The alsoran and his colleagues have spent a lot of time on his current project pondering the supposed Uncanny Valley, a theory put forward by Japanese roboticist Masahiro Mori that the more humanlike robots became, the more people are able to emphasise with them, seeing them in a human light. Paradoxically however, the closer they come to looking lifelike, the more we notice little mistakes due to each viewer having a lifetime’s experience analysing faces, to the point where the viewer is repelled by errors, finding them disgusting and creepy.

This has often been quoted as the reason why Final Fantasy: The Spirits Within was less accepted than the later Advent Children. Where the former tried to take on the awesome challenge of reproducing humanlike characters, the latter opted for a more familiar anime look, where the CG character actually come off looking much more convincing than their realistic elder siblings.

As we move closer to realistic game characters, we run the risk of falling into the Uncanny Valley. While this may seem like a way off in the future we now have proof that this is a real issue. At this years E3, the alsoran witnessed the PS3 demo of Heavy Rain from our old friends at French development studio Quantic Dream.

A victim of its own success, the sheer detail of the lighting and character model, based on actress Aurelie Bancilhon, leaves every single animation mistake plainly visible. Eyes that lazily move with the head, facial rigging that would be unacceptable in a student project, and lip-sycing that suggests the dialogue was initially recorded in Quantic’s native tongue all combine to produce one of the biggest warning signs to videogame animators everywhere. That as we move into the realms of more believable characters, we cannot afford to drop the ball on any element of our craft, be it modelling, lighting, rigging or animation.