Ahoy all!
Here is quick post-mortem about my LD24 entry called Deep Sea Fish Loves You Forever. The game is about piloting an underwater exploration robot and analyzing oceanic wildlife with a pink sonar.

What went well

Tech & Engine
I used the same technologies as I did during my other 2 Ludum Dare and this is really paying off. I am now very fast when it comes to prototyping and testing things in Bacchus (my custom game framework written in Haxe). The only tech problem I had was due to a recent rewrite the audio part of my framework, more about this below.
I don’t think I fixed a single engine bug during the whole weekend.

Timing
My schedule was similar to last time.
1. Do graphics on Friday night (I hate using programmer art so I start with gfx)
2. Write the core game mechanics on Saturday
3. Make a title screen and winning/losing conditions on Saturday night
4. Add audio, polish gfx and playtest on Sunday

I think it worked quite well, except that my (few) testers did not find too many bugs so I ended up submitting early. I could have added more species to the game but I was afraid to break the balance (which I thought was good (it’s not)).

Iterations
Testers were very useful not for finding bugs but for highlighting usability issues. It is thanks to them that there is a species counter in the GUI. This feature makes the game’s objective a lot more explicit, I’m really happy someone came up with it (thanks Phil!).

I also used them to improve the opening tutorial. My first testers didn’t understand much about the game because the tutorial had too much text and some of it was not explicit enough. I had enough time to redesign it and then to improve its texts later on. It went from this to this:

Assets creation
During LD#22 I wasted a lot of time doing pixel art (and an 8-frames long walk-cycle!). This time I painted things with basic Photoshop brushes and scaled them down. I think it looks quite good and it took a lot less time. The only thing where I worked on the pixel level is the GUI but it didn’t take too long because it’s very geometric.

Audio was a pleasure to make using Reason and my Akai MPK Mini. I had not done anything audio-related in 6 months so I spent a few hours reading and watching Reason tutorials the day before to get back in the mood. It was my first time writing some Electro-ish music and that was a lot of fun. I used a morse signal sound for the main the track intro (just for fun) and I was inspired by The Future Sound of London for claves’ sounds on top of the beat.

I didn’t have a microphone this time around so I made all the sound effects using synths, but this was ok because most of them are mechanical anyway.

What went not-so-well

Theme & Concept
Evolution is not one of the themes I was looking forward to and I had a hard time coming up with something. As always, I was very stressed at the beginning and settled for a vague idea quickly. I thought I would make a game where you play as a bacteria and evolve from species to species by consuming radioactive pills in order to solve environment puzzle (think Space Station: Sillicon Valley on the N64). Of course there was no time for that so I scaled it down from top-down to underwater and side-scrolling. Then I started doing art, without any other ideas in mind. By the end of the night I had drawn all the fishes that are in the game (and a radioactive barrel which isn’t) but I didn’t have any puzzle or gameplay ideas to accompany them. It was supposed to look like the following picture. I think this scale looks a LOT better than the one I ended up using but I didn’t want the player ship to be 5 pixels big. Maybe I was wrong.

The next morning I realized this was not going to work because I would never have the time to do environment puzzles, animations and gameplay programming for all the species. The challenge became to find another game concept where I could use all this art while keeping the evolution theme. After half an hour of panicking, I came up with the final concept. As a result, it doesn’t fit the theme too well (esp. in its gameplay).

Computer
My computer did shut down without a warning twice during the weekend. The second time it did, it decided not to turn on again so I went to bed. The next morning I had to open it and replug the hard drives, I think the SATA cables were twisted too hard or something. That was very scary and I thought I might have to go finish my game at the office (it was under version control so I wasn’t afraid of losing any work).

Food
I put some barbecue sauce in my sandwiches and omitted the eggs. BAD BAD mistakes.

Performance
I heard the game was very very slow on some computers =(
I didn’t pay any attention to performance because everything was running at a solid 60FPS on my machine. Some obvious optimizations could have been done to improve that. Not drawing a 2000 pixels tall half-transparent gradient on top of the screen would have been a good start.
Also, I found out that the new audio work I had done in my engine was painfully slow so I added a dirty one line hack to bypass it.

All in all, I had a lot of fun and I’m very happy (and proud!) I made this game, even though it’s far from perfect. Thanks a lot for reading and for being such a friendly community!

Tags: post-mortem

This entry was posted on Thursday, September 13th, 2012 at 3:08 pm and is filed under LD #24. 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.