Day 125: Today it’s early. I just wanted to stop early, and since I’m my own boss, I gave myself the permission. So, all that’s keeping me from my week end is this block of text. You (and by that I mean I) may wonder why I’m still writing this every day. It’s simple, it’s to prove my resilience. I don’t like quitting. And I think that forcing myself to do this every day without ever missing once is a good exercise. I know that it’s unlikely anyone will ever read it all. Hell it’s unlikely anyone will ever read today’s entry. But that’s okay.

Daily interesting stuff:

• Reddit post, “how to get beyond the hump?. I clicked because I had no idea what a “hump” might be. It’s just the point where the person want to quit. The more I see things like that, the more I wonder if it’s really that hard to finish a game. Based on the notion that a game is finished when you have a complete flow, I already finished around 10 games. I can’t say it was easy, but it wasn’t horribly hard too. I had time dedicated to that and I just worked until it was over. Now, making a commercially viable game, that’s another story. I think the most value I would pay for one of my game is 2$. And that’s a far stretch.

• Reddit post, how do you choose which game to make? It might be an interesting post. It had a lot of really long comment. But I was feeling too lazy to read them.

Tasks done:

• Fixed the freaking room exit automation. I feel like working on the Tiled importer is all that I’ve done this week. And it’s probably true too. But now it should work. I couldn’t test it with a lot of room since the mesh changes based on the tileset used, and that the new tileset isn’t ready to be used for the rooms. So for all I know my code could still be broken… Hey, I still have a lot of space left for today’s entry. So I should write some more sentences on that, to make you believe that I actually did some productive things today. Okay, so to fix this problem, I actually had to completely change how I find the exits. Before, I used a vertices based detection. I would check the position of a vertex and see if it was on the border and if it had a “jump” from the previous coordinates. A “jump” of 6 units indicated an exit. But there were a problem. The vertices can be in any order. So I can’t really detect a “jump”. So what did I do? I went back to the method I used before. I analyzed the colliders to see if there were some disparities indicating exits. I previously steered away from this method because it had problems to detect vertical exits and other weird things. But it turned out it was easy to fix. Instead of dividing colliders into horizontal ones and vertical ones, I just compared each collider to every other. Then with some condition I detect if both colliders are on the border and if there’s a coordinate difference of 6. It’s not the most beautiful code I ever wrote but at least it should work. Oh and if you’re thinking that a O(n)² solution is too slow, then you’re wrong. It’s an import tool that run once in the editor. And the whole code for collider detection and exit detection takes about 5ms.

• Did 76 rooms. I know, I know, it seemed like I was super productive and prolific. It sounds like I’m awesome and that I did in one day almost all the room content. HAHAHAHAHA, no. I did 76 rooms… on paper. I took a sheet and draw schematics for rooms. It’s much faster and should help me a lot while actually doing the rooms in Tiled. I also played Rogue Legacy and Entrer the Gungeon to see how they made rooms. One thing that I didn’t think of was that they do everything they can to break the “rectangle” aspect. And it can be as simple as having “bumps” in the outer walls. That gave me a boost of inspiration. Because I was running in circles with just some obstacles inside rooms.