This could be a new generation for the Junkyard Wars type of show, and I'm excited at the prospect. On the other hand, poor implementation or chasing a wide audience might lead to a lot more human drama than necessary and less technical topic.
Looks like Intel yet again trying to convince the world they're not being left behind by the Internet of Things. For those who aren't aware, the Intel Quark the show centers around is a die-shrunk and slightly modernised 486 that's trying to compete with modern ARM Cortex-M designs with predictable results. (Worse power consumption, worse IPC, pretty awful all round.) Hence PR exercises like this one.
And a terrible toolchain to boot. Intel has no clue what their plans are in this space. I heard the galileo team was like two full time people last year. I own both the galileo and the edison. They are such a pain to use that I hardly ever mess with them.
The Galileo was advertised as low-power, but the reality was rather uglier: https://learn.adafruit.com/embedded-linux-board-comparison/p... (And remember, that's comparing it to the original RPi with its already-ancient main CPU and inefficient linear voltage regulators. What's more, take a look at the performance comparison on the page before. Nasty.)
Also, the competing chips are fast enough and low-power enough that they essentially are dedicated pattern recognition engines with the right firmware. If you look at for instance Apple's M8 motion processor, that's just a Cortex-M3 chip, and not even a particularly power-efficient one by modern standards. I wouldn't even be surprised to find that Intel's just embedding a competing microcontroller core as their pattern recognition engine; it's how everyone else does it.
Makes the QFN packages of all the competing chips look positively hobbyist-friendly. (This seems to be aimed at the same market as stuff like the Nordic Semi nRF52 series and the TI CC2650. Which still aren't trivial to build boards for, but quite a few makers have managed it with the older nRF51822.)
I wonder if it's even possible to build a PCB for this without using expensive microvias.
I couldn't find much data on the Curie module, but it looks like the only pins required are power and slow io, so I can't imagine it's a super dense bga or anything. Even on the high end Intel socs with over 1k balls a good designer can escape all the signals without microvias.
Okay, seriously, who's with me to pitch the first Hackathon Reality show based on MasterChef and The Voice that features 48-hours of drama, suspense, beautiful graphics and working app at the end?
Has been tried for a game jam. Exploded spectacularly, mostly because the production company didn't understand the game dev culture.
This polygon article has an overview and links to individual positions: https://www.polygon.com/2014/3/31/5568362/game-jam-reality-s...
Or the Y(C)-Factor, where they just televise the YC process.
Given the bizarre effect the X Factor had on the entire ecosystem it operates in you could make a very strong case that it would be a massive net benefit to all participants.
I participated in a startup weekend hackathon that was just about that. They recorded the entire weekend, and specifically pulled people off our team for 'interviews' (which was really a blessing because then we could reach a decision and just move on). We won the hackathon with a beautiful working app. I don't think the show go picked up anywhere.