Intel logoIntel is just starting to get its feet wet in the smartphone space thanks to new low power Atom processors. But one day the chip maker hopes to ship 48-core microprocessors designed for smartphones and tablets.

And according to ComputerWorld, that day is around 5 to 10 years away.

If you look at the pace of advancements over the past few years that goal seems pretty reasonable. In fact, by the time 48-core smartphone chips are available, they probably won’t seem all that revolutionary. After all we’ve moved from single core to quad-core chips in mobile devices in just the last few years.

In fact, it’s only been about 5 years since the first iPhone went on sale, and today that device looks pretty dated.

While 48 processors cores might sound like overkill, there are good reasons to add multi-core capabilities to chips. It allows a phone, tablet, or other device to execute multiple instructions simultaneously without sharing resources. This leads to faster performance, better support for multitasking, and lower power consumption (since multiple cores can run at lower voltages than a single core trying to perform multiple tasks at once).

That doesn’t just mean it’ll be easier to listen to music while surfing the web. Applications designed to take advantage of multi-core chips could benefit as well. For instance games could use different processor cores to render different graphics effects, or a video player could decode multiple video frames at once.

While Intel marches towards 48-cores, I don’t expect competitors ARM or AMD to sit still. All told, the phone and tablet space could look as different 5 years from now as it did the day before the iPhone was unveiled.

Support Liliputing

Liliputing's primary sources of revenue are advertising and affiliate links (if you click the "Shop" button at the top of the page and buy something on Amazon, for example, we'll get a small commission).

But there are several ways you can support the site directly even if you're using an ad blocker* and hate online shopping.

Contribute to our Patreon campaign


Contribute via PayPal

* If you are using an ad blocker like uBlock Origin and seeing a pop-up message at the bottom of the screen, we have a guide that may help you disable it.

Subscribe to Liliputing via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 9,532 other subscribers

14 replies on “Intel wants to launch 48-core chips for phones… tablets within a decade”

  1. Too bad consumer software isn’t taking advantage of multiple cores as fast as hardware companies can add them. Many professional and scientific processing can be parallelized but most consumer applications can’t take much advantage of it beyond what’s mentioned in this article but those tasks tend to be offloaded to the GPU for parallel (the Tegra 3 has 12 cores) or dedicated processing.

    The OS can make use of multiple cores for running multiple apps and background services but mobile OS’s seem to take the route of minimizing background tasks to save power. So, again, only the foreground app will likely make full use of the cores and these apps are probably for specialized processing outside the consumer space.

    So, I guess, these 48 core tablets and phones are probably going to be targetted towards non-consumers or consumers who like showing off.

  2. “Intel wants to launch 48-core chips for phones …”

    I think what “Intel wants” is some attention and credibility in the area of phones and tablets. Right now they don’t have any. So they gave an interview to Computerworld about what they will do 10 years from now. As if they actually know. By then they could be just another ARM licensee.

    1. Intel used to be the biggest ARM licensee not too long ago. They may still have a license, although they may not be making any actual ARM CPUs. I suspect their license is similar to Qualcomm’s in that they can make pretty much whatever they want using ARM intellectual property.

        1. Yesterday ARM announced 16 cores very low voltage CPU for 2014 with ARM Cortex A50, today Intel announce 48 core for 2022, well, still a bit late for intel ^^.

  3. This makes even more sense if you think of that device less as a phone than as the replacement for your laptop/desktop.

    1. But as I said in my comment below, it’s not going to be — not on it’s own, at least.

  4. Most of the applications for these powerful multiprocessor chips haven’t been thought of yet.

    For example, to make self-driving cars a reality, we’ll need all the computing power we can get. The data field around a vehicle can be divided up into sectors, the more cores there are, the finer the sectors can be.

    Another example is developing portable search-and-rescue aids. A multicore device can drive multiple sensors to determine where trapped victims are.

    I’ll bet that military device makers are just salivating for this technology.

  5. Good points about multitasking and power consumption, but the idea of a 48-core processor does raise an interesting question of just how much processing power a mobile phone or tablet needs in the long run. Certainly, graphics capabilities (for games, “enhanced reality”, VR, etc.), and the efficient handling of HD photos and video still have some ways to do, but given the limitations of the small form factor (esp. for mobile phones) there are limits to what is needed.

    For just about everything else, it’s a question of data crunching, and that’s best done where the data is — in the cloud. Assuming the monopolistic tendencies of the telco industry is broken (and they won’t give up without a fight) then a decade from now the communications logjam will have burst and data will be fast, cheap, and ubiquitous, reducing any lag from cloud-based processing to be kept to a minimum. Google understands this, which is why they have so heavily invested in the cloud and are trying to ensure people will have fast access to it (Google Fiber).

    Thus our mobile devices won’t need to be powerful number-crunching devices like today’s high end desktops, they only need to the terminals, the presentation devices, albeit somewhat more sophisticated than the phones of today.

    1. I think the point is that processing power, $/cycle is going to be reduced drastically within the decade. The low-end stuff will be quad-core, octa-core. We’ll finally have “professional” processing power inside our pockets and we’ll be able to ditch the huge towers (perhaps some wires too). The cloud will be a big part of it, but I think there’s a trend towards de-centralization of computing power. Previously server-grade (multi-core) processors were expensive and hard to assemble, but the new cheap integrated solutions (as they increase in processing power) will be much more available to everyone. You’ll be able to build your own cloud without handing over your information to someone else’s server.

    2. “but given the limitations of the small form factor (esp. for mobile phones) there are limits to what is needed.”
      Phones won’t be just phones anymore. See my post above.
      Also, you can imagine a very small device (cpu/gpu and somme antennas) that can fit in a phone/tablet (or hybrid)/All-in-one shell.
      This concept exists already but is in its infancy.

    3. “but given the limitations of the small form factor (esp. for mobile phones) there are limits to what is needed.”

      5-10 years from now it’s much less likely that mobile devices would be as limited by their form factor as they are now.

      Something like the Asus Padfone may even seem quaint by what we may have 5-10 years from now.

      While there are many things yet to be developed for mobile usages.

      Something like Siri is still little more than a voice interactive search engine. But 5-10 years from now this kind of technology could start allowing for smart assistants for enough limited A.I. for more interesting applications than are possible now but offloading to the individual devices would allow for personalizations not possible for a all purpose cloud solution.

      While things like augmented reality really involves quite a bit of multi-tasking that today’s devices aren’t really up to handle extensively.

      It may take a few more years after introduction of actual products with these type of processing capabilities but eventually we always find uses for what devices are capable of doing…

    4. “Thus our mobile devices won’t need to be powerful number-crunching devices like today’s high end desktops, they only need to the terminals…” Maybe so, but I think the pendulum will continue to swing back and forth between light-weight clients and heavy-weight ones. It is hard to predict what kind of horsepower a client computer (as opposed to a server in the cloud) will need ten years from now.

Comments are closed.