You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+4Lines changed: 4 additions & 0 deletions
Original file line number
Diff line number
Diff line change
@@ -225,6 +225,10 @@ Once I implemented my third attempt at part 2, the only stumbling block was that
225
225
226
226
[On day 30] As tends to go, I had inspiration last night after I had shut everything down about how to fix this. I don't have time to work on it at the time of this writing, but I am committing the work I've done so far, even though I'm going to largely rewrite everything. One, it's bordering on unhinged, and I want to save it. :) Two...who knows, it might even come in handy later! I have seen part 2 yet, so maybe I'll get to make use of something unhinged after all. It's not a terrible commit. At least it's better than the version I was bughunting before where it turned out my numeric keypad didn't have a "1 2 3" row. :D
227
227
228
+
[A little later] It occurs to me that maybe I *can* salvage what I already wrote and just make it do less with a flag or something. Maybe I don't need a whole rewrite.
229
+
230
+
[A little later] GOD DAMN IT the algorithm wasn't the problem! I mean...it was, and still is, but it's not that the puzzle took up too much RAM, it's that I **added an extra keypad to the chain**. So...with the correct number of robots (and therefore keypads), my solution actually executes and looks like it's producing a result in the right neighborhood, but it's still wrong. It looks like whatever I was cooking up in there isn't passing test data. So *once again*, I am going to commit for posterity and then basically start over. I could halfway justify trying to salvage the unhinged implementation if it meant I could just shortcut the performance problems and get the right answer. But...I shortcut the performance problems and got the _wrong_ answer. Between that and the knowledge that the thing is already wildly overengineered for the puzzle ask, I might as well start from scratch. Sadly, I can't do it as I'm writing this, but I hope to give it another shot in the afternoon.
231
+
228
232
## Day 22: Monkey Market
229
233
230
234
I wasn't even going to start a new puzzle today, but the description for part 1 seemed so straightforward that I assumed either part 2 would be way complicated, or else I was fooling myself and part 1 would take forever with a naive implementation. I rolled the dice, though, and got part 1 relatively quickly. Then, of course, part 2 is *way* more complicated, so I'm not going to try that tonight. Until next time, monkeys!
0 commit comments