Final Words

The move to ARM doesn't fundamentally change the performance or usability of the Chromebook. It's still a slow (relative to more expensive notebooks), limited use notebook. That's not necessarily a bad thing, but you need to be aware of what you're getting into with this thing.

The display is functional but not very good. Build quality is decent but the new Chromebook is still a plastic notebook. Thankfully the keyboard and trackpad are both pretty decent.

For a basic web browsing and glorified typewriter platform, the Chromebook really does get the job done. It's a very focused, simple device that serves its purpose well. I don't know how big the Chromebook niche is, but Google has targeted it very well. At $249 if you need a physical keyboard and only need basic web browsing support, I don't know that there's a better solution. As an almost-disposable notebook for writing and browsing the web, I'm happy with the Chromebook. Good ultraportables are much more expensive, and even low cost PCs don't come with any sort of solid state storage. Admittedly the eMMC solution in the Chromebook isn't setting any records, but it does deliver consistent IO performance which is more than I can say for a cheap 5400 RPM 2.5" hard drive. 

Ultimately the Chromebook puts pressure on the traditional PC notebook ecosystem from below. Tablets have been doing that for a while but they are a physically different form factor. For those who still want a traditional notebook form factor, there hasn't been much low-cost competition. The Chromebook applies a good amount of pressure there. Compared to a $1000 notebook, the Chromebook's display isn't great but move down into the $500 - $700 range and it doesn't look all that bad thanks to many PC makers failing to invest in good quality panels. If a $249 Chromebook delivers a competitive keyboard, trackpad, display resolution and quality experience to your $499 PC notebook, it's time for a change. With Apple pushing at the top and Google working the bottom, the hope is that the entire PC notebook stack gets better.

On the SoC side, our first look at ARM's Cortex A15 is quite positive from a performance standpoint. I'm still not convinced on its power consumption for smartphone use (big.LITTLE must exist for a reason) but from a performance standpoint, it's going to make current smartphone/tablet SoCs feel very slow. And that's something we can all look forward to in 2013.

Battery Life & Power Analysis
POST A COMMENT

149 Comments

View All Comments

  • jeffkro - Wednesday, October 31, 2012 - link

    Krait, might loose out to this A15 in all out performance but it is pretty great for battery life especially on LTE phones. Reply
  • Krysto - Thursday, November 01, 2012 - link

    LTE has nothing to do with the chip. Those quad core S4 Pro's aren't even integrated with LTE yet. Latest LTE efficiency comes from the fact that it's made at 28nm, not at 45nm - the LTE chip itself that is. Reply
  • tipoo - Wednesday, October 31, 2012 - link

    Impressive that the A15 beats Atom across the board, and draws significantly less power doing that. Complete leapfrog. Reply
  • eddman - Wednesday, October 31, 2012 - link

    Not a valid comparison IMO when it comes to power consumption. That's an old 45nm Atom. Clover trail will do much better, thanks to the 32nm process and its new S0ix power states. Reply
  • tipoo - Wednesday, October 31, 2012 - link

    Fair enough, but the performance per clock won't go up significantly until Silvermont. Reply
  • silverblue - Wednesday, October 31, 2012 - link

    Still, the current Atom and Brazos are only mild updates (if that). The E2-1800 in HP's dm1 gets 547 in Sunspider, so clock-for-clock, Brazos 2.0 is ahead on this single test. However, the power consumption must be signicantly more. Still, slightly different markets.

    I would very much like to see the next Atom plus AMD's Jaguar; reduced power and better performance could make for a decent A15 competitor.
    Reply
  • lmcd - Thursday, November 01, 2012 - link

    Supposedly these were the leaky chips, so we could have a similar story in Samsung's favor. Reply
  • Krysto - Thursday, November 01, 2012 - link

    Brazos has a much higher TDP though. Was it 10W for CPU and another 8 for the GPU? It's crazy. Reply
  • Krysto - Thursday, November 01, 2012 - link

    Clover Trail is not a mobile chip. Only Medfield is, and they don't even have a dual core version yet. You will never see Clover Trail in smartphones (yet you will see Cortex A15), and the only reason you will see them in Windows 8 tablets is because Medfield would be too slow with Windows 8, and Clover Trail is more powerful, but also uses much more power, and they are willing to compromise on that, while lying through their teeth that it will get 8-9h of battery life. It will get nowhere close to that. Reply
  • wsw1982 - Thursday, November 01, 2012 - link

    the clover trail is about 3.5+ w tdp according to some review. the samsung a15 is a 6+w soc. what make you think a 3.5 w soc cannot go to mobile, but a 6+w soc can? Reply

Log in

Don't have an account? Sign up now