starting from a shortened down list of the 14 (is it?) spawn surface types.

Remember that all water is also a spawn surface for fish spots (from my mined data). I'm not sure if all open tiles are spawn surface for fireflies, too (didn't line those up in my sheet, maybe missed them). Although, maybe treated differently since their optimisation...

But I think the spawn algorithm may well be rolling every (open) tile all the time. With all possible mob types as percentages of the possible random number range.

It appears to decrement the time on each of the ~450 spawn cells that're in range too. Rather than having them pop at a set time in the future. Because they don't all seem to pop at the same time when I return to an area... 🤔

But I guess that could be because the ones I'm looking at are just in a queue, with the game sticking to processing only 1 every 3 seconds. However overdue their timers are.

Simple enough when teleporting and then standing still, but how this copes with moving about rapidly...? Is it constantly re-sorting a list of the in-range spawn cells? As it moves up to 24 in and 24 out, per ~second, while traveling at 14 tiles/s on a cart. (I guess it'd only need to do that once per 3 seconds, not so heavy vs other basic things.)

[inland compass]

460 spawn cells, 14 different surface objects, 256 tiles per spawn cell

And like i said, they probably check ALL spawn cells, but the first check would be within player range

[trim kestrel]

Check for what, when? The spawn algorithm only checks 1 cell per 3 seconds, I think

[inland compass]

You have to check the tiles in the spawn cell

256 of them

In theory you could have some sort of dirty bit set and only count the tiles if the dirty bit has been set

But I'm not giving them much credit on any type of optimization

[trim kestrel]

they probably check ALL spawn cells

Did you mean ones out of range. Every cell in the world? That's what I was asking what when?

[inland compass]

Oh yes, every cell in the world

But the first check would be distance to player

[trim kestrel]

But that's all that's ever checked for the vast majority of cells. If out of range, it doesn't care what tiles it has, or its event timer...

But it might easily avoid checking every cell in the world for range, every 3 seconds, if it's updating the list of cells within range as you move.

Which I think it does for every tile within 80 or 100 tiles, or whatever, for other things, like physics collision. Much heavier. 😵‍💫

[inland compass]

if they have separate event timers for all the spawn cells that's even worse

lol