LimiFrog - Ultra-compact prototyping. For IoT and much more.

[xavier-c]

Hi there -

I would like to tell you a bit about project "LimiFrog".
It is an ultra-compact, fully-featured programmable module to prototype and develop smart objects.
It is programmed in C (and a porting of MicroPython is in the works).
It's based on a very low-power flavor of the ARM Cortex-M4 (STM32L4) and includes an OLED screen, 64Mbit data storage, lots of sensors, BlueTooth Low-Energy (4.1), a LiPo battery.
Weight is under 25 grams.

It's live on KS now:
https://www.kickstarter.com/projects/765344573/limifrog-ultra-compact-prototyping-for-iot-and-muc

Hope you'll like it,

Xavier

[matseng]

Why are so many obsessed with smallness?  For prototyping and hacking I prefer parts & modules that are a bit larger and more easy to both handle, connect to and modify...  Unless you're doing a prototype that *must* be almost the same size as the final product of course.

[poorchava]

Agreed,  I'm now developing a product which is roughly of a size of a multimeter and first prototype was roughly the size of A3 sheet of paper.

[xavier-c]

Hi Matseng,
You are right of course, in some cases it's better to have larger parts.
But, as you note, in some cases you do want a demonstrator or prototype that is about the same size of the (small) final product. Or you're building something you want to squeeze into say a drone or a small robot. That's what this module can address. Not fit for all needs, but interesting for some I guess?

[tggzzz]

If you are pushing the "ultra compact" angle, then you should prominently state the dimensions. Let's stamp out meaningless adjectives!

You have more meaningless statements with respect to the battery lifetime: you mention hours, days, weeks and months without stating how large the battery is (are you presuming 55AH car batteries?)

You mention the current required by the processor - but that's irrelevant. What matters is the current required by the board.

No, I'm not going to watch the video - I've no idea whether it would answer any questions.

[farsi]

I like that the board uses an STM32. Potentially, it could be an easy point of entry for ARM programming.

Also, the connection with a real time operating system sounds interesting, but no experience with that.

[tggzzz]

I like that the board uses an STM32. Potentially, it could be an easy point of entry for ARM programming.

Also, the connection with a real time operating system sounds interesting, but no experience with that.

If you want hard realtime, you should look at the XMOS processors. The compiler will guarantee how long each piece of code will take.

[nidlaX]

Having Bluetooth now qualifies for a description with the IoT keyword?

[Howardlong]

Having Bluetooth now qualifies for a description with the IoT keyword?

I have yet to figure this out either, but this isn't the only example, plenty of supposed IoT devices don't connect to IP networks directly at all. To be honest, like the term Web 2.0 a decade or so ago, IoT doesn't really mean anything, its use has become so wishy washy.

And we have yet to see the killer app for IoT, whatever IoT actually is.

[tggzzz]

And we have yet to see the killer app for IoT, whatever IoT actually is.

They will come. Any app that relies on a single corporate service that might disappear, along the lines of Microsoft PlaysForSure (sic). Any app or device that can be subverted by the malefactors, probably because the manufacturer didn't apply security patches.

Or maybe you meant the other definition of "killer"

[farsi]

indeed, the IoT hype still needs to deliver - in the meanwhile, Uber looks like an interesting case for the first "IoT success" http://www.mckinsey.com/insights/high_tech_telecoms_internet/the_internet_of_things_five_critical_questions