saves some interobject/process communication

The buyer doesn't know the product the seller is going to offer

they can give a price in all 5 goods

I don't understand the scoring, but it sounds quite interesting

You just need to make sure that people will on average produce less than 5 products a year

@MartinBüttner actual scoring is just how long you survive without starving. the part about product scores is related to production, not score-score

there are 5 products, with an average score of 20

@Sparr oh okay... it's getting late over here

I should turn to video games instead of PPCG at this hour :D

furthermore, one of the base scores will be 3...meaning the max a player can produce is 5

@NathanMerrill check my math... the product with base "score" (you should choose a different word, how about "productivity"?) 3 will enter the game at a rate of 3 per player per turn if everyone chooses it, right?

average rate

yes

and the average for a given will be 4 per turn if all players choose that product

leading to an average of .8 of each product produced each turn

will those base scores be different each year? or game? or?

each game

ah I see

does the productivity vary per person?

I hope not ^^

sort of

the total productivity doesn't

well the base score doesn't right?

but the individual goods do

the base doesn't either

including base scores?

ah okay

phew

so only the variation about the base score is player and year specific?

no

not years

every year, every person can produce the same amount

oh, not even the variation?

k

but individual persons have different variation?

which stays constant for them thrughout the game?

so... if the base productivity is 3 for product A, then people will have values from 1 to 5

but from year to year, the same person will have the same productivity

aren't the people with a high variation on a product with a low base productivity favored?

yes

but that's why I play 50 games

why not make the variation change every year? that would probably also be more realistic

So, in terms of production choice, my bot has the same options every round of a game?

yes Phi

The point is to figure out what is rare I suppose

Martin, because then it becomes more obvious what values you under produce

yes overactor

ahh

you need to figure out what is rare AND what you have an advantage in

that's clever

what's the lowest base score for a product?

mmm

didn't think about that

probably 2

so you could have 0 productivity on a product?

yes

k

That's a maximum productivity of 4 of a certain product across the game

which means, maximum base score is 6?

no, Martin...you could have 2,2,2,2,12

yes overactor

oh okay

But people might overproduce it when they find out

I can see it balance out

...but its impossible to survive forever

Here's a problem I think could happen without careful planning:

What If everyone always produces their best product?

ooh...

no wait

20+2*5 = 30

your right

the one they have the best variation on

you're

@NathanMerrill you can edit chat messages by pressing the up arrow key ;)

I've done it before

habit from other chats :P

If I have high enough productivity levels, I would hardly have to trade, ever.

That would average a production of easily more than 5

If the productions levels are A B C D E, then we must ensure that 1/A+1/B+1/C+1/D+1/E < 1

@PhiNotPi why is that?

Otherwise, they would be able to avoid trading forever.

very high numbers for fulfill that inequality

I actually meant >

aha

that makes more sense

Didn't edit in time.

yeah

But, if the sum of the reciprocals is less than one, they never have to trade, ever.

In order to ensure fairness between players, you should make sure that the sum of the reciprocals is about equal for everyone.

it should be close to 1

so you get enough time to figure out what you should be trading

wow I'm really stumped... where do you guys get the reciprocals from?

oh that's how often you can produce to stay at one product per year on average

or something

yup

they all represent how often per turn you'd have to produce each product to have enough

if they'er all 1/5 that means you have to produce once every 5 turns

of the sum is smaller than or equal to 1, you can survive on your own, guaranteed

if it's lower, you can't

yeah

cheers

ok, back

so, I decided that each person will have to consume 2 of each product

and the total base score will be 30

okay

which means that if each person produces their top product

the total score will be 30+5*2 = 40

how is that different from reducing the base score to 15?

its not

but if I go down to 15, then +2 makes more of a difference

than at 30

less variation?

yes

ah yeah

So when do people necessarily start to drop?

so, average max product production each turn is 8/5 and consumed is 2

@Geobits you said earlier that [pick-your-side] might not fit when you could be paired up in coops once with each other player. I think that would not necessarily fall in the same category as the tag you originally proposed. that's still more of a KotH, because you'd declare a winner in the end. the point of the new tag is to signify that there is no single winner, but a group of people.

@NathanMerrill where does the 30+5*2 come from?

25 turns overactor

so, there is 30 base of each product

however, say each person produces the product they receive the +2 in

there are 5 products

so 5*2 is the max additional product

5*2?

is that the total over 5 players on average or something?

Yeah...let me explain it a different way

The easiest situation is where each base productivity is 6

there are 5 players, each with a +2 on a different product

the decide to trade with each other 1 for 1

they*

Oh, I got it

so, they are producing 30 of each product each turn

but if they do their maximums

they are producing 40 of each product each turn

but they consume 50 of each product each turn

Which is just enough to sustain the entire population

no...it takes 50 product to sustain 5 people

Oh right

50

not 25?

I think there may be a problem with there being only 1:1 trades and nothing else.

@Sparr he upped it to two of each product per year

oh

Yes, it is now 50.

why Phi?

got it

@PhiNotPi someone will start offering better trades when they are about to starve

okay, that makes sense

390625: (30, 2, 2, '[[[[[[]]]]]]  [[[]]]  [[[[]]]]', '5^2^3')

getting some results that make me happy from my new bracket numbers algorithm

How long did that take?

My personal victory is that my Rock-Scissors-Paper-Lizard-Spock bot was in second place earlier today. (Edit: it probably won't stay that way, though)

@NathanMerrill will people know how many other players are still alive?

@MartinBüttner not a useful metric, that's part of my table-generation step. I'm only generating up to 8+8 and 8*8 and 8^8 for right now. that's just 5^8 optimized

I'm just trying to track down bugs in my code before making bigger tables

No Martin

actually, I don't see why not

There should be more of an opportunity to explore the market I feel

I could allow the player to query the number of players currently trading before they decide whether to buy or sell

go on overactor

hmm

info about previous trades and offers would be useful, but maybe too much info

info about total amounts of each product offered or traded might be a happy middle ground

as a trader, though, you wouldn't know that

how about you get shuffled into an array

depends on the market. in a public exchange you know totals and prices.

and you can trade with the two people besides you?

no, I don't like either of those

how about the age of the products you're trading for?

that's realistic-ish

but a lot of data to track on the back end

how am I supposed to know that?

assume FIFO on products produced?

whenever someone eats a product, pick it at random

if I have 10 apricots, and I produce 4, and trade 3...which 3 did I trade?

I dunno, just brainstorming

the age isn't a bad idea

it is a lot to track, but I'm ok with it

if products are old on average, there must be more of them going around

...I would probably do a queue, not a stack