Pinetime Plus?

[Avamander]

For me personally, it has to be a Linux watch. I bought a LG watch Urbane for 50 USD on ebay, which is on the way to my place. There is also the Kingwear K88 Pro for 80 USD, which has a camera. There is an instruction availailable from Chandler Swift to install Linux Asteroid OS on the Kingwear. But the Kingwear Pro is not recommended for Bluetooth low energy and WIFI applications, since the Bluetooth stack of this watch appears to be a mess. I would spend 300 USD for a Ticwatch Pro when there is a Linux version available for it. Currently there is only Linux Asteroid OS available for the Ticwatch E & S.

You have said that very many times here already.

---

Is not Linux development easier than PineTime 1.0 development?

Not in such a constrained environment, follow the PineCube development to see how easy it is.

And is it not easier to find follower for a Linux watch than for PineTime 1.0?

Unknown. No Linux (not Android Wear) watch has been really popular for hackers as far as I am aware.

Is it not possible to build a “PineTime plus” with a more powerful CPU and more RAM which can run Linux software?

Possibly, but I am very sceptical that it has more pros than cons.

[Peter Gamma]

Peter Gamma Wrote: And is it not easier to find follower for a Linux watch than for PineTime 1.0?

Avamander:

“Unknown. No Linux (not Android Wear) watch has been really popular for hackers as far as I am aware. “

The Android wear watch TicWatch Pro has a custom ROM with a developer thread of more than 400 pages:

https://forum.xda-developers.com/smartwa...13/page409

is this not popular? There are many other Android Wear development projects around, for instance for Amazfit watches. But the problem is, that there are to many projects and too many watches.

Would not be one single hackable multi purpose watch with a single community, similar to the PinePhone with the Pine64 community, be more successful than a lot of individual projects?

[Avamander]

is this not popular? There are many other Android Wear development projects around, for instance for Amazfit watches. But the problem is, that there are to many projects and too many watches.

Not really no. My previous EU-market only phone had a similar level of support and it wasn't really popular.

[Peter Gamma]

But I still miss a Linux watch from Pine64. For those who miss such a watch too, I’m sharing my little experience, what is available and where it can be found:

http://petergamma.org/

I’m also Open, Friendly, and Community driven. I wrote some posts about Linux watches. It is a private project, and I do it for fun and in my spare time. And I miss a community for Linux watches. I was attracted by the PineTime. It was advertised as Linux PineTime companion. But the PineTime is definitely not a Linux watch. I made good experiences in the Asteroid OS IRC chat. They are professional and responsive. It is a pitty that there is no forum. They said that this will evolve at XDA developers, Reddit, etc.

And I don’t give up hope that there will be a Linux watch from Pine64 some day. The developers from Asteroid OS, which are much more professional than me, recommended to wait and see for the new Xiaomi MI watch or the Fossil watch, to put Asteroid OS on it. Information about Linux watches are scattered everywhere, and I miss a community for this. Join me.

[danielt]

Is not Linux development easier than PineTime 1.0 development? And is it not easier to find follower for a Linux watch than for PineTime 1.0?

I don't agree with either of these things.

There are 10+ watch projects for PineTime which is hardly evidence that the device is difficult to develop for.

Similarly it is only easy to find users for a watch when the price and specs are right. It's not clear that is is possible for a Linux smart watch to be all of cheap enough, capable enough, physically pleasing and with enough battery stamina to attract users. That's not to say that Linux smart watches will never reach that point... but its not clear now is the right time.

That means I shall focus my energy on devices that actually exist today rather than dreaming of a world that may never become our tomorrow. Unless the landscape changes a bit then, if asked, I would advocate for any PineTime successor to follow a similar strategy but with a slightly more powerful BLE device such as nRF52840 or even nRF5340 (if the later part can be acquired cheaply enough).

[ikonyshev]

For me good improvment for pinetime it is 5 pins (SWDIO,SWDCLK,3.3V,GND,5V) on back side of cover. Like of charging connection. For development without open cover of watch. It is cost less when 1 $. But quate usefull for developers.

[Arwen]

What I have noticed is that cheaper devices tend to sell well, and thus are developed faster, both hardware and software. It takes time to build up to the "pro" versions.

My personal opinion, is that while a Linux watch would be nice, and perhaps a camera, (for a Dick Tracy video phone watch!), that is not going to happen any time soon due to cost. (My opinion.) I'd guess such a watch would cost $250 to $1,000 to get something that ran non-embedded Linux. And had a reasonable feature set, (like a battery that lasted all day).

For the power savings, such a device might have to use current chips in 14nm to 7nm. Those cost more than older tech. Plus, just running Linux does not mean much. If the intent is to have such "smartwatch", (verses the "feature" watch PineTime), run random programs the user would want to install, then EXTRA memory is required. Both RAM & Flash. Most embedded devices have limited memory to both reduce cost and battery usage. This allows the planners to make a good estimate of battery life. Random programs run on smart phones do eat battery life, but we now have such phones last at least 1 full day. And generally longer.

All that means while I may see Pine64 coming out with a newer version of PineTime, I really doubt it will have the resources, (CPU, RAM, Flash & Battery), to run non-embedded Linux.

[Peter Gamma]

Linux watch dreams are not far from becoming real. My dream is only one step away from coming true. I have to install Linux Asteroid OS on my LG watch Urbane, which is one of the best supported watches for Asteroid OS (50 USD on ebay).

If I understood it right, all Linux software which can run on ARM based processors should also run on Linux Asteroid OS. Is adaption of size and user interface of Linux apps more difficult than adapting Linux apps to the PinePhone?

In the AsteroidOs IRC chat, people recommend to wait and see for the new Xiamo Mi watch or new Fossil watch to install Linux Asteroid OS on it (starting from around 100 USD each). Someone already managed to install Postmarked OS on LG watch R. Does the Linux watch future not look bright? Why should a Linux PineTime future not also look bright? Someone installed Linux Asteroid OS on LG watch G and had a battery life of two days:

https://www.youtube.com/watch?v=PRGt3Apkwxg&t=628s

I currently use Garmin watches with a Polar OH1 heart rate sensor and aquire heart rate data at the highest data acquisition rate. I use two Vivoactive 3 watches and two Polar OH1 sensors for this, one device charging and one is acquiring data. The Vivoactive 3 runs for more than 24 hrs with the built in heart rate sensor, the Polar OH1 8 hours. So I have to change the watch once a day, and the Polar OH1 three times a day. The price for this setup is affordable, it is not too uncomfortable, and it can compete with research grade devices which costs 10x more as far as accuracy is concerned.

[Peter Gamma]

.
 I also did some experiments with foot pods. My favorite is the Stryd foot pod, although it it is not cheap. It is around 300 USD. According to this review, it is the most accurate foot pod available, and it does not need to be calibrated:

https://fellrnr.com/wiki/Stryd

It did some experiments to use it with the android application a training tracker. With this application, the foot pod becomes a highly accurate step counter, which counts every single step very accurately. Wrist based step counters are less accurate. The Stryd also works at speeds slower than 4 km/h. Since I used it for walking meditation, this is very attractive. With a standard foot pod, and I guess also with a wrist based step counter, data are lost at slower speeds. But also affordable foot pods as for instance the Zwift Bluetooth low energy foot pod for 40 USD are very accurate when calibrated. These external Bluetooth low energy sensors can be added quiet easily to a Linux watch:

https://forum.pine64.org/showthread.php?...low+energy

but as far as I know not to a PineTime 1.0. A Linux PineTime would allow to add these external sensors easily, for building an open source sports watch which can compete with the sports watch market leader Garmin.

[Arwen]

@Peter Gamma You mention external sensor devices. And the intent to add the information from them to the PineTime.

Have you considered that another way to do this is an App on a smart phone, (like the PinePhone), and then make the information available to the PineTime?

That would save processing power, battery power and the PineTime watch would simply update the information every now and then. (How often would depend on what information...)
Anyone thinking of using Linux on the existing PineTime needs to review the hardware;

https://wiki.pine64.org/index.php?title=...t_overview

The current PineTime hardware has 64KBs of RAM. (Yes, that is KiloBytes, not MegaBytes.) Plus, has a single core 64 MHz ARM Cortex-M4F.