I don’t like the word “Post mortem”. It indirectly implies that something is dead on delivery. A game comes alive first when it is played. That said, here are a few key observations from creating Fjuffy teh kloud so that others can learn from my experience.
The silence is falling
Audio programming in general has always been my blind spot. The DIY compo rule makes I even trickier because sites like Open game art can not be used. The fact that there is no audio feedback at all makes time spent with Flyffy dry and stiff. Just a few “pings” and “swishes” would have made a lot of difference. I will have to put some extra effort into this part skill set before the聽 LD33.
Pomodoro? More like Pomodorocks!
The “Pomodoro technique” is a way of keeping focused. The rules are simple: Divide work into smaller task and work on one task at in intervals of 25 minutes (aka a “pomodoro”) with 5 minute breaks between them. Take a longer break after 4 pomorodos. Count your pomodoros. That is all.
Using this technique not only helped me keep focus. Counting the pomodoros also alerted me when something was taking to much time and this helped me re-plan when needed. This is something I will do again on LD 33 and I do recommend everyone who wants to be more efficient to try it.
Plan A and B
As the genius DevOps Borat puts it :
“Make of sure you have strong plan C. Plan A and B are never work.”
My original Plan A was to make an non-geometric game. By non-geometrical I’m referring to games that do not value positions, collisions, updates per second etc. Examples would be text-adventures, VN:s, some simulators and RPG:s. I was prepared fr this! But the first (and only) good聽 idea I came up was an action game, plan A failed.
Plan B was to make a game about encircling particles. You would fly around as a little cloud and make rings and other cool shapes. Unfortunately I have not dabbled with vector mathematics in over a year so having to quickly reboot those brain parts was not easy. Detecting what is inside a (possibly concave) shape is a hard problem. With only very basic vector code to reuse it was hard enough to just detect two lines intersecting.聽 This is why fluffy cleans the air near the jet stream intersection, not inside the loop. I think it works sort of well, even it it was not my original intent.
Minimal playable product
The day after LD was a regular 8:30-17 work day for me, so staying up coding untill the last minute was never an option. I did manage to cram stuff like a score and a “game over” screen in my very last pomodoros, making it an actual game (and not just unchallanging interactive art). I should probably have done this part sooner in the development process.
Firsts times a Charmander?
This was my first LD and my third Game Jam ever. I usually miss events like these and it always annoys me. Actually participating on the other hand was really a lot of fun! The time limit makes you scope down, and that means you finish your game. It got me fired up! I had ideas! I had to implement them all!
If you never been in LD before and you’re reading thing thinning “darn it, I missed it again.. oh.. well.. I might not have enjoyed it anyway.. Q-Q” then take this as my final advice for this time: The next Ludum Dare is in AUGUST, burn this into your MIND and CALENDAR.