day21 solution, boolean/assembly logic is weird

2026-05-18 19:13:27 +02:00 · 2020-08-08 12:51:58 -04:00
parent 4457349b8e
commit 0a3e7f8fab
5 changed files with 574 additions and 0 deletions
@@ -27,3 +27,4 @@ Day | Name | Type of Algo & Notes
 18 | Many-Worlds Interpretation | - Oof, this is the hardest algo so far I think... Combination of __Best Pathfinding Algo__ (going to use a BFS/Dijkstra's) and a weighted __Graph Traversal__ with "locked" edges <br> - I toyed with the idea of using some kind of bitmap to track keys that were found, but ditched it so I wouldn't have to write my own wrapper around bitwise logic... Time complexity comparisons are roughly similar for comparing two maps as comparing two sets of bits, the space complexity is probably much better for 27-bits (@ and a-z) compared to a hashmap, but OH WELL. <br> - The graph traversal required memoizing in order to run in any reasonable amount of time. <br> - My part 2 assumptions would not work for every example which is unfortunate :/ I assumed that within a particular quadrant, if the quadrant does not have the key for a door, just remove the door.
 19 | Tractor Beam | - Another Intcode outputting to a 2D grid <br> - __THERE NEEDS TO BE A CONVENTION FOR X AND Y WHEN WORKING WITH 2D GRIDS__ This is awful, should it be mathematical? Should it represent rows and columns? Who knows?! <br> - Little geometry (drawing squares) for part2 but nothing crazy
 20 | Donut Maze | __Breadth first search__ path finding algo. Dijkstra's Algorithm to find the shortest path given a maze with "portals." <br> Part 2 is kind of wild with the maze become 3D, but the solution is effectively the same, just more complex to implement <br> - I remapped this in my head to more of a 3D cube instead of a donut... <br> - That was really cool to get working... Dare I say fun...
+21 | Springdroid Adventure | Another simple-ish Intcode based problem, this time using some weird __Assembly language__ that gets inputs via writing ASCII values to the Intcode computer.