I like a good smart watch and I appreciate open source, but an ESP32 isn't a great pick when low power consumption is important and the device is going to be communicating regularly. I'm surprised LILYGO went that direction in a watch form factor.
Yeah one of the Nordic nRF chips would be nice. But I guess the advantage of ESP is the hobby developer community. It's not going to be as good on battery power but the barrier to entry for people wanting to tinker is really low.
an esp32 on an 1100mah battery will last years on deep sleep, and about a day with wifi on and in high power modes.
a pixel watch 4 says they last 30 hours , ambiguously. they use a battery less than half the size. in reality with constant use they'll drop dead in 6 hours.
the thing is clunky and heavy , anyway -- so if it lasts as long as an off the shelf watch who cares?
also, the primary reason : lilygo shoves ESPs into everything.
Several Garmin watches last for weeks (24 days full charge, actual 1-2 weeks with heavy gps and fitness tracking), and I struggle to understand why consumers accept anything less. It seems like consumers don't realize what's available.
I think cost is one factor. I have a Vivoactive 4 and I love it but it has a reported battery life of 8 days and I get maybe half that with regular run tracking. I'm guessing the 24 days/1-2 weeks is for a considerably more expensive model.
I'm not sure if my Pixel Watch 3 is much more efficient than your 4 or if by "constant use" you literally mean scrolling through it actively for hours at a time, but I only charge mine maybe twice a week. It's on at all times, connected to my phone via Bluetooth and to my Wi-Fi network when I'm home, and I actively manage any push notifications I get from it, but otherwise it seems to idle fairly efficiently.
I’ll need to check my notes on power consumption. I’ve spent quite a bit of time trying out different modes and configurations… it’s not great. I would not expect years at all. You gotta be very careful about what has to stay on and off.
Furthermore, bugs. To this time there’s random crashes that happen with sleep which limits their use
After careful optimization, the v1 got about 6 months out of a 1100 mAh battery. Later improvements and bumping to a 3300 mAh battery got me to 14 months, before my kid yanked it off the wall, total'd the panel and I rebuilt it. The test continues.
That said--op isn't wrong. If power usage is the metric you optimize for, there's much better BOM than an esp32.
Galaxy watch 7 (I think) weather here. Also about 30 hours. It's a charge every day thing, but allowing for some forgetfulness. It's not ideal, but it's manageable and certainly functional.
Nordic Semi, or maybe ST Micro. I've got an STM32WB on my bench at the moment with sensitive coulomb counting and it looks very promising but without all those radios. Of course with all those radios (ie, if you need LoRa on a watch... which is a design decision I'm also skeptical of) then Nordic has a good track record.
Different use case (environmental data recorder) but our current product uses an STM that turns on an ESP32 (not trusting that sleep mode) when it needs radio, to run on 2 AAs forever.
I like your hpdl watch better than the LilyGo watch. As a lover of analog and casio watches I can tell you that your watch can soon build you a loyal following in its current form if you only build 3 such watches and put them on Amazon, etc. and price them at 50$. Next batch of 3 you could try at 100$.
Don't worry about glass cover, our local watch repair guy can easily slap one in it
The nice thing about a project like this is that it shows how much taste matters in engineering. 'Works' and 'you'd actually use it' are very different milestones.
This is a product aimed at a very specific hobbyist segment of the smart tinkerer, who will probably get more joy out of it than most buyers get out of their Apple Watch. Is that a smaller market? Yes. That doesn't make it a bad product.
I think it's a relevant critique for something that goes on your wrist, especially when products like PineTime exist in the same market segment with aesthetics much closer to an apple watch (an aesthetic I'm not fond of either, but hey, it seems to be popular)
This watch shares a lot of visual characteristics with something like the original Casio GShock. Just because it doesn't look like a modern smart-watch, doesn't mean it's ugly. I quite like the rugged aesthetic.
No mention of battery life? I guess it depends on the software that you run. But it would be nice to have a benchmark for how long it would last in normal watch mode.
the same for me, however I can imagine that creating a good sleep analysing software based on the hypothethical HRM and accelerometer is mission impossible. I find the best software is Garmin at the moment of writing, or at least suits me well, and it blocks me from buying any opensource watch, unless I want to have two watches on my hands
There are only a few features I care about in a smartwatch:
1. O2 monitoring. I have sleep apnea and live at high altitude, so this matters to me.
2. Motion sensor. Also mostly for tracking sleep.
3. Vibrator for notifications.
4. A screen backlight.
5. Battery life longer than a week.
6. Waterproof enough to survive a splash in the shower/rain.
I consider GPS, cellular, AI, touchscreens, cloud-only sync and control apps, and just about everything else to be anti-features. There are no devices that really cover all this that I've found. A few Garmin and Amazfit/Zepp devices come close, but they have enough drawbacks for me to not be happy with them. The new Pebble is nearly perfect, but the lack of an O2 sensor is a dealbreaker for me :(
The Sensor Watch circuit board [1] inside the case of a Casio F91-W / A158W / A159W satisfies 2, 4, 5, and 6. Accelerometer or thermometer available as daughterboard.
Battery life measured in months, if not years. Although the simple LED backlight and the segmented LCD leaves a bit to be desired, and there is no wireless connectivity for notifications. Open source firmware.
The Ollee watch circuit board [2] is similar, better backlight but closed-source firmware and configuration over BLE in a smartphone app. Still no notifications over BLE though.
I'd think combining 1 and 6 (O₂ monitoring and waterproofing) would be difficult.
my experience with a sensor watch has been terrible.
imagine breaking a $3 watch that is not quite as indestructible as people think it is, but it is nonetheless pretty robust, and then trying to shove something 100x glitchier and 5x as expensive into its case...
Does anyone know if this has an accelerometer? I recently got a nice sports-oriented smartwatch (non-Garmin), to use it mostly for rowing, but it doesn't track the rowing-rate. It should be pretty easy to program one if the watch has accelerometers, but couldn't tell from the spec sheet (maybe that means no?)
This device looks capable of a lot of features and possibilities. Unfortunately nothing comes to my mind because I'm not good with diy hardware (once connected raspberry pi zero with led strips). Could someone tell examples of interesting and/or useful projects one can implement with this watch?
A Garmin memory-in-pixel eInk (MIP) watch? Lots of choices there, since about 2015. Modern versions (MIP not AMOLED, the category is getting confusing) last about 10 days on a charge.
Now we are talking! This one, in addition to all the other alternatives out there, makes the new Pebble dead on arrival. At least at the current price point.
To be honest there is not much to it, you buy the movement, put it in a case, and put the hands on it. you can get everything from aliexpress. it's easier and often cheaper to just buy a normal watch if you need one.
I had one, broke 2 screens before I gave up wearing it... It's really not meant to be work in daily life, which is sad because an e-paper screen makes a lot of sense on a watch.
The peripherals are great until you realize there's a dirth of software to use with them .. like, GPS is fun and everything, but not if all you've got is the coordinates ..
I really love the idea of LoRA in a watch though, so I hope that once this gets shipped, the software makes some leaps and bounds ..
have wished for decades now there was an open-source Garmin on the level of Cyanogenmod / LineageOS for Android
not sure if it will happen this decade but definitely next decade
proper running/cycling metrics are hard as demonstrated by how many well-funded competitors are somewhat close but not there 100% yet (Coros, Amazfit, etc)
someone once hacked and decompiled older Garmins but newer ones are encrypted/signed/locked-down
I'm very excited about this. GPS was the final piece of the puzzle.
I love(d) my bangle.js. Such a true hacker device. Really fun to use WebUSB and push JavaScript files as apps.
But the GPS on that device was a mess, honestly. I know this is a complicated problem but having to synchronize to satellites and recalibrate all the time was beyond me.
I really wanted it to work because I built my own toy run tracker visualization tool.
I am curious about this new lilygo device because it sounds like it has an alternative location sensor: "A u-blox MIA-M10Q GNSS module provides accurate location tracking..."
I'll need to look that up. Anyone have a summary on what's the difference between that and regular GPS?
The underlying Garmin platform is so old that it predates iPhone/Android. I think you can plug in many Garmins via USB without any special software and simply copy activities and data off the watch.
They had a segment of customers who wouldn't have or be allowed to connect a phone - triathletes, long-distance hikers, military. But it's been slowly changing as users want more modern features and the company wants to increase sales.
not sure if this will help you but there is a neat website that allows you to build free maps for older Garmin models that didn't have them at first like Fenix5
1990s is going way back though, they didn't even have mass-storage mode then, it was their proprietary "garmin mode" for usb which only things like BaseCamp can talk to
I like a good smart watch and I appreciate open source, but an ESP32 isn't a great pick when low power consumption is important and the device is going to be communicating regularly. I'm surprised LILYGO went that direction in a watch form factor.
Yeah one of the Nordic nRF chips would be nice. But I guess the advantage of ESP is the hobby developer community. It's not going to be as good on battery power but the barrier to entry for people wanting to tinker is really low.
an esp32 on an 1100mah battery will last years on deep sleep, and about a day with wifi on and in high power modes.
a pixel watch 4 says they last 30 hours , ambiguously. they use a battery less than half the size. in reality with constant use they'll drop dead in 6 hours.
the thing is clunky and heavy , anyway -- so if it lasts as long as an off the shelf watch who cares?
also, the primary reason : lilygo shoves ESPs into everything.
Several Garmin watches last for weeks (24 days full charge, actual 1-2 weeks with heavy gps and fitness tracking), and I struggle to understand why consumers accept anything less. It seems like consumers don't realize what's available.
I think cost is one factor. I have a Vivoactive 4 and I love it but it has a reported battery life of 8 days and I get maybe half that with regular run tracking. I'm guessing the 24 days/1-2 weeks is for a considerably more expensive model.
This. Why would I accept anything else than that?
I'm not sure if my Pixel Watch 3 is much more efficient than your 4 or if by "constant use" you literally mean scrolling through it actively for hours at a time, but I only charge mine maybe twice a week. It's on at all times, connected to my phone via Bluetooth and to my Wi-Fi network when I'm home, and I actively manage any push notifications I get from it, but otherwise it seems to idle fairly efficiently.
I’ll need to check my notes on power consumption. I’ve spent quite a bit of time trying out different modes and configurations… it’s not great. I would not expect years at all. You gotta be very careful about what has to stay on and off.
Furthermore, bugs. To this time there’s random crashes that happen with sleep which limits their use
I have some relevant experience. I built this a few years ago:
https://imgur.com/a/diy-automatic-e-ink-newspaper-using-rust...
After careful optimization, the v1 got about 6 months out of a 1100 mAh battery. Later improvements and bumping to a 3300 mAh battery got me to 14 months, before my kid yanked it off the wall, total'd the panel and I rebuilt it. The test continues.
That said--op isn't wrong. If power usage is the metric you optimize for, there's much better BOM than an esp32.
Galaxy watch 7 (I think) weather here. Also about 30 hours. It's a charge every day thing, but allowing for some forgetfulness. It's not ideal, but it's manageable and certainly functional.
What would you suggest instead?
Nordic Semi, or maybe ST Micro. I've got an STM32WB on my bench at the moment with sensitive coulomb counting and it looks very promising but without all those radios. Of course with all those radios (ie, if you need LoRa on a watch... which is a design decision I'm also skeptical of) then Nordic has a good track record.
Different use case (environmental data recorder) but our current product uses an STM that turns on an ESP32 (not trusting that sleep mode) when it needs radio, to run on 2 AAs forever.
Do you have any info you can share on this recorder? Sounds really interesting!
http://supermechanical.com/pickup Been in production hell thanks to tariffs (wrote a lot about it) but should ship by summer.
For an application using bluetooth low energy, a Nordic NRF52840 turned out to be the winner.
Not the newer more energy efficent nrf54 series?
Most of their lora devices are ESP32.
An advantage of ESP32-S3 could be that it seems possible to run Linux on it.
Not natively. It's not a supported architecture and lacks an MMU. Those who have run Linux on it have done so through a RISC-V emulator or similar.
I for one like it is ESP-IDF with FreeRTOS, at least we get some variation.
That's more a programmable watch than a DIY one :-)
I build mine from scratch, including the PCB and a 3D printed case.
For sure, that's not at all the same level of customability, programmability, capacity, nor quality. But It is really a DIY one.
For anyone interested: https://github.com/jblezoray/hpdl1414-watch
I like your hpdl watch better than the LilyGo watch. As a lover of analog and casio watches I can tell you that your watch can soon build you a loyal following in its current form if you only build 3 such watches and put them on Amazon, etc. and price them at 50$. Next batch of 3 you could try at 100$. Don't worry about glass cover, our local watch repair guy can easily slap one in it
The HPDL-1414 appears to be discontinued though still available from some suppliers in varying MOQs and leads.
Ok, we'll make the title say programmable instead.
Love it, especially your choice of digit display. Very nice hack! I'd wear that.
It's cool that the firmware is hackable but I think "DIY" is an imprecise way to describe that.
The nice thing about a project like this is that it shows how much taste matters in engineering. 'Works' and 'you'd actually use it' are very different milestones.
This is mean-spirited and daft.
This is a product aimed at a very specific hobbyist segment of the smart tinkerer, who will probably get more joy out of it than most buyers get out of their Apple Watch. Is that a smaller market? Yes. That doesn't make it a bad product.
I think it's a relevant critique for something that goes on your wrist, especially when products like PineTime exist in the same market segment with aesthetics much closer to an apple watch (an aesthetic I'm not fond of either, but hey, it seems to be popular)
It doesn't look dissimilar to the Apple Watch Ultra plus a Casio GShock/Garmin Fenix vibe
This watch shares a lot of visual characteristics with something like the original Casio GShock. Just because it doesn't look like a modern smart-watch, doesn't mean it's ugly. I quite like the rugged aesthetic.
No mention of battery life? I guess it depends on the software that you run. But it would be nice to have a benchmark for how long it would last in normal watch mode.
This is almost $80. The PineTime watch is less than half that price. Obviously the specs are different but that's quite a difference.
Where you can buy pinetine for 40$?
From PineTime.
https://pine64.com/product/pinetime-smartwatch-sealed/
https://pine64.com/product/pinetime-dev-kit/
Both of those products are out of stock.
So is the T-Watch Ultra.
Too bad it seems to not do health monitoring, my main use case of a watch now is sleep tracking with as long battery life as possible
the same for me, however I can imagine that creating a good sleep analysing software based on the hypothethical HRM and accelerometer is mission impossible. I find the best software is Garmin at the moment of writing, or at least suits me well, and it blocks me from buying any opensource watch, unless I want to have two watches on my hands
There are only a few features I care about in a smartwatch:
1. O2 monitoring. I have sleep apnea and live at high altitude, so this matters to me.
2. Motion sensor. Also mostly for tracking sleep.
3. Vibrator for notifications.
4. A screen backlight.
5. Battery life longer than a week.
6. Waterproof enough to survive a splash in the shower/rain.
I consider GPS, cellular, AI, touchscreens, cloud-only sync and control apps, and just about everything else to be anti-features. There are no devices that really cover all this that I've found. A few Garmin and Amazfit/Zepp devices come close, but they have enough drawbacks for me to not be happy with them. The new Pebble is nearly perfect, but the lack of an O2 sensor is a dealbreaker for me :(
The Sensor Watch circuit board [1] inside the case of a Casio F91-W / A158W / A159W satisfies 2, 4, 5, and 6. Accelerometer or thermometer available as daughterboard. Battery life measured in months, if not years. Although the simple LED backlight and the segmented LCD leaves a bit to be desired, and there is no wireless connectivity for notifications. Open source firmware.
The Ollee watch circuit board [2] is similar, better backlight but closed-source firmware and configuration over BLE in a smartphone app. Still no notifications over BLE though.
I'd think combining 1 and 6 (O₂ monitoring and waterproofing) would be difficult.
[1]: https://www.sensorwatch.net/
[2]: https://www.olleewatch.com/
> I'd think combining 1 and 6 (O₂ monitoring and waterproofing) would be difficult.
By O2 monitoring, they mean "blood oxygen through skin via LEDs" - there's no impact on waterproofing from that (as the Apple Watch demonstrates.)
The Pro version has a better screen (still segmented but more), RGB LEDs and an infrared sensor.
my experience with a sensor watch has been terrible.
imagine breaking a $3 watch that is not quite as indestructible as people think it is, but it is nonetheless pretty robust, and then trying to shove something 100x glitchier and 5x as expensive into its case...
Fitbit Inspire3 ?
That's awesome. If only it didn't look so tacticool.
I would love to see some extra sensors for heartbeat, temperature, blood oxygen and whatever else could be captured by the design.
Does anyone know if this has an accelerometer? I recently got a nice sports-oriented smartwatch (non-Garmin), to use it mostly for rowing, but it doesn't track the rowing-rate. It should be pretty easy to program one if the watch has accelerometers, but couldn't tell from the spec sheet (maybe that means no?)
It had a Bosch motion sensor with AI abilities it says in the description.
This device looks capable of a lot of features and possibilities. Unfortunately nothing comes to my mind because I'm not good with diy hardware (once connected raspberry pi zero with led strips). Could someone tell examples of interesting and/or useful projects one can implement with this watch?
LILYGO site shows pre-orders of all 3 versions are sold out unfortunately.
There's an older "Plus" model available on Amazon.[1] Surprisingly cheap, at US$66. This new model is the "Ultra".
It's amazing that the market is big enough to get the price that low.
[1] https://www.amazon.com/LILYGO-T-Watch-S3-Development-SX1280-...
When can we finally buy a smartwatch that can keep its (color) display ON during the entire day?
September 18, 2020. That's when Apple released their Watch Series 6, which has an always on color display, and enough battery to last all day.
You can also configure PimeTime to have an always-on display, and it should even last more than a day.
A Garmin memory-in-pixel eInk (MIP) watch? Lots of choices there, since about 2015. Modern versions (MIP not AMOLED, the category is getting confusing) last about 10 days on a charge.
Watches with color memory-in-pixel displays have existed since Pebble.
To add, the current Pebbles are estimated to last 30 days. 10-14 days for the round ones.
Anyone know what the battery life is likely to be like?
Bad.
Now we are talking! This one, in addition to all the other alternatives out there, makes the new Pebble dead on arrival. At least at the current price point.
s/Watch/Smartwatch
Regular DYI watches aren't big news...
(I would be over the moon for a DIY smartwatch with zero AI and e-ink screen.)
I would consider a DIY mechanical/analog watch to be far bigger news/more impressive than a smartwatch.
To be honest there is not much to it, you buy the movement, put it in a case, and put the hands on it. you can get everything from aliexpress. it's easier and often cheaper to just buy a normal watch if you need one.
It's impressive you start with a lathe and make the movement yourself!
Not nearly as impressive as designing and fabricating your own integrated circuits and display!
buying movement is like buying whole PCB
DIY analogy would probably be about acquiring individual gears
Is it different with a smartwatch? You buy the kit, it's not like you solder much as far as I understand.
I thought so too, but after quick research apparently there are kits. For various values of "DIY", I guess...
Sure but buying a movement kit is no different than buying a pcb. Writing code is not impressive any more.
Watchy fits the bill:
https://watchy.sqfmi.com/
I have one, its a bit bulkier than I'd thought it would be, but its a fine piece of timekeeping ..
I had one, broke 2 screens before I gave up wearing it... It's really not meant to be work in daily life, which is sad because an e-paper screen makes a lot of sense on a watch.
I doubt if esp32s3's power consumption can be used in real life.
This does look very cool. Every peripheral one could think of, even LoRA!
The peripherals are great until you realize there's a dirth of software to use with them .. like, GPS is fun and everything, but not if all you've got is the coordinates ..
I really love the idea of LoRA in a watch though, so I hope that once this gets shipped, the software makes some leaps and bounds ..
That stood out to me, too. Garmin should take a hint.
I would buy it if it had wifi. A (decent to wear) watch with wifi would be awesome. Tons of ideas for apps I would build for myself
Ofc, im excluding apple
It's an ESP32, it has wifi.
Yes. Though not all ESP32 variants have wifi.
https://en.wikipedia.org/wiki/ESP32#Features
Oddly, the article didn't mention it - but the watch does have wifi.
The article mentions it as "Wi-Fi" if you were scanning with Ctrl+F.
Preorders sold out already!
trun on and trun off
No heart rate sensor
have wished for decades now there was an open-source Garmin on the level of Cyanogenmod / LineageOS for Android
not sure if it will happen this decade but definitely next decade
proper running/cycling metrics are hard as demonstrated by how many well-funded competitors are somewhat close but not there 100% yet (Coros, Amazfit, etc)
someone once hacked and decompiled older Garmins but newer ones are encrypted/signed/locked-down
Have you looked at the specs for the upcoming PineTime Pro [1]?
[1] https://pine64.org/2026/03/28/pinetime_march_2026/
I'm very excited about this. GPS was the final piece of the puzzle.
I love(d) my bangle.js. Such a true hacker device. Really fun to use WebUSB and push JavaScript files as apps.
But the GPS on that device was a mess, honestly. I know this is a complicated problem but having to synchronize to satellites and recalibrate all the time was beyond me.
I really wanted it to work because I built my own toy run tracker visualization tool.
I am curious about this new lilygo device because it sounds like it has an alternative location sensor: "A u-blox MIA-M10Q GNSS module provides accurate location tracking..."
I'll need to look that up. Anyone have a summary on what's the difference between that and regular GPS?
Oh nice, didn't realize they were doing a second one. Loved the original but I took mine rock climbing and cracked it :(
> newer ones are encrypted/signed/locked-down
I have a garmin watch and didn't know this.
That said, I just used it out of the box, and never (on purpose) hooked it to wifi, bluetooth, garmin connect, etc. Can't do that with an apple watch.
The underlying Garmin platform is so old that it predates iPhone/Android. I think you can plug in many Garmins via USB without any special software and simply copy activities and data off the watch.
They had a segment of customers who wouldn't have or be allowed to connect a phone - triathletes, long-distance hikers, military. But it's been slowly changing as users want more modern features and the company wants to increase sales.
I have a garmin from the late 90s and am saddened by the lack of FOSS software to even sync a new map onto it
not sure if this will help you but there is a neat website that allows you to build free maps for older Garmin models that didn't have them at first like Fenix5
https://garmin.bbbike.org/
1990s is going way back though, they didn't even have mass-storage mode then, it was their proprietary "garmin mode" for usb which only things like BaseCamp can talk to
oh bud, mine doesn't even mention USB in the manual. I got the thing for like a dollar at goodwill haha