@domubpkm If you understood the workload involved to update Qt and it's submodules for an old version you would probably also understand that this is more than one should ask for in the first place.

The short answer is no, the long answer is: there is no sane point in keeping unsupported things around if nobody cares for (not "cares about") them. In other words: Suppose I care about pudding, I don't care for it, hence I want to consume pudding, not watch it grow mold.

In the end people need to move on, and as long as people stick with xenial just because they believe there is a point (that includes port maintainers, system integrators, app developers, and users) the resources will be spread too thin and the result for focal and what's beyond will look poor. The only real solution is to move on fast, period.

@lsitongia I hope you are aware that's 15% of ONE core

Two things:

• I found the reason for one particular A/V desync problem I could reproduce, and have proposed an MR: https://github.com/Halium/hybris-patches/pull/58

• I'm tweaking the job to clean up old GStreamer caches earlier now, which hopefully results in proper recordability and playback ability after reboots.

@emphrath you mean during playback on the device itself? Have you tried watching them on a PC yet?
Also, I would suggest against UT Media app, it doesn't support hardware acceleration.

@emphrath Okay I have pushed a potential fix now, tomorrow's devel update should include it. In the meantime, try to copy your beloved videos to your PC, try to open them, and attempt to fix them with ffmpeg if possible.

@emphrath If ffmpeg can handle them, then possibly. Other than that there is probably not much you can do. I know it sucks and probably hurts me the most (since I want to provide the features users care about) but yeah, don't know..

@emphrath Will take care of it real quick.

@uri Great. If it doesn't stay that way please let us know in a bug report, I can disable the offending piece of software permanently.

I've just tried to verify the issue, but I can't reproduce it. Have you attempted one more reboot after the upgrade? I had to change the video de- and encoding backend to account for wireless display support.

@peter-gamma You are aware that Aethercast, the Miracast compatible service on Ubuntu Touch, works now with some newer devices, including the Pixel 3a? There is no value whatsoever in supporting Chromecast other than refusing to buy a proper, open standard dongle.

The ability to stream the virtual display in desktop mode to a Chromecast receiver would first of all imply additional development resources and donations for me to consider it, and secondly probably some kind of NDA shenanigans to not open up the Chromecast implementation to the public (since none of the "blessed" implementations are open source).

Miracast is an open standard, Chromecast is not.

@dizzy Don't know what you're talking about, Bluetooth works pretty well here. Please be more specific, collect logs and let us know in a proper bug report what's not working as intended.

The fingerprint reader has been improved in accuracy in devel and will have the fix shipped as part of OTA-23. Problem was that it was always active in your pocket, causing the applet responsible for verifying the fingerprint to be confused.

@domubpkm Well since currently this is a device specific change it doesn't apply as-is to Volla, I actually don't know if that one has a Low Memory Killer activated.

In principle each device maintainer needs to make sure that low-memory situations are dealt with properly, one way or another. We currently don't have a unified approach to that.

The actual fix was this one: https://github.com/fredldotme/android_kernel_google_bonito/commit/944dd258c369ed224b37a9fa439a64c919ad4f27

@domubpkm that's not how development works usually, the tasks will be moved around when the features are ready and tested. Some are not ready and tested yet it seems. But it's basically "when it's done".

@3t_ed yes, at least we found out what happened with apps locking up everything, while in reality there was no low-memory killer active. I enabled it and it's available in devel or even release candidate already.

@aam42 without including the unlocking JingOS process itself, just a couple of minutes.

@flohack I would recommend going the Wayland way in case we want to care less about Halium 7.1 and lower in the long run.

I know it's not nice but it's "the future". For VSCodium I let everything run through XMir and it shows. It's unstable, has issues that are hard to fix and changes the experience immensely depending on form factor.

If X11 is the only way then we should focus on getting XWayland working well.

With that being said, I don't expect anyone to support Mirclient any longer. We depend on it hard but I would not recommend it for new toolkits. It can be done but IMO the energy is best reserved for getting Wayland to work nicely.

Strikes me as odd because the P3a XL should have worse battery life due to driving a bigger display.

I'm able to keep my phone running 1,5 to 2 days (granted, only for calls and occasional music, but still). So I'm surprised the battery life is worse for you.

EDIT: What I can tell you is: the battery life is better with WIFI as the 4G driver is waking up the device much more on mobile data, causing much higher battery drain.