An interesting concept, but the Ultrabook to date has been more a premium product Intel is trying to cost reduce to mainstream markets.
The performance and peripheral set at the IHS BoM would not make for a very interesting user experience.
Even Google's Chromebook Pixel which removes the Windows OS tax and a hard drive still sells for $1,200-$1,500 for a premium Web experience.
Is that chart listing the manufacturer's cost for those components or what would be a retail (to the customer) price for the components? If it's manufacturer's cost, that's a $600 - $800 retail device.
That said, I would think that a $200 retail laptop of sorts is quite possible. Looking at retail purchasable products, a PC can be put together at around the price point you're looking at. The compromises would need to be in the mechanics and the software.
With something like the Raspberry Pi at the heart, the OS would need to be thinned out a bit in order to get an acceptable level of performance for household tasks. In my opinion, the biggest problem with the first generation of net-books was the OS overhead. Take care of that and it's viable. Don't and it's not.
$200 seems quite an aggressive target for an ultrathin mobile PC. Instead, perhaps they should focus first on making those new generation notebooks roughly the same price as a tablet. I also agree with Rick that IHS's BoM seems very basic, and that most users will demand a more powerful and more full-featured machine.
It's a good thing Android doesn't rely on Intel and x86 chips at all, so it won't matter if Intel can do it or not. ARM chip makers will do it, and have already done it (hello Samsung Chromebook! - granted it's not Android, but same thing)
As Rick & Frank pointed out, a 200$ ultrabook with windows will be severely under performing device. whats the point of having such a device?
On top of that, why would Intel want to build such a device? to kill their 200$ CPU business with 20$ CPU business?
Wow, that would be an agressive price point alright. Perhaps what everyone is struggling with here is that Intel needs higher prices just to stay afloat. So perhaps, by inference, Intel should maybe be looking at changing its busness model entirely? To one of developing IP?
As to the repeated comment that PC sales "will never recover," please refer to the Windows 8.1 article in today's EE Times. It's only a matter of semantics. People need to broaden their perspective on what they call a PC, that's all. In the article on Win 8.1, check out how college students use Win8 "tablets," connected to a docking station, to do all their real work.
That system with a $200 BOM is not a $600 notebook, it's a $300 notebook. Hardware profit margins are pretty thin, both for manufacturers and for sellers, unless you happen to be Apple.
But it's also not Good Enough, at least not if it's going to run Windows. You really want at least an 11" display at 1366x768 resolution in the notebook form factor (where the screen is farther away than a tablet screen) and 4GB RAM. And you didn't include the cost of a Windows license. At that point you have an ASUS VivoBook X202e (except that it has a Core i3 instead of an Atom), which is currently selling for $400 and is a nice value at that price. Swapping in an Atom would give you less speed and more battery life and probably not change the price a lot; some users might prefer that tradeoff.
I don't think we can get to the $200 price point right now and build a compelling Windows notebook computer; we'd just be back to Netbook Redux. Perhaps in another year or two, as costs continue to come down. Meanwhile I think there are some sales to be made by $400 notebooks.
Alternative platforms like Chromebooks can reach lower price points because their hardware needs are smaller; 2GB and a small SSD are enough. It also helps that Google is willing to sell them at cost.