Second week: it's starting to look interesting

Hello, another week passed and a lot of progress was made on the Supernova 2 engine. On Monday I started with 4 out of 71 rooms programmed, now I am at around 22 rooms.

There was also a small change in the schedule. Originally I wanted to leave sound to the end, since I think it is the least important part of the engine and it wouldn't be that bad if I didn't manage to add it before the end of Google Summer of Code. Because now I am not that worried that I don't finish the engine in time, I am adding the sound alongside with everything else.

Today I got stuck on pretty interesting problem. I was working on a labyrinth inside a pyramid (it was labyrinth in game but also a labyrinth in code, so I had no idea about the overall layout of the labyrinth). Everything looked quite good, I could walk around and it looked like in the original game. But then I added a special room with puzzle:
and I thought it would be a good idea to actually try if the puzzle works by solving it. So I turned on the game, went to the pyramid and tried to find the puzzle. After like 30 minutes of wandering around with no success I started making a map. So thanks to that after a few more minutes I found the puzzle room. Now I just have to solve it, which is easy, right? It took me like an hour. Not that it was that hard, but while solving, I accidentally swapped 2 similar pieces (third row, second and forth column on the image) and couldn't figure out what was wrong. Overall I spend like 2 hours just doing this. But thanks to that I have pretty detailed map, so next time I won't get lost anymore.
That's all for this week, next week I hopefully manage to finish the pyramid and start working on museum, which is the last part of the game, that will be missing.


  1. If I remember correctly it was possible to get this sliding puzzle in an unsolvable state from the original engine (tiles were just randomly placed).
    So when you only had save games after the this puzzle was already generated you were out of luck.

    Would be great to have this fixed by generating the shuffled one in a proper way!

    1. There might have been a different version of the game, that could have this problem, but in the code I am looking at when writing the engine, the puzzle starting positions are hardcoded, so it is solvable every time. I hardcoded the same starting position into the engine too, so it is also always solvable. Maybe you made a similar mistake as me and switched the 2 pieces (3rd row, 2nd and 4th column on the picture), because they look almost identical and then you can get to a state, that looks like unsolvable.

    2. I likely have to same version: I just tested it and yes, the puzzle seems to start with the same preset order every time.

      Maybe it was because of those two tiles you've mentioned (they are really super similar!) or maybe I was thinking of another game...


Post a Comment

Popular posts from this blog

Mortville Manor TTS

ScummVM speaks