Day 95: I played around with the NetHack Learning Environment, a new project by facebook's research group (as much as I dislike them as a social media I was intrigued).

My 5 game win streak is over. After ascending a Healer, Valkyrie, Barbarian, Rogue, and Priest all in a row, my Samurai was struck down by a dastardly pair of soldier ants in Minetown. All in all, I'm happy with a 5 win streak since I have never even gotten 2 in a row before. And this puts me in 1st place for 3.7.0 win streak in the tournament! Sayonara johanv the Samurai.

Day 89 (yesterday): I got back into the development of ruegolike, my , by tweaking how things are displayed and the movement keys, and updating the roadmap.

Day 84: My friend and I rendered an animation of the man and... well, I'll let you see for yourself ;)

bonus: I finally fixed the rendering in the ! It was a matter of keeping track of which floor tiles have been discovered within rooms, to decide what a corner should be displayed as. (red tiles are ones that have not been discovered yet, for debugging)

Day 83: We finished another tutorial to create the face with different expressions. Turns out there are 2 more videos after this instead of this being the last one.

We finished most of the guy in and set constraints on the body parts so we can rotate them. There's still one tutorial video left, which is the face and animation

Day 82: I fixed the rendering of some of the rooms in the , but there are still some edge cases, such as when a corner gives away another room. This happens when there is a wall you have seen, but is hidden because the room has not been entered. Then the corner still counts that wall and connects to it, giving away the room.

noooo there are still rendering errors! I could fix this one with a special case, but i can think of a few that would be ambiguous, such as having a corner on both sides of another corner - which way would it go then?

Show thread

Day 81: I started messing with the code for the to try to get the rooms to render properly, but have fewer tiles to keep track of. Then I realized I needed to spend some time thinking through which methods of representing the room segments with tiles was the best. Having corners and walls be stored the same makes it ambiguous sometimes, so I'll have to store horizontal, vertical, and corners (no distinction needed between which type of corner).

Day 80 (yesterday): we made some more progress following the lego man tutorial, here's a render!

Day 79: we started learning by making a man using this tutorial: (not really code but it's for the game)

Day 78: we completed the last tutorial, which was to add coins to the game. once you collect 5 coins it brings you back to the menu screen

Day 77: we got some textures, enemies, and collision detection working in following the tutorial

Day 76: My friend and I are learning the open source game engine using this tutorial: amF1UrjVtccZ Here are some gifs of our progress after just the first 2 parts of the tutorial

bonus: I cleaned up the code, made walls block light, and turned off debug mode so this screenshot is what the player sees

Show thread

Day 75: Getting back into this challenge for the summer by making a in called "ruegolike" I've been working on it for a bit, but today I added generation of corridors between rooms. (I know the is supposed to be continuous, but... I guess I'm going for a piecewise continuous function.) Also yes, it is heavily inspired by , one of my all-time favorite games

got raycasting working! opted to just port the code over to go instead of implementing it myself. it actually uses shadowcasting instead of raycasting but it looks great!

currently trying to find a simple way to do on a 2D binary grid. the black squares are where there is a tile, and the red is the "shape" of the tile based on its neighbors. Based on this page:

Show more

Fosstodon is an English speaking Mastodon instance that is open to anyone who is interested in technology; particularly free & open source software.