OK, you probably know about ‘the uncanny valley’ (maybe start at the Wikipedia article if you don’t), but do you know about ‘the annoying valley’? If you’ve played a video game in the last bunch of years, I’m sure you do.

In an effort to make their AI players more human-like, a lot of game developers add little human touches, like a character saying something to you in a specific situation. But what’s a pain is that they often say the _exact_ same thing in the _exact_ same way (basically, the same sampled audio) in the same situations. Such that you hear it possibly dozens or hundreds of times in the course of normal gameplay. This is annoying, this is the annoying valley.

Some sounds in a game you expect to repeat exactly or nearly exactly, because they would in real life. But humans are so rarely that boring, so the fact that it’s supposed to be a human reaction just makes the annoyance level skyrocket. I suppose you could call that a personal problem, and I certainly understand the relative difficulty of making human speech/animation that sounds/looks good _and_ has lots of natural-seeming variability, but still…