Getting Ready for Episode 1

 I got a chance to test out the ARM build on a raspberry pi... It didn’t go well. I was able to get a very old build to run, but that version is not really playable. I will come back to this issue and try to do more testing in the future, but for now I want to focus on getting the win-x86 build done. Once its done it should be easier to troubleshoot issues on other architectures.

I put a bunch of work in on the first 4 levels. They just need some clean up and any map state triggers and start variables need to be added still. I worked a bit more on the sprites and textures as well.

Map state triggers are new and work kinda like (button) links, except instead of being activated by a button they are activated by the state of the map. Map states include things like “all buttons on”, “all buttons off”, “all enemies dead” and things like that.

I added trackers for all the map states and actions that can be called by the tracker. The states are available at level completion. To save cycles they are not actively tracked during play unless otherwise noted in the level file. The actions are things like “end the level”, “toggle lock”, “spawn item” and stuff like that.

Since I am tracking kills now I added stats to the level end screen. Kills and time are the only stats I am displaying. I’m not going to display “secrets found”.

The map file structure has also been expanded to include some flags and starting variables that are loaded and applied during the map loading method. This can be used to clear the players inventory at the start of a map, set the players position, and stuff like that.

I fixed the issue where doors could lose synchronization with buttons if the door is in the ‘open door buffer’ while the button is toggled.

I made my own random number method. Its not at all random. I picked a prime number around the 4000s and generated that many numbers and saved that to a table. I had to use the prime because I could see the repeating cycle of numbers on the screen melt screens as that uses lots of sequential calls to the random function. I wanted to use a table of numbers so I can play around with recording replay demos later.

I think that is most of recent updates. Right now I’m primarily working toward getting the features I need to get the first 10 levels working.