How did you guys die to gacha again

>>553985687
We need roguelike porn

>>553985687
o tabbed too furiously to goon

File: Screenshot 2026-01-15 213431.png (1.33 MB, 1843x985)

1.33 MB PNG

Damn it wasn't even that many titties in the catalog on the way over

>>553995041
It is just timing. I've seen the thread not hit page 10 in 4h
I feel the board has slowed down since they added the new captcha but when all the gacha threads simultaneously make a new thread then it's over

File: 500 tiles.png (78 KB, 1763x1048)

78 KB PNG

You gotta be fucking kidding me I'm 500 tiles from my home to the nearest refugee center? I haven't even cleared a quadrant of the megacity right above me what is this garbage lmfaoooo

>>553997358
better get driving

I'm new to ToME4. Just cleared the Maze but it didn't update my quest. It feels like there's an inaccessible pocket at the bottom of the map but I can't mine the walls. Anyone have tips for this?

>>554008232
This is likely the alternative layout with 4 floors (the default has 2), in which case you are locked out unless you get yourself a teleport rune (sold in shops) and just teleport randomly through walls

>>554008318
Thanks, I'll try that if I don't die in Old Forest.

>>554008232
Just get a pickaxe, they are relatively commong and you dont need to actually equip them to use the dig ability.

>>554008950
You can't dig in the maze specifically

>>553985958
erotic elements in a roguelike that is sexy and funny and not coomer garbage would be great

>>554009698
>sexy and funny and not coomer garbage
Where do you draw the line?

File: Screenshot 2026-01-15 235654.png (87 KB, 1920x1032)

87 KB PNG

>>553997358
that's the new normal, but i just ignore the "main quest" and NPCS and everything. solo unless I find someone out there.

>>554009994
remember how in the 80s and through to the 00s there were lots of funny movies with nudity and sex jokes? sometimes it was even a little erotic? that stuff we don't get in movies anymore?

stuff like that. or tasteful eroticism, though that's a tall order in an ASCII game. i have no idea what it would look like, but it would add a little zest to this nerd shit.

>>554009698
elona/elin has prostitutes fighting town NPCs to the death because they ran out of currency in their inventory to pay, and the prostitute usually wins because sex inflicts insanity debuffs
most player-made sprite replacements wouldn't fall in your camp though

>>554011430
that's pretty funny emergent gameplay, but i don't think that would make a good game. i am reminded of the first time i saw some complex NPC behavior play out in Postal 2. a guy allah ackbar'd me when i was standing near a cop and it aggroed the cop onto me and was generally chaotic and very funny.

maybe if NPCs had some romance options. kevin being kevin ruled out integrating AI chatbots into NPC convo and obviously none of those guys have touched a woman so I have, shall we say, no hope of this happening. and there are other ways you could do it beyond gag porn mags.

>>554012279
someone on reddit just announced/released some sort of branch of cdda with llm integration for the npcs. i havent looked at it, but i saw it just literally a few hours ago.

anyways, my game crashes everytime i aim a weapon. anyone had that happen before. different saves and everything.

>>554014356
https://github.com/josihosi/Cataclysm-AOL

>>554014356
never mind, update fixed it

>>554015386
Which update

>>554019836
the latest update

Continuing from yesterday, xom has too many ambiguities given only the wiki information so, yeah, I'm code diving now.

Everything else is dependent on "tension", so I've got to cover that first, for simplicity, you see.

Tension is dependent on monster-tension, which is separate, and is divided by a factor related to exp level.

"exp_needed()" is dependent on a function called apt_to_factor()
So, that's where I'm starting this journey.


In skills.cc:
>#define APT_DOUBLE 4
>float apt_to_factor(int apt)
>{
> return 1 / exp(log(2) * apt / APT_DOUBLE);
>}
Exp is apparently e^(input), so that's 1/ e^(log(2) * apt/ 4).
log(2), in c, seems to be ~.693, so it's ln(2). Seems about impossible to comprehend meaningfully.

Thankfully, apt can only be 4 values, so we only have to contend with 4 output values.

exp_needed() is in player.cc
> exp_apt = species::get_exp_modifier(you.species);
>return (unsigned int) ((level - 1) * apt_to_factor(exp_apt - 1));
exp_apt is retrieved from your species file, in the species folder, which is in the dat folder, which is in the source folder.
Demigod has -2 exp apt. 11 species have -1 exp apt. 12 have 0. Only kobold, formacid, and dijini have 1 as an aptitude.
exp_needed() subtracts 1 from aptitude before sending it to apt_to_factor, meaning it receives -3,-2,-1, and 0 as inputs.
For 0, it returns 1
for -1, it returns 1.189207
for -2, it returns 1.414214
for -3, it returns 1.681793

So that's that tiny function dealt with.

>>554010446
>that's the new normal
Is it really? I don't remember refugee centers from my previous runs suddenly spawning 500 tiles away from where I spawned that's for sure.

>>553997902
I have a default car and a 3x1 hit and run mobile, and I'm afraid only the latter can fit in relatively unscathed through all those mobs even if I drive perfectly and have the perfect driving conditions. This is so ass actually I'm only ever used to manually clearing things

just a thing i noticed in cdda. storing books on your phone, despite you gaining some knowledge and recipes from it, does NOT count as reading. ive stored 50 books, and only just now lost the conduct "self imposed illiteracy" by reading the trifex handling procedures. not that this really matters, just funny.

i guess if your running an actual illterate character, you can sorta cheese it a bit by learning recipes/storing reicpes by just scanning books

File: rlg.jpg (575 KB, 2048x1365)

575 KB JPG

It's very much over

>>554022786
i dont know what hes talking about. i have done 3-4 deaths in the past 2-3 days and the center spawning no more then 150 tiles max. i think your just getting unlucky.

>>554020667
exp_needed() is best represented in a table, which is probably why there's a commented out table which assumes exp aptitude zero. Rather, it assumes exp aptitude 1, because an aptitude of 1 is submitted to apt_to_factor as 0, which returns 1, and which is multiplied by the value submitted from the table. In other words, if aptitude is 1, then apt_to_factor does not modify the exp needed at each XL In other cases, it multiplies the total exp_needed.

The variable "level", through a series of else-ifs and switches, has the following values at each xl prior to modification:
1, 10, 30, 70, 140, 270, 520, 1010, 1980, 3910, 7760, 15450, 26895, 45585, 72745,108375, 152475, 205045, 266085, 335595, 413575, 500025, 594945, 698335, 81095, 930525, 1059325

However, the function actually returns:
> return (unsigned int) ((level - 1) * apt_to_factor(exp_apt - 1));
Which means level 1 returns 0, level 2 returns 9 instead of 10, level 3 requires 29 instead of 30. Rather than matching with this table or the wiki at aptitude 0, almost all instances of exp_needed() return the given value -1, assuming aptitude 1. Congratulations, you have exactly 1 less base exp which you need to get.

>tension, finally
In "religion.cc":
>int get_tension(god_type god)

It's a long function. Contains:
> // Divides by 1 at level 1, 200 at level 27.
> const int exp_lev = you.get_experience_level();
> const int exp_need = exp_needed(exp_lev);
> const int factor = (int)ceil(sqrt(exp_need / 30.0));
> const int div = 1 + factor;
> tension /= div;
So, tension is divided by a factor based on your current exp_needed(), based on XL and exp apt. This requires a near-unreadable and unfriendly text-based table.

Tension is divided by:
1+sqrt(exp_needed()/30), or
1+sqrt( ( (EXPN-1) * (1 / exp( log(2)*(APT-1)/4 )) ) /30)
where expn and apt are the variables

File: 1747465200447027.jpg (345 KB, 1169x1059)

345 KB JPG

do it again

>>554025960
bros less code talk and more feature talk if you want people to pay attention

>>554025960
Correction: at XL 25, the value is 810195, not 81095
column corresponds to apt 1, 0, -1, -2
row corresponds to XL
Tension is divided by:

1 1 1 1
2 2 2 2
2 3 3 3
3 3 3 3
4 4 4 4
4 5 5 5
6 6 6 7
7 8 8 9
10 10 11 12
13 14 15 16
18 19 21 22
24 26 28 31
31 34 37 40
40 44 48 52
51 55 60 65
62 67 73 79
73 79 86 94
84 92 100 109
96 104 113 124
107 117 127 139
119 130 141 154
131 142 155 169
142 155 169 184
154 168 183 199
166 181 197 215
178 194 211 230
189 206 225 245

So, in the most extreme cases, tension is halved between xl 1 and 2, and tension at xl 3 is 1.5* higher if you have apt 1 compared to any lower apt.

Other than comparatively, these numbers don't mean much until we know how much tension monster provide.
> int get_monster_tension(const monster& mons, god_type god)
This function is also in religion.cc.
Dead monsters return zero. Monsters that can't hurt the player, good monsters, neutral monsters, and monsters that don't give exp all return zero and so don't contribute to tension, which means most summons don't count.
Hostile god summons provide zero.
Friendly units reduce tension, providing negative tension equal to the positive tension they would have provided. Friendly god gifts provide double tension reduction.
Invisible monsters provide double tension. Monsters that can't see you provide half tension.
Confused and "caught" onsters provide half tension.
Slow monsters provide 66% tension
Haste monsters provide 1.5*
might provides *1.25
Breserk provides 1.5*

A monster's tension contribution is multiplied by their % remaining hp, so at 50%hp they provide 50% tension.

>int exper = exper_value(mons);
Their base exp value is used here, so we need examples of that. function exper_value() is in mon-util.cc

>>554020667
tension is horseshit, the best way to get a feel for it is to play Lava Orcs from the brief stint they existed in Trunk

Can someone explain what that anon is blathering about?

Or not I guess.

>D8
>joseph, pikel, prince ribbit, and FUCKING GASTRONOK
>not even half explored yet
what on earth is this nonsense

>>554045263
Those are all good uniques to have on D8. Gastronok is like half the speed of a player, by the way, and the rest are easy experience on D8.

>>554045263
Just use some consumables, easy xp.

Is Sigil of Binding bad?

>>554049647
On most characters, yes it's quite bad. It doesn't usually spawn where you want, and the effect is not that useful.

For what it's worth, Sigil of Binding is most useful on a Barachi because you can hop over your own sigils and it helps you walk away from dangerous monsters with your slow movement.

You guys are terrible at this.

roguelikes with a magic crafting system a la Oblivion except way more detailed because it's a roguelike and we use our imagination

>page 9
its so over

>>554063643
Crafting is thoroughly unsatisfying in most rougelikes unless it's really straightforward like in Sil where it acts more like a periodic scroll of acquirement when you find forges rather than holding onto a billion useless components

File: V2Alita 089.jpg (327 KB, 3840x2160)

327 KB JPG

And upgrades complete.
An anon in the last thread mentioned that helicopters get more efficient as they get bigger, from what I can tell, blimps are the same.
I haven't increased either the length or width of the vehicle but because I added more frames, panels etc, I was able to add more balloons and with them more solar panels, batteries and have more carry capacity.
I think optimally you'd want to make the main part square and then have a single row of doubled up balloons on every side.
Kind of makes me wonder how small you could make one and still be viable as a mobile base.

>>554039715
He's reading the code of DCSS to figure out the mechanics of the god Xom. But delved too deep and autistic. I'm sure he's doing this on purpose. Regardless, DCSS's source code might not quite reach the level of spaghetti code, but it's almost as head-scratching, because it's more like spaghetti design. Everywhere has multiple layers of janky modifiers and coefficients inserted into what was probably a simple formula at one point.

>>554063643
what is magic crafting in oblivion

>>554073864
Trying to "figure out" Xom is kinda contradictory to the point of picking Xom. I know video gamers love to optimize everything, but the point of Xom is to insert lolrandumb into your runs, so don't overthink it so much.

>>554079458
you can create spells with various effects or infuse staffs/wearables with a mix of those effects, it's really quite broad. want to make a spell that does damage, drains your opponents mana and makes you levitate? it's possible.
it's too bad such a powerful system is trapped in such a dogshit game

>>554081810
>it's too bad such a powerful system is trapped in such a dogshit game
It's mostly just Morrowind's magic system with some nerfs and small changes, so you can easily play with it in a far better game

>>554083265
>It's mostly just Morrowind's magic system with some nerfs and small changes
Glad I never bothered with Oblivion or Skyrim.

>>554079591
for you

>>554081810
Main issue with it in a roguelike game is that you don't typically have excess amount of ingredients like in those games
That or you want another sandbox like CDDA

>>554079591
Nah, there's no need to put Xom on a pedestal, he's a game mechanic just like every god, not a simp target.

>>554079591
But he's not random, that's the point. If he had a different flavor, like "Xertic, god of courage and trials" then you wouldn't be saying that at all, even if the mechanics stayed exactly the same.

>>554094179
>>554090287
>weird people trying to read my mind and psychoanalyze me
Where did this shit come from? Look, I'm just saying that even if you figure out a way to hack Xom and get only his good aspects, you're just doing a bunch of boring shit outside the game to break the mechanics, and then it's not going to be satisfying to abuse if you succeed. I'm not knocking you for trying it, but it sounds to me like you're optimizing the fun out of the game if you take the random god and try to set up deterministic outcomes. This guy >>554083625 is the only dude who makes any sense in here.

>>554094749
Do you consider searching for guides or looking up what an item/spell/monster does online to be boring shit outside the game? Plenty of people do that, and they still find satisfaction. The only difference is that the devs have told you this particular mechanic is lolsorandumb.

>>554095882
It's obviously better if that information is in the game itself. Of course I use wikis when it makes sense but it's not exactly the highlight of my gaming experience when I alt tab and open my web browser.

>>554096759
and all anon is doing is taking info from outside sources and using that to strategize
the blogposting is mildly annoying, but this is /rlg/

It feels like the Brogue late-game essentially boils down to whether or not you can avoid being lit on fire. Sometimes by enemies on the other side of the floor that you can't even see.

Are there any games more like Cogmind? Tired of DCSS level slop where devs just jam mechanics with zero consideration for overall design or balance. Preferably one where stealth builds aren't mega OP and combat is super difficult, that's my main gripe.

>failed 8 coinflips in a row
the idea that dcss isn't fucking with your rolls in dangerous situations is laughable, I don't care how they've obfuscated it

File: dcsshepshade.png (926 KB, 1557x957)

926 KB PNG

This Martyr spell apparently juices Hep into the stratosphere by doubling the ancestor's EHP. I sit back at the edge of LoS and these two dudes kill everything. I started out melee, but there's nothing left for me to bash when this bitch runs out and two shots any enemy in the game.

>>554103098
inb4 some faggot says "don't put urself in that situation in the first place, it was winnable if you played better" like every time someone complains about losing a 50/50 8 times in a row

>>554110339
1/256
you kill about 70 billion monsters in a run.

Using 8 coinflips each to kill 256 monsters, on average, kills you if you don't keep a retreat available.
Do you want to kill more than 256 monsters? Then you'll need to retreat 1 our of 256 times.

That means you'll have to spend your last turn on a blink scroll, which means you only get 7 coinflips, which means you need a blink scroll for every 128 enemies you face this way, That's a good 40 or something blink scrolls per run, which you're not getting. In other words, a coin flip is insufficient in the first place, if it's all you've got.

So no, I'll not be a massive asshole and tell you how you could win more.
I'll tell you you fucking suck. Your shit's retarded. Petulant entitlement will not win you the game.

>>554031087
>>554029083
Feature talk can only happen once I understand the code enough to know what features it has.

Pay attention when I get to the end of each section and summarize the results, if that's what you're interested in.

>>554034495
Seems to be, yep.

>function for monster experience value
> x_val = (16 + maxhp) * hd * hd / 10;
This is the base exp value of a monster. You'll notice that it uses maxhp, which seemed interesting, and so I set out to figure out how an Apis aura works with this, which sent me for a trip where I eventually found that the relevant apis aura started in .33, but I have been diving in .31 because that's just what came up first. So, now I would have to verify that my statements are still correct, which is to say you should add another grain of salt to all my conclusions here, as they might be wrong due to conflicting version information.

Also, in version .33, there's an extra line of code that accounts for the hp-doubling effect, meaning that I spent a whole lot of time looking for something that was already fixed.

>ANYWAY
Footnote, in .31, all information defining monsters was stored in a file called mon-data.h, but now it's been reformatted and I am again unsure where most of it is stored.
All unspecified information from this point forward refers to version .33

In mon-util.cc, exp_value()
>else if (avg_hp > 0 && !zombified)
>{
> x_val = x_val * 4 + x_val * maxhp / avg_hp;
> x_val /= 5;
>}
In .33, the formula seems massively different. For one, HD are no longer used anywhere in exp_value(). For another, maxhp now only scales a 5th of the value of exp_val, and only based on the variance among monsters of a particular type.
So, if you saw a green dragon which, by chance, somehow, had double the base HP of a normal green dragon, it would only supply *1.2 exp. A dragon with 1 hp would supply *0.8 exp. In other words, who cares?

> x_val = _mons_exp(mc);
x_val is basically determined by _mons_exp(), at this stage.

It is kind of sad that DCSS ruined the Freezing Lajatang that Battlemage Hep gets, since now it doesn't deal the right damage type to kill the other two orb types in the nu Zot 5. It's cool that there's new content, but the game sort of had a hierarchy of resistances where rF and rPois were mandatory for the player, while cold damage was the best damage type to deal to kill fire enemies in Zot. That whole balance is thrown off when there are different orb types.

I guess that's just one more consequence of the devs inserting their original content do not steal into a 20 year old game. I think the real smart move is just building a character that kills things at 7 range in the current gamestate, because you usually can't get all the resistances, get strong AC/EV/SH, and still deal good damage. Plus there is no resistance for doom so you might as well just put the harm scarf on and kill things faster. That's what the game really is about right now.

>>554116735

>static int _mons_exp(monster_type mc)
> return smc->exp;
Didn't know what this meant. smc just subs out for get_monster_data()

>monsterentry *get_monster_data(monster_type mc)
> return &mondata[mon_entry[mc]];
So the value this returns, from the array mondata, contains the exp value for a monster. This means mondata is an array containing the information on each kind of monster.
This variable is not declared anywhere in this file, or any file.

In "makefile":
>mon-data.h: dat/mons/*.yaml util/mon-gen.py util/mon-gen/*.txt
> $(QUIET_PYTHON)$(PYTHON) util/mon-gen.py dat/mons/ util/mon-gen/ mon-data.h
During compilation, a file "mon-data.h" is created which defines the array mondata[]
It does this using the files in "data/mons/*.yaml"

This is to say, there's a simple fucking folder full of monster data right next to the folder full of player species data and I missed it because I was originally on version .31, where all monster data was in a single .h file.

ball python: 1 base exp. Fucking 1.
quokka: 2 base exp
endoplasm 2
ork: 3
adder: 13
gnoll: 14
ork priest 42
bullfrog 90
ork warrior: 133
black bear: 128
acid dragon: 155
small abomination: 163
golden eye 190
fire bat 196
yak 204 insufficient
blorkula 224
raiju: 264
ancient zyme 324
cane toad 407
apocalypse crab: 454, which is fucking abysmal
draconian: 530
faun 538
ork sorcerer: 559
ork knight: 617
orb spider: 708
large abomination: 710
goliath frog 749
fire crab 764
ork high priest: 795
death yak 811
bolder beetle 828
alligator: 842
entropy weaver 898
fire dragon 910
ettin 929
Apis 1042
colored draconians 1054
All specialized draconians ~ 1200
azure jell 1337
ancient champion 1518
executioner 1531
frostbound tome: 1641
golden dragon 2457
bai suzhen 3094
Orb of fire 4638
ancient lich 4693

So, here's a list of somewhat representative enemies and their base exp values, before modification. I will be printing this list again, after modification.

>>554039715
Xir is trying to optimize playing Xom. Let xir cook. I personally find it quite funny we got quite the clown in our midst

>still no real Qud update since release
Didn't this game sell a bazillion copies? I know they want to milk it with paid DLC instead of free content updates, but they didn't even finish merging the digging claws update from 2 years ago.

>>554118058
Found source/util/mon-gen.py
It has the instructions which the game uses to turn a .yaml (text file) of a monster (yak.yaml) into data in the mondata[] array.
There is a variable "exp_is_mult" which is a boolean. If it returns true, a completely different formula is used to determine the end exp of the monster. This value does not appear in yak.yaml. Instead, the python script, in cases where it is unspecified, generates the variable as false by default.

pan lords, player ghosts, dancing weapons, ork apostles, mutant beast, and player illusions are the only characters which return true for this variable, thank god.

For these enemies, vaguely:
> x_val = _calc_exp_with_multiplier(mon, x_val);

In _calc_exp_with_multiplier:
>int hd_factor = hd * (25 + hd) / 25;
means hd factor is ~1 at hd 1, 6 at 5, 14 at 10, 24 @15, 36@20, and 50@25

diff is a difficulty modifier. It's value is clamped between 50 and 250.
For every spell an enemy has, if the spell is difficulty 5 or higher, it adds the "frequency"/2 of that spell to difficulty. Frequency is stored in "slot.freq". I failed to determine where slot is defined and stored, but the default frequency of spells is stored in mon-spell.h. It is unclear if this default value is normalized at some point, in which case the default frequency is a "weight" and a single spell of frequency 1 would have effective frequency 100. Assuming that's not the case, spell frequency varies a bit, but tends to be around 11, with some at 50 or 100. So, a powerful spell tends to increase difficulty by 5, or maybe 50, depending on how it actually works.

A melee bonus is added to difficulty for fast or normal speed enemies. For every physical attack they have with a max damage over 29, half that damage is added together. IF they are speed 10 (normal) then the damage is halved again.

Diff is then multiplied by speed/10, so 1 for normal speed, or *2 for double speed.

>>554129213

So, for a normal speed pan lord with a single spicy spell and a punch of 60 damage, that's
dif starts at 100
+ like 5 or 50 for the spell.
+15 for the punch
*1 for the speed
= diff of 120 or 165, which contributes *1.2 or *1.65 to the exp rating of the enemy.

> return exp_mult * hd_factor * diff / 100;
so, returns base exp * (like 24 or 36) * (between .5 and 2.5, but probably like 1.3)
or
base exp * 30-ish for these special monsters.

For all others, it's an average of x_val = monster default exp with some variance for max hp.

This system, just a few versions ago, was far more interactive, with the ability to somewhat manipulate these numbers by, for instance, getting regeneration on your enemy, or making sure you never cause an enemy to drop their weapon. Now, the whole difficulty mechanic seems to be just for a select few. The hd effect makes me think I've misunderstood something. Potentially, the default exp for these enemies is very low, which could mean they have more "reasonable" exp amounts overall, despite this multiplier.

exp is divided by the number of blob's in a large blob, or 12 for starcursed, which may imply that a starcursed mass can only divide into 12 blobs.

There's a stepdown function which does not run anymore.

And, finally, if you somehow get an exp value less than 1 at this point, the function returns 1.

In short, the exp value of a monster is, simply, whateverinthehell it says in their monster file.
It is horrendously un-interactive, but understandable.

>>554130268
In .31, the "difficulty" modifier applied to everything, was different, was more interactive and perhaps gameable. It just wasn't particularly discoverable through gameplay, how one might force enemies to give *1.3 exp on average or something. That dream is dead.

Forgot, also zombies give /4 exp.

>next section
in religion.cc, function get_monster_tension() :

First, it checks mons_can_hurt_player() in nearby-danger.cc. This returns true if a monster can path to you, if a monster has a ranged attack/spell, or if the monster "blows up"

this means, if a monster can't reach you with their ranged attack, but you can see them, then the monster can hurt you.
If the monster has a path to you, but somehow can never actually reach you, then this would return true, but that depends on how has_path_to_player() actually works.

As this function is also used for "travel interruption", you should be able to test this by clicking on the map and seeing if your character walks by the monster without stopping. For instance, a monster trapped behind glass will not stop autotravel. However, I don't know how autotravel would treat a monster which thinks it is trapped behind water, if that monster is incorrect.

If a melee monster trapped on an island can provide tension, then you may be able to create a condition which keeps xom perpetually active using dispersal or glorbula, which would mean you can, at any time, retreat to dungeon and let xom reset his shitty attitude before you go handle zot. This assume the trapped creatures represent significant tension, which we're still trying to figure out.
Just keep in mind that xom could try to kill your prey at any moment, unless we find a fix for that.

>Anyway
if monster can't hurt player, if monster is paralised/sleeping/stone, if monster is dead, then this returns zero tension contribution.
living spells and the like contribute 50 monster_tension

>Two Headed, Two Hearted, and Regeneration
Feels like I'm playing with training wheels. This is why I always go True Kin. Mutants are too easy if you actually take the good mutations.

>>554134860
Others begin modifying their tension based on their exp
> int exp = exp_value(mons);

> if (att == ATT_HOSTILE || att == ATT_NEUTRAL)
> exp = exp * (10 + (mons.hit_points * 10 / mons.max_hit_points)) / 30;
> else
> exp = exp * (10 + (mons.hit_points * 10 / mons.max_hit_points)) / 50;

means that exp for enemies is multipled by between 1/3rd and 2/3rd, proportional to missing % hp.
All others are between 1/5th and 2/5ths

Hostile god gifts do not contribute tension.
Friendly units contribute negative tension.
Friendly got gifts contribute double negative tension.
Neutral enemies contribute half tension.
"good neutral" gives halve negative tension (*-.5)

Sleeping and fleeing monsters give /20 tension.
Invisible enemies give double tension.
Enemies that can't see you give 66% tension.
Haste gives *1.5
Breserk/frenzie gives *1.5
"armed" enchantment is *1.2
"chaos lace" is 1.2
might 1.2
empower 1.2
recall 1.2
slow .66
vex .66
silened spellcaster/priests .66
confused/caught .5
fleeing .1
sleep/paralysis .05

That's a lot of little modifications. Notably, sleeping counts twice, as does fleeing. Seems like a mistake, but the bottom line is that paralyzed/sleeping/fleeing enemies contribute basically nothing.
As exp is the basis for monster tension, I think that player-summoned units would not take away from tension, but that needs to be somehow tested, perhaps with sword of singing. I somewhat think that's the real job of that artifact..

Anyway, after all those modifications, the value is returned, and that is the tension contribution of a single monster. It is initially reduced to 2/3rd. Haste and might give *1.8. Slow/silence/confuse/vex are useful, common, and deployable, but reduce tension. Anything that's asleep/paralyzed/fleeing is nearly meaningless. Friends negate some tension outright.

the dark side of roguelikes
talking to yourself about C code

I planted some seed potatoes in cdda, how long do they take to grow? I dont have any fertilizer.

>>554142847
Really fucking long time, I'm afraid.

I'll have you people know, I don't know python, or C.


Anyway,
>radius_iterator ri(you.pos(), LOS_NO_TRANS);

> const monster* mon = monster_at(*ri);
So, ri is being itorated and returning tiles, which monster_at turns into a monster. I could assume that this grabs each monster you can see, or each monster within rang to hypothetically see, However, this thing has a LOS variable, which could either set the range at which you can see, or could make this radius_itorator "ri" only return tiles within your character's "line of sight". If we could get a clear readout in game, this would be easy. I could just stick a monster behind another monster and see if the tension goes up when the monster steps around the other monster to get at me.

Instead, I've got this:

>radius_iterator::radius_iterator(const coord_def _center,
> los_type _los,
> bool _exclude_center)
> : state(RI_START),
> center(_center),
> los(_los)
>{
> ASSERT(map_bounds(_center));
> credit = get_los_radius();
> is_square = true;
> ++(*this);
> if (_exclude_center)
> ++(*this);
>}
A constructor for radius_iterator

I think the (_exclude_center) conditional pulls the first value out of the radius itorator, or increases the value of the pointer at *this by 1, and skips by the center tile. However, the for-loop in religion.cc, simply uses the radius_iterator as a boolean, while also incrementing it like an integer. It's slightly beyond me.

HOWEVER
los-type.h is a n enumerator which is filled with commented out inforamtion. LOS-type refers to what tiles are meant to block vision for the purposes of your particular radius itorator, and that means it has nothing to do with your character's line of sight. Mystery solved.

LOS_NO_TRANS means glass, walls, and 2 bushes/clouds in a row will block vision for the purposes of this calculation. An ettin behind glass does not contribute to tension. That's all.

>>554144124
>LOS_NO_TRANS
Despite the warning, they took over roguelikes anyway

>>554144124
total = tension value of all monsters in los, excluding transparent vision.

At least 1 monster must be able to attack, which is based on "attitude". Friendly, marionette, good-neutral and neutral return true and won't attack, but can still contribute to tension.

In abyss, for some reason, tension counts even when there are no visable enemies who can increase tension, which means that you may be able to, in abyss, have high tension despite having no enemies, somehow. Due to the nature of the abyss, that's not particularly abusable, but it may be possible to have a certain kind of enemy follow you around the abyss, and so always provide at least some tension.

Depending on % remaining hp, multiply total tension by
1@ 100%, ~1.14@75%, 1.333@50%, 1.6@25%, 2@ 0%
So, mostly nothing, like 1.1 normally, but 1.5 ish if you're already basically losing.

divide by your space on the table up here >>554031087 which represents your current XL scaling. (that's the same in .33)

Now, if on the orb run, tension is now raised to 3, if currently less.
If in abyss or unable to act, tension is now raised to 2, if currently less.

There are quite a few checks and related multipliers next:
If you.cannot_act() *10
If vexed *2
hav orb *2
caught *2
under vainglory *1.666
silenced *1.666
if fungus, pig, tree *1.666
abyss *1.5
attractive potion *15
-cast *1.5
DUR_NO_MOMENTUM] > 0 *1.5
Marked *1.5
ruin *1.5
slow *1.5
bat, wisp *1.5
-pot *1.333
-scroll *1.333
vitrified *1.333
contricted *1.333
petrifiying *1.3
fear *1.2
blind *1.2
mesmerise *1.2
DUR_WATER_HOLD] > 0 *1.2 (I think this is engulf)
Standing on water *1.2
pan *1.125
10% mana or less *1.125
haste * 2/3rd

>>554149325

For confusion, if -pot and confused, *2.5
if confused and you can drink, it's only tension * (9-xl/10)/4, so that's
mult @ XL,2.225@1, 2.125@5, 2@10, 1.875 @15, 1.75@20 1.575@27. This seems to contracict the comment text, which indicates that the point of this is to make confusion more significant at higher xl, but oh well.

Lastly, corrosion.
> if (you.duration[DUR_CORROSION])
> tension = tension * (10 + you.props[CORROSION_KEY].get_int() / 4) / 10;

Dur_corrosion is an enum from duration-type.h, which means it's an int and that duration[] is an array of every possible thing a player can have a duration for. Any non-zero int returned by you.duration functions as a "true" boolean.

you.props, as far as I can tell, is a catch-all for random nonsense which the program needs to keep track of. I cannot find the part of the code which takes you.props[CORROSION_KEY] and displays it in the upper right corner, so I do not know, for sure, that the value you see is exactly equal to the value stored here. However, I'll assume it is.

> tension = tension * (10 + you.props[CORROSION_KEY].get_int() / 4) / 10;
> tension = tension * (10 + X / 4) / 10;
mul@Corr
1.1@4, 1.2@8, 2@40, 4@120

This is to say, if you get extremely corroded, your tension will be multiplied by a large amount. As I have mentioned, you can intentionally corrode yourself, but there is also a trap you can step on which will corrode you.

You can also slow yourself.
Remember, however, that you're worshiping a god who will drop a quakka on yo while you're at -120 ac and slow. He probably won't. He should be happy with you.

>>554151737
from xom.cc
>xom_saves_your_life()
don't work if xom feels "nasy", which means penance or xom is "bored".
Must be above 1hp and xl1.
95%chance to fail

>if (death_tension < random2(5) || !xom_is_nice(death_tension))
> return false;
fails if current tension < a random number. I'm not sure if random2(int) returns 0,1,2,3,4 or instead 1,2,3,4,5. Generally, it's 0-4, but this might also be a float instead of an int or something.

The function is defined in random.cc, but calls .get_bounded_uint32(int) from pcg.cc, and I can't make heads or tails of that.

> bool xom_is_nice(int tension)
> const int tension_bonus = (tension == -1 ? 0 : min((MAX_PIETY - you.piety) / 2, random2(tension)));
If tension = -1, you get no bonus. While tension is often 0, it is almost never -1.

In effect, this means you can only get a peity bonus large enough to half the effective distance between your piety and 100% piety. at 50% piety, you can only receive 50 more effective piety on this roll. at 0% piety, you can receive 100 more piety from tension.

>random2(tension)
your attempted bonus is a 1d(tension), so you, in theory, want infinite tension for this nearly at all times.

>const int effective_piety = you.piety + tension_bonus;
> return x_chance_in_y(effective_piety, MAX_PIETY);
so % effective piety = % chance of nice xom, during a niceness check. at 200 effective piety, you get 100% nice xom

Oddly, when at 95% piety, you can only make use of 10 bonus piety, which means you don't need that much tension. However, at 0% piety, you want the full 100 bonus piety, which means you want well over 100 tension so that you're likely to get around 100 on a 1d(tension) roll.

>>554154752

>if (death_tension < random2(5) || !xom_is_nice(death_tension))
Anyway, if you have 5 or more tension, the left half of this will never trigger. At zero tension, some should never save you.
Xom "niceness" is % current piety witha bonus, which means it should average slightly above 50%, or 75% during infinite tension. You can, in general, just look and see when he's more ore less likely to save you.

However, it's already a 1/20 chance to even get this far, so making that 1/40 isn't that much worse.

Also, xom won't save you if you died from water, lava, draining, starvation, zotclock, rotting, or suicide

He gives you some hp back when he revives you.

So, that's one near unusable xom mechanic confirmed to be a joke, but if you've ever got less than a 1/40 chance of winning, consider an option where suicide might help you win a fight.

>void xom_tick()
>{
> // Xom now ticks every action, not every 20 turns.
> if (one_chance_in(20))
> {
Very important difference. Before, you could know that xom will only act every 20 ticks, which means you get a little time before he might act, and you can time his next action so that it might benefit you.
Now, that's gone in 0.33.

> if (old_xom_favour != new_xom_favour)
> {
> // If we entered another favour state, take a big step into
> // the new territory, to avoid oscillating favour announcements
> // every few turns.
> size += delta * 8;

this means, every time your *piety changes, your actual piety changes by 8 instead of by 1. This massively decreases the amount of average time it takes for your piety to max/bottom out, though the drift is still very slow on average. from 100 to 200, that's a free 16 piety, hypothetically. however, it works both ways.

This does mean that seeing your *piety change actually has a mildly significant effect on your %chance of nice xom actions, instead of being nearly irrelevant.

praise xom

>>553984225
I got into CDDA TLG lately. Its pretty cool i got the hang of driving. I need some help with something though. I can't make the debug menu work for the life of me. I put hotkeys on the menus that had debug on them and used them but one gives text and the other just brings up info.

>>554156262
As xom-tics are now random, and they also cause interest-decay now, xom's interest-decay is more random, meaning that you don't know for sure how many turns you earn with every point of interest. Supposedly, xom will warn you when you've got 1 point of interest remaining, meaning that you can instantly start spending resources or miscasting or using confusion on enemies, but it's still not as dependable as it was. Hypothetically, you could use up 200 interest in 200 turns, or you could use up your last interest 1 turn after xom warns you that he's getting bored.

Assuming that xom as "tic-ed" this turn,
> if (x_chance_in_y(2 + you.faith(), 6))
Assuming faith is zero, there would be a 1/3rd chance of this function continuing, instead of ending here.
I could not quite figure you.faith() out, though I did find the boolean fatih in player.cc.

Assuming that's this is the same, the function only seems able to return a boolean, which means "true" or "1" is the highest value it can return. The wiki says that an amulet of faith incenses the rate of xom actions by 33%, but it seems here that it would increase the odds from 33% to 50%. Regardless, the amulet of faith increases the odds of xom acting on each xom tic.

The variable "chance" is set to 1,2,3,4, or 5, at exactly 0, up to 5, up to 10, up to 20, and above 20.

If xom is bored, then high chance makes xom less likely to act. That's
%boredXomActs@ chance
25%@1, 21%@2, 15%@3 9%@4 and 5%@5

if xom is bored and fails to act, then xom has a chance to relent based on "chance"
0%@1, 1..25%@2 2.5%@3, 3.75%@4, and 5%@5

If you pass this check, you will get some interest.
> const int interest = random2(chance * 15);
Max 1d(75) @ 5 chance

Assuming xom doesn't act from boredom, the odds of a normal xom action are
odds%@chance
1%@1, 4%@2, 9%@3, 16%@4, and 25%@5

my fucking cars gone. i didnt die, didnt save scum, i just left it and took my bike instead and now its fucking gone. the little indicator on the map isnt there, i have no fucking clue where it went. what the fuck.
can the game just forget me saving a cars location?

>>554162345
oh christ okay i found it, it was waaaay farther away then i thought, but jesus fuck. psa: you CANNOT search on the map for the "remember this car" text. doesnt matter if when you hover over it it says "low-end cube van", searching van on the map will not find it. if your gonna do a big fucking massive 250 tile trek, fucking mark your car with a note.

File: V2Alita 090.jpg (1.49 MB, 3840x2160)

1.49 MB JPG

>>554162676
I guess this isn't a thing in that branch then?

>>554163134
i mean when you use "/" to search the map. i can search "pond" and tab through all the ponds within 900 tiles of me, but not either of my vans, or my bike

File: V2Alita 091.jpg (612 KB, 3840x2160)

612 KB JPG

Lol, wtf? They explode?
>>554163471
Yeah but original you said there wasn't an indicator there.

>>554165221
no the indicator was there, it was just WAAAAAY farther away. in my desperation, i took a screenshot of the map, and highlighted the blue color of the vehicle i had, and went into an image editor and removed all other colors to see what was left. it was there, just not in the 500 odd tile square i had first searched in. i was just a dumbass. thats after i had the idea of "oh well its just called a cube van, search that on the map" and it didnt find it because its not a searchable note.

>>554165221
hey if your playing on a 4k screen, why not use the sidebar that goes full autismo and shows ALL the stats? you could be seeing your body temp and the sun dial and all that

>>554165780
I'm just used to this one.

>>554159992
>I put hotkeys on the menus that had debug on them
Does that mean you added a hotkey for the debug menu, or something else? Because that should do it.
You can get to the debug menu from the Esc menu, but unless the hotkey is set, you can't actually access most of the functions.

>>554167587
It just displays info and lets me take screenshots. I moved to the free version and abandoned the character.

File: mywives.png (886 KB, 797x890)

886 KB PNG

>>554167678
>moved to the free version

>>554168634
I didn't know that it was made deliberately more difficult. Thought it was just a branch with more content. I don't have an issue letting in my library but the game is difficult enough as is.

File: anime shiggy diggy.jpg (36 KB, 405x431)

36 KB JPG

>>554167678
>It just displays info and lets me take screenshots.
Yeah, that's what it does when you haven't set the hotkey, which is why I'm asking if you did that.
>I moved to the free version
I certainly hope, for your own sake, that you didn't actually buy a free open-source roguelike from Steam.

>>554169264
No no no..i just downloaded it. I was ok with buying this version because it was cheap and i thought it can motivate me learn the game. Now that i can survive a bit i just moved to the github one.

>>554169264
i don't understand why those games are on steam at all. aren't they released under a creative commons license or some shit? surely some random person cannot monetize these games, so often they are just abandonware...

>>554169398
The license ensures that literally anyone can monetize them. It's not possible to restrict selling the game.

>>554169264
Another valuable lesson i learned is that inside the car does not mean storage space, only trunk means storage space. All the shit i put in the hippie car i found just disappeared. Found an air strip. I presume there are no flying vehicles in the game...got a bit disappointed because i wanted to use the fuel and apparently the station has fuel for a vehicle that doesn't exist in game.

>>554169371
>I was ok with buying this version
Lmao.

>>554169627
There are flying vehicles, but they're a bit finicky outside the BN fork. You need to pick the Helicopter Pilot trait at chargen, and the aircraft needs to be in mint condition. If it's damaged modified in any way, it no longer flies.

File: Screenshot 2026-01-17 075926.png (594 KB, 950x822)

594 KB PNG

>>554169640
he's only out $6. that's less than a happy meal these days.
TLG is available for free though, by the same person, is that correct?. it's a bit of an odd situation.

>>554169971
What does the BN fork have different that this one https://github.com/CleverRaven/Cataclysm-DDA

>>554170248
>no pockets, your inventory space is just added together
>no exodii, you can install bionics yourself by finding an autodoc
>you can build your own aircraft
>melee is easy
>less content

>>554169984
From what I understand, it was a different dev that monetised DDA.
As for:
>he's only out $6
It could be 6cent and I'd still find it funny that he is giving money to some rando and not the project.
Not that I give a shit about DDA but it's STILL funny to me.

>>554161393
nig niggit-y give me a recipe for a cake made out of chalk and eggs to bump this sad shitshow of a thread

I didn't think to save a screenshot but I had 6 orbs of fire on my screen today. I died. Been a long time since I died in Zot 5, honestly.

File: V2Alita 092.jpg (795 KB, 3840x2160)

795 KB JPG

Maybe it's just me but I kind of like the stupidly high dmg numbers you get for shooting flammable zombies with weapons that do fire dmg.

>>554117112
>I think the real smart move is just building a character that kills things at 7 range in the current gamestate
Is there anything like this except for a build that fills the screen with summons, or a build where you have to suffer until you finally reach manifold-assault storm-form ? I never had much luck with getting instant kills with blaster mages or archers.

File: 1744604410235296.png (1.53 MB, 1920x1080)

1.53 MB PNG

>>554176363
How did you get your sidebar like that? I want mine compact but I can't get the log to fully display, even if I hide other elements it just makes the pixel minimap bigger

>>554184330
There are plenty of good builds that hit out at range and instakill the kind of monsters that suck just having on your screen. Throwing javelins with Okawaru will always take care of max range threats. Manifold is the premium max range solution if you can get it online. I've also had good luck with bows in the current patch, particularly Draconians and Barachis with bows for the extra durability.

I'm not a blaster caster kind of player, but they are definitely in meta at the moment. With Vehumet you can get +1 extra range and guaranteed good spells.

File: achradamunja.png (830 KB, 1919x2157)

830 KB PNG

Really good damage on this prestige class.

>>554116198
How many of those enemies will need to be coinflipped within 8 turns, instead of beating a 95% chance 20 times in a row to kill you?
Overly reductive and if you don't know it you should've realized it.

>>554103098
Yeah dcss has some fucky bias going in the background, I swear now I fucking struggle to find a viable weapon of the type I started with everybsibgle fucking run, and the malevolent force is fucking retarded horseshit, I think I got the 3 level downward shaft more often than any other potential result. That shit was rare before, now its almost insured to hsppens once or twice per run.

>>554188360
nta but there's absolutely more than 256 enemies (or packs of enemies) over the course of a run who can kill you within 8 turns if you're not properly prepared

>>554161393
Reading this and comparing against the wiki I notice that the wiki implies Amusement and Interest are different things while being vague on how they differ, but this is just the result of the wiki writers unable to agree on a term for the same variable.

File: sidebar.jpg (703 KB, 3840x2160)

703 KB JPG

>>554186683

Finally finished outer dark for every prestige. Velite might be my favorite prestige, but frenzied one, dreamer and arsonist had some really memorable moments. Flameknight has such great single target damage it's amazing. Holy shit path of achra is amazing, how can one man be so fucking based?

>>554190673
Yeah it's really fun once you've memorised the abilities & items and can feel your brain expanding thinking of all the possibilities. I do wish it was less luck based on higher cycles as some runs boils down to praying you're not forced to fight enemies who hard counters your build. There's also the fact that some prestige classes are really underwhelming and the best way to play them is to ignore the class gimmick and use a generic build.

File: mind guy.jpg (548 KB, 3840x2100)

548 KB JPG

>>554187007
>Really good damage on this prestige class.
*laughs in mesmer*

What is this, a healthy conversation between BN, DDA and TLG?
I've been enjoying TLG a lot recently. Its much much harder than dda though so not beginner friendly. I'm surprrised my character is healthy considering he's consisting off 3 week old chocolate pretzels, clean water and multi vitamins. I should check my weight...

I really don't get this fascination with acting like other players couldn't possibly be acting reasonably. Like I would rather die than blink, like I didn't try nets or have (allegedly) higher success rate spells that ate my mana, like I haven't learned Leda's and Summon Forest and Malign Gateway and spammed Eringya's

Like all my other low-variance tools didn't fail before I tried, 8 times in a row, to win a 50/50 paralysis wand check against a faster enemy I can't manage to kill. Sometimes the backup-backup fails, man, you're a nigger if you're acting like the rest of us show up and try to o-tab through Zot with auto-train, 20 wand charges and terminal brain worms

>>554199503
Just post the morgue file so we can laugh

>>554193749
I wouldn't say that its luck based, it's based on whether he can see you at the start of the map or not. I guess inthat respect it is luck based. For example, I don't think I inuqqua is weak at all. You only lose freeze if you're unencumbered. The secret is to be encumbered and build your build around that.

>>554199503
Not reasonably
not unreasonably

Just
in a way that does not result in death.
You did not do that, because you died.

>>554199761
....Anon the last time I ownfaulted to 8 coinflips in a row was 255 games ago, back in November
...Wait

>>554189539
Wiki's fairly clear about the distinction, but isn't clear about what exactly tension is or does.

Amusement is just piety.
Interest is a countdown timer which you reset up to a new minimum based on taking a "funny" action, and its only effect is to cause "boredom" if it hits zero.
Intensity is fucking bullshit and is, due to simplified exp, far more simple now then it has ever been. It increases the chances of good things happening.

You'll note I can't define "good things" or how much of a difference it makes without being way more specific, so I can't blame the wiki for hitting the same limitation.
I'll get around to it, eventually.

>>554208927
I'm sorry anon. It's time.

>>554209939
Time for page 10.

What screenshot should i post to show my CLG character progression? Stats? Equipment? House? Fridge status?

>>554161393
Normal xom actions and bored xom actions use exactly the same formula:
> xom_acts(abs(you.piety - HALF_MAX_PIETY), maybe_bool::maybe, tension);
This is to say, when xom is mad at you, the normal xom act % formula is just as bad as the new reversed %action formula. Higher tension, and therefore chance, just means xom will be more likely to give you free tension, ending boredom.

>xom_event_type xom_acts(int sever, maybe_bool nice, int tension, bool debug)
>{
> bool niceness = nice.to_bool(xom_is_nice(tension));

The first thing the xom acts function does is call the niceness function, which we have already covered, in order to ask it if xom should beat you up. When bored, niceness returns false 100% of the time. Otherwise, it's % piety + a bonus based on 1d(tension), limited to half the missing piety

This calls xom_choose_action(bool niceness, int sever, int tension)

>if (niceness && tension == 0 && you_worship(GOD_XOM)
> && !x_chance_in_y(you.piety, MAX_PIETY))
When nice xom, no tension, and worshiping xom, you have an extra %piety chance to actually take the action, which is the same as missing%piety chance to fail the action and do nothing

> else if (!niceness && !_xom_feels_nasty() && tension > random2(10)
> && you_worship(GOD_XOM) && x_chance_in_y(you.piety, MAX_PIETY))
When mean xom, and xom isn't nasty, and tension > 1d10, and worshiping xom, there is a %piety chance to skip the negative action, and do nothing

There is a small if-else nest which retrieves one of 4 sets of xom actions. The list retrieved is good with tension, good without tension, bad with tension, and bad without tension. It then tries actions from the list randomly, removing any action considered currently invalid after attempting to use it. Individual actions have particular weights. It will try every action until either all actions are used, or it gets a valid action

An example action:
> XOM_GOOD_POTION, 500, 0, [](int /*sv*/, int tn) {return tn > 0;}

i cant breathe

>>554226492
>>554226492
That's ID, weight-with-tension, weight-without-tension, and some bullshit that becomes a boolean.

This means the category "good_potion" is selected or not selected at this stage. After this point, if somehow no potion is usable, xom would not be able to select a different category of action after the fact.

> if (!action->valid(sever, tension))
> return action, basically
The boolean "valid" does not check for targets, or tiles, or the ability to quaf potions. The boolean "valid", returns the value you see in the brackets at the end of the post above. If tn>0, potions are considered a valid category. Tn is just tension. Sv is severity, or distance from 100 (50%) piety.

Other actions may have more complicated "valid" checks, but that's on an individual basis.
>good actions

>XOM_GOOD_POTION,
>weighted 500 with tension, 0 with no tension
>valid if tension > 0

>Xom GOOD SPELL
>450, 0
>valid if tension >0 and _choose_random_spell(sv) returns at least one spell.
choose_random_spell() bumps any 0 severity up to 1.
It retreives _xom_random_spells, which is the following stack of crap:

SPELL_SUMMON_SMALL_MAMMAL,
SPELL_DAZZLING_FLASH,
SPELL_FUGUE_OF_THE_FALLEN,
SPELL_OLGREBS_TOXIC_RADIANCE,
SPELL_BATTLESPHERE,
SPELL_AWAKEN_ARMOUR,
SPELL_MARTYRS_KNELL,
SPELL_LEDAS_LIQUEFACTION,
SPELL_FORGE_BLAZEHEART_GOLEM,
SPELL_ANIMATE_DEAD,
SPELL_INTOXICATE,
SPELL_SUMMON_MANA_VIPER,
SPELL_WALKING_ALEMBIC,
SPELL_SUMMON_CACTUS,
SPELL_DISPERSAL,
SPELL_DEATH_CHANNEL,
SPELL_SUMMON_HYDRA,
SPELL_SPHINX_SISTERS,
SPELL_DIAMOND_SAWBLADES,
SPELL_MALIGN_GATEWAY,
SPELL_DISCORD,
SPELL_DISJUNCTION,
SPELL_SUMMON_HORRIBLE_THINGS,
SPELL_SUMMON_DRAGON,
SPELL_FULSOME_FUSILLADE,
SPELL_CHAIN_OF_CHAOS
for each of these, it runs
>spell_is_useless(spell, true, true, true)

>>554233680
f at least 1 of these has a use, then XOM_GOOD_SPELL is valid and can be selected as an action category with its full weight.

is_uselsss runs:
>string spell_uselessness_reason(spell_type spell, bool temp, bool prevent, bool skip_casting_checks)
This is in spl-util.cc
If you're out of bounds or not "on_current_level", then this returns false, which implies it is possible to turn off all xom spells through some glitchy fuckery.
Two other checks are skipped in our case, and then every indevidual spell has its own conditional logic. The logic is odd.

For polar vortex, ledas, and dragon call, it checks your associated "duration" variable to see if that spell is already active. A lot work like this.
For ozocubus, it does not check duration, and instead checks to see if you are wearing a specific artifact ring
some have fairly detailed conditions.
Small mamals will simply always claim to work.

And so, if I'm understanding this correctly, there is absolutely no point in doing this check for this purpose, other than letting the devs change xom later and have him still work.

>XOM_GOOD_CONFUSION
>440, 0
Valid if at least one "nearby" monster is able to attack. However, note that actions with relevant weight 0 do not get added to the list for consideration, so it is not possable to get an action with weight zero, even if it would hypothetically be "valid"


>XOM_GOOD_ENCHANT_MONSTER
>420, 0
valid if at least 1 nearby monster is "enchantable"

>XOM_GOOD_SINGLE_ALLY
>400, 60
valid if you can have allies and either there is any tension or 75% of the level is currently unexplored
return (tn > 0 || (_exploration_estimate(false) < 25)) && !you.allies_forbidden();}
Wiki says you can only get an ally before 75% exploration, which is wrong.

>XOM_GOOD_POLYMORPH
>275, 0
Valid if xom has a poly target. He has his own function just for this. I'm not checking it.

anon posts worst bump method, asked to leave thread.

>>554235584
I'm okay with the ausitm, but still, kek

>>554209558
>Amusement is just piety.
Are you sure? Because in the wiki, piety and amusement are definitely different and it's amusement and interest which are ambiguous.
http://crawl.chaosforge.org/Xom
>Piety starts at 100 and changes by 1 point roughly every 20 turns (about 49% chance to move towards 100 and 51% chance to move away).
>when you start worshiping Xom, its amusement is set to between 2 and 80. Every 20 turns, there's a 50% chance that Xom's amusement will be decreased by one, until it reaches 0.

>>554234904
>XOM_GOOD_ALLIES
>280, 0
valid if above 4 tension and you can have allies. This is the first action which requires higher than minimum tension.

>XOM_GOOD_FORCE_LANCE_FLEET
>230, 0
valid if above 4 tension and there are at least 6 non-solid spaces around you, at radius 2, I think, or radius 1.
>radius_iterator ri(you.pos(), 2, C_SQUARE, LOS_NO_TRANS, true)

>XOM_GOOD_CLEAVING
>160, 0
As this has a weight of zero during 0 tension, it's validity formula will always return true when it is actually prompted. As such, the tension > 0 check is meaningless. From here on, in such cases, I will simply say an action always returns valid.
Always valid.

>XOM_GOOD_FLORA_RING
>130, 0
returns valid if BOTH
at least 3 spaces within 2 tiles that can have a plant but are currently not occupied AND
>smiters < div_rand_round(you.experience_level, 6) - 1;
div_rand_round divides #one by #two and then has a 100*decimal value% chance of rounding up.
So, at XL 6, div_rand_round returns 1. AFter -1 that's zero, meaning that if there are more than zero smiters in view, you cannot receive flora_ring.
At XL 9, div_rand returns 1 50% of the time and 2 50% of the time, meaning that there is a 50% that xom will place a floral ring with 1 smiter in sight.
At XL 27, div_Rand returns and 4 50% of the time and 5 50% of the time, meaning xom will not add flora ring 50% of the time if there are 3 smiters in sight.

>XOM_GOOD_DESTRUCTION
>125, 0
valid if you see at least 1 non-projectile non-tentacle monster who isn't touching you when either they will attack you or xom feels nasty.

>>554237965
Yep seems you're right.
Piety is just called piety by the wiki and amusement = interest.

Having two words for it is a bit confusing, but they don't fuck up like I just did.

>You'll get a warning when the amusement gets down to 1 (at least 20 turns before it reaches 0): "Xom is getting BORED". If your interest reaches zero, your title will always be "A BORING thing", regardless of your piety level.
The first mention of interest happens when they describe that amusement is about to hit zero, and then say that interest does hit zero.

It's internally consistent, is what I'm saying, even if it might be a bit not-user-friendly.

tome4, for the second part of Anorithil's Darkest Light: "Converting X% of the damage [enemies] deal between Light and Darkness"?

How is the conversion handled? Does it select one or the other to fully convert the damage to or does it convert half to Light and half to Dark?

>>554243067
iirc it's split half half

>>554243887
Thank you

>>554238124
>XOM_GOOD_DOOR_RING
>110, 0
Multiple checks.
Within a 5 block radius, if there are at least 25 tiles 3 or more distance away from you which can be replaced with a door
AND
if the number of ground tiles at distance 6,7 or 8 from you is greater than or equal to the number of monsters which will attack and which are 1 or 2 tiles from you
Better stated as
potential doors > 5 AND closeDIckheads <= distant spots

>XOM_GOOD_MASS_CHARM
>87, 0
tension > 4 and at least 1 enchantable enemy

>XOM_GOOD_HYPER_ENCHANT_MONSTER
>70, 0
valid if there's an empty space in 2 tiles, or
valid if there's a "weak" monster according to _xom_find_weak_monsters(false)
I'm just assuming this grabs a weak ally and I'm not checking what xom finds weak.

>XOM_GOOD_FOG
>65, 0
Valid if you can have a cloud on you

>XOM_GOOD_FAKE_DESTRUCTION
>60, 0
valid if there's an enchantable monster "nearby"

>XOM_GOOD_TELEPORT
>40, 590
Good teleport is given a very high weight if you arn't currently in combat, meaning it is unlikly to bail you out, but it can. Don't wait on this. It's the rarest action in combat so far, I think.
Invalid if you're in abyss or can't teleport.
valid if over 20% of the level is unexplored
Between 20% and 0% of the level remaining unexplored, has a % chance of being valid anyway. Chance is explored%/110% of being invalid.
9.1%@100%
18.2%@90%
27.3%@80%
odd of being valid regardless. However, 80% won't actually use this part of the formula, since it was already declared valid in the prior step.

>XOM_GOOD_LIGHTNING
>21, 0
valid if at least 1 monster that will attack within 2 tiles of you, AND you are "in a dangerous place"
player_in_a_dangerous_place is in nearby-danger.cc. It take log(you.experience) and compares that to a gen_threat and a hi_threat

> _monster_threat_values(&gen_threat, &hi_threat, invis);
> return gen_threat > logexp * 1.3 || hi_threat > logexp / 2;

>>554246882

gen_threat adds the log(exp) of all nearby monsters together.
hi becomes the log(exp) of the largest nearby monster

If log(you.exp)/2 is less than log(biggest enemy.exp), you are in a dangerous place. Log seems to refer to the natural logarithm, or ln.
exp values tend to be 200 (low) 600-800-1200(mid) and like 2000-4500 (high) which means enemies provide around 5.3@low, 6.3-7@mid or 8..4@high.
The player, at differnt xl's, provides
4.9@5
8.2@10
11.2@15
12.5@20
13.3@25
13.6@27
So, even a single yak puts you "in danger" at level 12. 3 yaks will always put you in a dangerous place. A single specialized draconian puts you "in danger" no matter what.
Only little popcorn basic orks and the like don't put you in danger, ever, as ln(1) = 0.

So, during almost any reasonable combat after floor 1, good_lightning is valid.

>XOM_GOOD_WAVE_OF_DESPAIR
>15, 0
if at least 1 enchantable monster, it's valid.

>XOM_GOOD_SNAKES
>405, 0
Valid if there's a hostile snake, I think

>XOM_GOOD_LIGHT_UP_WEBS
>205, 0
Invalid if you are on a web with less than 27-4*fireResistancePips hp
Otherwise, valid if you can see a web.

>XOM_GOOD_ANIMATE_MON_WPN
>325, 0
valid if tension above 4 and there's a valid target andyou can have allies

>XOM_GOOD_DIVINATION
>420, 500
Under high tension (above 24), has a (unexplored% / 120%) chance of being valid.
Under lower tension, valid if at least 20% has already been explored.

Oddly, this means it works on mostly explored maps at low tension and it works on mostly unknown maps at high tension. This means you can maximize or minimize the odds of getting this action. It also means that it's unlikely to trigger twice during low tension, but very likely to trigger once on a new floor at low tension.

> XOM_GOOD_SCENERY
>120, 135
Only valid at 4 or lower tension.
otherwise, valid if _xom_scenery_candidates() returns anything, or else valid anyway 33% of the time.

>>554248675
>XOM_GOOD_CLOUD_TRAIL
>0, 375
valid if any part of the floor has been explored and you don't already have a cloud trail

>XOM_GOOD_BAZAAR_TRIP
> 60, 280
invalid if you're already in bazzar.
invalid if you have less than 777+sv*(4+ tripsSoFar*2)

This is the first use of severity, which ranges from 0 to 100.
if piety = 100 and so severity = 0,
you need > 777 gold to get this trip, always.

if piety = 0 or 200, severity = 100
you need
1100@0 trips, 1300@1 trip, 1500@2 trips, and so on.

Assuming you've got gold, this returns invalid between 66% and 83% of the time on floors you've explored at least 80% of. It's less likely to be valid on more explored floors.

At this point, If tension is above 28, this action is valid.
Otherwise, there is a 1/(7*trips) chance of this action being valid. zero returns valid here.

>XOM_GOOD_MUTATION, 32, 400
Valid if 3 conditions are all true.
If you can mutate safely
a 16 / you.how_mutated()% chance, which I think just grabs your current number of mutation levels.
1d(tension) is less than 5.

>XOM_GOOD_RANDOM_ITEM
>33, 775
Only valid if tension < 20
> XOM_GOOD_ACQUIREMENT
>21, 295
Only valid if tension < 20


And that's it for positive xom actions

heads up, if your cdda game saves during an autoloot/autosort, and you quit/exit, itll instantly crash when you go to load it

File: GrinDog.jpg (22 KB, 720x540)

22 KB JPG

>>554253781
heads up, dda sucks

>>554254420
yeah its not for casuals

>>554255714
>watching a green bar spin makes me hardcore
Thanks, DDA fans, you always make me laugh

I liked CDDA a year ago, recently got The Last Generation and didn't like it at all.
Whose idea was it to add random enemies with guns everywhere and reduce the amount of loot you get from a location at the same time?

>>554257631
probably the guy above

>>554259061
Nope, TLG is just DDA with extra autism, it sucks too. Bright Nights is the only Cataclysm fork that still retains any of the feeling of Whales' original game. Bright Nights kinda sucks in its own ways, but I like it best of all the forks I've tried.

>>554251890

>XOM_BAD_NOISE
>975, 415
valid if not under xom penance. Otherwise, valid if any tension.

>XOM_BAD_MISCAST_PSEUDO
>825, 715
valid if not under penance

>XOM_BAD_ENCHANT_MONSTER
>790, 0
valid if 1 or more enchantable monsters

>XOM_BAD_BLINK_MONSTERS
>680, 0
valid if any blinkable victim

>XOM_BAD_CONFUSION
>610, 0
valid unles you've got clarity

>XOM_BAD_CHAOS_UPGRADE
>590, 0
I think valid if any nearby monster can receive a chaos brand

>XOM_BAD_SWAP_MONSTERS
>540, 0
valid if any _rearrangable_pieces(), I think.

>XOM_BAD_TELEPORT
>475, 1625
invalid if in abyss while unable to teleport,
If xom is nasty and you've explored more than 40% of the floor, this is invalid.
If xom is nasty and tension is above 10, this is invalid.
Otherwise, there's a % chance that this will be valid. at exlored% 100% this is invalid. So, assuming the other
valid%@ exloration%
0%@99%
50%@79%
100%@59%
Here's the logic tree for that
> return ( ( !player_in_branch (BRANCH_ABYSS) || _teleportation_check() ) )
> && ! ( (_xom_feels_nasty() && (explored >= 40 || tn > 10) )
> || (explored >= 60 + random2(40) ) );


>XOM_BAD_POLYMORPH
>380, 0
valid if any poly target available


>XOM_BAD_SUMMON_HOSTILES
>170, 670
valid if less than 49 tension

> XOM_BAD_TIME_CONTROL
>133, 0
valid if you don't have haste, slow, or paralysis

>XOM_BAD_BRAIN_DRAIN
>100, 0
valid if more than 3 mana

>XOM_BAD_FAKE_SHATTER
>75, 0
valid if more than 3 doors and diggable tiles in your vision

>XOM_BAD_CHAOS_CLOUD
>58, 0
valid if there is not a cloud on you


>XOM_BAD_GRANT_WORD_OF_RECALL
>, 80, 0
invalid if tension is above 24
invalid if player XL is below 4
invalid if player XL is below 6 while under penance
valid if 1 or more nearby monsters can be forced to use recall

>XOM_BAD_SEND_IN_THE_CLONES
>45, 0
invalid if tension above 24
valid if at least 3 spaces to put player illusions within 2 tiles

>>554270297
>XOM_BAD_DOOR_RING
>38, 0
invalid if less than 4 tension
invalid if less than 31 door-replaceable tiles at 3,4, or 5 distance from you.
Adjacent space, I think, is equal to the floor tiles you can see which don't have a monster capable of attacking.
adjacency_cap is 2+ XL/6, with rounding dependent on the decimal remainder. minimum 2 maximum 7.
Adjacent monsters is all monsters you can see within 2 tiles distance.
Moveable_hostiles is every other hostile you can see

invalid if there are more monsters within 2 tiles than your adjacency cap.
invalid if there are more hostiles visible at distance greater than 2 than there are visible floor tiles not currently occupied by a hostile monster.
After that, valid if at least 2 adjacent monster, or at least 2 movable monsters.

>XOM_BAD_TORMENT
>6, 0
valid if you don't resist torment. Wiki says "immune" to torment, but that might be what this means. Function is you.res_torment(). Check it or don't. Probably, just test it in game by existing with rN+. Very rare.

XOM_BAD_MUTATION
>210, 1010
valid if tension under 9 and you can "safely" mutate.

>XOM_BAD_DRAINING
>6, 160
Valid if not rN+++

>xom_maybe_reverts_banishment(true, true)
>2, 1
Valid if you're not in abyss, or on the abyss "stack" (which I think means bazarrs or ziggs started in the abyss) and _allow_xom_banishment() returns true.

allow-banish is mostly complicated for the joke.
Under penance, he'll banish.
If nasty, he rolls 5/XL% chance to disallow banishment.
Xom checks if you're XL is over 6 if during penance or over 9 normally.
If your level is not high enough, xom has a chance of allowing banishment, but it's a fake banishment so the odds don't matter.
If your level is high enough, xom rolls 5/XL% chance to disallow banishment.

>>554270432

>XOM_BAD_MOVING_STAIRS
>215, 0
valid if there's at least one _nearrby_stairs()

>XOM_BAD_FIDDLE_WITH_DOORS
>95, 5
invalid if xom is nasty
invalid if there are no valid doors
valid if either 96% of the map is explored or tension is above 4

>XOM_BAD_CLIMB_STAIRS
>55, 0
invalid if no nearby stairs
invalid if you're standing on a shop... for some reason
valid if you're currently on "stairs".

In this case, the folloiwng are valid "stairs"

portal_entrance branch_entrance portal_exit branch_exit altar
DNGN_STONE_STAIRS_UP_I:
DNGN_STONE_STAIRS_UP_II:
DNGN_STONE_STAIRS_UP_III:
DNGN_ESCAPE_HATCH_UP:
DNGN_EXIT_HELL:

DNGN_STONE_STAIRS_DOWN_I:
DNGN_STONE_STAIRS_DOWN_II:
DNGN_STONE_STAIRS_DOWN_III:
DNGN_ESCAPE_HATCH_DOWN:
DNGN_ENTER_ABYSS:
DNGN_EXIT_THROUGH_ABYSS:
DNGN_EXIT_ABYSS:
DNGN_ABYSSAL_STAIR:
DNGN_ENTER_PANDEMONIUM:
DNGN_EXIT_PANDEMONIUM:
DNGN_TRANSIT_PANDEMONIUM:
DNGN_TRANSPORTER:
DNGN_ENTER_SHOP:

So, all of those allow this xom action, other than the shop. Shops are stairs.

And that's the end of validating xom actions. Now, tims to see how each action actually gets implemented, where that matters.

File: oh shit.jpg (272 KB, 764x808)

272 KB JPG

oh cool ive never seen lava before. oh fuck im dead arnt i

>>554271745
You've got a hazmat suit, just stay back from the heat and you'll be fine

>>554270584
So, for selecting valid actions, the only use of severity is to make you require more money to reach a bazaar.

>xom_take_action(action, sever);
This does not receive tension at all, only severity, meaning that tension is not relevant past this point.


If xom gives you a bad action while bored, there may be a message about it, based on an "option" which I think is a setting the player can change somehow.
yet another function handles the actual action, so just assume some "did something" at this point.
> _do_xom_event(action, sever);

If xom was bored and took an action, xom will now give the player some interest, solving the boredom. New interest is 2d(badness*60)/2 basically, with a max of 255. So, on average, badness*30.
If xom's action itself killed you when he isn't bored and you're not under penance, he should revive you.
If worshiping xom, there's a 1/5 chance for xom to randomize your piety. 1d(200).
If xom was previously bored, it updates your favor status.

That's the end of a xom_take_action.

_do_xom_event() grabs a "map" which was built in some earlier part of the program which maps each action name to a specific function someplace, and runs that function with severity


>XOM_GOOD_POTION
runs
>_xom_do_potion(
Attempts to drink potions from a weighted list, removing each failed potion.
weight 10, CURING,
10, HEAL_WOUNDS,
10, MAGIC,
10, HASTE,
10, MIGHT,
10, BRILLIANCE,
10, INVISIBILITY,
5, AMBROSIA,
5, ATTRACTION,
5, BERSERK_RAGE,
1, EXPERIENCE);
Having full hp, full mp, no curable status effect, -berserk and maybe clarity all raise your odds of getting invisibility, haste, or experience. Highest odds of exp pot are 1.96%. 1.6% without clarity or -berserk

Potion power, whatever that is, is 40 for exp or 150 for everything else. Severity irrelevant.

>>554274006
yeah, uh, i wasnt wearing one until i just found it in the room next door

>>554277097
It's fine, it's not radioactive, just gives off massive amounts of heat, and I think some noxious gasses? Been a long time since I went up to one to grab some chunks of sulfur for fungicide

File: file.jpg (141 KB, 392x309)

141 KB JPG

>AI generated pesudo codebabble bumps

>>554276480
>XOM_GOOD_SPELL
>_xom_random_spell()
>spell = _choose_random_spell(sever);
will cover that in a moment
>power = sever + you.experience_level * 2 + get_tension() + you.runes.count() * 2;
The spell xom casts has power based on your XL, the current tension, and your rune count.

Assuming XL max and all runes, that's severity + tension + 84. Tension, I think, tends to be around 5 or 20 or something. We haven't concluded what to expect from tension during different branches. Severity is 50 on average, but you can see clearly when it's high or low.
So these spells tend to have around 89 to 100 spellpower, which may function like normal spellpower. At endgame, that's 150 spell power or a slight bit more when it counts.

>_choose_random_spell()
To find your list of valid spells, start with the list of all xom spells in order, seen way above.
Then, only take the first (severity) number of those spells. There are only 26 spells, so an intensity above 26 will grab every single spell. An intensity of 1, however, will only grab "summon small mammal"
Then, remove all spells which are not valid by their own special definitions.
Lastly, cast one of the remaining spells at random, with the spell power from earlier.

>XOM_GOOD_CONFUSION
for monsters in view which will attack, has a 95% chance of attempting to cast confuse
>_confuse_monster(*mi, sever);

> if (mons->add_ench(mon_enchant(ENCH_CONFUSION, 0,
> &env.mons[ANON_FRIENDLY_MONSTER], random2(sever) * 10)))
I believe this confuses the monster for 1d(severity)*10 duration, so up to 100 turns and proportional to severity%. Or, I'm wrong.

>XOM_GOOD_ENCHANT_MONSTER
>_xom_good_enchant_monster(
Good enchant monster and bad enchant monster call the same function, which handles them in parallel.
>_xom_enchant_monster
Paying attention only to the positive version
cap = 1 + div_rand_round(xl, 7);
So, cap is 1, and 2 at xl 7, and 3 at xl 14, 4@21.
Cap is then set to 1d(cap-1)+1

File: dcssannoyingfollower.png (339 KB, 640x590)

339 KB PNG

>play Armataur
>entire race designed around charge ability
>get forced into Beogh by smite RNG on Dungeon 2
>this dumb follower constantly gets in the way of my charge so I can't get my regen stacks
Fuck it. I should have just taken my game over instead of going through this Beogh humiliation ritual.

>>554279920
In the description for good-spell, I said intensity when I meant severity a few times.

Also, good_enchant refers to the negative set of enchantments, which means I must outright redo that section.

>retcon
>XOM_GOOD_ENCHANT_MONSTER
>_xom_good_enchant_monster(
Good enchant monster and bad enchant monster call the same function, which handles them in parallel.
>_xom_enchant_monster
Paying attention only to the GOOD version

cap = 1 + div_rand_round(xl, 5);
Cap is eventually set to a random number from 1 to 1+(xl/5), so 1 to 7 at most.
the 4 potential enchantments are weighted based on XL, so that's best represented in a table.

enchantment, @XL0, @XL27
petrifyi, 200, 308
paralyz, 300, 165
vitrified, 300, 165
slowed, 200, 308
That means xom's helpful enchants get somewhat worse as your XL goes up, though they are still almost as good, and are a death sentence half the time.

Xom distributes (cap) number of enchantments to monsters.

The duration of the enchantments is based on severity:
for petrify and paraly:
duration 30+ 1d(severity/10), so like 3 turns period.
for vitrifi and slow:
duration 200 + 1d(severity), so 20 to 30 turns, generally 25.

>>554280503
You can use t to force your follower to retreat in particular directions. That way, when combat starts, you can get your charge while he returns from the side.

Imagine having permanent flight and falling in a shaft anyway.

>>554220912
Depends on what you got I guess. Cool bionics? Cool guns? Cool car?

>>554276480
But what is badness?

File: V2Alita 093.jpg (552 KB, 3840x2160)

552 KB JPG

>You can keep me safe
Lol.
I do wonder how the formula actually works these out.

File: V2Alita 094.jpg (672 KB, 3840x2160)

672 KB JPG

>>554293482
That's... actually not to bad.
No annoying things like mood swings or stuff like that, I've seen much worse and I've got an entire stack of CBMs to gathering dust I can install to make her even better.

>>554291939
Badness is a variable present in the "map" object called xom_events. A single event in xom_events contains:
1: action name, which corresponds to the action names used by the xom_chooos_action section I have described in detail above.
2. function name, which links an action name to an actual function somewhere in xom.cc
3 for "bad" actions, it contains an int which is "badness". Badness is, at most, 50.

xom_event_badness() returns badness/10 for most negative events, and either 3 or 1 for BAD_Teleport.

in other words, 2d(badness*60)/2 is, at most, 2d(5*60)/2, or 2d(300)/2, but you can't have more than 255 gift_timeout, which is also the same as xom's "interest" timer.

I forgot to clarify that, so thanks.

Whats the point of having all of these fucking mods available if none of them work on the latest version. From 40-50 mods i installed with the launcher of CDDA i had to restart world creation 10 times just to go through the stuff that the game can't even generate. I'm glad that i am able to customize the plunger with which i can bash my head in but the conversion mods or whatever else just give an error because it requires a version of the game from 98. Fuck these lazy assholes. I had a better time troubleshooting my skyrim install and that one has over 1000 conflicting mods. Is it that hard to not make things conflict in an ASCII game?

File: survivable.jpg (127 KB, 506x538)

127 KB JPG

fug

>>554307173
How did you manage that?

>>554295306
But what is baldness?

I think I am ready for a proper roguelike. Does Bright Nights work with Magiclysm?

>>554271745
Lava is great to have around, it destroys any item you put there. Now that in CDDA a pile of ceramic shards instagib a road roller drum and you can't destroy items that way anymore, I always map edit a tile of lava somewhere.

>>554307173
i dont play dda what is going on here

>>554321438
Bright Nights has its own fork of it, but I think it's the same thing more or less.

File: cata-2026-01-18-22-02-04.png (49 KB, 1920x1080)

49 KB PNG

>environmental storytelling

>>554332476
What is the story

>>554339847
That the people who lived at this farm ate people.

>>554339985
Just creeps, though?
https://www.youtube.com/watch?v=XFkzRNyygfk

>>554073435
four tiles like a long motorcycle can just barely fly under its own power fyi
>>554171073
yeah but that's TLG, not DDA, you retard
>>554186683
have you tried having a 4k screen
>>554253781
autosave is very broken, yes

>>554261673
none of them have the constant ambient spawning due to noise and needing to keep your noise level below 80 for vehicles, among other things
static spawns were a mistake

>>554294667
oh yeah, juice that girl full of cow or whatever and she'll be right

>>554345949
>constant ambient spawning due to noise
I'm fine with that, zombies constantly teleporting down (and usually "behind u") from the Starship Zombieprise was janky and annoying.

>>554346268
yeah but I'm acutely missing it right now for some insane reason
presumably brain damage

>>554346384
I think a good way to bring some of it back would be if zombies had a dormant mode, where they'd basically go into hibernation and wait. Maybe they'd want to crawl into or under things first.

Then instead of hundreds of ambulatory eyes keeping you out of the city, you'd have a bunch of sleeping zeds you could creep past (making use of the underutilised stealth system) as long as you didn't make any noise, and as long as no NPCs or dogs or Mi-go came strolling by making noise and attacking everything.

Maybe have an occasional watchmen zombie, either specific subtype or just a random pick, who stays on his feet like a lookout, so you have to avoid line-of-sight for him, but it's more doable because it's just one instead of thirty or more.

>>554349512
Like, imagine slipping into a city to try to reach a garage and get a wrench, and there are dormant zombies and other blob monsters snoozing here or there, some of them hidden under cars or whatever. Don't prybar that door, the noise will wake the one nearby! Can't loot this building, the alarm will ensure everything within four map tiles will wake up and you'll be surrounded.
A handful of zombies spot you? Can you break LoS and hide somewhere until they go dormant again in X number of hours?

Could be fun.

File: cata-2026-01-19-01-14-15.png (48 KB, 958x915)

48 KB PNG

rate/hate

>>554352921
Four foot tall bird girl melee specialist with deft, night vision, and solid stats? Not much good at carrying your burdens, but otherwise a good choice for companion. B+, would give a katana to.

>>554352921
she needs at least eight doses of bird serum, lead by a shot of purifier serum
I've never seen a survivor that didn't make me want to kill myself before, that's incredible

>>554281216
>XOM_GOOD_SINGLE_ALLY
>_xom_send_one_ally
"strength" = 1d(5*severity), so 0 to 500

xom_random_pal(strength) returns the type of monster you're getting.

So, in xom_random_pal():
Strength is either <= 130, <=300, or >300, and that's all that matters about it.

Based on that and XL, you get a "variance" modifiire, which is between -4 and 3.
Here is a table of "variance" at different "strength" and "XL"s:
strength, Xl<10, Xl<19, Xl>=19
<=130, -4, -4, -4 to -3
<=300, -2 to -1, -2 to-1, -2 to 0
<=500,-1 to 1, 0 to 2, 1 to 3

There isn't really a reasonable English way to say that.
Take a variance value from the above table. Add your XL. Potentially add a value from the following three cases
+ 1 to 4 if spawning an ally
+2 to 4 if xom bored
+4 to 6 if xom penance

Lastly, if variance is greater than 33, set variance = 33.

> monster_type mon_type = xom_picker.pick(_xom_summons, variance, MONS_CRIMSON_IMP);
This is yet another function that gives the monster type we're getting to the function which called it, so we're 3 functions deep at this point.

_xom_summons is a vector, which is basically an array, containing the following information:
{ -4, 26, 5, FLAT, MONS_BUTTERFLY },
{ -4, -1, 5, FLAT, MONS_QUOKKA },
{ -3, 6, 60, SEMI, MONS_CRIMSON_IMP },
{ -1, 6, 40, SEMI, MONS_IRON_IMP },
{ 2, 8, 40, SEMI, MONS_SHADOW_IMP },
{ 3, 8, 50, SEMI, MONS_UFETUBUS },
{ 3, 8, 40, SEMI, MONS_WHITE_IMP },
{ 4, 8, 5, SEMI, MONS_MUMMY },
{ 4, 10, 75, SEMI, MONS_PHANTOM },
{ 5, 10, 10, SEMI, MONS_BOMBARDIER_BEETLE },
{ 5, 12, 90, SEMI, MONS_SWAMP_DRAKE },
{ 5, 12, 25, SEMI, MONS_WEEPING_SKULL },
{ 7, 14, 50, SEMI, MONS_ORANGE_DEMON },
{ 7, 15, 95, SEMI, MONS_SHAPESHIFTER },
{ 8, 14, 40, SEMI, MONS_ICE_DEVIL },
{ 8, 14, 40, SEMI, MONS_RED_DEVIL },
{ 8, 15, 60, SEMI, MONS_BOGGART },
{ 8, 20, 155, SEMI, MONS_CHAOS_SPAWN },
{ 9, 14, 60, SEMI, MONS_HELLWING },

>>554356536

{ 9, 14, 5, FLAT, MONS_TOENAIL_GOLEM },
{ 9, 14, 30, SEMI, MONS_VAMPIRE },
{ 9, 16, 50, SEMI, MONS_YNOXINUL },
{ 10, 14, 40, SEMI, MONS_HELL_RAT },
{ 10, 14, 30, SEMI, MONS_KOBOLD_DEMONOLOGIST },
{ 10, 15, 90, SEMI, MONS_ABOMINATION_SMALL },
{ 10, 16, 50, SEMI, MONS_UGLY_THING },
{ 10, 17, 50, SEMI, MONS_SOUL_EATER },
{ 10, 18, 85, SEMI, MONS_RUST_DEVIL },
{ 11, 17, 50, SEMI, MONS_SMOKE_DEMON },
{ 11, 17, 30, SEMI, MONS_DREAM_SHEEP },
{ 11, 20, 80, SEMI, MONS_WORLDBINDER },
{ 11, 22, 50, SEMI, MONS_NEQOXEC },
{ 12, 16, 30, SEMI, MONS_OBSIDIAN_BAT },
{ 12, 18, 30, SEMI, MONS_TARANTELLA },
{ 12, 18, 15, SEMI, MONS_DEMONSPAWN },
{ 13, 19, 75, SEMI, MONS_SUN_DEMON },
{ 13, 20, 75, SEMI, MONS_SIXFIRHY },
{ 13, 20, 15, SEMI, MONS_GREAT_ORB_OF_EYES },
{ 13, 21, 105, SEMI, MONS_LAUGHING_SKULL },
{ 14, 22, 90, SEMI, MONS_ABOMINATION_LARGE },
{ 14, 22, 105, SEMI, MONS_GLOWING_SHAPESHIFTER },
{ 15, 22, 50, SEMI, MONS_HELL_HOG },
{ 15, 23, 1, FLAT, MONS_OBSIDIAN_STATUE },
{ 16, 22, 10, SEMI, MONS_BUNYIP },
{ 16, 23, 50, SEMI, MONS_RADROACH },
{ 16, 25, 75, SEMI, MONS_VERY_UGLY_THING },
{ 16, 24, 25, SEMI, MONS_SPHINX_MARAUDER },
{ 17, 24, 35, SEMI, MONS_GLOWING_ORANGE_BRAIN },
{ 17, 33, 35, SEMI, MONS_SHADOW_DEMON },
{ 17, 33, 65, SEMI, MONS_SIN_BEAST },
{ 17, 33, 15, SEMI, MONS_CACODEMON },
{ 18, 25, 50, SEMI, MONS_GUARDIAN_SPHINX },
{ 18, 33, 35, SEMI, MONS_REAPER },
{ 18, 24, 30, SEMI, MONS_DANCING_WEAPON },
{ 19, 26, 1, FLAT, MONS_ORANGE_STATUE },
{ 20, 33, 30, SEMI, MONS_APOCALYPSE_CRAB },
{ 21, 33, 30, SEMI, MONS_TENTACLED_MONSTROSITY },
{ 22, 33, 1, FLAT, MONS_STARFLOWER },
{ 23, 33, 30, SEMI, MONS_HELLEPHANT },
{ 24, 33, 5, SEMI, MONS_MOTH_OF_WRATH },
{ 25, 33, 5, SEMI, MONS_NEKOMATA },

xom_picker is a monster_picker defined in mon-pick.h
A monster_picker is a type of random_picker
random_picker is defined in random-pick.h

>>554311147
>>554332047
screwing around with debug. was checking some stuff out and needed to pass time, so set it as invincible, invisible, and set it to wait 24 hours. on the second 24 hours i guess the shadow showed up and started mass spawning amalgamations. theres already 400-500 dead ones there.
strangely, it didnt slow down the passage of time. you know how if your near a refugee center and sleeping takes forever? this still passed at 10 seconds per every 4 seconds so my computer sounded like some horrifying gangbang with all the sound effects going on. BLURGH every 15 milliseconds basically. im guessing the shadow brought the necroboomer? because i was nowhere near anything with anything still alive

>>554332476
the hunting lodge sometimes has a LOT of cooked long pig in it

>>554357251
Regarding the values in the table above.
The 5th value in each entry is just the name of the monster which will be returned if selected.
The 4rth value, flat or semi, decides which of two functions is used to calculate the "rarity" of a particular monster given the supplied variance value.
The 1rst value is the minimum variance value which is required in order for a monster to be considered in a "pick" calculation.
The 2nd value is the maximum variance value which is allowed before a monster is no longer considered in a "pick" calculation.
The 3rd value is a default "rarity" of an entry in a random picker.

If a monster is "flat" and "variance" is equal to or between minimum and maximum, then the rarity of that creature for this calculation is just equal to their rarity value.
So, if you submit a "variance" of -2, then the "quokka" would be considered valid and would have a rarity of 5.

Rarity is a weight. The "quokka" in the above example would have a 5/totalRarity chance of being selected.

If a monster is "semi" instead of "flat", then the rarity, for this calculation, is base_rarity * between 100% and 50% depending on the exact variance. Specifically,
rarity * (the distance of variance from the midpoint between minimum and maximum)/(the distance between minimum and maximum)
This is not the actual formula, but the result is equivalent. Keep in mind that variance can never be greater than max or smaller than min, and so can never be further from the midpoint than (max-min)/2.

If you're unsatisfied with that, here you go
>ln is 2* max-min, or 2 if it would be zero
>rarity* len- abs(min+max-2*depth) /len
> return rar * (len - abs(pop.minr + pop.maxr - 2 * depth))
> / len;

>>554367741

Anyway, an array is constructed which associates each valid monster with it's current rarity value and adds that value to "totalrar"
We wet totalrar = 1d(totalrar)
We subtract, in order, the rarity of each monster from total rar until totalrar <= 0, at which point we select that monster type.

So, it's just a weighted random result.
> monster_type mon_type = xom_picker.pick(_xom_summons, variance, MONS_CRIMSON_IMP);
Returns the above weighted random result.


Then, depending on branch, xom has a chance to just replace this monster type with a different monster type.

In zot and zig, there's a 1/13rd chance of xom rolling:
MONS_DEATH_COB @ weight 5
MONS_KILLER_KLOWN, @ weight 4
MONS_PROTEAN_PROGENITOR @ weight 3
MONS_CURSE_TOE @ weight 2

in pan or zig and the previous roll failed, there's a 1/13rd chance of xom rolling:
60% MONS_DEMONSPAWN_BLOOD_SAINT
40% MONS_DEMONSPAWN_CORRUPTER

in zig, if the previous roll failed, there's a 25% chance of xom rolling:
5/7ths MONS_PANDEMONIUM_LORD
2/7ths MONS_PLAYER_GHOST

And that's what _xom_random_pal(strengthRoll, true); does, which will matter to other functions later.

Finally, back in
>_xom_send_one_ally(
It seems the summon has a default duration of 6, however long that actually is.
> mg.set_summoned(&you, MON_SUMM_AID, summ_dur(6));

Xom adds 100 to 200 points to the summon timer, gives 1000 to the regeneration timer of that monster, and gives a "chaos upgrade" to the weapon of any reaper he may have just summoned.
It's not hard to test, but your monster was just given summ_dur(6) and then +100 to 200 extra summon timer, which could mean 16 to 26 turns, or 10.6 to 20.6 turns, or something else. If this were a guide, I would go get a xom ally and count how long he lives for sanity checking. For now, I don't know this.

File: bird.jpg (10 KB, 340x200)

10 KB JPG

>>554369597

File: 1000_F_465527563_MwxxIcDU(...).jpg (199 KB, 1000x667)

199 KB JPG

just noticed if you hit shift + b to butcher an zed, choose the zed, but then hit escape when it asks you what type of butchery option you want, it pulps the zed in 0 seconds and with zero effort. i GUESS you could exploit this, but good lord the autism required to do that every time

File: eid-mubarak.png (70 KB, 393x285)

70 KB PNG

>>554369597
>there's a 1/13rd
>1/13rd
>13rd
>thirteen-erd

>>554280503
I will never convert to Beogh at gunpoint, kill me first furfags

So, an example.
AT XL10 with 50 severity, as is typical

strength = 1d(5* sev) = 1d(250) = 125 on average
Variance from the table is initially -4. Add xl and it's 6. Add "ally" random value and it's 8, this time.

Simplest way to calculate rarity is rarity*( ((varaince-min) + (max-min)/2 ) / (max-min) ). In English that's "distance from edge plus half the range divided by the range, times base rarity"
If the number is closer to max than min, switch that part out for (max-variance) instead.

Don't worry about decimal fuckery. The actual number is inflated to account for that.

butterfly with 5 current rarity
shadow imp 20
ufetubus 25
white imp 20
mummy 2.5
phantom 62.5
bombardier beetle 9
swamp drake 83.57~
weeping skull 23.21~
orange demon 32.14~
shapeshift 31.25
ice devil 20
red devil 20
boggart 30
chaos spawn 77.5

total rarity appears to be 461.67. As this isn't a range where "near the middle" has any particular meaning, I can't just say it's average 230.835. However, swamp drake, chaos spawn, and phantom are the most common results at this XL and severity. Everyone's odds are current rarity / total rarity, which is the mildly useful information.

Assuming I'm in a zig, for some reason, there's a 7.69% chance that my summon is instead replaced using replacement table 1
There's a 7.1% overall chance of replacement using replacement table 2
There's a 21.3% chance of replacement using replacement table 3.

The monster is summoned with sum_duration 6.
It is enchanted with average 150 extra duration, which may not mean 156 total duration and may not mean 210 duration.
It is enchanted with 1000 duration regeneration, which I suspect means "exactly 100 turns"

And, finally, it is done, and it is confirmed that you now have the tools needed to, by hand, figure out what xom can summon given your XL, severity, and location.

>>554375982
I really don't think he's a bad god, but the army playstyle is extremely dull. The redesign made it much more tolerable than old Beogh, but it's still not fun.

File: silent bike.jpg (57 KB, 862x433)

57 KB JPG

huh, this is new, never seent dis befoah. a (nearly) silent motorbike with a generator that converts all fuels into electricity? hell yeah im putting that shit into my pickup

>>554382715
What version of cataclysm?

>>554383650
completely vanilla

>>554384317
DDA isn't vanilla, it's Kevin's special chocolate

>>554385482
oh okay, i guess the most popular and played version of dda, then

File: I Don't Know What I Expected.jpg (30 KB, 462x456)

30 KB JPG

>>554385665
>the most popular and played version of dda, then

>>554385951
yeah me either. anyways, its pretty useful. finally a reason to collect all this diesel

File: wheelbarrow lol.jpg (38 KB, 287x286)

38 KB JPG

lmao and now wheelbarrows parked in parking spots? thats new

>cant wear the condoms
fuck this game

File: 1754370876256043.png (583 B, 437x68)

583 B PNG

>>554392056

File: hungry.png (101 KB, 1460x1057)

101 KB PNG

I just realised CLG doesnt have any portal storms. Peace and quiet, ahhhh bliss.
Here's my fridge, i am not eating well, its been cheese and malt milk balls. I barely have enough survival to make a butchery rack, then i'll find some poor beaver or something. Just crafted a stonebow, took 3 in game days but now i have a better weapon than a slingshot.

File: i feel nothing.png (190 KB, 1190x285)

190 KB PNG

>>554393870
Nvm i am sorted.
There's been this cow loitering around my base, i tried to feed it straw but you cant do that. I had a can of corn though, made cattle fodder and now i have infinite milk.
I also butchered a mutant frog and have been eating that, the toxins (25U) in the cooked meat cant be that bad can they?

>>554397131
Toxins aren't too bad as long as you're not eating only that. It will build up over time and start to give you unpleasant effects, first minor, then nastier as the toxin level rises.
It goes down over time, so you can totally supplement your food supply with a little mutant meat and be 100% fine, or eat a lot of it for a while when you lack other sources.

>>554377206
>>554374494
AK-

>>554377206
>XOM_GOOD_POLYMORPH
>_xom_good_polymorph(
ends up at
>_xom_polymorph_monster(*mon, helpful);
which runs
>_xom_mons_poly_target(
to get a target and
>_xom_polymorph_monster(*mon, helpful);
to do the effect.

In xom mons poly target, a "monster_near_iterator" is used to grab all the monsters "near you", and I have failed to understand this object, so I'm going to assume it grabs all visible monsters and move along. It's in act-iter.cc
It won't poly "firewood" which I guess means bushes.
It won't poly monsters for which _choose_mutatable_monster(monster) returns fails, which I think excludes monsters which cannot, by default, be polymorphed.
It won't poly monsters over water or lava if your XL is under 5
It won't, I think, mutate plants unless your XLis over 8

Note that "you can see" in these cases ignores monster invisibility. I just mean it's tile is on screen.

Once it has a list of all valid targets in sight, it randomly picks one and returns it.

Once xom_polymorph_monster(*mon, helpful) receives a monster,

If the monster is not a "ghost_demon", is not a shapeshifter, is natural, and either you're above XL4 or xom is nasty, there is a 12.5% chance of xom adding a "shapeshifter" enchantment. Specifically, shapeshifter 66% of the time and glowing shapeshifter 33% of the time. I am not sure if this makes them an actual shapeshifter, which is instead a monter-type.

Xom will attempt to polymorph while lowering XL if the monster will attack during good polymorph, or if the monster won't attack during bad polymorph.
If not that, then Xom will instead attempt to polymorph while raising the XL when you're above XL 4 during bad polymorph.
Otherwise, xom will attempt to polymorph while leaving the XL alone.
During bad polymorph, if your XL is under 10, xom will malmutate the monster he just polymorphed.

>mons.polymorph(PPT_val)
calls this function

>Polymorph(PPT)
It's in mon.cc. It receives a single value which is the same as an int and can only be MORE, SAME, or LESS.

If the monster can't be polymorphed, it returns false.
If the monster is an uglything, it just changes the color. This means that xom can't change what an ugly thing is.
Polymorphing a monster with the shapeshifter enchantment returns it to its original shape.
Slimes are special-cased to polymorph their parts separately.

"targ" is RANDOM_POLYMORPH_MONSTER if PPT_same was used, but otherwise is RANDOM_MONSTER

> return monster_polymorph(this, targ, PPT);
this function is in mon-poly.cc

This fails if target is using stairs.
Fails with message if monster has "demon tier -1", which seems to mean pan and hell lords
Fails on "pet slime" with message

>targetc = _concretize_target(
gets the target species

In concretize_target(
It calls _poly_from_set(mons) if targ was Random_polymorph_monster earlier, which grabs a random entry from a list of monsters that is specific to every monster and so does not have to be explained here, because it can't be.

If targ was RandomMonster earlier, because zom tried to increse or decrese HD, then _fill_poly_weights() is called instead.

>_fill_poly_weights() and
>_poly_weight

For every monster in the whole monster list, it disqualifies glitches, duplicates, and the monster type we are polymorphing away from.
It disqualifies monsters which cannot live in all the same habitats as the original monster. That means target flying monsters are always allowed. Any non-flying monster is an invalid target if the original monster was a flyer. After flight, the list of all valid terrain for a target monster must contain every valid terrain for the original monster, regardless of what tile they are currently standing on. A creature which could live in lava, for instance, would have very few targets they could polymorph into. However, almost everything would be able to polymorph into a lava crab.

Then, it sets target HD to current HD -2 or +3, depending on if PPT was LESS or MORE. HD goal is capped at 1 or 27.

It obtains the difference between target HD and new HD. If new HD is higher than target HD and PPT was LESS, it doubles the HD difference.
It obtains the difference between current Demon Tier and new Demon Tier, and adds the HD difference to the Tier difference.
If the total difference is greater than 8, the monster is disqualified.

Otherwise, the monster is added to the weighted list. The weight of the monster is 2^(8-total difference). So, the highest weight occurs with zero difference and is 256. Each point of difference makes a monster half a likely to appear. At 8 difference, the weight is 1.

Once it has a weighted list of every valid monster, it rolls one.

Back up in monster_polymorph, the target monster is received and change_monster_type() is called.

Jiva does something we don't care about..
A lot happens in order to make this transformation while maintaining details about the monster, so only a few are of note.
fearing and "beholding" are interrupted, so I think this cures mesmerize, maybe.
Deletes associated tentacles.
Dumps TRJ minions out.
zombification is maintained.
Godless new priests are given gods.
Monster drops objects.
Unknown shapeshifters are marked as shapeshifters.
Constriction, both direction, is broken.
Engulfing others is broken

>mons->speed_increment = 67 + random2(6);
Which I think means that a polymorph target doesn't move for 7 turns, which I don't think can be a correct interpretation.

At the end of monster_polymorph(, a function xom_is_stimulated(int) is called.
If a shapeshifter was targeted, it uses 12 for its value.
If poly morph tried to weaken a monster, or the monster was friendly, it uses 25.
If polymorph did not attempt to change the XL, it uses 50
If polymorph attempted to strengthen a monster, it uses 100.

>>554404652

>xom_is_stimulated()
This function is the function which generally increases xom's current interest in the player. The first thing the function does is attempt to check what actor took an action. If it was xom, it does not change xom's interest level.
However, only xom can specifically attempt to increse the XL of a polymorph target, as far as I can tell. This indicates that either part of this code is pointless, or else the game does not credit xom as the actor which is preforming his polymorphs. So, this method of detecting what actions xom caused is likely inconsistent, from the perspective of a player. For polymorph, this is somewhat testable, with difficulty and a lot of time.

Also of note, the wiki does not mention that polymorphing gives xom interest, I think, and it does.

Piety_scale is a function which vaguely multiplies a number by 1.25 when the user has a level of faith. It was used earlier for the normal drift of piety up and down by 1, which means that having an amulet of faith causes it to drift up/down by 2 25% of the time.

In xom_is_stimulated:
>int interestingness = random2(piety_scale(maxinterestingness))
So, the interest-value of an action may be multiplied by 1.25 if you have faith, with random rounding. Then, interest is set = 1d(interest)

This means the actual amount of interest which you get is random. This means you cannot, at all, in any way, assume that an amusing action will actually amuse xom. Those conclusions were wrong, and that's why this slow compilation of all xom's features doesn't belong on a wiki.

Interestingness is capped at 200, so you can't actually roll 255 interest from use-IDing potions.

>>554404735

If interestingness is less than 10, it does nothing.
If interestingness is 10 or more, and interestigness is greater than Xom's current interest value, then Xom's interest is set equal to this interestingness.

Every time this happens, when the interest goes up, xom will give you a message about it.

So, the wiki says "Spell miscasts (the level)", meaning that a miscast has an interest value equal to the spell level. The highest level spell is 9. Miscasting a spell, then, can never amuse xom, because it cannot roll greater than 9 on a 1d(action's interest).
However, with a level of faith, 9*1.25 = 11.25 and 8*1.25 = 10, meaning you have a small chance, when casting very high level spells and miscasting, to amuse xom.

So, lets check to make sure it isn't as bad as the wiki indicates.

In spl-miscasts.cc
>miscast_effect()

> if (target.is_player())
> xom_is_stimulated( (level / 3) * 50);
Level is an int. Level comes from spell_difficulty(spell). Spell_difficulty returns _seekspell(spell).level. Seekspell returns a spell_desc, which must have a "level" variable.

spl-data.h contains the spells. spl-util.cc contains a "struct" which matches turns the information in spl-data into a set of spell-desc objects. The 4rth variable in the struct is called "level". The 4rth variable in each spell's definition in spl-data is the spell level which should be expected. For instance, "cause fear" is level 4 and the wiki says i is level 4, meaning this level is the same as waht we normally call levels.

so, miscasts of level 1 and 2 don't give any interest because level is an integer. 3,4, and 5 all give 50 interestingness, which gets turned into interest if you're lucky. 6 7 8 give 100. LVL 9 miscasts give 150 interestingness, which is sizeable.
Contamination is 2*(LVL^2*fail% +250), just like it says a bit ambiguously on the wiki. so, for a level 9 spell, you get under 5000 contamination for a miscast at 27% spell failure rate.

>>554408053
It's 35%@lvl 8. 45%@lvl7. 62%@6, 89%@lvl5 and lower spells are always safe for at least one cast at zero contamination.

Trusting this formula from the wiki, for now, damage incoming is
(Spell Level)d(Spell Level + Raw Failure Rate) / 9
9d(9+99)/9 = max 108 damage, potentially elemental and resistible. Any late game character should be able to afford that every 1500 turns on average, if they are willing to spare the spell slots.

>>554382715
and 0 noise when riding. kinda op really

>>554408419
>XOM_GOOD_ALLIES
>_xom_send_allies

Uses the same strength roll and functions from one-ally above.
Also gets a number of monsters to summon from _xom_pal_counting.

Uses the same strength number for this:

Strength, xl<4, xl<10, xl<19, xl>18
<151, 2 to 3, 3 to 4, 3 to 5, 4 to 5
<301, 1 to 2, 1 to 2, 2 to 3, 2 to 4
300<, 1, 1, 1 to 2, 2 to 3
Meaning higher tension, in this case, causes fewer monsters to appear.

If you have 5 or more runes, add runes/5 monsters, with random rounding based on the decimal value.
If xom is nasty and summoning enemies, *1.5 monsters.

Since summon allies was already covered, that's all.
Fuck that's nice. Polymorph was a lead brick.

>>554409310
>XOM_GOOD_FORCE_LANCE_FLEET
>_xom_force_lances
Pushes things within 2 distance of you away "harmlessly"
Attempts, again, to push away "adjacent" monsters.

If xom is nasty, attempt to summon (2+ xl/6 + 1d(xl/15)) lances
They would have xl/6 HD, at least 1.

If xom is not nasty, attempt to summon (2+ xl/2 + runes/8 1d(XL/5 + rune/20) )
They would have (XL+runes/4)/2 HD

>XOM_GOOD_CLEAVING
>_xom_cleaving
Adds 10+ 1d(severity) cleaving duration, so maximum of 11 turns. Seems to low. Needs a test.

>XOM_GOOD_FLORA_RING
>_xom_harmless_flora
Ignores severity. Random radius of ring tends to increase with XL.
radius 2, weighted 54-XL
3, weighted 27
4, 13+XL

Has 13+XL/54 chance of using demonic plants instead of toadstools, which might have different HP or something.

50% chance that random 1/3rd of the plants in the innermost ring will be missing.
Does not, I think, replace monsters or push them away or anything.

>XOM_GOOD_DESTRUCTION
>_xom_real_destruction
For each monster "near" you, if it isn't a projectile or a tentacle,
2/3rds of the time
when that monster isn't next to you, unless when xom feel's nasty in which case he'll do it anyway
Xom attempts to cast a spell.

"Dice" is a value = 2 + XL/13 with random rounding. Random rounding, by default, always means decimal-value*100% chance of rounding up.
if you have more than 4 runes, add runes/3 with rand rounding to "dice"

Beam is a "sticky fireball" and I don't know if it's resistable.
Damage is dice_def(Dice, 4+severity)
Beams explode with size 1, which I think means no splash damage but also no evasion.

This dice_def() creates a dice with Dice d (4+severity) damage. At XL 26, that's 4d( avg 54) or an average 108 damage on 66% of visible monsters. At XL 1, that's an average of 54 damage, likely fire type.

>>554408672
You can build your own diesel hybrid though
>b-but muh fuel compatibility!
Fuel isn't rare, and if you've adjusted the drop rates to make it rare, you probably shouldn't be counting on finding a specific unique motorbike.

>>554413646
>XOM_GOOD_DOOR_RING
>_xom_good_door_ring
>_xom_door_ring(bool good)

Xom attempts to move monsters away from you. First, he moves the monsters which are inside the central square of the door ring. Then, he attempts to move the monsters on the potential door tiles away.

Xom places the door ring, radius 3, 4, and 5. Xom can't replace "critical" things. He can replace unruned doors, solid floor, or diggables.
He apparently fails to replace living creatures still in his door-zone, though I'm not sure why.

>XOM_GOOD_MASS_CHARM
>_xom_mass_charm
Duration is 500+1d(severity)

The average HD of all nearby monstes is obtained.
The first checked monster is always charmed, which means there's value in figuring out the order in which nearby_monsters will check tiles.
Then, for every monster, a random 1,0,or-1 is added to their hit dice. If this value is <= their average HD, the monster is charmed. This stops when 50% of monsters are charmed, or when all monsters have been checked.

>XOM_GOOD_HYPER_ENCHANT_MONSTER
>_xom_hyper_enchant_monster
Will not buff a monster with haste, invis, might, or empowered spells. Will only buff friendly monsters. Will not buff "peripheral" monsters.
Maximum HD is 3+ XL*7/20, so 12 at most.
If more than 3 runes, add runes/3 randRound to HD, so 17 at most.

For every nearby mosnter, add (-2 to 1) randomly to Maximum HD. If their HD is less than Max HD, add them to the list of monsters which may potentially receive buffs.

If there is no available friendly monster, xom will summon one.

Under xl 7, iguana25% or cerulean imp 75%
Otherwise, if XL < 13, 14, or 15 randomly, black bear25% or hell rat 75%
Otherwise, if XL < 20,21,22 randomly, 25% wyvern or 75% hell hound
Otherwise, 33% elephant or 66% toenail golem
sumon_dur(3) for this summon.

Your list of valid monsters is randomized.
If xom feels nasty, he will sort your list in ascending XL order, I think. Check if you want:

>>554425739
> shuffle_array(targetable);
> if (_xom_feels_nasty())
> {
> sort(targetable.begin(), targetable.end(),
> [](const monster* a, const monster* b)
> {return a->get_hit_dice() < b->get_hit_dice();});
> }

For only the first monster in this list, the game will attempt to apply each of might, haste, invis, empowered spells, repel missiles, resistance, regeneration, willpower, toxic radiance, double hp, mirror damage, and swiftness.

>XOM_GOOD_FOG
It's fog. Nothing special. No XL or severity.

>XOM_GOOD_FAKE_DESTRUCTION
>_xom_fake_destruction
>_xom_destruction(int sever, bool real)
same function as real destruction, except it doesn't do anything. Might, maybe, show the location of an invisible enemy for 1 frame.

>XOM_GOOD_TELEPORT
>_xom_good_teleport
Will standard-teleport you at least once.
After every teleport, has a 10% chance to stop.
After each teleport, if the 10% chance fails, there is another ((teleports - 6)*20)% chance to stop, so 100% chance at 12 teleports.
Lastly, if the player is not in a "dangerous place", there is a 25% chance to stop teleporting.

This seems extremely insufficient. While the good teleport makes an attempt at leaving you in a safe place, it simply fails. Xom gameplay should, in part, be designed around avoiding this effect. The bad version, through worse, isn't that much worse.

>XOM_GOOD_LIGHTNING
>_xom_throw_divine_lightning
Does 3d30 damage.
Sends out a random number of bolts from 4 to (tension/5, minimum 5 and maximum 27)

For every nearby monster, has a 33% chance to throw a bolt at them in particular.
Will not throw more than the number of bolts described above. If any bolts remain after trying each monster, it will throw them at a random spot in a 13x13 tile area centered on the player.

>>554422637
yeah but this lets you do it in a single tile, could make it foldable as well

What's the best real roguelike I can play if I want something mechanically (and content wise) simple and non-bloated, but still want all the crazy stuff with daisy chain reaction abilities and emergent gameplay possibilities?

>>554428635
Path of Achra.

>>554428635
Yeah Achra is basically all about making builds, 90% of characters will be doing the equivalent of O-tab or spamming wait.

>>554428635
>What's the best real roguelike

Dungeon Crawl Stone Soup

>>554428635
Caves of Qud

File: V2Alita 095.jpg (659 KB, 3840x2160)

659 KB JPG

She's looking much better now.
I'm just worried that she'll be too stupid to turn the power armor on in combat.

File: CBN_aftermath.png (1.5 MB, 1600x900)

1.5 MB PNG

I lived.

File: CBN_corpses.png (3.97 MB, 2059x1521)

3.97 MB PNG

>>554445584
Lotta bodies. Not including another horde off-screen to the left or the ones I converted directly into meat.

Can i somehow disable my items falling from the car in cdda?

>>554448015
What do you mean? Falling how?

>>554448015
>>554456521
It can look like you're dropping items "into the car" but if it's not installed in a storage area (quarterpanel, door, trunk, trunk door, cargo space, seats even- I think aisles MIGHT work, but I can't recall for sure), mostly meaning bare frames, it'll fall through onto the ground.

>>554448015
>>554457684
Use the advanced inventory management menu instead of dropping stuff.

>>554446260
Messy.

Is there a roguelike that ISN'T entirely reliant on rng for "difficulty"? Just lost 2 lives in tome4 to a fucking lvl 12 snake of all things because first time around, it put me to sleep and strike me 3 times without me waking up, then the second time around it struck me with crits on repeat eating all my shields, then 3rd time I won in 2 turn because I crit it twice.This is fucking retarded, tactical skills are fucking irrelevant when the rng can just decide to fuck you over on a dime. I dont even feel good about beating the little shit because it was purely luck, the dices decided I won and thats it. In my 350 hours of tome4 to date im pretty sure over 95% of my deaths were the result of bad rng, and im starting to genuinely feel fatigue over it and it hamper my enjoyment of the game. Its to the point I cant even be bothered to think of a new character or new build because I just know ill die to some dumb bullshit that I cannot use tactics to counter.

Didnt think there would be a time where I think xcom is fair, and it isnt like I have better luck with rng in those games.

>>554477223
If you have good knowledge in tome4 then even on Insane you should win like 80% of the runs. Probably 100% if you are very vigilant.
Tome4 is the most deterministic of all roguelikes since it allows you to choose from a lot of areas and you don't have to clear much to win.
Your mistake there sounds like you tried to reengage after knowing that it can kill you easily the first time it slept you. Could have just run away.

>>554428635
Don't listen to these faggots anon. Path of Achra is a glorified puzzle game.

Get Golden Krone Hotel, it's basically DCSS (and on a similar level of simple) except with all the autism removed and it actually plays and looks good.

What is the dark souls of roguelike?

>>554493254
Unfortunately, I don't know any roguelikes with a dodge roll button. The dodging is usually just a stat check.

>>554493757
My ragulike will have a dodge button
In fact it will have 8 dodge buttons

>>554425739
fucik
HYPER_ENCHANT_MONSTER
Duration of buffs is randmo from 200 to (200 + sever * 2)
so average 250.

>XOM_GOOD_WAVE_OF_DESPAIR
_xom_wave_of_despair(
Makes dead skeletons around the player, I think on every empty tile within distance 2. Potential necromancy use?

pow = 50 + random_range(sever / 2, sever), so 50 + (avg 50 to 25), so avg 87.5


For every monster around, it gives 1 level of vulnerability for pow *BASELINE_DELAY
in defines.h
>BASELINE_DELAY = 10;
So that's 87.5 turns of vuln.
Gives lvl 2 draining for pow duration, so that's 9-ish turns, minimum 5.
"annoy"s monsters that won't attack.

Gives you vulen for 87.5-ish turns.
And lastly, you cast fear with pow*5 power, or 437.5 average power, max 750 min 250. I think a scroll of fear is 1000 power.

In beam.cc:
>mass_enchantment(enchant_type wh_enchant, int pow, bool fail)
fail is false be default, from beam.h
Mult pow by 1.5, so that's 655 for us with rounding.

> pow = min(pow, 200);
Here, we have a line which caps power at 200. The smallest value xom's fear effect can submit is 250. This means that there is no effective variance in the power of xom's fear effect, and that it is always as strong as possible. If that is the case, why use a variable instead of just using 1000 like what the scroll of fear appears to do?

Further, any mass enchantment with a power of 134 or higher is as strong as possible, in terms of chance to occur.
This is absurd enough that, despite seeing no flaw in my reasoning, I must note it and warn that I may be incorrect about some part of how mass_enchantment( works or what function it calls.

This was such an unacceptable conclusion that I have now compiled the game with a small printf() statement meant to trigger every time mass_enchantment is called. I have now compiled the game. When I use a scroll of fear, my command prompt reads 200. When xom uses wave-of-despair, the command prompt reads 200.

It's legitimately this stupid. I am now a modder over this.

>>554498536

I also pumped an int 12 tengu to xl27 and skills 27 and cast discord.. Discord had 69 out of 200 power. 69*1.5 = 103. When I cast discord, the console printed 103, so casting also uses this same function, but is much less likely to hit max effective power.

Anyway, power = 200 at this stage.

"_monster_resists_mass_enchantment()" mostly just removes invalid targets and then calls mons->check_willpower(&you, pow)
It might do something interesting with "mass anguish" not affecting friendly targets, which might mean scroll of torment maybe, but that's somebody else's problem to figure out.

check_willpower() is in actor.cc.
Will-immune are unaffected.
Effective willpower is reduced by 1 pip if the player is in sphinxform and 2 pips if the player has "guile", so 3 at most with no further stacking.
Willpower minimum is 0.
Casting monsters with ATT_MARIONETTE reduce the effective willpower of their target by *.5 instead.

Spell power is stepped-down. runs "stepdown(pow, 35)" which returns "step * log2(1 + value / step)". In our case, pow is value and step is 35.
So, that's 35 * log2(1 + power / 35)
new power@power
96@200
84@150
68@100
44@50
27@25
0@0

The monster's degree of resistance to this specific enchantment is 100 + effective willpower - newSpellpower - (0 to 99) - (0 to 100)
If all numbers are maximum, that's 100 + 200 - 96 - 99 -100, which is 3, meaning that an enemy with perfect willpower cannot possibly be enchanted.

That value, "res_margin", is used to create a message in game which tells you how much your target resisted you by. Any value above 0 means the enchantment was resisted. Any under 1 means the enchantment worked on that target.

Path of Achra is bretty fun. A little easy to break by just summoning a billion allies, but you can just not do that.
I like that it can be fully controlled with a mouse. I have been spoiled by modern times ;_;

roguecels

>>554493254
>>554493757
Dwarf Fortress Adventure Mode has an active dodge. Also blocking and parrying. And bossfights that kill you instantly.

can you not drill/jackhammer through floors in cdda? its only walls?

>>554513281
ah fuck is it really just "mine downstairs". i cant just bash through the floor or anything?

>Find 2 bear traps
>Find pupating brute
>Trap it
Haha, yes, now taste my pebble crossbow!
>Breaks free both in seconds and regenerates all health
Vivec, give me the CHIM powers i need.