*Everyone* will be infinitely doomed if you make a mistake, so at least we’ll all be in the same boat.

@@anteshell what

@@user_hat i kinda understand what they're saying, but also can't

@@anteshell I think they meant the dream was recurring, not the shape.

@@anteshell are you trolling?

@@anteshell The tube is infinitely long, and the nightmare kept showing up. Recurrent nightmare, tube going on forever

I know what you mean; the other ones felt like hard logic puzzles, while this one feels like an exercise in recursion (and, more specifically, the recursive-based puzzles).

My mind was bended. Goddamn it was funky.

I don't know, I'm on level 14 of infinisweeper and the recursions have gotten so much worse, as in a recursion with mulitiple outer border recursions, and other wacky stuff.

It's like Understand all over again! 😆

I love watching him do it, it’s quite satisfying.

R/verysad

I prefer when Tyler understands a game better than I do. Because if I'm watching a let's play of a game that I understand more than the person I'm watching, I get a ton of facepalm moments, which is not enjoyable for me.

@@WanderTheNomad yea, that was his Celeste play through for me, love that game to death.

Seems like you’re a man of culture

True

@@theendofthestart8179 well it used to be everyday but now since he is moving? (I think?) it is going to be once every 2 days for a month

You might like Patrick's parabox. He played it a while ago

The "no mine on first click" was actually added later. But you're mostly correct

It's only a thing from windows 7's minesweeper and up, In this game I only made sure that the first click isn't a mine

@@youtubehandlesux oh dang, I've seen game devs comment on Tyler's videos before, but this is the first time I've seen one that was already a fan of Tyler, or at least watched some of his videos before

@@youtubehandlesux It's from before windows 7...at least if you select custom difficulty. I know because I cranked the mines up so high trying to find the safe sectors was like playing keno crossed with russian roulet, but the first click was always still not a mine.

@@WanderTheNomad I'm a kinda OG fan, was here since his poly bridge 1 videos, didn't comment much on the videos though

Tyler's minesweeper and minesweeper-like games series never really last that long anyways

minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of minesweeper inside of mineswepper...

I think for the starting spot you want to take a spot with the fewest amount of adjacent cells. In the original minesweeper, that would be an edge and I think it mostly applies here too. Since edge cells are adjacent to bigger cells, there is less space for more cells. EDIT: Just want to say that this logic doesn't fully apply to minesweeper games that start with a 0, there is still optimal starting spots but they are different.

Pattern recognition to be more precise

choosing an embedding/visualisation of that graph that makes sense to the player would be a challenge (considering the number of connections shown in some of the puzzles in this video)

you could represent minesweeper as a vertex colouring or as a weighted graph (AKA an edge colouring) in the vertex colouring vertices are either A or B, if B it's a mine, if A it's not (this is the most 'normalised' form of the graph) or you could also colour the vertices by a number that is the number of adjacent mines, but keeping this updated if you change the graph requires searching the graph (a 'denormalised' or 'cached' form of the graph). for a weighted graph you could weigh each edge with special values as to say if it's between a mine and a non-mine, or otherwise. this doesn't determine where the mines are but it does mean if the player chooses a vertex you can force this to be a non-mine and then generate the graph from there (maybe there are problems with this?). you could also have more values for the weights to represent an edge between two mines, or an edge between two non-mines, or a wildcard for any pairing, etc. or you could implement it many other ways but I'm trying to think of ones that don't add much extra structure compared to a traditional graph.

after some fiddling on my whiteboard I found a directed graph representation that has an interesting topology and simple properties of it allow you to calculate the number of adjacent mines without actually tracking the mines or this number. let's start with an undirected graph that gives us the underlying topology that we want to base our minesweeper on, and turn this into a directed graph where every edge in the original graph is bidirectional in our directed graph. we should suppose our original graph is connected since otherwise it would effectively have multiple separate minesweeper boards within it. this method will not work if we start with a general directed graph, but I'm not sure what that would even represent in terms of a usual understanding of minesweeper, in this case the normalised or cached way still works (and in either case is probably still the best computational choice). then let's decide some of the vertices are mines, there should be at least one mine and at least one non-mine. to make a vertex a mine, reduce its in-degree to 0. you could do this by looping over all the directed edges and removing those that end in a mine. or we could instead reduce the out-degree of mines to 0, which if we represented the graph as an adjacency list could be done by removing the mines from the indices of the list, but this will give us a topology with a different meaning. for either choice we could also decide to keep any mine to mine edges, so that if we permissively converted the directed graph to an undirected graph we would get the original graph, this would not reduce the respective degrees to 0 but the following formula will be unchanged since the mine to mine edges are bidirectional. once we have done this, for every vertex v we can define its 'mine count' as M(v) = deg+(v) - deg-(v) where deg+ is the in-degree, the number of edges directed into v, and deg- is the out-degree, the number of edges directed out of v (or if we chose to reduce the out-degree of mines instead of the in-degree, negate this formula so it is instead M(v) = deg-(v) - deg+(v)) for all non-mines this is actually the number of adjacent mines. an explanation is that for each directed edge between a mine and a non-mine, when we remove that edge that decreases the negative term of the mine count of the non-mine by 1, so it increases the mine count by 1, so if there were n mines adjacent to that non-mine, the mine count will be increased by n, and the mine count began at 0 because each edge began bidirectional, so the mine count is 0 + n which is n. if we try to use this same formula for mines it just gives us the negative number of adjacent non-mines, but minesweeper never needs to know anything about what's adjacent to a particular mine, so this is fine and we can even leave M undefined for mines. if we removed mine to mine edges we can use this to determine if a vertex is a mine, since if it is not isolated it will have a negative 'mine count' unless it is isolated in the new graph which case it must be a mine since every non-mine was not isolated in the original graph cannot be made isolated in the new graph (with isolated in a directed graph meaning having both 0 in-degree and 0 out-degree). so we could forget which vertices are a mine now and still be able to retrieve that information. but if we kept mine to mine edges then it is possible for a mine to have a 'mine count' of 0 but not be isolated, the same way it is possible for non-mines. this makes it harder to tell if non-isolated vertices are mines, we would have to search the connected component in the strictly converted undirected graph of our directed graph, since all the vertices in such a component will be the same type (mine vs non-mine), for any vertex that has a monodirectional edge in the directed graph, since one must exist as the original graph is connected and not all the same type, and the direction of that edge (based on our choice of which degree to reduce) determines the type. as for the topology, non-mines that are connected through the directed subgraph induced by only non-mines of the original undirected graph are also connected in this directed graph, if we decided to keep mine to mine edges then that also means connected mines in the only mine subgraph remain connected in the directed graph as well. if we removed mine to mine edges and reduced the in-degree of mines, then any path in this directed graph ends on a non-mine, so it represents all the ways to move that apart from your starting position avoids all mines, so it prevents you from moving from a non-mine to a mine, that keeps you 'safe' from mines. if we kept mine to mine edges and reduced the in-degree of mines, then each monodirectional edge always takes you from an island of mines to an island of non-mines, making non-mine vertex sets disconnected to any mines, keeping the 'safe' meaning. if we still removed mine to mine edges and reduced the out-degree of mines, then any path that visits a mine must end on that mine, that works well with the idea that the mines are deadly if encountered since you would be 'trapped' there. if we kept mine to mine edges and reduced the out-degree of mines, then each monodirectional edge always takes you from an island of non-mines to an island of mines, making mine vertex sets disconnected to any non-mines, keeping the 'trapped' meaning.

@@MagicGonads Oh almost missed this. I started some implementation steps for it (nothing to show yet really). I decided to go with what amounts to vertex coloring for the representation of the graph in memory. The player will not just see X or O for mine or clear of course though, they will see the number of mines among the neighboring cells (/vertices) if the cell is uncovered. I will calculate that number on the fly. I plan to add some quirks later on like duplicate edges between vertices, weighted edges which affect the displayed mine count (so a duplicate edge would essentially be the same as weight 2), and directed edges which are all things a regular game of minesweeper usually doesn't have of course.