Lights – Post Mortem
Challenge accepted
When we were presented with this project we were 3 months into our game design studies and none of us had actually made a game before. Of course we had a basic theoretical understanding of what needed to be done, but that was as far as our knowledge would take us. As it turned out, due to licencing issues, we would not be able to use the engine we had been using during our studies (UDK). The only other engine any of our crew had touched was Unity, so that was the one we decided to go for.
Even though Unity wasn’t completely unknown to us, we still didn’t have any profound knowledge of it’s inner workings and we basically had to learn everything as we went. Knowing this and then finding out that we would only get two months to finish the game, where almost two weeks would get swallowed by christmas, it was clear that this would not be easy. We would probably have to work arround the clock during the entire project to finish on time. Still, we all eagerly and without doubt accepted the challenge and set to work as soon as we possibly could. After all, this was what we wanted to do with our lives.
What went right
Great minds think alike.
From the very start of this project it became clear that we all had very similar idéas and visions about what we wanted to do. Writing the document and getting the concept in place was a very straight forward process as soon as we had gotten the «go» from the guys in Eye Emma Jedi. Also when we had to make cuts or changes during the production it was very painless and without much discussion.
Pressure
Having all three of us living and working together, seeing each other every day with very little time to get out of the house, the chance of pressure building up and ending in explosion was allways pretty high. Luckily we managed to keep the peace and avoid disaster. With our tight deadline there wouldn’t have been much room for distractions and so having a friendly and peaceful environment was important to keep work and morale up.
Eye Emma Jedi
Working with the guys of Eye Emma Jedi was a dream. As of what clients go, they certainly were of the best kind. They had complete trust in us from the beginning and were very understanding of how little time we actually had to make their game. Very few of the changes we had to make were on account of them. We mostly got free hands and they were allways exited and happy about the things we showed them. This was an important part of why the production went as smootly as it did.
What went wrong
Too much freedom
When we first started working on the project we didn’t get any guidelines for the visual design. We got 100% free hands and we came up with a design we felt quite good about. When we showed it to the client they weren’t as confident about it as we were and we ended up redesigning the whole thing. Luckily this happened quite early and it wasn’t a big fix for us to do.
Hardware trouble
As we got into building our level it started demanding more and more from our computers. The biggest problem was not having enough ram and the computers would frequently crash or freeze for long periods of time to calculate changes. This made the process very slow and it ate a lot of valuable time. We managed to make it better by cutting down on the number of pick-ups, but there was only so much we could remove. This was probably the biggest problem we encountered.
Bugs
During production there were a few small bugs we had trouble rooting out. We had some issues with some of the assets jumping along with the character without any obvious reason. Some things were just a quick fix in the code, but in other cases they seem to vary from computer to computer which made them hard to do anything about.
Music and timing
Timing the flow of the game to the music turned out to be a bit more difficult than we had anticipated. We didn’t have any tools that could automatically place pickups by the beat of the music and so we had to do this by playing through and ajusting everything little by little by hand. The fact that the character for no reason seem to speed up a little when it jumps causes the timing to vary a little bit from time to time. We tried timing it as close as possible, but it still varies each time it’s played.
Murphy
A couple of weeks before the deadline, I (Yngvill) caught the flu and got so ill that I couldn’t even sit up straight. I lost two whole days of work, which in this case was quite a lot. Luckily I managed to catch up when I got better, and I promise never to speak of Murphy ever again.
Still breathing
After delivering the game to Eye Emma Jedi we were all exhausted, but we felt pretty good about what we’d accomplished during the past couple of months. Even though it had been hard, we were all happy that we had taken the project. The game still has a few bugs and isn’t perfect, but we’re still happy that we actually did it, and that we finished on time. And the most important thing is that we managed to make the client happy and in the process we’ve gained valuable knowledge about the thing we hopefully will be doing for the rest of our lives.
Read more about the production of Lights at the now available devblog.
Post Mortem
