{"author_name":"huga","cat":"LD #29","comments":[],"epoch":1398744540,"likes":0,"metadata":{"p_key":"30276","p_author":"huga","p_authorkey":"0","p_urlkey":"65748","p_title":"Worm Simulator 2012: Post Mortem","p_cat":"LD #29","p_event":"LD29","p_time":"1398744540","p_likes":"0","p_comments":"0","p_status":"UPD5","us_key":null,"us_name":null,"us_username":null,"event_start":"1398384000","event_key":"22","event_name":"LD29"},"text":"<p>This was my fourth attempt at Ludum Dare, and my first time working in a team. How did things go? Well, play the game and you&#8217;ll see. Let&#8217;s just say that they didn&#8217;t go perfectly. While I think the concept was the game&#8217;s strong point, it was probably also a mistake. The thing about simulators is that you&#8217;re simulating one thing to a high level of detail, which in my case meant that 90% of the game was in a single file. This resulted in my teammate spending most of his time on reddit as there was only really one thing that could be worked on\u00a0at a time.\u00a0However, by far the biggest problem was the game&#8217;s performance. When I\u00a0added a level\u00a0for the worm to explore, 32&#215;32 tiles, the game would run at 5 frames per second. This was my first time working with the Java 2D api, so I had basically no idea what to do at this point. I ended up having to use the Slick2D library for graphics, but I screwed up a bunch of stuff while adding that and I didn&#8217;t get the game back to where it was until Sunday afternoon. The end product was basically fun to screw around with, but nowhere nowhere close to being a\u00a0game. However, we like the idea of this game and we think we&#8217;ll keep working on it.<\/p>\n <p>&nbsp;<\/p>","time":"April 28th, 2014 11:09 pm","title":"Worm Simulator 2012: Post Mortem"}