It’s irrelevant to the puzzle, just the random display output 
I’ve found all, logic, direction, information, board. But only 974Pts… Why my algo Follow Wall doesn’t work ?!
You are probably missing something. And this is likely “when does a run ends?” or “when does my score goes up?”
Around 330 on testing M1L1, and 974/980 on Submission. I discover CodingGame recently and unable to find how I can see the Rank of all player, like other game “PacMan” etc.
you have general leaderboards here: https://www.codingame.com/leaderboards/general/global
compete > leaderboards
and you have a leaderboard for each multiplayer game
Thank’s
Haaaaaa 1938 !!! I know now this game, but hard to … 
Maybe last things to add.
Is it possible to view my output on 2 Failed test after submitted or have a much harder training test to understand why ‘failed’ so close to success?
someone explain what this game does so I can code it properly. Been scratching head from 2 hours
The whole point is that you have to figure it out by ‘reverse engineering’ the problem
It was hard but after 2 days I got 1126 points with my first algo and now (3 days later) I have 3025 points !
I’ll try to improve my algo a little bit
Anyways thanks for this challenge !
1 Like
Very funny to try this hard reverse enginnering, i’m having fun by discovering small rule of the game one at the time, but, look like codingame doesn’t like me spamming the test each time i put a new rule on trial XD
This is a great challenge. At first, nothing made sense. But after digging a little bit in the Console and playing with the results, it actually made a lot of sense and it’s quite a fun game.
Late to the game, but I have to say: great challenge!
I managed to figure out the name of the game rather quickly but was slightly upset to discover later that the hints basically give it away – well, not fully – but I think what is suggested in the hints is exactly what is fun and rewarding if one figures it out oneself. It’s OK, I just felt less smart than five minutes before that discovery.
Be this as it may, I am quite sure I have discovered a bug in the reference engine (not the exploitable one regarding the final state) . This is hard to phrase w/o spoiling anything…
I tend to make assertions about my assumptions and raise an exception when my assumptions are violated. This has happened occasionally in the IDE test cases – as usual, mostly due to me making a mistake.
But eventually things stabilized, and I still got a few exceptions. One particular case is M7L2: at some point one of the mysterious pairs of numbers (7, 23) in the per-turn input indicated something that was later contradicted by character input 2. This was not the last entry in the list of pairs. I think this should never happen. The work-around was simple to implement and also made some occasional failures on the M3 validators go away (and yes, I know – I wasn’t making assumptions about initial input 3).
I would be interested to hear whether someone else made a similar observation and/or whether this is intended behavior.
P.S. I’m sure anyone who frequents this site will be interested in this:
ICFP Contest 2020
I can’t believe this topic has been dead for 4 years ? (Edit: my bad, I can’t read dates properly since they hide the year of the last reply if it’s the same year). It’s one of the most fun puzzles in my view. Though I was surprised when the “hiiring” popup came up (totally freaking me out).
Also being able to succeed all validators without trying felt extremely weird. Like being rewarded for putting no effort. There could a threshold…
Also the C initial code is wrong (mentioned above).
All that being said, the optimization can be pushed very far but you can do well without doing the hard work. It’s a shame so few people try this one.
Hi
Need advice how to debug the situation when I pass all testcases in IDE but after sumbit there is a single validator (M7L1) which fails for unknown reason. I’ve tried to reduce complexity of calculations to minimum and still it fails. I wonder how can I investigate what’s wrong if available testcases doesn’t cover the fail situation
In this puzzle, as long as you output a valid instruction (A,B,C,D,E) you should pass all test cases. So likely, you need to check in your code if there are edge cases where you don’t output any of these values or if in some scenario you are either trapped in an infinite loop or a loop that would spend more than 100ms to execute
2 Likes
12,346 people have submitted a solution
An image randomly appeared on the output screen when i was running the code. I tried going back but could not find it!
Then on another run it appeared again and I was able to capture it:
Sorry if this is a spoiler but I could not understand what it is, except a guess that it might be the playing field!
Might be useful for someone to figure out the game.
Ask your grand parents. For once, they will know more than you. Not sure their answer would be of any use though.
P. S. It’s also quite possible your grand parents need to be French too…
1 Like
Thanks for reminding me that I’m getting old. It’s not a spoiler.
3 Likes
So it was there just for distraction??
I feel so angry 
2 Likes