In grid, dijkstra IS bfs, because distance between every grid cell is 1. When you have distances though, dijkstra is going to find the optimal path and bfs is just going to find the path with the smallest number of road

Pretty major flaws in the comparison. The red dot always being in a corner mitigates some of the flaws of poor path finding algorithms. Also, having only 1 possible path eliminates some flaws of poor pathfinding algorithms.

This was amazing, dfs performance was a surprise too

Green means stop, Red means go, got it. Also, this irrationally infuriates me.

Thank you

All one take, eh?

would be cool to see a version of this with more relevant modern algorithms like ANYA, Block A*, Sub-TL, Lazy Theta* w/ Optimizations, etc.

The modern sequel is a pay2win mobile game garbage, it's really sad and disheartening to see one of my favorite childhood IP being violated like that.

Playing Metal Slug in an arcade cabinet in Mexico will always be one of my fondest memories. :D

I bet that Japanese arcade doubled in profit that day that we played cause I died every 3 seconds

Sorry for the later half of the video but I had a bit of a busy month with some IRL stuff and wanted to still post this month, but I should be good and back with a slightly changed up style of video for April

There is also a much simpler and faster algorithm which was used in early RTS games like Dune 2, Warcraft and Doom. It involved casting a ray into the direction of the goal, and if the ray hits an obstacle, they the algorithm either uses a right-hand side rule to go around it or shots a ray into random direction. It isn't guarantee to find the shortest path, but has constant memory usage and usually finds the path faster. You can also run it in parallel by shooting several rays. It also works when you have to navigate an actual drone without a pre-made map.

The heuristic used in A* doesn't necessarily need to be the Manhattan distance. It works well with regular grid-placed nodes but in real life geometric scenarios the Euclidean distance would be the preferred choice.

One of the games I hold with the highest regard in terms of both fun and variety, really hoping that Path of Exile 2 lives up to the same hype as it's predecessor!

Add loops to the maze!

i wonder like are there diffrent pathfinding algorythms for multi dimensonal "roads"? like you could say as long as one block is all around move forward?

This is such a good video and such a relatable topic. I used to love seeing box art of all the games my mother and father bought when they would go out. Thanks for doing this video.😊😊😊

The last video essay of the year! I'll be back next year with some more videos like these, but in the mean time I'll be uploading a recap of this year coming sometime in December

Great video but 3:11 “Hueristic is Manhanntan” triggers me. 😂

Too bad you didn't try bidirectional search. That is the only one I thought would have some chance of beating A*

3:44 "on average" cool cool, but where's the average?

good vid man :)

A second video in more a video review fashion this time talking about a more popular game that I've been playing a lot recently. I've been getting a lot more used to the format/style of these types of videos so hopefully better edited ones coming soon! Two days after this Valve added back community maps so nice

Bro, its very hard to understand what youre saying. Try to speak slower, not "swallowing" words.

Could you make a video on heuristics soon?

Dijkstra is bad for graphs with unweighted edges.

i like your editing style it’s very pleasant and chill

don't use top-left use bottom right instead and make texts a bit smaller (and you dont have to write the developers)

Dusting off a game I've had on the backlog for a while, hoping to review some more games down the line too

Testing out a new style/editing aesthetic for this vid, new vid out in October with something more normal

DFS and Dijkstra benefit is 1 cell is visited only 1 time. Your mazes doesn't present "cycles" and it gives DFS a clean benefit (because your maze "forces" DFS to visit 1 cell only 1 time... something really unrealistic. A symthom of this problem is DFS finds the "shortest" path in your mazes (because there is only 1 possible path between the 2 nodes)

Since he doesn't mention this, DFS stands for depth-first search and BFS is for breadth-first search. An added benefit of the heuristic with A* can be seen in pathfinding of video games. Due to various terrains, the shortest path might not be the fastest. (accounting for elevation changes or perhaps a glue pit, things that might slow down or speed up character movement) Also, you talk a bit too fast like a nervous kid giving a speech in front of his class.

How you visualize this algo ? You make this video with programming language or a video edittor ?

am i the only one who gets confused by the inverted color scheme of the maze

Would you mind If I use your code to help me with a project?

Here's my naive algo: move in a straight line until you can't anymore. Then move randomly for a while and try again. I totally am not using this is to make an rts game 😂

almost all labyrinths have near to straight path solution, but in real maze you could go for example right on full width and then top on half height and then left on full width etc., or like a spiral or something more difficult, i suppose some of the algorithms could be better without heuristic.

I do realize that HON had a massive impact on the development of MOBAs (esp before Dota 2 was announced). But I chose not to include it since I didn't have that much exposure to it 😔

Nice! I think though that putting starting point in the corner of the maize is not ideal.

Untal di pangan ben kaya terus jek ngemis.

Iki dolanan opo rak mudeng aku

Skyrim: Straight line to objective. If there's anything in the way, walk in a circle until the player goes away.

kzbin.info/www/bejne/mqe4inWIf5Znbrs

I wonder how the algorithms would fare if you flipped all of those mazes on the upper left to lower right diagonal, and had the start in the top right cell. I get the impression that they were all designed for a "general case" of having the start in the lower left of a maze.

A and B could be very close to each other while the linking path could be extreme winded and long. So distance of A and B does not matter at all. Only if there is more than one linking path considering the distance should be taken into account.

Sorry, but I was unable to listen to you, due to your speed of talking, the unwillingness to seperate the words and to be understood.

Isn't the biggest advantage of A* the fact that you have a cost function? The better the cost function the better is A*. On big maps A* is usually combined with precomputed region data anyways... right?

Scam

Try right hand rule through the maze where you put your hand on your right wall and walk where it reaches to