That would be interesting to know. Helium is remarkably low-density by the standards of inert gasses(~.18grams/Liter vs. ~.9 for Neon and 1.25 for the ignoble-but-pretty-well-behaved Nitrogen), so it may actually be substantially different from any of the other options.

$6,000? AMD's 64-bit-ARM Opteron A1100 reference board is half as much and (while obviously not very 'embedded') is a rather bigger piece of gear...

It isn't 100% clear: All 'Silvermont' SKUs appear to be 64 bit capable, and this board has a Silvermont-based Atom on it(two of them actually); but Intel's Edison Native Applications Guide definitely appears to be walking you through setting up a build environment for 32 bit x86.

Whether that means that Intel actually lasered support for 64 bit execution off when they were designing this chip(which isn't like most of the other Silvermont devices, which have a GPU and more PC-style I/O, so clearly some cutting was done), or whether it means that they decided to default to 32 bits for everything because the device only has 1GB of RAM soldered on and no support for more, is not clear from anything I've been able to find.

TFS is simply incorrect; but may have been confused by Intel's "Galileo" board, which is based on Quark (at 400MHz). Curiously, 'Galileo' is much more Arduino-esque (a bit more raw punch, weaker hard-real-time bitbang); but also has a full PCIe lane(brought out on a miniPCIe connector) and 100Mb ethernet(optional PoE in Gen2); but no RF.

The much more PC-like, or at least BB Black/rPi-like Edison has a substantially punchier CPU; but no PCIe, wired ethernet, and includes wifi and BT.

I'm not sure if we are misjudging intel through the lens of existing products, or if Intel can't quite decide what niche to hit; but I find this mix a trifle confusing personally...

The teeny little one has the high powered CPU (relatively speaking); but not the high speed expansion bus or wired networking and PoE options. The relatively big one has high speed expansion (but severely limited GPIO, making compatibility with MCU projects that depend on bitbanging rather tepid); but the weak CPU and limited RAM.

I'm interested, and it's always worth keeping an eye on Intel; but I'm a bit confused about what they are aiming at here.

I'm definitely open to suggestions if you have any cool ones to point me to; but I've had some fairly unpleasant experiences with a lot of the theoretically-capable-and-crazy-cheap 'android stick' type sticks and boards. The rPi is indeed quite feeble; but what I/O it has is actually mostly broken out, and (while not exactly FSF-level 'open') is crazy open by the standards of something broadcom based, and enjoys a very strong community of people who can answer questions about any obscure points.

The beaglebone black is a bit more expensive and supply-challenged; but has nicer networking, a bit more punch, ans the SPUs.

Do you know of some things that exploit the various unbelievably cheap smartphone parts; but don't require a working knowledge of chinese datasheets, or have absolutely all signals but HDMI exposed in lilliputan test points? I imagine that there are some; but the couple of times I've dived in without too much research I've ended up with something that...mostly...runs the included Android-build-of-mystery; but is otherwise effectively unusable.

(There are, of course, some nice options over in router-land, in terms of price and network connectivity, sometimes enough GPIO for light work; but those aren't going to be doing much video anytime soon.)

Intel has the Product Brief up. Unless they specifically decided to hide the fact (which would seem unlikely), video appears to be absent.

Wireless connectivity looks pretty nice (wifi and bluetooth out of the box, though BTLE is mysteriously 'in Q4-14', which makes one wonder if perhaps the driver situation has a few grim little stories that we should know about...) and the inclusion of both 2x Atom cores at 500MHz and 1x Quark core at 100MHz is potentially interesting, depending on how easily and elegantly the Atom and Quark 'sides' of the device can talk to one another and either share control of, or at least transfer control of, the various I/O lines.

Not going to shove the rPi out of the way for video-pushing applications, and I suspect that PWM and bitbang-heavy applications may still be underimpressed by Intel compared to the classic microcontroller options; but it could certainly be a contender for a lot of the 'arduino-connected-to-the-network' applications which don't lean too hard on squeezing every last drop out of bare-metal-MCU work; but which could really use a bit more punch on the networking and storage/logging side.

Certainly nothing obvious comes to mind, in terms of features.

The only class of complaint that would be nice to see addressed (not exclusive to the heart, also covers skeletal muscle, bone density, tissue healing, and assorted metabolic processes) is that most of the parameters are still tuned for a relatively high-exertion, high-risk, environment with a strong risk of at least occasional malnutrition. As your basic first-world 'press buttons at work, hired because he knows what buttons to press' type, I can afford plenty of calories and adequate supportive care if injured. Since that's the case, storing spare energy on my waistline is largely unappreciated, maintaining muscles at relatively feeble standby states, rather than defaulting to growing them for when I do have use for them, and prioritizing speedy; but scar-prone, healing over slower but more elegant regeneration are all kind of annoying.

In the same boat; but more of an issue for women, is calcium handling: bioavailable calcium is cheap as dirt these days; but we still suffer skeletal embrittlement over time.

I can't think of any big, dramatic, neat features, just a variety of optimizations that were adaptive under a somewhat different environment than the one I now live in that would be nice to bump a bit.

What? The NHS pays for is medicines and technology at prices negotiated with the pharmaceutical companies. It doesn't get them for free.

It is potentially useful data; but the trouble is that detecting 'NSA-like' activity is just plain hard.

A large-scale exploit attempt (while it is something that an intelligence agency might try, under certain circumstances) is really what you'd expect from someone with purely commercial interests: Find a nice bug, try to hit a lot of targets as fast as possible and cash out before the guys playing defense (or your competitors) catch on to the new toy and either the targets start to harden or your competitors start cleaning them out before you can get them.

An intelligence agency, on the other hand, has less use for large numbers of low-value compromises; but likely has a much shorter list of very high value targets that would receive attacks targeted with considerable care and precision and tailored to be minimally intrusive(if the purpose is observation) or maximally damaging(if it's a Stuxnet-style sabotage operation). Such uses would be unlikely to show up in a broad survey of mostly-low-value targets; particularly if the survey requires any cooperation on the part of the site operator, which is more likely in the case of random commercial outfits who depend on security vendors, less likely in the case of paranoid high profile targets.

I'd hate to have one sneak up on me, leaving me mentally incapacitated before I knew what hit me; but I'd say that "enjoy the bodily condition of my early 20s until cognitive indicators start to look worrisome, put affairs in order, administer euthanasia procedure co-developed by veterinary experience and recreational pharmacology enthusiasts" is a strategy that would easily exceed virtually all historical quality-of-life outcomes...

That statement sounds like gutless PR posturing: Obviously you want to learn as much as you can from a failure, especially an unexpected one; and make appropriate modifications based on what you find; but there's a reason why people say mean things about sample sizes of 1. It's entirely possible that you won't be able to nail down exactly why things went wrong in a single unit. Given the continued supply of patients who Will Die, Period, without replacement organs that aren't available, it hardly seems worthwhile to stop just because you lack a perfect determination (so long as you do, of course, make whatever improvements/modifications your imperfect determination suggests are needed).

You aren't going to make progress without experiments, some risky; but you have a population of (adult, of-sound-mind) patients who will definitely die without intervention. You don't want to act in reckless disregard for human life; but when you've got people who will definitely die and the possibility of producing improved treatments, excessive avoidance of uncertainty is disregard for human life. It's a pity that the PR flacks didn't have the guts to say that.

Certainly. If you lack an important organ, even a pretty lousy reproduction looks like a great idea (see also, kidney dialysis, diabetes treatment, etc); but we have a rather long way to go before any transhumanist puffery about 'advantages' becomes justified. Even reaching approximate parity with a normally functioning organ is a pretty heroic challenge. If you don't have one of those, your standards are lower; but you aren't getting an upgrade.

Given the (lack of) alternative, it's certainly an improvement in this case. I was objecting to TFS's "More than just pumping blood, future artificial hearts will bring numerous other advantages with them. They will have computer chips and wi-fi capacity built into them. We'll control our hearts with our smart phones, tuning down its pumping capacity when we want to sleep, or tuning it up when we want to run marathons." blather.

Since there aren't enough donor organs, and those tend to have their own problems, a deeply imperfect artificial apparatus beats the alternative; but characterizing its defects as 'numerous other advantages' is just idiotic. It makes about as much sense as describing an insulin pump as an 'improvement' because you can arduously fiddle with delivery profiles, rather than using lame, old-school, feedback systems.

When the alternative is death, many imperfect substitutes are good ideas; but that is very different from their being 'improvements' except in the strictly local case of patients in markedly worse than normal shape.

There's also the fact that having a control interface isn't really 'cool' as much as it is 'we couldn't duplicate the sensory and feedback systems that made the old one adjust itself automatically, so here's a dial to futz with!'

Compared to out-of-control, control is nice; but compared to 'just fucking works' it's a thankless chore and a good opportunity to make mistakes.

Flattering; but I suspect that he would have actually done the math, at least to the level of plausible approximation, rather than handwaving that part in favor of some dreadful puns.

That said, if anyone with a knowledge of railguns wants to calculate(or test, be sure to record) exactly how lousy a salmon would be as ammunition, I'd read the hell out of it.