Some progress and some more complaining

pmdevita on 2017/02/25 05:02 of Clarissa/Peter Team

We had a brainstorming session and have come up with a game idea.

The game starts with a man walking down the street. Suddenly, an evil scientist flies in and takes either on of his arms or legs - the choice is made by the player and it is the lesser of two evils tie in. Should you sacrifice attack power or mobility? The man then goes on a quest to retrieve his missing appendage and return to his normal life.

Considering how we have less than 24 hours at this point, it will probably just be the "final" level.

But more worrying is that it turns out cocos2D isn't just poorly maintained and sloppily documented - it's straight up broken. You can't import a .tmx map without putting it into an eternal loop, which means I either have to write a converter for it into the XML style map or we write the XML style map by hand.

The python game situation really is not good. pygame is an unpythonic SDL wrapper. cocos2D is a steaming mess and a lot of other stuff is abandoned. A Unity-esq full featured editor and engine for making python games would be rather nice to have.

I had started work on my own stuff on top of pyglet about a month ago before I realized unreleased/personal codebases weren't allowed. I was going to try and create a graphics engine similar to Scratch. Doing graphics in Scratch is simple and accessible but it still has a well fleshed out feature list.

Panda3D seems well made though. Maybe I should have tried using it.

(log in to comment)

Comments

Buffalo974 on 2017/02/25 11:14:

what do you think about pySFML or pyglet ?

pmdevita on 2017/02/25 15:28:

pySFML seems interesting but their documentation is currently under construction. pyglet is alright, I was using it to create my own engine before I realized I couldn't bring in personal codebases. Its documentation lacks a good description of how it operates at a lower level though, which meant I had to step through it's code a lot to figure out how to properly interface to it.

I think at this point I'm just going to write the game straight on to pyglet. It won't be as elegant but it's probably the best way to get results.