So we did it. For our very first participation, that’s quite a success. 72 hours (a bit less because we had to attend school on Monday) was shorter than expected and, as any newbie I think, we did a lot of poor choices.
Yet the game is out, it’s kind of fun to play and it was a very nice experience overall.

Common! Play with us!

What went right?

• The timed roguelike concept turned out to be funnier than expected. It was our only valid idea when we ran out of time to choose a game design. We had thought of “you can only get one true love” where you’d be in a room with a lot of people and you’d have to reduce your choice to one; stuff like that, but nothing seemed really fun to play… Turns out that adding a time pressure on the classic genre makes it all new again, and most people we’ve seen play were screaming when they had only a few seconds left to the counter.
• The hall of fame/Score system has started a new competition inside our classroom, with amazing scores! If we had an online leaderboard (we had enough time to add it to the game), the emulation would have been even more active.
• The stuff system with no inventory and the fast-choice system is great, and totally in line with the rest of the fast-paced experience.
• No major game breaking bug that we know of, at least. And that is some sort of miracle given the fact that 30 minutes before release, the game crashed on an 8th of the equipment.
• The monster variety was really a big worry during the development, we had only 3 sprites for the enemies… Turns out that displaying their equipment was enough to forget about this limitation.

What went wrong?

• Coding. Before we started, we had settled on GameMaker Studio. It should have allowed us 3 to work on the game together in a perfect world filled with unicorns… The harsh reality made us go back: we lacked of experience with GM:S and during the first night we switched to Pure Basic. The bad part is only one member of our 3 men crew had any experience with programming (and none with game), and a lot of poor code design choices were made: we ended up with the worst code ever. But at least the game was done.
• The shield mechanic does not work, and the reason is obvious: you’ve got to wait in a time limited situation. Pretty rubbish.
• A huge lack of feedback, that is mainly a time issue, but most of our feedbacks couldn’t make it to the release; you don’t know when you are short on life or on time, and death is especially surprising.
• Our workflow is pretty much non-existent : we had nothing prepared, no file sharing system and not even a complete development pipeline…
• Balancing SUCKS. We couldn’t do any testing till the game started to be a bit playable (in the very last 4 hours), and most of our formulas were not even remotely working. We settled on too-easy and almost game breaking ones by default, but we now know how to think them. Next time we’ll do better, for sure.
• Level generation is boring. This fact is counterbalanced by the fact that no-one in our team had any idea of how to make a level generator when ours was coded, but we should have done some in-depth studies on existing generators before rushing to code.
• The controls, especially on windows with a keyboard are not working. We should have tested it to avoid using alt as default button (you can use wasd and q/e/space as an alternate and much better solution).

Aaaand that’s all for us. We are amazed that you guys played our little game. Many, many thanks for that; and our next participation will be much better, I swear.

Till then, we are going to redo rogue rush with all the cut-out features ;3

Cya next time

The SchizoKoa Team

This entry was posted on Friday, December 20th, 2013 at 11:40 am and is filed under LD #28. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.