thibpat: I'm streaming my progress on Ocean Of Code on https://twitch.tv/thibpat
yfprojects: @AntiSquid I meant this one: https://www.codingame.com/ide/demo/758100a939c77b58a4a37fc56ab749a6cbced5
Quidome: Huh, does pacman already exists?
Quidome: in codingame?
Uljahn: does Automaton2000 already exist on CG?
Automaton2000: i don't see it on the forum
AntiSquid: oh hi yfprojects
chucknorris: https://www.codingame.com/replay/446738484
**yfprojects greets AntiSquid
mark7: Hello!
AntiSquid: what is the point of myID variable in the input?
mark7: If you are first or second
mark7: But it doesnt matter much
AntiSquid: you get NA first turn as opponent message if you are first
Ifthel: Grasping at straws, but it lets you plan around being first or second when you pick your starting position.
AntiSquid: ok, now to figure out when that matters :thinking:
wlesavo: AntiSquid i use it to alternate fixed seed in self play
Acem1: (╯°□°）╯︵ ┻━┻
Acem1: /me
**Acem1 slaps around a bit with a large fishbot
JFB: wlesavo - good point
chucknorris: can you please stop recking me pb 4:(
pb4: chucknorris let's make a deal. Get jolindien to stop wrecking me, and i'll stop for you :)
chucknorris: sounds like a fair deal :D
MJSS: Chucknorris making a deal? WHAT?
2571540: I want a deal too
AntiSquid: ya chucknorris should wreck them all
AntiSquid: by 10 points difference ofc
wlesavo: lol now thats a tactic https://www.codingame.com/replay/446751711
pb4: what frame ?
Nagatwin: im doing the same wlesavo
Nagatwin: the end pb4
pb4: oh wlesavo's surface to go back ?
wlesavo: pb4 i mean surronding by mines
pb4: haha
pb4: I was thinking the surface made perfect sense
pb4: From the torpedo point of view
pb4: if you go forward, you get hit by a torpedo for 2 HP
Battat: hey, does someone know if the "force-field" of the pods is determined by the speed/thrust also?
wlesavo: i would like to go back but i have a mine in my knee
Nerchio: I can do move -> silence -> shoot in 1 turn?
Nagatwin: yes
pb4: Battat : ""force field"" increases mass by 10x
Nagatwin: Everytime I open the IDE a different date shows up for the legend leage, I have to refresh cache to get the right date :/
[CG]Thibaud: weird
wlesavo: not every time but sometimes happens to me too
Noobmaster69: Hello world
aCat: Hello Noobmaster69
amurushkin: is it legal to create a company team like "RUCHAT"?
MadKnight: #ruchat
MadKnight: #amurushkin
Nerchio: is there any way to simulate X games against someone and see the results
MadKnight: simulate ?
MadKnight: how much is X ?
Nerchio: idk 5?
Nerchio: just want to see how I am doing against AI in like 10 games instead of waiting for 10 to finish
MadKnight: u can download AI binary and run as many games as u want in local
MadKnight: if they give u their binary
MadKnight: u will even be able to run 1000 games in parallel on GPU
MadKnight: CG servers are really busy playing other players games they can't run lots of games for u that quickly
MadKnight: Nerchio in what language do u code ?
Nerchio: java
MadKnight: oh, u will probably have to rewrite in C to run on GPU
Nerchio: it's fine i will do without it i guess :D
MadKnight: are u sure? u can test like 1000 games in seconds
MadKnight: u can make multiple versions of your own bot to fight against each other
Nerchio: yea it's my first competition I don't need to go that far
aCat: No you definitely don't
MadKnight: will u go that far on your second competition ?
Nerchio: maybe, we will see :)
MadKnight: what's your current approach for this game ?
MadKnight: what does your bot do ?
Nerchio: it survives more than 10 turns that's a success
MadKnight: that's nice
MadKnight: u wanna try an easier competition first ?
Nerchio: don't worry I am in silver soon it's not that bad ^^
MadKnight: nice
MadKnight: hey aCat and how's your bot ?
aCat: somewhat silver python approach
Nerchio: my bot semi-well tracks the enemy and shoots =/ how sophisticated
aCat: and I'm rewriting everyting and making new one in C#
aCat: as it grows bigger and I need some proper ordering in code
Piemert: 900 lines of javascript mess here :x
Nerchio: what league are you in?
Piemert: gold, rank 92
Nerchio: that's not bad in my eyes i am in bronze :D
Nerchio: and ranking up to silver has been a pain the last 2 days
Piemert: my script is probably not the most elegant, but I can't see anyone get into gold with just a few lines of code ;)
Nerchio: i have 650 in java so far
Floryan: im at almost 800 in java
Piemert: I'm not sure why I didn't pick typescript.
AntiSquid: more like depends on the user, of course you should have picked tpyescript if that's what you use at work
AntiSquid: 800 lines c++ btw :p
kovi: wow, huge lead pb4
pb4: I think it's a lucky streak
pb4: repushing the same AI to check
pb4: but I did like seeing this :D
Nerchio: so you can shoot around the corners but not through islands yeah?
pb4: Yes
pb4: but the blast can go through islands
Nerchio: any easy way to check line of sight that can't go through islands? :D
2571540: Do a bfs from your position
wlesavo: pb4 do you consider a space control aspects already?
pb4: Nope, and I won't
wlesavo: ok, just thinking about it, but probably wont either
pb4: space control involves long term planning, which I have no way to introduce. I'm current only planning on the very short term
wlesavo: same
wlesavo: wonder if anyone will get to it
2571540: I tried a bit
2571540: But short term does most of the job on top10
wlesavo: at least for now
Nerchio: it's so satisfying to see your bot do the job :D
Nerchio: i got promoted to silver hell yeah :D
Riyuk: gg Nerchio!
chucknorris: I'm sure long term area control planning will be a thing in the final top10
yfprojects: chucknorris what do you use to keep you path this compact?
chucknorris: bitmap
yfprojects: thats very expressive!
Nagatwin: "this" compact ?
yfprojects: why?
yfprojects: his path seems very compact in comparison to other top players
chucknorris: ahh okay. I'm not trying to keep it compatc btw. I try to maximize my available space, the count of locations where the enemy will think I am and few other things
AntiSquid: just got a letter with Boris' signature to stay at home ... wonder how many millions they spent to send these letters out to everyone
infinitetundra: Boris', as in life of boris?
Nagatwin: Zoning might be a good strat actually
Nagatwin: https://www.codingame.com/share-replay/446807667
Nagatwin: I guess it does not work that much on low gold
AntiSquid: as in "belonging to", left the s out since it's s's at the end otherwise
yfprojects: In the FAQs is says that my program is run on a multi-core architecture.
yfprojects: So am I able to use these cores?
Nagatwin: Never tried using multi_core programs but I think you can try
Nagatwin: my simu has no respect of others when it can win https://www.codingame.com/share-replay/446809829 (last 5 frames)
AntiSquid: ready for approval if anyone is interested:
https://www.codingame.com/contribute/view/4950a2e3becd8c81885ea8cb112d3b43bbf7
wlesavo: contest is out lol
Nagatwin: AntiSquid what's that ?
Nagatwin: next contest ?
AntiSquid: no, just old multi
AntiSquid: never got published
Nagatwin: oh
Nagatwin: isn't next contest Pacman too ?
wlesavo: i guess we can train on this one :slight_smile:
wlesavo: perfect timing btw
AntiSquid: it's not going to be this game
Nagatwin: Yeah I hope so would be unfair :p
aCat: Looks great!
aCat: I would like to see source later.
Is there only one level?
AntiSquid: i will upload referee later
aCat: oh, there is download button :D
AntiSquid: one level
AntiSquid: will add more levels later if people request it i guess
aCat: but sorry for now - ooc-focused ;]
wlesavo: AntiSquid did you deleted it?
AntiSquid: Thibaud said he doesn't want it shared
AntiSquid: first he said sure, so i thought it's ok, but i guess they might want to release it officially for pacman celebration?
AntiSquid: along with the real contest maybe?
wlesavo: oh, i see
Nagatwin: :'(
wlesavo: too bad i was not fast enough to see the grapichs
darkhorse64: I was, hehe
blasterpoard: wlesavo https://i.imgur.com/JMNygWd.png
wlesavo: wow, thx
[CG]Thibaud: sorry guys :(
AntiSquid: lol i got too eager to share it also
darkhorse64: IP problem or is it just temporary ? Not that we are starved atm
[CG]Thibaud: we've got an issue. We're looking into it
andrei00001111: I have a problem. I've received a mesage that said I was promoted to the Bronze League, but I didn't actually get promoted. Any ideaa how to resolve this?
joeltonyf: Is the codingame multiplayer working?
andrei00001111: Oh, the gaame is Coders Strike Back.I have no idea is thge multiplayer is working. I've found this site 2 or 3 days ago
Zac__likes_fortnite: (╯°□°）╯︵ ┻━┻
mat_cloud: :pig:
Nagatwin: Too bad AntiSquid I was up for that training :D
Nerchio: if somebody saw my code i feel like they should ban me from coding for the rest of my life
MadKnight: don't worry Automaton2000 seen worse codes
Automaton2000: i watched a lot of draws
MadKnight: if u get banned from coding how will u improve your code Nerchio ?
Nerchio: I guess I won't it will forever be a silver bot
MadKnight: u wanna try another contest ?
MadKnight: an easier one
MadKnight: https://www.codingame.com/ide/puzzle/coders-strike-back
MadKnight: practice is practice, Automaton2000
Automaton2000: why are u not playing csb
MadKnight: exactly Automaton2000
Automaton2000: you need to play against me
MadKnight: then write a bot Automaton2000
Automaton2000: wonder if i can figure out how to write a ga
Nagatwin: Hahhaaa
Nagatwin: Mk and csb is some kind of a love story
MadKnight: Nagatwin teach Automaton2000 to write a GA so Nerchio could play vs him
Automaton2000: that makes it more complicated
MadKnight: Nagatwin u got another easier contest for a guy ?
Nagatwin: Csb is not that easy
AntiSquid: omg CSB spam again
MadKnight: just use -3vel and u're golkd
Nagatwin: :D
MadKnight: AntiSquid join us
AntiSquid: there are far easier games on CG
MadKnight: Nagatwin maybe u just don't like vectors ?
Nagatwin: haven't found that 3vel yet stuck in silver xD
MadKnight: AntiSquid your dota isn't easier
AntiSquid: nerchio better play code a la mode it's more fun
Nagatwin: dota ? :o
AntiSquid: it's more fun than CSB MadCarpet_7f43
MadKnight: yea there is dota on CG
AntiSquid: MadKnight
Nagatwin: MadCarpet :D
MadKnight: don't be a mad carpet Automaton2000
Automaton2000: you can do it as a win
Nagatwin: gota find this one
Nagatwin: dota find this one*
AntiSquid: Nagatwin https://www.codingame.com/multiplayer/bot-programming/botters-of-the-galaxy
Nagatwin: looks cool
Nagatwin: gotta try someday
Nagatwin: still lacks jungle camps
MadKnight: fix your dota AntiSquid
AntiSquid: it has camps Nagatwin
AntiSquid: they spawn later during the game
AntiSquid: see the (brown thingies) groots at the top
MadKnight: can u camp a camp spawner Automaton2000 ?
Automaton2000: trying to make my bot a little bit :p
AntiSquid: and the top down view wasn't our idea, we just had to go with it nagster
AntiSquid: lol i keep pinging the wrong people, sorry folks
Nagatwin: I'll try this one mb bewteen pacman and ooc
Nagatwin: what are the PM strats like ?
AntiSquid: no trees, just bushes
Nagatwin: bush search algorithms ,
Nagatwin: ?
Nagatwin: next level data structure
AntiSquid: heuristics i think
AntiSquid: there are some sims
Nagatwin: I kinda like heuristics
AntiSquid: do you mean in regards to gameplay or algorithms?
MadKnight: yea most are heur
Nagatwin: Both if you know :D
AntiSquid: because there are lots of different weird tactics when it comes to gameplay
AntiSquid: don't think it's that well explored yet
Nagatwin: Is this contest old ?
AntiSquid: 2017 ? not sure anymore
AntiSquid: or 2018 ?
Nerchio: should i refactor my 800 lines of code it will take like 1 hour at least or just leave my bot in silver forever :D
AntiSquid: Nerchio i have 800 lines of ugly C++, stop worrying so much about it and rewrite if you have to, i am rewriting right now since i really need cleaner code and different structure
Nagatwin: I had to rewrite 1500 lines for this context to revert the way my bottom structure was working :'(
Nagatwin: Was worth it, learned a lot
elsinnombre2d2: what a codes xD
elsinnombre2d2: i only have 500 lines
Nagatwin: I like verbosity and having pretty debug functions
Nagatwin: probably 30% of it is mostly useless
elsinnombre2d2: that explain it
Nagatwin: But saves time for debug
Nagatwin: http://chat.codingame.com/pastebin/ffc052b7-9872-4788-bbfc-4a4c105a7f27
elsinnombre2d2: is there any way of watch the best codes when the competition ends?
eulerscheZahl: no. just read text descriptions
Nagatwin: No (un?)fortunately the code others do it not shared
elsinnombre2d2: sad
Nagatwin: you only have acces to the posts some publish describing their strat
eulerscheZahl: example for the previous contest: https://www.codingame.com/forum/t/unleash-the-geek-feedback-strategies/133878
Nagatwin: I still wish there would be more code sharing / reviewing on this website
Nagatwin: It's easy to discover algorithms & stuff here but hard to have some feedback to improve
eulerscheZahl: the game still continues. no fixed deadline with glory and possible prizes to win. when someone shares a top bot. many will read it and probably learn something from it
eulerscheZahl: but a few just copy-paste it into the arena, ruining the fun for those who still want to play the game
eulerscheZahl: the 3rd placed hypersonic contest bot is on github. with the consequence of having almost 100 bots saying "SAYONARA" when they die
Nagatwin: eulerscheZahl do you mind giving the link in pm ?
AntiSquid: kinda considering to share my data structure for OOC though, i wonder if there's a simpler way to build this thing
Nagatwin: Antisquid I'm up to this kind of things too. I did 4 different structure, each one having different up/downsides
eulerscheZahl: i changed my representation too
Path from List<Cell> to bool[]
AntiSquid: ^ ugh i wonder how you manage to cope with just that ... would have tought vector<Cell> would be best
eulerscheZahl: bool[] is faster to check if a cell is already visited
eulerscheZahl: and an extra var to store the current position (which used to be the last entry of the list)
Nagatwin: arrays save class allocation time + make garbage collector's job easier I guess
Nerchio: i use boolean[][] and int[][] but i am beginner
eulerscheZahl: and for C# a 1D array is significantly faster than 2D (at least on Mono, which CG still uses)
eulerscheZahl: for .net core I didn't measure a difference iirc
Quidome: I think the rating system is somewhat broken. When you resubmit you drop 30 places (after initial submit matches), 12 hours later you climb gradually 30 places. This is really strange
Quidome: (with teh same version)
Nerchio: I thought ever submit drops you to the end of your league :D
Quidome: yeah that is true, but after the recalc you shot roughly be at the same spot
Quidome: and that is not the case
Quidome: And i suspect it is only happening when you somewhere at the end of the list
Quidome: at least in the last 50% part
AntiSquid: yes there are better ways to sort out ranking
Nagatwin: 2 submits may have bad/good luck and vary a lot
Nerchio: so if i get to 100% of battles i won't get recalculated anymore?
AntiSquid: just hope the last submit is the best :p
Quidome: Yes, but it is always this same pattern for me. Submit, wait 12 hours and then i am back
eulerscheZahl: i observe such behavior when there's something fundamentally wrong with my bot (e.g. occasionally crashing)
eulerscheZahl: otherwise my ranking is mostly stable on resubmit usually
AntiSquid: also you can drop way below your real rank if stronger bots keep being submitted
Quidome: Hm that is interesting
AntiSquid: if weaker ones get submitted you get pushed up
Quidome: @antisquid, true, but that is not the issue here
Nagatwin: eulerscheZahl On top20 this is true, but below 50 the last battles have a big influence on where you end up
eulerscheZahl: hm..
i rarely play in that regions :innocent:
Nagatwin: Haha
Quidome: @eulerscheZahl, this fundamentally wrong remark is interesting
AntiSquid: part of the issue imo, if you submit when a stronger bot is climbing you occasionally fight it and get dropped while submit is still not even finished
AntiSquid: Quidome
Quidome: Yes, but here are the numbers.
Quidome: I am at 130, then I resubmit and after the complete resubmit i am at 160. Wait for 12 hours and back at 130
eulerscheZahl: the fundamentally wrong part makes you lose vs all kinds of bots, even lower ranked ones
eulerscheZahl: when you wait long enough after your submit, you mostly play against those in your region, which are pushing you up slowly
Quidome: Yes, but how will this result in a grudual climb?
eulerscheZahl: race condition or do i still have to explain it?
dbdr: AntiSquid why would strong people submitting affect you more than the people around you?
AntiSquid: you lose more dbdr
eulerscheZahl: because you are the special snowflake
Nagatwin: dbdr when a stong one submits, he crushes you with low elo
Quidome: YOu play against yout region, how is that workoing?
Nagatwin: so you basically loose 1-5 more battles depending on your luck
dbdr: you lose more than what? than the people around you? then you deserve to go down
Quidome: How is it working anyway? I assumed a resubmit plays evryone on the list
eulerscheZahl: others submit and play against similar ranked bots most of the time (including your bot with a fundamental bug that makes you lose 10% of the games for whatever reason)
AntiSquid: ah missread, well dbdr it's random, sometimes you don't lose, since you dodge, but depends on luck i guess
dbdr: definitely there is noise
eulerscheZahl: but in 90% you play well and win most of the matches. so you just need more games to climb in ranks than you should need without the fundamental bug
dbdr: I agree with eulerscheZahl, an unstable rank for your bot does not necessarily the ranking system is broken
Quidome: Aha, that is interesting
dbdr: it can be you play well against some kind of bots and badly against others
dbdr: the solution is to improve your bot ;)
Quidome: The thing is there is something going wrong sometimes. So your analysis works on my case
eulerscheZahl: that too. and when the rock to your scissors keeps submitting all the time, your rock will sink like a stone
dbdr: but it's easier to complain about the system ;)
AntiSquid: and also i mentioned both situations, including the time when weaker ones submit and you get to climb
dbdr: as well as everyone else?
AntiSquid: but really you can't be sure if you really are above or below player x
dbdr: everyone goes up? :D
Quidome: I am not complaining, I am a happy camper. Just trying to understand
AntiSquid: dbdr randomly
AntiSquid: some get more bad match ups some better
dbdr: right
inoryy: dbdr say you're top50 and resubmit at the same time as top5, you end up meeting around top100. first, there's double the chances of playing (since you and him request a game from the same region), second, when you lose it will affect your std.dev much more than somebody who's already stabilized. so you can randomly crash 100 ranks and then have to re-climb.
dbdr: it will affect it less
AntiSquid: you can 30 spots today, drop 60 tomorrow, but nobody really improved
dbdr: because the system knows that player is not yet placed with certainty
chucknorris: with a fresh submit your sigma will be bigger and wont affect less your opponents points
inoryy: xD
dbdr: the main problem is the ranking does not display the variance, just keeps it internal
AntiSquid: hard to see since you expect everyone improved, but after resubmit you might end up at 30+ spots above what you should be and you wonder ...
chucknorris: will*
dbdr: so we easily misinterpret what we are shown
inoryy: that's not how it works, dbdr
dbdr: not saying the system is perfect. but it's better than many people think
eulerscheZahl: :popcorn:
dbdr: it is, I read the TS paper
inoryy: I also did and have bsc in stats
redet-G: hello
redet-G: world
AntiSquid: did you play on halite? it's a good example of a more accurate leaderboard
dbdr: good. so you agree it's hard to reason when you are shown avg - var only, no?
Quidome: Thnxs for the discussion guys! Really helpfull :)
inoryy: sure, but it doesn't invalidate my explanation
Quidome: I will go back, fixing bugs :P
inoryy: double the chances of meeting and much wider sigma swing on a loss
dbdr: you are saying losing against someone with a high variance makes you go down more?
dbdr: that sounds backwards to me
AntiSquid: this was supposed to be a quick random mention, not a fulfledged debate, it's not that great and there are some better ways already out there. that's all i wanted to say
inoryy: your sigma is high as well, it jumps wildly on losses
dbdr: yes, but that's just sigma
dbdr: and we interpret it as going down
inoryy: just do a resubmit and track your progress, every loss in the first ~30-40 games will drop you like crazy
dbdr: because that's wha CG shows, but it's not
dbdr: yes, your rank, not your hidden avg
**dbdr afk
inoryy: your mean drops because your sigma is high....
geouwu: getting gold league is a good thing right
geouwu: i started using this app a few hours ago
Nagatwin: yes it is
dbdr: no, your mean - sigma drops because sigma is high :)
dbdr: well, both actually
dbdr: there, everyone's right! :D
Nagatwin: the mean could go higher with a high sigma too
AntiSquid: company leaderboard is an interesting way to motivate colleagues to join eh ?
Nagatwin: yeah
Nagatwin: Im noo sure my startup colleagues are too technical but thats for sure a good way to motivate
Nagatwin: not*
AntiSquid: so University of Wroclaw is the best?
JFB: I think that it is a lot of aCat's students. Probably he will give them points in egzams or simply result of the contest will be egzam's score
AntiSquid: is there a pretty way to split strings by character in C++ ? nothing looks elegant when compared to what other languages have
darkhorse64: http://chat.codingame.com/pastebin/70cf277d-ce50-43fa-8331-5274b7f3656f
tranculent: I'm not sure if we can ask questions in regards to the Ocean of Code challenge but how did you get all the 4 cells around (x,y)? I've implemented it but I am using 4 loops which I don't think is very optimal.
Nerchio: just check (x-1, y) (x+1,y) (x, y+1) (x, y-1)
Nerchio: :p
JFB: In my representation each cell is a bit in 256 bit number.
elsinnombre2d2: why do u use this representation?
AntiSquid: i use dir array tranculent
Nerchio: in theory you can surface -> move -> torpedo -> sonar -> silence -> mine in 1 turn? :D
AntiSquid: https://youtu.be/z5YhwAqT7Iw?t=1604 i find it convenient tranculent
JFB: Nerchio - not nly in teorey - in practice too
elsinnombre2d2: a good approach could be to use silence after use torpedo
tranculent: thank you AntiSquid, that will definitely help (y)
Nerchio: why elsin?
elsinnombre2d2: he
elsinnombre2d2: coz when u shoot is easy to know where u are
Nerchio: i guess they can pinpoint your location more
JFB: Becouse torpedo inform about your position - and silence will mask your position
JFB: Addionally you can faster fun from your opponent and you will add time to recharge torpedo
JFB: If you lunch torpedo you think that you are neer you opponent. So afer torpedo you are without torpedo neer your opponent - it not looks as good for you is not it?
JFB: ;-)
Nerchio: smart guy :D
Nerchio: I just hope my opponents can't track and i am free to do whatever i like
JFB: In silver and for sur - in gold - it is not valid assumption that ypur opponent can not track
Nerchio: yea i know i am in silver
elsinnombre2d2: im trakcing the opp since bronze, with a very good accurate
Nerchio: I still hope my opponent is bad at tracking :D
elsinnombre2d2: the silence is the best tool to avoid that
elsinnombre2d2: its not really hard to track the opponent
Nerchio: when i tried to do perfect tracking which accounts every possible silence location it takes too much time for me
JFB: Using only sillence is enoug to promote to gold. But it is only enough to be aroud the end of gold.
JFB: ~160
JFB: Maybe !180
JFB: Track + silence = #180 in gold (only)
KANEKI: yeoooo
JFB: You have to add mines at least to better position
KANEKI: If you have N buckets, and the distance between them is D, and each time you move the buckets you split S water for units of water
KANEKI: How do you find the total spill
KANEKI: Assume n/2 buckets are filled
JFB: Add each spill ? ;-)
KANEKI: If odd, round down
KANEKI: So what would be the eq
JFB: Why to move bucket? If it is no move = spill is 0 - :-)
KANEKI: You have to move buckets
KANEKI: Only time spill is 0 is if N is zero
KANEKI: which is the number of buckets
KANEKI: Or if its 1 because only n//2 is filled
KANEKI: How would you find the total spill, help
JFB: So final result is all woter in only 1 bucket?
JFB: You explained starting point (N buckery, N/2 filled with woter) but you do not explained the final result
JFB: Without knowing the goal - it is not possible to answer your question (what will be total spill)
Nerchio: do you guys use sonar?
Nerchio: if you have superb tracking it's probably not that useful
eulerscheZahl: when the opponent makes clever use of surface, it still helps
eulerscheZahl: but currently my sonar is deactivated
AntiSquid: probably sonar is good if you want to track everything and reduce the possible paths Nerchio
pb4: sonar is good to find the opponent early
Noobmaster69: Anyone here interested in Tron battle?
Noobmaster69: The bot programming game
TERMINATE: mashiron noob
MadKnight: Noobmaster69 u wanna rekt somebody ?
Noobmaster69: Haha
kovi: is there a define for c++ to differentiate between ide and submission?
Noobmaster69: Yeah I wanna reck Thor
pb4: kovi no idea
chucknorris: some1 told that there is a debugger attached to the ide environment kovi
chucknorris: my code is much slower in the ide
pb4: oh ?
eulerscheZahl: but afaik there is no way to check in code if you are in the IDE or arena
eulerscheZahl: except if you code in Rust
kovi: thx
JFB: I just checked NDEBUG - it is not set in IDE and in submition too
JFB: In some environmend NDEBUG is set in release - but not in CG it looks
pb4: kovi : why would you want to detect if you're in ide or submission ?
Illedan: I guess you can crash yourself in the IDE to prevent others from benching against you :P
kovi: actually dbdr does that
kovi: pb4 - debug messages
Illedan: he timeouts?
kovi: hmm, i havent checked
kovi: could be
RoboStac: probably not intentionally, thats just how rust works on cg
dbdr: kovi not on purpose...
Illedan: Maybe it's possible to run an IDE match in Release mode, just not in the UI? Anyone knows?
kovi: i c
dbdr: the way I know it to send a compiled binary
kovi: uh, so heuristical game
dbdr: no, in the arena ist's fast
dbdr: it's
eulerscheZahl: don't let the binary be your final submit. that can cause a disqualification
dbdr: I know
pb4: ^
dbdr: I never submit a binary
eulerscheZahl: pb4 knows...
dbdr: that's why it times out for you in the ide :(
eulerscheZahl: was Thales soccer, wasn't it?
pb4: Yup
dbdr: that one had prizes too, no?
pb4: It was more a reflex than anything else -_-
eulerscheZahl: it had prizes
Illedan: Would you get a prize? :P
pb4: sending a binary doesn't improve performance
pb4: actually in some cases it reduced it
eulerscheZahl: robo won a bicycle i think
but then got the money instead
pb4: (vindinium)
dbdr: fix your compiler ;)
eulerscheZahl: or was that decathlon?
too many contests
RoboStac: it was decathalon, but that was the soccer one
dbdr: pb4 why do you do it then?
eulerscheZahl: thales was the CSB2 then?
RoboStac: the ctf one I guess?
eulerscheZahl: yes
pb4: for NN dbdr
dbdr: ah, for the data?
AntiSquid: the weights i guess
eulerscheZahl: i don't like to call it CTF, as i connotate that with finding security issues in a system
pb4: I spent a lot of time on CSB so the process of "run the script, put on CG" is a bit ingrained in my fingers :D
dbdr: capture the flag is a phisical game you can play outdoors, originally ;)
pb4: The script is really convenient, it already puts the binary data in your "copy/paste buffer"
eulerscheZahl: running my build script merging multiple code files into one for submitting is even part of my project settings
eulerscheZahl: so i can understand that you didn't really think about it
dbdr: I do the same
Nagatwin: pb4 for NN do you have a preferred way of embedding weights ?
dbdr: # generate file
cat file | xclip -selection c
pb4: yup, that's it :)
dbdr: in bash the script
Nagatwin: Apart bins :p
pb4: Nagatwin : directly within the compiled binary
Nagatwin: pb4 and when you should not upload compiled bins ?
eulerscheZahl: hm, sounds conventient...
but when i'm debugging i don't want to replace my clipboard
pb4: constexpr array<float, 1000000> weights = {0.21231,0.23213, 0.3213...}
dbdr: eulerscheZahl use a clipboard history manager
namaska: Can you help me with my problem into my code ?
eulerscheZahl: i never heard about it but now i need it
dbdr: you 8definitely* need it
dbdr: incredibly convenient
pb4: never heard about it either
dbdr: I use Parcellite
pb4: on which os ?
Nagatwin: pb4 don't you reach the maximum file size ?
eulerscheZahl: linux
eulerscheZahl: let me apt install it quickly
pb4: nagatwin : yes you're still limited, but you can pack much more
pb4: 1 float in text = mabe 9 characters
eulerscheZahl: installed, how to use it?
dbdr: it sits on my taskbar
pb4: 1 float in binary = 32 bits, so only 4 characters
dbdr: let's see how I start it
eulerscheZahl: i have no taskbar
Nagatwin: Yeah
dbdr: probably in login script
pb4: there is also some regularity
pb4: most floats are around 1
AntiSquid: idk, it's quite limited, how many floats do you get in total pb4 ?
dbdr: i3wm
eulerscheZahl: status bar
eulerscheZahl: i3status
dbdr: that one
pb4: so you can zip the weights and they'll be reduced even more
pb4: AntiSquid : I don't remember but quite a bit
AntiSquid: yes that's what i mean, quite a bit, isn't it too easy to exceed the limit?
pb4: 15k floats is relatively easy to pack
pb4: You can also use less precision on your floats
dbdr: eulerscheZahl I have
exec parcellite
in ~/.config/i3/config
pb4: keep only 18 bits for example
pb4: etc...
AntiSquid: how low would you go on the decimal count for weights
eulerscheZahl: ok, got it running already
eulerscheZahl: thanks
pb4: many techniques to store more
dbdr: pb4 what kind of network sizes do you typically use?
pb4: AntiSquid I've gone as low as only 12 bits per float and didn't see any adverse effect
dbdr: oh, 15k weights?
eulerscheZahl: from the man page:
Hotkeys
The available hotkey modifiers are <Ctrl>, <Alt> <Shift>, <Release>, <Meta>, <Super>, <Hyper>, <Mod[1-5]>
eulerscheZahl: i don't know half of those keys
dbdr: :D
wlesavo: <Super> <Hyper> <Release>
dbdr: sounds naughty
eulerscheZahl: man page from 2011
either poorly maintained or it works that well that it doesn't need any updates
eulerscheZahl: i'll hope for the latter
dbdr: works fine for me
AntiSquid: "man page" :thinking:
eulerscheZahl: for me too, so far
eulerscheZahl: haven't tested much
eulerscheZahl: do you want a woman page AntiSquid?
AntiSquid: you called it that twice :D
eulerscheZahl: 2 women pages
AntiSquid: must be a really manly page
dbdr: if you don't know how to use it:
man man
eulerscheZahl: i never thought about entering that command. but now it makes sense that it exists
dbdr: gcc gcc.c
dbdr: half the fun in CS is self-referential stuff :)
eulerscheZahl: we need a quine puzzle in an esoteric language here on CG
pb4: when is legend by the way ?
struct: friday
kovi: yay, lucky rolls for me
pb4: nice all green
dbdr: all green?
AntiSquid: all tests ok ?
pb4: he has >50% against everyone
dbdr: ah :)
duckquackduck: what kind of pathing algorithm does the Silver boss use? It is sooo smooth
AntiSquid: in ocean of code?
duckquackduck: yes
eulerscheZahl: no idea, it's a user who was selected as a boss
eulerscheZahl: so only that user and CG staff can tell
dbdr: wasn't it inoryy?
duckquackduck: ah I see. Its very nice though
eulerscheZahl: inoryy = bronze boss
eulerscheZahl: silver is Risen i think
inoryy: oh what
eulerscheZahl: you are the bronze boss
inoryy: how did I miss that
wlesavo: lol
eulerscheZahl: roll back to your old bot from bronze opening and test yourself
skal: rip
dbdr: and your path is smooth :)
eulerscheZahl: bronze players optimize against your bot and then destroy you in silver :D
inoryy: it all makes sense now
dbdr: rip
dbdr: oh, so it's quite bad to be boss gold too then?
eulerscheZahl: i think so
eulerscheZahl: i hope i won't be the boss :P
dbdr: never heard that theory b4
dbdr: but it makes sense
eulerscheZahl: especially in a rock paper scissors game
dbdr: make sure to be higher or lower ;)
dbdr: of the unknown rank
eulerscheZahl: for a sim heavy game like mean max i would care less
wlesavo: make them to optimize against a different version and then destroy everybody
Quidome: @R4N4R4M4, i am debugging my opponent trace and using your debug messages. At some point you have a different answer and I think I am right but not sure of course, interested in a debugging session?
inoryy: haha https://www.codingame.com/share-replay/446914756
eulerscheZahl: Quidome try another bot that prints stats too. to see if it's you or r4
Quidome: hm, can do that, but that will be a differnet match :)
dbdr: inoryy nice. I was thinking, game is ot symmetrical, they should diverge
eulerscheZahl: of course it will be a different match
Quidome: :P
eulerscheZahl: https://www.codingame.com/share-replay/446915569
eulerscheZahl: my numbers don#t match with r4
Quidome: Aha, thnxs a lot
eulerscheZahl: i print me tracking the opponent first and then the opponent tracking me
eulerscheZahl: paths + cells
eulerscheZahl: (current cells)
Quidome: cool, going to try yours then :)
eulerscheZahl: zenoscave is printing stats too (or was the last time i checked at least)
Quidome: great, thnxs
R4N4R4M4: Hi everybody
Quidome: Hi
R4N4R4M4: Quidome, my detection is not really good, I'm working on it
Quidome: Aha, ok, thnx
eulerscheZahl: mystery solved
R4N4R4M4: Because I'm using heuristics and not simu
Quidome: I will put in stats myself for comparison
kusrabyzarc: Sorry, but can you check it, please: https://www.codingame.com/contribute/view/4941a1bc0861e1fc3aec3cfc2872f54f9f53
eulerscheZahl: my tracking is finished (including mine explosions that reduce candidates for previously visited cells)
eulerscheZahl: so if you get discrepancies, there is a bug in one of our bots
eulerscheZahl: if it's me, i'd like to know about it ;)
R4N4R4M4: eulerscheZahl, it is a great contest you made, and a tough one :D
Quidome: OK i am going to look into it now
eulerscheZahl: yeah, which "I" made :smirk:
eulerscheZahl: but i like it too
eulerscheZahl: my part was the smallest of the 3 users mentioned
blasterpoard: kusrabyzarc the statement is unclear and gramatically incorrect
blasterpoard: imho it shouldn't be approved in its current state :(
eulerscheZahl: and now write the same on #ru
Quidome: mine explosions for previous visited cells...
Quidome: Didn't thought of that yet
Quidome: hm, more to do :D
Octothorpe: hi
kovi: hmm, maybe not so lucky rolls this time
Nerchio: wait
Nerchio: can yo do silence with 0 range?
Nerchio: "This allows you to move 0 to 4" omg :D
aCat: yep
aCat: thats a dirty trick
Nerchio: i just found this out and i am in silver
aCat: cause you ignore that you visited this square
kovi: i player silence n 0 till silver
Nerchio: i am changing my structure and my bot doesnt track silence yet but it tracked the enemy properly because silence 0
Nerchio: haha
Nerchio: so it's like waiting a turn for your bot?
Nerchio: i know enemy sees it different but for you
DEIZ21: is it possible to drop a mine and trigger it in the same turn?
mark7: NO
DEIZ21: thx
DEIZ21: and surface and move in the same?
DEIZ21: ?
eulerscheZahl: why would we need both commands then? they are not the same
mark7: Yes
aCat: why no?
aCat: you can't place mine and trigger
mark7: NO you can't place mine and trigger same mine
DEIZ21: cause i tried dropping a mine and trigger it in the same turn but it doesnt work even it seems ok
eulerscheZahl: why "why no?"?
aCat: I mean it makes sense - place and fire
aCat: bu :(
DEIZ21: yes but not allowed by the referee
mark7: aCat DEIZ21 Please read decription
eulerscheZahl: because rules disallow it, that's why
DEIZ21: already done
eulerscheZahl: rules of the original game disallow it too i think
eulerscheZahl: "You can't place and trigger a mine in the same turn."
DEIZ21: i didnt see any thing speaking about triggering the same mine in the same tunr
mark7: eulerscheZahl can you post replay of our bots?
aCat: OK, there is in triger description !
eulerscheZahl: https://www.codingame.com/share-replay/446934342
DEIZ21: ok my bad
mark7: And maybe some advice too.. :grin:
DEIZ21: what advice?
mark7: How to improve
DEIZ21: so tell me :)
mark7: As now i can successfully track my possiblities in opponent and opponent possibilites
DEIZ21: yes it s on my agenda
DEIZ21: cutting paths with mines too
DEIZ21: not implemented yet
DEIZ21: i lost 30 rank in 2 days
DEIZ21: so i need to improve
eulerscheZahl: do you try to avoid my mines mark7?
JBM: hey eZ you still around?
eulerscheZahl: https://www.codingame.com/share-replay/446935048
i think you don't
JBM: what's match end condition wrt winner decision?
JBM: i had the following two in mind
mark7: Yes Kind of but its not working ... Main problem with my bot is path finding :-(
JBM: A brings B to 0, then triggers mine and falls to -1
JBM: what's the match outcome?
eulerscheZahl: 2nd question: do you have a killer move when you can enforce a victory?
eulerscheZahl: that helps quite a bit
mark7: no..
mark7: Like optracked, or silencing to go near opponent?
eulerscheZahl: see how i win in the first game. i don't even know exactly where you are. but i know that i win
mark7: eulerscheZahl No I dont have that
mark7: Added in todo list to implement
eulerscheZahl: add it. that's my advice
eulerscheZahl: low hanging fruits
mark7: Hehe!
mark7: http://chat.codingame.com/pastebin/b361cf26-b532-48ed-9b58-27562a24609d
eulerscheZahl: dude...
eulerscheZahl: it's a contest, don't share code
pb4: brb, let me put this into my code :D
mark7: Its already mess
mark7: pb4 it wont help you much :-)
pb4: "JBM 21H09
A brings B to 0, then triggers mine and falls to -1
what's the match outcome?
"
pb4: A wins
pb4: game stops as soon as one player is down to 0
JBM: kthx
JBM: i recqll their was some kinf od ambiguity as to how turns split, but i can't get it back
JBM: that's way too many typos, lemme try again
JBM: i recall their was some kind of ambiguity as to how turns split, but i can't get it back
mark7: Silver #2 Timesout as soon as I use Silence , Still at 2 position
eulerscheZahl: then that bot is doing something else much better than you are
redet-G: https://www.codingame.com/clashofcode/clash/1179964ba3a85be6b3c8bdf695a19b5aece80da join me
eulerscheZahl: no
mark7: Maybe only timesout when I use silence as first powerup
mark7: bewating #3 https://www.codingame.com/share-replay/446940445
DEIZ21: mark7 see this
DEIZ21: https://www.codingame.com/replay/446940617
DEIZ21: your bot knew where mine was it moved and charged torpedo but didnt fired
DEIZ21: you need to fix this
YannT: yo kovi, back in the game :)
DEIZ21: frame 240
mark7: DEIZ21 Submitted another bot try that
Illedan: Is the top real? Or does it feel random?
Illedan: (ranks)
kovi: real with uncertainty of 3-5
chucknorris: +- 1 point
kovi: yesterday i was around 5, no around 2
Illedan: Same code?
wlesavo: i liked pb4's submit today
MadKnight: hey Illedan do setters in c# return value ?
MadKnight: do they return value or they call getter ?
eulerscheZahl: yesterday he did it on his own
eulerscheZahl: had a broken bot this morning
Illedan: Setters do not return
kovi: nah, i had promising improvement. i submit multiple times to be sure
Illedan: aiit
wlesavo: no i mean when he was 2 points above top#2
Illedan: Looking forward to legend
Illedan: more game
MadKnight: so u can't do var x = obj.Prop = 3;
?
Illedan: s
Illedan: Sure, if it has a getter
MadKnight: so it just calls getter ?
MadKnight: and it doesn't reuse value passed to setter ?
eulerscheZahl: you can have a property with getter and setter
eulerscheZahl: a setter doesn't have to call anything as it doesn't return a value
MadKnight: but will it use Prop's getter ?
Illedan: https://tech.io/snippet/OOaIxK7
MadKnight: oh
kovi: 35.59
kovi: now that is lucky rolls
struct: is the luck on the enemy you get?
struct: or on the maps?
pb4: wlesavo : I repushed the same code right after that, was #5
wlesavo: pb4 yeah, but was nice for a sec :slight_smile:
JBM: pb4: were your answers to me "statement says so" or "referee says so"?
pb4: None of the above
pb4: Referee author says so
pb4: :D
JBM: could be mistaken
JBM: wouldn't be a first
pb4: right. Since I do not intend to stand on a mine and trigger it when I've won, I decided not to investigate more
JBM: sounds qbout right
JBM: $% ^&* knd layoutsb
JBM: (*kbd layouts)
dbdr: referee checks victory condition after every frame
JBM: that's a better argument
JBM: not that i know what a frame means, but i can kinda infer
struct: 1 frame = 1 action
struct: MOVE N TORPEDO | TORPEDO x y | SILENCE N 0
should be 3 frames
JBM: That's better
JBM: you guys rock
dbdr: :guitar:
JBM: https://www.youtube.com/watch?v=3rzgrP7VA_Q
dbdr: I did wonder if you could surface, get to 0, and still kill your opponent to get a draw :D
dbdr: my conlusion was that you can't
DEIZ21: guys what s better than fluidfill for this bot?
JBM: (ahmerdecestpaslebon)
JBM: like, anything
JBM: maybe this one https://www.youtube.com/watch?v=e7z1PuDRCko
JBM: tea much better
JBM: *yea
dbdr: alternative rickroll?
JBM: so much better 'an a rr
DEIZ21: what can i implement instead of fluidfill?
dbdr: DEIZ21 I was top 10 with a floodfill
ZarthaxX: not only a floodfill i guess dbdr lol
DEIZ21: cool but i think i'm found verry offen du to my movement
ZarthaxX: also some search mixed
DEIZ21: exact Zart
JBM: the word on the streen right here today was hamiltonian paths
JBM: *st
ZarthaxX: what?
ZarthaxX: hamiltonian paths for this?
dbdr: that won't take care of being discovered, will it?
dbdr: actually makes it "easier"
dbdr: theoretically
ZarthaxX: tru
ZarthaxX: dbdr doing search or wont talk? :P
dbdr: I'm doing some kind of search, yes
ZarthaxX: i see
ZarthaxX: tried smth but possible sequences of actions exploded
mark7: When you make a move do you only calculate one move ahead or how many move ahead?
dbdr: you could always go the way of recurse in his locam stream, MC
dbdr: or prune
ZarthaxX: oh.. tru
ZarthaxX: i can check that, thought it would miss key moves
JBM: anything predictable makes it more likely to be discovered
JBM: just have enough alternatives and pick at (unpredictable) random
ZarthaxX: lol the way to go JBM
JBM: one of'em
reCurse: What did I do
dbdr: locam stream
JBM: the bitstream stuff?
JBM: man, i still hear about that now
reCurse: No he actually talks about MC
dbdr: yeah, but not that
reCurse: Which is the part no one actually remembers
reCurse: So I'm surprised
dbdr: crazy right?
JBM: who knew
dbdr: wow, I found the winner of the contest
dbdr: https://www.codingame.com/replay/446916405
dbdr: frame 322
wlesavo: if he would also surface and triggered that turn i would agree
dbdr: ah true, not complete
dbdr: who'll take the challenge?
JBM: just do it
dbdr: it's more impressive if it happens without planning it
inoryy: what am I looking at there, the fakeout trigger?
dbdr: bonus points for 4 damage and victory on the turn
JBM: who are we to decide that
dbdr: all powers used in a single turn
Illedan: This is why we dissalowed triggering on the same turn as dropping a mine :P
DEIZ21: :D
JBM: well, move's not really a power is it
JBM: didi say say anything sensible recently? carry on
icecream17: well, the only way to charge is by moving
Astrobytes: On the rum again JBM? :D
JBM: not tonight, thx for asking
Astrobytes: :P
JBM: well not right now anyway
JBM: sharing :beers: with the good folks on #Fr
Astrobytes: Santé
JBM: noice
JBM: how are you even allowed to do that
Astrobytes: What, use a French word in World?
JBM: shame on you
JBM: you should see the CM when we dare use some anglicism on #Fr
Astrobytes: Yeah I've dropped a few English words in #Fr once in a while, it's such a French response :D
Astrobytes: *words/phrases
JBM: naw, you're way below the threshold
Zenoscave: Do you guys look ahead more than one move for mine avoidance?
Zenoscave: or path finding even
ghassen: the further u see the better, usually
Astrobytes: JBM has anyone ever used "possiblement" in #Fr? :D
JBM: nope
JBM: we say "eventuellement" for that
JBM: one of the greatest source of en/fr confusion ever
Astrobytes: lol, good. No I know, it's just a lot of my French colleagues back when I was in London used to confuse their relatives on the phone by using it
Astrobytes: Too Anglicised you see
JBM: eventually, actually
JBM: top false friends
Astrobytes: Possibly
JBM: possibly isn't really in french
Astrobytes: Yes, that's the point
JBM: yup
Astrobytes: "possiblement" "huh?" "PUTAIN. Fucking English"
Astrobytes: Frequently heard phone conversations
Astrobytes: Never mix with the English speakers. It will ruin you.
JBM: Never
aCat: Name it boring game:P https://www.codingame.com/share-replay/446974916
Astrobytes: That was... yep. Boring :D
aCat: Shhh its all has it purpouse
Astrobytes: One of these accounts for the students? Or just testing?
aCat: Testing
aCat: I'm tweaking LOCm account so I can train to track exact one thing
Astrobytes: Shocking aCat. 2 accounts. Naughty cat.
struct: !ban
Astrobytes: :D
aCat: It was accidentally!
aCat: really
Astrobytes: hehehe
aCat: believe this cat, this cat is an honest programmer
Jlivingstonsg: I have join codingame because I found a DOS game called BlockOut from 1989 that we played and today you can play it in your browser. But I have still not found a version that can register points on a global database so that is what I plan to do. Does anybody here know about BlockOut? Its a 3D Tetris
aCat: LOCM acc was made for the stream purposes - I just now fiound the way to use it
Astrobytes: Yeah I know, all good :D
struct: Jlivingstonsg I just checked
Astrobytes: Jlivingstonsg have you seen the codingame SDK for creating games?
struct: The problem for me with games like tetris on CG
struct: is that on tetris pieces start moving after some time
struct: For example on 1vs1 tetris
struct: you can place 3 pieces and enemy 0
Jlivingstonsg: I will check SDK later
struct: here is link if you want to save https://www.codingame.com/playgrounds/25775
Jlivingstonsg: Google: blockout online and try one in your browser
struct: Do you want to make it a solo player game?
struct: single*
Astrobytes: I know it, not sure if it would be a viable game for CG purposes but do feel free to try
Jlivingstonsg: I am just here to practice coding. If it is possible to code BlockOut here then I will have a go at it here.
Astrobytes: struct, how's your OoC?
struct: Im not doing it astro
Astrobytes: Not feeling it?
struct: yeah
JBM: afraid are you
struct: Im just not good at these types of games
struct: So I dont even try
Astrobytes: I'm not massively into it myself but I'm gonna try a bit now I'm feeling better
Astrobytes: Spent some time refactoring so at least I got rid of the mess and can do something!
struct: what league are you in atm»
Astrobytes: Silver
icecream17: MESSAAAAGE! :smile:
JBM: well get crack'n
Astrobytes: Didn't submit for a couple of days until tonight, and gained 150+ places
Astrobytes: yes icecream17 :D
Jlivingstonsg: Its quite fun. I work as math and physics teacher on high level and I always challenge students on schools in BlockOut/Out of control and they have no chance. I am 56 now and have score over 100.000 points and kids today that play games all the time can often never get over 20000 points. I thouught kids today were professional gamers but in spatial thinking they really suck.... :)
JBM: what points?
Astrobytes: JBM BlockOut (3d tetris)
JBM: oh wow
JBM: that's, like, Old
Astrobytes: Yup
JBM: i loved that game
Astrobytes: Think it's doable with the CG SDK?
JBM: nope
struct: too much work
JBM: too 2d-centric
JFB: Jlivingstonsg :-) At least sb older that me (I'm 55)
Astrobytes: That's what I thought too
Astrobytes: Ah, I feel so young now at 38
struct: And I feel like is under impression that here you code games
struct: :p
Astrobytes: That too struct
JBM: naw, nobody writes those
JBM: they just come out of the ther
JBM: *ether
JBM: come ON kbd, what's with ya tonight
Astrobytes: PEBKAC
Astrobytes: :P
JBM: trying to push it lower
JBM: but it comes back biting me
DEIZ21:
Astrobytes refactoring is never bad thiing; i just finished refactoring my code and guess what?! i found a bunsh of ugly stupid bugs, i fixed them and submited. i didnt add any new feature and guess what? i gained 50+ in rank from 135 to 82 for naw
JBM: err... lol
JBM: refactoring can totally be bad thing
DEIZ21: true
JBM: one anecdote doesn't a rule make
DEIZ21: actually i began yesterday
Astrobytes: absolutely
DEIZ21: bad went in bad direction and was tired
Astrobytes: But I spent 2 days and tested everything
JBM: you could also have introduced a new bug
JBM: that happened to cancel one of the old ones
DEIZ21: so i restarted fresh today and kept it simple
JBM: no generalization holds
DEIZ21: no reason to make complexe things
JBM: yeah, no ruls there
JBM: show me your code and/or stream, i'll tell ya
JBM: Astrobytes: same for you ;p
DEIZ21: i was upset telling my self poeple are adding new stuff and features and i'm loosing my time on refactoring
DEIZ21: but as i said it was totally worth it
Astrobytes: I would be deeply ashamed to code on stream lol
Might sell you it for 5 euro though :P
JBM: why the shame?
JBM: everybody's got to learn fro; everybody
Astrobytes: DEIZ21 I've been ill so taken me some time to do it, but definitely worth the effort
Astrobytes: Would be a very long stream JBM :D
DEIZ21: exact
DEIZ21: typing like ..
DEIZ21: a dumb
JBM: that's the fun of it
JBM: i still got no idea how twitch thinks 626 people watched my stuff
JBM: probably not paying enough
DEIZ21: jbm search for illedan
JBM: why?
DEIZ21: he streamed the basic things
JBM: i listened to it
DEIZ21: he type quicker than i could speak :D
JBM: was too seesharpy for me
Astrobytes: lol
JBM: nope that'd be errichto
DEIZ21: me too
DEIZ21: thx for the creator of pause/play :p
JBM: then again errichto didn't go that high
DEIZ21: who?
struct: he only coded for a few hours though
JBM: illedan moreso, but twas too long for me to listen to the whole of it
DEIZ21: errichto
Astrobytes: I can type quick, but as a biologist streaming programming I'd feel like an impostor, frankly
struct: his youtube channel is getting quite big
JBM: wouldn't be
JBM: * not that many real "this is all i cna do" programmer people exist anyway
JBM: * rankings matter a helluva lot more
JBM: gtg spk to ya tomorrow
DEIZ21: guys i need a more efficient way to move
DEIZ21: now i m using fluidfill
DEIZ21: any idea?
Astrobytes: gn JBM
Astrobytes: I'm still using floodfill DEIZ21
Astrobytes: But you're gold right?
DEIZ21: yeh
DEIZ21: they told me there is a better way
DEIZ21: but didnt indcate any thing
Astrobytes: I'd be looking at moving either away from or towards the enemy, based on your tracking
Astrobytes: So, generating possible paths
Astrobytes: (for the enemy)
Astrobytes: And moving based on that
DEIZ21: how do you type with red
DEIZ21: cant mention any person here
DEIZ21: how the hell
Astrobytes: DEIZ21 Type the whole username, or the first few letters + TAB
DEIZ21: Astrobytes ah that was th tab the missing thing
DEIZ21: doesnt work
Astrobytes: That worked
DEIZ21: ah
DEIZ21: ok work only for the person?
DEIZ21: not stupid
Astrobytes: It's only red if someone tags your nick
DEIZ21: k thx
Astrobytes: no worries :)
AntiSquid: look at SuperNit's messages :D
AntiSquid: silly
Astrobytes: Quoting Portal no?
kovi: 36.74
struct: Seems the margin is getting higher
DEIZ21: how adding 3 chars in the code could make u loose 100 in ranking :p
struct: What are these chars?
DEIZ21: d>3
icecream17: ugh. this random person in 438th somehow always exploits obscure hard to find bugs in my code
icecream17: yes! I used cg sync! should've pressed that obvious button earlier
yfprojects: @AntiSquid I meant this one: https://www.codingame.com/ide/demo/758100a939c77b58a4a37fc56ab749a6cbced5
