[Balance] malnutrition does stuff

[dissociativity]

Summary

SUMMARY: Balance "Malnutrition is not good for you"

Purpose of change

PK rebalancing has had working penalties for malnutrition for a while, to my eyes they seem fairly accurate to real life, the scientific articles and 5000 word essays on what values should specifically be had at a given level can be had later, for now we have a missing feature existing and working from readily available json changes.

Describe the solution

Malnutrition didn't do things other than iron deficiency due to the blood system PR, now it has negative effects on the player for more than iron deficiency.

Describe alternatives you've considered

Allow players to eat only iron rich diets, and ignore all other vitamins in shame forever.

Testing

play game, eat terrible diet or too many multivitamins, suffer

Additional context

using data from #33776

Data:
includes VitaA deficiency, VitaB deficiency, VitaC deficiency, & Hypervitaminosis.

Vitamin A deficiency's base level lowers per by 1, if it progresses it can lower per by more, the PR includes night blindness but I am unable to verify if the code is worth adding as I'm pretty much a novice at very basic C++
Vitamin B12's deficiency at base level causes dex and int to lose a point , fatigue, and a health stat debuff, progression makes all of this worse, the PR this is based on adds dementia resulting in skill rust, though once again, I'm unable to verify how good the code is, that can be up to others:
"max_intensity": 3, "base_mods": { "int_mod": [ -1 ], "dex_mod": [ -1 ], "fatigue_min": [ 5 ], "fatigue_max": [ 10 ], "fatigue_chance": [ 900 ], "h_mod_min": [ -1 ], "h_mod_min_val": [ 0 ], "h_mod_chance": [ 900 ] }, "scaling_mods": { "int_mod": [ -2 ], "dex_mod": [ -2 ], "speed_mod": [ -5 ], "fatigue_chance": [ -200 ], "h_mod_min": [ -1 ], "h_mod_chance": [ -200 ] },
Vitamin C deficiency results in pain, fatigue, and overall negative health, this gets worse with progression
"max_intensity": 3, "base_mods": { "pain_max_val": [ 35 ], "pain_chance": [ 900 ], "pain_min": [ 1 ], "pain_max": [ 5 ], "fatigue_chance": [ 900 ], "fatigue_min": [ 10 ], "h_mod_min": [ -1 ], "h_mod_min_val": [ 0 ], "h_mod_chance": [ 900 ] }, "scaling_mods": { "pain_chance": [ -300 ], "fatigue_chance": [ -200 ], "h_mod_min": [ -1 ], "h_mod_min_val": [ -100 ], "h_mod_chance": [ -300 ] },
Hypervitaminitosis does nothing for now as it's implementation is very vague and abstract, I'm considering changing it to have a minor health stat debuff as generally it's probably not ideal to have too many vitamins, but it would be very mild given how vague it is as a concept.

[dchieh222]

Looks promising, but for purposes of realism check, can each of the vitamin effects be listed in the PR, preferably under Additional Context? This will allow its effects to be compared against what we know of vitamin deficiency and excess.

Appears at the moment to be VitaA deficiency, VitaB deficiency, VitaC deficiency, & Hypervitaminosis.

Also to be considered: audit of present foods for accurate vitamin values.

[dissociativity]

alright, updated that.
I haven't found on a vegetarian diet in-game so long as it's supplemented with mushrooms and/or nuts that there are nutritional issues, those that depend almost exclusively on meat may get scurvy.

As far as I'm aware there's vitamin values that are decent in terms of realism in gameplay more or less for everything, and food ingredients cause a recipe to inherit the vitamins.

[Venera3]

Correct me if I'm wrong, but all this was already in the game in 0.C and got removed for being all kinds of unreasonable.
Having a common "hypervitaminosis" for different vitamins is not how it works. Fat soluble vitamins (DEKA) get stored in fat, so you wouldn't see any acute effect till all your fat is saturated (depending on the vitamin, in the multigram range). Water soluble ones can generally do more damage and they are the ones getting peed out, but that doesn't mean they make you pee more.
The Vitamin A deficiency is an important cause of preventable blindness in the developing world, but the player character can be assumed to have enough stored amount to shore them up for a few months and only cause night blindness long after that. You also don't get stupider from it.
Scurvy is fine and it could also do some proper damage at higher intensities.
Hurt_chance causes direct damage, in the case of nondeclared bodyparts to the torso which is pretty overkill for a b1 deficiency. That one also has the problem of the charcter starting with enough to last a few months.
Generally aside from vitC you'd need to redo the stored amount/decay rate of the vitamins pretty significantly to have anything approaching realistic. Otherwise I'd just include scurvy for now.

[Fris0uman]

Here's another PR that was going to fix the thing but the author didn't have time to finish. It has pretty well laid out plans you can use as guide #33776

[dissociativity]

okay I can't do that, I can just do this, I am not a c++ wizard

[dissociativity]

the effects for this are fairly mild at lower deficiencies and scale with more, it was 'unreasonable' in 0.c because one couldn't actually get vitamins without multivitamins due to food not having nutrition specified, which it does now, please do not misinterpret the past like this, I was there playing too.

[dissociativity]

I can nerf hypervitaminatosis to only affect thirst, but I think the rest are fairly reasonable values that won't break a game/run or character individualy, and it's been in PK rebalancing fine without complaints for years at this point.

Did you mean I should use the stat modifiers from that PR as a guide?

[Venera3]

The maluses being mild does not change the fact that a healthy adult couldn't develop any acute vitamin deficiencies (other than as said, scurvy) in the first year of playtime. If you want it to be reasonable you can nerf the depletion/raise the starting vitamin storage to reflect that. The only people who would be affected would be junkies, alcoholics and the like, but I don't think vitamin levels can be defined on a per-scenario level.

[dissociativity]

The maluses being mild does not change the fact that a healthy adult couldn't develop any acute vitamin deficiencies (other than as said, scurvy) in the first year of playtime. If you want it to be reasonable you can nerf the depletion/raise the starting vitamin storage to reflect that. The only people who would be affected would be junkies, alcoholics and the like, but I don't think vitamin levels can be defined on a per-scenario level.

mind you, I believe if the jsons between traits for stat buffs and debuffs and such can, one can make alcohol, meth, ect have a vitamin malus that drains the body of nutrients at a higher rate (while under the influence of said drugs) as far as I'm aware, I'm unsure if this is a new or old json feature however, but I can work on this particular part.

Do you think moving minor deficiencies for the other types towards doing nothing and the 'scaling' later-malnutrition end of malnutrition can be moved to more or less what the 'minor' malnutrition is for now?

I don't want to have the feature not work period, even though it's technically implemented as a status that tells the player that they're suffering from malnutrition.
I could however rework the descriptions to tell the player that it's more of a mild deficiency with no effects, more of an encouragement for the player to eat more of X or Y, rather than an actual damaging malus, so the game is no longer lying to the player, which is what would be happening if I left it as is.

Any drug or alcohol vitamin malus, which I'll describe vaguely here:
Essentially there's a value that can alter (per minute afaik) how much of x or y vitamin the body consumes, it's used for traits that allow eating mutant meat, as the blob inside is a 'vitamin', but can apply to any, those traits as far as I'm aware don't make it physically imposible to get blob psychosis per say, but substantially harder.

So let's say the player is drunk, I'm not currently sure of what vitamins that deprives a person of, so I'd need to do research, but a seperate PR would make this actually apply, but this would make the player consume more of X and/or Y vitamin per minute vs stock rate, which might justify making drugs a tad more realistic (meth doesn't just last for an hour or so and make you have to sleep for several days irl, it allows you to stay up for several days, and you're tired both from the stimulant abuse AND the staying up for incredibly long you've already done, makes sleeping very difficult.

As a side,note, where would I go other than a hacky permanent status effecty for a 'healthy' player to slow vitamin drain for everything but C, or more realistically to increase the starting vitamin stockpiles for b12, iron, A, calcium, as as you said the body has a large store of these that, as I've heard irl with a vegetarian diet, B12 and Iron take very long to deplete even if you get inadequet intake.

[Venera3]

Sorry, I should have been clearer: the only people being affected in the first year should be pre-Cataclysm junkies, alcoholics and critically ill. It takes literal years of drinking heavily and malnutrition for an alcoholic to get to the level that could cause problems within months/weeks. Hardcore drug users get there faster, but still not overnight.
To affect vitamin stores you can edit the amount and the decay rate in vitamin.json, as well as the disease threshholds. Currently everything is set to run out in ~90 days completely, which you could take for scurvy.
Ediited to add: I'm honestly of the -probably unpopular-opinion that vitamins should really become a thing when the usual run length gets into the multi-year range. The system itself is nice to have for toxins/diseases and the like but "regular" vitamins (other than C )shouldn't realistically come into play outside of certain backgrounds or mutations.

[I-am-Erk]

Venera3 has already expressed my main concerns. Unless someone is starting the game malnourished, these need extremely long timers. Even with a population thriving on chicken tendies and fries, we don't see these deficiencies often irl.

[dchieh222]

If we work in presumed hypermetabolism from blob infection, does this increase vitamin drain?

Comment withdrawn due to error in assumption/understanding, reply kept for archival purposes.

[Fris0uman]

But we don't assume hypermetabolism from blob infection.

[dissociativity]

I'm very busy lately but will slowly make the proposed changes under all of your guidance

[dissociativity]

So:
I can simply make this PR only add scurvy, but modeled more realistic as of the linked PR that did a pretty good job,
or I can make all effects modeled as close to in-json as the more realistic linked PR
AND
either:
make the deficiencies trigger at much lower vitamin amounts (other than perhaps scurvy), so players will generally not see these effects without long-term, year-long+ bad dieting
or:
Alternatively increase starting vitamin stores for all vitamins but c? substantially could be an option

A drug PR to cause drugs to temporarily increase vitamin drain while the drug effects are active may also be included seperately or together, also look into diseases and other status conditions' malus on vitamins

Opinions?

[I-am-Erk]

I think increasing starting vitamin stores to high level would be the way to go, that's literally what it is irl. Finding data on exactly how much to add would be a bit of a trick but should be very feasible.

[dissociativity]

right, so with this PR, B12, Iron, VitA all take 4x longer to deplete if I'm understanding the vitamins json and the logic given to me correctly.
Given a player's starting stores of decent nutrition should last 90 days by default, this should mean it would take a full 360 days to deplete the vitamins that are not C, I'm unsure if even longer should be necessary and can easily change it to be thus, but it does mean the features 'work', I did not apply this to vitamin C as it seemed from context that C and Scurvy was not problematic, I ported the json effects from the old PR over as well as they seem far better researched than the ones I've been using.

I also removed the symptoms from hypervitaminatosis for now since it's too vague and abstract as implemented, as suggested.

An alternate implementation could instead of decreasing the rate at which the vitamins are depleted, instead making the thresholds for deficiency to offer noticeable symptoms to be lower.

Do let me know if vitamin C stores should actually last longer than they do (presumably the 90 days default) and by how much it should be increased.

I think 360 days is reasonable given that mild deficiencies in average diets aren't exactly uncommon, so it's not like we assume the player gets regular blood tests and sees a nutritionist.

[I-am-Erk]

If you eat no vitamin C at all, it takes about 4 weeks to start to develop scurvy. It's not one where stocks last long.

b12 deficiency takes 2-5 years to develop and is tricky, it takes a pretty extreme diet to run out of b12.

I am having troubles finding data on adult vitamin A deficiency. it's exceedingly rare in the developed world, which means it's likely similar to b12 in timelines.

Iron deficiency through diet is a complex one. Most anaemia is developed through external loss of iron, but some extremely low-iron diets can produce iron deficiency through lack of dietary intake. I would estimate that from diet alone, it would probably take around 6 months to develop symptomatic iron deficiency. Our players are more likely to be developing it by losing blood, obviously.

Does that help?

Now, in terms of deficiencies, the b12 deficiency you've described is way too fast. it's an insidious deficiency

• First, fatigue increase and stamina decrease
• Later, dex begins to decrease
• Later still, int and per begin to decrease
• Later still, str begins to decrease
  It should also cause anaemia but that's complex because it's not an iron deficiency anaemia and will probably require code that links it to blood recovery rates.

[Venera3]

Yeah, I meant the -0.2 scaling on an effect with a max intensity of 3, sorry. I typed that up at work and didn't want to ninja edit.

[Venera3]

I spent some time looking up nutritional values for an unrelated PR, and I have to say that the assumption that the vitamin values for existing foods is accurate is...very optimistic. Maybe eggs are just an unfortunate exception, but the vitamin content will need a reality check before we arm any of the deficiencies ( at least the ones that get activated ).

[dissociativity]

while lots of food doesn't have completely accurate nutritional data, it's rather trivial to eat a diet that won't result in any deficiencies as far as I'm aware, it's only players that purely eat meat or purely eat bread that would be adversely effected by this.

[dissociativity]

Not to mention that iron deficiency exists with or without those changes, so wouldn't this be an argument to remove iron deficiency because of insufficiently accurate food nutritional data?

[Venera3]

Yes, anaemia because of an alimentary iron deficiency shouldn't be a thing, or at the very least should have a multi-year latency. Ideally we'd have a way for vitamins and effects to influence each other, so that you'd need an iron-rich diet to recover after heavy blood loss, but that requires code support. Then we also wouldn't need two separate anemia effects.

Hypocalcaemia in a healthy adult should not be a thing, ever and that vitamin could be cut. Alternatively, we can keep it commented out until the game has a Postapocalyptic Lunch Lady scenario, where you have to put together balanced meals for the kids in the refugee center's basement, complete with growth modelling.

Malnutrition of macronutrients should definitely be a thing, but micronutrients are way harder to justify in the game's context. Most of them have more of a public health impact irl, especially in regards to adolescent nutrition and healthy development, but that's not really part of Cata's gameplay.

[int-ua]

Do I understand correctly that skill rust is no longer part of this PR?

[dissociativity]

it affects int still, so if skill rust is enabled, as far as I'm aware int makes it worse?

[int-ua]

AFAIU these rates should've been tuned to represent IRL values to be synced with RDA vitamin values of all the comestibles. Were they previously? Are they now? Don't get me wrong, I like this PR and will enjoy a playthrough with realistic vitamin requirements. But I wouldn't like if they were irrelevant again.

[int-ua]

Maybe just change initial values?

[dissociativity]

rough aproximations of reasonably realistic values suggested by other users here, I'm not gettign a PHD in nutrition to make a CDDA PR, by default it took 90 days to lose all nutrition one started with, I simply multiplied it, later I will look into what drugs should do to vitamin levels, make them drain them more rapidly while under their effects, this should keep them relevant

[int-ua]

Sorry, there are just too many comments to read here. And part of them must be obsoleted by later ones. Please try keeping the original description relevant so that people don't have to read the entire history. Anyway, I'm not the one who will decide if this gets merged, just trying to help. Resolve this thread if you feel it should be.

[Venera3]

You indeed don't need to do a PhD for a pull request, but some cursory googling and listening to the MDs in the audience would not be inappropriate.

[Venera3]

This causes scurvy on Day10, way before any reasonable timeline for presentation ( symptoms should start at 60-90 days). A conservative 40-45 min decay tick would be a better fit, to allow for errors in the nutritional values.

[actual-nh]

Rather too soon, I agree. There are some traits that should cause a lower starting level of various vitamins (Hates Fruit for C, Hates Meat for B12 - see below); will this require C++ coding, or can it be done via JSON?

[Venera3]

I'm pretty sure it would have to be C++. Traits/mutations can mess with vitamin uptake and decay rate, but AFAIK nothing JSON-able influences the vitamin storage at start. It could be a nice parallell trait, something like "Strict Vegetarian Background", which could be obligatory to some professions ( are any religious figures strictly vegetarian?) but would on its own not influence the vitamin uptake/calorie absorption. You could also then slap a harder malus onto Junky, Frail (or whichever starts you in the hospital), so that they have malnutrition effects already at start but nothing keeping them from recovering.

[Venera3]

A normal 80 kg adult has a literal kilo of Ca stored in their bones, with the blood levels being very tightly controlled through bone remodelling/deposition. No intake would lead over years to osteoporosis, with consistently normal blood Ca levels. How you would go about not eating a basic element of all living things is a whole another question, unless you are in a roguelike where not everything has its nutritional value audited. Whenever bone density / breakage and mutations affecting those variables is implemented calcium will have a place as a relevant nutrient, but until then it should be of no consequence.

[Venera3]

See above. Developing hypocalcemia, meaning lowered blood levels because of low intake just does not happen in real life, unless you have severe chronic GI problems (repeated episodes of pancreatitis or the like ), or kidney failure. Both of those are failure conditions in the postapocalypse, and otherwise tracking calcium intake would only be of consequence if bones get properly implemented.

[actual-nh]

Yes, anaemia because of an alimentary iron deficiency shouldn't be a thing, or at the very least should have a multi-year latency. Ideally we'd have a way for vitamins and effects to influence each other, so that you'd need an iron-rich diet to recover after heavy blood loss, but that requires code support. Then we also wouldn't need two separate anemia effects.

Code support would also be needed for the iron/vitamin C interaction - non-heme (non-meat-source) iron is not absorbed well without vitamin C.

Also regarding vitamin C, even what is usually regarded as a subclinical deficiency can impede wound healing - see this article and many others going back decades.

Hypocalcaemia in a healthy adult should not be a thing, ever and that vitamin could be cut. Alternatively, we can keep it commented out until the game has a Postapocalyptic Lunch Lady scenario, where you have to put together balanced meals for the kids in the refugee center's basement, complete with growth modelling.

I'd have it commented out until some interaction with bone healing can be coded in.

Malnutrition of macronutrients should definitely be a thing, but micronutrients are way harder to justify in the game's context. Most of them have more of a public health impact irl, especially in regards to adolescent nutrition and healthy development, but that's not really part of Cata's gameplay.

A character with Hates Meat (especially + lactose intolerance) may start with a low enough B12 level to run into problems; see this PubMed Central article. (As a vegan, I am perhaps more aware of this than others...)

[kevingranade]

The starve test vitamin c failure needs fixing.

[dissociativity]

apologies, work has picked up more again, but I'll be back home early february and well rested enough to continue this PR assuming things go as planned.

[dissociativity]

alright, I'm back home, and starting to feel more alive, I'll look into implementation of your suggestions soonish

[I-am-Erk]

Afaict if the vitamin c test is fixed and the merge conflict fixed, we could merge this and touch up the special cases as we go.

[Night-Pryanik]

@dissociativity could you please resolve conflicts?

[dissociativity]

apologies, been studying, will try to work on it soon, I'm pretty behind on study though

[dissociativity]

oops. I'm just going to make a new rebased PR for this I think