Jump to content

claude0001

Members
  • Content Count

    554
  • Joined

  • Last visited

  • Days Won

    66

Posts posted by claude0001

  1. 21 minutes ago, wrc08 said:

    Is there any way I can get up to LineageOS 18.1

    If you haven't updated your LineageOS 16 in two years, chances are you using one of @tdm's unofficial builds. Not sure if there are supported upgrade paths from there. I'm still on (a modded) LOS 16 myself, others might know better if upgrading to 18.1 can be done tranparently in your case.

    However, if you never updated, regressions in LOS cannot be the reason for your troubles anyway, so what would be the point of upgrading now?

  2. 4 hours ago, wrc08 said:

    If I reset the factory settings, then everything is gone ...

    Before wiping everything, I would first try to just uninstall MS Swiftkey, and see if that helps.

    I do not know that software from personal experience, but from what I read, it does have functions related to hardware keyboards. Maybe there is some conflict with native keyboard handling in Lineage.

    Also, there has been quite some development around the keyboard driver in LineageOS (18.1) during the last few months. While all these changes were of course meant to improve things, regressions for heavily customised set-ups are always possible, I guess. Which release and version of Lineage are you using?

  3. 2 hours ago, suicidal_orange said:

    It wouldn't take 5 minutes to tell everyone what they already know

    Considering how people here and on IGG analyse (and maybe over-interpret?) these monthly reports, I think they spend quite some time writing those up. I can imagine they plan very carefully what to publish and how to present it in a way such as to cause least damage to the project. It is easy for us to ask for more honest reports and estimates, but unfortunately it is also a fact that many customers need the harsh reality to be "embellished" for them, or else they jump ship.

    Some people understand that by supporting Fxtec, they support an interesting enthusiast project: a concept of a smartphone all big players have abandoned for a reason (profit), and they are ready to go through some lengths for that dream. But many (most?) of their customers do not care about such things, and never thought of the possibility of Fxtec actually failing on the marked and them loosing their money -- simply because that is not what they are used to when buying from big manufacturers. If you had honestly told those people they just payed 700 bucks to get a 2019 SoC in 2022, they'd have said: "f**k you, gimme my money back, I'll get an iPhone!".

    One year ago, I did not believe one second Fxtec would be able to develop and deliver what amounts to a completely new Pro1 model within 7 months. And I guess they, too, knew that their proclaimed delivery in August was an overly optimistic estimate. But Just imagine they had, in January 2021, openly admitted people would probably not get their phones before mid-2022. Would the project still be alive? I honestly do not know. But this is the kind of questions they have to answer for themselves whenever they need to communicate another setback.

    • Like 6
    • Sad 1
  4. 2 hours ago, Rob. S. said:

    Comments on IGG

    I like most the ones that tell me F(x)tec is a fake company and that my Pro1 does not exist. 🙃

    Thinking that through, it follows that we in this forum as well as all other people publishing Pro1 videos on the web, discussing bugs and hardware issues, making custom ROMs, etc., are just actors hired to distract from the fact that Fxtec have long moved to the Caribbean where they spend their backer's money on expensive cigars. One wonders if setting up such an elaborate smoke screen would really be easier (or cheaper) than just producing the phone after all ...

     

    • Haha 3
  5. On 6/22/2021 at 8:15 PM, claude0001 said:

    For the case anyone else still has interest in LineageOS 16.0, I've made them available here:

    http://findus.zwergenschaenke.net/~puma/linux.html#lineagepro1

    I finally made a new image (20220129-UNOFFICIAL) with the January 2022 security patch level.

    Full list of local mods with respect to upstream Lineage-16.0 tree (practically unchanged since December):

    • Modified gps.conf which (for me) enables the Pro1 to get its initial fix much faster.
    • Corrected QoS powerhint signals. Backport from Lineage 18.1.
    • Keyboard driver: Enable high keycodes Backport from Lineage 18.1.
    • Keyboard driver: Fn-Tab acts as Alt-Tab. Cherry-picked from an experimental patch proposed by @Slion against LOS 18.1. Was never merged into any official distribution.
    • Keyboard: Prevent unwanted multiple keystrokes: Backports of LOS 18.1 patches 318274 and 318496.
    • [New:] Some purely cosmetic changes to hardware keyboard settings panel, inspired by LOS 18.1.

    The source code changes with respect to upstream are now also available as a patchset at the above link.

    Have fun.

    • Like 6
  6. I've had one such dot since day one. It has not worsened one bit in 1.5 years. It seems to be a quite common QA problem with this kind of screen, not limited to the Pro1. I wouldn't worry too much about it.

    • Thanks 2
  7. On 1/25/2022 at 9:13 PM, LordV said:

    I was looking for a way to automatically turn it on, but everything I tried had no result.

    When I place a script named e.g. "99hwoverlays" with the following code

    #!/system/bin/sh
    #
    # Set "Disable WH Overlays" developer option at startup
    #
    bash -c "service call SurfaceFlinger 1008 i32 1" &

    into /data/local/userinit.d/, the "Disable HW Overlay" option gets set automatically after boot.

    Remember to make the script executable (chmod +x). Also note that user startup scripts in /data/local/userinit.d/ are no longer executed by default in modern LineageOS, you have to install some helper app like RunUserinit from F-Droid.

  8. 13 minutes ago, LordV said:

    Do I understand correctly that you have no problems with the animation? I have a very visible slowdown when scrolling through lists in all places without activating this feature.

    It's always difficult to discuss topics that are somewhat subjective ...

    It would probably be good if you could provide a specific and (hopefully) reproducible example of an app that does not scroll smoothly for you.

    I can scroll smoothly (as far as my expectations go) when browsing through a thread in QKSMS, when scrolling down a web page in Firefox, when displaying a thumbnail gallery in the default Lineage picture app, when displaying folder contents in MaterialFiles, and when browsing through a lengthy program in Acode.

  9. 1 hour ago, LordV said:

    If in the developer settings turn on the function "off hardware overlays" and "gpu rendering" then the problem disappears until you reboot the device.

    I have never played with those options and am certainly no expert in tuning the Android graphics stack to its max.

    However, there seems to be consensus that those options are not enabled by default for a reason. That's especially true for the "Disable HW Overlays" flag, which seems to worsen battery life (as more CPU load is generated). I'd rather not mess with those defaults in my ROM.

    I can confirm that also for me "Disable HW Overlays" is not persistent across reboots (while "Force GPU" is), which is kind of strange -- although it could also be an indication that this is really a debugging feature only that should not be enabled all the time on a production system. I could look into how to make that setting persistent for you, but it's going to take a while, probably.

    If you are rooted, there seem to be ways to enable "Disable HW Overlays" on boot automagically. Would that be an option for you?

  10. 6 hours ago, VaZso said:

    Right, but they planned that Pro1-X will be 100% identical to Pro1 inside

    That's a good point. You are right that the Pro1 could -- technically at least -- have had the upgrade to the Pro1X's Android 10 for "free" if the SoC thing had not happened.

    However, it is also true that updates of stock Android 9 stopped long before the mainboard change. As far as I know, they ran into trouble with their contractor making the OS, independent of the other issues with the SoC manufacturer. Let's hope they (and we) will have more luck with the Pro1X ... 

    • Like 4
  11. 2 hours ago, VaZso said:

    I think at the beginning, they wanted to do it differently but a lot of things have happened.

    I do not know from personal experience, but people close to the Lineage porting effort have hinted F(x)tec at least suspected they would not be able keep an official Android OS afloat -- long before the 835 thing happened. Seems like keeping a certified Android OS up-to-date is not so cheap after all, let alone porting newer Android major releases to the device ... 

  12. I cannot really comment on the state of the stock Android OS as I have been on Lineage since day one. However, especially those who jumped-off stock only recently seem to not look back at all. Considering how unhappy you seem to be with your stock Android 9, I'd suggest you just give Lineage (or some derivative) a chance.

    I think it is by now clear that F(x)tec more or less intended the Pro1 to be used with alternative ROMs. Yes, I fully agree that they should have either stated this more clearly from the beginning, or properly supported the original OS their phone was delivered with. But I think that ship has sailed.

    The good news is that your Pro1 is probably quite easy to sell even at full purchasing price, if its hardware is in decent condition ... 😉 

  13. 18 hours ago, Hook said:

    Wait. What? We have to convince our spouse??

    I know you are joking, but I meant that quite seriously.

    My wife brings home half of the family income, but, for some reason, I thought I need a phone that is 10x more expensive than hers (while having worse audio quality 😉 ). Of course I explained to her why I wanted that Pro1 so badly before throwing all that money at a company that, even then, I was not sure would be able to deliver at all. If you manage your entire household from a common financial pool, it is imho a question of mutual respect to discuss major expenses prior to engagement -- especially if they are about what most people will see as a luxury item ... 

    Back on topic: I think F(x)tec have shown that, with the Pro1, they can fulfil the demand of a certain niche marked. However, we have also seen that, already now, they are plagued by problems typical for a small-volume manufacturer: uncertain supply chains, lacking quality assurance, need to source second-choice hardware (uncalibrated screens).

    If, by trying to include too many fancy new ideas (adaptive keyboards), F(x)tec push the price of the individual device further up, their already niche marked will shrink, which will further aggravate their problems. So I think this is really not the way to go.

    Rather, they should build on what they have achieved, and incrementally improve the (after all) successful concept of a smartphone the Pro1 is. That's what the big players, mostly, do too.  

    • Like 2
    • Thanks 2
  14. 6 hours ago, VaZso said:

    [...] and should be solved also on Pro1 [...]

    You are right, but let's not further misuse this thread for discussing known issues of the original Pro1 (again).

    Lineage may be able to fix a few more things on the Pro1, but I think we have to accept the fact that, from F(x)tec's perspective, that one is done. Any software issues related to the binary blobs will not be fixed. Imho, we will be lucky to even get spare parts for components unique to the Pro1.

    Let's see how the Pro1-X comes out and how long F(x)tec plan to keep it available for purchase. If it improves on some of the issues already in the first production batch, and if they are willing to keep fixing remaining bugs in later batches, the Pro1-X may end up having a quite long product life before a Pro2 becomes necessary.

    • Like 2
  15. I very much enjoy the discussions here, and hope that also F(x)tec uses them for inspiration on future projects.

    But considering the features of a hypothetical Pro2, I think we should also be realistic: The Pro1/Pro1-X are very expensive phones even without including exotic technologies like e-ink-keyboards. I think 1000 €/$ are a magical landmark, where even enthusiastic keyboard-phone lovers will have a hard time convincing their spouses that they really need that device.  

    A Pro2 should probably focus on getting an up-to-date flagship SoC and improve on the known shortcomings of its predecessor(s). For the Pro1 those are (imho and in order of decreasing severity): main camera quality, keyboard quirks, telephony audio quality, and wifi signal strength. Possibly, the Pro1-X will improve significantly on some of these points, so that it should serve as the new benchmark for an all-out re-implementation of the concept.

    • Like 3
  16. Some time ago, I recommended Devuan as my GNU/Linux distribution of choice for installation in a chroot of LineageOS.

    Today, I wanted to upgrade my xrdp-pulseaudio modules (providing seamless sound forwarding to Android when using my X11 desktop) and found out that installing them on Devuan is not possible in a straightforward way using the tools provided on neutrinolabs' GitHub repo.

    The reason is that the drivers need to be compiled against the pulseaudio sources matching the version of each respective distribution. On Devuan, the command:

    # sudo apt build-dep pulseaudio

    fails, probably because of some unresolvable (and likely unnecessary) dependencies on systemd.

    So how to proceed? Disregarding the (non-)dependence of some packages on systemd, Devuan 3 (which I use) is binary-compatible with upstream Debian 10. So, if I had an ARM64 version of the latter, I could build the drivers and just copy them over to my Devuan 3 system.

    Luckily, there is schroot. It allows to seamlessly install and use a a foreign Linux distribution within one’s main OS. We install a minimal Debian 10 ('buster') system in an directory /var/chroot/buster. Note that all this will (temporarily) require at least ~1 GB of SD card space, make sure you have enough.

    # sudo apt install debootstrap schroot
    # sudo mkdir /var/chroot/buster
    # sudo debootstrap buster /var/chroot/buster http://deb.debian.org/debian

    After that, create the config file for the new distribution in /etc/schroot/chroot.d/buster.conf:

    [buster]
    description=Vanilla Debian 10 'buster' for package compilation
    directory=/var/chroot/buster
    root-users=<YOURUSER>
    users=<YOURUSER>
    type=directory 

    Enter your new Debian 10 system as root:
     

    # schroot -c buster -u root

    install sudo in Debian 10:
     

    > apt install sudo 

    Then, edit the the sudoers file in the chroot
     

    > visudo

    adding a line like
     

    <YOURUSER>    ALL=(ALL:ALL) ALL

    Save the sudoers file, exit the schroot and re-enter as regular user
     

    > exit
    # schroot -c buster

    For clarity, we use "#" to symbolize the Devuan 3 CLI prompt, and ">" for that of the Debian 10 'guest OS'.

    And, yes, we are in fact running a Debian 10 inside a chroot of a Devuan 3 which is, itself, running in a chroot of LineageOS. If you are not confused at this point, consider yourself hardcore. If you are, do not worry, we hold you. Either way, take a minute to meditate on the fact that the concepts taking us this far in 2022 were developed ~42 years ago in UNIX v7.

    Install the minimally necessary dev tools in the schroot:

    > sudo apt install build-essential dpkg-dev libpulse-dev git autoconf libtool

    Get the pulseaudio sources for Debian 10:
     

    > sudo echo "deb-src http://deb.debian.org/debian buster main" >> /etc/apt/sources.list
    > sudo apt update
    > cd ~
    > apt source pulseaudio

    Get the build dependencies for pulseaudio (as we are in vanilla Debian 10, this should work now):
     

    > sudo apt build-dep pulseaudio

    Configure the pulseaudio sources:
     

    > cd pulseaudio-XX.Y 
    > ./configure

    Clone the GIT repo with the xrdp driver sources:
     

    > cd ~
    > git clone https://github.com/neutrinolabs/pulseaudio-module-xrdp.git

    Compile the driver modules:
     

    > cd pulseaudio-module-xrdp
    > ./bootstrap && ./configure PULSE_DIR=~/pulseaudio-XX.Y
    > make

    Of course, running the provided “make install”-script from within the Debian 10 chroot does not make sense, as that would install the drivers only in the Debian 10 'guest', but not in the Devuan 3 'host', which is what we want. Either leave the schroot and run the installer script in Devuan 3, or just copy the drivers to /usr/lib/pulse-XX.Y/modules/ and configure pulseaudio manually, which is what I do. The compiled .so’s are in ~/pulseaudio-modules-xrdp/src/.libs/ .

    After successful compilation and driver installation, you may delete the Debian 10 schroot in /var/chroot/buster to free up some SD card space. You might also choose to keep it around as a dedicated building environment, allowing you to (re-)compile more stuff without polluting your 'main' distribution with hundreds of dev packages.

    Have fun!

    • Like 2
    • Thanks 2
  17. 7 minutes ago, VaZso said:

    I think if you may connect a wired mouse to the phone using an OTG conversion then you can give input pattern using a mouse pointer.

    Yes, just tested that: USB mouse works for pattern-based unlocking, even if one connects the mouse only while the phone is already locked.

    Bluetooth mouse could work, if it is already paired with the device and BT is switched on. Otherwise probably not ...

    • Like 1
    • Thanks 1
  18. 1 hour ago, toast said:

    Seems like (not very surprising) /e/OS came out on top, followed by LineageOS

    Thanks for sharing this article.

    Note that the comparison is quite unfair with respect to LineageOS in that, in this study, the LOS device had GApps installed, which then -- unsurprisingly -- phone home to Google in the same way they do when installed on a commercial AndroidOS. The Lineage Project does not ship GApps in their ROMs -- users have to install them separately (and willingly) from a 3rd-party source after flashing the OS.

    I think it is safe to assume that a LineageOS with MicroG (as also used by /e/) or a vanilla LineageOS (with neither GApps nor MicroG installed) would come out essentially on par with /e/ in this study.

    • Like 5
  19. 14 hours ago, manjaro said:

    I did not want to totally dismiss what has been done with Ubuntu Touch. I can see that a lot of work has been put into it. I just think, and this has happened to me in my career, the target and goal were off the mark. They are trying to meet Apple in terms of GUI standards, it appears. There is no way that will come to pass. Apple is vertically integrated whereas Ubuntu Touch would need to adapt to whatever hardware it gets installed onto.

    That could not be more true.

    Canonical (at the time they still backed the project) would have liked us to think about Ubuntu Touch and the Ubuntu distribution for PCs as two flavours of the same OS -- as in iOS vs. MacOS. Many users unaware of technical details may believe that to this date.

    The truth is, for all but a few exotic devices (like the PinePhone), Ubuntu Touch does not even use an Ubuntu Linux kernel! It bundles a minimal version of the device-vendor supplied Android system, including its kernel and driver blobs. It then uses compatibility layers (libhybris) to interface the AndroidOS (bionic) from the Ubuntu userspace (GNU). As a consequence, only apps specifically developed against libhybris can take advantage of all Android APIs, while standard programs developed for the desktop arm64 Ubuntu distribution are limited in their system calls to what the vendor kernel allows directly. That's the reason why, e.g., X11 programs running in Libertine containers cannot have hardware accelerated graphics.

    To my knowledge, the limitations one encounters when running standard Unix software on Ubuntu Touch are identical when compared to running the same software in a chroot of Android. I hence gave up on waiting for Ubuntu Touch to mature and settled for a chroot-on-Lineage solution with my Pro1. That way, I have the best (or close to) of both worlds: a fully-functional Android smartphone environment plus a Debian desktop distribution for doing serious work.

    I agree that having a native GNU/Linux system would be preferable, but the Pro1 is not that kind of device. Neither will be the Pro1X.   

    • Like 2
    • Thanks 1
  20. After reading a little, I am under the impression that /e/ is indeed more about the default software distribution than about the underlying OS. Not to turn down that effort -- sane defaults that just work out of the box are certainly a good thing.

    I'll keep using vanilla Lineage (without Gapps or MicroG) for now. The Aurora store also displays tracker information from Exodus btw. But most of my apps are from F-Droid anyway...

    Thanks for sharing.

    • Like 2
  21. 13 hours ago, manjaro said:

    No way to install common unix software like curl. W - T - F

    I believe you have fundamentally wrong expectations regarding Ubuntu Touch. It is not designed as a traditional Unix, but as a smartphone OS, sharing quite a few concepts with Android. This includes app confinement (sandboxing), which actively breaks core Unix philosophy, as data "belonging" to one program purposedly cannot be easily picked up by another.

    Installing a true Unix environment on UbuntuTouch is possible, but involves quite similar techniques than doing the same on Android, the difference being that with UbuntuTouch you use an LXC container while on Android we rely on traditional chroot'ing. 

    If you seek a "standard" GNU/Linux OS for phones, your Manjaro on PinePhone was actually a much better bet. Sailfish may be the closest you can get with standard Android-devices (like the Pro1).

    I agree with @EskeRahn that this discussion has little to do with the Pro1(X).

    • Like 1
×
×
  • Create New...

Important Information

Terms