Jan 21, 2024
Rogues of Europa - greyhound
Sorry for the missed week last week. I was very ill and not in any state to work or even type out a devlog. However I have recovered fully and will resume pushing out weekly devlog once again.
This couple of weeks was spent working on the mad scientist towers, particularly implementing the physical spaces the puzzles would occupy.
My initial instincts for making towers was just a big square. Plenty of room for activities, so to speak, reasonably sized, nice and utilitarian.
Like father, like son.
I might put some mobs inside the rooms, or loot drops or something. But the real meat and potatoes are obviously going to be the puzzles.
Much like in monster lairs, mad scientist towers will have a randomly generated set of encounters on each floor. This will probably either be two puzzles which must be solved to advance (one door puzzle and one floor puzzle) or some monsters and a door puzzle.
As I made my building, which has a relatively complex interior, I finally got absolutely fed up with the previous way I had been doing pathfinding in the overworld. I fixed up an extremely inefficient cost function and implemented A* instead of the previous greedy algorithm. Now everything uses A*. If you're an indie developer out there, write this down: do things the right way, not the easy way. If you just play the game, you will likely see some slight pathing changes which are guaranteed to be the most efficient route, making your moves take slightly less time. The previous greedy algorithm could enter infinite loops, which happened more frequently in challenging geometry, but the new A* algorithm is guaranteed to terminate.
I then started making the actual levels of the tower. I'm planning on having a more open plan than the ground floor (the ground floor itself may change). From the player perspective, the tower is fundamentally the same as a lair. It only goes up or down in your imagination. With that in mind, I just copied existing code, and after a few tweaks, got tower levels to generate.
Turns out mob generation was still turned on. From the program's perspective, the tower is merely a special case of lair with only one room and no paths.
I fixed the mob generation issue and got on to creating an actual level for the tower. My plans for the towers are to have a primary room or set of rooms (decided as an encounter) and a secondary room which leads to the stairs up. This room will be blocked off with a locked door.
The clock will be ticking as you solve puzzles which means that monsters will encroach on you as you solve them. It might be a good idea to fight them off first. Any damage taken while in a puzzle will automatically end the puzzle.
My goal is to have the towers wrapped up and a good update out by the end of January, so keep on the lookout for that. Remember to email any questions, comments, and prayer requests you have to roguesofeuropa@protonmail.com.
This couple of weeks was spent working on the mad scientist towers, particularly implementing the physical spaces the puzzles would occupy.
My initial instincts for making towers was just a big square. Plenty of room for activities, so to speak, reasonably sized, nice and utilitarian.
Like father, like son.
I might put some mobs inside the rooms, or loot drops or something. But the real meat and potatoes are obviously going to be the puzzles.
Much like in monster lairs, mad scientist towers will have a randomly generated set of encounters on each floor. This will probably either be two puzzles which must be solved to advance (one door puzzle and one floor puzzle) or some monsters and a door puzzle.
As I made my building, which has a relatively complex interior, I finally got absolutely fed up with the previous way I had been doing pathfinding in the overworld. I fixed up an extremely inefficient cost function and implemented A* instead of the previous greedy algorithm. Now everything uses A*. If you're an indie developer out there, write this down: do things the right way, not the easy way. If you just play the game, you will likely see some slight pathing changes which are guaranteed to be the most efficient route, making your moves take slightly less time. The previous greedy algorithm could enter infinite loops, which happened more frequently in challenging geometry, but the new A* algorithm is guaranteed to terminate.
I then started making the actual levels of the tower. I'm planning on having a more open plan than the ground floor (the ground floor itself may change). From the player perspective, the tower is fundamentally the same as a lair. It only goes up or down in your imagination. With that in mind, I just copied existing code, and after a few tweaks, got tower levels to generate.
Turns out mob generation was still turned on. From the program's perspective, the tower is merely a special case of lair with only one room and no paths.
I fixed the mob generation issue and got on to creating an actual level for the tower. My plans for the towers are to have a primary room or set of rooms (decided as an encounter) and a secondary room which leads to the stairs up. This room will be blocked off with a locked door.
The clock will be ticking as you solve puzzles which means that monsters will encroach on you as you solve them. It might be a good idea to fight them off first. Any damage taken while in a puzzle will automatically end the puzzle.
My goal is to have the towers wrapped up and a good update out by the end of January, so keep on the lookout for that. Remember to email any questions, comments, and prayer requests you have to roguesofeuropa@protonmail.com.
Share: 
