This is a good example of a poor web site design. If you, like me, do not know what Phosh is and go to their website, it will tell you not much beyond "A user interface for your mobile phone," which could mean pretty much anything. Is it a UI level on top of Android? Is it an idependent mobile OS? How it is better that competition? What are key features and design goals?
I think it’s more a failing of a link aggregator site like this one dropping you into a release announcement page without context. If you don’t already know what Phosh is, why would you read their 0.56.0 announcement? Normally you wouldn’t. But here we are. Putting “The Phosh project aims to provide a daily-usable, robust and easy to use graphical user environment for mobile devices running mainline Linux” on every release page feels like a waste of time.
As it is I didn’t find it at all difficult to find the answers to your questions by going to the “about page”.
> The Phosh project aims to provide a daily-usable, robust and easy to use graphical user environment for mobile devices running mainline Linux. The name is a portmanteau of phone and shell as phosh was one of the first components developed by the project. It hence coined the whole project’s name and is still one of its core components. All of Phosh is entirely Free Software.
Linux has long been the most practical laptop OS for me, but I can't see it ever being competitive with mobile OSes, and that's coming from someone who wants it to succeed (I've installed postmarketOS on a OP6T). I just don't see how it will overcome the various issues (app support, tap-to-pay, camera quality, etc).
I've never understood tap-to-pay being a dealbreaker issue. It takes me just as long to pull out my credit card as it does to pull out my phone, and you can use them on the exact same terminals.
App support and camera quality I can understand more. I'm on a Linux phone using Phosh (FLX1s), and there's Android app compatibility, but it is a little rough (and of course things that rely on Play Integrity won't work). I've managed to avoid tying myself to anything that requires Google for now, but I acknowledge that I'm lucky there.
With tap-to-pay, you can store multiple cards in your digital wallet, and you don't have to remember any of their PINs. You can use your fingerprint to sign transactions. I believe this makes it faster.
There's also thresholds where a simple tap of a card won't work and you need to insert the chip, not the case with the phone. Phone is much much easier, don't even have to open the wallet app, just unlock the phone with your fingerprint and tap.
Tap to pay isn't itself a dealbreaker, for me it's more a representation of the status-quo and what is to come, specifically how more and more of the modern world will rely on integrations with proprietary software and hardware on phones. Tap to pay with credit cards isn't too big of a deal, but the wallet as a whole (i.e. boarding passes for airplanes, gift cards, ChargePoint tap-in) is a major feature.
I often leave the house without my wallet these days. It’s great. Especially riding the subway, where Apple’s transit functionality means you don’t have to do anything other than tap your device without even unlocking, is a very nice convenience when you’re carrying groceries or something similar.
> It takes me just as long to pull out my credit card
It can be a lifestyle difference.
I personally don't being my wallet in most daily trips, have no use of it. I used to stick a credit card in my phone case but also got rid of it as more stores reliably offered wireless and QR code payments.
Mind you this comes with a specific environment I don't expect everyone to live in or long for, I'm just explaining.
See, I have to have my driver's license, but if I could have that on my phone as well, I might do this. Running out of battery is largely not a concern for me as I already carry an external battery with multiple days of charge.
In Denmark at least we have this, both our drivers license and social security cards are apps on our phones. I only need my old physical drivers license when I leave the country.
I just put my debit card in the pocket in my phone case. Job done. If yours doesn't have a pocket, you can try putting it between the phone in the case. In that scenario, I suggest turning off NFC on the phone so it doesn't keep trying to read the card.
That's why I never accepted the app from my bank. It still work most of the time but it's cumbersome for some uses (like adding a new recipient for wire transfers, it takes two days).
Unfortunately not. A lot of UK banks require using your mobile phone as the 2nd factor to log on to online banking on your laptop. They used to issue fobs/card readers, but have moved to using the app for this now.
As long as every phone distro is just a desktop distro shoehorned on a small screen, that's not gonna happen.
IMO it's not a matter of tap to pay and camera quality, rather a matter of whole system paradigm. Having millions of disconnected services in the "do one thing and do it right" spirit and using text based communication and hundreds of python and shell scripts is relatively maintainable and relatively easy to use, but very inefficient when it comes to CPU cycles - and on a handheld every cycle counts.
And of course every app is optimized for desktops/laptops... but I guess that's a chicken-or-egg problem: once there is a working distro, there will be apps too. And once there will be apps to use, there will be a working distro. Maybe.
From personal experience, many apps work suprisingly well on Phosh. The thing keeping me from using a Droidian phone with Phosh was a seemingly device-specific issue with answering regular phone calls. Kind of ironic that the linux phone could do many apps from FlatHub flawlessly with touch and everything, but not regular phone stuff. Nevertheless, the smartphone part is handled quite well with the Settings app, the quick toggles, the support for flatpaks, and so on, especially with Wayland where the android ecosystem is within reach.
If I read correctly, you identified the lack of smarthpone-level apps and distro as the limiting factor. In my opinion, the lack of sufficiently powerful but still open hardware is what we miss. Mainline linux with proper hardware support is pretty good but not complete on the Snapdragon 845, a 9 y/o platform. Anything newer? Nearly impossible without some android-specific layers (such as libhybris in the case of Droidian). Currently mid-high level hardware with PinePhone-like openness would probably let Phosh (and Plasma-Mobile, SXMO etc.) distros strive. The smart features, as in the apps, are mostly there.
Oddly enough I am using Win10 right now on my laptop. On my
main computer I use linux but I also got tired having to set
up things specifically for the laptop or be locked down in
a specific distribution; plus, I also have to run various
software on the laptop and when the rest of the class or
group uses Windows, and you are the sole Linux person, it
feels very lonely. So I fake being a win user in that case.
I've been using GNU/Linux on my phones for the last 18 years, initially as a portable hacking platform back when I was a teenager and now as the most practical mobile OS for me. I often see people mentioning these areas as something they "can't see becoming competitive", but this doesn't really match my experience. It feels very much like what running GNU/Linux on a PC without having Windows around felt like 10-15 years ago or so and it very noticeably keeps moving forward over time.
My experience today is that I have all the applications I need daily, and the odd ones I don't have natively I can run in Waydroid. There are currently some limitations when it comes to access to hardware from within the container, but nothing that couldn't be overcome with some moderate effort. The only real blocker is remote attestation, but I see it as a threat to the whole fabric of society, not just Linux phones, and it should be opposed regardless of platform one uses.
I don't really get tap-to-pay. I never thought it's something I want or need to have on my phone (which doesn't even have NFC, though it could be added as an extension). I just use a card. I get it that it may be slightly more convenient, but definitely not worth changing the OS over, let alone giving others control over my phone. That said, I have full access to my bank account from my phone anyway. Practically everything except TOTP payments (Blik) can be done from the Web (and with Epiphany's webapps it's just there as an app to launch), and Blik can be used from the Android app in Waydroid. Frankly, if I couldn't access the bank from my phone I would rather change the bank than the phone, and would make sure to let the customer service know very clearly why I'm doing that.
And when it comes to camera quality, matching the state-of-the-art overprocessed mess would be hard, but having played with camera processing a lot recently I can say that perfectly adequate quality is absolutely well within the reach (if not there already for daylight photos). I only researched absolute basics of photo processing and implemented some essential stuff while completely ignoring others and I'm already quite happy with the photos I get (https://social.librem.one/@dos/tagged/shotonlibrem5) and there's enormous room for improvement still that only needs someone who knows what they're doing to sit down and implement things that are missing. Some of that stuff is already happening around libcamera as more people get involved there thanks to laptops with webcams that also need similar software handling. The particular phone I'm using doesn't have hardware video encoder (only a decoder) so video recording will stay low res there forever unless limited to short clips, but it's not a limitation inherent to the platform, just to this specific device.
I have briefly carried an Android device as a secondary phone some 10 years ago out of necessity (N900 was starting to get too old to handle the Web and Librem 5 did not exist yet) and it felt quite miserable, it seemed like an appliance rather than a personal computing device despite of equipping it with microG, F-Droid, rooting it etc. When you're used to being able to script a simple thing right on your phone using whatever technology you already feel comfortable with, it's hard to give it up. Just try to patch a simple thing from the system's core on an Android phone - yes, it's possible if you're really determined, spend a lot of time on it and don't mind attestation issues, but I can do that with dpkg-buildpackage on my phone's screen within minutes, send a patch upstream and have it actually merged in time it would take my laptop to produce an Android image. I may not exercise this ability a lot in my daily use, but having such possibility is incredibly empowering. I can't see Android becoming competitive there, quite the contrary ;) The only actual obstacle I see on the path to growth is the threat of remote attestation gaining widespread use and this affects any existing or future platform from outside the duopoly, not just GNU/Linux in particular, and I believe this still isn't a lost game.
I've installed this on my Surface Go 2 64GB. Runs smooth! Absolutely the best tablet experience for Linux. The support is also wild: My silly questions are answered within hours.
I have a oneplus6 and use a mobile version of Debian called "Mobian". postmarketOS is a really good choice, and they have a wiki of supported devices: https://wiki.postmarketos.org/wiki/Devices
Seemed interesting until I read that Phosh pulls in GNOME - gnome-settings, gnome-session etc. Seems like a very strange bundle to bring in for an extremely power constrained device, where every % of increased battery drain is noticed by the user
Why? What's particularly heavy in these gnome tools?
Like the particular programs are no issue, but the whole UNIX-userspace as done in the mainframe era and still is. Like you definitely need cooperative program suspend/resume like on Android for any kind of sane battery life, but that's unfortunately completely missing in case of GNU/Linux.
Nothing. On my phone there's 770MB RAM consumed by the entire system with Phosh running right now. The rest is free to be eaten by the browser.
If anything, it's GTK that can be rather inefficient when it comes to rendering (both GTK3 and GTK4, but for different reasons). There are some tricks that phoc (Phosh's compositor) could still learn too, but it will get there.
Not in a million worlds. Android is by far the most optimized OS (as a whole, including user space, graphics stack everything) for mobile devices. It's almost like the most widely used mobile operating system has had quite a bit of dev hours spent on it.
Trust me, you wouldn't be saying this if you had actually tried modern Android on low powered devices. low-mid range phones drop frames more often in 2026 than they did in 2015 with Android 5-7's much simpler UI.
edit: oops, the topic was RAM usage. in 2026. of course everyone, it's definitely the 200MB usage surplus that makes your phone feel heavy. let's add 1TB of RAM to a C2D and it will be ready to be paired with a 5090.
> of course everyone, it's definitely the 200MB usage surplus that makes your phone feel heavy.
It actually is though? My Pixel 9a has a perfectly serviceable CPU but I am often frustrated by its 8GB of RAM. Switching to the Revolut app to generate a disposable card number consistently evicts my browser tab from RAM. ~100% of the time this happens and I get extremely frustrated by it losing my state in the checkout flow.
Well, the market has consistently raised the low end of what constitutes "decent" hardware for a smartphone, and it really is not expensive today to have good enough hardware.
Why should software not adapt to reality?
As another data point, take a look at what "embedded" development means today. It is increasingly common to just simply have a device running full Linux, as it is getting cheap enough to do so.
Add a bunch of fat, semi-desktop binaries that actually provide some kind of functionality to make it remotely comparable, and then you just have a worse, fatter system that runs hot and wastes the battery.
Android is a unified system working together to make the device race to sleep as fast as possible, with as few wakeups as feasible (e.g. batch together events that would wake up the device).
Lighter as SXMO will run on a low budget 1GB of RAM armv7 handheld released in 2014 (maybe even 512MB of RAM), even without 3D acceleration and Android 17 just won't, no matter how much you tweak it
And what? I mean, I can also just run Linux on a raspberry pi and connect a screen to it. That in itself will have zero relevant features, it's a Linux with a screen that you can toy around with, not a phone.
I don't see how it's relevant if you want something that works on said device?
In terms of resources I think it's roughly going like this, from lighter to heavier:
SXMO < Lomiri (Ubuntu Touch) < Plasma mobile < Android < Phosh < Gnome mobile
I could also add XFCE but they officially don't support mobile.
Android isn't the lightest mobile environment by any conceivable means, maybe it's the most featured but certainly not the most optimized for low-end devices, the days of Android 11 are over.
You need 4GB of ram, a good CPU and full hardware acceleration to run modern Android.
And if the next Android bumps the requirements again, I would put it at the end of the list.
I've been using openbox for over a decade. That's completely false. Xfce is not a memory hog. And it's not minimal either it's fully functional.
Gnome is a bloated mess of a thing and I hate it. Why would anyone want their desktop to use over 1gb of ram. I have a 32gb laptop and I still loath the idea of throwing away memory on such a bloated awful thing.
It doesn't run GNOME Shell, which is the main memory hog of GNOME.
It uses some GNOME services, namely so it doesn't have to invent it's own. None of these services are memory heavy and all have a purpose (e.g. managing Bluetooth)
Do you have any numbers? The last numbers I remember seeing, had XFCE around 500 MB and gnome around 700 MB. I'm trying to find some current numbers, but it's a pretty tough thing.
Without any different numbers, I think saying a massive memory hog is a little hyperbolic. Applications in use, especially browsers, are going to dwarf the desktop environment anyway. Having the polish is well worth it to many people, myself included.
I would definitely like to see less memory requirements for the various desktop environments, but at the end of the day I don't pay for any of this
I'm a certified gnome hater, but a phone is basically the perfect application for it. As far as resource usage goes, I have been dailying an FLX1 for a year and a half now and Firefox is the only noticeable resource hog
Phosh is not based on gnome-shell and has its own settings and apps, but it does use parts of gnome, no reason to reinvent the wheel.
> Seems like a very strange bundle to bring in for an extremely power constrained device, where every % of increased battery drain is noticed by the user
I'm sure you can make your Frankenstein version that would be 10% as usable and secure as phosh by removing everything but for most users, 100mb more ram and 1% more battery drain for an OS aiming to be a daily driver is something that's worth it.
This is a good example of a poor web site design. If you, like me, do not know what Phosh is and go to their website, it will tell you not much beyond "A user interface for your mobile phone," which could mean pretty much anything. Is it a UI level on top of Android? Is it an idependent mobile OS? How it is better that competition? What are key features and design goals?
I think it’s more a failing of a link aggregator site like this one dropping you into a release announcement page without context. If you don’t already know what Phosh is, why would you read their 0.56.0 announcement? Normally you wouldn’t. But here we are. Putting “The Phosh project aims to provide a daily-usable, robust and easy to use graphical user environment for mobile devices running mainline Linux” on every release page feels like a waste of time.
As it is I didn’t find it at all difficult to find the answers to your questions by going to the “about page”.
From that same website:
> About Phosh
> The Phosh project aims to provide a daily-usable, robust and easy to use graphical user environment for mobile devices running mainline Linux. The name is a portmanteau of phone and shell as phosh was one of the first components developed by the project. It hence coined the whole project’s name and is still one of its core components. All of Phosh is entirely Free Software.
Linux has long been the most practical laptop OS for me, but I can't see it ever being competitive with mobile OSes, and that's coming from someone who wants it to succeed (I've installed postmarketOS on a OP6T). I just don't see how it will overcome the various issues (app support, tap-to-pay, camera quality, etc).
I've never understood tap-to-pay being a dealbreaker issue. It takes me just as long to pull out my credit card as it does to pull out my phone, and you can use them on the exact same terminals.
App support and camera quality I can understand more. I'm on a Linux phone using Phosh (FLX1s), and there's Android app compatibility, but it is a little rough (and of course things that rely on Play Integrity won't work). I've managed to avoid tying myself to anything that requires Google for now, but I acknowledge that I'm lucky there.
With tap-to-pay, you can store multiple cards in your digital wallet, and you don't have to remember any of their PINs. You can use your fingerprint to sign transactions. I believe this makes it faster.
There's also thresholds where a simple tap of a card won't work and you need to insert the chip, not the case with the phone. Phone is much much easier, don't even have to open the wallet app, just unlock the phone with your fingerprint and tap.
Tap to pay isn't itself a dealbreaker, for me it's more a representation of the status-quo and what is to come, specifically how more and more of the modern world will rely on integrations with proprietary software and hardware on phones. Tap to pay with credit cards isn't too big of a deal, but the wallet as a whole (i.e. boarding passes for airplanes, gift cards, ChargePoint tap-in) is a major feature.
Both Passes [0] and KDE Itinerary [1] work well on phones and can handle plenty of this stuff.
[0] https://flathub.org/en/apps/me.sanchezrodriguez.passes
[1] https://flathub.org/en/apps/org.kde.itinerary
I often leave the house without my wallet these days. It’s great. Especially riding the subway, where Apple’s transit functionality means you don’t have to do anything other than tap your device without even unlocking, is a very nice convenience when you’re carrying groceries or something similar.
> It takes me just as long to pull out my credit card
It can be a lifestyle difference.
I personally don't being my wallet in most daily trips, have no use of it. I used to stick a credit card in my phone case but also got rid of it as more stores reliably offered wireless and QR code payments.
Mind you this comes with a specific environment I don't expect everyone to live in or long for, I'm just explaining.
See, I have to have my driver's license, but if I could have that on my phone as well, I might do this. Running out of battery is largely not a concern for me as I already carry an external battery with multiple days of charge.
In Denmark at least we have this, both our drivers license and social security cards are apps on our phones. I only need my old physical drivers license when I leave the country.
I just put my debit card in the pocket in my phone case. Job done. If yours doesn't have a pocket, you can try putting it between the phone in the case. In that scenario, I suggest turning off NFC on the phone so it doesn't keep trying to read the card.
The main issue is lack of banking app support for me. Without that (which the banks will never allow) you would always need two phones.
That's why I never accepted the app from my bank. It still work most of the time but it's cumbersome for some uses (like adding a new recipient for wire transfers, it takes two days).
At the risk of sounding really old can't you use your computer/laptop for that?
Unfortunately not. A lot of UK banks require using your mobile phone as the 2nd factor to log on to online banking on your laptop. They used to issue fobs/card readers, but have moved to using the app for this now.
Ah interesting. We're still on the email/sms 2FA
Some new fintech banks only truly support smartphones.
As long as every phone distro is just a desktop distro shoehorned on a small screen, that's not gonna happen.
IMO it's not a matter of tap to pay and camera quality, rather a matter of whole system paradigm. Having millions of disconnected services in the "do one thing and do it right" spirit and using text based communication and hundreds of python and shell scripts is relatively maintainable and relatively easy to use, but very inefficient when it comes to CPU cycles - and on a handheld every cycle counts.
And of course every app is optimized for desktops/laptops... but I guess that's a chicken-or-egg problem: once there is a working distro, there will be apps too. And once there will be apps to use, there will be a working distro. Maybe.
From personal experience, many apps work suprisingly well on Phosh. The thing keeping me from using a Droidian phone with Phosh was a seemingly device-specific issue with answering regular phone calls. Kind of ironic that the linux phone could do many apps from FlatHub flawlessly with touch and everything, but not regular phone stuff. Nevertheless, the smartphone part is handled quite well with the Settings app, the quick toggles, the support for flatpaks, and so on, especially with Wayland where the android ecosystem is within reach.
If I read correctly, you identified the lack of smarthpone-level apps and distro as the limiting factor. In my opinion, the lack of sufficiently powerful but still open hardware is what we miss. Mainline linux with proper hardware support is pretty good but not complete on the Snapdragon 845, a 9 y/o platform. Anything newer? Nearly impossible without some android-specific layers (such as libhybris in the case of Droidian). Currently mid-high level hardware with PinePhone-like openness would probably let Phosh (and Plasma-Mobile, SXMO etc.) distros strive. The smart features, as in the apps, are mostly there.
Oddly enough I am using Win10 right now on my laptop. On my main computer I use linux but I also got tired having to set up things specifically for the laptop or be locked down in a specific distribution; plus, I also have to run various software on the laptop and when the rest of the class or group uses Windows, and you are the sole Linux person, it feels very lonely. So I fake being a win user in that case.
I've been using GNU/Linux on my phones for the last 18 years, initially as a portable hacking platform back when I was a teenager and now as the most practical mobile OS for me. I often see people mentioning these areas as something they "can't see becoming competitive", but this doesn't really match my experience. It feels very much like what running GNU/Linux on a PC without having Windows around felt like 10-15 years ago or so and it very noticeably keeps moving forward over time.
My experience today is that I have all the applications I need daily, and the odd ones I don't have natively I can run in Waydroid. There are currently some limitations when it comes to access to hardware from within the container, but nothing that couldn't be overcome with some moderate effort. The only real blocker is remote attestation, but I see it as a threat to the whole fabric of society, not just Linux phones, and it should be opposed regardless of platform one uses.
I don't really get tap-to-pay. I never thought it's something I want or need to have on my phone (which doesn't even have NFC, though it could be added as an extension). I just use a card. I get it that it may be slightly more convenient, but definitely not worth changing the OS over, let alone giving others control over my phone. That said, I have full access to my bank account from my phone anyway. Practically everything except TOTP payments (Blik) can be done from the Web (and with Epiphany's webapps it's just there as an app to launch), and Blik can be used from the Android app in Waydroid. Frankly, if I couldn't access the bank from my phone I would rather change the bank than the phone, and would make sure to let the customer service know very clearly why I'm doing that.
And when it comes to camera quality, matching the state-of-the-art overprocessed mess would be hard, but having played with camera processing a lot recently I can say that perfectly adequate quality is absolutely well within the reach (if not there already for daylight photos). I only researched absolute basics of photo processing and implemented some essential stuff while completely ignoring others and I'm already quite happy with the photos I get (https://social.librem.one/@dos/tagged/shotonlibrem5) and there's enormous room for improvement still that only needs someone who knows what they're doing to sit down and implement things that are missing. Some of that stuff is already happening around libcamera as more people get involved there thanks to laptops with webcams that also need similar software handling. The particular phone I'm using doesn't have hardware video encoder (only a decoder) so video recording will stay low res there forever unless limited to short clips, but it's not a limitation inherent to the platform, just to this specific device.
I have briefly carried an Android device as a secondary phone some 10 years ago out of necessity (N900 was starting to get too old to handle the Web and Librem 5 did not exist yet) and it felt quite miserable, it seemed like an appliance rather than a personal computing device despite of equipping it with microG, F-Droid, rooting it etc. When you're used to being able to script a simple thing right on your phone using whatever technology you already feel comfortable with, it's hard to give it up. Just try to patch a simple thing from the system's core on an Android phone - yes, it's possible if you're really determined, spend a lot of time on it and don't mind attestation issues, but I can do that with dpkg-buildpackage on my phone's screen within minutes, send a patch upstream and have it actually merged in time it would take my laptop to produce an Android image. I may not exercise this ability a lot in my daily use, but having such possibility is incredibly empowering. I can't see Android becoming competitive there, quite the contrary ;) The only actual obstacle I see on the path to growth is the threat of remote attestation gaining widespread use and this affects any existing or future platform from outside the duopoly, not just GNU/Linux in particular, and I believe this still isn't a lost game.
The fact I thought it was a custom UI over stock android means they got this well rounded.
I've installed this on my Surface Go 2 64GB. Runs smooth! Absolutely the best tablet experience for Linux. The support is also wild: My silly questions are answered within hours.
What are the best phones/distros to use phosh with?
I have a oneplus6 and use a mobile version of Debian called "Mobian". postmarketOS is a really good choice, and they have a wiki of supported devices: https://wiki.postmarketos.org/wiki/Devices
https://phosh.mobi/faq/#so-what-phones-are-supported
Seemed interesting until I read that Phosh pulls in GNOME - gnome-settings, gnome-session etc. Seems like a very strange bundle to bring in for an extremely power constrained device, where every % of increased battery drain is noticed by the user
Why? What's particularly heavy in these gnome tools?
Like the particular programs are no issue, but the whole UNIX-userspace as done in the mainframe era and still is. Like you definitely need cooperative program suspend/resume like on Android for any kind of sane battery life, but that's unfortunately completely missing in case of GNU/Linux.
> What's particularly heavy in these gnome tools?
Nothing. On my phone there's 770MB RAM consumed by the entire system with Phosh running right now. The rest is free to be eaten by the browser.
If anything, it's GTK that can be rather inefficient when it comes to rendering (both GTK3 and GTK4, but for different reasons). There are some tricks that phoc (Phosh's compositor) could still learn too, but it will get there.
Gnomes' a massive memory hog
I was looking at this and thinking maybe it would improve a cheap android phone. But now I know it's running gnome I won't even consider trying
> maybe it would improve a cheap android phone.
Not in a million worlds. Android is by far the most optimized OS (as a whole, including user space, graphics stack everything) for mobile devices. It's almost like the most widely used mobile operating system has had quite a bit of dev hours spent on it.
Trust me, you wouldn't be saying this if you had actually tried modern Android on low powered devices. low-mid range phones drop frames more often in 2026 than they did in 2015 with Android 5-7's much simpler UI.
edit: oops, the topic was RAM usage. in 2026. of course everyone, it's definitely the 200MB usage surplus that makes your phone feel heavy. let's add 1TB of RAM to a C2D and it will be ready to be paired with a 5090.
> of course everyone, it's definitely the 200MB usage surplus that makes your phone feel heavy.
It actually is though? My Pixel 9a has a perfectly serviceable CPU but I am often frustrated by its 8GB of RAM. Switching to the Revolut app to generate a disposable card number consistently evicts my browser tab from RAM. ~100% of the time this happens and I get extremely frustrated by it losing my state in the checkout flow.
Well, the market has consistently raised the low end of what constitutes "decent" hardware for a smartphone, and it really is not expensive today to have good enough hardware.
Why should software not adapt to reality?
As another data point, take a look at what "embedded" development means today. It is increasingly common to just simply have a device running full Linux, as it is getting cheap enough to do so.
Every desktop environment is a massive memory hog. Do you really want something minimal like xfce on a touchscreen?
SXMO is pretty minimal in that regard, but it doesn't force you to use the touchscreen. Can also navigate through menus via the volume buttons...
Even that will be a massive energy vampire compared to android, that had top notch engineers working on it for close to 2 decades...
Yes and no, SXMO will work without hardware acceleration, I've not tried but I doubt that Android will.
Additionally, minimum requirements of Android 17 are way above what SXMO allows.
With Phosh you would have a point but SXMO is lighter than a modern Android.
Lighter for what? Displaying a clock?
Add a bunch of fat, semi-desktop binaries that actually provide some kind of functionality to make it remotely comparable, and then you just have a worse, fatter system that runs hot and wastes the battery.
Android is a unified system working together to make the device race to sleep as fast as possible, with as few wakeups as feasible (e.g. batch together events that would wake up the device).
Lighter as SXMO will run on a low budget 1GB of RAM armv7 handheld released in 2014 (maybe even 512MB of RAM), even without 3D acceleration and Android 17 just won't, no matter how much you tweak it
And what? I mean, I can also just run Linux on a raspberry pi and connect a screen to it. That in itself will have zero relevant features, it's a Linux with a screen that you can toy around with, not a phone.
I don't see how it's relevant if you want something that works on said device?
In terms of resources I think it's roughly going like this, from lighter to heavier:
SXMO < Lomiri (Ubuntu Touch) < Plasma mobile < Android < Phosh < Gnome mobile
I could also add XFCE but they officially don't support mobile.
Android isn't the lightest mobile environment by any conceivable means, maybe it's the most featured but certainly not the most optimized for low-end devices, the days of Android 11 are over.
You need 4GB of ram, a good CPU and full hardware acceleration to run modern Android.
And if the next Android bumps the requirements again, I would put it at the end of the list.
I've been using openbox for over a decade. That's completely false. Xfce is not a memory hog. And it's not minimal either it's fully functional.
Gnome is a bloated mess of a thing and I hate it. Why would anyone want their desktop to use over 1gb of ram. I have a 32gb laptop and I still loath the idea of throwing away memory on such a bloated awful thing.
Running gnome on a phone. Yeah... No
It doesn't run GNOME Shell, which is the main memory hog of GNOME.
It uses some GNOME services, namely so it doesn't have to invent it's own. None of these services are memory heavy and all have a purpose (e.g. managing Bluetooth)
weird, I'm running GNOME on my laptop right now and I checked and it's using 275mb of ram. I would not call that particularly bloated.
Phosh is not based on gnome-shell and nobody cares if you want to run openbox on your phone.
Do you have any numbers? The last numbers I remember seeing, had XFCE around 500 MB and gnome around 700 MB. I'm trying to find some current numbers, but it's a pretty tough thing.
Without any different numbers, I think saying a massive memory hog is a little hyperbolic. Applications in use, especially browsers, are going to dwarf the desktop environment anyway. Having the polish is well worth it to many people, myself included.
I would definitely like to see less memory requirements for the various desktop environments, but at the end of the day I don't pay for any of this
I'm a certified gnome hater, but a phone is basically the perfect application for it. As far as resource usage goes, I have been dailying an FLX1 for a year and a half now and Firefox is the only noticeable resource hog
Phosh is not based on gnome-shell and has its own settings and apps, but it does use parts of gnome, no reason to reinvent the wheel.
> Seems like a very strange bundle to bring in for an extremely power constrained device, where every % of increased battery drain is noticed by the user
I'm sure you can make your Frankenstein version that would be 10% as usable and secure as phosh by removing everything but for most users, 100mb more ram and 1% more battery drain for an OS aiming to be a daily driver is something that's worth it.
It works kind of okay for recent devices as phones are very powerful nowadays, Phosh on a <2015 device is much more painful though.
Terrible name. It's going to fail on those grounds alone.
Not that it would really succeed otherwise. You need Android app compatibility to stand a remote chance.
> You need Android app compatibility
That's not phosh's problem; waydroid is a pretty much independent component.
The name sounds like someone driving by at high speed ...
a