-
• #2
Excellent news! I've just ordered!
Did you ever find out whether you can share the datasheet for the VC31, and do you have the one for the VC31B?
-
• #3
"Specification" tab has error: "2x Buttons"
-
• #4
@malaire thanks!
@myownself I haven't found out yet but the datasheets I have are basically useless anyway. They're for the physical device (schematic, solder pads, etc) and in Chinese, and they don't mention registers or how it actually works anywhere
You'll be able to tell from the source in GitHub though I've spent ages on reverse engineering so the vc31 driver is at least a bit better documented now
-
• #5
Ah fair enough. It is only in the past couple of years that I've done anything (minor) with electronics, so I'm still getting it into my head that datasheets are not that useful. I'll have a closer look at the source sometime, I glanced at it a while ago but I know you've made a lot of improvements to the HRM in the past week or so.
This is all so exciting!
-
• #7
Oh - exciting. How many do you have for sale this time. It will be great to see a new inrush of developers and new apps ! Great that the Firmware updater is working in the App Loader. That should reduce support posts a lot.
-
• #8
in chinese.
I dont suppose there are any UK or American manufacturers of smart watches - or was that a dumb question as we just ask the Chinese to build them anyway. Nothing against China - just that I'm too old to learn Chinese.
-
• #10
It's hard to explain - just tried putting them side-by-side for a video but if I put them both on my wrist the performance depends on which order they're in.
Here are two screenshots...
- Red is the filtered value (which is then peak detected)
- Purple is the raw HRM value (with DC bias removed)
- Green is the raw value from the HRM
The one on the right is the new VC31B. The actual signal varies less with heart rate, but it's much 'cleaner' - I believe the sensor itself probably has a higher dynamic range.
I run the new one at 25Hz rather than the old one's 50Hz as well, which helps with power consumption. Even at half the rate it seems to perform slightly better
2 Attachments
- Red is the filtered value (which is then peak detected)
-
• #11
I run the new one at 25Hz rather than the old one's 50Hz as well,
Hmm. That should be fine for counting BPM, but other metrics will probably require a higher sampling rate. What is the range of tge sensor? In theory, could apps request a higher sampling rate for specific purposes?
-
• #12
This is something we should really start a new thread for, but yes, you should be able to request a higher sampling rate (you can on the original Bangle.js 1 and 2, but it doesn't work yet on the new sensor as some more reverse engineering is needed).
-
• #13
Yes, sorry, when we get there we can discuss in the PPG thread.
On topic for this thread, I just realised I didn't get an email notification when the new bangles came in (I had put my email in to be notified). Maybe you haven't sent them yet as you have enough to deal with from people seeing the forum, but if you have, it might be worth double checking they've gone.
-
• #14
Thanks - yes, I haven't sent the emails out yet, and I probably won't for another 3 weeks.
I've just swapped the shop over to using IOSS (with us paying the VAT for Europe) and I wanted to make sure everything was working fine before everyone ordered. Also Christine (who's doing the programming and shipping of the Bangles normally) is on holiday for a bit from next week, so while I plan to ship out Bangles from here while she's away, there are 3000 emails to send out, and if even a fraction of those order a Bangle I'll be spending the next 2 weeks doing nothing apart from packing boxes (again!).
We just received our next batch of Bangles, and we're now ready to start selling them!
You can get them from the Espruino Shop, and several distributors will be getting stock soon too!
These are almost the same as the KickStarter devices, but with a few small tweaks: