Embedded Tech Journal on Atom and hamsters
Embedded Technology Journal currently has two articles that caught my attention.
One very nicely discusses the Good, the Bad, and the Ugly of Intel's new Atom processors which are designed to bring x86 technology to more embedded scenarios. Atom certainly makes some very interesting designs possible, see this new Kontron board, e.g. And from a Microsoft perspective, one of the greatest advantages of Atom is that it can enable Windows Embedded Standard to run on new, smaller, power-aware device classes which were typically driven by Windows Embedded CE - which has its own advantages, but is somewhat harder to program, and only has very limited Windows compatibility. (And we're confident to have the Atom chipset drivers in the box when WES ships.) So, for me, the Good in Atom outweighs the Not-So-Good by far.
The second article, titled "Why Hamsters Cannot Save the Planet", discusses what the embedded industry, and esp. chip makers, are doing to help reduce power consumption of devices. They are taking practical approaches and discuss why we don't need wishful thinking about all kinds of renewable energy sources (incl. hamsters producing electricity by spinning their little wheels) which cannot produce even a significant fraction of the energy from fossil fuel which they are supposed to replace, but instead we should apply clever engineering to make sure our devices consume less energy in the first place, without sacrificing on functionality or comfort. This is, btw., another very welcome (at least to me) voice of reason emphasizing that we need "numbers, not adjectives" in our discussion of how to solve the energy crisis that we currently seem to face. Just a few days ago I found this article about a book by David MacKay that is currently being written, in which the author really puts some serious calculations behind typical scenarios for replacing fossil by alternative energy - and finds out that most of what's suggested is simply utopian. Recommended reading.