Intel believes that the meta-verse will need a thousand times more computer capacity

Intel issued its first statement on the metaverse on Tuesday – its first public acknowledgment of the sometimes hazy future of computers, which promises an ever-connected virtual world that exists in parallel with our physical. But while the chip company is optimistic about the possibilities of metaverse in abstract, Intel raises a key issue in realizing any metaverse ambitions: There is not nearly enough processing power to go around.

“Metaverset may be the next big platform in computing after the world wide web and mobile,” begins an editorial by Raja Koduri, a senior vice president and head of Intel’s Accelerated Computing Systems and Graphics Group. But Koduri is quickly pouring cold water on the idea that the meta-verse is just around the corner: “our computer, storage and networking infrastructure today is simply not enough to enable this vision,” he writes. The crucial thing is that Koduri does not even think we are close. He says there is a need for a 1,000x increase in power compared to our current collective computer capacity.

Much of the metaverse hype is built around what you want to do there, be it virtual reality meetings, digital concerts and of course blockchain and NFT based integrations. And there’s plenty of excitement surrounding the future of virtual and augmented reality headsets, whether it’s Meta’s Quest products (formerly known as Oculus) or Apple’s long-rumored headsets.

But the real building blocks of the meta-verse will not just be software and virtual spaces (which, of course, is its own struggle, given that today’s digital worlds are extremely independent) or even headsets and gadgets that people carry to “get there. It will be in the computers and servers that run the huge shared virtual worlds that the meta-verse exhibits as the future of technology.And that’s where Intel has the biggest reality check: Today’s computers are simply not powerful enough to make those dreams come true. is not even close.

On the one hand, the statement here is almost ridiculously obvious. Meta’s flagship VR space, Horizon Worlds, has a maximum of 20 participants for one space, and it’s too basic, Roblox-style animated worlds. The latest in VR still requires thousands of dollars of PC gaming hardware, with lots of drawbacks (like requiring a tied headset and graphics that still can’t match what 2021’s best flat screen games can offer). And even the biggest traditional video games it is it not handle the additional requirements of VR as Fortnite or Battlefield 2042 can only handle up to 100 to 128 players at a time.

As Koduri notes in his leader, we can not even say that thaw people in a truly detailed virtual environment with today’s technology. Consider what it takes to place two people in a social environment in a completely virtual environment: compelling and detailed avatars with realistic clothing, hair and skin tones – all rendered in real time and based on sensor data capturing real 3D objects, motion, sound and more; data transfer at super high bandwidths and extremely low latencies; and a sustainable model of the environment that can contain both real and simulated elements. ”

And it’s only for two people – upscaling to the hundreds of millions of users, as one Ready Player One, Snow controlled, or Matrix-style metaverse concept would require much, much more computer infrastructure.

Of course, Intel also has a vested interest in saying that we need more and better computers and servers. After all, Intel makes CPUs (and soon GPUs) for both consumer devices and data centers. And if the meta-verse – the hottest buzzword technology of the future – needs a literal 1,000x increase in computing capacity, well that’s just fine for business. It is no coincidence that Intel explicitly named both its client computer and cloud processors and graphics products in its metaverse brief.

The problem, though, is that even Intel do not think the hardware alone will make us 1,000x. As Koduri explained in an interview with Quartz, “We believe that a standard type Moore’s law curve will only lead us to about eight or 10 times growth over the next five years.” (Moore’s law is generally defined as a doubling of computing capacity every two years, following the eight to 10 times growth that Koduri predicts.)

Instead, Koduri optimistically predicts that algorithms and software enhancements will close the gap. Things like machine-learning neural networks or AI-enhanced computing techniques of the kind that Intel is already using for things like its Deep Link technology or the upcoming XeSS supersampling, which it plans to debut with its Arc GPUs early next year . However, that’s a big question – Intel is counting on algorithms or AI to offer a hundredfold (or more) improvement in computer capacity, all on top of the growth that its existing hardware roadmap offers.

Koduri notes in the same Quartz interview that improved software and algorithms will not only be needed to close the gap in the ambitious five-year timeframe he lays out; they will be crucial in helping to mitigate the increased energy consumption that attempts to brute force the problem would create, something which he compares to the current problems of cryptocurrency mining today.

It’s easy to just wave your hand and say that software will fill in any gaps left by the hardware (especially for a company like Intel that primarily makes the hardware). Lots of big technology companies have flocked to the idea that AI and machine learning will solve their computational problems, for everything from making smartphone cameras better to offering upscale gaming visuals, and it’s appealing to think that they can. But it still seems like a big order to rely on them for 100x tomorrow’s computing, which is only expected to see a 10x leap based on hardware improvements alone.

However, the fact that Intel is thinking about all this now – and stating the problem – is an encouraging sign. It’s easy to drive on the hype and start pitching great ideas on selling NFTs that will follow you from place to place in different games and virtual environments. It is less sexy to improve the server infrastructure and work on reducing latency; but as Intel’s presentation shows, there is much more basic work to be done in the coming years to pave the way if the meta-verse is ever to achieve its sci-fi ambitions.


Related:

Leave a Comment