So I recently decided the best way to get a bunch of foods and their nutrition information for a my game, was via the usda database of foods. So a 3.9gb json later and now i have 466,563 foods with all the nutritional information on each.
The question is how much food do i actually need, and what nutritional information should i keep?
Not sure i need 3.9 gb of ram filled with just food(or maybe i do, this is a weight gain game after all). Feel free to share your thoughts!
the amount of unique food depends on the game itself… I see it this way: if the game is long, 1 hour for example, then you can cram quite a lot of unique food (maybe 10 or so), so that the player discovers something new for some time, which helps to retain.
but if we talk about a short game of about 15-20 minutes, then I think 4-5 unique food will be enough, there is no need to give it to the player every minute
in any case, it’s all subjective, you can’t say 100% what kind of food, how much you need to show the player, you just need to calculate and think for yourself
I’d imagine that 466,563 foods is, indeed, too many. Granted, if that’s the same database I’m thinking of, you don’t necessarily have 466,563 unique foods, but rather 466,563 UPCs, many of which will be differing sizes, brands and variations of the same foods (I think I looked up Tomato Soup and there was roughly 300 different UPCs for that).
I suppose I’d have to ask what kind of game the food would be in, and thus what sort of context. A simulationist slice-of-life game? A post-apocalyptic survival game? A cottage-core farming game? An oddball roguelike? Each player will want something different from the food they encounter (raw numbers vs. believable availability vs. appeal), and that will inform how much or what kinds of food you put in the game.
Just remember that, unless the game is specifically about eating every kind of food, players will likely not seek out every kind of food, and you shouldn’t bother making a thousand different food items to put in the game when you can expect a player to only see/use maybe 40 or 50.
Its a simulation rpg style game. The intent was not to add half a million foods rather to add probably 1,000 food items will full nutrition information. The player actually does not get to much choice on which food items, rather its to prevent repetition. Seeing you eat a pastry every other day gets boring fast. Also im trying to maximize replay-ability by allowing a wide variety of foods.
as I think, there is nothing wrong with having the same food in your inventory, variety is good, but you need to remember that the player does not really need to get new food every step, with different characteristics and effects, most food and other items will be repeated not because the developers are too lazy to come up with something new, but because there is no need to come up with a unique item, it will be enough for the player to have 5 pieces of unique food with different situational effects that he remembered, than to have 15 unique ones in his inventory, it will not be so easy to remember what each one does and you will have to read every time what they do
but in any case, the choice is always up to the person who makes the game, he wants 1000 unique food, for God’s sake, the only question is whether he can finish the job or not
For me, resources like this are interesting, but only useful as far as they help support the experience you’re trying to create.
I don’t think most games would benefit from this sort of massive amount of information, but perhaps one where becoming an expert on food minutia might.
This is just a hypothetical, but perhaps something like managing a supermarket with a weight gain element. Get your customers hooked on surprisingly high-calorie foods and make money while keeping their esoteric demands served.
I think the bigger question is then how you present that much information to a player and have them manage that distinction. If you have too many items, do they need to scroll a bunch? Do they need to search? How do they tell each item apart easily to make sure they get or use the one they want? Etc…
Two thoughts
First, raw quantity of food items is less important than number of distinct food items. As a player, I don’t want 10000 bowls of oatmeal, I want different items to be recognizably different
Second, food items have an opportunity for characteriztion. Is the feedee/gainer a foodie, wanting to try everything with little repetition? Are they the opposite, picky and sticking to a small subset? Do they have allergies or intolerances? Do they prefer high-end foods? Are they from certain parts of the world and prefer foods from that region? Lots of ways to get variety even if you don’t have half-a-million items
I would go about it the other way around and see how many distinct food items the game needs and can support, without making it more boring and annoying to deal with.
if the items provide basically the same benefit then why even bother engaging with the system and if there are like 1k food items you have to sift through just to find which one is the best that doesn’t sound like something fun to do.
there also need to be enough challenge to the game where it feels like there is a point to engage with the system. otherwise you might get the zelda BotW issue where you can just munch on raw ingredients because there isn’t enough reward to make engaging with the system worth it
same goes for the nutritional info, how many things does the game support. I can only really thing of 3 nutritional stats that affect gameplay. you could have the main 3 stats have sub-stats that need to be filled but that feels like making the system more complex “just cause”
there is also just that much you can do with real life information. like what vitamins do I need to resist magic damage? I guess you could tie each vitamin and whatever to some in game stat but then you might end up in a situation where the balance of the game goes out the window unless you change the stats of the items or come up with a clever system to prevent it.
personally I would just use that sort of data to make “tiering” items make more sense, like the highest calorie item out of x,y,z heals the most
I appreciate all the helpfull thoughts, and look forward to more, keep in mind,
Is the reason this game this game and almost every mechanics exists. Like the fact that i have made the ai able to process 10,000 npcs turns(game is turn based) in milliseconds(there are >~13,286,025,000,000 possible choices for a turn for each player/npc). Do i need 10,000 npcs no, did i want 10,000 npcs to fill a city, YESSSSSS, therefor i made it happen.
The point is not to have
its to have a few highly detailed bowls of oatmeal, hence the usda database with all the nutritional info.
Why do i want the nutritional info, well see this topic belly-queue-design-concept-for-text-adventure-and-or-rpg-games I have designed a in depth digestive simulation base on the idea in this topic. And it would be fun have alll sorts of interesting effects based on whats in a food.
Lastly a major goal of my game is not to have players micro manage so much as macro manage, so, players would be picking food categories with the game ai selecting the specific item for them. Pick a corroborated beverage and be bloated, by what who knows you’ll find out when playing.
To add my two cents, and with what the others mentioned, why not take some inspiration from Chub, Chomp, Chill and add a cooking mechanic? Basically, focus less on complete foods and try to go back to scratch. Think of what the people have to cook and work with.
Now granted, this would probably focus more on world building. However, food could be in great service to this.
If you take the setting into account, it can greatly influence the diet of your society; two examples of this that come to mind would be: this video:
And an Isekai story called “Ascendence of a Bookworm”.
Heck, looking into the history of various food sources that have fallen in and out of popularity over the millennia can be fun in and of itself: for instance, did you know that for a long time, tomatoes were not consumed because they were believed to be poisonous, like other nightshades? It wasn’t until a statesman in Salem, New Jersey in 1820 publicly ate an entire basket that they were considered edible. Otherwise, they were merely decorative.
@zdeerzzz Here’s a vague description of a mathematical approach:
Every listed food exists in a high-dimensional space, where each dimension represents the %DV of some nutrient. So the appropriate amount of calcium tablets is pretty much the vector [0, 0 … 1 … 0], where that 1 was in the calcium dimension, and a more normal nutritionally balanced food would be nearer to the [0…] → [1…] diagonal.
To provide players and algorithms the most ability to make nutritional trade-offs, pick points that fill the area as evenly as possible. So, pick some subset of the half-a-million kinds of food that maximizes the variety of angles from the origin in this nutrition space. How to formalize and calculate this I leave to you.
You may also want to have a dumb script pick 10 items totally at random, then check the results for kinds of things you just don’t want at all.
As for how many kinds of food in the game? I’d say try 10, 100, and 1000 and see whether you notice any differences, then go from there.