[C#] Maze Solver

[DeathArt]

Ok, I made this a while ago, but seeing alot of C# interest, I figured some people might be interested.

Basicly, what I've done is implemented the Breadth-first Search (BFS) algorithm along with a image-analyser to solve a given maze.

Here's a snippet from wikipedia, explaining about the algorithm.

In graph theory, breadth-first search (BFS) is a graph search algorithm that begins at the root node and explores all the neighboring nodes. Then for each of those nearest nodes, it explores their unexplored neighbor nodes, and so on, until it finds the goal.

Download/Test: http://web27190.web08.talkactive.net/archive/BFS2.exe
- Note: Requires .NET Framework 3.x

Screenshots:

Test with numbers:



Test without numbers:


Source
- http://www.dragons-lair.org/reposito...h-First_Search

[MentaL]

[Sequence]

Just downloaded and tried it, it's really nice ;)

If you want to extend it, then you should add a sort of project files, so it would be able to open a file and then do the BFS thing.

[Daevius]

Nice one. The second image looks like one of the things I did too:
http://www.gamedev.net/reference/art...rticle2003.asp

This one included diagonal paths aswell, I tried to build it. It's very complicated (in C++), it sort of worked but not fully, so I'm using a library now for it...wondering which method is better or faster.

Nice one.

[PixelPro]

Nice release. I think it looks good, I tried it out myself. + Damn, it requires the .NET Framework. My laptop doesn't have it installed.

[DeathArt]

Nice one. The second image looks like one of the things I did too:
http://www.gamedev.net/reference/art...rticle2003.asp

This one included diagonal paths aswell, I tried to build it. It's very complicated (in C++), it sort of worked but not fully, so I'm using a library now for it...wondering which method is better or faster.

Nice one.

The code for diagonal paths are actually in my source, it's just commented out ;-)

[Aurora]

Question, isnt this just based on A*plus pathfinding ?

[DeathArt]

Question, isnt this just based on A*plus pathfinding ?

Yes, it is

The BFS algorithm is one of the most common used in Pathfinding. Someone else been at University eh? ;-)

[Daevius]

I've been doing A*...but whats the difference between A* and BFS?

[DeathArt]

A* is more complex, or that's actually wrong. BFS is more simple, since it's pratical a special version of A* (simplified).

I'm not expert on the subject, and the only paper I had on pathfinding was regarding BFS.

[Daevius]

I've tried to make A* in C++, took me several weeks...it _worked_, but nothing more. It took longer paths or other weird things. Really difficult to make it work good and efficient, so I took an existing library xD.

[DeathArt]

Took a evening to write this application :p C# is SO much easier to create things in than C++ (and I'm not reffering to UI stuff here)

[Aurora]

I agree, c# rox.

I made alot of tools/game stuff/networking and emulator in C# so far, like

Patch creator (byte comparison)
Virtual file system (pak files) /w encryption
Map editor
Client software (directx)
Packet loggers
Network library TCP/IP

ect ect, its just keep amazing me how much C# rox. :)

[Daevius]

Took a evening to write this application :p C# is SO much easier to create things in than C++ (and I'm not reffering to UI stuff here)

I guess BFS is easier somehow, I used classes in C++ just like in C#, so it can't be much different. All functions had logic coding, not functions that a language could replace somehow...

[DeathArt]

I guess BFS is easier somehow, I used classes in C++ just like in C#, so it can't be much different. All functions had logic coding, not functions that a language could replace somehow...

C++ don't have properties, nor foreach iterations.
Not to mention, I could likely optimize it with LINQ.

Thus C++ is much more complex to code a final solution in. Including that writing the image analyser, would take half a day, compared to about 30 lines in C# :-)

[Daevius]

Hehe, yeah, maybe thats true :)