#towntaster

The past two weeks have been filled with sickness, being out of town, and generally being too sick and busy to really do much work. There was no devlog last week due to the fact that I was out of town. So what did get done? A little bit, small things I've been meaning to do for a while. I finally added the 4 player server cap(which I may make alterable, I might not). I also reduced the number of threads used for client server communication. Gameplay wise, nothing got done. I have no screenshots to show, and nothing exciting to talk about. Not much work got done, and that's ok. Making sure to take care of yourself is key in maintaining motivation on a project, and I've gotten some good rest. I have 4 days off from work this week(including today), and I plan on using them to rest from my everday life and get a little game development done. Next week I'll have something more exciting lined up, hopefully. I'm not sure what I'll work on, really it'll be whatever I decide to jump into at the time.

World generation, and not much else Getting the world to generate in the structured way I envisioned was about all I could muster this week, even though I wanted to work on ambiance as well, like saved lighting arrangements per room, flickering lights, well lit rooms, not so well lit rooms, etc. But I am happy with how the world generation turned out. it just needs a few more tweaks and a couple more additions. What you're actually looking at in the screenshots is as follows: Top: in-game map of a partially(well, considering it's infinite, it'll always be partial) explored area, where the structure of tunnels and sectors becomes more apparent Middle row: Not much to do with what I did this week, but I wanted to make up for my lack of apparent progress. These are screenshots of the game at a higher resolution, so you can see more details. Bottom row: Visualizations of the data that gets stored by the game of a generated region. Yellow is roads, and the other colors are sectors. The sectors will not appear perfectly square, as its still random the shape WITHIN the sector. Roads connect to the edges, to be able to enter new regions. So world generations is pretty good now, and it generate pretty fast. it takes a split second for the server to respond after you enter a new region, and that's running on my dual core, so pretty standard hardware. Still, I'll see if I can find any optimizations, or ways to hide that split second from the players. This week, I'm either going to work on ambiance(alongside the world generation it needs), or a bit more on some element of gameplay, perhaps on constructing buildings. I'm probably not going to put too much more polish on world generation for a while though, I'm a little burnt on it. Bonus! I also made a mini-game this week, particularly Saturday. Suggested on an episode of steam train, I liked the idea of a comedic game about a horse on wheels. If for some reason you want to play it, you can check it out here: http://towneater.itch.io/horsepower It uses the same engine as Hardboot, but is much glitchier, because I intentionally left some of the comedic bugs in.

Top: Actual screenshot, in game inventory(Plus a glitchy version of a new way to render text I was working on. I fixed it, but it makes a REALLY nice giberish generator) Bottom: Not a screenshot, a mockup of what the main menu(while in development at least) might look like. I'm not sure if FailureIO will be the final title for the game, nor do I know if that will be the final icon. It's all up in the air. Development status: Working on GUI, HUD, and a bunch of bugfixes. Bugfixes: After sending a prototype to UniverseUsurper, and they could hardly run it, I figured it was time to stop ignoring stability. My main problem stemmed from so many threads trying to interact with the same data at the same time. I changed the networking threads so instead of actually acting on the game's data based on information received, they simply pass what they find along to a buffer, where the main class deals with it in the main tick. I also completely reworked how the engine handles scene structure, and how the renderer interacts with that. Now, the renderer does not look directly at the scene, but requests a "snapshot" of the scene, so if changes are made mid-render, the renderer wont freak out and stall, flicker, or crash. So I cleared up a lot of bugs, and now everything is a lot more stable and clean. GUI and HUD: Some of my least favorite things to do, but they need to get done. I am actually excited for them this time though, as for once I might be doing a decent job. I'm starting to make some breakthroughs on the visual style of the interface, some of which should have been obvious from the start. Mainly the bulk of the work is going to be implementing my ideas, taking them from screenshot mock-ups to actual screenshots. To recap, things are finally starting to come together. Systems for basic engine functionality, basic gameplay functionality, and systems for large amounts of diverse content are coming together in unison. Soon, screenshots should be getting far more interesting, as all the basics will be done, and content creation can go forth!