Ultrabook: Behind How Intel is Remaking Mobile Computing

Heading Into the Fourth Generation

Touch issues aside, Intel knew that the key to bridging the tablet and clamshell worlds would lie in system resume performance. This is, in part, why Intel required a drop in resume-from-hibernation times with the touch-enabled Shark Bay—from seven milliseconds down to only three. Ultimately, to serve as a tablet, the PC would have to wake up like a tablet. That includes the idea of “fresh data,” so when a user opens his or her convertible in tablet mode, information such as current email is already refreshed and waiting.

“The second major vector [after touch] was around responsiveness,” says Rob DeLine. “This is not the speeds and feeds of the processor and graphics. Responsiveness is about how this thing wakes up and whether it feels snappy when I’m in the flow doing my workload. We spent a lot of time and reliance on using solid-state memory devices and cache to get the system to wake up fast and multitask quickly, so it would snap between applications.”

Over 90 Ultrabook designs arrived with the Ivy Bridge generation. Now, with Haswell and the fourth generation of Ultrabook, we have more in store. Although Haswell left us disappointed on the desktop, we already knew that the architecture's feature set was molded around certain mobile goals. Improved efficiency means that a battery's capacity goes that much further, even before taking new design elements into consideration. According to Intel, most devices will see over nine hours of runtime, and some are already peaking over 12 hours. Intel bills this as a 50% battery life leap, the largest such jump in its product history. To put it in perspective, the Ultrabook platform figured on a 35 W TDP envelope in 2010. Looking into the Bay Trail generation now launching in the second half of 2013, that number plummets to 6 W.

If you caught Chris Angelini’s June deep dive into Haswell, then you’re already up to speed on the architecture's subtle improvements to IPC and faster graphics. All of that feeds into the next generation of Ultrabook (except for ultra-low power models that will be based on the Silvermont Atom design). Expect the number of models available with touch to triple in this generation. For the 2012 holidays, there were five 2-in-1 convertible Ultrabooks on the market. This year for the holidays, expect more than 50.

What else? Fourth-gen Ultrabook will have Thunderbolt—better late than never. You’ll get the Intel Identity Protection and Anti-Theft features enabled via the security improvements in Haswell. (Since I’ll be carrying my two-pound Ultrabook around and forgetting it everywhere, or having it filched by my kids, I want an app that can help me locate it inside my house.)

Not least of all, new Ultrabooks will ship with Intel WiDi built in. WiDi is Intel’s spin on the Wi-Fi Direct and Miracast standards. In a nutshell, WiDi provides an easy way for devices to stream content directly to a TV, monitor, or similar supporting display device. Samsung has dabbled with such capabilities in a proprietary way, but perhaps Ultrabook will push the technology to its tipping point and into the mainstream. Again, I should point out the importance of having big wireless pipes (particularly of the 802.11ac variety) in order to make this an enjoyable experience.

  • techtalk
    I am just half way through the article. I am compelled to comment here. "What an Article" Amazingly well written, superb flow and great content.
    Reply
  • outlw6669
    11271822 said:
    The battery always comes out first.

    Words to live by.
    RIP brave little Ultrabook.
    Reply
  • zodiacfml
    Nice Toms. It's so good....I wanted to read more.
    Reply
  • nibir2011
    Ultrabook aims to make its platform so compelling that, frankly, you’d be a fool to consider an under-performing, over-priced, feature-limited high-end tablet.

    It will be only possible if Intel and AMD goes hand in hand. Mobile sector is so lucrative to OEM that eventually they will go there until there is a great product. If intel only thinks about their own business then it will be like what microsoft did to desktop. No software developers do not want to make consumer application, as app developing is business friendly.

    It has to be a joint collaboration.
    Reply
  • nibir2011
    Ultrabook aims to make its platform so compelling that, frankly, you’d be a fool to consider an under-performing, over-priced, feature-limited high-end tablet.

    It will be only possible if Intel and AMD goes hand in hand. Mobile sector is so lucrative to OEM that eventually they will go there until there is a great product. If intel only thinks about their own business then it will be like what microsoft did to desktop. No software developers do not want to make consumer application, as app developing is business friendly.

    It has to be a joint collaboration.
    Reply
  • kartu
    Huge, fictional article on what is supposed to be a tech site.

    Everyone has a notebook. Most of them are more than fast enough.
    Now what can a company that excels only at CPUs do about that?

    It sure takes a genius to notice that people like lighter thinner thingies, right.
    I'm sure Steve Jobbs (I guess that's The Genius to the article's author) absolutely had to take part in this astonishingly far sighted decision to go lighter and thiner, it is soo far sighted, nobody else could have imagined that.

    People prefer thinner and lighter, cooler looking things... What a frucking surprise...
    Reply
  • williamvw
    11274396 said:
    Huge, fictional article on what is supposed to be a tech site.

    Fictional. I'm not sure that word means what you think it means.

    11274396 said:
    Everyone has a notebook. Most of them are more than fast enough. Now what can a company that excels only at CPUs do about that?

    That is an excellent question. You may wish to review pages 1 through 6 for answers. Pages 10 through 13 aren't bad, either. None of the content in them is fictitious, in case you remain unsure.

    11274396 said:
    People prefer thinner and lighter, cooler looking things... What a frucking surprise...

    Thank you for your input. Your skepticism is even more warranted than it is well-stated. Of course, just because people want things doesn't mean that those things actually exist. Or are affordable. Or can be serviced. I mean, at least that's the case in the real world. In fictional scenarios, where the Tooth Fairy delivers ultralight notebooks from the future, tiny companies can move product ecosystems with the same ability and effectiveness as large ones, and unicorns soar majestically through pink and purple treetops, I suppose anything is possible. In the real world, though, this article describes how things actually get done.
    Reply
  • superduper
    Although a well written article, I still have reservations regarding some of the context:

    The battery life "ballooning" had very little to do with Ultrabooks but rather silicon that was more frugal, particularly at idle and the density of battery packs. Your average Ultrabook often sacrifices Li-Polymer battery capacity to remain thin and svelte (MBA an exception). As a result, the 35W notebook with the bigger battery will get better battery life than the 17W ULV (vast majority of computing is spent at idle).

    The facial and speech recognition software seems very nifty, but it's still not something that's an Ultrabook specialty. The software is available to tablets and smartphones as well (the Moto X uses a specialized core specifically to handle the speech recognition).

    I'm far more interested in what Intel is looking to offer in 2015 for the Ultrabook platform than I am about the rather weak software additions that don't differentiate it. What is Intel going to offer me in exchange for the extra $200-$300 dollars out of my pocket for an Ultrabook? What am I getting beside a thinner chassis to warrant that much cash? In some ways, to me it seems like Intel and its OEMs have become victims of their own success. The Ultrabook is there to revive some lost sales to the mobile market, but outside of a higher price tag and a dedicated keyboard, I don't see what the bonuses are.
    Reply
  • williamvw
    11275162 said:
    Although a well written article, I still have reservations regarding some of the context:

    The battery life "ballooning" had very little to do with Ultrabooks but rather silicon that was more frugal, particularly at idle and the density of battery packs. Your average Ultrabook often sacrifices Li-Polymer battery capacity to remain thin and svelte (MBA an exception). As a result, the 35W notebook with the bigger battery will get better battery life than the 17W ULV (vast majority of computing is spent at idle).

    The facial and speech recognition software seems very nifty, but it's still not something that's an Ultrabook specialty. The software is available to tablets and smartphones as well (the Moto X uses a specialized core specifically to handle the speech recognition).

    I'm far more interested in what Intel is looking to offer in 2015 for the Ultrabook platform than I am about the rather weak software additions that don't differentiate it. What is Intel going to offer me in exchange for the extra $200-$300 dollars out of my pocket for an Ultrabook? What am I getting beside a thinner chassis to warrant that much cash? In some ways, to me it seems like Intel and its OEMs have become victims of their own success. The Ultrabook is there to revive some lost sales to the mobile market, but outside of a higher price tag and a dedicated keyboard, I don't see what the bonuses are.

    Excellent points, and I agree with all of them. You are absolutely correct about the battery issue. Intel has no influence that I know of over Li-Ion battery efficiency; it can only try to reduce the platform's drain on the battery that's there. That was the company's challenge: how to get every component to consume less power. Obviously, some have more leeway than others.

    I also share your curiosity about 2015, as I indicated in the conclusion. It might be fair to say that if Centrino's mission was to cut the Ethernet cord, Ultrabook's (at least initial) mission is to cut the power cord. The thinner and lighter business just goes along for the ride.

    To be totally honest, 2012 Ultrabooks were not enough to interest me. It wasn't different enough from what I already owned. But you have to start somewhere and implement change in stages. The first design that really grabbed me was the Yoga. The convertible thing works for me and my needs, and the design is superior to, say, a tablet wrapped in a keyboard case.

    Your big question, of course, comes back to MIPS, and this is really a religious issue. Do we want our MIPS in the cloud or on our lap? There are good arguments both ways. Obviously, Intel's substantial PC group prefers them in our lap. My daily struggles with Google Voice tell me that this is a worthwhile thing. Now, if carriers improve and whatnot, and I'm able to get the same class of perceptual computing performance from the cloud on my phone that I can get on my lap in an Ultrabook, I think the weight of judgment must finally fall in the cloud's favor. It's more efficient on all counts. (I'm ignoring security concerns for the sake of argument.) But when I'm using my phone to compose notes or story chapters or whatever, which I do every day, then all I care about is accuracy, speed, and my total productivity. If the Ultrabook effort fosters a notebook ecosystem in which I can get better results for my needs from a two-pound convertible, then I'm all about the convertible and totally behind Ultrabook. I'm selfish that way.

    In short, we may find that the Ultrabooks of 2015 don't offer you enough extra value to justify your extra $200 or $300. However, I'd wager that at least some of the benefits you will enjoy in your non-Ultrabook, mainstream laptop of 2015 would not exist at their then-current level of development without Intel having made the investments in Ultrabook I've described in this article.

    And what if Google and Apple and whomever manage to saw Intel's legs off and leave the notebook paradigm in the dust? Well, that's how it goes. The market decides what has value and what doesn't. That trend has already started. The question now is whether it will continue.
    Reply
  • williamvw
    11277068 said:
    How much did Intel pay you to write this ? , we all know that ultra books sales are below freezing
    Oh, my gosh -- ANOTHER accusation of bribery! How novel! Well, since you managed to deduce that much on your own, geeze, lemme think... How much did they offer to pay me? Oh, I remember! It was http://www.youtube.com/watch?v=-DJtHL3NV1o!!! Because that's how globe-spanning $115 billion companies get to $120 billion, by putting their reputations on the line and bribing little journalists like me to write articles about historical developments just like this one.
    Reply