The C64 version of Flooder (
https://josipretrobits.itch.io/flooder) is a game I've been working on-and-off on for about a year and I've decided that I think I'm happy to finally quit working on it so thought I'd dump everything into it here. All files I have used will be linked at the bottom
To give a little information on the game: it functions similarly to the GB version in the sense that the goal is the same and the idea behind the game is the same. It is a board that needs to be made one colour by changing the top left pixel to that of surrounding pixels which can then change adjacent pixels to the growing 'flood'. This game is structured differently, as it contains levels with an increasing number of colours, rather than just being one board that can be played again and again. Its flood mechanic is also different, as it visually floods the board one pixel at a time rather than being instantaneous, and this can make a difference when it comes to finding the fastest solution.
As for finding the fastest solutions for each board, the first step is finding the optimal boards. Unlike the GB version where its board configurations are preset, this game's boards can be manipulated using inputs, so my method has been just running a script to input randomly as the board forms and hope it gives a good board.
A good board can usually be spotted fairly easily after running through enough possibilities, though a method of solving the board needs to be found to deal with any suspicions. After running through a number of potential methods, I eventually settled on writing a script that would take the board configuration handily formed in a useable format through another script and then find the shortest possible solutions. As there are usually multiple solutions of the same amount of moves, it shows every possible solution. This has worked well up until about the level the .bk2 file linked gets too. It still works, it just takes far too long on my machine and with the game already running slowly, this is why I decided to call it quits. It was becoming a process that was far too long and tedious to enjoy working on.
Having multiple possible solutions also meant that I could run through all of them and find the actual fastest, as how the board floods is also a factor determining the speed the level can be solved in. In earlier levels, I would try to find multiple boards that each had the same lowest number of moves and find the fastest there but I cannot guarantee they are the absolute fastest solutions. What I can guarantee is that all but one of the levels (probably) have the lowest amount of moves possible. The recurring pattern was that with an increase of one colour each level, the amount of moves required also went up by one. As mentioned, all but one level followed this rule, with this level requiring two more moves than the former. This is the level I spent the most amount of time on so I eventually just gave in with it.
I think letting go is going to feel like a massive relief and so I'm going to leave everything here should anyone ever wish to continue the project in the future.
.bk2:
https://tasvideos.org/UserFiles/Info/638723761875212228
Watch file:
https://tasvideos.org/UserFiles/Info/638723762797864834
Board reader:
https://tasvideos.org/UserFiles/Info/638723763920964939
Board manip script:
https://tasvideos.org/UserFiles/Info/638723764524161960
Solver script:
https://tasvideos.org/UserFiles/Info/638723765819061460
