Halo 3

Yesterday Wired ran an article about Randy Pagulayan's team at Microsoft Games User Research and their user testing of Halo 3.

It's an entertaining read with a few little tidbits of information on how they conduct their usability evalusation process. Highlights for me include the observation that dying can be fun. This is clear to a game player or developer, but might come as a surprise to a usability specialist focussed on effective and efficient interfaces. That "skin meters, cardiac monitors, and facial electromyographs" were needed to deduce this is quite surprising. I'd have thought that a questionnaire would suffice. Of course, this depends on the kind of game. If you get to see an entertaining animation, or a tactically-useful slow-motion replay of your death then there's clearly 'value' for the player who died. However, in a multi-player deathmatch game where you have to sit and wait for several minutes until all of the other players finish their game, this is clearly going to be a frustrating experience - unless there is some activity to take part in such as chatting with other dead characters, or trailing other live players.

Most interesting for me is the description of the tools Pagulayan's team devised to capture and visualise play throughs. This is where I see a double programmer / usability expert as being extremely useful to game development. Conditionally compiled code could be added to a particular build of the game to enable data tracking, perhaps offloaded to disk or to a database on the local network. The other side of this process would be to construct tools to analyse and visualise the data. It would work like a semi-independant tools and QA team, providing analysis back to designers.