Chat:World/2022-06-21
jwpapi: I’m rank 1 at wood 1 league, how do I rank up?
antiwonto: [auto] Hey jwpapi, here is a :taco: for loggin in today while it is quiet. You now have 1 tacos
noidealol: hello guys I just recently had an interest of learning how to code but to my surprise it's so hard! it's like alien language haha This is not for me.
jwpapi: noidealol you probably try too much of a challenge, this website is not for absolute beginners
miobyte: who can give me salt?
5DN1L: What is salt?
5DN1L: Automaton2000 do you know salt?
Automaton2000: as you can see the code that is
5DN1L: This? https://en.wikipedia.org/wiki/Salt_(software)
waterproofsodium: gm
antiwonto: [auto] Hey waterproofsodium, here is a :taco: for loggin in today while it is quiet. You now have 49 tacos
skelesans: hello :)
Nerchio: give me taco
Nerchio: :rage:
j4at: :taco: Nerchio
antiwonto: [auto] j4at has awarded Nerchio 2 tacos. Nerchio now has 3 taco. j4at now has 11 taco
Nerchio: :yum:
jacek: :taco: Nerchio
antiwonto: [auto] jacek has awarded Nerchio 4 tacos. Nerchio now has 7 taco. jacek now has 48 taco
Nerchio: I will be rich soon
skelesans: :money_with_wings:
Leeward: taco
antiwonto: [auto] Hey Leeward, here is a :taco: for loggin in today while it is quiet. You now have 4 tacos
TINOUAINANI: Hi
ArianPunk: it's funny when you submit the puzzle and some tests fail without letting you to at least see the error ...
jacek: some tests are invisible to prevent hardcoding
ArianPunk: cause you might hardcode them! it don't make sense.
jacek: !
ArianPunk: you didnt read carefully
jacek: :nerd:
ArianPunk: it's ok. too hard for you.
Uljahn: you can enable custom testcases in the settings on the left (mode -> expert)
ArianPunk: and i covered some cases too by custom mode, yet no progress when submitting.
Uljahn: which puzzle?
jacek: i cant find it
iamsofia_rless: hy
ArianPunk: found the issue in forum.
jacek: :tada:
emh: I want to optimize a lookup table for a matrix. how can I use symmetry of reversing columns and/or rows to divide the table by around 4 (so that it takes less than 1 second to init lookup table)?
emh: I tried to maintain the invariant that first row should be less than last row and first row should be less than reversed first row
emh: maybe I should think about first column and last column instead
emh: my first invariant worked for around 10M out of 11M matrices
emh: well.. I don't know that for sure, but at least the ones that didn't hit unitialized matrices
jacek: 10M matrices? oO
emh: out of 1<<24. I want to Smash the Code with 3x4 row (where row is 6 cols) connected component lookups
emh: I know how to connect the 3 lookups fast
emh: just the problem is to init the table fast enough. I could use 3 rows for 18 bit and 4 lookups instead to simplify
emh: or maybe I should use a partial cache instead of initializing the whole table up front
emh: I'm able to do 23 bits in 950 ms or so, that's why the last bit is annoying me haha
jacek: try 23.5 bits
emh: hehe
emh: could just loop as far as I get in 1 second and do the rest dynamically
emh: that would be 23.5 I guess
jacek: afaik msits does his endbook tables in 1st turn in 1 second and few turns after
emh: ahh ok, but then search suffers for it
emh: maybe a good search is not so important for first few turns
jacek: or ask him. i bet he can do everything in O(1)
emh: hehe
jk_java: looking for approvers on a music CoC game
jk_java: https://www.codingame.com/contribute/view/20739e7af87e7682761095576937ba9196bba
Astrobytes: spelling mistake in your stub text 'trasnposed', other than that looks good. Won't approve as I don't clash
Astrobytes: be patient and approvals will come
geppoz: I know, question already asked 1000 times, but:
geppoz: first THROW then GIVE? or viceversa?
putibuzu: first throw, then give
struct: throw -> give -> and then you get the card
geppoz: ty
ThomasGottschalk: Im the real one
ThomasGottschalk: Are there any girls here
Astrobytes: Many. Now go away.
Crosility: .. I'm sure there are. But it just feels creepy to ask.
struct: thomas behave
Violinie: :joy: is this a dating site?
ThomasGottschalk: Im a girl too
Crosility: I kinda think of the people here as a collection of like minded brains, not.. "what are you made of"
ThomasGottschalk: can someone help me with level 1
Kanawanagasaki: hmm, codingame still use C# 8.0
ThomasGottschalk: Mad Pod Racing
struct: follow the tutorial
Violinie: hint: read the code carefully
ThomasGottschalk: ok ok thanks
jacek: Y Y or X Y :thinking;
Ayza:
ThomasGottschalk: I cant doi level 1
ThomasGottschalk: its already advanced
ThomasGottschalk: Can someone help me with Mod Pod Racing
Crosility: Then back up, do some simpler puzzles.
Crosility: At greater levels, there's a lot more complexity. If you aren't getting stage 1, you should definitely get some foundational understanding.
Violinie: I'm stuck too :grimacing::sweat_smile: where are easire ones?
Crosility: Algorithms Tree
Crosility: Horse Racing Duals, MarsLander, Power of Thor, The Descent..
Crosility: To name a few.
Violinie: I'll try it (again), thanks
ThomasGottschalk: Hey Violinie
Crosility: Keep trying, if still struggling, under Activities tab at the top there's a "Learn" Section.
Crosility: If you sort by popular, lot of intro/beginner guides in your language of choice.
Violinie: Great! Thank you :yellow_heart:
Crosility: No problem.
geppoz: desk 0 and desk7 are considered near in terms of GIVE ?
therealbeef: yes
geppoz: ty
DomiKo: so I can't calculate how many cards are on board? or can I?
struct: you can
struct: sum of all cards
struct: from 0-7 5-sum
DaNinja: 5 - all your cards - all opp cards
Astrobytes: it's that question again struct :D
DomiKo: ehhh
struct: I should have automated it
Astrobytes: heh heh
DomiKo: do it!
struct: for bonus is 36 - sum and 100 - sum for tech
DomiKo: thx
DomiKo: I'm just starting to read eveything ehhh
Astrobytes: plenty time DomiKo, also at least the bugs are all fixed now
struct: I dont remember how to add commands to antiwonto
DomiKo: Yeah, I've waited for that
struct: Are you going to commit to the contest now DomiKo?
struct: or you still don't have time?
DomiKo: I'm going to
struct: gl
Astrobytes: that's the spirit!
DomiKo: I need some time for preparation for internship, but I believe I have plenty left
DomiKo: But that game looks hard to make something more than if-else
Astrobytes: Priorities DomiKo, priorities! :P
DomiKo: :sweat_smile:
struct: Some people at the top have sim
DomiKo: a little of sim?
struct: last time I saw msmits he was bruteforcing all moves
struct: + playing a few more
DomiKo: that make sense
DomiKo: I will try exactly the same
DomiKo: should be easy to debug for 5th app
Astrobytes: or just output first given move until silver ... :P
struct: and some moves are easily pruned
MSmits: yeah, me and euler do the same, just eval differs
DomiKo: Astrobytes that's what I did :D
struct: Like you dont want to automate cards that you alredy have 2 off
Astrobytes: lol :D
eulerscheZahl: hi
struct: you dont want to train if you cant draw
struct: hi
Astrobytes: hi
MSmits: hi
struct: and a few more stuff
struct: that should be obvious
DomiKo: a lot of ifs to write
struct: I only have 187 ifs
ThomasGottschalk: EinsAchSiebeeen
struct: But I have repeated code atm
Astrobytes: 71 ifs here
struct: MSmits I have one question for you, when you do coding do you draw all cards, or you draw 1 card and create a state for each?
struct: If you don't want to reveal it's okay
struct: No pressure, :hammer:
Astrobytes: :D
eulerscheZahl: 76 if. close to astro
eulerscheZahl: and i solved my timeouts i think
struct: nice
MSmits: draw all cards?
MSmits: you can only draw 1 card
eulerscheZahl: zobrist hashing
struct: so you draw 1 and create a state for each of them?
struct: I tried to draw all, I knew it was a bad idea, my sates exploded
struct: :D
eulerscheZahl: but for now i don't compute incremental hashes, so it's quite expensive
MSmits: i use a recursive algorithm, like minimax
MSmits: i dont create states
MSmits: i modify
Astrobytes: a dfs
Astrobytes: ?
MSmits: yeah dfs
eulerscheZahl: :D
MSmits: backpropagating the result
DomiKo: good old dfs
struct: Yeah but coding draws 1 card
MSmits: it does
MSmits: and then it's on to the next phase
eulerscheZahl: better known as single-player minimax to some
struct: But which card do you draw?
Astrobytes: :D
MSmits: i loop over all possibilities
struct: ah ok
struct: Thanks
MSmits: http://chat.codingame.com/pastebin/d46c9d1e-48be-44a5-80ad-19b61ca32347
MSmits: ah doesnt work
MSmits: sec
MSmits: for (int i = 0; i < singles.size(); i++) {
MSmits: hand[singles[i].card]++; draw[singles[i].card]--;
MSmits: totalProbability += singles[i].probability * Play<normal>(p + 2, false, depth + 1, 0);
MSmits: you figure it out
struct: Yeah, I understood it, thanks
MSmits: it's a loop over all possible cards
MSmits: and then i go deeper into the tree and weigh each result by probability
MSmits: drawing 1 card means probability is easier
MSmits: drawing 2 cards is more complicated, you dont want to split this into 2 phases
MSmits: combining them into 1 probability of a cardpair is much cheaper
struct: yeah
Astrobytes: isn't it eulers job to leak ideas? :D
Tien118: hi
struct: hi
DomiKo: hi
StevenV: Invalid paste id, perhaps it expired?
eulerscheZahl: i don't have ideas to leak currently
Astrobytes: Smits has got your back, all good
DomiKo: I have ideas on how to steal ideas :D GUI now helps with that :D
eulerscheZahl: opening book for the first few turns
darkhorse64: euler is watertight
Astrobytes: It's that famous German engineering!
MSmits: it's not much of a leak really
MSmits: i already said before you need to try all possibilities and weigh by probability
Astrobytes: nah I know, just poking fun while we still have a chat
MSmits: ahh ok
DomiKo: "still have a chat" :sob:
struct: probabilities, I need to email a cat to get algos that do well with that
eulerscheZahl: you are pulling us down
MSmits: I'll share a lot of math after the contest. I have more probability stuff like this. It's mostly common sense though, I don't use fancy stuff
darkhorse64: meanwhile, discord CG rooms are really quiet.
MSmits: you can work a lot of probability calculus out on your own with a pen and paper without needing a math class
DomiKo: discord CG rooms?
eulerscheZahl: i gave up linking other contests there, little to no response
Astrobytes: the CG discord DomiKo
Kowskaner: I cant even solve the Easy practise. I feel so dumb
DomiKo: I thought they made some special rooms
darkhorse64: that's wat I said
Astrobytes: and the other Special CG Discord Room is for post-chat removal
darkhorse64: that's what I said
darkhorse64: oops
eulerscheZahl: that's what she said
Astrobytes: :D
eulerscheZahl: sorry, jacek isn't here
DomiKo: someone had to do that
jacek: but i am
Astrobytes: then shame on you
Astrobytes: for not outputting your normal responses
darkhorse64: Could we write a jacek2000 ?
Astrobytes: it would have to be a discord bot
Astrobytes: (and be careful what you wish for!)
struct: plain text over here
Astrobytes: we have salt by the kilo here
MSmits: jacek are you training us like a nn?
struct: no MSmits
struct: dont give him ideas
MSmits: sorry
Astrobytes: :D
jacek: :D
putibuzu: funny thing: I just realized that each player only draws 4 cards per turn, thought for some reason it was 5 :sweat_smile: luckily I didn't hardcode it (too much) :D
geppoz: at start 4
geppoz: but along the game that can become 5 or more
geppoz: permanent skill ARCHI
jacek: so permanent
darkhorse64: permanent until release
miobyte: is there anyone knows k8s?
putibuzu: initially I thought the game had a bug and I spent 1+ hours reading the ref and trying to spot the "bug" :D :cry: :D
borrel: there are bugs for sure
struct: I don't think there are more bugs
struct: maybe the viewer can be confusing
struct: But I haven't found anything else
putibuzu: but I agree the viewer could be better
Astrobytes: it looks pretty at least. though it's better with the scrollbar app tooltips
darkhorse64: There a minor ones: possible moves do not show trade bonus/skill, pile are not shown cleared in the outputs
darkhorse64: piles
putibuzu: wdym pile are not shown?
Astrobytes: fix those ankles
darkhorse64: For instance, when the discard or draw pile is emptied, it does not appear in the inputs
putibuzu: but that is intentional from what I understand
putibuzu: not sure why the dev wants it that way though :thinking:
darkhorse64: Yes, but I think it's bad practice. Implicit behaviour always leads to problems
Astrobytes: afk
struct: Maybe he went over the limit of how many inputs he could send in total
struct: But it's unlikely
struct: I think the limit for sdk is around 100k?
putibuzu: no idea :sweat_smile:
darkhorse64: In this particular case, I was under the assumption that the whole state was published and did not clear my piles until I saw that my card counts were sometimes wrong. I moved 25 ranks higher in the top 100
darkhorse64: when it was fixed
darkhorse64: true, i should have read the statement more carefully and use more defensive programming
eulerscheZahl: while i didn't make the same bug in my code, it surprised me as well that some input is missing
eulerscheZahl: for me it was just null and i crashed
struct: for me it was never a problem tbh
struct: I didnt even notice it
jacek: neither did i
darkhorse64: All in all, it's a very nice job from [SG] guys. The artist is also outstanding
struct: Yeah
struct: The bugs were fixed pretty quickly
GNye22: Is there a way to do clash of code with a single language? As in, all players must utilize the same language? Cuz it's almost impossible to beat people using Python when I'm using Java
struct: GNye22 only on private clashes, but sometimes there are clash events where everyone has to use the same language
j4at: boblob is beating everyone with C :p
pedrosorio: GNye22 start using Python
putibuzu: Agreed darkhorse64. I'm having so much fun with this :)
struct: jacek what was the max number of moves?
jacek: 198 for referee
struct: thanks
jacek: without TASK_PRIORITIZATION n 8
jacek: so 206?
MSmits: yeah have to add 8
struct: thanks
MSmits: what do you want to do with that information?
MSmits: I haven't needed that
jacek: reserve()?
MSmits: dunno what to use it for. I guess for a policy network to get the number of output nodes
jacek: :thinking:
jacek: you sometimes have weird ideas
MSmits: i do
struct: You caught me MSmits
MSmits: :)
putibuzu: on the NN topic: is someone training or working on an eval NN+ search?
geppoz: what did you mean with "TASK_PRIORITIZATION n 8" ?
MSmits: if they are, they will keep it secret
geppoz: you cannot swap with bonus?
MSmits: geppoz the input doesnt give all possible actions
MSmits: swapping with bonus is not included
MSmits: but possible
geppoz: ok ty
putibuzu: MSmits hmm ok, I sometimes forget this is a serious competition :D
putibuzu: I'm very curious about the PMs then :D
geppoz: and also swapping bonus with n ?
MSmits: all 9 cards with all 9 cards yes
MSmits: excluding TD
geppoz: ok ty
emh: arghh.. somewhere in my code there is a 64 bit operation lurking among the 128 bit ones. chops my components at the 64 bit boundary. I want to find it. but how.. just read and read and read I guess
putibuzu: refresh my memory please: CG starts timing my bot after all inputs have been accepted by my bot right?
struct: no
struct: after 1st input
darkhorse64: after the first read
putibuzu: what happens if I have computationally intensive code before the first read? :sweat_smile:
darkhorse64: untested by me
struct: you will use time anyways
struct: and you cant really measure that time
struct: because the process is frozen
putibuzu: I guess that is why my bot is timing out right now
putibuzu: What is good practice to time the bot? start the timer after the first read, then read everything and do computations afterwards?
struct: yes
putibuzu: Ok great thank you very much :)
putibuzu: Let's see if I can make my bot work :sweat_smile:
eulerscheZahl: new strategy: collect training+coding cards. immediately pass them on to your opponent and hope for timeouts
darkhorse64: stuff your opponent
kovi: i like that eulerscheZahl
struct: this makes me remember of pb4 test on ooc
therealbeef: training + coding + task prio
putibuzu: yeah task prio branches heavily :D
geppoz: I think there is a missing input
geppoz: the "play_count"
geppoz: in player informations, when phase is "PLAY_CARD"
darkhorse64: implement a counter that you clear on move
putibuzu: in principal you can deduce it from preceding actions
geppoz: y, ofc, but since it seems there is the assumption that all state is given...
darkhorse64: wrong assumption
geppoz: :D
PatrickMcGinnisII: can u cont_int a bonus skill card?
Astrobytes: Yes
Naak11: how does this chat work?
Astrobytes: People type, and people type responses. Like a conversation
PatrickMcGinnisII: tx Astrobytes
Astrobytes: np
Naak11: wow, i hadnt realised tht
Astrobytes: :D
Astrobytes: xmpp
Astrobytes: but don't get too attached to it
Astrobytes: it's being removed later this month
Hadd2: won't happen if we riot
Astrobytes: lol
Astrobytes: the French spirit! :D
Hadd2: devs don't try us
Naak11: why is it being removed?
struct: Got an interesting error in c++
putibuzu: they should make it a challenge: develop the best CG chat, easier to maintain, more mod tools etc etc community and CG will vote on the winner
struct: Never saw it before
Astrobytes: costly to maintain/replace, outdated, that kind of thing
struct: /usr/bin/ld: /usr/lib/gcc/x86_64-linux-gnu/11/../../../x86_64-linux-gnu/Scrt1.o: in function `_start': (.text+0x20): undefined reference to `main' collect2: error: ld returned 1 exit status
struct: this in cg
Astrobytes: linker error, weird
struct: only happened once
Astrobytes: did it happen quickly after the previous play?
struct: I guess it can be that
Astrobytes: only thing I can think of rn
struct: thanks, I'll report if it happens often
Astrobytes: it's an odd one for sure
struct: to me it's irrelevant, but if it happens to a top player if might matter
Astrobytes: indeed. afk again
jacek: :iphone: :door:
TINOUAINANI: hi, guys, I have a question, is there a way to know which 2 cards TRAINING will give you from draws ? or it's completely random?
eulerscheZahl: random
TINOUAINANI: no specific seed in their source code?
jacek: you can manually give seed
jacek: options under ide
jacek: and well, knowing seed would be cheating eh
eulerscheZahl: still some unknown, what the opponent did
TINOUAINANI: how do they generate their seed?
eulerscheZahl: CG magic in the backend
TINOUAINANI: ok thanks guys
jacek: they use nuclear waste decay
struct: seed += 1
TINOUAINANI: another question, what's happened if the draws empty and you used TRAINING?
MSmits: you get nuthin'
MSmits: well discard would be reshuffled first
struct: unless you have cards in the discard pile
TINOUAINANI: ok thanks
TINOUAINANI: so if it's empty you get two cards from discard?
struct: if the discard has enough cards
struct: yes
TINOUAINANI: thanks
MSmits: better to put it like this: When your draw it empty, the referee puts your entire discard pile into your draw pile and then draws from your draw pile. It's also worth nothing that cards in your played pile stay there until the next turn
TINOUAINANI: so I should put all discard in the draw pile before I draw two cards from it? what about if there is just one card in the draw pile do I need to get the second card from discard pile or just one that I find in the draw pile ?
MSmits: you get the one from draw, then empty the discard into the draw and then get one from there
MSmits: thats what the referee does
TINOUAINANI: ok thanks MSmits
MSmits: np
MSmits: if I download the referee and I use skeleton main, will all the bug fixes be in there?
jacek: yes they are in master
MSmits: master?
jacek: https://github.com/societe-generale/GreenCircle
MSmits: yeah i am there, but there is a test folder. I am just wondering if that is something separate that did not get the fixes
jacek: who cares about test
jacek: but it got bugfixes as well
pedrosorio: Msmits what do you mean something separate? The test folder contains unit tests
MSmits: oh ok
MSmits: skeleton main is in test
MSmits: thats what i use for running games locally
PatrickMcGinnisII: hmm, back up to about rank 250. Still haven't gotten all the possible commands tightened up
PatrickMcGinnisII: er working
PatrickMcGinnisII: i got one replay where i was starved down to 1 card
PatrickMcGinnisII: holy hell
MSmits: that takes effort
struct: Skeleton main will run the the referee
struct: I think thats how it works
MSmits: ah ok
eulerscheZahl: i've built a brutaltester version of the referee
MSmits: nice
eulerscheZahl: but my param tuning failed me
MSmits: mine gave me some improvement, but not a lot
eulerscheZahl: right, wanted to create a bug report at brutaltester. doesn't compile anymore
MSmits: oh
struct: what do you mean doesnt compile?
eulerscheZahl: wait, wrong
eulerscheZahl: new log4j version throws a runtime exception
PatrickMcGinnisII: any1 got the ref link handy?
eulerscheZahl: but still compiles
eulerscheZahl: the green box with starter kits right at the top of the statement
eulerscheZahl: PatrickMcGinnisII
struct: https://github.com/societe-generale/GreenCircle
PatrickMcGinnisII: tx
eulerscheZahl: don't undermine my educational attempts struct
struct: lol
PatrickMcGinnisII: that helped rules.bigApplicationsCount = 12;
PatrickMcGinnisII: starting to see trend in early game v. mid game, and of course endgame
PatrickMcGinnisII: hmm 1.6B ways of choosing 12 out of 28 possible
jacek: making opening book?
PatrickMcGinnisII: oh, non repeating
jacek: i think its 30421755
PatrickMcGinnisII: 30421755
PatrickMcGinnisII: beat me to it
PatrickMcGinnisII: a bunch of symmetry but the play card abilities break it up
PatrickMcGinnisII: game is becoming more interesting as you get into it, but the burden of 'entry' is high
PatrickMcGinnisII: offset by random command to keep it interesting, w/o it ... who would play...smh
eulerscheZahl: for the first few turns the application pool doesn't really matter. you have to collect boni first
Astrobytes: "boni" :D Always coming with that Latin
jacek: boni?
PatrickMcGinnisII: mia culpa es phallus
Astrobytes: correct Latin plural of bonus jacek
jacek: quidquid Latine dictum sit, altum videtur
PatrickMcGinnisII: back to Monty Python... dingus
PatrickMcGinnisII: so profound
eulerscheZahl: wow jacek, that's so deep
eulerscheZahl: how would you say instead of "boni", astro?
jacek: -That. - She
DomiKo: XD
Astrobytes: bonuses is more acceptable these days in English euler, though boni is correct Latin
eulerscheZahl: as long as boni isn't wrong in english, i'll stick to it
eulerscheZahl: 1 visum, 2 visa or 1 visa, 2 visas?
jacek: this is proper english too https://en.wikipedia.org/wiki/African_wild_ass
Astrobytes: 1 visa
PatrickMcGinnisII: no wonder technical_debt is so high, always forcing your boni in it
eulerscheZahl: but not even 2 visae? that's sad spell check already tells me i'm in the wrong here
Astrobytes: and bonuses is technically more correct than boni, since we're using English and not Latin
jacek: :rage:
MSmits: If you're using bonussen it is even more correct because it is Dutch
Astrobytes: :D
eulerscheZahl: mutemus!
MSmits: a mute mus?
Astrobytes: We'll all just speak Latin regardless of language now then
eulerscheZahl: "let's change that"
MSmits: o
eulerscheZahl: in German we say Boni btw
PatrickMcGinnisII: 5dn_l is the Muter
Astrobytes: you can tell by the mut part that it's something to do with actively changing
Astrobytes: e.g. mutation
Astrobytes: or mutate
eulerscheZahl: mutare
Astrobytes: yes
eulerscheZahl: mus = we
struct: I'm lost
Astrobytes: would you like directions in Latin?
PatrickMcGinnisII: :coffee:
struct: Never heard about Latin
eulerscheZahl: this card game gives me a hard time
struct: What's so hard about it?
jacek: :no_mouth:
Astrobytes: The boni presumably
Astrobytes: :door: I left
eulerscheZahl: the eval
struct: oh my
struct: I havent reached that part yet
eulerscheZahl: clueless
struct: Please don't scare me
eulerscheZahl: i might even do some if-else for parts of the game
eulerscheZahl: hard to make the eval work in early and late game
Illedan: Word
struct: just have an eval per turn
PatrickMcGinnisII: ^
struct: As you can see I'm an expert on writing evals
eulerscheZahl: topcoder marathon should be up at this point but i can't find it
eulerscheZahl: bad timing anyways, i'll stay on CG
struct: Usually my eval is bool hasPlayerWon(int player)
eulerscheZahl: i have a feeling that google translate failed me
Marathon Match 137 scheduled to start from 6/22 26:00
Illedan: In 6 hours
Astrobytes: 26 o'clock
eulerscheZahl: at 26 o.
eulerscheZahl: that
Nerchio: it starts today and tomorrow at the same time
Astrobytes: like your parcel euler, omnipresent and time-travelling
eulerscheZahl: oh wait, it's tuesday
Nerchio: is it?
Astrobytes: until tomorrow, yes
eulerscheZahl: the parcel didn't glitch to me yet
Astrobytes: hehehe
struct: I only know there is 5 days and 14 hours remaining, no idea what day it is
Astrobytes: :D
PatrickMcGinnisII: trying to stay in top half of silver b4 gold opens... how long until it does, ne1 know?
eulerscheZahl: monday morning for those with a healthy sleep schedule
struct: in 20 hours patrick
jacek: tomorrow
eulerscheZahl: top 50% won't be enough for gold presumably
eulerscheZahl: i predict 150 to gold
jacek: just print second action
jacek: and third action for legend
Astrobytes: lol
eulerscheZahl: if there are no 3 actions? wrap around, modulo?
struct: If everyone does 3 I probably can get rank 1
Astrobytes: 150-200 for gold, hopefully the lesser, since it's a smaller silver league
jacek: silver is too crouded
jacek: crowded even
eulerscheZahl: but i don't want a hardcore legend promotion
eulerscheZahl: my bot sucks
struct: last time it was like 2k silver
struct: or something
struct: on day 1 of silver
Astrobytes: something like that yeah
eulerscheZahl: last time we didn't have 50% stuck in wood
Astrobytes: how many currently? (can't be ar$ed to look)
eulerscheZahl: 680-ish in wood
eulerscheZahl: 750 in bronze/silver
struct: good thing I made it on day 1
Astrobytes: lol
struct: I probably would have quit
PatrickMcGinnisII: I didn't have to play cards to get in silver
PatrickMcGinnisII: just didn't have to make dumb moves
Astrobytes: nah you would have done something last minute struct
PatrickMcGinnisII: basically avoid opponent +/- 1
struct: I dont even know what my bot does yet
struct: and I dont want to know
Astrobytes: hahaha, as long as it works right
jacek: me neither
jacek: i question reality everyday since the contest started
Astrobytes: because you're still playing it?
Illedan: Give me a number to test?
Astrobytes: 42
jacek: 42069
eulerscheZahl: pi
jacek: 8008135
jacek: 2137
Illedan: Big numbers :thinking:
Illedan: Searching 42 moves ahead might break things
Astrobytes: pi x a lot
eulerscheZahl: i see what you did there. except 2137, i don't get that
Astrobytes: 2137 is lost on me
jacek: its quite polished thing
Nerchio: i submit my day1 code lets see where it ends
Illedan: hm, 42 didn't crash
Illedan: :O
eulerscheZahl: 73: 21st prime. and 37 is the 12th
struct: Do I need a syncedBeam like I did for SC21?
Astrobytes: Illedan :D
eulerscheZahl: that's something to consider. not knowing your next hand is painful though
Astrobytes: the numbers 21 and 37 mean much more: it is the time of death of the Polish Pope John Paul II
Astrobytes: ^ 2137
eulerscheZahl: the last german pope is still alive
Nerchio: its a meme in poland buts I am from Poland and I don't quite get it
Astrobytes: I don't know why it's memeworthy
jacek: then are you really a pole
Nerchio: maybe not idk would have to check birth certificate
eulerscheZahl: and investigated again, as he supposedly hepled to cover catholic activities
struct: why is my code not working if (0) ...
Illedan: Thx Astrobytes
Astrobytes: lol euler, all the other ones were just better at it
Astrobytes: np Illedan
Illedan: 42 made me realise 3 was too small. 8 was better ^^
Astrobytes: glad I could help :P
Illedan: Maybe breaking top 10 :eyes:
eulerscheZahl: catholic activities in regards to children. but let's get back to other numbers
Illedan: *100
Illedan: :(
Astrobytes: eulerscheZahl: yes, my statement still applies
eulerscheZahl: the user i reported 1 or 2 days ago got deleted btw. i think you missed it, jacek saw it
Astrobytes: I missed this I think yes
jacek: i saw many things
struct: poor jacek
Illedan: draw vs emh :tada:
Astrobytes: nice :)
Illedan: With 10
Illedan: :P
Illedan: On submit
Miki09: lol hardcoded first 4 moves pushed me up 200 places :-D
Illedan: First 4 moves?!
struct: like a book?
Miki09: go to 4, than 5 than 6 than 7
Miki09: every time I do this
Illedan: :/
Nerchio: insane strat
Illedan: Don't reveal that now
Miki09: ok
Illedan: :P
Astrobytes: lol
Nerchio: Illedan your submit is 1 position near my day1 bot
struct: what if the opponent goes to 6 Miki09?
struct: what do you do?
Miki09: nothing
Miki09: I have to work on that
Illedan: Nerchio, my day1 bot is top 50
struct: my day1 bot is silver
struct: There are many copies of it
Nerchio: i submitted day 1 to see that I am sane and going into the right direction lol
Illedan: I never look back :D
Nerchio: but it pretty much knew how to play player1 only
Nerchio: i like how codingame asks me if I want to submit without testing my code and I click yes every time, i test in production
Astrobytes: It is the CG Way
Suenodk: Haha testing in production nice :grin:
Miki09: place #193, enough for today :-)
Miki09: gn everyone
Astrobytes: gn Miki09
struct: gn
Suenodk: that's high! Gn :)
Nerchio: the amount of games played in a submit is really nice
struct: How many is it?
struct: 200?
eulerscheZahl: 2*90 i think
MSmits: man this is the creepiest thing ever. I just found out my windows connected to my work account since yesterday and it started controlling some settings
MSmits: compiled bots started getting deleted because they were untrusted exe files
Astrobytes: :scream:
MSmits: had to disconnect it
MSmits: but sheesh
Astrobytes: Yeah... keep that disconnected
MSmits: I don't know what i did to connect it
Astrobytes: do you VPN into your work system using that one?
MSmits: nope
MSmits: I do use one-drive to synchronize some files
struct: o.o
Astrobytes: and you were worried about smartphones!
MSmits: The other day they turned on one-drive on my work laptop. I had some private files on there (some cpp files and such) and they just started synchronizing the entire my documents folder, instead of letting me decide which folders (how it was before)
MSmits: lots of colleagues had their private files synchronized to the cloud
Astrobytes: d'oh!
MSmits: it's not like we were not allowed to have some private files on there, it was specifically allowed to have some limited private usage
MSmits: doesn't mean they can just force us to put it in the cloud...
Astrobytes: Yeah but it shouldn't come with consequences like that
MSmits: and now this crap.... pisses me off
MSmits: I'm thinking of going outside my school system entirely. Feels safer to me
Astrobytes: backed up? Your deleted bot exes I mean
Astrobytes: and yeah, you're not wrong
MSmits: ahh they are not important really. I can just recompile. It's just kind of annoying when my brutaltester version crashes because there is no bot
Astrobytes: ever so slightly yes
jacek: good thing they didnt delete oware database
MSmits: i have that backed up a few times :)
jacek: humanity treasure
MSmits: indeed
Astrobytes: I'm surprised you haven't memorised it by now :D
Astrobytes: Or haaaaave youuuu
MSmits: 98 GB of pure binary code... sure, here goes:.....
MSmits: 101010111010111...
struct: you are joking but today I saw a video of a guy that wanted to be chess world champion, by training just 1 month
Astrobytes: damn right! :D
struct: his plan was to memorize all chess positions
Astrobytes: wtf
jacek: plot twist: its chess960
Astrobytes: and how far did he get with that particular endeavour?
MSmits: did he understand what it meant to say "all chess positions" ?
struct: I don't think so
MSmits: btw, it's not enough to memorize all chess positions
MSmits: that's like a map with keys and no values
struct: ofc he would eval them aswell
MSmits: ahh i see
struct: his idea was to play like a computer
struct: https://www.youtube.com/watch?v=MFNv-FJFGTg
struct: Not worth the time
PatrickMcGinnisII: one of my hosts that i pay, shutdown a couple websites that i run my own encrypted text files as data storage (customer email addys mostly) ... host identified txt files as malware and they wouldn't turn the sites back on til i deleted them...I was so pissed. still pissed.
Astrobytes: oh haha, I watched part of that video years back
MSmits: thats annoying PatrickMcGinnisII
Astrobytes: as long as they're not 'text files' that decompress a load of powershell when opened and then hide in your registry to do Bad Things :P
MSmits: what a dumb guy, he could have at least used a professional chess engine
struct: I can't take it seriously
struct: This probably must be satire or something
MSmits: maybe
Astrobytes: I don't think he expected to win tbh
Astrobytes: or to get very far
eulerscheZahl: i see a youtube link. did you do your homework already and watch Hasan Minhaj?
struct: not yet euler
Astrobytes: oh the game dev thing, yes I did
Astrobytes: I had seen just over half of it apparently
Psyho: it was probably asked like 20 times already, but how do people run the games locally?
Psyho: i.e. what's the easiest way of running the tester locally
eulerscheZahl: single-threaded: you clone the referee and insert your bot into the main class
jacek: double click on .exe file
eulerscheZahl: https://github.com/societe-generale/GreenCircle/blob/master/src/test/java/SkeletonMain.java#L21
eulerscheZahl: for multi-threading there is a tool called brutaltester. but you have to adjust the referee. i can share a pre-built jar of referee+game engine if you want
MSmits: that would be great eulerscheZahl
MSmits: i dont have the multi variant, just single
eulerscheZahl: https://easyupload.io/y86u11
MSmits: thanks
struct: thanks
struct: here is the brutaltester link https://github.com/dreignier/cg-brutaltester
eulerscheZahl: execute it with something like java -jar cg-brutaltester.jar -r "java -jar GreenCircle.jar" -p1 "dotnet SocieteGenerale2022.dll" -p2 "dotnet SocieteGenerale2022.dll" -n 100 -t 8 -s
eulerscheZahl: i added the brutaltester to the upload
MSmits: 1.0.0 is latest?
eulerscheZahl: i built from source yesterday
MSmits: no i meant the link struct shared
eulerscheZahl: java -jar cg-brutaltester.jar -r "java -jar GreenCircle.jar -s" -p1 "dotnet SocieteGenerale2022.dll" -p2 "dotnet SocieteGenerale2022.dll" -n 1 to run a single game with local server
eulerscheZahl: will be available at http://localhost:8888/test.html
eulerscheZahl: i increased the timeout from 50ms to 150ms and enforce bronze+ rules. if you don't like those changes, you have to compile your own referee
Astrobytes: with blackjack, and hookers
MSmits: oh you added the brutaltester jar too, nice service
eulerscheZahl: as i mentioned ;)
eulerscheZahl: oh, maybe it was a bad idea. GPL and i didn't upload the license file
Astrobytes: shhh
eulerscheZahl: let's hope Magus won't sue me other this :D
Astrobytes: lol, I think he'll be understanding of the situation
jacek: they will fight https://www.youtube.com/watch?v=UoPqoctg21k
Psyho: late thanks!
MSmits: eulerscheZahl does your brutaltester give 1 second on first turn?
eulerscheZahl: of course
MSmits: weird, crashes for me
eulerscheZahl: you can't even disable that i think
MSmits: yeah i didnt think so either
eulerscheZahl: did you run the 2nd command i shared? with 1 game and server mode
eulerscheZahl: can so see the replay on that localhost site?
MSmits: ok will try sec
MSmits: tried: java -jar cg-brutaltester.jar -r "java -jar GreenCircle.jar -s" -p1 "1.exe" -p2 "2.exe" -n 1
MSmits: crashes
MSmits: maybe i should not do it from a bat file and use a cmd window
eulerscheZahl: "crashes" - as accurate of an error message as some of my colleagues
MSmits: yeah that worked
eulerscheZahl: yay
MSmits: i think i should have added a line to my bat file to prevent it from stopping
MSmits: https://imgur.com/a/oKlImZ6
MSmits: i was wrong, didnt work
MSmits: it gave the typical endgame result, but it does that on an error too i guess
eulerscheZahl: you can ignore that message
eulerscheZahl: http://localhost:8888/test.html
eulerscheZahl: does that show anything?
eulerscheZahl: i didn't bother to investigate about the log4j message as the rest is still working fine
MSmits: no, but i have an idea
MSmits: what i fi have my own brutaltester running on my system
MSmits: (I do)
eulerscheZahl: did you add it to your PATH variable?
MSmits: it says another process is using codingame assets
MSmits: HUD.json
MSmits: i will wait till my other process is gone
eulerscheZahl: sounds like another instance running
MSmits: yeah could be that.
eulerscheZahl: i'm off for today
Astrobytes: gn euler
eulerscheZahl: after helping the guy who's already 2-3 score points ahead of me
eulerscheZahl: i always make the same mistakes
struct: gn
struct: gn
MSmits: hehe sorry euler, i'll figure it out no worries :)
MSmits: you go do some more tuning
MSmits: or gn even
JamesWrank: Hey guys I've been stuck at first place in wood ligue 1 for ages. Not sure what to do ? ^^
Astrobytes: Get to Bronze. That might help :P
JamesWrank: Yeah okay but how ? :D
Astrobytes: Are you far from the boss points-wise?
jacek: thats in DLC, pay $$$
JamesWrank: not sure how you check that
JamesWrank: boss was at 3rd place for a while
Astrobytes: open the Leaderboard from the button on the left
jacek: what game
JamesWrank: Oh okay from that leaderboard I'm 3 points away from the boss
jacek: you need to be better than him ~
MSmits: btw... i realized something today. I thought we didn't have the full information on our opponent because we can't see his hand
JamesWrank: Mhh right I thought my script was enough as I was winning like 95% of the fights
MSmits: but we do. He has no hand
Astrobytes: Yeah. I was gonna say if it was a little bit you might be pushed but you're gonna have to improve a little
JamesWrank: Alrighty, I'll do that tomorrow :)
MSmits: there's no fog at all in this game, just random
JamesWrank: GN people
MSmits: gn
struct: The training makes it a bit harder MSmits you were right
struct: hi
MSmits: Yeah
putibuzu: hmm MSmits we still don't know what opponent played
KINGKRINKLES124: sorry if innterupting
MSmits: does it matter putibuzu?
putibuzu: yes, I think it does
MSmits: if you know the full state, what does the past matter?
MSmits: his full turn is done, you know his score, his automated, all his cards, etc.
kovi: you dont know his hand
putibuzu: Well, all I know is what deck in total the opponent has but not where his cards are
MSmits: you do
MSmits: his hand is empty
MSmits: his turn is done
MSmits: everything is discarded
MSmits: so you know everything
putibuzu: only his hand and played cards
struct: You know the cards opponent played?
MSmits: yes, so everything is discarded
putibuzu: the deck is not fully shuffled
MSmits: which means you know the full state
struct: ah i missunderstood
MSmits: what does it matter what he had in hand during his turn, it is no longer there when his turn starts
putibuzu: not sure if I understand what you mean MSmits
MSmits: let me rephrase. you know the exact location of every one if his cards
pedrosorio: Msmits, you don't know which cards are on the opponents (draw, discard) piles
MSmits: you get this as input
MSmits: OPPONENT_CARDS
MSmits: OPPONENT_AUTOMATED
pedrosorio: opponent_cards = draw+discard, no?
putibuzu: but if you knew what he had in hand, and his discard and draw pile before his turn you would be able to compute probabilities of his next hand (i.e. his next ) turn and then you could easier predict what he does
MSmits: opponent_cards is everything he has
MSmits: and since everthing is in the same place, you know where it is
pedrosorio: Msmits, draw and discard are different piles
MSmits: ohh, right your discard is not shuffled into your draw pile
putibuzu: true, but you don't know where those cards are, i.e. some are in discard, some are in draw, but where exactly you don't know
MSmits: now i see my error
MSmits: this only happens when draw is empty
Astrobytes: only when draw is empty
MSmits: yes thanks
MSmits: this cleared up a misconception for me.
putibuzu: I would love to have that additional info btw, like in a real boardgame. seeing what the opponent did
MSmits: yeah, i think it would be good to have it, this game does not need randomness and fog both
putibuzu: btw in my simulation the opponent works exactly as you described
putibuzu: take all cards and draw some at the start of turn randomly
MSmits: take all cards?
struct: all opponent cards
struct: I guess
MSmits: ohh
MSmits: yes
MSmits: i tried this but doesnt do much fo rme
struct: What about mcts but he has all cards in hand?
putibuzu: hmm I'll see if I finally manage to submit :sweat_smile:
struct: maybe he would end up playing too well
Astrobytes: i'm not optimistic about that approach tbh
Astrobytes: I could be wrong
putibuzu: yes, I'm hoping to find out :sweat_smile:
putibuzu: otherwise I don't know how to track opponent's cards properly
MSmits: if he has all cards in hand he can probably just finish any app with bonus cards and stuff
putibuzu: I mean the only things difficult are additional draws
pedrosorio: 50ms is very short, simulating the opponent may just not be that useful in this contest
MSmits: you mean 2 out of draw pile and then another 2 from discard?
MSmits: additional draws?
putibuzu: ah no TRAINING and CODING actions
MSmits: oh i got those working
putibuzu: they mess with the opponent tracking
MSmits: ah yes for opponent thats a lost cause
putibuzu: all other actions can be tracked perfectly right?
MSmits: aside from the fact that most of the time you have no idea what he has in hand
MSmits: i mean you dont even know the possibilities
putibuzu: difficult game, this is
MSmits: yeah i guess it is
CASPER_000: You have to find the sum of the first x powers of a number n starting from 0. You will find that this puzzle will be easier if you know the trick
CASPER_000: any explaintiont
struct: My code is a mess for this one
CASPER_000: any explainetion
MSmits: most of my code is reasonable except my eval stuff, that's starting to get worse and worse
MSmits: a lot of stuff with parameter = 0 i am afraid to delete
geppoz: CASPER_000 there is a math formula for that
CASPER_000: ok what is it?
TINOUAINANI: :smiley:
geppoz: the finite equivalent of the infinite one that leads to 1/(1-x)
putibuzu: the eval gets messy so easily :D one starts adding stuff here and there and suddenly there are 100 free parameters :sweat_smile:
MSmits: ye
MSmits: the last few submits, every time i had an improvement it wasnt because of a new parameter i think
MSmits: mostly because i found a bug and fixed it :P
putibuzu: during spring challenge 2021 I had a buggy arena, which always said 100 % winrate for the first bot and I was always putting my new version first, so I thought every change I made was incredible
putibuzu: then I wondered how I was not #1 :D
MSmits: heh sounds familiar
putibuzu: found out during the last hour or so :D
MSmits: ouch
putibuzu: that's why people always say tests are important I guess :D
MSmits: yeah
KINGKRINKLES124: you want to be friends
KINGKRINKLES124: :smile::sweat_smile:
MSmits: Astrobytes i figured it out. Now when you log into One-drive it has a little checkbox that asks if you want your organization to manage your computer
Astrobytes: MSmits: ahhhh that would explain it
MSmits: not sure if i checked it or if i already automatically logged in when they added the checkmark in an update (and set it to default to checked)
Astrobytes: Nonetheless nefarious.
MSmits: yeah it s
MSmits: is
PatrickMcGinnisII: i disable Onedrive in startup
PatrickMcGinnisII: so it can never sync
struct: This is why I still use windows 98
Astrobytes: 98? Noob. 95 is where it's at.
struct: I heard that one is not supported anymore
PatrickMcGinnisII: yall r whack, NT
PatrickMcGinnisII: heh
struct: :(
PatrickMcGinnisII: old argument, like 20 yrs ago
Astrobytes: NT workstation 4.0 was the greatest
Astrobytes: (seriously)
PatrickMcGinnisII: I'm hearing WIN11 noise
PatrickMcGinnisII: Win 8.1 uses the NT kernel... the kernel was the best in the day
Astrobytes: Yup.
PatrickMcGinnisII: nt came on 4 3.1" floppies, lol
PatrickMcGinnisII: 4MB OS imagine that
struct: I get like 10k moves on 2nd round, is this normal MSmits?
struct: with all possible combinations
PatrickMcGinnisII: people on YT bitching about encrypted HD issues
Astrobytes: malware-encrypted?
PatrickMcGinnisII: WIN11 bitching about bitlocker
Astrobytes: oh bitlocker
PatrickMcGinnisII: i gotta stop watching YT
PatrickMcGinnisII: and MS, they might geet mad and brick my pcs
MSmits: 10k moves ??
MSmits: struct do you mean on move selection?
struct: I just got lke 4Million on one...
MSmits: also do you allow infinite training/coding depth?
struct: no
MSmits: how much depth/
struct: 1 depth so until release
struct: from move to release
MSmits: i mean
struct: all moves, like move 0-8
MSmits: how many played cards max
struct: no limit
MSmits: ahh ok, i do 3 max currently
struct: ah
MSmits: i can do better if i optimize
struct: i guess I can try that
MSmits: but i also use some tricks
MSmits: i'll give you one
Astrobytes: you might be repeating states too with that many
MSmits: for cards that are not training or coding, make a playing order
MSmits: reset the order on draw
MSmits: so for example, you can make an order so that you are never allowed to play architecutre after review
MSmits: this prevents the situation where you have a state where you first play review and then arch and a state where you play arch, then review
struct: oh
struct: I guess that can happen in my code atm
MSmits: zobrist can do this too, but i dont like using it
Astrobytes: that's what I was getting at with repeating states
struct: ah
MSmits: just make sure to only fix orders for cards where the play order doesnt matter
MSmits: and reset the order after a draw
Astrobytes: makes sense
MSmits: yeah, you have to be real careful not to prune too much with the repeating states
MSmits: if repeating states contain random, they are not truly repeating
Astrobytes: naturally
Michael_Howard: Is Task Prioritization really rarely useful to play, or am I just doing something wrong? :thinking:
MSmits: it can be really powerful
MSmits: especially if you need 1 specific card to finish the last app
struct: well limiting draws put me from 4M to 20k
struct: or something
MSmits: remember, if you use task prio, it is still your card, it just goes back in your deck
MSmits: you only really lose the traded card
struct: yeah
Michael_Howard: I expected it would be powerful on last turn, but it never seems to be. I can't figure out why.
MSmits: should be useful in some situations on last turn
MSmits: the thing is, last turns are selections out of many possible turns that could have happened due to draw
MSmits: oftentimes you are lucky enough on a last turn that you dont need prio to make it better
MSmits: you should test it differently
PatrickMcGinnisII: I'm still trying to debug detecting a release beforre a potential release phase... yall r so far ahead
struct: https://www.codingame.com/replay/644127136
MSmits: look at earlier turns that werent a last turn and ask yourself, if i had a prio now, would i have won
struct: Here I timeout bcause I have like 19 cards on draw pile
struct: I think ill need to go back to the drawing board for this one
MSmits: training can only ever make you branch 55x max
MSmits: per played training card
MSmits: usually it is less
struct: 55?
MSmits: yes
struct: why 55?
MSmits: you take 2 cards out of 10
struct: ah ok
struct: I forgot training is not coding
MSmits: 2 the same * 10 + 9 + 8 + 7... + 2 + 1
Astrobytes: don't do that irl
MSmits: I'm being pushed hard, 4th now, but so many people submitting i guess
MSmits: so i am not actually 4th
Astrobytes: doesn't matter much at this point
Astrobytes: "high up" is enough til gold
MSmits: well it gives me confidence i will get legend, that is the most important goal for me
Astrobytes: True, depending how much more time you have to spend that is sensible
MSmits: yeah, i have a lot of time left. Not sure if my boss agrees but he doesnt have to know i put all my work off till monday
Astrobytes: I like your style :sunglasses:
MSmits: this is the first contest in like 2 yrs i played very seriously
Astrobytes: I noticed
Astrobytes: :D
kovi: the only boss you need to care is the gold one
MSmits: yeah kovi
Astrobytes: May I ask why? The probabilities thing?
MSmits: they are usually terrifying
Astrobytes: last question @MSmits
MSmits: i do like the probabilities thing, but i already decided beforehand i was going to do this seriously
Astrobytes: Nice. I'm glad about that
MSmits: It started when i had a pretty annoying cold last week and i just kept working day and night to make sure i had no work last weekend
MSmits: even though ifelt sick, so thats motivation
Astrobytes: And dedication
MSmits: yeah, i didnt get much notice on the contest
MSmits: otherwise i would have prepared sooner
Astrobytes: I had 6 days I think, I have no idea when it was put up on events page
MSmits: me neither, not long anyways
Astrobytes: was thanks to 5DN1L that I saw it
Astrobytes: woops, didn't mean to ping
MSmits: havent seen 5DN1 L
jacek: but did you see Automaton2000
Automaton2000: i could look at the leaderboard
Astrobytes: I logged in after a few weeks, literally right on time as he posted the events page link
Astrobytes: (5D, not that bot)
PatrickMcGinnisII: MSmits you bot goes straight to cont_int to automate bonus, and then skips releases to avoid debt...did you setup a debt ceiling? holy crap
MSmits: i dont know, i make eval and fit parameters. My bot figures out the rest
MSmits: i did have a debt ceiling i think, dont want to have more than 2 TD debt on release
MSmits: but i changed it to 3 or 4 and it had no effect on the bot
MSmits: eval prevents those releases
MSmits: i havent watched a game in days PatrickMcGinnisII :P
PatrickMcGinnisII: I'm using cont_int on bonus cards... it is so necessary for endgame
MSmits: yeah i guess so, if you see good bots do this
PatrickMcGinnisII: But i don't skip any releases and get debt piled on
MSmits: i see
PatrickMcGinnisII: prettu much i geet 3 releases very fast
MSmits: it would not surprise me if my bot does not do that cont_int thing if it has to give a card to get there
PatrickMcGinnisII: then last two r slow af
pedrosorio: MSmits how do you fit params for eval? Local tournaments between your bot variants?
MSmits: pedrosorio yeah nothing complicated
MSmits: i was going to do something more complicated, but my bot changes too fast and i spend most time quickly testing changes
MSmits: maybe last weekend i will do something like this
MSmits: smarter param fitting
MSmits: but i doubt i will have the time
jacek: smrter grid search?
pedrosorio: jacek simulated annealing where temperature decreases to 0 as end of contest approaches
MSmits: yeah there are many methods
Astrobytes: lol pedrosorio
jacek: GA!
antiwonto: [auto] 'GA' was defined as ' Genetic Algorithm '
Illedan: Damn play limit -.-
struct: Damn brain limit -.-
struct: Illedan are you testing localy?
Illedan: Just in IDE
struct: euler pasted here in chat referee for brutaltester
Michael_Howard: Is it better to continuous integrate a bonus or a skill card?
Astrobytes: the former in a lot of cases it seems
Illedan: Depends on the skill card vs applications left
Illedan: Early game => Bonus
PatrickMcGinnisII: no wonder i'm not detecting releasable apps, I'm not counting automated, ARGH!
Astrobytes: again?
Astrobytes: that's twice you've done that!
Illedan: Almost top 50 :tada:
Astrobytes: grats Illedan
Illedan: I had a fat big TODO in my Throw sim
Illedan: xD
struct: My bot is not normal
Illedan: Wonder where that comes from
Illedan: :smirk:
struct: ah ok
struct: Its just user that is dumb
struct: my bot is fine
jacek: he copied his NN into the bot
Astrobytes: pebcac!
struct: I was evaling only Continuous integration cards
struct: and was wondering why my bot chose to go to 3 instead of 5
Astrobytes: hehe nice
struct: I forgot you can get a card from 5 with card 3
struct: https://www.codingame.com/replay/644163191
struct: my blue bot doesnt even eval score
struct: lol
struct: and managed to win because it was the first action
Astrobytes: does the first action thing still work?
struct: no but since it was release phase
Astrobytes: ah
struct: my move gen gives release actions before wait
struct: so since the score was the same it prints the 1st
Astrobytes: I see
Illedan: Lol, emh automated 3 Code_review xD
struct: Finally I destroyed jacek
struct: in one game
struct: I can delete now
Astrobytes: eh, even I did that earlier
geppoz: I suppose never automate more than 2 of same card
Illedan: Maybe in Legend geppoz
Illedan: When you want to automate a card so that you wont use a draw on a card you don't need
Illedan: But emh is not checking that
Illedan: no way
struct: and maybe you dont want to automate if its not in apps
Illedan: True too
geppoz: there are opposite requirement :D
geppoz: automate what it is more in apps VS automate what it is not in apps
Astrobytes: "room for further improvement" ;)
struct: I just want to make little progress at a time
struct: as long as I see progress im happy
Astrobytes: worked for me since last night
struct: This is why I only submit after adding 1k lines
struct: But my todo list keeps getting bigger and bigger
Illedan: Can I have some todos? I'm out
struct: its mostly simple stuff, the one that requires the most work atm for me is the states exploding
struct: due to cards drawn
struct: and I have the problem where I can have alot of repeated states
Astrobytes: TODO: get 1st place
struct: also I need to write an eval
emh: Illedan hehe
struct: but I leave that to last days
Illedan: ^^
Illedan: New IF incomming
emh: nahh I've been working on StC all day
Astrobytes: if (not_winning) win();
Illedan: stc?!
antiwonto: [auto] 'stc' was defined as ' Smash The Code https://www.codingame.com/multiplayer/bot-programming/smash-the-code '
Astrobytes: ^ :o
Illedan: ok?
Illedan: csb?
PatrickMcGinnisII: :taco: Astrobytes
antiwonto: [auto] 'CSB' was defined as ' Coders Strikes Back (aka Mad Pod Racing) '
antiwonto: [auto] PatrickMcGinnisII has awarded Astrobytes 10 tacos. Astrobytes now has 66 taco. PatrickMcGinnisII now has 60 taco
Astrobytes: BotG
emh: even with a 24 bit lookup table it was slower than bit bfs, so I wasted the whole day
Astrobytes: Hah :P
struct: emh and I have the same problem with STC
struct: We optimize but never make a working bot
emh: true true hehe
Astrobytes: are you STILL doing that one struct?!
struct: no
Astrobytes: surely you've been at it longer than emh
KINGKRINKLES124: im doing mad pod racing
struct: I went to try and avx CSB instead
struct: i got 600k sims only
KINGKRINKLES124: what are yall doing
struct: by avxing collisions
Astrobytes: oh right, you mentioned that
struct: was expecting 1M
struct: but meh
Astrobytes: not bad though
struct: ill go back to it later once contest ends
Astrobytes: that smitsi?
struct: no, just random search
emh: my last contest submit was 1 day ago
struct: but I didnt submit since i want to try smits
Astrobytes: 600k ain't bad
KINGKRINKLES124: are you doing clash of code
struct: emh come back to the contest
Astrobytes: No KINGKRINKLES124
struct: STC can wait
emh: I don't have a lot of ideas for new ifs
KINGKRINKLES124: didnt mean to make mad
struct: What about less ifs?
emh: you mean CMPXCHG?
Astrobytes: Or old ifs :sunglasses:
KINGKRINKLES124: :disappointed:
pedrosorio: Astrobytes what is smitsi?
struct: Smitsimax
Astrobytes: aka UCT Forest
struct: https://www.codingame.com/playgrounds/36476
MSmits: on my profile. Some ucb1 based search I came up with for multi agent games with not too many agents that weakly interact. Yeah uct forest
MSmits: i do not use it in this game
MSmits: just to be clear
Illedan: Whaaat?!
struct: ok i know what to use now
MSmits: i don't lie :P
struct: me neither, I just dont always tell the truth
Astrobytes: *you are economical with the truth
Astrobytes: truth when necessary, troll otherwise
MSmits: lol
Astrobytes: a true art form imo
struct: MSmits I think I understood what you meant on the order of actions to not repeat
struct: So lets say I play action 2[daily routine]
struct: Ok its a bit messy to explain
struct: let me see if I can come up with an example
MSmits: yeah it is difficult to explain
struct: ok if I play 4[architecture]study, in theory I already tried all actions before it
struct: so when I do move gen I only do after 4 right?
struct: something like this
struct: 4 being architecture
struct: and before being task and daily
struct: as long as they dont draw a card
MSmits: yeah
struct: thanks
Astrobytes: wahey, one win vs emh :P
MSmits: i did this by passing a parameter "int order" down the dfs recursion, i only allow moves of the same order or higher. The order resets to 0 on draw
emh: hehe
struct: oh that makes it easier I guess
struct: instead of doing it on move gen
struct: smart
MSmits: i did this before on locam i think. To much greater extent. You can look at my PM
MSmits: it's a long time ago though
struct: your PM is ancient by now
Astrobytes: struct think of it like using moves from a transposition table or iterative deepening ab
struct: if it doesn't include the word NN
MSmits: it doesn't :)
MSmits: yeah, i came up with this because i didn't know how to hash stuff :)
Astrobytes: ie. you know the previous moves from before so just check what's available at this current depth
struct: I see astro, I think I got it
struct: oh
Astrobytes: lol MSmits
struct: Im generating moves everytime
struct: every state
Astrobytes: I do love your inventiveness
struct: maybe thats not optimal
MSmits: yeah why learn something new that's easy when you can invent something yourself that's much more complicated
Astrobytes: it's the scientific way! :D
Astrobytes: at least you don't have to argue about grants, funding and ownership too :P
Astrobytes: *Additional funding
MSmits: yeah, I'm glad I have never had to do that
MSmits: got into teaching quckly enough to avoid
Astrobytes: likewise, job came up before research opportunities :D
ruessin: does this website have mock assessments with 2-4 problems?
struct: Sounds like certificates
ruessin: i guess algorithmic certificates, if they exist. not looking to test knowledge of different languages
struct: Then I guess the answer is no, but there are plenty of puzzles I guess
geppoz: question: you have to output every single action?
geppoz: I mean if i already have MOVE+GIVE
geppoz: can I output 2 rows in one turn?
struct: bad idea
geppoz: or I need to wait new inputs?
struct: wait for new inputs
PiterYeh: nope, it's one action per output
geppoz: (experimentally seems not)
geppoz: ok
struct: in theory it works
struct: but you will still get the inputs
geppoz: no, it doesnt
struct: So it will go out of sync
Astrobytes: Output for One Game Turn 1 line containing one of the following actions depending on the game phase:
geppoz: Line 1: gamePhase the name of the current game phase. It can be MOVE, RELEASE
geppoz: :D
geppoz: where are THROW_CARD, GIVE_CARD?
Astrobytes: they failed at that part
Crosility: Howdy! :D
struct: hi
Crosility: How's things going? Everyone still focused on the contest?
Nerchio: no i am here to read the chat
Crosility: (Seriously? Or sarcastically?)
Psyho: It turned out the less I submit the higher I'm on the ranking so I stopped submitting
Nerchio: I like the chat but not that much let's just say that
struct: lol
struct: Crosility I think most are focused on the contest
Astrobytes: RAIC is happening
Crosility: Haha, okay Nerchio. I agree struct. That's what I was thinking. The contest isn't bad or anything. Just not my thing. Card games. And a "I don't know what to do - random() out" combined - made it so I quit submitting at bronze.
Astrobytes: 7th July start
Crosility: RAIC? (Royal Architectural Institute of Canada?)
Nerchio: Astrobytes i saw that, might participate
Astrobytes: No, hahaha
Nerchio: depends on the game
Astrobytes: Nerchio: yeah same
Astrobytes: Crosility: Russian AI Cup
Astrobytes: https://cups.online/en/contests/coderoyale
Crosility: Thanks, my google results weren't giving me any context. Haha!
Astrobytes: lol, the acronyms never end
struct: Duck as been on rank 1 for a long time
Astrobytes: very Duck-type contest, showed up first day too
struct: Maybe I should submit
Astrobytes: #1 or quit
struct: I woudln't even start :p
Crosility: Haha.
PatrickMcGinnisII: an app releasr can use shoddy skills from a card that isn't required?
struct: yes
MSmits: yeah, but i dont think you should ever do that
struct: ever?
MSmits: i mean it means over half your skills are shoddy
MSmits: > 4
struct: ah right
PatrickMcGinnisII: ok, that is why my prog is calculating the debt wrong
geppoz: well you cannot know from which hand card the release takes his shoddy
MSmits: thats also very bad
MSmits: actually no it isnt
MSmits: it's all discarded anyway
geppoz: you can only know how many shoddy are needed, then it takes it from any
struct: Is it worth ever to use cards if they wont give you anything?
MSmits: yeah it doesnt matter, the only thing that matters is how many shoddy you are using
geppoz: exactly
MSmits: struct nah, just the 2 skill type cards + bonus imho
struct: Imagine using refactoring but you have no Tech debt in hand?
MSmits: it does nothing
struct: what is that?
struct: 2 kill type cards?
struct: skill*
MSmits: for the app
MSmits: i misunderstood you
MSmits: about the refactor, no there is no reason to play it for nothing
struct: I see
PatrickMcGinnisII: I prioritize releases...if it can't be released it's still in the sort...so a debt is staying at say 3, when it could be played with 5. When phase happens, its still at 3, dumb me, thought u had to have at least 1 card in hand of each type that is needed
struct: I do the same for multiple
MSmits: it's gonna go in your played and will be discarded at the end
struct: I dont play training if I cant draw any card
struct: It makes no difference I guess
MSmits: good, it;s useless
struct: Since they are all discarded
MSmits: ye
struct: Lots of small things I need to work on
struct: Closing chat, ping me if its neded
Nerchio: i hate losing to the bots that end up with 20 debt at the end of the game xd
MSmits: going to bed
MSmits: ignore me if i snore
Astrobytes: lol gn
PatrickMcGinnisII: :head_bandage:
Crosility: Sigh, when compiler thinks you're doing one thing, but you're doing another. So it throws an IDE red squiggly error. But it's perfectly legal. <.<
Crosility: And compiles, and functions flawlessly.
Astrobytes: c#?
Crosility: Basic C.
Crosility: clang is an annoying "error checker".
Crosility: http://chat.codingame.com/pastebin/3524868c-bff0-4a89-8348-28df118920c6
Crosility: If curious
Astrobytes: ahhh I thought you meant the online IDE
Astrobytes: (CG IDE)
Crosility: No, it is. It's what they use here.
Crosility: They use Visual Studio with clang instead of GCC. for C.
Crosility: Fields must have a constant size: 'variable length array in structure' extension will never be supportedclang(typecheck_field_variable_size)
Crosility: All I'm doing is using a previous declared variable, to determine size of array in a structure. Nothing dynamic. Lol.
Astrobytes: If you can keep reproducing that you should make a report or pull request on the Monaco repo
Astrobytes: or the clang, or both
Crosility: I believe, from what I say, it's been fixed for a couple years now.
Astrobytes: huh
Crosility: I am under the suspicion CodinGame is just using a outdated version.
Astrobytes: It was updated fairly recently since they made and had accepted a couple of PRs to the Monaco repo
Astrobytes: *fairly*
struct: so you think they are using msvc for intellisene?
Crosility: https://www.codingame.com/playgrounds/40701/help-center/languages-versions (sec, actually verifying against their wiki)
Astrobytes: VSCode, ie. Monaco
Astrobytes: and yeah, gcc not clang
Crosility: Then .. interesting I get clang-based error reporting.
Crosility: I'm seeing that too.
Crosility: As it's not been an issue for GCC.
Astrobytes: Peculiar
Crosility: If you wanna reproduce. it's happening at as low as a level as this :
Crosility: int w = 5;
struct TEST { char line[w]; } test[5];
Crosility: struct TEST throws a IDE error/warning for "line", due to variable array size.
Crosility: Though, it's definitely defined before test[] is created.
struct: #define w 5
struct TEST { char line[w]; } test[5];
struct: This was your error
struct: error: variably modified ‘line’ at file scope
struct: If you want to look it up
Crosility: For future reference, where did you get that error? (Since I'm compiling fine)
struct: I ran it locally
struct: the c++ and C error messages are broken on cg
Astrobytes: and it's gcc
Astrobytes: #ifdef __GNUC__ //GCC printf("GCC\n");
- endif
struct: but the messages also appears on cg
struct: when you compile
Astrobytes: the only alternative to #define is enum I think (good catch btw struct)
Crosility: Interesting. It's weird, because functionally, it works. But it's throwing a "code-sense" error/IDE, not any compiler errors. But, looking up what you said, yes, I'm using a variable to define size of array in a struct.
struct: I was getting compile error
struct: both locally and on cg ide
Crosility: That's weird you are - but I'm not?
struct: or are you use c++?
Crosility: Nope, just C.
struct: on C it doesnt compile for me
struct: int w = 5; struct TEST {
char line[w];
} test[5];
struct: this code just to be sure
struct: we are on the same page
Crosility: Perfect.
Nerchio: nice submit MSmits you trashed my bot :sob:
Astrobytes: yeah, doesn't compile my end either
struct: Nerchio it's time for revenge
Nerchio: Still glad that I managed to break into top20 again, time to sleep
Nerchio: cya tomorrow
Astrobytes: aha he didn't really go to bed
struct: gn
Astrobytes: gn Nerchio
struct: Maybe he got up, I sure did yesterday to write stuff on paper
Astrobytes: I did the same, but it was to remember to buy ginger and garlic today that I wrote down :P
Crosility: http://chat.codingame.com/pastebin/9682a238-1abd-4356-aa6a-375a83df5049
Crosility: These pastebins continue to fail.
Crosility: int w = 5;
struct TEST { char line[w]; } test[5];
Crosility: test[0].line[0] = 'a';
fprintf(stderr, "%c\n", test[0].line[0]);
Crosility: Standard Output Stream: http://chat.codingame.com/pastebin/e21963b5-3bc3-4ace-b354-9ef8badb2ab9
struct: pastebin is broken
Crosility: Output > a
struct: locally or on cg?
Crosility: CG, bout to run locally.
Crosility: So, I use code::blocks with GCC. Don't hate. But, no error.. either.
struct: paste the full code on pastebin
Crosility: Copy+paste code, new project.
Crosility: https://pastebin.com/gDrBS5zb
struct: oh
struct: I thought all the time that the struct was outside main
Crosility: No, sorry. Because then, yes, that would be outside of scope declaration.
Astrobytes: I still get an error
struct: not if the struct is inside the main astro
Astrobytes: got it
struct: Anyways I don't know why intelisense doesnt like it
Crosility: So, now on the same page. See what I mean?
struct: Yes
Astrobytes: yeah, Monaco bug
Crosility: Weird. Well, now.. is the editor color-changing in Codingame broken?
Crosility: (Tried setting the squiggly line color coding to something less.. well, eye-sore inducing)
struct: no idea tbh, i dont really use the cg ide
Astrobytes: never actually tried tbh, I only use it for short puzzles or param tweaks
struct: I run locally and use CG local
Crosility: CG local?...
Astrobytes: IDE-browser bridge
struct: https://www.codingame.com/forum/t/cg-local/10359
Crosility: Thanks struct, lmao. I wonder if IDE's other than Visual Studio is supported.
jbiz: anyone else think comparing code length is a bit unfair on clash of code, considering it's comparing different languages?
struct: you can use any ide Crosility you cant run anything locally
struct: it just so you dont have to copy paste the code
Astrobytes: it just uses the file that you specify
Astrobytes: and you can let it update based on changes or based on your consent
pedrosorio: jbiz you can use any language you want, use short languages for shortest challenges
Astrobytes: ^ this all day
Astrobytes: Right tool for the right job
struct: Master ruby and you will never lose a shortest clash again
struct: unless someone uses bash or perl
struct: But mastering bash and perl is much harder
jbiz: yeah but I prefer to write in c#. I mean I can, and have written in python, but I have to look stuff up because I don't use it that often. C# however just generally has longer syntax.
Crosility: Aw, perfect. Once I progress into some harder puzzles. I'll be 100% looking to start doing it off-site. @jbiz, totally agree. I'm a C main. It's not even a competition. I'm the loser if I do size comparisons.
Astrobytes: I wouldn't say perl is harder, stranger maybe
struct: Its one of the goals jbiz to make you look up stuff
pedrosorio: jbiz learn a concise language or accept you're going to lose on the shortest challenges, that's the way it goes
struct: you can use this cheat sheet I guess
struct: https://github.com/siman-man/ruby-golf-style-guide
struct: Anyways its time for me to go back to coding
Crosility: This CGLocal is awesome, reading the page. I can just dump c files (or whatever file) I tested locally to the game.
Astrobytes: yep
Astrobytes: I use that or just copy paste from VS/whatever I'm using
Crosility: (Non compiled, obviously), but hey, this is way better than continuously finding/editing Visual Studio and browser tweaks.
Crosility: --> Just so you know, in a few hours, I've added like 3 bookmarks from you guys giving incredible help. Thank you.
Astrobytes: you can send compiled too but don't do that in a contest, whatever you do
Astrobytes: yeah, chat's real useful. Shame it's going discord only
Crosility: Discord isn't bad-per say. Just this "small chat-right next to the thing you're doing" is very helpful.
Astrobytes: Yes, easier to give direct help in context of the issue etc
Astrobytes: And nice for shooting the sht / talking random stuff
Astrobytes: lots of peeps can't/won't use discord though, so we're gonna lose a lot of good people
Crosility: I wish I enjoyed this site as much as I do now, 4-5 years ago, when I initially joined.
Astrobytes: yeah, I joined in 2015, didn't realise there were contests until 2018 lol
Crosility: Haha. I don't remember anything about the site. But I KNOW it wasn't as user/beginner friendly before.
Crosility: The slow-roll in quest system. To sorta "reintroduce"/set pacing.
Astrobytes: Yeah it was a right old mess way back, before the quest map it was not the best navigation-wise but better-than-before, quest map and a few other tweaks seem to work for most people
elderlybeginner: What's your general strategy for Green Circle?
Crosility: Watch other people talk about it :joy:
Astrobytes: Right now I can't say I have anything specific tbh elderlybeginner
struct: elderlybeginner some people bruteforce all combinations
struct: some people have heuristics
Crosility: Well, again, it may have not been the solution I was looking for/expecting. But it's a million times better and will probably implement tomorrow. (The external IDE). Thank you again. Off for now. :]
Astrobytes: Nice one Crosility :) See ya later!
struct: cya
Astrobytes: elderlybeginner: release as many apps as you can, don't get too much tech debt otherwise
Astrobytes: about as general as I can get
Astrobytes: currently my bot is doing the equivalent of automating technical debt :rofl:
struct: Im going now for my V4
elderlybeginner: I have a few ifs right now and I wonder if I even want to continue that contest
Astrobytes: submit time?
Astrobytes: elderlybeginner: go as far as the ifs will take you
Astrobytes: it's kind of a love/hate contest in terms of if you like it or not I guess
elderlybeginner: yep, certainly not in my type
struct: its hard at the beggining but it gets better
struct: you just need to understand the rules
KingWhitfield2011: can somebody help me with how to play green circle
struct: It's a very hard game KINGKRINKLES124
KingWhitfield2011: i dont really know how to play it
Astrobytes: there are easier games/puzzles
KingWhitfield2011: OK.thanks for your help
struct: its not very beginner friendly imo
struct: If you dont have experience in other multis you will struggle a lot here
Astrobytes: Yeah, definitely not
Westicles: don't listen to them KingWhitfield2011, they just don't want you to win their phone
elderlybeginner: the rules are overwhelming and that's what's opposite to great contest. I prefer simple rules and deep strategy
KingWhitfield2011: you are funny elderlybeginner
KingWhitfield2011: and Westicles
Astrobytes: elderlybeginner: since it's kind of an 'unofficial' sponsored contest I feel that's okay
Astrobytes: But the statement is fairly horrifying
PatrickMcGinnisII: lost 70 ranks on the last sumbit, competition getting stiffer
PatrickMcGinnisII: my full move sim is bugged
Astrobytes: PatrickMcGinnisII: getting stiffer? That's what she said
PatrickMcGinnisII: back to latin IC. ;)
Astrobytes: :D
KingWhitfield2011: :laughing::thumbsup:
Astrobytes: anyway, my continued presence in the coding environment is doing me no favours, keep getting to 128-138 or miles below with every change. more bugfixing tomorrow. gn all
PatrickMcGinnisII: nite
bobth: Have You Guys Heard Of The Hit Game Among Us?
PatrickMcGinnisII: my 15 yr old used to play it
KingWhitfield2011: i hate the people who get you out even when they did it
Westicles: hard to tell the game name with all caps
bobth: lmao
Westicles: My new puzzle doesn't have any 1-stars, very unusual. Maybe wait another day
Westicles: Maybe the power is out in france or something
KingWhitfield2011: you r in france
j4at: Lets make among us like game in codingame :o
AbrahamJLR: Can someone guide me to transform an exercise that recently came out in clashofcode and I managed to finish in an inefficient way in my opinion?
I'm trying to avoid using the while loop and get the result by a formula, but I haven't been able to figure it out yet.
KingWhitfield2011: group project
bobth: j4at yooo ill make a CoC
j4at: nice, I'm talking about bot programming tho. 8 players among us game. That will be very sick.
bobth: oh oh
bobth: I'll still make one anyways
j4at: good luck :)
bobth: rhanks
bobth: t
KingWhitfield2011: can you do team projects'
Salyx: hello
Salyx: hi Coco
struct: hi
jbiz: is there any way to choose which type of clash of code game you play? Ie. Shortest
AbrahamJLR: Just in private rooms.
Glyptod0n: moving to desk 0 while opponent is on desk 7 count as "on a desk next to the opponent" right
jbiz: so, i got to the milestone "reach top 1000 players"...
jbiz: umm, is that even possible?
jbiz: looks like there's a billion players on here
jbiz: how long's it gonna take me? a million years?
jbiz: i can't access the other milestones without doing this one :/
Westicles: clash rank decays if you stop playing. you are probably competing with 4k people and 7/8 of them are 11 year-olds
jbiz: is it the rank under "crafter"? if so i'm 13,215th
Westicles: that's contest rank, you are talking about the clash achievement?
dickhead3: can someone teach me how to code
antiwonto: [auto] hey dickhead3 looking for some intro tutorials on programming? This isn't the place probably. Try codecademy.com , 'The Coding Train' on youtube, or first levels in codewars.com