The Nineteenth Byte - 2018-04-20 (page 3 of 3)

« first day (2638 days earlier) ← previous day next day → last day (2190 days later) »

3:17 PM

the internet just told me that the rapture will be on monday

you all ready?

personally i want to get some space pants before that happens

@Poke please no, next monday is a holiday here D:

I planned on staying home, not being enraptured

but it's okay because today is 4/20

WHICH MEANS

Super Troopers 2, babyyyyyyyyyyy

@Poke TEA TIME

Is LQP queue automatic or manual or both?

Looks like both.

Is there a way to differentiate them?

@Poke What is that?

3:24 PM

@Pavel a movie about the funnies

@Dennis Thanks.. it's remarkably hard to commit account suicide :)

Lets not be that morbid

@Pavel it's the sequel to Super Troopers

@Pavel I didn't mean real suicide..sorry

@JoKing If that's my fault, sorry.

3:28 PM

@Dennis so who has to look at it if not you?

My thought process: "Well that joke's kinda in bad taste" => "wait should I actually be concerned right now" => "...nah" => "but what if?" => "ok, no"

@Lembik You mentioned a CM by name in chat. I didn't see anything else.

@Dennis meta.stackexchange.com/questions/309260/… is how far I got

"All in all: everything's working as expected, and someone just needs to have a look at your profile "

just wondering what sort of person if not you

A community manager. Mods can only schedule an account deletion after a certain rep threshold, much like you can do yourself.

got you thanks

and now I am thinking of a ppcg challenge. Have we had one for just counting up on the spot? That is printing 1, 2,3,4,5 etc. to the console but in place, so one number replaces the one before?

if not, maybe there is time for one last challenge :)

3:49 PM

Is there any issue with reposting in the sandbox if the challenge has been significantly updated, and all of the comments are now completed/invalid?

the only side effect that I can think of is that it gets reposted in here, but that seems to be a good think to me

the challenge was posted 2 days ago

0

Q: Number of surjections

Task Given 2 positive integers n and k, where n > k, output the number of surjections from a set of n distinguishable elements to a set of k distinguishable elements. Example When n=3 and k=2, the output is 6, since there are 6 surjections from {1,2,3} to {1,2}: 1↦1, 2↦1, 3↦2 1↦1, 2↦2, 3↦1 1...

code-golf arithmetic set-theory

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Bot Wars at the Auction king-of-the-hilljavascript I have seen KOTH challenges where the bots fight each other, and I have seen KOTH challenges where the bots are in an auction. So I came up with this: The Challenge: You must build a javascript bot function that will fight other bots using we...

@NewMainPosts That was a CMC.

+1 for the impatient upvote-demanding implicit in a line of code that basically reads “plus one, now!”. — Janus Bahs Jacquet Nov 14 '15 at 14:39

@Lembik I feel that it's just "count up forever" + "use \r as the separator".

Random question: Why do you want to delete your account?

Yo @NathanMerrill you around? I have a few KOTH points for consideration

:)

4:03 PM

So originally my arena was a square with the origin at the center

However, since it's symmetrical

I realized that having bots have to worry about which side they start on (Which is undefined) is a nuisance

Do you get what I mean? e.g. one bot starts at (800,0) and the other at (-800,0) under this coordinate system

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

king-of-the-hill The following is an idea I hope to eventually make into a full king of the hill. I've never written or participated in a king of the hill so help would be appreciated. Here is how this is going to work. You are going to be able to write the brain behind a bot. Your bot will ...

So at first I was thinking of having the bots have separate origins -- either the middle of the far left or far right side

@quartata so yeah, there are two ways to solve this: 1. You don't care (and it doesn't matter a ton). 2. You invert all coordinates for a bot

It's more problematic for this KOTH, because movement is relative to the direction you're facing

If you start on RED (left side), moving to the left moves you along positive Y. If you're BLUE (right side), it would move you along negative Y

Because they're facing each other

so?

4:12 PM

So their lefts are different, and the bots have to be able to handle that

can bots not change direction?

They can

then they already have to consider that

if they turn 30 degrees, then their left is a totally different direction

let them calculate that

So having a fixed origin at the center is OK?

Is there a place where I can read about writing koths?

4:14 PM

There is a separate benefit for that, which is that the control point will always be at (0,0)

Well, technically (0,0,400)

10

Q: Tips for King of the Hill challenges

Inspired by this question, what are some tips you guys have for creating king of the hill challenges? What are some things to keep in mind in the planning and implementing of the rules and the controller, and advice on what makes for a good challenge?

discussion challenge-quality

They don't have to worry about that thougj

I prefer my answer, but you can read the rest as well :)

I should probably explain that

@quartata yeah, that's fine. The origin does't really matter, as they are going to be far more advanced math

4:15 PM

@NathanMerrill Thanks! I had been searching koth not king of the hill.

So aside from killing the other bot, there's a control point at the center that unlocks after a while. If you stand on it for long enough to "capture" it, you win. It's supposed to eliminate stalling

I should expand it though

And brings both bots into closer range

@WhatWizard I really recommend asking yourself all of the questions under "Refine your idea"

There are also two places to heal at (0, X) and (0, -X). Can't remember the number precisely

4:17 PM

it's easily the most information-dense part

But there's stuff around the origin is the point

yeah, so I'd argue that making that the origin is a benefit, not for the bot's calculations, but because it makes explaining far easier

I can always show a picture of course

@WhatWizard this doesn't feel like a KotH to me. There's very little interaction between the bots

OK, so the only other thing was that although I can explain the physics of almost everything, there's one thing that's somewhat chaotic

4:20 PM

@NathanMerrill What do you mean? They have to interact.

When a grenade lands, it rolls because it has angular momentum

I can't explain the rolling very well

@WhatWizard oh, I literally just read the sentence that says you that they have to work together

I'd bold that

it's critical

The rest of the physics is very simple except for that

@WhatWizard ok, so I'm trying to think through strategies here: The initial strategy is "work together to get it to the goal". As long as the agar gets to the goal, everybody has infinite energy

However, if say, I get some agar, and simply drop it, then I've wasted all of the energy of everybody who passed it to me, unless they can pick it back up

and get it to the goal (which they don't have enough energy)

No, energy is still limited if you make it to the goal

4:24 PM

how so? Moving without agar is free

There is no way to get back spent energy

...you say you get it back

huh, where?

oh, I misread it

"a point for every energy expended in moving that agar to the bots who moved it"

I thought it said "a point of energy"

so, the goal then is to minimize your effort

because there's a fixed bonus of N points

not quite

you get 10 points for every agar as well.

4:26 PM

(but you still want to ensure it actually gets to the goal)

@WhatWizard yeah that's the fixed bonus

so, if I move it 1 square, I get 11 points

if I move it 10 squares, I get 20

Oh I see, yes that is the case

it's far more energy efficient to move it 1 square

you just have to make sure it actually gets where it is going

exactly

so, this game seems far more interesting with human players

hm

4:29 PM

like, with humans everybody will try to be lazy eventually

but that won't be true with bots. There will be bots that will happily move it as far as possible

simply because there's always a variety of them

so, if I write a bot that simply wanders around, and moves agar 1 square, I'm not sure what beats it

Well it seems unlikely that much of your agar will make it to the end

true, I may move the same agar more than once, so there is some logic to ensuring that doesn't happen

@WhatWizard that depends on where I wander around

I'll just wander around near the end, guaranteeing that I get a part of all of the close stuff

in fact, the end is going to be very populated, as that is the most valuable agar

I'm assuming you are going to allow bots that specifically work together?

yes

I'd definitely add that to the spec, just to be clear

I also don't think the bonus will quite be 10 as well. I was thinking more like 2 or 3.

4:33 PM

I'm not sure how I feel about Purescript: It's a good choice, if people knew it

@WhatWizard yeah, the size of the bonus (relative to the size of the arena) has a large effect

if you really want a purely-cooperative KotH, remove the bonus

that makes this a game of "how well can you work together"

also, efficiency doesn't matter in this game.

The problem with that is if there is no bonus and you reach the goal the winning strategy is to just expend all your energy moving back and forth in place.

yeah

Which isn't really very cooperative.

ok....what about this. Each agar is given a fixed bonus

that bonus is split between the players that moved it based on how far they moved it

a player can spend all of their energy back and forth

Hm, this is an interesting idea

4:37 PM

they get a bigger split, but players are less likely to work with them, and they won't get multiple bonuses

and make the split easily divisible, like 420

so that you have to deal with decimals (or ceil/floor) as little as possible

I'd just use fracs anyway. I don't like floats

Another thought an agar might have a point limit, given to bots on a first come first serve basis. That way you are incentivised to get in on it early but you still want to leave enough so that others will be willing to pick it up.

So after like 100 steps the agar stops giving points to new steps.

I think I am going to go with your idea for the time being.

Seems most interesting.

A couple of other questions: how is agar passed?

or, in a better way: If I place it down, I can place it any where?

does it cost energy?

aka, is there free movement by placing it down?

so, you need picking it up or setting it down to cost energy

otherwise, you can move agar for free the entire way

You put it down where you are standing

and pick it up by standing on it.

I should include this

of course you can stand on agar without picking it up too

4:55 PM

I didn't realize this...but that vision is so small

you don't even give them a proper 3x3 :)

I'm not saying that as a bad thing, I'm just surprised.

I don't know whether it'll be a bad or good thing

it might be too small, but limited information games are interesting

oh man...if you actively memoize stuff, you could get your controller really fast

I don't know if purescript will do that for you

limited info games are very intresting. RE: Formic Functions

~~goddamnit i keep mistaking black hole for a food pellet early on~~

@betseg Wrong tool. This is what excel was made for.

i think matt parker did that

@flawr i thought it was for being unusually good at codegolf challanges.

@betseg I love that video

5:04 PM

@NathanMerrill did you see my other question? We kinda had a bunch of conversations going

er no?

Oh so for the spec I've tried pretty hard to explain the physics of everything (e.g. the actual numerical forces involved)

right, but rolling is hard

But the one thing that I don't really know how precisely it works is when a grenade lands and begins to roll -- it has angular momentum, but I don't really know how much or why

what's the question?

5:06 PM

Can I just say "don't make any assumptions about rolling"?

I feel that's unfair because humans can aim "rollers"

yeah, and bots will be even better at that

It's not that hard, I just can't quantatively explain it

@NathanMerrill but in order for them to be able to, I have to say what the initial state of the grenade is

I think you have to, though

I have translational not rotational

oh, so grenades have rotational velocity too?

5:08 PM

OK so I do need to figure that out then

do they have control over that?

@NathanMerrill I'm pretty sure. They have peculiar bouncing and rolling behavior because of it

@NathanMerrill you can't change it. You click a button and it creates the grenade with xyz initial state

I think you have to get more details about it then.

hmmm...maybe not?

like, this is basically like saying "Go to website and do X"

you don't have to describe what the website does

hmmm...not sure

this is grey area

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Simultaneous Horizontal / Vertical Shift This isn't really finished right now, but I've had this CJam program sitting open for a while now: q~{_S*n1$N*nNo(+(@+(@+}100*]; It prints 100 iterations of the following 2 steps: The horizontal top "rotates" once, moving the first letter to the end; ...

0

A: Sandbox for Proposed Challenges

Output Infinity code-golf math Challenge Output infinity in any convenient format (symbol, word, image, ASCII art,...). Possible outputs 1- ∞ 2- Infinity 3- infinity 4- ,od8888bn. ,.od88bo, d8P' '*88bn. , 'Y8b 88' '*888b. 'D8 Y8b ,'*Y8bn. ,d8P '*Y8bn,. ...

5:23 PM

@NathanMerrill This is a video demonstrating them: youtube.com/watch?v=7at8nwTwMQ0

maybe that would help. You can see they're fairly complicated, but I have most of the behavior formalized

There's no air resistance, so they simply arc through the air until they hit something, then they bounce off and roll

They deal a flat 100 damage if they hit something directly, and a linear scale based on distance if they detonate on the ground

if you can formalize the distance they roll, and simply say that they roll in a random direction, then that should suffice

The fact that they're not point bodies only comes into play when they collide

They're very simple in the air

The thing is that knowing the direction they go is more important than how far

There's no real obstacles for them to collide with, so they're either going to hit the floor or hit directly

Esolanging Fruit

@ETHproductions I'm trying to use addict; how do dynamic names work?

I think I can explain it

Esolanging Fruit

i p[x] fails to parse

5:30 PM

If they had air resistance I wouldn't be able to explain it at all:

The Magnus effect is an observable phenomenon that is commonly associated with a spinning object that drags air faster around one side, creating a difference in pressure that moves it in the direction of the lower-pressure side. The most readily observable case of the Magnus effect is when a spinning sphere (or cylinder) curves away from the arc it would follow if it were not spinning. It is often used by soccer players, baseball pitchers and cricket bowlers. Consequently, the phenomenon is important in the study of the physics of many ball sports. It is also an important factor in the study of...

that would be way too hard, thankfully they dont

what does it mean when a flag is "disputed"?

does this mean it hasn't been processed?

Somebody counterflagged it, most likely from a review queue.

In particular, it probably hasn't been handled by a moderator. VLQ and NAA flags go through a review queue first.

@Dennis Thanks for the quick reply. Next question: Will someone else look at it or not?

NAA as in not an answer? because it's that type of flag.

No, a disputed flag won't show up anymore. NAA is not an answer and should only be cast on non-answers, not wrong answers.

Whoopsies, guess I flagged a lot of answers with the wrong type then..

What type of flag would be appropriate for answers that don't fulfill the criteria?

5:38 PM

@BMO Custom.

Custom is best because it ensures that a mod will handle it rather than the review queues

@DJMcMayhem Ok, next time I'll know. Sorry about that..

0

Q: Fill out the Bubble Sheet

Today was AP exam registration day at my school, and while I was meticulously bubbling in the pages and pages of information required, the idea for this challenge hit me. So, given a string of letters and numbers, output an appropriately filled out bubble chart. Rules: For each character in th...

code-golf ascii-art

No worries!

6:17 PM

@DJMcMayhem You're just saying that so you can better monopolize the few things that do show up in the review queues. :p

I've actually stopped using the review queues since I got steward. It seemed like other people wanted to be able to review things as well

6:49 PM

0

Q: Word Search Solver

I got to wondering yesterday if I could write a program to comb through a given word search and output the answers. It was actually surprisingly easy. Now I wonder just how small we can get. Rules Your first input is a string of n lines, each of which is n characters long Your second input is...

code-golf word-search

7:06 PM

> To simplify the process you can use Excel to generate these commands. Get a list of users in CSV format, and then use this to generate the move request command for each user. Once you have your list of commands you can then either put them into a PowerShell script to be run as a group, or run them individually by pasting them from Excel into the shell.

^ Someone that doesn't know how to PowerShell.

7:20 PM

@AdmBorkBork You're probably the only one here who does, please explain

PowerShell can read directly from CSV files, and you can use that plus pipes to create the commands on-the-fly.

The "advice" there completely ignores the concept of variables.

It would be equivalent of not knowing there's a for command in Python and doing something like x=1; do_stuff x; x=2; do_stuff x; x=3 ...

8:08 PM

@EsolangingFruit Looks like they only work in aliases. Not sure why it's like that though

8:40 PM

Is this challenge idea good or bad? Given a positive integer, output a brain-flak snippet using only (...) and {} that multiplies the TOS by that number

I guess it would have to be output a minimal snippet, other wise you could do print('('+'({})'*input()-1+')'), but then it'll need to clearly specify what minimal means

maybe metagolf?

Guys I did it, I found a way to make an Asynchronous execution model for Funky.

Yeah, that could work, but it's not too hard to find the optimal snippet, so all decent solutions would end up tying

At least, I expect it is. I suppose I've never tried with larger integers

you are now aware that your jaw has weight

I was aware of that before.

8:55 PM

My feet also have weight

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Well-formed Parentheses Gone Wild The object is to determine if the "parentheses" in a string are well-formed, in the sense that they are balanced and well-nested. For example [(ab(c!{})n)] is such a string, while [(ab(c!(d)x){)}] is not (balanced, but not well-nested due to the last closing )...

9:49 PM

@DJMcMayhem It would likely beat the only answer here. (Unless that answer does the same as you)

10:05 PM

@H.PWiz Oh, I didn't realize that challenge was posted

Anonymous

10:25 PM

So I created a really crappy tool to solve a problem some people were having at work: github.com/Mego/Google-Review-Link-Shortener

Anonymous

They complained about the crappiness, so I decided that it should be as crappy as possible while still being functional

Anonymous

So TNB, do your worst. I will only reject PRs if they make it non-functional or inappropriate for a workplace environment.

Anonymous

Also submit PRs to the best branch, not master, so I can make sure that stuff doesn't break before it breaks the live version

@NathanMerrill What do you think of obstacles? The tips page says to remove unnecessary complications and sources of randomness, which the obstacles seem like but I am unsure.

@WhatWizard a good idea

10:38 PM

Ok, I'll do that. Thanks

Now the game is pretty much deterministic except for turn order.

and starting location

right?

that seems like the biggest source of randomness

as I said before, the best spot to be is near the end

I haven't determined how starting location will be determined

It is also very easy to get where you want to go regardless of where you start.

yeah, but that is wasted time

well if you want to be at the end you still have to wait for the agar to start to get there

fair enough

10:48 PM

I'll think about this more carefully once I have a controller

hmmm....I feel like we need something

for example, right now a really good strategy is to take a single agar as close as you can

like, if you get it 80% there, and hope that somebody else will take it the remaining 20%

I don't know you are putting all your eggs in one basket

you'll get 80% of an agar-point, which is the maximum

Also I thought the arena would be about twice the max distance, so only 50%

ok, so 50%

regardless, an egg is more than 50% likely to be picked up at 50% than at, say, 20%

oooh!

what if, instead of a global energy limit, have a maximum energy limit per agar

10:52 PM

hm

and make it like 20% or something

There is also not much incentive to move laterally

except for obstacles

unless you are searching for agar people have dropped

Oh well I had removed obstacles

oh, by "a good idea", I meant that obstacles were a good idea

10:54 PM

Oh ok.

I'll go ahead and put them back then

because obstacles provide interesting situations

yeah I think so.

Ok....in regards with scoring/energy/other fundamental rules: What's the goal here

like, obviously collaboration is a goal

but right now, collaboration really isn't required

you simply move stuff closer to the goal and hope somebody else will pick it up

well the 1 goal is to make it interesting and non-trivial. the number 2 goal is to make it so that collaboration is rewarded.

I was thinking that a way for bots to communicate might be a good idea

but I don't know how they could.

communication often times ruin things

it can work, but wait to see if it fits the challenge

so, for collaboration to be required, you need the ability for players to do something the wrong way

taking the real world for example, collaborating with somebody to carry a wooden pole means walking in step with each other

one person walking too fast or too slow is always problematic

but if you were to look from a single-player's perspective, a person walking fast is totally ok

this is where the interesting stuff comes from collaboration

a single-player view doesn't work

11:04 PM

what if we had two agar piles, and two goals, such that agar from pile A must go to goal A and pile B to goal B.

yeah...

this is clever

have them make an X

oooh, or have one set going north/south, and the others going east/west

with no way to differentiate the agar once they leave the piles.

yep

I like this.

...this is also feeling like a team challenge

but you can't identify your teammates either

11:07 PM

hm, should I assign them teams you think?

its a drastically different game, very similar the ants game and red vs blue

I am not familiar with either of them.

127

Q: Red vs. Blue - Pixel Team Battlebots

This contest is officially over. The Blue Team won! I autoran two sets of 50 battles and amazingly, Blue won all 100 of them. Looking at the stats, it is clear that the cooperative entries of PhiNotPi and Sp3000 were the real heroes. Great job you two! In fact, if you disqualify every other memb...

game king-of-the-hill grid javascript

90

Q: Formic Functions - Ant Queen of the Hill Contest

Watch live | Active answers | Add new answer | Chat room | Source code | Leaderboard Thanks to Best of PPCG 2017, this challenge will have a 500 rep bounty posted at some point soon (probably during March or April 2018). I will run parallel tournaments until then for each new player/update that ...

grid game king-of-the-hill javascript

Oh I have seen the ant game

that is what made me want to try this out.

ok...so if you make a team game, many things have to change

you need a bigger vision

11:10 PM

I don't think I am going to make it a team game.

ok

...are answers that try to ruin others competitive?

like, lets say there was a bot that simply moved pellets back into piles

that bot obviously isn't trying to win, so I'd say no

I don't know. I do plan on limiting things to one entry per person.

however, what if there was a bot that would identify formations of piles, and would move piles he liked forward, but other piles back?

It might be hard to determine what is intentionally antagonistic and what is poor design. For example a bot the intentionally puts agar in the wrong piles vs, one that does so accidentally.

I mean, we have the same issue with other "submissions must be competitive" situations

like, we point to certain things (like not removing whitespace), but other times, we simply have to give them the benefit of the doubt

anyways, I think that that 40% is a good energy limit

it's basically 33%, but gives them some wiggle room

New Sandboxed Posts

11:16 PM

0

A: Sandbox for Proposed Challenges

♫ I see a window and I want it painted black ♫ Yes, I know this is a popular mishearing of the lyrics. But instead of a red door, I really do want an (application) window painted black. Your standalone program should launch an application window at least 400x400 and fill it entirely with black....

@NathanMerrill Ok. Thanks for all the help btw.

yep!

other ideas: 1. You may want to consider allowing multiple instances of the same bot (and even placing them together) 2. Consider allowing bots to choose their starting location

yeah, I'm thinking I might want 3 or so instances of every bot.

@WhatWizard oooh! make it so that players can't just ignore one of the sides. Make their score the minimum of their two agar scores

Hm.

that could be pretty hard

11:30 PM

It's important

consider what happens if a set of successful bots started going N-S

bots are more likely to also go N-S than E-W

I could make it so bots can't tell the difference

true, but then teams that depend on each other won't work

because they might be on opposite sides

that's a decent option though

another option is to scale the score of a certain agar based on how popular it is

that's a good one too

like, if 20 N-S pellets were delivered, then E-W pellets are worth 20 points at the end

something also to consider is the map size

if you have large maps, then you have more room for formations

yeah. I might want to make it scale with size as well

11:34 PM

if you have small maps, then players have more of a chance of doing both sides

well, I was thinking that a map of size super small (like 12) could be interesting

a fixed number of players per map

that is something I feel like I just have to play around with.

yeah, for sure

if you also do small maps, your controller could do a genetic algorithm or something

try to find the teams that work the best together

oh yeah that's a cool idea

@betseg plz no

« first day (2638 days earlier) ← previous day next day → last day (2190 days later) »

Transcript for

Apr '1820

The Nineteenth Byte

The Nineteenth Byte: General discussion for codegolf.stackexc...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile