**** BEGIN LOGGING AT Tue Nov 24 03:00:53 2015 Nov 24 06:19:17 DocScrutinizer05: cellphone is happily running now Nov 24 06:19:22 cheers :3 Nov 24 06:19:26 yw Nov 24 06:19:45 (as apposed to before the reflashing) Nov 24 09:36:22 hello Nov 24 09:39:48 KotCzarny: How exactly does the volume control work in oscp Nov 24 09:40:05 changing the volume seems to have no effect Nov 24 09:40:34 salamisami: it uses softvol, it doesnt touch system volume Nov 24 09:40:56 might be bugged because lately i've added libav resampler which might skip softvol in oscp Nov 24 09:41:24 thx for reminding me Nov 24 09:42:43 until i upload the fix you can just click status bar near the clock and change system volume Nov 24 09:46:24 not that straightforward with ssh :P Nov 24 09:46:31 ahm Nov 24 09:46:52 still, its a bug, and i've added it to the TODO Nov 24 09:47:13 right, good to know Nov 24 10:03:02 softvol means? how does one adjust it? Nov 24 10:03:56 sicelo, softvol means every sample is multiplied by some float by oscp, and that float is changeable by user (via -/= keys or clicking on vol bar) Nov 24 10:06:01 freemangordon: ping Nov 24 11:37:29 KotCzarny: oscp stops playing after a track finishes Nov 24 11:37:41 have to manually start new one Nov 24 11:37:42 always? Nov 24 11:37:46 yeah Nov 24 11:37:50 it gets stuck loading next one Nov 24 11:37:56 or did i press a wrong button? Nov 24 11:38:03 hmm, mp3 ? Nov 24 11:38:06 yep Nov 24 11:38:38 what happens if you press ']' (default next track button i think) Nov 24 11:39:22 hmm Nov 24 11:39:30 this seems to only be a problem with certain files on my playlist Nov 24 11:40:14 can you upload any of them somewhere? Nov 24 11:40:58 i got them by using mp3fiber to download them from youtube since i couldnt find them elsewhere Nov 24 11:41:25 offtopic, use youtube audio/video downloader for firefox Nov 24 11:41:49 it allows to download audio tracks of youtube clips easily and with selectable qulity Nov 24 11:41:51 *quality Nov 24 11:42:01 or cutecube2 on your N900 Nov 24 11:43:11 and note that oscp can play webm/mp4/m4a/ogg/opus etc Nov 24 11:43:17 so no need to reencode to mp3 Nov 24 11:43:27 alright Nov 24 13:14:21 ~seen pali Nov 24 13:14:21 pali <~pali@Maemo/community/contributor/Pali> was last seen on IRC in channel #neo900, 20d 13h 41m 47s ago, saying: 'campaign for what?'. Nov 24 16:03:20 something went haywire with my N900 .. would connect successfully with my home network, but not be able to send/receive data. in the end a modprobe -r wl12xx helped. weird Nov 24 16:21:57 almost normal Nov 24 16:23:03 anyway thanks for the modprobe hack Nov 24 16:23:55 I every once in a while notice WLAN locking up when walking out of range of an associated AP Nov 24 16:25:05 it seems to desperately try to reconnect to the no longer available AP and never check for new ones, neither switch to other connectivity channels i.e. GPRS Nov 24 16:25:20 I guess disabling/enabling wireless from the status-menu would work as well Nov 24 16:25:28 yes Nov 24 16:26:59 if you have the applet :) stock does not have enable/disable wireless on status menu Nov 24 16:27:17 hmm Nov 24 16:27:27 it has 'select new connection' Nov 24 16:27:40 and there it has 'disconnect connection' Nov 24 16:28:15 and i think disconnecting from wifi disables it (if you dont have search for networks enabled) Nov 24 16:29:39 well, that didn't help Nov 24 16:29:43 airplane mode Nov 24 16:29:45 KotCzarny: as far as I can tell, no, it doesnt Nov 24 16:30:20 anyway, never got that symptoms so i cant check Nov 24 16:30:26 *those Nov 24 16:38:20 ok, that's about as much of module-nokia-voice as I can do at this point :) Nov 24 16:38:38 now I need freemangordon :) Nov 24 16:38:52 or some other RE guru if one exists in this community :) Nov 24 17:15:16 My N900 had the WLAN lock up for the first time this week in a similar way. Very odd. Nov 24 17:15:38 jonwil: hmm? Nov 24 17:15:55 oh , the voice module :) Nov 24 17:16:01 yep Nov 24 17:16:13 I have done all I can at this point Nov 24 17:16:36 the other functions (see status.txt) all have things that I cant seem to figure out Nov 24 17:16:44 WOW, great amount of work on your side Nov 24 17:18:24 yeah, I did a lot :) Nov 24 17:19:29 need to do something more in regards to xprot temperature.c but to do that I need to pull in libbmeipc-dev and I cant seem to do that since scratchbox doesnt have kernel-battery-blah or whatever libbmeipc0 from pali wants Nov 24 17:23:44 where can i take a look at your job, jonwil? Nov 24 17:24:37 https://github.com/community-ssu/pulseaudio-nokia/ is the current pulseaudio-nokia tree Nov 24 17:27:30 anyhow, I have done all I can, time for zzz :) Nov 24 17:27:48 hopefully Freemangordon can pick it up from here and do some more stuff :) Nov 24 17:28:09 cya guys Nov 24 17:28:54 cya! Nov 24 17:55:59 freemangordon: what was the latest pali/mainline kernel that worked with ofono (or with maemo) for you Nov 24 17:56:02 call wise - I mean Nov 24 17:56:26 I have 4.1.0-rc4+ (Pali's branch) and I cannot get ofono to work yet Nov 24 18:26:13 re xprot I suspect it's quite a bit overengineered. I honestly doubt you need to ask BME(!!) about any temperature to determine whatever XPROT does to protect the speakers Nov 24 18:28:02 or would the IHF audio shut down when battery gets too hot? to protect battery from discharge-induced overheating? Nov 24 18:28:39 more like glue on speakers getting hot Nov 24 18:28:56 (and thats why it has to limit displacement more) Nov 24 19:17:47 couldn't they have just used better speakers or something? sounds like the speakers inspired convoluted systems Nov 24 19:19:44 yup Nov 24 19:27:21 Wizzup: never tried to make phone calls with mainline kernel, and for GPRS I used what is in maemo (csd or whatever) Nov 24 19:27:59 DocScrutinizer05: or most-probably they have had the same/similar problem with some other device and just reused the code Nov 24 19:29:43 freemangordon: did you see anything ""relevant"" in xprot? Nov 24 19:30:07 "relevant" as in? Nov 24 19:31:05 any function that seems to do something "useful" Nov 24 19:31:13 and/or that we understand Nov 24 19:31:45 well, the thing xprot does is a HPF Nov 24 19:32:00 fmg, isnt its more than that? Nov 24 19:32:01 freemangordon: right, so not ofono. well regardless, currently I don't get working gprs (problems in kernel modules) Nov 24 19:32:18 ie. limiting cone displacement is not only hpf Nov 24 19:34:02 freemangordon: highpass filter? Nov 24 19:34:43 KotCzarny: well, I didn't done thorough analysis of what xprot does, but check pngs in http://46.249.74.23/xprot/ Nov 24 19:35:18 haha, my browser marks the two jpeg links as "already visited" :-) Nov 24 19:35:32 bencoh: yes, highpass filter Nov 24 19:35:52 ~xprot Nov 24 19:35:52 xprot is, like, http://wiki.maemo.org/Porting/Audio/Q_and_A_RE-PA and http://talk.maemo.org/showthread.php?t=91325 Nov 24 19:36:13 KotCzarny: http://46.249.74.23/xprot/xprot.png is the input, http://46.249.74.23/xprot/xprotraw.png - the output Nov 24 19:36:41 hmm, I guess the speaker response curve is the opposite/inverse, or I dont see how we're supposed to listen to music with that Nov 24 19:36:49 (unless speakers were never meant to listen to music) Nov 24 19:36:57 KotCzarny: or you can take the 2 raw files and do whatever analysis you want :) Nov 24 19:37:14 bencoh: wait, why? Nov 24 19:37:58 What is xprot displacement? Nov 24 19:38:01 XProt displacement - don't remember. Something about the speaker coil safe traversal probably. Nov 24 19:38:16 for some reason they mantion it and not a simple hpf Nov 24 19:38:19 wait, xprot.png is input? Nov 24 19:38:20 *mention Nov 24 19:38:34 sure, this is not a simple hpf Nov 24 19:38:52 but functions similarly to Nov 24 19:39:07 would be meaningful to draw sines/triangles before/after xprot? Nov 24 19:39:19 maybe square waveforms too Nov 24 19:39:35 sines doesn't make sense Nov 24 19:40:09 squares might make sense, but still, I expect output similar to that of an HFP Nov 24 19:40:53 also, there is a controlled gain, I guess based on the predicted coil temp Nov 24 19:41:27 isn't it more of a bandpass if the two pictures are input/output? Nov 24 19:41:43 centered around ... 1k-2k? Nov 24 19:41:55 yeah, why not :) Nov 24 19:42:12 but see the input dB Nov 24 19:42:12 just "enhancing" human voice Nov 24 19:42:49 (well, more amplifying than enhancing) Nov 24 19:42:57 more like constant filter up to ~800hz, then linear catch up to no-filter at 2khz Nov 24 19:43:03 freemangordon: what about it? Nov 24 19:43:20 bencoh, look at the scale Nov 24 19:43:36 at 2khz they both are -30db Nov 24 19:43:48 also, we don;t know the electro-acoustic parameters of the speakers and their case, so we don;t really know what is the real output Nov 24 19:43:57 true Nov 24 19:43:57 :nod: Nov 24 19:45:06 also, if you manage to find speakers that fit with almost no hacking into n900 that are without the flaw, it would be cool Nov 24 19:45:39 Butterworth filter? Nov 24 19:47:13 but I don;t have the phase response to judge with certainty Nov 24 19:47:54 not to say it was a while I was student :D Nov 24 19:48:55 if you keep learning, you just go on a self-directed student program Nov 24 19:50:33 the point is that I knew lot of this stuff (analog/digital filters, freq/phase diagrams, poles, etc, etc), but seems I have forgotten most of it Nov 24 19:51:15 freemangordon: ((most-probably they have had the same/similar problem with some other device)) hardly, iirc they developed XPROT in a hurry while delaying shipping of N900 since they found out "the speakers will break on your device, and you don't want that to happen. So please stay patient until we deliver the fixed device to you" Nov 24 19:51:27 I hope we'll be able to build a make a brain snapshot someday Nov 24 19:51:41 (not really hoping, but .... that'd be cool) Nov 24 19:52:12 bencoh, pray that we will not Nov 24 19:52:24 otherwise humans will get annihilated Nov 24 19:53:13 DocScrutinizer05: could be, who knows. but still, I think the filter code and coil temp prediction was borrowed from somewhere else and just adopted Nov 24 19:54:39 that's possible, and they used an academic model that counts in irrelevant environment 'variables' like "ambient temperature" from BME Nov 24 19:54:55 my point exactly Nov 24 19:57:43 the only feedback it gets is temperature? Nov 24 19:57:47 bencoh: ((isn't it more of a bandpass)) no, it's a highpass that cuts below some 800(?)Hz and finally applies linear attenuation below 400(?) HZ Nov 24 19:58:09 bencoh: there is no feedback Nov 24 19:58:16 :D Nov 24 19:58:18 freemangordon: ^^^, using white noise instead of pink noise would yield more legible diagrams Nov 24 19:58:21 ;-) Nov 24 19:58:23 then I really dont understand Nov 24 19:58:34 +1 for the white noise Nov 24 19:59:25 DocScrutinizer05: iirc it was you who wanted freq response diagram for pink noise :D Nov 24 19:59:35 :D Nov 24 19:59:53 bencoh: just tilt your head 45° to the right before watching both diagrams Nov 24 20:00:14 freemangordon: yes, but *I* can read it ;-) Nov 24 20:00:47 that's why I provided it, for you to read it :P Nov 24 20:01:14 and pink noise is way more 'natural' so any dinamic parameters in XPROT will adjust to more natural usual values with pink Nov 24 20:02:02 freemangordon: for bencoh you could provide a drived plot: $output - $input Nov 24 20:02:32 or he can just grab those 2 .raw files and feed audacity with them ;) Nov 24 20:02:32 logarithmic X and Y scale Nov 24 20:03:04 and then drop the base Nov 24 20:03:33 and he can even feed matlab with the data to recover the transfer function :D Nov 24 20:03:39 i dont remember which freq pattern resulted in constant cone displacement Nov 24 20:04:08 maybe feed some signal with non zero dc? Nov 24 20:04:19 bencoh: the filter has a __/'''''''''' characteristic, with __ being some -20dB or whatever, and "/" between ~400 and ~800Hz Nov 24 20:04:37 DocScrutinizer05: :-)) Nov 24 20:06:20 bencoh: in "output" draw a line on left "\" and a parallel line on right "\" you then consider the right line to be the X-axis with a Y of 0dB Nov 24 20:06:56 laymen's plot analysis Nov 24 20:07:53 DocScrutinizer05: riiight, but as freemangordon said I'd prolly rather go through the audacity process :) Nov 24 20:08:20 oops, the two corner frequencies are 2000Hz (upper) and ~750Hz (lower) Nov 24 20:08:41 for the "/" section of the filter Nov 24 20:09:51 below 750Hz it applies a constant attenuation of around -6dB Nov 24 20:10:03 bencoh: input is 32bit, output is 16bit. IIRC Nov 24 20:10:18 with this particular input parameters like volume, coil temperature etc Nov 24 20:10:31 or input is stereo 16bit, output is mono Nov 24 20:10:39 left channel only Nov 24 20:16:15 it would be way more interesting to test burst impulse response over time or volume sweep over time, e.g. for an input consisting of a 5 seconds pink noise +6dB burst or a input sweep from -infinity to +9dB over a 20s Nov 24 20:16:28 and then all the way back during next 20s Nov 24 20:16:53 and have the output plotted as 3D-graph Nov 24 20:17:06 or as color map, or whatever Nov 24 20:17:26 X:time Y:freq Z:amplitude Nov 24 20:17:39 yeah, that should reveal how HFP is tuned with input volume and freq Nov 24 20:17:49 and temperature! Nov 24 20:17:50 :) Nov 24 20:17:53 yes Nov 24 20:18:14 and temperature... which is never close to the reality :D Nov 24 20:18:26 hotairstation to the rescue Nov 24 20:19:29 hmm Nov 24 20:19:51 KotCzarny: I dont really think we need to test that on a real device :p Nov 24 20:21:35 test what? Nov 24 20:21:50 we can't measure coil temp Nov 24 20:22:43 i wonder what would happen to the audio when fake bme would report 100C Nov 24 20:23:35 iirc I tested various temperatures (fake), don;t recall something spectacular Nov 24 20:24:48 KotCzarny: but I can provide the code I used to test xprot, if someone is interested Nov 24 20:25:54 ohmy, are people going to TEST an irrelevant parameter like ambient temperature now? Nov 24 20:26:31 I suggest to completely kick out the temperature parameter and replace it by a constant of 55°C Nov 24 20:53:30 hey Nov 24 20:53:44 is anyone working on the telepathy stuff? Nov 24 21:04:43 DocScrutinizer05: how much do you think shipping an N900 from Germany to South Africa would cost? rough estimate? Nov 24 21:17:57 rtcom-messaging-ui seems weird Nov 24 21:22:13 Sicelo: the german n900s don't have 4 arrow keys Nov 24 21:30:11 still an N900 :) Nov 24 21:32:31 https://www.dhl.de/en/paket/pakete-versenden/weltweit-versenden/paket.html Nov 24 21:35:56 oh god! so i'm in zone 4 .. that means EUR41.99? Nov 24 21:37:17 how can i get the source for telepathy-gabble? Nov 24 21:38:02 erlehmann: apt? what is weird with messaging ui btw? Nov 24 21:38:31 Sicelo it is written in html and js Nov 24 21:38:37 http://wstaw.org/m/2015/11/24/plasma-desktoprN3616.png Nov 24 21:39:44 Sicelo: ^^^ Nov 24 21:40:11 Sicelo how do i get the source with apt? Nov 24 21:40:17 > 504 Host wstaw.org lookup failed: Host not found Nov 24 21:41:02 o.O Nov 24 21:41:11 WFM Nov 24 21:42:51 thanks DocScrutinizer05. seems much more reasonable. the N900 is from someone who does not need it anymore .. he's willing to give it to me for free (sans shipping), if no one else comes up with a price Nov 24 21:43:05 i want to make rtcom-messaging-ui show images in xmpp messages. can someone help me where to start? Nov 24 21:43:14 or is there an xmpp client that can do images in xmpp messages? Nov 24 21:43:28 needless to say, i wish no one makes an offer, haha :D Nov 24 21:43:51 lel why Nov 24 21:44:02 then i get it Nov 24 21:44:37 i understand. what about message receipts? Nov 24 21:44:54 erlehmann: i think Pali fixed your MUC stuff? so he should know more .. and since he was able to fix it, i surmise that source is available. try `apt-get source rtcom....` Nov 24 21:45:11 ah Nov 24 21:47:32 for scripting languages like HTML (incl JS) it seems to me the program is the source Nov 24 21:49:15 well, almost Nov 24 21:50:18 DocScrutinizer05 the point is that i want to patch telepathy-gabble and have no idea where to start Nov 24 21:50:30 like where do i get the current source for maemo gabble? Nov 24 21:53:17 used this http://maemo.org/development/sources/ Nov 24 21:54:14 i was like „apt-get source telepathy-gabble“ Nov 24 21:59:03 you can ddl the tar.gz http://repository.maemo.org/pool/maemo5.0/free/t/telepathy-gabble/ Nov 24 22:00:40 http://maemo.org/packages/view/telepathy-gabble/ Nov 24 22:00:48 dang, too slow Nov 24 22:01:28 I suggest http://maemo.org/packages/package_instance/view/fremantle_ssu_pr1-2_armel/telepathy-gabble/0.8.11-0maemo2+0m5/ Nov 24 22:01:43 i see http://lists.freedesktop.org/archives/telepathy-commits/2012-December/018091.html Nov 24 22:03:50 oops nevermind Nov 24 22:05:14 ? Nov 24 22:07:23 shit! Nov 24 22:07:26 IroN900:~# ll /usr/lib/telepathy/telepathy-gabble Nov 24 22:07:28 -rwxr-xr-x 1 root root 680080 2015-03-21 19:35 /usr/lib/telepathy/telepathy-gabble Nov 24 22:07:29 IroN900:~# file /usr/lib/telepathy/telepathy-gabble Nov 24 22:07:31 /usr/lib/telepathy/telepathy-gabble: ELF 32-bit LSB executable, ARM, version 1 (SYSV), for GNU/Linux 2.6.14, dynamically linked (uses shared libs), stripped Nov 24 22:07:43 what is shit? Nov 24 22:08:12 this is not exactly a HTML file Nov 24 22:08:24 Hello all ! Nov 24 22:08:50 hi Nov 24 22:09:35 pfiou it s strange to come back here after years :p Nov 24 22:10:10 hi Khertan Nov 24 22:10:20 people are still using the n900 Nov 24 22:10:25 erlehmann: http://repository.maemo.org/pool/maemo5.0/free/t/telepathy-gabble/telepathy-gabble_0.8.13-0maemo3%2b0m5.tar.gz Nov 24 22:10:49 Khertan: indeed, long time no see :-) Nov 24 22:11:06 klopsi-u3, my n810 is still used ... Nov 24 22:11:10 DocScrutinizer05 i found it Nov 24 22:11:20 :nod: Nov 24 22:11:27 Khertan: mine got stolen Nov 24 22:11:50 ok i used as a bot with a display :p Nov 24 22:12:03 i use it Nov 24 22:12:25 the n9 is a dedicated secure bitcoin wallet Nov 24 22:12:50 erlehmann: don't miss the debian/patches/*.patch Nov 24 22:17:09 quilt crap Nov 24 22:17:54 DocScrutinizer05 the patch does not cleanly apply :/ Nov 24 22:18:26 sorry nfc. See https://wiki.debian.org/debian/patches maybe it helps Nov 24 22:19:23 DocScrutinizer05 i mean the patch from http://lists.freedesktop.org/archives/telepathy-commits/2012-December/018091.html does not cleanly apply to that version of telepathy-gabble. Nov 24 22:19:33 is it possible to just recompile current telepathy-gabble? does it use same interfaces? Nov 24 22:21:04 good question, maybe just try? Nov 24 22:26:06 I would get original source same version and make a patch to get maemo changes. After this make a patch from old original to new original source code. Look into both patches and you will get a idea what to do ;-) Nov 24 22:30:58 or just look into the changes in git between the maemo version and the recent one that the patch is meant for? Nov 24 22:32:08 probably tghe complete patchset from mainstream gabble_0.8.13 to recent should be applicable to maemo gabble_0.8.13 Nov 24 23:36:36 . Nov 24 23:36:58 hi salamisami Nov 24 23:37:13 why did nokia give-up with maemo? Nov 24 23:45:02 you asking me? Nov 24 23:48:04 yeah Nov 24 23:48:11 you're finnish you should know Nov 24 23:50:02 afaik meego-harmattan was the hybrid that evolved from nokia's maemo and intel's moblin being thrown together Nov 24 23:50:19 so maemo wasn't sorta given up on Nov 24 23:51:20 as for meego there's Elop and his burning platforms oration Nov 24 23:52:20 that was bad Nov 24 23:52:33 i think other forces shut down maemo Nov 24 23:52:58 nokia apparently got too big Nov 24 23:53:09 the communication inside house was not good Nov 24 23:53:31 theres a text about all that stuff Nov 24 23:53:53 hm Nov 24 23:53:55 http://muropaketti.com/artikkelit/the-story-of-nokia-meego Nov 24 23:54:02 illuminating read Nov 24 23:54:14 anyway no use crying about spilled milk Nov 24 23:54:20 n900 is as good as its ever gonna get Nov 24 23:54:50 sadly yes Nov 24 23:56:33 "Nokia intended to release a phone codenamed Lauta, with a QWERTY keyboard immediately after Lankku in the late 2011. The shell of the phone was made out of the same polycarbonate as the N9, the only difference being the slide-out QWERTY keyboard. Inside, Lauta uses the same OMAP 3630 SoC by TI as the N9." Nov 24 23:57:25 "It would have been the successor of the N900 awaited by many but it never saw the light of day as a finished product. Nokia usually gave the official model name for a device at the last minute before release and the official name for Lauta is not known, if it was ever even decided." Nov 24 23:58:09 that part kind of reminds me of obi wan talking to anakin in star wars III end Nov 24 23:58:41 :/ Nov 25 00:15:07 hi Nov 25 00:15:22 hi Nov 25 00:19:03 hi Nov 25 01:01:47 huh **** ENDING LOGGING AT Wed Nov 25 03:00:53 2015