Is it possible for Indy to move from one room to the other through the BOTTOM? I know it’s not given in the problem statement, but should I consider this case? There are many room types (like Type 1) which will allow entrance through BOTTOM by rotation.
No Indy can’t enter in a room by BOTTOM. Only LEFT, TOP and RIGHT. But he can exit a room by BOTTOM of course (and enter in the next one by TOP).
Came here thinking that the puzzle was “flawed” a bit like I had mentionned for Skynet here, then realized that unlike Skynet there is a very hard version for Indiana, which seems to stress test a lot more our algo 
It’s funny that while they have the exact same rules, I get 100% on the hard, and 0% on the very hard. More work for later 
Hi,
I got 100% using BFS. At every node, I tried without change, turn clockwise, turn anticlockwise (if clockwise and anticlockwise do not give same node) and save it. If I have a preceeding “WAIT” move I try double turn.
This passes all tests except for the last 2.
For last 2, I ran my fastest movement list, ans when I got a “WAIT”, I tried eliminating rocks.
This was enough.
3 Likes
Hi everybody !
I was wondering something about this puzzle … I just solved the Episode 1. For Episode 1 you only have to compute which way to go when you enter a room. Now I read the Episode 2 description and I see that there are several paths, I got to find the good one, it might need to rotate some rooms and rocks can appear and block eligible paths.
The way I see it, I need to 1) compute the map, connecting paths using room types 2) find the best path for Indy (using probably BFS), then 3) consider I can rotate some rooms to look for more solutions and 4) at every turn I might need to consider rocks movements to avoid collision (so some rooms are “forbidden” at specific timeslots).
What I find strange is the difficulty gap. For Episode 1) you need to find the right path to take while for Episode 2) you need a BFS with possible rotates and rocks. It looks like a 10x increase of the difficulty level. Am I wrong ? Did I miss something there ? I hope this question does not look stupid.
I already got this impression with Mars Lander, Episode 1) so easy and Episode 2) so hard in comparison.
3 Likes
You are correct to sense the 10X difficulty increase. For EP3 there is another 10X increase again. Great fun ahead.
3 Likes
I write this just to let you know that this level is just about connecting tubes , no need to care about room’s type…
this is boring and should be mentionned in the description… spend so much time for no reason on this …
The structure of the tunnel hasn’t changed. To review the types of rooms, click here.
Link implied to be under “here” is absent…
1 Like
I finally solve this one thanks to your comment ! I’ve been having the same issue 
Hello! This puzzle has been around quite for a while, but I am somehow new to codingame and I have just started it yesterday.
I have a problem with the validation. I have had them before in other puzzles and there is always a reason, but I cannot find it with this one. I pass the “rocks” test without problems, but the validators give me error. It just happens in step three, with no rocks around that I can see, and I do not have information on why it is chrashing.
Could anybody give me a hand? why is it failing?
Thanks to all
Not sure but maybe you tried to rotate a cell with rock in it.
I will double check that, thanks!. It is a possibility though not having anything hardcoded I cannot understand why it would happen since the validation and test do not seem so different.
Found, thanks… It had to do with rocks and is there where I started looking, but it was an “out of bounds” error when finding the downpath of the rocks. :-))
1 Like
Hi all,
I find this puzzle quite interesting but a little frustrating because it lacks a way of debugging when validating a submitted code.
Initially I have completed 100% tests when running from IDE. And it works for validation tests without rocks as well. It actually works for one with rocks too.
BUT!
The very last test ironically crashes at the very last step of the game (just when Indy is about to leave through the exit). It’s odd as by this moment I have no rooms to rotate: the pathway is ready and all the rocks are blocked except the one chasing Indy from the beginning.
I was able to beat the validation with 100% using ugly hardcoded plug that prevents trying to deal with rocks after step 10 (I know it’s safe for all tests). This plug is only needed for validation as it works like a charm in the IDE without it.
Still I just want to know what is wrong with the function dealing with rocks (eliminateRock). It’s hard to do so as I can see no info from console whatsoever.
Here is my solution in Javascript:
If anybody feels like helping to debug it in blind mode I’d really appreciate this.
Thanks!
Hi there,
I have some problems with the puzzle. Currently I do a depth first search and the problem is that I get a timeout at the 4th test (Broken Mausoleum) with is quite a big level.
For the next states I look, if I have to rotate the next room when it can be rotated to not crash against a wall.
Else I just rotate all the rooms below me and try them all out. I also make sure that I don’t rotate a room back to it’s original rotation because that would be useless.
With all the possibilities I need to go through it’s clear that I get a timeout. How can I reduce the amount of next states? Or is my approach flawed alltogether?
maybe this article will help:
Thank you. Yeah this is very helpful.
Would a greedy method work for this? That is, find the first “problem” that Indy will encounter and rotate it so that he doesn’t do it anymore?