| sicelo | looks like they exercised their freedoms extensively, given they owned qt by then ;-) | 00:15 |
|---|---|---|
| uvos | maybe conversations could just be a sphone ui modlue, then it would not have to worry about any backend stuff, and scope would be quite limited at first - to the stuff sphone supports | 01:34 |
| uvos | recent qt style changes have broken at least smplayer btw | 01:36 |
| Wizzup | broken how? | 01:38 |
| freemangordon | uvos: how was smplayer broken? | 07:03 |
| freemangordon | menu text colors? | 07:04 |
| freemangordon | uvos: keep in mind smplayer uses its own stylesheet | 07:47 |
| freemangordon | if you choose maemo5 style from options it looks ok to me | 08:18 |
| uvos__ | freemangordon: right @smplayer i tripped over that, its fine | 09:21 |
| Wizzup | where is this option? | 09:21 |
| uvos__ | Wizzup: options->settings->interface | 09:22 |
| freemangordon | we have an issue with filechooser, it is not native. I am on it | 10:47 |
| norayr | freemangordon: i tried, but i got no vibration on notifications. i am updating again. | 14:00 |
| norayr | to try. | 14:00 |
| norayr | btw i did something with my terminal and it now shows very smaaaallll letters. | 14:00 |
| norayr | i think i was able to increase decrease fonts but now i cannot figure out how i did that in past. | 14:00 |
| Wizzup | volume butons? | 14:04 |
| uvos__ | norayr: device? | 14:06 |
| uvos__ | what sceanrio dose it not vibrate? you have vibration enabeld in the profile right? | 14:06 |
| norayr | uvos__: d4 | 14:08 |
| norayr | yes, it vibrates when i receive mails. | 14:08 |
| norayr | i was asking fmg to make it vibrate when i get notifications from dino. | 14:08 |
| norayr | because i miss what my friends write me when i walk. | 14:08 |
| norayr | even when i look at the phone every 10 minutes. | 14:08 |
| norayr | though it is very exhausting i try to do that. | 14:08 |
| norayr | but then it happens that i looked not every ten minutes, but every 20-30 | 14:09 |
| norayr | and i miss what they write. that is a big problem for me. | 14:09 |
| norayr | and since dino already sends notifs via libnotify, and maemo shows those. | 14:09 |
| norayr | i asked fmg if he can do something about it, so that other programs, that send notifs via libnotify | 14:09 |
| norayr | would vibrate. | 14:09 |
| norayr | very short. he said in the beginning that it doesn't correspond to freedesktop.org spec, but well, i need to get notifs from regular linux apps. | 14:10 |
| norayr | so he said to test it after some time, but i tested and dino doesn't cause vibration again. | 14:10 |
| norayr | uvos__: i have something to tell you. i discovered an Xfree function that | 14:10 |
| norayr | adds a key to the current layout. | 14:11 |
| uvos__ | i know of this function | 14:11 |
| norayr | that can solve our keyboard problem, maybe with just one function call. | 14:11 |
| uvos__ | but thank you | 14:11 |
| norayr | oh you read what i wrote? | 14:11 |
| norayr | do you think it can help? when the key is pressed, the function would add that key to the current layout? | 14:11 |
| uvos__ | no i mean this function was not news to me, but thank you anyhow. | 14:11 |
| norayr | i found it from svkbd source. | 14:11 |
| uvos__ | this corrispondes to the solution of generateing a keymap on the fly | 14:12 |
| norayr | yes, instead of doing setxkbmap every time. | 14:12 |
| uvos__ | sure | 14:12 |
| uvos__ | but you still cause a lot of wakeups | 14:12 |
| uvos__ | because all of the applications need to get the new sybol translation table | 14:12 |
| uvos__ | so you dont want to do this all the tie | 14:12 |
| uvos__ | *time | 14:12 |
| uvos__ | svkbd is a good hint | 14:13 |
| uvos__ | ill look at what they do, essentally you would have to have some region reserved for him | 14:13 |
| uvos__ | and then place the next x chars you need there | 14:13 |
| norayr | well if that solven a kbd issue, i want. because currently our keyboard without english is only useful with maemo apps or if one knows they need to do setxkbmap by hand. | 14:13 |
| uvos__ | and then send the codes and change it to the next x keys | 14:13 |
| uvos__ | to avoid wakeing up all x clients on eatch key input | 14:13 |
| norayr | but even that is a problem because when i do setxkbmap by hand, some symbols from armenian keyboard layout, let's say the hyphen don't work. | 14:14 |
| uvos__ | so anyhow, yeah you can do this | 14:14 |
| uvos__ | but it needs some tought | 14:14 |
| norayr | before i discovered the func i was thinking to create a map - maemo layout - xkbmap | 14:14 |
| uvos__ | that imo also makes sense | 14:14 |
| norayr | and do setxkb by program when the layout is switched. | 14:14 |
| norayr | but | 14:14 |
| uvos__ | since its the best solution power wise | 14:15 |
| uvos__ | updateing the syboml table is very expensive | 14:15 |
| uvos__ | *symbol | 14:15 |
| norayr | then as i say, if i do setxkbmap, symbols from latin, that i use in armenian layout, don't show. | 14:15 |
| norayr | so i need to do setxkbmap us. | 14:15 |
| norayr | so svkbd's approach i really like. | 14:15 |
| norayr | i discovered it because when i created armenian layout for svkbd, which i use on sxmo/pinephone | 14:15 |
| uvos__ | well you could have a layout for each supported langugage x eatch device with hwkbd + ibm pc layout | 14:15 |
| uvos__ | you have 251 keycodes available | 14:16 |
| norayr | it had a problem, and i looked in to sources, and found that function, and i have another patch which i need to send them, maybe today i will send. | 14:16 |
| uvos__ | this is sufficant for 2 languages | 14:16 |
| uvos__ | this is the "best" soluton | 14:16 |
| uvos__ | but requires one to create a bunch of maps | 14:16 |
| uvos__ | (you can automate this ofc= | 14:16 |
| uvos__ | ) | 14:16 |
| norayr | hmmmm... | 14:16 |
| norayr | okay, i might do that. | 14:17 |
| norayr | so i found that func in svkbd, they do it, and it works. and i thought that might be best solution. | 14:17 |
| uvos__ | it works | 14:17 |
| norayr | but if u really think the best is to have maps i may try to generate maps. | 14:17 |
| uvos__ | its realitvely low effort | 14:17 |
| norayr | yes. | 14:17 |
| uvos__ | idk if its the best way | 14:17 |
| uvos__ | this is why i say it needs some tought | 14:17 |
| norayr | but for now it would solve the problem. | 14:17 |
| norayr | and later we can think of replacing that. | 14:17 |
| norayr | just it is a torture to write in any non-english layout currently. and most of the users don't even know they need to do setxkbmap. | 14:18 |
| norayr | plus if one don't have hw kbd, it is harder to do setxkbmap. with hwkbd u always have latin to type the command. | 14:19 |
| norayr | but on pinephone it is not the case. | 14:19 |
| norayr | thank you for discussing it. | 14:19 |
| norayr | sorry what was the way to increase font size in terminal? | 14:40 |
| norayr | freemangordon: i just checked dino again, it receives the message, shows maemo notification, in separate window, but does not vibrate. | 14:43 |
| norayr | folks something happened. not latest version of photolightmeter is in repos. | 14:56 |
| norayr | it is 1.9.8-m7, but that is very old. it should be 2.0.0.1-leste1+2m7 | 14:56 |
| norayr | oh i did something wrong when i was making chimaera build. | 15:07 |
| buZz | eh huh | 15:38 |
| buZz | my d4 was on usb power all night | 15:38 |
| buZz | yet, battery percentage is absent? :O | 15:38 |
| buZz | prior i even saw '100% battery' in /sys , while battery voltage was 3.9V :/ | 15:38 |
| buZz | powercycling now to see if that fixes it | 15:39 |
| norayr | rebuilt photolightmeter with normal recent changes. | 15:41 |
| norayr | yes, terminal font is controlled by volume up/down keys. | 15:42 |
| norayr | so how is leste on n900? i have a dev friend, i have n900, i want to give him that n900 as a present, though the blood will pour from my heart. (: | 15:42 |
| norayr | in hope that he will contribute. | 15:42 |
| norayr | i had another n900, i gave it to other dev friend, and he looked so excited, but didn't do anything. :/ didn't even install leste. :/ | 15:43 |
| norayr | so pity. | 15:43 |
| norayr | this friend doesn't have any enthusiasm about mobile linuxes or phones at all, and has no phone. | 15:43 |
| norayr | and he lives in other city and other country. i am thinking to send him other n900. :/ | 15:43 |
| norayr | the HAM shows the new version | 15:44 |
| norayr | ok i hope the users will get the update. | 15:45 |
| buZz | norayr: ask the n900 back from the first user | 16:02 |
| buZz | :) | 16:02 |
| buZz | weird, i dont believe my phone now ever gets to 'Fully Charged' :O | 16:43 |
| buZz | its charging at 10mA now | 16:43 |
| Wizzup | battery unhappy/dead? | 16:45 |
| buZz | battery seems totally fine | 16:45 |
| buZz | 4.28V now down from 4.31 , while still charging | 16:46 |
| bencoh | uh, voltage doing down while charging doesn't sound normal | 16:58 |
| bencoh | going* | 16:59 |
| buZz | bencoh: pretty normal while we dont switch to USB power | 17:05 |
| buZz | the phone isnt powered by USB while charging, or at least, not exclusively | 17:05 |
| buZz | while the hw -can- do that | 17:05 |
| bencoh | oh right | 17:06 |
| buZz | already back at 4.3v now , it keeps going up&down, but never reaching 'fully charged' | 17:06 |
| buZz | so no percentage displayed :( | 17:06 |
| buZz | i -thought- this was all happening in kernel? | 17:07 |
| buZz | so i dont really understand what/why this is happening | 17:07 |
| Wizzup | what makes you tink it is not happening in kernel | 17:19 |
| Wizzup | V4.28 sounds like just about the max | 17:19 |
| buZz | well , its been >4.2 for -hours- now | 17:19 |
| buZz | cpcap_battery cpcap_battery.0: calibration done: 0x0000 | 17:21 |
| Wizzup | what do you expect? | 17:21 |
| buZz | huh | 17:21 |
| Wizzup | all of this seems normal | 17:21 |
| Wizzup | like, in line with things | 17:21 |
| buZz | Wizzup: i expect to get 'fully charged' and see a percentage in /sys/class/powersupply/battery/uevent | 17:21 |
| buZz | i mean, this works fine, there's just no percentage shown | 17:21 |
| buZz | its been on charger all night, and most of today, but never reached charged | 17:22 |
| freemangordon | norayr: did you reboot after upgrade? | 17:49 |
| arno11 | Wizzup: n900 voice call news: now my patch is working with states at the right time with no external troubles. Only one bug remaining: if it takes too long to answer a call a weird latency happened during few seconds. | 19:39 |
| arno11 | and i think it is because of the ringtone running at 44100hz. | 19:41 |
| arno11 | to be sure i will test a 4000hz ringtone to avoid 2 different rates at the same time. | 19:42 |
| Wizzup | arno11: great progress! | 20:47 |
| uvos | why would voice and the rinnger ever be active at the same time? | 20:51 |
| arno11 | uvos: good question. in fact record and play are not starting exactly when you answer the call | 21:00 |
| arno11 | and anyway 3 pulseaudio process are running at the same time | 21:01 |
| arno11 | cmtspeech is really tricky... | 21:02 |
| arno11 | per default record starts when the phone starts ringing | 21:02 |
| Wizzup | arno11: 3 PA servers? | 21:06 |
| arno11 | yes probably | 21:06 |
| Wizzup | that seems really wrong to me | 21:07 |
| Wizzup | you sure? | 21:07 |
| arno11 | not 100 percent sure | 21:07 |
| arno11 | but 80 lol | 21:07 |
| arno11 | i just doublechecked: 3 server daemons | 21:15 |
| Wizzup | are they just threads of the same server? | 21:16 |
| arno11 | probably one for general stuff. one for record/play. and one for buffer/dynamic latency ? | 21:16 |
| arno11 | htop shows 3 different daemons | 21:17 |
| Wizzup | dinner time | 21:17 |
| Wizzup | I can check on my n900 in a bit | 21:18 |
| arno11 | ok enjoy | 21:18 |
| Wizzup | how's latency looking? | 21:18 |
| arno11 | latency is fine but tricky: PA constently try to balance I/O latency | 21:19 |
| arno11 | for example if incoming and outgoing latency are both 200ms that's sound like zero ms | 21:20 |
| arno11 | the diff is the real latency | 21:20 |
| arno11 | so sometimes when local latency is 20ms and incoming is 40 the result is not better | 21:22 |
| arno11 | so i think the most important thing is I/O latency stability | 21:29 |
| arno11 | uvos: freemangordon: Wizzup: sicelo: for me the main concern is why Xorg burns 35-40% cpu during calls (and 5 without...) | 21:38 |
| Wizzup | yeah, that's weird for sure | 21:59 |
| arno11 | yep | 22:01 |
| uvos | is any non-obivous x client also using a lot of cpu in call? | 22:47 |
| norayr | freemangordon: yes, today i did. and no result. :/ | 22:51 |
| norayr | last time i also did. | 22:52 |
| norayr | buZz: yeah right? | 22:52 |
| arno11 | uvos: apparently not | 22:55 |
| buZz | norayr: whats right? | 22:57 |
| norayr | right to ask him to return n900. | 22:59 |
| norayr | but it is very hard for me... | 22:59 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!