Intel connects 64 Loihi 'brain chips' together and hands them off to researchers

Intel's current brain network isn't quite as smart as a frog. Next up: mole rats.

Credit: Intel

Intel said Monday that its “Pohoiki Beach” neuromorphic system, consisting of 64 of its Loihi chips connected together to simulate 8 million neurons, is being made available to the broader research community. A 100 million-neuron system, known as Pohoiki Springs, is on track to debut later this year, as originally planned

In 2017, Intel began defining Loihi, the first of its neuromorphic “brain chips” that simulate how the human brain works, in terms of the synapses that connected them. Now, the company has returned to just talking about simulated neurons.

If a Wikipedia summary of the number of neurons within a particular animal’s brain is assumed to be correct, that would put the level of intelligence of the Pohoiki Beach system at about half that of an average frog, or slightly less than an adult zebrafish. One hundred million simulated neurons would be on par with the Ansell’s mole rat, described as a species of rodent native to Zambia that's noted for its tunnels that can stretch over two kilometers long. 

nahuku neuromorphic board Intel

One of Intel’s Nahuku boards, each of which contains 8 to 32 Intel Loihi neuromorphic chips, shown here interfaced to an Intel Arria 10 FPGA development kit. Intel’s latest neuromorphic system, Pohoiki Beach, is made up of multiple Nahuku boards and contains 64 Loihi chips. (Credit: Intel Corporation/Tim Herman)

Intel is looking at neuromorphic or probabilistic computing as an alternative to conventional silicon architectures. It’s not quite clear what role Intel sees for a neuromorphic chip in a world governed by x86 PC processors, but the company is making the chips available to researchers to test out neural-inspired algorithms such as sparse coding, simultaneous localization and mapping (SLAM), and path planning.

The University of Waterloo has been able to run a real-time learning benchmark on top of a Loihi chip at 109 times less power than a GPU, which have been adapted to run dedicated algorithms in past years, researchers said. Scaling the network by fifty-fold only increases the power consumption by thirty times, they said, while still maintaining the algorithm in real time. 

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.
Keep up with the latest tech news, reviews and previews by subscribing to the Good Gear Guide newsletter.
Mark Hachman

Mark Hachman

PC World (US online)
Show Comments

Most Popular Reviews

Latest Articles


PCW Evaluation Team

Tom Pope

Dynabook Portégé X30L-G

Ultimately this laptop has achieved everything I would hope for in a laptop for work, while fitting that into a form factor and weight that is remarkable.

Tom Sellers


This smart laptop was enjoyable to use and great to work on – creating content was super simple.

Lolita Wang


It really doesn’t get more “gaming laptop” than this.

Jack Jeffries


As the Maserati or BMW of laptops, it would fit perfectly in the hands of a professional needing firepower under the hood, sophistication and class on the surface, and gaming prowess (sports mode if you will) in between.

Taylor Carr


The MSI PS63 is an amazing laptop and I would definitely consider buying one in the future.

Christopher Low

Brother RJ-4230B

This small mobile printer is exactly what I need for invoicing and other jobs such as sending fellow tradesman details or step-by-step instructions that I can easily print off from my phone or the Web.

Featured Content

Don’t have an account? Sign up here

Don't have an account? Sign up now

Forgot password?