For comparison: Texas Instruments sell something called the EZ430 Chronos watch, which has a much lower-res display, a proprietary RF protocol with a USB transceiver, and a bunch more sensors including 3-axis accelerometers, pressure and temperature.
It is based on the MSP430 microprocessor which is much less powerful than this thing, its capped at 16mhz or something IIRC and only has around 8K of program space. It is however about a third of the price.
We choose one button to keep it simple (and quicker to get to market). You can do a lot of things with just one button, just look at the single button on the iPhone! For example, you can do two actions, like click and click-hold.
I've experimented with using a single button (a Griffin Powermate, actually) to control all of my electronics: one tap toggles the lights, three taps changes lighting scenes, press and hold/five taps turns the projector on or off, eight taps toggles "party mode," ... I'm quite convinced that multiple buttons would be better :).
Ah - sorry - I was after physical dimensions - i can kinda figure it from the videos, but I'd like to know if it would actually fit on my writs comfortably.
I really wish there was - I may get around to asking for this info, but I bet you've got to sign some type of license or pay fees. I don't know - I haven't tried.
Are you interested in making a case or something similar?
Hey, I know this must be there in the sdk docs somewhere, but can the knob on the side of the watch be used as an input device? Also, I wish I had more upvotes to give you.
EDIT: Oh, never mind, just saw that its a button, not a knob.
Just a heads-up: we've got a limited supply of the first batch (as some intelligent commenters below have deduced). If you'd like to get hacking soon, I'd suggest picking one up! The first batch starts shipping tomorrow...
We're trying to make it as easy as Arduino to develop an app for your watch! To dodge the arm-gcc toolchain issue, we've built a cloudcompile system which allows all platforms Mac/Ubuntu/Windows to have an equally easy time compiling. Simply code your app in C, then run our python script to compile your code and load it onto your watch, wirelessly!
Also, your terms of service (http://www.getinpulse.com/terms/) states that you claim rights over information I upload to sites run by Allerta. That must include code I upload to be compiled, correct?
From the terms: "hereby grant to Allerta a perpetual, irrevocable, non-exclusive, worldwide, royalty-free license, with the right to sublicense, to reproduce, distribute, transmit, publicly perform, publicly display, digitally perform, modify, create derivative works of, and otherwise use and commercially exploit any text, photographs or other data and information you submit to the Website (collectively, User Generated Content) in any media now existing or hereafter developed, including without limitation on websites, in audio format, and in any print media format."
As I mentioned, we're releasing the entire SDK with instructions on how to compile using the full arm-gcc, right on your local machine. We just started off with this cloudcompile service to make it super easy for everyone to get Hello, Watch! (http://getinPulse.com/apps/hello_watch) running.
We are developers ourselves and committed to providing the best experience possible. We'll work fast to get the full toolchain out there.
Your python script uploads my code to your server and sends me back a binary in return. What happens when that server goes down?
Getting people up and writing code on devices quickly is awesome; but sending code (unencrypted, even) off is pretty shady, particularly since you're not disclosing it to anyone.
Arduino manages to have a cute little barebones IDE based around an actual compiler; and they provide full hardware docs, links to datasheets, the works. You're doing exactly the opposite of that. Even Apple will let me compile code for my own iOS devices -- after I've paid the $99/year fee or jailbroken, of course...
It is interesting to note that the EZ430-Chronos can last about 6 months on a charge if continuously reading BlueRobin data (which is a one-direction wireless data) while this inPulse aparently can only last 2 to 3 days recieving data over bluetooth.
Less powerful processor and cheaper display might be a good thing, depending on your uses.
What is the battery life like?
With aggressive power management (turning the screen off most of the time and leaving bluetooth disconnected), it can be up to 3 days. If bluetooth is constantly connected with the screen turning on periodically for alerts, you can expect 25 - 30 hours. If you have a processor-intensive application and are using the full capabilities of the screen non-stop, you can expect 2 - 3 hours.
So unfortunately it seems that the battery life is the biggest issue with inPulse
The _state_ of battery technology is the biggest issue _for_ inPulse. There's not much they can do, given the lack of viable battery technology available.
They _might_ be able to save a decent amount of battery life by using e-paper, but you lose color, and I'm not sure in practice you'd end up saving much power over the OLEDs they're currently using.
Seriously. When one has a 60x lead in lifetime over the other, you don't blame 'poor battery technology', especially when they use the same 'poor battery technology'!
Blame other interesting things you might want to wirelessly communicate with for standardizing on power-hungry Bluetooth rather than using a lower-power wireless tech (zigbee, ANT, etc) instead/additionally.
I bought 3 EZ430s. While they're interesting concepts, they're not without their problems (most notably not keeping time very well on the default firmware). You do come up against limitations and they're more of a proof of concept than anything else.
It is based on the MSP430 microprocessor which is much less powerful than this thing, its capped at 16mhz or something IIRC and only has around 8K of program space. It is however about a third of the price.
http://processors.wiki.ti.com/index.php/EZ430-Chronos