I was scared off attempting to use it when, one day I was doing something in a terminal and it locked up. I was clicking all over the screen to try and bring up the task manager (or any response), when the phone called my other phone! Input must have been going through to other apps but the screen was showing a blank terminal.
It's the most disappointing phone ever. A 10-year old Android performs better and is more stable.
I wasn't expecting it to work well, but it was far worse than I imagined. I tried several ROMs and the software is starting to improve slightly, though.
A pure Linux phone could obviously benefit from being as polished and streamlined as its iOS and Android counterparts.
But thats not its main mission. Thats not what it's selling.
Total user control, like on real desktop Linux, for those who can wield it, because thats literally what youre getting.
For now this space is for a very specific audience. Maybe in time (like Ubuntu did for desktop Linux on PCs) there will come a distro which is usable enough OOB to have the same effect on mobile.
We mobile Linux users will still be a minority though, and we probably wont mind. Just like we dont mind being minority desktop users either.
I was hoping that there would be a reasonably rock solid "feature phone" platform we could start building from, but it seems everybody wants to scale down the full linux desktop.
I just want a feature phone with a browser.
I would even be keen to see KiaOS running on it, but I don't think you can right now.
If you want to, you can have that; just take postmarketOSs new stable release when it's out with your prefered mobile shell and put the bare minimum of apps you want/need on there and you'll be set.
As for KaiOS, I doubt we'll see that soon, as IIRC like FirefoxOS it's designed to run on Android kernels. With GloDroid being mostly dormant and incomplete, I doubt we can pull that off.
> A 10-year old Android performs better and is more stable.
A decade-old phone would run Android 2.3.7 (Gingerbread), have a dead and unchargable battery, no longer support modern SIM cards which is fine because its odds of connecting to any modern networks approach zero as the old spectrum gets recycled, be unable to run any recent apps (ancient OS version, no memory) including not being able to run any browser capable of opening modern websites, and would have a slow clunky UI on a cramped little screen. The PinePhone has a long way to go, but it would destroy any decade old phone in very nearly every comparison.
I dug out a Nexus One last week, and even after upgrading it to Lineage on Android 2.3.7 it's basically useless.
The stock browser crypto algorithms are too old to connect to most sites, practically zero apps of use support anything that old, you aren't going to be running Firefox or Chrome unless you grab an ancient build. No WhatsApp, Signal, etc. The majority of stuff on F-droid doesn't run, the camera is terrible, it has basically no app storage and you have to use that old mod which allows an ext partition on your SD for apps, which is _slow_. I'm really not sure could find a single practical use for it.
On the other hand, I'm still using my Galaxy Note 3 (2013). Only some months ago did I eventually get around to flashing it from Lollipop to latest Lineage (android 10/Q). The battery is fine because I replaced it. 4G works fine. 5 GHz wifi works fine. A few apps were unstable or had performance issues which led to me finally upgrading and they went away. I can watch YT just fine, play various games, whatever, it doesn't feel slow.
I'm (hopefully) getting a Librem 5 phone sometime this year but I already know it's going to disappoint even as I've been expecting as much from the start. You're not getting these phones for their practical superiority on any metric, but as a weak signal to the world of an approach you'd rather others took, and maybe this will help form a better bedrock for yet another attempt in another 5+ years especially given how much upstreaming is going on vs. what Mozilla or Ubuntu did. I think the Librem is even more disappointing than the PinePhone (which, given its much more reasonable price, I'm tempted to buy for the same reasons I'm getting a Librem) and that's only comparing with my Note 3 let alone some of the monsters out there now (16 GB of RAM on a phone exists). The Librem/PinePhone have as much RAM as my nearly 8 year old phone, same number CPU cores (but lower frequency and probably lower performance), same internal storage, in the case of Librem actually more money in nominal prices, both lower screen resolution, uncertain whether Librem'll survive carriers moving to VoLTE (my Note 3 will)...
My only point in this mini-rant is that while it may not be a 10 year old phone, at 8 years my Note 3 is still looking better than these 'new' things on feature set. GP is right that these open phones are disappointing, even if in some cases there's nothing to be done because of the higher level project goals.
My 9yo Samsung runs 7.1 (I have an 8.0 image, but it's less fast and stable) with a couple gigs of ram, and has an up-to-date Firefox and Brave on a comfortable screen. Sure it's painfully slow, but so is my HP mini netbook, and I like using that (with an SSD).
It's tri-band, but I only use it on wifi. I might be able to compile newer firmware, but the Android SDK has never worked for me even in the most casual use, so I never tried.
Android Gingerbread was my first real phone. I think I had an LG Nitro. It was awesome. But yeah it wouldn't be great in the modern world at all. Still better than my OnePlus 8. What a complete piece of trash! At least old android gave me clear access to see what was running on my device. New android doesn't do that.
10 year old Android means roughly a single-core 1GHz CPU, 512MB RAM and a 480p screen. Would it be more stable? Maybe. But certainly not better performing, as that's more or less Raspberry Pi 1 hardware.
Pricing is tough to guesstimate right now with the component shortage. Given that hardware has not changed in a while, I don't see what should hold you back from getting it now, and the remaining hardware bugs are acceptible.[0]
The experience has improved quite a bit since I got my first PinePhone in June. That said, this device is never going to be really fast. It's acceptible now, though, and it turns out, that you can adjust to things being a little slow every then and now. Most distributions don't use the acceleration capabilities the SoC offer as much as they could, and there are other areas where improvements for battery life and smoothness are still happening. It's been serving me fine as my only smartphone on weekends (and as a secondary device on workdays).
I've been tracking the Linux Phone landscape on my blog[1], in multiple videos[2]. I have also adopted an App List, to list all the apps that run on these nascent Linux Distributions and are optimized for smartphones.[3]
BTW: The coolest feature is that you can just put another distribution on a microSDHC/XC card and boot it. Like this you can carry multiple phone operating systems that fit certain circumstances.
Absolutely fantastic resources! Thank you. I just want to get out of the major eco systems of apple and android, and gain the feeling of owning my computer. Having a Raspberry has been an enlightening experience in that regard.
I have the PostmarketOS version. It's not something you can hammer all day, but I'm really happy with it as a secondary phone (I carry one for work and one for me. This is the one for me.) It already does stuff my Android and iOS phones never did (like turning into a half decent Linux desktop with a dock), but if you NEED something that only comes as an app for those two OSes, then don't worry about it for a while.
In December they said they were thinking about opening up a reseller program so they don't have to support end users [1]. Then they announced the end of the community editions [2]. And now this announcement. I think they're trying to evolve into something more sustainable.
I think the problem is that the demographics are already niche for the existing Linux on mobile software ecosystem on a device with as many software issues as it has. Then the issue of what little extra money they could make doesn't make it easier to find components that match their philosophy, much less performant parts, making a high price an even harder sell. I like their current strategy and it seems to be working for them. Slowly build an audience while helping the ecosystem to develop while maintaining a sustainable business that isn't prone to demand falling off a cliff because your phone is suddenly a year old and people want the new hotness.
Pine has decided that selling their products as close to cost as possible is how they can best achieve their goals of spurring open development and a having a vibrant, international community. It seems to be working for them.
In a sense yes, but PinePhone is also focused on economically feasible for developers. The hardware may be solid, but the firmware and software on it still need a lot of work. It's a frustrating experience. Others mention freezes when making or receiving calls. I have the KDE Plasma version and I sometimes get kernel panics when taking it on and off the charger. Enthusiasts (like me) will pay $200 to support the project and have a secondary handheld device running Linux, but at higher prices, not so much.
Also, I'm not trying to knock the pinephone - I have one and fully support them. Their goal is to get more linux phones into the hands of more people, both devs and end-users. They're using open-source distributions and their approach will (in theory) lead to better software. The experience should get better over time.
There is the Librem 5 by Purism. They do target the free as in freedom crowd, and it starts off at $800. They also have the same phone made entirely in the USA for $3k. The PinePhone and Librem hardware have fairly similar specs. The Librem 5 does have a better camera and a slightly larger battery, but the real difference is going to be that they put a lot more polish into the system.
As far as I can tell, this looks like the same hardware theyve sold before for the earlier community edition phones.
Im guessing that calling it "BETA" signifies an increased confidence in it actually being usable, due to recent software improvements in the Linux mobile community?
While thats great in itself, Im looking forward to them moving to a beefier SOC.
(Posted from a PinePhone PostmarketOS community convergence edition phone)
Yes, it was in the march community update [0], they feel the software is now beta quality: 'The upcoming production-run, as well as the next two or three following it, will be referred to as the Beta Edition. The term “Beta” indicates the present software status'
Make a screen+mobo+speakers+... (the top 1/3 of a phone), give it an SD slot, and make it a unit that clicks onto a generic battery. Case makers will fill in the rest.
Actually I think a better idea could be a common chassis. The common chassis could provide power, (touch)screen, buttons, audio, IMU, NFC, light/distance sensors, barometer, thermometer, cameras and leds; it could connect to the motherboard through pogopins using USB-C standard. This allows video, power and USB hub to expose the rest of the hardware to the mobo.
The mobo could have slots in the back similar to librem-5; theses slots would hold sd, wifi and modem cards.
I foresee 3 groups for whom this maybe interesting: pine64, purism and fairphone. I really think there is a market for a common platform to make linux phones competitive against alternatives on price too.
I'm not an EE, I really don't know... Maybe a standard USB-C could connect the motherboard to the chassis then. Still, it is possible with currently available tech.
Hardware still uses a single band WLAN. Unfortunately, I won't be able to use it because of this. The 2.4GHz band in my apartment complex is severely degraded due to the sheer amount of channel conflicts. My Netgear router can only handle 2-5Mbps.
I put an elevator flyer that simply had a cartoon router saying: "Is your wifi slow? We're all yelling over each other. Switch to channel 1,6, or 11! Email <me> if you want help."
And then in 12pt I explained it with a QR code for a web page.
Of the 30 or so SSIDs on random channels, we got half off them in about three months. Performance was significantly improved. Many were happy. Some couldn't tell. One was angry at me for apparently causing something completely unrelated.
I can see 120 SSIDs from my balcony. I gave up on trying to do this. The only solution is to run my choice of channels in the 5100 to 5250 range, which mostly blasts over the noise floor caused by my neighbors because 5.x GHz doesn't go through concrete floors/ceilings and walls very well.
And then I moved the 2.4 GHz into a 20 MHz wide channel, the cleanest I could find (it's all equally bad), narrower channel cuts through the noise slightly better than 40. Of course at the expense of throughput. But the only things that are 2.4 only here are like, a raspberry pi zero w.
At some point it probably makes sense to line each apt in rf blocking sheets and then have only your own wifi inside. Would cause problems with guests on 4g though.
It's not illegal to have copper wallpaper or whatever, if you want to go full faraday cage. Passive signal blocking is legal. There might be a few situations (movie theatre?) where you would want to install a big obvious sign stating that wireless devices will not communicate outside the room. It's really no different than being in the third level sub-basement of a concrete parking garage.
Jammers are not legal, unless you're operating them in a shielded RF test chamber.
It would also drain the battery of any phone you forget to switch to wifi-only. Most phones, when dealing with tenuous or absent xG signal, will try harder and harder to transmit.
However I have all my APs on very low power so that roaming works well within my apartment. In fact some of the neighbors' APs are stronger in my living room than my own ceiling mounted one!
I'm glad I don't need 2.4 much anymore for performance devices but most IoT gizmos don't do 5.
Right, but the page itself is very sparse on details, and there's no way for me to find out that this phone is "hacker friendly". That's really what I was trying to get at. It would have made more sense to link to somewhere on the site itself, rather than a preorder page. This just came off as something using HN as a vector for marketing.
As someone with a pinephone, I think there needs to be solid development put into two things:
1. Phone calls need to consistently work
2. Signal needs to handle texts
I plan on contributing when I have a tad more time, but I’d actually use the phone full time if it could make and receive calls without error. Can’t do that yet.
Signal only exists officially as an Android app and, for desktop, an Electron app that is not adapted for a small mobile screen at all. The PinePhone community has no say in these two pieces of software, and Signal's devs are strongly opposed to anyone creating third-party Signal clients. So, this is out of the PinePhone community's control. There is also the option of running Signal over the Android emulator Anbox, but that is very RAM- and processor-heavy, and the PinePhone has very weak, underpowered hardware.
For a handful of unusually dedicated nerds, it does look like maybe one can set up signal-cli on the phone (which, again, is not an official Signal client and disapproved of by Signal's devs), but this is not a full-featured, polished solution.
"Signal's devs are strongly opposed to anyone creating third-party Signal clients."
Hmm, that might have been the case. There was that famous blog post from 2016 about no third-party clients. That has never been retracted, but they got so much flac over it that they seemingly changed their mind. What I hear from the developers of Whisperfish for Sailfish OS is that the people from Signal are quite helpfull in bringing that app towards a good functioning client.
Don't forget Axolotl[0], an unofficial client for Ubuntu Touch that has been unofficially packaged up for Debian (and that package can in turn be installed with a PKGBUILD script on Arch and Manjaro). A working official client would be a lot better though.
The problem with Axolotl and signald mentioned in another comment here, is that they don't have a very strong reputation for security like the official Signal apps.
Unfortunately, in practice everything is quite slow on the pinephone.
(But I don’t want to sound so negative, it is a wonderful project and the advances in the last year is fantastic. Unfortunately the amount of RAM, and sort of old hardware doesn’t help)
Bought a Manjaro version in December. The thing just fried itself when I pressed the restart button in the GUI. With no sign of life, several tries, and only 5 minutes af total use.
Did you email Pine64? They are a touch slow to communicate, but they are generally pretty good about replacements and sending parts even if they list them as out of stock on their website.
After digging, I found it was because the charging logic has a bug that fails hard when the battery gets to a certain voltage.
The fix, and I'm not kidding here, is to remove the battery and plug/unplug it a few dozen times; then put the battery back in and plug it in. Repeat until it works again.
I did this for five minutes and it came back to life; I've read of it taking a half hour for some folks.
Living somewhere where MMS is not used, I have not been following this to closely. There have been various scripts that apparently work well enough. At least for the Phosh/Gnome on Mobile side this issue [0] is the one to watch.
Wow this is great news! Does anyone know if they are considering variants with more RAM? Given how hungry Gnome/KDE and the web browsers are, 3 GB does not seem enough... I'd definitely shell out for more RAM.
I've been using the Pinephone with Manjaro Phosh since beta5 as my daily phone and tablet(Angelfish ftw). Works just fine for me. Calls work a lot better recently, due to improving modem software. Can't imagine ever going back to a walled garden ecosystem anymore!
71 comments
[ 3.3 ms ] story [ 158 ms ] thread[1] https://www.youtube.com/watch?v=fCKMxzz9cjs
edit: added details
I was hoping to find some time this weekend to try some alternate UI's
I wasn't expecting it to work well, but it was far worse than I imagined. I tried several ROMs and the software is starting to improve slightly, though.
But thats not its main mission. Thats not what it's selling.
Total user control, like on real desktop Linux, for those who can wield it, because thats literally what youre getting.
For now this space is for a very specific audience. Maybe in time (like Ubuntu did for desktop Linux on PCs) there will come a distro which is usable enough OOB to have the same effect on mobile.
We mobile Linux users will still be a minority though, and we probably wont mind. Just like we dont mind being minority desktop users either.
I just want a feature phone with a browser.
I would even be keen to see KiaOS running on it, but I don't think you can right now.
As for KaiOS, I doubt we'll see that soon, as IIRC like FirefoxOS it's designed to run on Android kernels. With GloDroid being mostly dormant and incomplete, I doubt we can pull that off.
Genode is working on a port. I'd gladly run seL4 instead of Linux.
A decade-old phone would run Android 2.3.7 (Gingerbread), have a dead and unchargable battery, no longer support modern SIM cards which is fine because its odds of connecting to any modern networks approach zero as the old spectrum gets recycled, be unable to run any recent apps (ancient OS version, no memory) including not being able to run any browser capable of opening modern websites, and would have a slow clunky UI on a cramped little screen. The PinePhone has a long way to go, but it would destroy any decade old phone in very nearly every comparison.
The stock browser crypto algorithms are too old to connect to most sites, practically zero apps of use support anything that old, you aren't going to be running Firefox or Chrome unless you grab an ancient build. No WhatsApp, Signal, etc. The majority of stuff on F-droid doesn't run, the camera is terrible, it has basically no app storage and you have to use that old mod which allows an ext partition on your SD for apps, which is _slow_. I'm really not sure could find a single practical use for it.
I'm (hopefully) getting a Librem 5 phone sometime this year but I already know it's going to disappoint even as I've been expecting as much from the start. You're not getting these phones for their practical superiority on any metric, but as a weak signal to the world of an approach you'd rather others took, and maybe this will help form a better bedrock for yet another attempt in another 5+ years especially given how much upstreaming is going on vs. what Mozilla or Ubuntu did. I think the Librem is even more disappointing than the PinePhone (which, given its much more reasonable price, I'm tempted to buy for the same reasons I'm getting a Librem) and that's only comparing with my Note 3 let alone some of the monsters out there now (16 GB of RAM on a phone exists). The Librem/PinePhone have as much RAM as my nearly 8 year old phone, same number CPU cores (but lower frequency and probably lower performance), same internal storage, in the case of Librem actually more money in nominal prices, both lower screen resolution, uncertain whether Librem'll survive carriers moving to VoLTE (my Note 3 will)...
My only point in this mini-rant is that while it may not be a 10 year old phone, at 8 years my Note 3 is still looking better than these 'new' things on feature set. GP is right that these open phones are disappointing, even if in some cases there's nothing to be done because of the higher level project goals.
It's tri-band, but I only use it on wifi. I might be able to compile newer firmware, but the Android SDK has never worked for me even in the most casual use, so I never tried.
The experience has improved quite a bit since I got my first PinePhone in June. That said, this device is never going to be really fast. It's acceptible now, though, and it turns out, that you can adjust to things being a little slow every then and now. Most distributions don't use the acceleration capabilities the SoC offer as much as they could, and there are other areas where improvements for battery life and smoothness are still happening. It's been serving me fine as my only smartphone on weekends (and as a secondary device on workdays).
I've been tracking the Linux Phone landscape on my blog[1], in multiple videos[2]. I have also adopted an App List, to list all the apps that run on these nascent Linux Distributions and are optimized for smartphones.[3]
BTW: The coolest feature is that you can just put another distribution on a microSDHC/XC card and boot it. Like this you can carry multiple phone operating systems that fit certain circumstances.
[0] https://wiki.pine64.org/wiki/PinePhone_v1.2b
[1] https://linmob.net
[2] https://linmob.net/videos/
[3] http://appl.ist
That's me at least.
I'm tempted to buy one just as a sign of support. We really need PinePhone to grow.
1: https://www.pine64.org/2020/12/02/pine-store-community-prici...
2: https://www.pine64.org/2021/02/02/the-end-of-community-editi...
Also, I'm not trying to knock the pinephone - I have one and fully support them. Their goal is to get more linux phones into the hands of more people, both devs and end-users. They're using open-source distributions and their approach will (in theory) lead to better software. The experience should get better over time.
There is the Librem 5 by Purism. They do target the free as in freedom crowd, and it starts off at $800. They also have the same phone made entirely in the USA for $3k. The PinePhone and Librem hardware have fairly similar specs. The Librem 5 does have a better camera and a slightly larger battery, but the real difference is going to be that they put a lot more polish into the system.
https://puri.sm/products/librem-5/
Im guessing that calling it "BETA" signifies an increased confidence in it actually being usable, due to recent software improvements in the Linux mobile community?
While thats great in itself, Im looking forward to them moving to a beefier SOC.
(Posted from a PinePhone PostmarketOS community convergence edition phone)
[0] https://www.pine64.org/2021/03/15/march-update/
The mobo could have slots in the back similar to librem-5; theses slots would hold sd, wifi and modem cards.
I foresee 3 groups for whom this maybe interesting: pine64, purism and fairphone. I really think there is a market for a common platform to make linux phones competitive against alternatives on price too.
And then in 12pt I explained it with a QR code for a web page.
Of the 30 or so SSIDs on random channels, we got half off them in about three months. Performance was significantly improved. Many were happy. Some couldn't tell. One was angry at me for apparently causing something completely unrelated.
If you want to meet the neighbours, do this.
And then I moved the 2.4 GHz into a 20 MHz wide channel, the cleanest I could find (it's all equally bad), narrower channel cuts through the noise slightly better than 40. Of course at the expense of throughput. But the only things that are 2.4 only here are like, a raspberry pi zero w.
Jammers are not legal, unless you're operating them in a shielded RF test chamber.
Or just a municipal wifi that everyone in town could share.
But hey, capitalism, gotta have everyone paying $39.99/month to clog up the airwaves with redundant SSIDs.
However I have all my APs on very low power so that roaming works well within my apartment. In fact some of the neighbors' APs are stronger in my living room than my own ceiling mounted one!
I'm glad I don't need 2.4 much anymore for performance devices but most IoT gizmos don't do 5.
This phone, however, is very hacker friendly and does things that other phones don't, so we (HN) are interested.
1. Phone calls need to consistently work
2. Signal needs to handle texts
I plan on contributing when I have a tad more time, but I’d actually use the phone full time if it could make and receive calls without error. Can’t do that yet.
For a handful of unusually dedicated nerds, it does look like maybe one can set up signal-cli on the phone (which, again, is not an official Signal client and disapproved of by Signal's devs), but this is not a full-featured, polished solution.
Hmm, that might have been the case. There was that famous blog post from 2016 about no third-party clients. That has never been retracted, but they got so much flac over it that they seemingly changed their mind. What I hear from the developers of Whisperfish for Sailfish OS is that the people from Signal are quite helpfull in bringing that app towards a good functioning client.
Also, there are laws being worked on in Europe that popular platforms will have to allow third-party clients, not much different from the web and its browsers. The Digital Markets Act: https://ec.europa.eu/info/strategy/priorities-2019-2024/euro...
[0] https://github.com/nanu-c/axolotl
(But I don’t want to sound so negative, it is a wonderful project and the advances in the last year is fantastic. Unfortunately the amount of RAM, and sort of old hardware doesn’t help)
On my Android 10 phone they don't anymore, and nobody cares. I guess we're over the whole "using smartphones for phone calls" thing now.
After digging, I found it was because the charging logic has a bug that fails hard when the battery gets to a certain voltage.
The fix, and I'm not kidding here, is to remove the battery and plug/unplug it a few dozen times; then put the battery back in and plug it in. Repeat until it works again.
I did this for five minutes and it came back to life; I've read of it taking a half hour for some folks.
[0] https://source.puri.sm/Librem5/chatty/-/issues/30#note_14822...