Embedded x86: Intel Atom Versus Via Nano, Which Way Are You Gonna Go?
I swear to you, when several months back I picked the topic of last Thursday’s cover story, I didn’t realize that its publication date would predate by only a few days the start of Computex, one of the world’s largest PC-focused trade shows! Truth be told, in fact, I didn’t cop a clue about Computex’s timeframe proximity until I got to Taipei last week and saw all the AMD banners hanging from light poles. In my (limp) defense, my limited travel budget precludes me from attending the show, so its dates don’t hit my radar screen like those for AES, CES, IDF, NAB, SIGGRAPH, WinHEC, etc…
But how timely, huh? Computex predictably, acted as the coming-out party for Intel’s Atom processor, and OEM partners (or, as Intel referred to them as in the pre-show teleconference I attended, ‘ecosystem co-travelers’…ahem) showcasing prototype Atom-based systems were present in abundance. MSI’s Wind (running a pre-production dual-core version of Atom) looks particularly tempting to a long-time thin-and-light system user like myself (here’s anuddah looker), and ASUS is also aggressively converting its Eee PC product line from Celerons to Atoms. Although Atom looks to be a winner, in small-form factor computers today and perhaps in smartphones in the future, I’m still hung up on the thought I raised in my cover story:
Perhaps the biggest question on the company’s road map for the remainder of the decade is the degree to which Atom will cannibalize existing Intel CPUs in a manner that is fiscally unattractive to the company, instead of broadening the overall x86 market at the expense of competitors, such as ARM, as Intel hopes.
And then there’s Via’s Nano product line, referred to in my article by its ‘Isaiah’ project moniker. Although initial product sampling is in progress, judging from a few (contradictory) benchmark test suite results which have hit the Intertubes, it’s not clear to me just how ‘real’ the product line is at this point. Few ODMs and OEMs were reportedly showing Nano-based systems, and Via itself unveiled an "open-source" subnotebook reference design one day prior to the Nano launch…based on the prior-generation C7 CPU architecture. Very confusing. Although ASUS’s chairman was publicly skeptical last week about Via’s chances at breaking into the PC in a big way, I personally wish the ‘scrappy underdog‘ the best of luck; as I’ve said any number of times, vigorous competition is beneficial to suppliers and consumers alike.
Check out Ars Technica’s Atom-vs-Nano spec-comparison analysis, which echoes another comment I made in last week’s writeup:
Ironically, as Intel has stripped out-of-order execution and other superfluous features from its previous architectures, thereby resulting in an approach somewhat reminiscent of Centaur’s nearly decade-old vision, Glenn Henry, president of Via’s Centaur subsidiary, and his design team are poised to unleash Via’s first three-way superscalar out-of-order architecture.
Here’s anuddah. And anuddah. While I aspire to do a competitive analysis against a Via Nano-based board, pragmatically I’m not holding my breath, considering that my last experiment with the company’s hardware was…underwhelming (ahem). Nonetheless, I remain eternally optimistic; stay tuned to this space for any updates. I’ve also got a Via C7-based ARTiGO box that I hope to fire up soon (check out Microsoft’s ARTiGO-centric presentation and reference Windows CE build from the Embedded Systems Conference). And in the interim, I welcome your Intel-vs-Via thoughts!
p.s…in light of the fact that my C7-based NAS finally flamed out a few weeks back, I thought you might enjoy seeing a similar no-cooling experiment which began in mid-May. As of early last week, when I previously checked out the live video feed, the system was still kicking, but this morning the video stream is dead. Network problems, or failed Pico-ITX hardware?