In terms of realism, features such as occlusion, which is what happens when an audio source moves into another room and shuts the door, or attenuation, the drop-off in volume over distance, have long been modelled in audio engines. The problem is that they tend to be computationally expensive to calculate as they require knowledge of the 3D geometry of the game environment.

This has made them difficult to bring to online games, especially those which have high concentrations of fast-moving players in small locations. When the choice boils down to realistically modelled audio or a higher framerate, the eye-candy still wins, even in these days of quad-core processors.

Tullis says Axon is designed so gamers who want great graphics and sound can have their cake and eat it, too. “We have our own proprietary codec, which enables us to do some processing on the client side, as well as ensuring we’re very efficient in terms of what we send for processing and mixing on the server. Basically, we minimise the number of transforms we have to calculate on the server, while the codec allows us to make the dialogue high quality at low bandwidth. We can do a full 3D surround sound scene with as little as an average of 16 kbps per player,” Tullis explains.

On what Tullis refers to as a ‘modern PC’, the CPU overhead for Axon is around one per cent. On the minimum specification (which is equivalent to World Of WarCraft’s minimum: an 800MHz Pentium 3 with 512MB of RAM), the overhead is less than ten percent.

“We try to keep the gamer in mind all the time, because people are automatically concerned about the impact on their CPU performance. They don’t want lag or frame drops. It’s the same reason that we will require minimum configuration from players in terms of them fiddling with Axon’s settings. If there’s clipping, we’ll automatically lower the voice at input. If someone’s set their microphone level too low, we’ll boost it as much as we can,” Tullis says.

But while Axon should solve plenty of boring technological problems for audio engineers (and gamers), the thing that should be really exciting them, and is certainly enthusing Tullis, is its potential for gameplay. No developers are officially announced for Axon yet, but over a dozen are evaluating it. Examples of its use could include the ability to attach an in-game microphone to a location or an object. In a typical deathmatch scenario, this would enable you to lure an opponent into a trap such as a mined location or ambush point.

More prosaically, you could just drop a microphone next to your flag in Capture The Flag. Alternatively, another studio working on a more social-oriented title is experiencing with voice-weapons, which enable players to morph the voices of those opponents they frag: this uses another of Axon’s features, the ability to use voice fonts so you can customise the dialogue of in-game characters.

Tullis says he hopes such features and flexibility will allow designers and audio engineers to be more creative. “It took a long time for Dolby to get to the stage where nine out of the top ten console games were using Dolby Surround to enable a more cinematic experience,” he argues. “And I think Axon will also be a differentiator because it enables you to deliver high-quality voice, and from a developer’s point of view we’ll also expand their creativity. And hopefully, the games that use Axon will be more fun.”