Engineers teach their 1-milliwatt neural chip to play Doom, say this is serious work yo, everyone nods
"It unloads the clip, but then it figures out that’s not a good strategy."
The NDP200 is an ultralow power chip that was not designed for your gaming rig. This thing is designed to essentially monitor video via neural networks while using barely any power and, when it spots something of interest like a person walking past, wake up beefier systems to do the 'real' monitoring. AI boffins being AI boffins, of course, they decided to show off its capabilities by teaching it to play Doom.
During the 2023 IEEE International Solid State Circuits Conference in San Francisco last week Syntiant, the chipmaker behind the NDP200, showed how it had been trained to play VizDoom (thanks, IEEESpectrum). The latter is a stripped-back and lightweight version of Doom that's often used in AI research and reinforcement learning.
The level shown has the straightforward title Defend the Circle and features a round room that constantly spawns enemies which, when killed, are immediately de-spawned (so no death animations). The NDP200 was trained on a neural network consisting of several layers in order that the chip's 640 kilobytes of onboard memory could identify the demons of hell, aim at them, and fire.
IEEE fellow and Syntiant's recently departed chief architect David Garrett says the whole reason for training a chip like this to play Doom is to show "you can do meaningful detection and actions at this scale." The NDP200 also had to learn over time that spamming fire is not a good idea. "After its first kill, it unloads the clip, but then it figures out that’s not a good strategy," said Garrett.
I don't think I can quite wrap my head around 1 milliwatt of power being used in this fashion, but assume this was a conference highlight. Training an ultralow power chip to play Doom is a new one anyway, even in a world where getting Doom to run on anything that even vaguely has a pulse is almost a religion. You can run it on your motherboard's BIOS, a pile of potatoes, a tractor, a Lego brick, a home pregnancy test, hell you can play Doom in Doom. My favourite of recent times, though, is the genius who trained rats named Carmack and Romero to play Doom.
The biggest gaming news, reviews and hardware deals
Keep up to date with the most important stories and the best deals, as picked by the PC Gamer team.
Rich is a games journalist with 15 years' experience, beginning his career on Edge magazine before working for a wide range of outlets, including Ars Technica, Eurogamer, GamesRadar+, Gamespot, the Guardian, IGN, the New Statesman, Polygon, and Vice. He was the editor of Kotaku UK, the UK arm of Kotaku, for three years before joining PC Gamer. He is the author of a Brief History of Video Games, a full history of the medium, which the Midwest Book Review described as "[a] must-read for serious minded game historians and curious video game connoisseurs alike."
'Destiny has a long history of reinventing itself in response to feedback': Assistant director teases a Metroidvania-inspired future, talks weapon crafting and vault space, but fails to address the shocking number of bugs
Ballistic, Fortnite's new tactical FPS mode, is a deeply unserious Counter-Strike clone that's going to be huge anyway