Skip to main content
  1. Home
  2. Computing
  3. Trash
  4. News

Why multi-chip GPUs are the future of graphics power

Add as a preferred source on Google

Multi-chip module (MCM) graphics cards just might be the future, bringing upcoming generations of GPUs to a whole new level. Combining multiple dies instead of relying on a single chip could provide better performance without being hindered by the current hardware limitations.

According to recent leaks, both AMD and Intel may currently be working on MCM graphics cards. AMD’s rendition of this technology may be just around the corner — in the upcoming RDNA 3 GPUs.

Patent: Position-based rendering apparatus and method for multi-die/GPU graphics processing – Intel

Intel MCM GPUs is coming…

More details: https://t.co/GIkfwrXGzV pic.twitter.com/sXGt9nbJ1S

— Underfox (@Underfox3) February 3, 2022

The term “multi-chip module” describes the architecture used to assemble electronic components, such as semiconductor dies or other types of chips. MCM involves integrating multiple GPU modules into a single package. This could potentially result in a lot more GPU power without drastically increasing the size of the chip, thus hopefully improving manufacturability and power consumption.

Today’s leaks come from two different sources and both point (somewhat strongly) toward AMD and Intel both exploring MCM architecture for their upcoming consumer-level graphics cards. A new Intel patent has been uncovered by Underfox on Twitter and later shared by Wccftech, talking about using several GPUs combined via MCM for image rendering and the benefits this would provide.

Intel’s patent talks about using tile-based checkerboard rendering to achieve far more efficient scaling on multi-chip graphics cards. Although it’s unlikely (if not impossible) to see this technology in first-generation Intel Arc Alchemist graphics cards, we may see MCM GPUs from Intel in one of its next-gen discrete graphics cards.

The leak regarding AMD (first shared by PCGamer) comes from an unsuspecting AMD engineer who shared their current projects on their LinkedIn profile and was spotted by blueisviolet on Twitter. The engineer is a principal member of technical staff and works on Infinity Fabric. Information pulled from the profile strongly indicates that some of the next-gen AMD RDNA 3 graphics cards will feature an MCM design.

this why on some my previous twit
i said rdna3 will probably start with 5nm
amd pssst linkedin pic.twitter.com/ZfdfrvgwTO

— blue nugroho (@blueisviolet) February 4, 2022

Multi-chip module GPU technology is something that Intel and AMD have previously dabbled in, but Nvidia is also no stranger to the architecture. As early as 2017, Nvidia has published a paper titled “Multi-Chip-Module GPUs for Continued Performance Scalability.” Since then, rumors have emerged indicating that Nvidia’s next-gen Hopper graphics cards will feature multi-chip GPU designs.

AMD has already released an MCM graphics card before, the monstrous Instinct MI200 HPC GPU, designed for high-performance applications (such as data centers.) The card offers up to 3.2TB/s of bandwidth, 128GB of HBM2e memory, and up to 14,080 processors. However, it’s likely that MCM technology will now make it into the consumer market with AMD’s upcoming RDNA 3 graphics cards. Intel also has a data center MCM GPU in the works. Dubbed Ponte Vecchio, the card has not yet been released.

MCM architecture certainly has a lot to offer, and even if we’re not quite there yet, it seems that we may soon start seeing its benefits on the consumer market as part of the best graphics cards. AMD RDNA 3 graphics cards are set to launch later this year. The same can be said of Intel Arc Alchemist, and although we don’t have any certain release dates for either, Intel is rumored to launch the first Alchemist GPUs before the end of this quarter. Meanwhile, AMD is also rumored to release a refresh of previous RDNA 2 cards within the next few months.

Monica J. White
Monica is a computing writer at Digital Trends, focusing on PC hardware. Since joining the team in 2021, Monica has written…
OpenAI is locking its most advanced AI models offline after a string of security scares
OpenAI’s fake internet is a safety fix and an admission.
openai-anthropic-logo-on-screen

Part of me is relieved that an AI lab is hitting the brakes. The other part wishes it hadn’t taken real incidents to get there.

OpenAI said Friday that its most advanced models will now train without live internet access. Instead, they’ll practice on a cached, simulated copy of the web.

Read more
Your Windows touchpad is about to make scrolling long pages way less tiring
Three touchpad tricks that make long documents feel a lot shorter.
Windows new scrolling feature

Today I learned about a touchpad tip that I think every Windows laptop user should know. Microsoft shared it as its Friday Windows tip of the day, and it’s all about scrolling.

If you read long documents, browse webpages, or write code on a Windows laptop, you know the drill. You swipe, lift your fingers, and swipe again until you reach the part you need. Microsoft has three touchpad features that cut down on all that swiping.

Read more
This app turns the iPhone Duo into a pocket-sized Mac
Headroom uses macOS's built-in Screen Sharing to let users control their Mac from the foldable iPhone, without installing additional software on the computer.
Headroom App on Apple iPhone Duo controlling a Mac

The iPhone Duo has barely been out for a day, and someone's already found a way to run their Mac on it. Well, sort of. Headroom, a remote desktop app from MusicFlow, lets users control their Mac from Apple's new foldable phone, with an interface that adapts as the device opens, closes, and folds in half. Think of it as turning a pocket-sized phone into a miniature MacBook, minus the actual MacBook.

Headroom turns the iPhone Duo into a mini Mac

Read more