Week Eight Observation by Jeremy Kang (Academic Staff at Republic Polytechnic, Singapore)

An almost deafening silence permeated the GAMBIT Labs for most of the week, as the air was thick with tension. Aside from the occasional laughter and chatter that marked the lunch hour at GAMBIT (with a hilarious discussion and demo of Big Rig Racer thrown in for good measure), the interns mostly locked themselves up in their respective rooms and were fully focused on getting the builds up to a Feature-Frozen state by Thursday, and Gold for the following Tuesday.

The pressure was definitely on, as there were many furrowed brows and glares intense enough to stare through the monitors. The momentum built up over the summer had definitely reached its highest point, and every team's game was coming together much faster than they ever did.

Snowfield probably went through the most change, as the team and their Product Owner sat down on Monday morning to re-think the direction of the game, and agreed on a way to incorporate more gameplay into the game using the elements that were already present. The other teams had their work cut out for them as well - some with more man-hours needed to go into the overall visual and audio polish of the game, while others had to focus more on building content on top of the established mechanics, and yet others had to focus a bit more on usability issues in their UI.

A definitely trying week for the interns; and for most of them, this would probably be the first time that they are going through the pressure-cooker of a game development environment. If they can come out of this alive, they would probably gain a multitude of experience points... but that's the story for next week.

Week Eight by Douglas Finnigan (Temasek Polytechnic)

Today, once most of the teams had reached Gold, I asked a bunch of students to let me pester them with questions about their Gambit experience. The discussion was very interesting, and helped to bring a few things into better focus. I thought I might discuss their thoughts about the research questions given to each team. I was curious to know if they had a good understanding of their research questions, as well as an understanding of what research is in general.

The Squeezicks team saw themselves as working to demonstrate their product owner's soft-body physics engine, and so were more technical in focus. They also had to consider the requirements of their client, the Museum of Science, which introduced an educational aspect to the game. One Fab team member felt that their research question related to minorities in general. When asked to state his understanding of the research question, he felt that it was to explore if LGBT representation could be positive as well as meaningful in a game. My feeling was that they saw research as a means to an end, with that end being a game, rather than as a way to gain a deeper understanding of some physical or social process. This might be a result of the results-oriented education they receive in Singapore, or of how they see themselves and their roles in GAMBIT, or a combination of these. It may also be that a working game is the intended outcome of the research. One thing the students emphasized, though, is that they felt involved in the research, with GAMBIT staff supporting the teams' decisions and frequent changes of direction. They obviously saw themselves as more than GAMBIT lab technicians, and more as collaborators. When asked if they had gained a better understanding of the research questions from developing their games, the LGBT team member noted that he hadn't gained a deeper understanding, as such, but now had a different view of how games as a relatively new form of media were part of the wider media representation of gender. (In fact, this wider view does represent a deeper understanding of the research question.)

Most students said that they had no prior experience of research, but that their GAMBIT experience has made them see how games can be used as a research tool, and that they were motivated to do more games research back home in Singapore. I hope the momentum can be sustained.

Going Gold in GAMBIT by Mark Gossage (Singapore Polytechnic)

Although on paper Thursday was the deadline for 'feature freeze' I don't seem to recall that happening. The game was meant to have nothing new added. I seem to recall Friday's standups still had a 'get the last battle in' or similar. Fixing the text is fine, second passes on art/sound is ok, bug fix is essential. But still there was this last minute push to get that little bit more in.

Having said that, the weekend was quite quiet, next to no coding happened. I was in the lab for quite a bit of the time. Students were around, but chilling. Monday was almost all test-bug fix. Audio bugs, UI bug, gameplay bug, probably buggy bugs. It was a full day of hardening the build and all hands were on testing. This year no team managed a gold candidate on Monday, one team finished late Monday night, but Tuesday morning they were still testing. It seems they there is a little rivalry on who can get the gold up first. Why not! Final deadline is Tuesday 4pm. Most of the teams submitted around Tuesday lunchtime. Part of the reason is the game director & producer pushing the team to fix only the stuff that needs to be done. Do quote one of the directors: Nobody wants to be in the situation "I fixed something optional, it broke something essential."

As soon as the first build was in, it was all QA's (and Mark) to the testing lab. First up was verifying each item, each NPC, each sound, each interaction was accounted for. In one of the games a QA just sat with the checklist while we all played the game and shouted out every item we saw. Towards the end it was "Ok, has anyone seen the Ice Kachang and the Curry rice?" as we hunted for those last few odd items. After that it was what strange and crazy things can we do to the game? How will it respond if I don't have a name or try to do tasks in the wrong order or hammer random keys on the off chance.

Tuesday came and went, submission 6/6, passed 2 ½ /6. 2½? Well two were clear passed, and one managed to pass on one of its platforms, but not the rest.
What went wrong? Bad credit screens, buggy loading screens, occasional crashes. Generally I feel that the QA team just listed all the bugs & the game directors (not owners) decided if they were sufficiently minor to ship with. By end of Wednesday, I have started to get a good feel for which were the 'we cannot ship with this' and the pass rate was up to 3½ /6. Some teams had run 3 gold candidates and still had problems.

By 5:15 on Wednesday, the QA team was dead beat, we must have done a good 6-8 QA runs (each game about twice). The one high point of the day was when they told team Fabulous that their game had gone gold. We heard the cheer through two walls!