Well now. It was another balls-to-the-wall pyweek. I'm sure they are all going to be like this. It started with a bang and maintained the pace all week. Trying to remember when we first had the core completed. I think late Monday evening. I say "we", but for a few small exceptions I was the only coder.

The team was worse for manpower this time around. I knew Cosmo was a great asset in pyweek 12, but it was not until pyweek 13 that I really understood. I really missed his energy, enthusiasm, and creativity.

In addition we had no graphics artist, so we faced a choice. The deciding factor was that DR0ID had an interest in Blender. So DR0ID spent the last five days learning Blender. I daresay he did a superb job. The critters, avatar, fire, ice, and exits are his frames, rendering, and animations. The critter and avatar lack textures, but we felt it was important to show what we accomplished rather than substitute a stranger's art work farmed from the Internet.

Tiles are another story. I used some of the very cool tilesets from The Mana World, and spent most of yesterday and today doing map design, fixing bugs, and fulfilling last-minute coding requests from the team.

JDruid came awake on the last day and whipped out some SFX and music, and wasn't even breathing heavy! :P Maybe I'll take up sound. :)

It was a great learning experience for my pygame library, Gummworld2. I discovered some aspects that really need improvement or redesign. By and large, it performed admirably and except for one design flaw--incompletely leveraging the Tiled map for layer rendering order--it did not require any major workarounds. I can see myself getting a lot of use out of it.

In the end we did not have time to add much content and polish what we had. Menus and credits are plain. Combat should have been made more exciting, and the core has a lot of untapped potential. In retrospect, had I known DR0ID would not be coding I would have preferred a more humble game idea. Another part of me is fairly confident that if you have time to kill during a pyweek you need to think bigger.

Despite all the challenges, or maybe because of them, it was fun! Well, no. Strike that. My car breaking down and sucking four freaking hours out of my valuable coding time was *not* fun. But the rest sure was. :)

Well, here was a rude moment of truth. GameClock breaks down when the host's CPU is challenged, and it's not properly sacrificing frames in favor of ticks. To make matters worse, this happened last pyweek, too. Another possibly related mystery is that DT is not computed accurately under stress, and game mechanics lose huge chunks of time when the CPU can't keep up.

The practical result is that some people will not be able to complete the second map unless they use the final2 release. I radically hacked GameClock to get it to function well enough for *this* game, and increased the spawn timer by two seconds on the tricky shroom. Conversely, people with faster computers will find this level way too easy. But at least hopefully no one will be frustrated by a DNW now.

Back to the drawing board with GameClock... again... with a steady stream of invective-laiden grumbling.