I've bought a Sleeptracker Pro from Amazon in Oct. '08. I've tested it for ca. 3 weeks in both US and EU, including the few jet-lagged days in between. Bgnd info on the tester: I buy and test sundry bio-feedback devices that may help me to improve, or document, my sleeping patterns.
My 1-st take on Sleeptracker Pro: A good & simple idea with a flawed --alpha-- implementation. Good enough for a prototype, useful for geeks and novices, yet a let-down for folks expecting hassle-free results, or for advanced users.
The idea is simple: Use a motion sensor to inferr the wearer's current sleep stage. While it can't tell the diff between deep sleep and resting (not moving), or, a longer meditation session, the concept is a practical approximation. The execution, though, is unnecessarily limited. Rarely i've actually stumbled upon a potentially useful device so unfinished as this...
Items that can be improved.
1. Please consider adding a "Free Sleep One-touch" record function, w/o having to set both the "to bed" and the "alarm" time. This type of acquisition window reduces the consumer benefit (to simplify the development effort?). I'd like to press once to start the sampling, and once more to terminate it. Man people would like to analyze their sleep w/o actually using the alarm.
2. Extend the user memory (e.g., by 100-1000x), enough to store a few weeks of sampled data. Currently --unless one doesn't download daily to a PC-- each night will over-write the previous' session data. While FLASH memory costs cents/MB, the S.Pro stores merely a few 'user' bytes, forcing one to either daily download or loose data. When a 4GB USB stick costs a few $, one would expect an $180 device to squeeze more than 0.1KB of user data.
3. Relocate the "Mode" button to make it accessible while the USB clip (the 3-prong gotic device with USB cable) is attached. BTW, it's quite an irony to have the *only* button needed for USB transfers being blocked by the USB clip... Or just use a mini-USB port (water may become a problem, but it should actually matter for this device).
4. Software: Reminiscent of the 80s + GUI. While the GUI is relatively clean, it's mostly an empty shell devoid of the needed functionality. Please enhance the s/w beyond its still alpha version where it's stuck since '07. At least pls. allow the data to be deleted, added/inserted, shifted and displayed in more useful ways.
5. Pls. allow the user to set the "eagerness" of the alarm, which currently starts always on the FIRST motion during the window. Explain clearly with examples or a picture how the alarm window works. E.g., the 20min interval is NOT centered on the alarm time, but it's ahead of it. With larger windows, this means that the user must accept to sleep less; a few times the alarm woke me up nearly an hour earlier. Perhaps I could've slept 1-2 more deep episodes before the final alarm...
6. Allow once per month downloads (repeat) w/o over-writing the data.
7. Provide more suitable statistics. The average can be misleading when placing 5 billionaires (Gates, Buffet etc.) i the same room w/ 95 homeless and concluding that this must be super-rich's assembly. Median and variability could better describe a sleep made of 3-4 long segments (90min. each) plus a dozen 8-12 min. tracks.
Net: A good idea w/ a coarse 1-st implementation (actually this is the 2nd or 3rd generation), slow in technically improvements. Marketing seems to trump technology thus far, because with a tad more thought the hardware and the software could be greatly improved - to reach new markets.
Does it actually work? Yes, with limitations. The commercials have fuelled some false expectations, of more than the actual device can do. This is a basic alarm watch with a built-in motion sensor (like those used in most 2.5" hard-disks to avoid crashing the heads) that will go on at the 1-st move detected during the alarm window. The trade-offs are obvious...
Is it worth your money? Yes, if you have some moderate sleep problems and enough discipline in your daily schedule to work on them. No, if you expect an true sleep monitor (headband sensors) or a reliable, thoughtful and straightforward tool.
IMHO it's a thoughtless raw design that could do much better. I rarely had to resort to adjectives such as 'thoughtless' until now, yet there's little engineering attention paid to this product. A shame. I gave it 3 stars to reward the idea, which by now is common in many other products.
Update Febr.'09: The Glow button no longer works w/o pushing really hard - but it's still functional. A few s/w bugs have caused the watch to display wrong Data A, and to repeat the same readings for 2 nights.