Today I got the AI submitting decisions. Right now they just insert their decisions last, after all human players. My plan was to have them inserted randomly in between humans, but since the order based rooms are going to get deprecated, whats the difference?
Tonight I got the profile switch, weighted option switch, and bot-health all coded. From here on out it should be pretty easy, now that I know the parsing seems to be working ok. Testing is still a problem, it takes a lot of clicking around and code changes (that I just realized I accidentally committed oops) to run a room. My old room testing script wasn't optimal and doesn't support bots; it was pretty bare bones just for user based content creation. I need to write a terminal based script, but the problem is that its a lot of code to write and I'm not sure if its worth it, given this may be my last big push on this project.
Yeah, I guess that's another thing; if this new crop of features (ui, fast join, ai) isn't a hit with some playtests, I don't know how much longer I can work on this. The code base is getting unweildy because it wasn't designed to be as big as it is; when I started this years ago I thought it would be a few php scripts max. Not sure why. I was way off. But I thought I'd quit a long time ago, and yet here I am! I am still having fun with it.
Ok so list of what needs to happen before May 8th:
- finish parsing rules for AI
- rewrite rooms to support AI and a variable number of players. this means getting rid of certain types of rooms (ones that require strict ordering). I also want to take the opportunity to hone in on what rooms were fun (target rooms, consensus rooms) and ax what isn't fun (order rooms, trait rooms).
- testing. lots of testing with the bots and maybe some pre-playtest with a person or two.
- Deployment. game needs to be redeployed. This is fairly easy but probably a night or two of work.
2 hours
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Today I got the AI submitting decisions. Right now they just insert their decisions last, after all human players. My plan was to have them inserted randomly in between humans, but since the order based rooms are going to get deprecated, whats the difference?
Tonight I got the profile switch, weighted option switch, and bot-health all coded. From here on out it should be pretty easy, now that I know the parsing seems to be working ok. Testing is still a problem, it takes a lot of clicking around and code changes (that I just realized I accidentally committed oops) to run a room. My old room testing script wasn't optimal and doesn't support bots; it was pretty bare bones just for user based content creation. I need to write a terminal based script, but the problem is that its a lot of code to write and I'm not sure if its worth it, given this may be my last big push on this project.
Yeah, I guess that's another thing; if this new crop of features (ui, fast join, ai) isn't a hit with some playtests, I don't know how much longer I can work on this. The code base is getting unweildy because it wasn't designed to be as big as it is; when I started this years ago I thought it would be a few php scripts max. Not sure why. I was way off. But I thought I'd quit a long time ago, and yet here I am! I am still having fun with it.
Ok so list of what needs to happen before May 8th:
- finish parsing rules for AI
- rewrite rooms to support AI and a variable number of players. this means getting rid of certain types of rooms (ones that require strict ordering). I also want to take the opportunity to hone in on what rooms were fun (target rooms, consensus rooms) and ax what isn't fun (order rooms, trait rooms).
- testing. lots of testing with the bots and maybe some pre-playtest with a person or two.
- Deployment. game needs to be redeployed. This is fairly easy but probably a night or two of work.
2 hours