Guest224 | is there any "EU-money" in project left to buy to open some some sources? I don't think Nokia ask lot of money when tell that that code is over 10 years old. | 00:04 |
---|---|---|
Guest224 | and latency issue...original code use DSP? | 00:07 |
Wizzup | the foundation has some money but it doesn't seem realistic that they would do this | 00:07 |
Wizzup | you can ask them, but I doubt anyone still even has the source there | 00:08 |
Wizzup | the latency is probably something else, I don't have any details atm so it's hard to speculate | 00:08 |
Wizzup | but pulseaudio by default (say parec) has a big latency | 00:08 |
Wizzup | 1-2 seconds | 00:08 |
Wizzup | this can be tweaked | 00:08 |
Wizzup | but I'd have to check with pavel about that | 00:08 |
Guest224 | hmm..even some people use in x86 PC realtime kernel-version to put down latency in audio use....hmm i have seen some documentations about that...searching... | 00:12 |
Guest224 | ok..there was something at, but not anymore: http://www.pengutronix.de/software/linux-rt/debian_en.html | 00:19 |
Guest224 | anyway realtime kernel patches might not be optimal solution for mobile. | 00:36 |
Wizzup | yeah, this is likely a different problem | 00:41 |
Guest224 | anyway 1-2 seconds latency sounds very big...when my 20 years old 0.6 Ghz Mac has only 1ms at 24bit/96khz audio in/out latency (ok..i have some special PCI-card :) and not in pre-emptive multitasking OS) | 00:48 |
Guest224 | GSM 06.10.-audio codec operates with 160 bytes and needs 20 ms even with zero network latency. | 01:02 |
Wizzup | right, it's a known problem, probably with a simple fix .. ultimately, once we know what is wrong | 01:03 |
Guest224 | but good to know GSM 06.10 decoded sound is only 8khz samplerate...should drop soundquality to that quality when phone calls "playing" so no need nice big conversion buffer and filttering...hmmm..is there good low level documetation about N900 audio? | 01:10 |
Guest224 | it will be ugly hack for problem, shut off pulseaudio when phone call is aswered and use directly N900 hardware :) | 01:20 |
Guest224 | to sleep -> | 01:21 |
sicelo | interestingly, my d4 that's not connected to USB in any way reports charging in sysfs, and /sys/class/power_supply/usb/online = 1 | 09:11 |
sicelo | in fact in the last few weeks, i notice some flakiness when it comes to charging/usb behavior - at some point, it did not detect usb connection at all, until i rebooted it | 09:12 |
sicelo | i doubt that it's failing hardware, e.g. cable or port, since normal operation is restored after reboot. | 09:13 |
* sicelo has just noticed, for the first time, that uevent files are writable ... wonder what's the reason for that | 09:15 | |
Kevan | hello all! | 11:48 |
Kevan | I saw a video of Maemo-Leste running on a Unihertz Titan.. anyone have an idea on which download was used and what the support was like? | 11:48 |
Wizzup | hi | 11:51 |
Wizzup | I believe that video was someone running maemo leste in a chroot on the unihertz titan | 11:52 |
Wizzup | we helped him set that up in here, but I am not sure if he produced a final log of what he did | 11:52 |
Wizzup | I assume you mean this video? https://www.youtube.com/watch?v=OqFHivcPIRM | 11:53 |
Kevan | impressive how responsive it was in a chroot though | 11:53 |
Kevan | yeah thats it | 11:53 |
Wizzup | he probably will respond at least on the youtube comments, I am not sure if there is a better way to contact him | 11:53 |
Kevan | I'm just glad you're on IRC too | 11:53 |
Kevan | everything going discord makes my heart hurt | 11:54 |
Wizzup | we'd probably ideally want it running natively at some point, but atm we don't (even) have an image for chrooting, but you could try to do it yourself and document it, I guess | 11:54 |
Wizzup | we might at some point have some form of support for libhybris too, but right now we're quite mainline oriented, that is, we want the kernel to be new and without too many (if at all) patches, from kernel.org | 11:55 |
Kevan | I may be interested in doing that later - device is in the mail. I'm a developer myself, work on a small slackware based distro no one knows about.. busy with 2 jobs and a toddler but a fully functional Linux phone would be a time saver | 11:55 |
Kevan | what version of the kernel is it running at the moment? | 11:56 |
Wizzup | on unihertz, I don't know, some android thing, ours are 6.1 / 6.2 | 11:57 |
Wizzup | where ours is n900, pinephone, droid 4, droid bionic, etc | 11:57 |
Kevan | wow | 12:01 |
Kevan | quite frankly I would be happy with 4.19 | 12:02 |
Kevan | lol | 12:02 |
Kevan | the droids are closed source arent they? how do you even begin to make your images work with that? Seems the unihertz would be in the same ballpark | 12:03 |
Wizzup | Kevan: do you mean the motorola droid series? | 12:12 |
Wizzup | well some kernel hackers did a lot of work on it (sre, tmlind who is here), and there is a way to boot the new kernels | 12:13 |
Wizzup | and from there you boot a distro and work on it ... for 5+ years :) | 12:13 |
freemangordon | :) | 12:16 |
Kevan | haha nice! | 12:16 |
Kevan | I started messing with stripping down and modifying Slackware back in 2008~.. Did builds for years but didnt release anything for a long time. Developed against -current64 until 15.0 and now I'm trying to switch all the stuff I'd written from Python to Go.. | 12:17 |
Kevan | https://sourceforge.net/projects/allegianceos/ super alphas really | 12:18 |
Kevan | (as in very alpha not AWESOME) lol | 12:18 |
Wizzup | the gtk3 theming changes were a pain for sure | 12:20 |
freemangordon | Kevan: definitely we lack manpower, so any help would be appreciated | 12:20 |
* Wizzup back in a bit | 12:20 | |
Kevan | yeah I've basically dropped gtk3 as the base | 12:20 |
Kevan | all the upstream .. crap.. has made it hard to make things look uniform | 12:21 |
freemangordon | I am going to buy pro1 at some point, so that's another up-to-date device we will hopefullt support | 12:21 |
Kevan | and my UIs were tied to pygtk3 and whatnot.. so just trying to go with Golang executables that are independent | 12:21 |
freemangordon | isn;t that too heavy for mobile? | 12:21 |
Kevan | thats a 64bit computer OS not mobile | 12:22 |
Kevan | i have almost no experience working on mobile | 12:22 |
freemangordon | well, yeah. | 12:22 |
Kevan | last time I was close was these old Hitachi Visionplates with Transmeta Corp processors, 128mb CF cards and Firefox 1.0.0.0.12 Beta.. lol | 12:23 |
freemangordon | my point was that it will be too heavy to be used on mobile, in the context of leste | 12:23 |
freemangordon | like, because we use outdated gtk2 and at some point will have to move on | 12:23 |
Kevan | oh no doubt | 12:23 |
Wizzup | golang is fast, but mixing too many ui toolkits doesn't help | 12:23 |
* Wizzup bbl | 12:23 | |
freemangordon | it is not only about speed, but how it would affect battery usage etc. you know | 12:24 |
* freemangordon have to run as well | 12:24 | |
freemangordon | bbl | 12:24 |
Kevan | alright thanks for the info guys :) | 12:24 |
Wizzup | sure thing, stick around for lots more info daily :p | 13:02 |
Kevan | haha I'm sure | 13:03 |
Kevan | getting into handheld dev has been sort of a nagging desire ever since those Visionplates | 13:03 |
Kevan | spent 10 months rewiring Slackware 10.2 to work on them | 13:03 |
Kevan | then Goodwill'd the lot lol | 13:03 |
Wizzup | I had gentoo on my phone before this project, but not in any way usable of course | 13:09 |
dsc_ | https://paste.debian.net/plain/1269947 :(( | 17:21 |
Wizzup | dsc_: do you get this after dist-upgrade too? | 18:02 |
dsc_ | Wizzup: yeah | 18:04 |
dsc_ | Linux devuan 5.10.0-20-amd64 #1 SMP Debian 5.10.158-2 (2022-12-13) x86_64 GNU/Linux | 18:05 |
bencoh | is that oldstable -> stable? | 18:09 |
bencoh | or just dist-upgrade on stable? | 18:10 |
dsc_ | good q | 18:16 |
dsc_ | i forgot, but most likely the last | 18:16 |
dsc_ | https://maedevu.maemo.org/images/virtual-machines/20221213-chimaera/ is this latest? | 18:18 |
dsc_ | as in, I can update from this image | 18:18 |
dsc_ | (and that would work) | 18:18 |
dsc_ | try installing `libmafw-shared0-dev` on chimerea i guess | 18:21 |
dsc_ | maybe you will also get the error ^^ | 18:21 |
Wizzup | dsc_: will try to see if I can fix this, weird that ck doesn't satisfy this | 18:43 |
Wizzup | freemangordon: ^ | 18:43 |
dsc_ | Wizzup: thx | 19:13 |
freemangordon | Wizzup: gvfs-backends is the one that pulls something | 22:27 |
Wizzup | freemangordon: and it pulls elogind even with ck installed? | 23:00 |
freemangordon | gvfs-daemons depends libelogind0 :( | 23:00 |
freemangordon | thats crazy | 23:01 |
freemangordon | hmm | 23:02 |
freemangordon | this is not an issue | 23:02 |
* freemangordon is still checking | 23:02 | |
freemangordon | Wizzup: udisks2 | 23:08 |
freemangordon | see https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=888892 | 23:15 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!