**** BEGIN LOGGING AT Sat May 17 02:59:57 2008
May 17 03:01:09 <Navi> scratchbox
May 17 03:01:12 <Navi> er
May 17 03:01:31 <Navi> there's a vmware image you can use
May 17 03:02:16 <mysc> aha, thanks, will google.
May 17 03:17:33 <Navi> mysc, wait, what bug?
May 17 03:27:21 <mysc> Navi, the bug introduced by the debian developer
May 17 03:27:28 <Navi> That was openssl
May 17 03:27:45 <mysc> yes, openssh uses openssl's random number generator
May 17 03:28:23 <Navi> Oh
May 17 03:28:41 <Navi> well just compile from the beautiful untouched source
May 17 03:29:10 <mysc> i'll try :|
May 17 03:29:28 <Mikho> It's a rare thing to see beautiful source
May 17 03:29:31 <Navi> Debian maintainers do such nasty things to source
May 17 03:30:41 <mysc> they sure did in this case...
May 17 03:30:49 <Navi> They always do
May 17 03:30:58 <mysc> and this has trickled down to all xandros (eeepc) and ubuntu users
May 17 03:31:19 <mysc> the latter is a _huge_ userbase
May 17 03:32:04 <Gracana> Gah. vtwm does not allow me to declare functions, so I can't have functions that reference each other.
May 17 03:32:37 <Navi> heh
May 17 03:32:37 * Gracana had a bind keys and unbind keys function, and a key that would call bind keys. When bind keys would call, it would remap that key to unbind keys, and vice versa.
May 17 03:33:01 <Gracana> that was a convoluted sentence, but whatever
May 17 03:33:20 * Gracana install something sane like fluxbox
May 17 03:33:28 <Navi> go ahead :P
May 17 03:33:34 <Gracana> Or maybe openbox.
May 17 03:33:44 <Navi> go ahead :P
May 17 03:34:11 <Navi> I set up my debian build environment five hours ago and still haven't done what I wanted to
May 17 03:34:45 <mysc> Navi, you're setting up scratchbox too?
May 17 03:35:07 <Navi> Yes
May 17 03:35:29 <Navi> I took two naps in the four hours
May 17 03:35:39 <Gracana> Navi, make a waterfall spectrograph that can read from the microphone input
May 17 03:35:51 <Navi> No :P
May 17 03:35:56 <Gracana> It's what the people want!
May 17 03:36:00 <Navi> right
May 17 03:36:03 <Gracana> And by "people" I mean me.
May 17 03:36:21 <Navi> I wanted to compile xcompmgr and transset to make the matchbox-keyboard translucent
May 17 03:36:30 <Gracana> hah
May 17 03:36:42 <Navi> sure it'd take up more ram, but I'm only using 50MB at the moment so who cares?
May 17 03:37:26 <Gracana> matchbox panel has pseudo-transparency
May 17 03:37:35 <GeneralAntilles> The RAM police, Navi.
May 17 03:37:42 <Navi> Gracana, meh
May 17 03:38:02 <Gracana> I am using 9MB of RAM right now >.>
May 17 03:38:04 <Navi> I'd rather true transparency so I can maximize the matchbox-keyboard and see exactly what I'm typing underneath
May 17 03:38:11 <Gracana> right, that makes sense
May 17 03:38:46 <Navi> congratulations mister vtwm
May 17 03:38:53 <Gracana> Navi, perhaps you could just make a shaped window with holes in it.
May 17 03:38:59 <Navi> ha
May 17 03:39:14 <Gracana> That is a semi-serious suggestion.
May 17 03:39:47 <Gracana> I think it could be done nicely. Just display the keys as outlines or small dots with characters on them or something
May 17 03:39:47 <Navi> weird
May 17 03:39:58 <GeneralAntilles> Anybody else have trouble making mplayer die after finishing a video?
May 17 03:40:01 <Navi> matchbox keyboard does weird things with my openbox layout
May 17 03:40:09 <Gracana> Like what?
May 17 03:40:10 <GeneralAntilles> It always pops back up a few times before it dies for good.
May 17 03:40:23 <Navi> Gracana, it's resizing all of my maximized windows
May 17 03:40:27 <Gracana> I have never used mplayer, so I can't say
May 17 03:40:29 <Gracana> o.o Navi
May 17 03:40:59 <Navi> woohoo
May 17 03:41:03 <Navi> I have fbreader
May 17 03:43:20 <ricko73> evening
May 17 03:44:02 <ricko73> Finally used the gps on my n810 with Roadmap
May 17 03:44:14 <Navi> How'd you like it?
May 17 03:44:24 <ricko73> besides draining the battery and taking a while to get the initial sync, it worked nicely
May 17 03:45:09 <ricko73> I'd guess about 5 minutes ish to get the initial location, which doesn't make a whole lot of sense, since the gps console showed 4 satellites almost immediately
May 17 03:46:03 <Navi> That isn't bad
May 17 03:46:21 <ricko73> I finally have it mounted in my vehicle though so we'll see how much practical use it will be in the long run
May 17 03:46:35 <Navi> at least you're not one of the people that have to wait for hours to lock on
May 17 03:46:56 <ricko73> I've always liked the roadmap program because it uses actual vector data
May 17 03:47:03 <Navi> mhm
May 17 03:47:07 <ricko73> I know that doesn't help many people outside of the US
May 17 03:48:08 <ricko73> I used Gpsdrive back in the day (2001 ish) and remember running scripts to grab maps from mapquest or yahoo before going on a trip
May 17 03:55:00 <Navi> johnx, from now on, your name is jackx
May 17 03:57:12 <dookdook> i'm not sure if this is the proper forum for it, but i have a nokia n800 and am trying to compile a program under scratchbox.  its giving the following compilation error: 'unit_test.c:1: error: bad value (arm1136j-s) for -mtune= switch'
May 17 03:57:42 <dookdook> i could have sworn i got this compiled and working a month or so ago, and i've updated some things (scratchbox i believe) since then...so i'm not sure if thats important
May 17 03:57:45 <dookdook> any ideas?
May 17 03:58:10 <KotCzarny> dookdook: try -march instead of -mtune?
May 17 03:58:16 <KotCzarny> or mcpu
May 17 03:58:44 <dookdook> i have an -mcpu=arm1136jf-s line already...
May 17 03:58:58 <KotCzarny> then i think you don't need mtune
May 17 03:58:59 <KotCzarny> :)
May 17 03:59:03 <Gracana> jf-s or j-s?
May 17 03:59:13 <Gracana> The one you pasted is j-s..
May 17 03:59:51 <dookdook> jf-s for mcpu, and j-s for mtune
May 17 04:02:17 <Gracana> oh really :P
May 17 04:05:31 <dookdook> taking out mtune gives:  `-mcpu=' is deprecated. Use `-mtune=' or '-march=' instead.
May 17 04:05:31 <dookdook> idct_armv5te.c:1: error: bad value (arm1136jf-s) for -mtune= switch
May 17 04:05:47 <KotCzarny> then use only march
May 17 04:06:18 <dookdook> and then putting in -march=arm1136jf-s gives:   error: bad value (arm1136jf-s) for -march= switch
May 17 04:07:02 <KotCzarny> heh
May 17 04:07:08 <KotCzarny> let me check my flags
May 17 04:08:48 <KotCzarny> export CFLAGS="-Os -s -pipe -fomit-frame-pointer -mtune=arm1136jf-s -mfpu=vfp -mfloat-abi=softfp"
May 17 04:09:17 <KotCzarny> if it doesn't work check if you are compiling for the right target
May 17 04:09:22 <KotCzarny> ie. armel vs x86
May 17 04:09:23 <KotCzarny> :)
May 17 04:09:42 <dookdook> how do i check that?
May 17 04:09:49 <KotCzarny> what's your prompt?
May 17 04:09:58 <dookdook> [sbox-CHINOOK_X86: ~/proj/src] >
May 17 04:10:01 <dookdook> ach!
May 17 04:16:42 <dookdook> ok, thanks guys.  sorry bout the dumb questions...
May 17 04:17:09 <ricko73> there are no dumb questions
May 17 04:18:16 <Gracana> Hey, at least it was a simple fix.
May 17 04:18:52 <Gracana> Is the x86 target worth using? I've never touched it.
May 17 04:19:18 <Gracana> Is it just there for performance?
May 17 04:19:41 <Gracana> ie get everything going on x86 and then compile the final project on arm?
May 17 04:20:00 <KotCzarny> it's for testing in sdk
May 17 04:20:10 <KotCzarny> ie. you compile app and test it on your pc
May 17 04:20:34 <KotCzarny> in emulator
May 17 04:36:31 <AStorm> KotCzarny: why the float abi?
May 17 04:37:08 <AStorm> we do have VFP in N8x0
May 17 04:37:27 <KotCzarny> astorm: otherwise it's not compiled for vfp
May 17 04:37:39 <KotCzarny> at least that's what i've been told
May 17 04:37:49 <AStorm> softfpu is SOFT floating point, dummy
May 17 04:37:58 <AStorm> this way you compile for *no* fpu
May 17 04:38:00 <AStorm> :-)
May 17 04:38:17 <KotCzarny> astorm: i know the logic, but
May 17 04:38:26 <AStorm> if you don't give an option, with jf-s you build for vfp
May 17 04:38:30 <KotCzarny> that's what i've been told by local gurus
May 17 04:38:33 <AStorm> (checked, target says that)
May 17 04:38:43 <AStorm> with j-s you build for soft
May 17 04:38:44 <KotCzarny> btw. jf-s is vfp one
May 17 04:38:47 <AStorm> yes
May 17 04:38:58 <KotCzarny> but you need that abi switch
May 17 04:39:03 <AStorm> actually, not
May 17 04:39:04 <KotCzarny> and don't argue with me
May 17 04:39:09 <KotCzarny> ask ssvb, kulve
May 17 04:39:10 <KotCzarny> etc
May 17 04:39:18 <AStorm> someone sent a lot of FUD, really... check what binutils say
May 17 04:39:29 <AStorm> they're *not* wrong about targets
May 17 04:39:42 <AStorm> the kernel *can* use VFP in its softfp, yes, but that's irrelevant
May 17 04:39:58 <AStorm> gcc can use it better
May 17 04:40:16 <KotCzarny> astorm: have you done any benchmarks?
May 17 04:40:18 <KotCzarny> :)
May 17 04:40:25 <KotCzarny> that could convince me
May 17 04:40:28 <AStorm> I've done checks with binutils.
May 17 04:40:32 <KotCzarny> otherwise you are doing fud too
May 17 04:40:33 <KotCzarny> :)
May 17 04:40:39 <AStorm> Benchmarks may be wrong if VFP is worse than softfpu
May 17 04:40:41 <AStorm> ;P
May 17 04:40:59 <KotCzarny> benchmarks are benchmarks
May 17 04:41:02 <AStorm> (it may be possible on some code, although unexpected)
May 17 04:41:15 <KotCzarny> if softfpu is better, then there's no point in vfp
May 17 04:41:17 <KotCzarny> right? right
May 17 04:41:30 <AStorm> no no, it only means gcc is futzed
May 17 04:41:53 <AStorm> as the kernel may use VFP for softfloat anyway
May 17 04:42:07 <KotCzarny> give me numbers please
May 17 04:42:08 <KotCzarny> :)
May 17 04:42:20 <AStorm> 1) objdump
May 17 04:42:26 <AStorm> happy with the number?
May 17 04:42:43 <KotCzarny> you know what i mean
May 17 04:43:13 <AStorm> Sure, I'll get them. (from lame MP3 encoder, preferably)
May 17 04:43:33 <AStorm> I don't trust no stinking benchmark. Real code is a better test.
May 17 04:44:14 <KotCzarny> :)
May 17 04:44:26 <KotCzarny> benchmark == timed real code
May 17 04:44:35 <AStorm> that I can do ;P
May 17 05:00:33 <lopz> night
May 17 05:01:06 <mschoolbus> nighty
May 17 05:16:19 <Navi> Gracana, it's never worth it to build the x86 target :P
May 17 05:17:09 <dmsuperman_> !log
May 17 05:17:11 <dmsuperman_> !logged
May 17 05:17:17 <dmsuperman_> is this channel logged anywhere/
May 17 05:17:23 <AStorm>  /topic
May 17 05:17:57 <AStorm> dmsuperman_: I meant, it's in the topic
May 17 05:18:17 <dmsuperman_> yeah I found it :D
May 17 05:18:59 <dmsuperman_> Alright, the .deb that penguinbait created to partition and clone your OS to an MMC uses a boot menu. Does anybody know how to change the time it waits before auto selecting an item?
May 17 05:19:23 <GeneralAntilles> Download fanoush's initfs flasher
May 17 05:19:30 <GeneralAntilles> bootmenu.conf.n8x0.example
May 17 05:19:35 <GeneralAntilles> Fix it up how you want it
May 17 05:19:43 <AStorm> dmsuperman_: for that, you have to mod bootloader.conf or the script itself
May 17 05:19:44 <GeneralAntilles> rename it bootmenu.conf
May 17 05:19:50 <AStorm> oh, bootmenu.conf, yes
May 17 05:19:53 <GeneralAntilles> Flash it
May 17 05:20:01 <GeneralAntilles> See the README for more details.
May 17 05:20:19 <dmsuperman_> And this won't interfere with my current setup or anything?
May 17 05:20:25 <GeneralAntilles> No
May 17 05:20:30 <dmsuperman_> Alright
May 17 05:20:32 <dmsuperman_> Thanks
May 17 05:20:32 <infobot> gern geschehen, dmsuperman_
May 17 05:20:42 <dmsuperman_> ...?
May 17 05:20:56 <GeneralAntilles> infobot luvs u
May 17 05:21:06 <Navi> Anyone try out iNES?
May 17 05:21:15 <GeneralAntilles> RST38h I'm sure. :P
May 17 05:23:23 <dmsuperman_> GeneralAntilles, The warning says don't run it from SD card, but that's where my OS is. Do I have to reboot into the flash one?
May 17 05:23:53 <GeneralAntilles> No, I don't believe so.
May 17 05:24:03 <dmsuperman_> alright
May 17 05:24:04 <GeneralAntilles> My guess is that he means don't run it from a FAT partition
May 17 05:24:09 <GeneralAntilles> Because it wont run from there. :P
May 17 05:26:19 <penguinba> whats up
May 17 05:27:51 <penguinbait> dmsuperman what are you installing
May 17 05:29:23 <Navi> Lol
May 17 05:30:51 <penguinbait> ?? seems there was some confusion?
May 17 05:31:33 <dmsuperman_> penguinbait, I was just going to try to edit my boot menu's waiting time
May 17 05:31:38 <dmsuperman_> I used your .dev earlier today
May 17 05:31:42 <dmsuperman_> .deb
May 17 05:31:48 <dmsuperman_> I think I was on as cjohnson then
May 17 05:31:58 <penguinbait> did you uninstall the deb?
May 17 05:32:05 <dmsuperman_> no, it all works fine
May 17 05:32:15 <dmsuperman_> I just wanted to change the boot time from 30s to something shorter
May 17 05:32:26 <penguinbait> if you have not yet removed the deb
May 17 05:32:41 <penguinbait> you can update bootmenu
May 17 05:32:48 <dmsuperman_> howso?
May 17 05:32:58 <penguinbait> from /root/install_tools
May 17 05:33:11 <penguinbait> edit bootmenu.conf
May 17 05:33:31 <dmsuperman_> is that all I have to do?
May 17 05:33:40 <penguinbait> run initfs_flash
May 17 05:33:52 <penguinbait> follw promts
May 17 05:34:00 <penguinbait> you will reboot
May 17 05:34:08 <penguinbait> close all apps
May 17 05:35:13 <dmsuperman_> no prompts, no reboot
May 17 05:35:19 <dmsuperman_> but no errors, looks like it went well
May 17 05:35:37 <penguinbait> oh yeah,thats my doing
May 17 05:35:56 <penguinbait> if I could only remember what I do?
May 17 05:36:02 <penguinbait> so just reboot
May 17 05:36:10 <dmsuperman_> haha
May 17 05:36:18 <dmsuperman_> what is this usb network recovery mode?
May 17 05:36:46 <penguinbait> I never tried it, It has a telnet server
May 17 05:36:51 <penguinbait> ???
May 17 05:36:55 <AStorm> or ssh dropbear server
May 17 05:36:56 <AStorm> or both
May 17 05:37:22 <dmsuperman_> what does it have for a username and password?
May 17 05:37:26 <penguinbait> if you want ssh you have to download fanoushes original package
May 17 05:37:37 <AStorm> dmsuperman_: root w/o pass afaicr
May 17 05:37:49 <dmsuperman_> ah
May 17 05:38:15 <penguinbait> dont run from flash, and you really dont need it
May 17 05:38:49 <dmsuperman_> you mean ssh from the recovery mode?
May 17 05:38:56 <dmsuperman_> or from the live maemo
May 17 05:39:29 <AStorm> recovery mode
May 17 05:39:39 <AStorm> maemo one has rootme as the default pass
May 17 05:39:40 <dmsuperman_> oh
May 17 05:39:58 <dmsuperman_> yeh, installing openssh allows you to overwrite it when it's installed
May 17 05:39:59 <penguinbait> I mean if my main os in os boot from sd, if flash becomes corrupt, I just reflash, install bootmenu again and start up my already configured os
May 17 05:40:24 <GeneralAntilles> Why install again?
May 17 05:41:04 <GeneralAntilles> Just --flash-only=rootfs
May 17 05:41:10 <penguinbait> I need it to fsck the boot fro sd after crashes
May 17 05:41:33 <penguinbait> right, that too
May 17 05:41:51 <penguinbait> again I am confused
May 17 05:41:56 <Navi> heh
May 17 05:42:33 <penguinbait> its almost 2am her and I am crashing
May 17 05:42:53 <penguinbait> I am not in college like GA
May 17 05:43:03 <KotCzarny> are you old?
May 17 05:43:08 <dmsuperman_> is it normal that mmcblk0p2 is mounted as both /mnt/initfs and /mnt/initfs/tmp ?
May 17 05:43:23 <penguinbait> 36
May 17 05:43:26 <penguinbait> yes
May 17 05:43:28 <KotCzarny> :)
May 17 05:43:32 <AStorm> dmsuperman_: uhm, the .../tmp should be tmpfs, but yes, it's ok
May 17 05:44:07 <penguinbait> I have 4 kids and feel older
May 17 05:44:20 <dmsuperman_> oh wait I misread that. /dev/mmcblk0p2 is mounted both to / and /mnt/initfs, and "none" is mounted to /mnt/initfs/tmp
May 17 05:44:20 <penguinbait> does that count for anything?
May 17 05:44:24 <KotCzarny> :)
May 17 05:44:35 <KotCzarny> pb: sounds like fun
May 17 05:44:48 <dmsuperman_> penguinbait, how old?
May 17 05:44:59 <penguinbait> 36?
May 17 05:45:04 <dmsuperman_> 4 kids
May 17 05:45:13 <penguinbait> oldes is 15
May 17 05:45:24 <penguinbait> youngest is 9
May 17 05:45:28 <dmsuperman_> wow
May 17 05:45:36 <dmsuperman_> i bet that's a bundle of fun
May 17 05:45:36 <Navi> So you had your first fairly young, and in quick succession
May 17 05:46:22 <penguinbait> it took me a while to figure out what was causing it
May 17 05:46:42 <dmsuperman_> it was mayonaise i bet
May 17 05:46:42 <penguinbait> then we got a mail-lady
May 17 05:46:46 <penguinbait> hehe
May 17 05:47:57 <penguinbait> alright I am out, sleep calls!
May 17 05:48:03 <dmsuperman_> me too soon
May 17 05:55:36 <dmsuperman_> I didn't mount it, but is there a problem with sftp into the n800 from a linux desktop?
May 17 05:57:57 <GeneralAntilles> No
May 17 05:58:28 <dmsuperman_> Awesome
May 17 05:58:35 <dmsuperman_> Then I found the perfect way to transfer in
May 17 05:58:38 <dmsuperman_> I didn't even mean to
May 17 05:58:49 <AStorm> there's fuse sshfs too
May 17 05:58:50 <dmsuperman_> I definitely didn't install an FTP server on the n800
May 17 05:58:57 <dmsuperman_> unless openssh includes ftp?
May 17 05:58:58 <AStorm> which is made of win
May 17 05:59:19 <GeneralAntilles> sftp is openssh
May 17 06:00:41 <dmsuperman_> ah
May 17 06:46:32 <RST38h> back
May 17 06:51:49 <molkko> how one gets commandline in n810. in my n800 it was "xterm". searching xterm on maemo.org gives nothing??
May 17 06:54:28 <GeneralAntilles> It's included by default.
May 17 07:01:12 <molkko> what it is called
May 17 07:01:34 <molkko> ok. found it
May 17 07:02:37 <molkko> i did not notice that I must scroll down. I though all the items under utilities are visible immediately
May 17 07:16:50 <Blastur> Hey guys. Quick question. My n810 keeps pulsing the blue light whenever it sits idle, no apps running, display is off and in offline mode. What is it trying to tell me?
May 17 07:17:18 <Blastur> Kinda annoying if you wanna sleep and it keeps pulsing
May 17 07:17:20 <Sargun> its telling you its bored, and sleeping.
May 17 07:17:27 <Sargun> hehe
May 17 07:17:32 <GeneralAntilles> Is it plugged in?
May 17 07:17:36 <Sargun> just put something over your N810, or put it to sleep.
May 17 07:17:38 <GeneralAntilles> do you have anything in the notification area?
May 17 07:17:49 <GeneralAntilles> See Control Panel -> Display -> LED Tab
May 17 07:17:54 <Blastur> Not plugged in
May 17 07:18:12 * RST38h ported SlideRule to Maemo but can't figure out the best way to do controls
May 17 07:18:26 <RST38h> Damn thing does not even have a dialpad =(
May 17 07:19:11 <Blastur> Ah "device on" notifier maybe the cause?
May 17 07:19:37 <GeneralAntilles> Likely
May 17 07:19:43 <Blastur> Thanks
May 17 07:19:43 <infobot> sure thing, Blastur
May 17 07:19:47 <Blastur> Cya
May 17 07:19:48 <GeneralAntilles> I don't recall if it just shows that when it's on and screen off or when it's actually charging.
May 17 07:36:59 <Navi> whee
May 17 07:37:58 * qwerty12 just realised i'm not as crap as I thought with programming
May 17 07:38:04 <Navi> woo
May 17 07:38:28 <qwerty12> hehe, anything good happen :)
May 17 07:38:33 <Navi> ?
May 17 07:38:50 <RST38h> qwerty: You appletized Conky?!? =)
May 17 07:39:19 <qwerty12> RST38h, Nay. Because I don't understand, I'm gonna try and take a clunkier approach :/
May 17 07:39:40 <Navi> It'd be nice if we could just get some pipe applets
May 17 07:39:58 <GeneralAntilles> ssh + imagemagick + http server + web shortcut applet. :D
May 17 07:40:07 <Navi> :|
May 17 07:40:20 <qwerty12> I'm gonna try and make conky draggable and make another applet that doesn't display anything but will load conky when ran and close it when the applet is disabled.
May 17 07:41:05 <qwerty12> ATM though, I just need to finish off the keybindings on this game I got to run :)
May 17 07:48:37 <qwerty12> Just a quick q, do SDL_BUTTON's refer to the mouse buttons?
May 17 07:51:25 <trickie> qwerty12: well you can test for a mouse button click using that macro IIRC
May 17 07:51:43 <qwerty12> Ok, thanks, it answers my question :)
May 17 07:53:07 <qwerty12> Sigh. I love my N800. It keeps restarting for fun. No hacks done to it. It's probably checking the jffs2 blocks :/
May 17 07:53:40 <trickie> qwerty12: are you grabbing coredumps ?
May 17 07:54:03 <qwerty12> Yes I am come think of it.
May 17 07:54:08 <KotCzarny> 2coredumps1qwerty ?
May 17 07:54:09 <KotCzarny> :)
May 17 07:54:19 <qwerty12> hehe
May 17 07:56:07 <qwerty12> Core dumps are actually enabled at initfs time so I could take a look through them. thanks :)
May 17 07:57:06 <qwerty12> Sigh. It was my N800 realising too late it was out of battery power :/
May 17 07:57:54 <Navi> lol
May 17 07:58:45 <qwerty12> Most N800's (mine at the beginning) would shut off before it starts booting from rootfs. Mine decides to boot. :/
May 17 07:59:03 <qwerty12> And then gets shut of later leading me to think the worst.
May 17 08:06:40 <Navi> bleh
May 17 08:32:27 <rm_you> johnx: yo?
May 17 08:32:33 <johnx> yo yo
May 17 08:32:39 <qwerty12> wag1
May 17 08:32:44 <rm_you> :P
May 17 08:33:01 <rm_you> johnx: did you say a while ago you might be interested in going in together for hosting?
May 17 09:08:03 <Navi> johnx, hi
May 17 09:27:51 <Cptnodegard> are there any emulators whatsoever that run full speed on the ITs?
May 17 09:28:36 <Navi> iNES supposedly does
May 17 09:28:36 <glass> i'm pretty sure theres some nes and such
May 17 09:28:40 <GAN800> Yes.
May 17 09:28:40 <qwerty12> Can SCUMMVM be classified as a emulator?
May 17 09:28:48 <glass> scummvm while not an emulator still runs at fullspeed
May 17 09:28:56 <qwerty12> Ah, i see.
May 17 09:29:07 <GAN800> dosbox runs t
May 17 09:29:09 <glass> qwerty12: game engine is better word i think?
May 17 09:29:10 <qwerty12> RST38h's emulators are brilliant
May 17 09:29:16 <GAN800> at fullspeed
May 17 09:29:16 <Navi> scummvm runs a bit slow depending on the game
May 17 09:29:18 <qwerty12> glass, yes :)
May 17 09:31:45 <Cptnodegard> :/ nothing useful iow
May 17 09:32:02 <GeneralAntilles> What do you want, PS1? :P
May 17 09:32:47 <RST38h> qwerty: you are welcome :)
May 17 09:32:58 <qwerty12> :)
May 17 09:33:00 <glass> i find scummvm plenty useful
May 17 09:33:07 <RST38h> Cptn: All of my stuff except for VGBA should run at full speed
May 17 09:33:28 <RST38h> VGBA will run at 60-75% of GBA speed on ITs
May 17 09:33:37 <Cptnodegard> RST38h: any snes?
May 17 09:33:46 <RST38h> I haven't done SNES, sorry
May 17 09:34:02 <Cptnodegard> what then?
May 17 09:34:16 <RST38h> Cptn: http://fms.komkon.org/
May 17 09:34:24 <qwerty12> RST38h, any chance of ever implementing GBA bios option? GPSP ran brilliantly on my PSP with GBA bios.
May 17 09:34:29 <Cptnodegard> thx :)
May 17 09:34:45 <Navi> johnx, why is the display variable set to :10.0 by default?
May 17 09:34:58 <RST38h> qwerty: GBA BIOS is emulated, but if you have a real BIOS file in /usr/share/vgba. VGBA will use the real BIOS
May 17 09:35:09 <qwerty12> RST38h, brilliant, thanks :)
May 17 09:35:12 <RST38h> Real BIOS is slower than emulated though
May 17 09:35:26 <qwerty12> Ahh :/
May 17 09:35:57 <RST38h> So, unless you have a game that ABSOLUTELY wants it or you really like the GBA opening animation, don't bother
May 17 09:36:27 <qwerty12> Ah, ok then. No worries, at least I can play GBA games at acceptable speed :)
May 17 09:38:29 <lardman> bmidgley: ping
May 17 09:49:39 <lardman> Anyone know why the SBC code assumes the input stream will be in biendian fomat (i.e. Sun .snd)?
May 17 09:49:48 <lardman> s/biendian/bigendian
May 17 09:52:37 <RST38h> biendian would be nice...
May 17 09:52:44 <lardman> why though?
May 17 09:52:51 <RST38h> works both ends
May 17 09:53:13 <lardman> oh :)
May 17 09:55:44 <lardman> I just thought it was a bit odd that the sbcenc expects to see bigendian input data, as presumably the data is usually receives is little endian
May 17 10:03:21 <lardman> ah, it all becomes clear
May 17 10:03:54 <lardman> I was only looking at the test code there, the wrapper around the sbc functionality, and they'd chosen to use SUN .snd
May 17 10:07:06 <thopiekar> hi @ all
May 17 10:07:17 <thopiekar> qwerty are you there?
May 17 10:07:19 <thopiekar> ^^
May 17 10:07:23 <qwerty12> Yeah, give me a min.
May 17 10:07:49 <thopiekar> ok^^
May 17 10:09:22 * qwerty12 is waiting for standard kernel to compile first...
May 17 10:11:18 <Blastur> Whats the esiest way to send files to someone over wifi on n810? Can only find scp cmdline util but its too much work. Is there a ftp gui or such?
May 17 10:11:43 <GeneralAntilles> From or to the N810?
May 17 10:11:52 <Blastur> From
May 17 10:11:56 <GeneralAntilles> maemoftp
May 17 10:12:23 <GeneralAntilles> Maybe Bluetooth for small stuff.
May 17 10:12:24 <Blastur> Ok hmm. Searched for ftp, no hits in app mgr
May 17 10:12:29 <Blastur> I look again
May 17 10:12:32 <GeneralAntilles> It's not in the repos
May 17 10:12:33 <GeneralAntilles> google
May 17 10:12:34 <qwerty12> .Check on ITT
May 17 10:12:41 <qwerty12> Or google like GeneralAntilles says
May 17 10:12:47 <GeneralAntilles> http://www.student.oulu.fi/~jualasal/gb/maemoftp.html
May 17 10:14:21 <thopiekar> < is away for around 5mins..
May 17 10:14:57 <qwerty12> Just as soon as kernel built :p. I'll try the module next now.
May 17 10:15:38 <GeneralAntilles> He only charges $49.99 a download.
May 17 10:17:24 <qwerty12> thopiekar, when you get back, I have asix.ko waiting for you ;)
May 17 10:17:35 <qwerty12> Compiled against standard nokia kernel.
May 17 10:18:31 <thopiekar> I'm back..
May 17 10:18:37 <thopiekar> ohh nice^^
May 17 10:18:48 <thopiekar> can you upload it to the iTT
May 17 10:18:53 <qwerty12> Ok, will do :)
May 17 10:18:58 <thopiekar> into the thread
May 17 10:18:59 <thopiekar> thxx
May 17 10:19:01 <thopiekar> ^^
May 17 10:19:37 <thopiekar> yeah this community roxxx
May 17 10:19:41 <thopiekar> ^^
May 17 10:20:31 <qwerty12> Np, I've posted it :)
May 17 10:22:10 <thopiekar> what do you mean with:
May 17 10:22:11 <thopiekar> Remove the .zip extension.
May 17 10:22:15 <thopiekar> rename it??
May 17 10:22:20 <qwerty12> Yeah
May 17 10:22:26 <Blastur> Thanks guys it woeks now
May 17 10:22:26 <infobot> Blastur: bitte
May 17 10:22:29 <thopiekar> ok thxx again
May 17 10:22:51 <thopiekar> "bitte"??
May 17 10:23:02 <thopiekar> thats german, isn*'t it
May 17 10:23:04 <thopiekar> ^^
May 17 10:24:47 <GAN800> Somebody kill infobot
May 17 10:25:02 <qwerty12> infobot, die.
May 17 10:25:03 * infobot takes two shots to the head and crumples to the ground, lifeless.
May 17 10:25:06 <qwerty12> infobot, die.
May 17 10:29:19 <term-sig> guys anybody from uk?
May 17 10:29:41 <qwerty12> I am :/
May 17 10:30:21 <term-sig> qwerty12: u know something about SC (Security clearance)?
May 17 10:30:51 <qwerty12> erm, no sorry :(
May 17 10:31:16 <GAN800> Nobody is from the UK
May 17 10:31:20 <GAN800> The UK doesn't exist
May 17 10:31:27 <GAN800> It's all a lie.
May 17 10:31:32 <qwerty12> http://www.certes.co.uk/security-check-clearance-jobs/ & http://en.wikipedia.org/wiki/Classified_information_in_the_United_Kingdom#Security_Check_.28SC.29
May 17 10:31:35 <qwerty12> All I can find.
May 17 10:31:52 <GAN800> Stop spreading lies, qwerty12
May 17 10:32:12 <qwerty12> I can't help it. I think I will wave my Union Jack too.
May 17 10:32:32 <term-sig> qwerty12: hehe i found this same but no info how pass that cert..
May 17 10:33:04 <qwerty12> Hehe, I was born here and I've never applied for job yet so I don't really know sorry :(
May 17 10:33:09 <qwerty12> Maybe lcuk will
May 17 10:33:32 <GAN800> a LAZY liar. . . .
May 17 10:33:45 <qwerty12> I'm still in school yet :P
May 17 10:33:57 <term-sig> dont be so lazy qwerty12 ;P
May 17 10:34:01 <GAN800> 14 is the appropriate age to start working. :P
May 17 10:34:04 <qwerty12> I don't have any qualifications yet to apply for a job with yet
May 17 10:34:20 <qwerty12> Meh, can't be arsed. I live of bank of Mom and Dad.
May 17 10:34:25 <GAN800> That's why you go bag groceries.
May 17 10:34:43 * GAN800 worked 40 hours a week his senior year. :P
May 17 10:34:53 <lardman> Go write computer systems for your Dad
May 17 10:35:06 <lardman> or someone anyway :)
May 17 10:35:42 <qwerty12> hehe. My Dad don't know nothing about computers except www.ebay.co.uk :p
May 17 10:36:07 <lardman> :)
May 17 10:37:41 * qwerty12 is happy. I got OpenBOR compiled. I remember playing on Dreamcast. Now to see if it runs...
May 17 10:38:21 * lardman is also happy not being at work today
May 17 10:38:34 <lardman> I say stick with the holidays :)
May 17 10:38:55 <GAN800> Party party party!
May 17 10:39:50 <thopiekar> im gonna leave
May 17 10:39:53 <thopiekar> cu
May 17 10:39:59 <qwerty12> bye
May 17 10:44:02 <pupnik> doot doot
May 17 10:47:20 <lardman> pupnik: how's it going?
May 17 10:48:25 <pupnik> purdy good.
May 17 10:48:52 <pupnik> someone did an OPL pass-through driver for dosbox
May 17 10:48:56 <pupnik> checking that out today
May 17 10:49:09 * lardman saw OPL and thought of Psion for a moment :)
May 17 10:49:19 <glass> lardman: hah me too..
May 17 10:49:32 <lardman> I was quite excited until it sunk in
May 17 10:49:34 <glass> lardman: shame that it wasnt ported to symbian 9
May 17 10:49:45 <lardman> glass: the source is available
May 17 10:50:09 <glass> i guess most people intrested in it would have gone for python by now
May 17 10:50:46 <lardman> yep, that's my conclusion; means I need a book by my side now though; I knew that programming manual inside out :)
May 17 10:51:12 <pupnik> anybody want a dismantled netbook?
May 17 10:51:24 <pupnik> otherwise it's goin up on ebay.  motherboard power connector broke off
May 17 10:51:27 <lardman> :)
May 17 10:51:35 <lardman> too old now really
May 17 10:51:40 <pupnik> yeah
May 17 10:51:41 <GAN800> Solder it back on.
May 17 10:52:01 <lardman> I still like the Psion series 5 form factor though
May 17 10:52:02 <pupnik> i lost the SMT resistor it needs.  if i could figure out what to put there, i'd be ok
May 17 10:52:20 <GAN800> Use it to build your doomsday device.
May 17 10:52:56 <pupnik> something nice about a umpc with leather on the case..
May 17 10:53:17 <pupnik> yeah the 5mx was nifty... keyboard just not quite big enough for my hamfingers
May 17 10:53:41 <lardman> bigger than on n810 though
May 17 10:54:03 <lardman> screen was bigger too, physically, good for the eyes :)
May 17 10:56:55 * aquatix has a 5mx laying around
May 17 10:57:09 <aquatix> nice device
May 17 10:57:14 <aquatix> don't use it though
May 17 10:57:18 <aquatix> needs wifi :P
May 17 10:57:30 <lardman> nor do I anymore, except for the crossword/scrabble cheating app
May 17 10:57:37 <aquatix> :)
May 17 10:57:59 <aquatix> previous owner used it among other things for taking notes at meetings
May 17 10:58:10 <aquatix> for writing up the summary
May 17 10:58:22 * aquatix finds the keyboard a bit cramped though
May 17 10:59:15 <aquatix> just slightly too small, otherwise it'd be awesome for typing
May 17 10:59:25 <lardman> it's not too bad for quick bursts of typing
May 17 10:59:29 <aquatix> indeed
May 17 10:59:31 <lardman> thumb typing that is
May 17 10:59:42 <GAN800> All of our problems would be solved if we only had a 10 letter alphabet
May 17 10:59:49 <aquatix> better than the n810 i guess, but the n810 holds better in your hands while typing i think
May 17 10:59:54 <aquatix> GAN800: :)
May 17 11:00:04 <lardman> GAN800: more room for symbols in the standard ascii set
May 17 11:00:13 <pupnik> it's 2008 and our computers are still too stupid to be voice driven
May 17 11:00:35 <GeneralAntilles> Voice is inefficient and irritating.
May 17 11:00:36 <zap> GAN800: The 2 digit alphabet is at your service
May 17 11:00:39 <qwerty12> Wait until 3008
May 17 11:01:03 * aquatix doesn't want voice driven irc in the train ;)
May 17 11:01:56 <aquatix> time for some smart-laid-out small keyboard around my left wrist
May 17 11:02:08 <aquatix> then i can type with my right hand while holding the tablet with left
May 17 11:02:10 <pupnik> here's an invention, two wristbands that can detect fingermovement through the tendons, type on any surface (knees) with chording
May 17 11:02:16 <RST38h> Why do you need to talk to computer at all?
May 17 11:02:26 * aquatix plugs usb in ear
May 17 11:02:35 <aquatix> almost fits
May 17 11:03:04 <aquatix> pupnik: yeah, but how do you know which `keys' you are hitting?
May 17 11:03:12 <lardman> aquatix will take a long time to charge with only 100mA
May 17 11:03:13 <molkko> searching VNC in maemo.org downloads section. does it mean vnc client nor server is available are there other corresponding SW for n810 or is it possible to x server in ubuntu laptop
May 17 11:03:29 <aquatix> lardman: :P
May 17 11:03:32 <pupnik> chording - i.e. not querty but finger contraction combinations
May 17 11:03:33 <lardman> :)
May 17 11:03:37 <aquatix> maybe i'll be the powersource ;)
May 17 11:03:41 <aquatix> pupnik: ah indeed
May 17 11:03:45 <aquatix> read about that
May 17 11:03:54 <aquatix> intersting technique
May 17 11:04:02 <GeneralAntilles> http://vncviewer.garage.maemo.org/ http://mike.saunby.googlepages.com/x11vncfornokia7702
May 17 11:04:29 <pupnik> yeah with a nice display in glasses, and some form of chording, you could be all 'wearable' mobile
May 17 11:04:44 <GeneralAntilles> What did they call those guys in Snow Crash?
May 17 11:04:52 <aquatix> erm
May 17 11:05:01 <aquatix> dammit, almost remember :/
May 17 11:05:33 <RST38h> In Hyperion, they cold those "federal employees"
May 17 11:05:38 <RST38h> s/cold/called/
May 17 11:05:38 <infobot> RST38h meant: In Hyperion, they called those "federal employees"
May 17 11:05:57 <aquatix> RST38h: heh, indeed
May 17 11:06:01 <aquatix> and that hacker guy
May 17 11:06:13 <RST38h> Willing to work 16 hours a day for meager salary and surplus electronic equipment
May 17 11:06:24 <aquatix> which died in the prcoess of retrieving the AI
May 17 11:07:51 <RST38h> Another cute idea was an internet-based parlament, de facto powerless and dominated by demagogues
May 17 11:08:17 <aquatix> GeneralAntilles: meh, can't remember how they called people plugged in to the metaverse :/
May 17 11:08:29 <GeneralAntilles> The interwebz are failing me, too.
May 17 11:08:31 <GeneralAntilles> Oh well
May 17 11:08:34 <GeneralAntilles> Somebody'll think of it.
May 17 11:09:14 <pupnik> i liked 'fragments of a hologram rose'
May 17 11:12:28 * aquatix is currently reading Asimov
May 17 11:12:32 <aquatix> Foundation series
May 17 11:12:55 <pupnik> great stuff.  he incorporated a couple of other novels into the foundation 'universe' as well
May 17 11:13:03 <aquatix> yeah
May 17 11:13:16 <pupnik> though 'foundation's edge' was a bit of a letdown imo - maybe it was just me
May 17 11:13:20 <aquatix> i read caves of steel a while ago
May 17 11:13:39 <aquatix> currently reading the second of the 7-book core
May 17 11:14:45 <aquatix> [Forward the Foundation]
May 17 11:15:52 <GeneralAntilles> The first 4 of Foundation are really exceptional
May 17 11:15:57 <GeneralAntilles> Then it just goes in the crapper.
May 17 11:16:49 <aquatix> GeneralAntilles: Prelude till Foundation and Empire?
May 17 11:17:14 <GeneralAntilles> Something like that
May 17 11:17:17 <GeneralAntilles> maybe first 5
May 17 11:17:18 <GeneralAntilles> been a while
May 17 11:17:31 <aquatix> Second Foundation :)
May 17 11:17:34 <aquatix> http://en.wikipedia.org/wiki/Foundation_series
May 17 11:17:46 <aquatix> http://en.wikipedia.org/wiki/Foundation_series#List_of_books
May 17 11:22:33 * RST38h didn't like Foundation very much, although Azimov's popular chemistry books are nice
May 17 11:23:22 <GAN800> You sure you read the right ones? <_<
May 17 11:23:32 <RST38h> Pretty sure
May 17 11:23:49 <RST38h> Unless there is a second set for specifically enlightened readers
May 17 11:24:35 <lopz> morning
May 17 12:00:21 <eichi> maemo mapper often su**s the map database file...someone knows this?
May 17 12:24:24 <RST38h> eichi: Translate "sucks the map database file"
May 17 12:25:03 <eichi> maemo mapper doesn start anymore. it crashs if the second cam, the map will shown
May 17 12:25:33 <eichi> it i eject the sd card, with the map database on it, it starts with black background
May 17 12:25:42 <eichi> then it works
May 17 12:25:53 <eichi> but the database is now corrupt
May 17 12:26:30 <RST38h> Ok, this one is relatively easy to fix
May 17 12:26:59 <RST38h> Run xterm, go to where database is, delete the database.
May 17 12:27:18 <RST38h> You can always reload it off the Net, so other than some lost time it will not harm you
May 17 12:53:56 <practisevoodoo> is there any way to configure the menus on an n810, or anything else like personal menu
May 17 13:01:20 <pupnik> practisevoodoo: you can move menu items around
May 17 13:01:52 <practisevoodoo> yeah but you got to scroll around all over the place
May 17 13:02:01 <qwerty12_N800> pipeline's linkmaker is a brilliant way to add stuff too.
May 17 13:02:18 <qwerty12_N800> then use personal menu like you said
May 17 13:03:12 <practisevoodoo> ok cheers i might muck around with that then
May 17 13:03:19 <eichi> RST38h, thats no fix ;D
May 17 13:03:58 <eichi> i redownloaded the maps now several times...its no fun anymore
May 17 13:04:09 <qwerty12_N800> eichi, see if  card isn't getting i/o errors
May 17 13:04:41 <JackCrow> Good morning
May 17 13:04:49 <qwerty12_N800> afternoon
May 17 13:08:30 <JackCrow> does anyone know how to turn off the "type-ahead" bar when using a keyboard?  Some keypress brings it up, but I can never get rid of it w/o rebooting
May 17 13:09:25 <qwerty12_N800> i think setting is in gconf but text input settings may be able to do it
May 17 13:10:46 <eichi> qwerty12_N800, maemo mapper says after crash: gdbm fatal: read error
May 17 13:11:57 <JackCrow> There is a key combination that initiates it.  Something with shift, i think.  And I am /almost/ positive that i turned it off once with a key combo, but not sure
May 17 13:11:58 <qwerty12_N800> does sound like i/o error,  a lot of apps would  lock up with old card of mine. try an fsck on it  and  untar metasploit on-the-jobit for testing
May 17 13:12:15 <qwerty12_N800> s/on-the-job/on
May 17 13:12:29 <qwerty12_N800> stupid text prediction
May 17 13:12:34 <JackCrow> it wouldn't be too bad, except i don't get the on-screen keyboard when I have the bluetooth keyboard turned off
May 17 13:14:54 <eichi> qwerty12_N800, how can i install fsck for ext2 ?
May 17 13:15:32 <qwerty12_N800> It comes with the ext2 tools package. i'll search up pkg name...
May 17 13:16:19 <eichi> e2fsprogs i think so
May 17 13:16:31 <qwerty12_N800> yes, that is it
May 17 13:26:25 <eichi> fsck.ext2 on device or on mountpoint? device or? which one is the extern sd slow?
May 17 13:26:27 <eichi> slot
May 17 13:30:49 <qwerty12_N800> on device. /dev/mmcblk1
May 17 13:30:55 <qwerty12_N800> or /dev/mmcblk1p1
May 17 13:31:13 <eichi> yes, i know, but: is this the same?
May 17 13:31:27 <qwerty12_N800> should be
May 17 13:35:17 <eichi> yeah, many errors
May 17 13:39:06 <konttori> has anyone compiled kde 4 for maemo?
May 17 13:39:11 <konttori> just out of curiosity
May 17 13:39:26 <qwerty12_N800> nay, broken qemu arm emulation
May 17 13:44:15 <eichi> stupid fsck...i had to press the "y" now more then 500 times...doenst seems to stop
May 17 13:44:22 <Stskeeps> fsck -y
May 17 13:44:23 <Stskeeps> ?
May 17 13:44:23 <Stskeeps> :P
May 17 13:45:53 <eichi> Stskeeps, i tryed -p ;) you saved me ;D
May 17 13:47:28 <eichi> okay, after this check, the content of my sd card is nearly empty ^^
May 17 14:20:02 <briand> has anyone else seen problems loading the gronmayer page?  I get about one screen of info on my n800, then a "Fatal error: Call to undefined function: curl_init()"
May 17 14:21:28 <bmidgley> lardman I see you figured out the sbc thing
May 17 14:21:31 <bmidgley> endian
May 17 15:11:37 <lardman> bmidgley: I was actually going to ask you about whether 64bit types are really needed, but they seem to have disappeared in the newer code
May 17 15:12:00 <lardman> though I may find them again if I'm unlucky ;)
May 17 15:15:56 <zap> briand: it's so for several days already
May 17 15:22:31 * lardman has just compiled the first attempt at a DSP SBC encoder
May 17 15:22:37 <johnx> woo!
May 17 15:22:40 <lardman> coffee break, then time to test
May 17 15:22:55 <johnx> lots of fun things happening on the tablets these days :D
May 17 15:22:56 <lardman> far too easy thus far, something is bound to go wrong
May 17 15:23:03 <qwerty12> SBC is that A2DP stuff init? :)
May 17 15:23:04 <lardman> bbiab
May 17 15:23:05 <johnx> I know that feeling
May 17 15:23:10 <johnx> qwerty12, yup
May 17 15:23:18 <qwerty12> Thanks johnx and lardman :)
May 17 15:23:18 <infobot> de nada, qwerty12
May 17 15:23:27 <qwerty12> infobot, wtf
May 17 15:23:41 <johnx> lart infobot for being infobot
May 17 15:26:44 <johnx> man sony really bungled that "mylo" thing badly...
May 17 15:27:06 <thopiekar> hey qwerty can you please quick explain me howto keep up usb.lan-networking
May 17 15:27:09 <Cptnodegard> does anyone have any tips to good podcasts? preferably gadget related
May 17 15:27:14 <thopiekar> with my asix.ko?
May 17 15:27:18 <Cptnodegard> been searching but found shit
May 17 15:27:21 <qwerty12> thopiekar, No idea, never used it.
May 17 15:27:28 <thopiekar> thx
May 17 15:27:32 <qwerty12> I'd try replacing one of the modules with the asix
May 17 15:34:06 <benh_> hi
May 17 15:34:06 <infobot> que tal
May 17 15:34:20 <benh_> somebody knows of some kind of HOWTO to replace the internal micro-SD of the N810 ?
May 17 15:34:34 <johnx> it's a soldered on chip
May 17 15:34:47 <benh_> oh !
May 17 15:34:54 <benh_> I though it could be updated
May 17 15:35:15 <johnx> With a desoldering kit and some godlike soldering skills maybe ;)
May 17 15:35:20 <benh_> nah
May 17 15:35:25 <benh_> not for me then
May 17 15:35:47 <qwerty12> Or just shoot the micro-SD with glock
May 17 15:45:33 <Navi> The Mylo could of been great if it didn't suck
May 17 15:45:49 <johnx> yeah...avoiding suck is sometimes a difficult trick
May 17 15:46:38 <Navi> My friend's sister had a Mylo
May 17 15:46:45 <Navi> I thought it was interesting
May 17 15:46:46 <qwerty12> Was it ever "hacked" more by their community? wikipedia says it runs linux
May 17 15:47:07 <johnx> qwerty12, nope. Sony locked it down after the 1st firmware and no one got it re-opened
May 17 15:47:15 <Navi> yeh
May 17 15:47:18 <qwerty12> That really sucks.
May 17 15:47:22 <johnx> no community software, just a whole bunch of kids on forums whining
May 17 15:47:38 <briand> zap: ah, thanks.  :/
May 17 15:48:09 <Navi> not only that, but it's butt ugly
May 17 15:48:12 <johnx> there's probably *some* type of vulnerability in netfront or even lower in the linux stack
May 17 15:48:12 <briand> actually, i was trying to find where gstreamer was, so I could install it.  found it, installed it.  turns out, it didn't help with the issue i was having, anyway.  :(
May 17 15:48:16 <Navi> so it'll never gain mainstream success
May 17 15:48:43 <johnx> Navi, yeah, the "aimed at kids"-look kind of meant it missed the linux hackers I think
May 17 15:48:51 <Navi> indeed
May 17 15:49:02 <Navi> Oh yeah, I got xcompmgr and transset working
May 17 15:49:07 <Navi> it was extreme luls
May 17 15:49:11 <johnx> :D
May 17 15:49:15 <johnx> painfully slow?
May 17 15:49:23 <Navi> not painfully, but yes, slow
May 17 15:49:27 <Navi> not where you'd think though
May 17 15:49:34 <Navi> moving it around was bearable
May 17 15:49:38 <briand> issue being:  videocenter doesn't play ABC news videos or Discovery Channel videos very well at all.  I had tried the "-ao gst -ac dspmp3" options for mplayer, and video played great, but no audio.  found i should probably have gstreamer on the n800 for that...
May 17 15:49:51 <Navi> however, when you actually want to use the keyboard, CPU spikes from 0 to 100
May 17 15:49:58 <johnx> interesting
May 17 15:50:07 <briand> okay, so now i have gstreamer, but i can't get audio using that config...
May 17 15:50:14 <Navi> and there's a split second delay between pressing a \key and it coming up on the screen
May 17 15:50:17 <qwerty12> briand, gstreamer is installed. mplayer needs recompiling for gstreamer
May 17 15:50:32 <qwerty12> Actually scratch that, mplayer comes with it afair
May 17 15:50:44 <Navi> dspmp3 is disabled
May 17 15:50:56 <briand> i noticed i didn't have the lib in /usr/lib, so now I do...
May 17 15:51:03 <Navi> the builds in the repos don't have it enabled
May 17 15:51:19 <Navi> at least, that's what I got from the last discussion about it
May 17 15:51:22 <briand> anyway.. i'm kinda disappointed that Discovery Channel video isn't watchable.  :(
May 17 15:51:30 <briand> ..at least, for me (so far)
May 17 15:51:52 <qwerty12> When it comes to video and N800, I'm not disappointed anymore. I'm used to it.
May 17 15:52:04 <Navi> transcoding is fine with me
May 17 15:52:12 <johnx> media-serv++
May 17 15:52:21 <Navi> I still find the screen impressive, actually
May 17 15:52:24 <qwerty12> transcoding is always an option. I did that with my phone.
May 17 15:52:41 <Navi> I do it with... everything
May 17 15:53:05 <johnx> I'm just not big on mobile video I guess
May 17 15:53:09 <ricko73> grrr.  claws keeps crashing on the new rev
May 17 15:53:13 <briand> Navi: transcoding isn't an issue, really.. but c'mon - if I subscribe to a feed in Video Center, shouldn't I be able to play it directly, without having to download the video, cart it over to the desktop, transcode it, and then put it back on the n800?
May 17 15:53:22 <qwerty12> I didn't have to do it at all with my PocketPC. Although I guess it did have 520Mhz arm processor
May 17 15:53:33 <Navi> briand, for stuff like that, I have a cron :D
May 17 15:54:23 <Navi> I've never recommended the N8X0 for video
May 17 15:54:29 <johnx> qwerty12, not much to do with processor and everything to do with that stupid lcd controller bus...
May 17 15:54:29 <Navi> but I'll probably do so for the N900
May 17 15:54:52 <qwerty12> johnx, yeah.
May 17 15:54:57 <johnx> I'll wait to see how the pandora and the N900 compare :D
May 17 15:55:04 <Navi> The processor can decode your normal xvid/divx at full frame rate
May 17 15:55:10 <qwerty12> I'm probably gonna buy another Pocket PC after the N800.
May 17 15:55:11 <Navi> johnx, they're almost the same
May 17 15:55:11 <lardman> re
May 17 15:55:18 <briand> Navi: understood, again, it's not an issue of my not being able to do it... but why should I have to require the presence/usage of my desktop for this?  If I'm at my desktop, I can just watch it -there- for heaven's sake!
May 17 15:55:27 <johnx> Navi, I'm starting to care more about openness these days
May 17 15:55:30 <Navi> Well, assuming that the N900 is going with OMAP3430
May 17 15:55:56 <Navi> johnx, well, pandora's probably going to win :P
May 17 15:56:25 <Navi> Nothing stops us from stealing their OS though
May 17 15:56:27 <Navi> :D
May 17 15:56:50 <johnx> Navi, well user land stuff yeah
May 17 15:57:06 <Navi> Nah, I'm talking about the more intimate stuff
May 17 15:57:08 <johnx> but dragging a binary wifi module along could be painful
May 17 15:57:27 <Navi> the 3520 and 3430 are nearly identical :P
May 17 15:57:38 <Navi> or uh, whatever the pandora was using
May 17 15:57:48 <johnx> right, but the wifi chipset isn't part of the omap
May 17 15:57:53 <Navi> Oh
May 17 15:58:10 <johnx> From what I understand
May 17 15:58:18 <briand> qwerty12: fwiw, I have mplayer 1.0rc1-maemo.27.n8x0 installed... is that not the best option for me on my OS2008 n800 ?
May 17 15:58:53 <johnx> so if Nokia uses one with closed drivers and leaves us with a binary kernel module tied to kernel version x.y.z then we're back in the same boat we are now
May 17 15:59:14 <qwerty12> briand, I think it's fine. You *may* get better performance out of the rc2 but you know?
May 17 15:59:28 <Navi> We could pray they go with an open chip :D
May 17 15:59:37 * lardman has just eaten too much pecan pie
May 17 15:59:46 <qwerty12> Nokia listens to money, not God.
May 17 15:59:52 <johnx> Navi, they'll go with the cheapest one that fits their requirements
May 17 15:59:59 <Navi> :/
May 17 16:00:09 <Navi> Not unless we complain
May 17 16:00:09 <johnx> but like I said, I'll wait until both are out
May 17 16:00:11 <ricko73> Nokia doesn't listen to the Ceiling Kitten?
May 17 16:00:12 <johnx> now rush this time
May 17 16:00:16 <Navi> riot in the streets!
May 17 16:00:37 <johnx> Nokia is always welcome to prove me wrong...in fact, I look forward to it!
May 17 16:00:50 <Navi> hell, if they prove you wrong, I'd be pretty happy :P
May 17 16:01:32 <Navi> I don't know what they'll do
May 17 16:01:48 <johnx> neither do I
May 17 16:01:59 <johnx> hence: wait and see :)
May 17 16:02:05 <Navi> we should hit up some nokia employees and force them to tell us
May 17 16:02:25 <johnx> in the meantime my next project is getting hildon stuff running better in debian
May 17 16:02:44 <johnx> or maybe angstrom
May 17 16:02:47 <johnx> who knows
May 17 16:03:00 <Navi> you could do both at the same time :P
May 17 16:03:18 <johnx> I'll probably at least prototype in debian
May 17 16:03:31 <Navi> a lot of the hildon libs are in the repos
May 17 16:03:36 <johnx> yup
May 17 16:03:39 <Navi> I installed them accidentally
May 17 16:03:44 <johnx> and they have started packaging tinymail
May 17 16:03:49 <Navi> yeah
May 17 16:04:03 <johnx> maybe I'll try and get modest running
May 17 16:04:09 <Navi> I was working on mplayer
May 17 16:04:22 <Navi> but my debian setup in scratchbox stopped working
May 17 16:04:35 <Navi> so I was like, fookdatsheet
May 17 16:10:51 <Proteous> fucked data sheet?
May 17 16:12:50 <unixSnob> check this out.. gps software that only requires 2MB memory for maps => http://shop2.frys.com/product/5539850
May 17 16:13:29 <unixSnob> That s/w could probably run in the garnetvm
May 17 16:17:28 <jku_> 2MB is definitely not enough for USA street maps (vector or raster) in any usable scale...
May 17 16:19:09 <jku_> That's about 4 bits of data for every mile of paved road in the US
May 17 16:22:00 <lardman> well it asked for data, just didn't give any back :)
May 17 16:23:13 * qwerty12 really likes switchdesktop
May 17 16:23:19 <lardman> ah, probably failed to identify the header, need to check that - bloody endianness issues
May 17 16:26:29 <thopiekar> hey qwerty... the readme says that I have to "insmod usbnet.ko"
May 17 16:26:42 <thopiekar> where is the file stored in??
May 17 16:27:13 <qwerty12> no idea, can't see it on my N800
May 17 16:27:21 <lardman> in the initfs iirc
May 17 16:27:31 <thopiekar> path?
May 17 16:27:50 <unixSnob> jku_: what's going on w/ that s/w then?  How can they advertise 2mb needed for maps?
May 17 16:28:06 <lardman> not there actually. I thought here though: sftp://root@192.168.2.3/mnt/initfs/lib/modules/2.6.21-omap1
May 17 16:28:16 <lardman> ignoring the sftp stuff of course
May 17 16:29:32 * qwerty12 hates sites that make you enter "www."
May 17 16:30:32 <jku_> unixSnob, maybe they just don't mention that handheld versions only have a small map area on them (so you'll have to update before going to another town)...
May 17 16:31:11 <thopiekar> qwerty12: can you make me a "usbnet.ko", too. <please?>
May 17 16:31:30 <qwerty12> no idea where it comes from and I'm quite tired atm :/
May 17 16:31:31 <jku_> unixSnob, Windows version map data seems to be 1.1 GB ...
May 17 16:31:33 <thopiekar> that would be very nice ;)
May 17 16:31:50 <qwerty12> just insmod the asix thing and see if dmesg spits out any errors
May 17 16:32:28 <thopiekar> so just type dmesg into the console?
May 17 16:32:36 <qwerty12> yea
May 17 16:32:48 <qwerty12> after insmod <path>/asix.ko
May 17 16:32:54 <thopiekar> as user?
May 17 16:33:02 <qwerty12> you can't insmod as user
May 17 16:33:08 <qwerty12> but dmesg will work as user
May 17 16:33:14 <thopiekar> i mean dmesg
May 17 16:33:19 <thopiekar> oh
May 17 16:33:22 <thopiekar> ok^^
May 17 16:34:56 <johnx> jeez, ITT is being slow
May 17 16:35:02 <thopiekar> so.. I rebooted my device now.
May 17 16:35:50 <qwerty12> I think usbnet is compiled in kernel, at least I don't have .ko when I compiled nokia_2420_defconfig :/
May 17 16:36:25 <thopiekar> host mode started... insmod asix.ko... and dmesg...
May 17 16:37:07 <thopiekar> how can i find now problems in the dmesg.output????
May 17 16:53:31 <thopiekar> hello @ all , again
May 17 16:53:51 <thopiekar> is here someone called "Rob Williams"??
May 17 16:57:39 * lardman ponders, sbc_encode() returns with an error about a lack of space - related to handling sizes in terms of 8bits and 16bits at the same time
May 17 17:05:50 <Sakkath> Hi, I can't upgrade to os2008.
May 17 17:06:08 <Sakkath> I passed -f -R as options and when I turned on the device it says sw-version: <no version>
May 17 17:06:12 <Sakkath> then returns back to my prompt
May 17 17:06:28 <Sakkath> in other examples i've seen it then proceeds to flash the OS . . . not in my case
May 17 17:06:37 <qwerty12> flasher-3.0 -f -F <path to file.bin>
May 17 17:06:46 <Sakkath> -f -F?
May 17 17:06:51 <Sakkath> the Maemo wiki page said -f -R
May 17 17:06:57 <Sakkath> I'll try, hold on.
May 17 17:06:57 <qwerty12> Yes, -f (flash) (FIASCO)
May 17 17:07:08 <qwerty12> *-F  (FIASCO)
May 17 17:07:16 <Sakkath> Oh.
May 17 17:07:55 <Sakkath> qwerty12: http://maemo.org/community/wiki/HOWTO_FlashLatestNokiaImageWithLinux?highlight=%2528CategoryFlashing%2529#312936fe52de4398657f61dbf7c06747 <-- then that is invalid
May 17 17:08:13 <qwerty12> It's been said before. No one really cares tbh :/
May 17 17:08:29 <qwerty12> I learned how to use flasher by the one line sentence of GeneralAntilles :p
May 17 17:08:50 <Gracana> How is that invalid?
May 17 17:09:32 <Gracana> I'm pretty sure that's what I've always used
May 17 17:10:08 <Sakkath> Gracana: well qwerty12 is telling me to use -f -F as opposed to -f -R
May 17 17:10:27 <Gracana> Yes, but that line says ./flasher -F FIASCO -f -R
May 17 17:10:31 <qwerty12> Why not just try it and see if it works :/
May 17 17:10:43 <qwerty12> All -R does is reboot after flashing.
May 17 17:10:48 <Sakkath> Oh, that it does.
May 17 17:10:52 <Sakkath> Please forgive me.
May 17 17:10:54 <Gracana> Which says -f(lash) and -R(eboot) using the -F(IASCO image) FIASCO
May 17 17:11:04 <Sakkath> I didn't notice the -F before the bin file
May 17 17:11:06 <Sakkath> I'm so sorry :D
May 17 17:11:11 <qwerty12> hehe
May 17 17:11:11 <Gracana> ah, I wondered
May 17 17:11:23 <Sakkath> God damn, it's always something _so_ stupid
May 17 17:11:23 <Sakkath> :(
May 17 17:11:35 <Sakkath> Well, I'd rather it be this than my n800 is messed up :P
May 17 17:11:47 <Gracana> Yes, certainly.
May 17 17:14:06 <thopiekar> qwerty: do you know what the output of dmesg "menelaus 1-0072: Setting voltage "VMMC" to 300mV (reg 0x0a, val 0xb8)" means??
May 17 17:14:15 <thopiekar> -.-" leaved
May 17 17:14:17 <thopiekar> ...
May 17 17:15:28 <thopiekar> can someone else tell me what this message means?
May 17 17:15:31 <Sakkath> Nice, thanks.  See ya later.
May 17 17:23:06 <toresbe> hey
May 17 17:23:10 <toresbe> is carrierdetect here?
May 17 17:28:38 <Proteous> I heard typing /names works well
May 17 17:31:04 <Proteous> amazon.com has a nice price drop on the n810 today
May 17 17:31:33 <Proteous> it's normal price on there is $419 today it's $385
May 17 17:31:59 <Proteous> hmm
May 17 17:32:18 <Proteous> might be because it is shipped from someone other then amazon
May 17 17:32:29 <Proteous> so no free shipping
May 17 17:32:31 <Proteous> nm
May 17 17:32:49 <RST38h> hehe
May 17 17:32:53 <RST38h> like ebay!
May 17 17:33:10 <kulve> KotCzarny: AStorm: I made a little test about the mfpu and mfloat-abi options: http://tuomas.kulve.fi/blog/2008/05/17/gcc-options-and-floating-point/
May 17 17:36:34 <AStorm> Let's see :-)
May 17 17:38:45 <kulve> actually now I noticed that I forgot to say something important about the gcc command I used..
May 17 17:40:03 <kulve> AStorm: I added there "-O2 and -march=armv6 were given in all cases"
May 17 17:41:51 <AStorm> Why nothing works with hard and vfp?
May 17 17:41:53 <AStorm> it should!
May 17 17:42:03 <AStorm> you messed up by using that march
May 17 17:42:14 <kulve> "vfp.c:1: sorry, unimplemented: -mfloat-abi=hard and VFP"
May 17 17:42:23 <AStorm> intriguing
May 17 17:42:31 <AStorm> apparently my GCC is newer
May 17 17:42:41 <AStorm> (4.2.3 to be exact)
May 17 17:42:45 <bmidgley> lardman multiplies are 32x32=64 bits iirc
May 17 17:42:46 <AStorm> I'll recheck it.
May 17 17:43:07 <kulve> yeah, the maemo gcc is quite old. But I'm not sure if it has some tweaks backported to it
May 17 17:43:34 <AStorm> No, it doesn't. It's the plain CodeSorcery one
May 17 17:43:52 <kulve> based on what information?
May 17 17:43:57 <bmidgley> lardman I really wish we didn't provide for big-endian input...
May 17 17:44:22 <AStorm> kulve: notice the CodeSourcery tag
May 17 17:44:40 <AStorm> then find the compiler and compare :-)
May 17 17:44:54 <bmidgley> lardman it complicates zero-copy (the codec does a lot of copying currently)
May 17 17:46:43 <Whisker_> @denny: hi denny, i'm new to irc and trying to link nicks, but allways receiving errors. can you help?
May 17 17:50:15 <Gracana> Whisker_: /msg nickserv group nick password
May 17 17:50:26 <Gracana> I think that's what you're talking about
May 17 17:50:37 <Gracana> You can always /msg nickserv help group
May 17 17:51:09 <Gracana> FreeNode nickserv doesn't do grouping, apparently. Or it calls it something else..
May 17 17:51:23 <Gracana> Oh, it calls it linking :P
May 17 17:51:44 <Gracana> Same syntax, just use the word link
May 17 17:52:09 <AStorm> Gracana: too late
May 17 18:10:27 <toresbe> Hrmmm
May 17 18:10:44 <toresbe> Could anyone do me a fairly huge favour?
May 17 18:11:32 <RST38h> how much?
May 17 18:12:02 <toresbe> 3 kilograms of favour
May 17 18:12:08 <RST38h> no.
May 17 18:12:20 <qwerty12_N800> 3 kg of weed!?!
May 17 18:12:23 <toresbe> I need to get a simh binary running on my N800 soon, for a quick presentation
May 17 18:12:56 <toresbe> but I don't want to install the development stuff
May 17 18:13:08 <toresbe> could somebody please build a package of it for me on the N800?
May 17 18:13:59 <qwerty12_N800> If you  had asked an hour ago ; i'd have tried but i'm off computer  now.
May 17 18:14:35 <unixSnob> for 3kg of weed, qwerty12_N800 should be willing to fly to toresbe, and provide onsite support
May 17 18:15:08 <qwerty12_N800> lol
May 17 18:15:27 <Gracana> toresb, give me a minute.. :)
May 17 18:16:18 * Grackle on the desktop machine
May 17 18:16:28 <unixSnob> Does the n800 have a voltage regulator?
May 17 18:16:43 * qwerty12_N800 goes  back into diablo repo...
May 17 18:18:24 <unixSnob> Anyone try feeding just 500mA@5V to the NIT?  does it cause any problems?
May 17 18:18:46 <Grackle> toresbe: Waiting for my virtual machine to boot. Any specifics I should know about? What OS are you running?
May 17 18:25:21 <qwerty12_N800> can anyone dump default values of gconf /apps/osso/xterm/key_labels and /apps/osso/xterm/keys
May 17 18:33:18 <Grackle> Compiled, toresbe. Uploading.
May 17 18:39:57 <Grackle> toresbe: http://nyus.joshuawise.com/upload/data/simh_bin.tar
May 17 18:40:09 <Grackle> That's untested. Good luck. :P
May 17 18:51:34 <trickie> unixSnob: yes it does have a regulator
May 17 18:52:40 <unixSnob> What voltage range can it tolerate?
May 17 18:52:56 <glass> probably up to 8 or so if it's like other nokias
May 17 18:53:02 <AStorm> woohoo
May 17 18:53:06 <unixSnob> i'd like to connect it to my cars 12V
May 17 18:53:11 <glass> 5 is fine
May 17 18:53:13 <AStorm> that'd mean I can get even higher efficiency
May 17 18:53:23 <trickie> unixSnob: http://focus.ti.com/lit/ds/swps021d/swps021d.pdf
May 17 18:53:25 <AStorm> which regulator are we talking about?
May 17 18:53:26 <trickie> i think thats it
May 17 18:53:29 <glass> 12v is probably too high
May 17 18:53:45 <glass> unixSnob: just get a cheap cellphone car charger
May 17 18:54:00 <glass> that'll probably have a regulator to down it to 5
May 17 18:55:03 <unixSnob> Well I have a USB port in the car, w/ a max current of 500mA.  The NIT will overdraw that I think
May 17 18:55:09 <unixSnob> the nit's PSU is 890mA
May 17 18:55:54 <AStorm> glass: sorry, only 5.5V max in
May 17 18:56:00 <AStorm> ;P
May 17 18:56:10 <glass> AStorm: some old nokia chargers were 6v iirc
May 17 18:56:12 <AStorm> although you can get 10%
May 17 18:56:18 <AStorm> which means just above 6V
May 17 18:56:22 <glass> AStorm: and some phones can tolerate 9v in
May 17 18:56:36 <glass> from a 9v batt without extra reg
May 17 18:56:37 <AStorm> this chip isn't a phone converter
May 17 18:57:20 <AStorm> but, wait
May 17 18:57:24 <AStorm> that's for battery input
May 17 18:57:28 <AStorm> the 5.5V... ?!
May 17 18:57:46 <AStorm> this means... that I could just plug the NiMH in directly
May 17 18:57:55 <AStorm> after somehow disabling the charger
May 17 18:58:21 <AStorm> although it's recommended to stay below 4.5 V
May 17 18:59:57 <AStorm> NiMH will easily tolerate that 880 mA draw
May 17 19:00:18 <AStorm> (note, it really can be that high - so that wimpy Nokia AC-4E charger can't supply enough ;P )
May 17 19:00:42 <AStorm> the q still remains: how to disable the charging, then maybe later reenable it?
May 17 19:02:00 <unixSnob> perhaps by removing the battery, and putting it back in
May 17 19:03:01 <AStorm> nopes
May 17 19:03:07 <AStorm> it's enabled by software
May 17 19:03:16 <AStorm> so I guess it can also be disabled
May 17 19:03:49 <RST38h> Well, the software says "disconnect charger to save electricity"
May 17 19:03:50 <AStorm> killing the control program seems ungraceful
May 17 19:04:00 <RST38h> It would have to say it if it could disable charging
May 17 19:04:09 <qwerty12_N800> dbus sends out signals on charger connect. Maybe option in sysfs or proc?
May 17 19:05:32 <AStorm> it's dsme stuff
May 17 19:05:42 <RST38h> s/would/wouldn't/
May 17 19:05:42 <infobot> RST38h meant: It wouldn't have to say it if it could disable charging
May 17 19:05:55 <qwerty12_N800> ahh
May 17 19:06:33 <RST38h> Also, the battery itself may be keeping track of its charge to avoid meltdown
May 17 19:07:49 <AStorm> RST38h: it can't obviously disable the charger itself!
May 17 19:07:55 <AStorm> so it'd still suck power
May 17 19:08:20 <AStorm> yes, quite a lot, like 10%
May 17 19:08:41 <RST38h> yep...
May 17 19:09:41 <Navi> whee
May 17 19:10:04 <AStorm> TWL92230 has a backup battery port
May 17 19:10:09 <AStorm> I smell a nice mod
May 17 19:10:39 <RST38h> heheh
May 17 19:12:25 <AStorm> not sure yet if it only powers the chips, or the whole thing
May 17 19:15:34 <AStorm> nah, it'll only keep RTC and such up
May 17 19:20:17 <AStorm> there must be another charging chip anyway
May 17 19:20:59 <EruditeHermit> hey has anyone ever got the SIP video working with the tablet?
May 17 19:21:06 * unixSnob considers putting one of these behind the dashboard, and wiring the 5V output to a USB hub: http://www.powerstream.com/mini-itx.htm
May 17 19:21:12 <EruditeHermit> I can never get the video to work with ekiga
May 17 19:21:49 <unixSnob> EruditeHermit: I've heard some folks have success w/ gizmo and video (which is using the SIP protocol)
May 17 19:22:47 <unixSnob> but it doesn't work well enough to be practical.  It tends to just work if two NITs are connected from the same LAN
May 17 19:23:27 <EruditeHermit> thats one of the problem areas with Linux
May 17 19:23:36 <EruditeHermit> its difficult to get videoconferencing to work
May 17 19:27:38 <lardman> bmidgley: long long on the DSP is only 40bits
May 17 19:27:50 <AStorm> lardman: what a pity :-)
May 17 19:27:54 <lardman> bmidgley: I can implement 64bit types, but would prefer not to if pos
May 17 19:28:07 <RST38h> lardman: should have called it "longer" =)
May 17 19:28:12 <AStorm> this Menelaus chip is quite powerful
May 17 19:28:14 <lardman> well 2x32bits in a struct will get around it
May 17 19:28:35 <lardman> well long is 32bits and int is 16bits, so long long is long long really :)
May 17 19:29:17 <lardman> wasn't an endianness issue in the end, something to do with byte counting getting confused; looking at it now
May 17 19:29:55 <AStorm> it should be at least 48bit to be useful ;P
May 17 19:29:55 <lardman> bmidgley: nice that the codec doesn't really do any allocation though, makes life much easier
May 17 19:30:00 <qwerty12_N800> AStorm, Is Menelaus to do with mmc cards?
May 17 19:30:11 <AStorm> qwerty12_N800: with power supply and MMC, yes
May 17 19:30:12 <lardman> AStorm: oi! stop saying nasty things ;D
May 17 19:30:20 <qwerty12_N800> AStorm, Thx
May 17 19:31:40 <lopz> re
May 17 19:31:43 <KotCzarny> yawn
May 17 19:31:46 * AStorm starts to hate BGA
May 17 19:31:49 <KotCzarny> astorm, got any results?
May 17 19:31:50 <KotCzarny> :)
May 17 19:32:03 <AStorm> KotCzarny: right now nope, still too busy
May 17 19:32:16 <KotCzarny> i have seen kulve's link
May 17 19:32:46 <KotCzarny> apparently it's much faster with softfp
May 17 19:33:01 <AStorm> yes, but VFP didn't work in hard mode for him
May 17 19:33:03 <AStorm> :>
May 17 19:33:14 <KotCzarny> pity he didn't post the code
May 17 19:33:16 <kulve> AStorm: prove me wrong
May 17 19:33:21 <KotCzarny> hi kulve
May 17 19:33:22 <kulve> KotCzarny: the code is there
May 17 19:33:24 <AStorm> I'll try.
May 17 19:33:28 <KotCzarny> ahm, sorry
May 17 19:33:44 <EruditeHermit> I got a gizmo account but I can't find out what the suffix for the account is
May 17 19:33:48 <KotCzarny> i have overlooked :)
May 17 19:33:49 <EruditeHermit> @gizmo.com?
May 17 19:33:49 <kulve> AStorm: be sure to check that you don't optimize that for loop away ;)
May 17 19:33:55 <AStorm> hmm
May 17 19:33:57 <kulve> it might happen easily with newer gcc
May 17 19:33:57 <AStorm> -O1?
May 17 19:34:07 <AStorm> that shouldn't remove dead code
May 17 19:34:09 <kulve> AStorm: I used O2 but old gcc
May 17 19:34:13 <AStorm> mhm
May 17 19:35:00 <lardman> bmidgley: interesting; sbc_pack_frame() takes an input argument, len, which doesn't appear to be used at all?
May 17 19:35:57 <kulve> AStorm: but you need to remember that I'm using the official toolchain for chinook. If you are using something else, you need to specify it.
May 17 19:36:06 <KotCzarny> also
May 17 19:36:19 <lardman> bmidgley: Another random thing I noticed (as I had to create typedefs): sbc_pack_frame() has two u_int32_t variables, while everywhere else has uint32_t
May 17 19:36:21 <KotCzarny> it's interesting how gcc optimization helps the code
May 17 19:36:30 <AStorm> kulve: ofc I will :-)
May 17 19:36:32 <KotCzarny> so checking for misc -O options is also useful
May 17 19:36:41 <AStorm> A cross compiler actually.
May 17 19:37:02 <KotCzarny> script for checking all interesting combinations is easy :)
May 17 19:37:10 <AStorm> KotCzarny: it's called acovea
May 17 19:37:11 <AStorm> ;P
May 17 19:37:23 <KotCzarny> i usually use /bin/bash
May 17 19:37:24 <KotCzarny> :)
May 17 19:37:25 <AStorm> it's smarter than your average script
May 17 19:37:32 <kulve> KotCzarny: depends on the case. There can be huge difference on some codecs between no -O option and -O3 (talking about speex here=
May 17 19:37:33 <AStorm> AI and all
May 17 19:38:29 <AStorm> kulve: do you know if Speex resampler could be further optimized for ARM?
May 17 19:38:51 <AStorm> KotCzarny: btw, I'm still waiting for that pulseaudio
May 17 19:38:57 <kulve> resampler? I think that's something quite new and it propbably could be optimized
May 17 19:39:03 <kulve> especially for ARMv6
May 17 19:39:04 <AStorm> until I'll have the time to do that myself
May 17 19:39:09 <KotCzarny> astorm: hmm? :)
May 17 19:39:27 <qwerty12_N800> I compiled quick pulseaudio, no idea if it works
May 17 19:39:43 <AStorm> qwerty12_N800: with Speex support?
May 17 19:39:53 <qwerty12_N800> Erm, no :/
May 17 19:39:58 <AStorm> that resampler should be the fastest of the ones available
May 17 19:39:58 <kulve> AStorm: I also got the speex encoder running on DSP but with lousy perf. According to Mr. Speex it should be much faster even when unoptimized
May 17 19:41:01 <lardman> kulve: yeah but he didn't have to hack it into life did he
May 17 19:41:32 <lardman> kulve: have you done any more with it?
May 17 19:41:39 <kulve> lardman: not sure what you mean :) But he was mainly comparing MIPS there
May 17 19:41:42 <kulve> lardman: no :/
May 17 19:41:56 <lardman> kulve: I was meaning that it wasn't a nice thing for him to say
May 17 19:42:44 <kulve> lardman: I didn't see it like that. He was just surprised to hear that it was 5x slower than on ARM side so there must be something odd happening
May 17 19:42:56 <AStorm> like gcc fail
May 17 19:43:08 <lardman> kulve: I read it the wrong way then, my fault
May 17 19:43:11 <AStorm> time to look into asm
May 17 19:44:15 <AStorm> and profile it
May 17 19:45:38 <lardman> oops, yet another watchdog reboot
May 17 19:45:48 <AStorm> mwhahaha
May 17 19:45:49 <lardman> can't quite work out what I'm doing to kill it
May 17 19:45:50 <KotCzarny> bad watchdog, bad
May 17 19:45:58 <AStorm> it barks and bites
May 17 19:45:58 <kulve> lardman: it logs it
May 17 19:46:18 <lardman> kulve: Oh I know I'm resetting the DSP, just not quite sure why :)
May 17 19:46:29 <kulve> ah, ok :)
May 17 19:46:48 <AStorm> does anyone need a simulator?
May 17 19:46:53 <lardman> my DSP resets are bad as dsp_dld can't be reloaded as I'm sharing the framebuffer too
May 17 19:46:55 <kulve> you can also disable the "reset the device if daemon X dies too many times" feature
May 17 19:47:13 <AStorm> kulve: wrong watchdog, he's talking about DSP one? :>
May 17 19:47:15 <lardman> kulve: yeah, but at least it comes back up with the DSP in a normal state
May 17 19:47:51 <lardman> AStorm: actually I think the DSP reset probably tries to reload dsp_dld which won't which then triggers the dsme watchdog
May 17 19:47:53 <kulve> AStorm: no, problems in DSP results to restarting dsp_dld and when it restarts X times, the dsme restarts the whole device
May 17 19:48:00 <AStorm> hehe
May 17 19:48:18 <AStorm> mad software
May 17 19:48:23 <kulve> or, if it's the watchdog and not the restart count, then my comments are just bs..
May 17 19:49:07 * lardman was wondering why an Nvidia driver was tainting the kernel on his N800; didn't notice the ssh connection had died; doh!
May 17 19:49:17 <kulve> :)
May 17 19:49:18 <KotCzarny> lol
May 17 19:49:19 <AStorm> which driver? wtf?
May 17 19:49:31 <AStorm> ahem
May 17 19:49:48 <AStorm> use some more readable prompts then
May 17 19:49:51 <lardman> kulve: probably a restart count as it can't restart it when the fb is shared (which may be why the 770 used to crash more often)
May 17 19:50:26 <AStorm> btw, is OMWeather fixed now?
May 17 19:50:55 <qwerty12_N800> yes
May 17 19:51:01 <AStorm> great
May 17 19:51:07 * AStorm runs to update
May 17 19:51:21 <KotCzarny> kulve: left menu is a little broken on your site
May 17 19:51:36 <KotCzarny> on a window 900x450
May 17 19:51:38 <KotCzarny> :)
May 17 19:51:42 <KotCzarny> *in
May 17 19:52:46 <kulve> KotCzarny: yeah, I know.. If you mean that it doesn't fit entirely?
May 17 19:52:56 <KotCzarny> yes
May 17 19:52:59 <KotCzarny> and doesn't scroll
May 17 19:53:42 <kulve> yeah, the not-scrolling is a feature. And it works OK on most non-IE browsers with decent window sizes, so I haven done anything to it..
May 17 19:53:46 <kulve> maybe I need to
May 17 19:53:53 <KotCzarny> firefox 2.x here
May 17 19:53:54 <KotCzarny> :)
May 17 19:54:22 * qwerty12_N800 wishes i was back on ff 2
May 17 19:54:34 <kulve> yeah, it bugs with all browser with "too" small window size..
May 17 19:55:10 <KotCzarny> kulve: i assume than it will happen on nit too, then :)
May 17 19:55:31 <qwerty12_N800> it does :/
May 17 19:57:17 <KotCzarny> *that
May 17 19:57:49 <kulve> yes, because it doesn't scroll..
May 17 19:58:00 <kulve> maybe I need to change it to regular scrolling thing..
May 17 19:58:04 <KotCzarny> qwerty: you can simply unpack firefox 2.x somewhere and add to menu :)
May 17 19:58:25 <KotCzarny> kulve: yeah, stuck menu looks nice but is inpractical i think
May 17 19:58:53 <qwerty12_N800> will do thanks :), don't want to use deb incase something breaks during upgrade
May 17 19:59:08 <lardman> oops, another bus error, but it looks like it might have even done something :)
May 17 19:59:26 <KotCzarny> qwerty: just grab tarball from mozilla.org
May 17 20:00:02 <lardman> http://maemo.pastebin.com/m11cddf5f
May 17 20:01:24 <lardman> hmm, doesn't seem to have actually saved anything though
May 17 20:01:36 <Gracana> What happened to toresbe? ...
May 17 20:01:58 <Gracana> Looks like I didn't get that package up in time.
May 17 20:15:36 <rm_you> lardman: I wish I could hug you :P
May 17 20:15:43 <KotCzarny> lol
May 17 20:15:58 <rm_you> DSP-SBC == <3
May 17 20:16:47 <bmidgley> lardman: the len is supposed to be used to avoid buffer overflow looks like
May 17 20:16:54 <qwerty12_N800> I'll <3 for dsp sbc (joke) :p
May 17 20:17:02 <rm_you> lol
May 17 20:20:22 <lardman> bmidgley: It doesn't appear to be used though, just thought I'd point it out (otherwise I'm making a mistake)
May 17 20:20:50 <bmidgley> lardman: right, it's a feature we've yet to finish
May 17 20:20:56 <lardman> cool
May 17 20:21:06 <lardman> just wanted to check I wasn't messing up :)
May 17 20:21:46 <lardman> God only knows where my data has gone; I wonder if my ARM-side code has a problem
May 17 20:21:50 <lardman> http://maemo.pastebin.com/m3fcb9a9c
May 17 20:36:09 <rafael_> somebody is brazilian in here?
May 17 20:36:33 <KotCzarny> try /who #maemo and look for .br
May 17 20:39:57 <rafael_> haha, thanks, however I asked because easier for me
May 17 20:40:47 <rafael_> .br
May 17 20:41:39 <rafael_> however, like that is me
May 17 20:41:42 <rafael_> no problem...
May 17 20:43:04 <rafael_> someone I can respond if there is any utility that I can use to create a network with my USB N800?
May 17 20:46:37 <KotCzarny> yes
May 17 20:46:53 <rafael_> where I can find?
May 17 20:47:49 <KotCzarny> google for maemo usbnet
May 17 20:47:51 <KotCzarny> or something
May 17 20:50:30 <AStorm> rafael_: the utility is gadgetfs usb module
May 17 20:50:45 <AStorm> you get a usb0 interface then, use ifconfig or udhcpcd to set it up
May 17 20:50:59 <AStorm> s/gadgetfs/gadget/
May 17 20:51:05 <rafael_> uhmm
May 17 20:51:08 <AStorm> gadget_eth or something
May 17 20:51:09 <rafael_> gadgets :)
May 17 20:51:18 <rafael_> ok
May 17 20:51:26 <AStorm> the module is on the device already
May 17 20:51:28 <zap> g_ether
May 17 20:51:30 <AStorm> I think.
May 17 20:51:33 <AStorm> yes, g_ether
May 17 20:51:55 <zap> it doesn't work reliably
May 17 20:52:00 <AStorm> nah, it does
May 17 20:52:02 <zap> for some reason
May 17 20:52:11 <zap> well it works ok for the first time I plug USB in
May 17 20:52:22 <zap> and if I do it second time, it usually hangs the n810
May 17 20:52:28 <AStorm> not here :P
May 17 20:52:49 <rafael_> really thanks for this hint AStorm
May 17 20:53:10 <zap> rafael_: the easiest way is to do from command-line:
May 17 20:53:14 <zap> (as root)
May 17 20:53:29 <zap> rmmod g_file_storage (or something like that, do a lsmod)
May 17 20:53:43 <zap> then insmod /mnt/initfs/lib/modules/`uname -r`/g_ether.ko
May 17 20:53:55 <zap> then ifconfig usb0 192.168.0.202 up
May 17 20:54:00 <zap> then insert the USB cable into PC
May 17 20:54:26 <rafael_> ok, but, on PC is easy
May 17 20:55:26 <rafael_> but, I see today something software that created a graphical interface to this
May 17 20:55:39 <zap> you can try it, but it hangs for me
May 17 20:56:31 <rafael_> I'm know that come (dependence) of Pluthon
May 17 20:56:48 <rafael_> however, I don't find :/
May 17 20:57:22 <zap> https://garage.maemo.org/projects/usb-otg-plugin/
May 17 21:24:29 <Jaffa> Interesting, a maemo Google Summer of Code project from the VLC guys: http://wiki.videolan.org/SoC_2008/New_VLC_interface_for_Maemo
May 17 21:25:43 <Cptnodegard> nice
May 17 21:26:00 <Cptnodegard> but isnt vlc rather flacky compared to mplayer on maemo?
May 17 21:26:08 <Cptnodegard> flaky*
May 17 21:26:32 <KotCzarny> wouldn't it be nice to have generic video gui instead?
May 17 21:26:44 <KotCzarny> that could do mplayer, vlc or any other player
May 17 21:26:53 <Jaffa> Indeed, I wasn't aware of a well-optimised port. And articles like http://www.freesoftwaremagazine.com/columns/2008_google_summer_code_21_projects_im_excited_about assume that because it can decode anything on high-CPU x86 boxes, it'll work for anything on a 400MHz, resource-constrained, h/w-limited ARM
May 17 21:26:55 <RST38h> IMHO mplayer is flaky as well
May 17 21:27:16 <Jaffa> KotCzarny: both Canola and MediaBox use both gstreamer or mplayer as appropriate for a given video.
May 17 21:27:39 <KotCzarny> jaffa: but those are quite heavy
May 17 21:27:48 <KotCzarny> and use medialibrary
May 17 21:27:49 <Jaffa> And technically the built-in player *is* generic through gstreamer, we've just got one implementation of gstreamer plugins on maemo ATM
May 17 21:28:30 <Cptnodegard> i really dont see how they can ever compare to mplayer so why bother
May 17 21:28:33 <RST38h> Canola is most likely going to move away from using DBus sooner or later
May 17 21:28:49 <RST38h> They are having too many sync problems when using DBus
May 17 21:29:23 <oDesk> hey all, i've installed Flash 9 on os2006 it works but no audio, any idea ?? i read of lib file for sound where i can find it ?
May 17 21:30:58 <oDesk> i'm on n770
May 17 21:33:14 <zap> vlc can stream media, but who cares
May 17 21:33:49 <zap> Jaffa: what you mean "we've got one implementation" ? Maemo gstreamer lacks an ffmpeg plugin, which is very bad.
May 17 21:35:27 <Jaffa> zap: we've got one set of gstreamer plugins, but as you say, gstreamer can have an ffmpeg plugin and then you've effectively got both mplayer and VLC.
May 17 21:35:46 <zap> yep, what puzzles me is that nobody did it yet
May 17 21:36:40 <zap> as far as I know, there are many ARM assembler optimizations that were made for mplayer but will work with gstreamer too
May 17 21:36:48 <zap> in ffmpeg I mean
May 17 21:37:25 <ssvb> zap: gstreamer has a very heavy overhead itself
May 17 21:37:49 <zap> indeed, I have measured ogg decoding and it's about 50% overhead (150% vs 100% mplayer)
May 17 21:37:53 <ssvb> zap: you can try benchmarking some video playback with oprofile
May 17 21:38:12 <zap> I just played the ogg to the null device and measured the time
May 17 21:39:13 <ssvb> zap: wrapping ffmpeg into gstreamer will also result in poor playback
May 17 21:39:25 <KotCzarny> ssvb: would it be possible to have mplayer using dsp decoding asm ?
May 17 21:39:43 <KotCzarny> without gst, plain dsp code
May 17 21:40:02 <KotCzarny> well, not exactly asm, but code in general
May 17 21:40:23 <zap> ssvb: I understand, and still, gstreamer is a well-designed multimedia system, while mplayer is full of dirty hacks
May 17 21:40:42 <ssvb> KotCzarny: probably yes, it's better to ask lardman
May 17 21:40:53 <KotCzarny> zap: but if you are short on cpu you need hacks over beauty of the code
May 17 21:41:07 <zap> sad but true :)
May 17 21:41:11 <KotCzarny> ssvb: what would be needed from mplayer side?
May 17 21:42:59 <ssvb> KotCzarny: most likely nothing special, mplayer has support for many audio and video decoding libraries, it even has a loader for win32 dll's
May 17 21:43:22 <ssvb> KotCzarny: so I guess it's very flexible
May 17 21:43:31 <EruditeHermit> is it possible to run hildon on the desktop?
May 17 21:43:32 <KotCzarny> ssvb, hmm
May 17 21:43:38 <zap> hehe, this fits into a standard 19" server rack: http://www.canford.co.uk/commerce/productdetails.aspx?productid=11-231
May 17 21:43:57 <KotCzarny> ssvb, as i understand the only way of using dsp in mplayer is gst
May 17 21:44:13 <KotCzarny> hmm
May 17 21:44:24 <KotCzarny> i think i'm confused
May 17 21:44:54 <KotCzarny>  dsptasks are gstreamer, yes/no ?
May 17 21:45:01 <kulve> no
May 17 21:45:14 <kulve> gstreamer (and others) use dsp tasks
May 17 21:45:28 <kulve> in gstream you need to do a gst plugin that uses the dsp task running in the dsp
May 17 21:45:33 <kulve> anybody can create and use a dsp task
May 17 21:45:40 <KotCzarny> so it's a matter of writing proper -ao for mplayer that would use dsp task device, right?
May 17 21:45:54 <kulve> but playing audio directly from dsp is nokia propriatary stuff
May 17 21:46:17 <kulve> KotCzarny: yes (I think)
May 17 21:46:42 <ssvb> KotCzarny: not quite, there is a difference between decoders and audio output modules
May 17 21:47:26 <KotCzarny> but /dev/*/mp3dec decodes and plays without returning decoded data
May 17 21:47:29 <zap> how can DSP play the decoded audio directly? Does it have direct access to ALSA devices?
May 17 21:47:32 <KotCzarny> am i right?
May 17 21:47:47 <ssvb> KotCzarny: implementing -ao gst as audio output module was a forced solution and dirty hack
May 17 21:47:49 <kulve> zap: there's no proper alsa device in itos2008
May 17 21:47:57 <zap> doh
May 17 21:48:09 <zap> so bad?
May 17 21:48:44 <kulve> KotCzarny: yes, mp3dec plays the audio directly. But I'm not sure if its api is open either..
May 17 21:48:57 <KotCzarny> hmm
May 17 21:49:17 <KotCzarny> it would save some data moving around
May 17 21:49:26 <ssvb> KotCzarny: exactly, everything would be much easier and more "well-designed" (tm) if mp3 gstreamer element could return decoded data back
May 17 21:49:31 <kulve> the gst plugins that uses the dsp tasks are open, but they depend on some headers that are closed
May 17 21:49:42 <ssvb> KotCzarny: and even better would be to do this without gstreamer
May 17 21:50:06 <ssvb> KotCzarny: gstreamer is the only public API for using dsp tasks
May 17 21:50:13 <KotCzarny> hmm
May 17 21:50:28 <KotCzarny> lardman uses dsp tasks directly, or not?
May 17 21:50:30 <kulve> passing the data back from the dsp and then back to dsp for pcm playback introduces some overhead
May 17 21:50:53 <kulve> KotCzarny: the dsp task API is open, the nokia tasks are not
May 17 21:51:19 <zap> hmm so the only way to output sound is esd?
May 17 21:51:24 <kulve> i.e. you can make your own task and use it but you can't use the nokia's tasks
May 17 21:51:36 <kulve> zap: esd feeds the data to dsp too. Just like gst
May 17 21:51:54 <zap> but what device is dsp? I can't see either /dev/dsp or /dev/snd/something except the mixer
May 17 21:51:59 <kulve> zap: there's also an ALSA plugin that does the same
May 17 21:52:05 <KotCzarny> zap: /dev/dsptasks/
May 17 21:52:07 <KotCzarny> or something
May 17 21:52:15 <ssvb> kulve: how much is this overhead?
May 17 21:52:18 <zap> /dev/dsptasks/ are the way to output sound???
May 17 21:52:29 <kulve> zap: the OMAP cpu includes ARM core and DSP core
May 17 21:52:45 <zap> ah, /dev/dsptask/pcm[0-3]
May 17 21:53:08 <kulve> ssvb: I did some tests and it was <1ms for one data write (1-500 bytes) to one direction. But that's just my test, nothing sure
May 17 21:53:17 <zap> pretty exotic way to output sound )
May 17 21:53:32 <kulve> zap: yes, very propriatary
May 17 21:54:07 <ssvb> kulve: and how much is that in ARM core cpu cycles?
May 17 21:54:31 <ssvb> kulve: is ARM core busy during this time, or could do some other data crunching?
May 17 21:55:10 <kulve> ssvb: not a clue. What I did was gettimeofday, send data, receive "ok", gettimeofday (so not exactly like I just said)
May 17 21:55:28 <RST38h> wait wait gentlemen
May 17 21:55:29 <ssvb> kulve: latency is actually not so important as it can be compensated
May 17 21:55:45 <RST38h> do you mean I can use /dev/dsptasks/pcmx the same way I use /dev/dsp on desktop linux?
May 17 21:55:55 <KotCzarny> rst: not exactly
May 17 21:56:05 <kulve> ssvb: yeah, the latency is not an issue, but I think the ARM cpu is busy during than time, so it takes some time..
May 17 21:56:08 <RST38h> sans ioctl()s?
May 17 21:56:09 <ssvb> kulve: and we already have a high latency with current audio output, but nobody has died from it (though seeking is slow)
May 17 21:56:20 <EruditeHermit> johnx: does the debian port have openoffice?
May 17 21:56:23 <KotCzarny> rst: i think it needs some initialization
May 17 21:56:34 <KotCzarny> when you open it
May 17 21:56:51 <kulve> RST38h: you can read(), write() and ioctl() to the dsp task. But nobody knows what Nokia's tasks want
May 17 21:56:59 <RST38h> heheh
May 17 21:57:13 <kulve> RST38h: /dev/dsptask/* do not have anything in common with /dev/dsp
May 17 21:57:18 <RST38h> my guess is these are TI's not Nokia's
May 17 21:57:20 <kulve> just bad naming
May 17 21:57:37 <KotCzarny> yeah, signal vs sound :)
May 17 21:57:41 <kulve> RST38h: could be Nokia's too. Who knows.
May 17 21:58:44 <KotCzarny> hmm
May 17 21:58:53 <kulve> KotCzarny: basically the initialization is done automatically by the plugin when it gets loaded. They might need some setting up, like telling the channel count, bitrate etc.
May 17 21:59:06 <KotCzarny> kulve: is spying on ioctls possible?
May 17 21:59:07 <kulve> s,plugin,task,
May 17 21:59:21 <kulve> KotCzarny: yeah, it should show up in strace just fine
May 17 21:59:23 <KotCzarny> ie. moving pcmtask device node somewhere and inserting spy device of ours?
May 17 21:59:31 <zap> kulve: but you said the sources for Nokia plugins are available, just not some private headers?
May 17 21:59:41 <KotCzarny> then just doing dome basing playing
May 17 21:59:58 <kulve> zap: yeah, for the gst plugins. But iirc they didn't contain anything usefull
May 17 22:00:08 <lardman> re
May 17 22:00:16 <zap> but can't you sniff the expected data format?
May 17 22:00:19 <KotCzarny> ah, all the hackers are here
May 17 22:00:21 <KotCzarny> rejoice!
May 17 22:00:21 <KotCzarny> :)
May 17 22:00:29 <lardman> ssvb: Did you see that yuv420 output code I was fiddling with - post to maemo-developers
May 17 22:01:42 <lardman> Someone was asking out audio output from the DSP?
May 17 22:01:45 <kpel> hi
May 17 22:01:45 <infobot> moin moin
May 17 22:01:53 <KotCzarny> i should subscribe to maemo-developers
May 17 22:02:27 <kulve> KotCzarny: you can also read the archives. But I see the list useful
May 17 22:02:33 <ssvb> lardman: green color from your screenshot seems like zero U and V data
May 17 22:02:44 <lardman> ssvb: regarding returning the decoded data, my impression is that this will happen in the end
May 17 22:02:49 <ssvb> lardman: maybe something is not initialized
May 17 22:03:12 <zap> or maybe overflow
May 17 22:03:22 <lardman> ssvb: I will probably feed it some fixed value YUV data (different value for Y,U,V etc.) and return the framebuffer data to debug it
May 17 22:03:34 <KotCzarny> lardman: 0 uv == signed vs unsigned?
May 17 22:03:39 <ssvb> lardman: that's a good idea
May 17 22:03:55 <lardman> KotCzarny: should be unsigned, I think
May 17 22:04:15 <lardman> ssvb: just strange that most of the image is good, with those strange glitches
May 17 22:04:42 <zap> perhaps YUV to RGB conversion can give negative values, which should be treated differently
May 17 22:04:55 <zap> I don't remember the formulas though
May 17 22:05:22 <lardman> zap: that was YUV output converted from YUV input
May 17 22:05:29 <ssvb> zap: there should be no conversion there, just reshuffling of bytes
May 17 22:05:33 <lardman> zap: just from planar to interleaved form
May 17 22:05:38 <zap> YV12 to YUV420?
May 17 22:05:45 <lardman> yep
May 17 22:05:49 <zap> aha
May 17 22:06:12 <ssvb> lardman: maybe DSP just fails to correctly write to some area
May 17 22:06:31 <lardman> zap: http://lists.maemo.org/pipermail//maemo-developers/2008-May/017705.html
May 17 22:06:41 <lardman> ssvb: perhaps, but the RGB output looks fine
May 17 22:06:56 <lardman> i.e. take the Y plane and output as a single RGB colour
May 17 22:08:08 <KotCzarny> lardman: do the colordcube
May 17 22:08:09 <KotCzarny> :)
May 17 22:08:41 <kulve> about the gst dsp plugins I mentioned: https://bugs.maemo.org/show_bug.cgi?id=2271
May 17 22:08:45 <lardman> well I did red, then green then blue, then got bored ;)
May 17 22:08:50 <kulve> (without the s in http)
May 17 22:09:04 <kulve> for some reason I can get it downloaded atm, I just get timeout
May 17 22:09:07 <KotCzarny> lardman: how about rgb pixels?
May 17 22:09:14 <zap> SHBUF_LENGTH is 384k, which is 800x480x1 byte per pixel -- isn't this too little?
May 17 22:09:22 <KotCzarny> well pixel red, pixel green, pixel blue
May 17 22:09:23 <KotCzarny> etc
May 17 22:09:30 <lardman> zap: DSP code? 16bit char
May 17 22:09:36 <zap> %-O
May 17 22:09:38 <KotCzarny> with brightness depending on y+x
May 17 22:09:49 <zap> DSP has 16-bit wide char type?
May 17 22:09:55 <kulve> zap: yes :)
May 17 22:09:59 <zap> omg :)
May 17 22:10:15 <lardman> KotCzarny: could do, but wasn't too bothered. I know the data are ok as I can display them, the conversion from YV12->YUV420 is the reason to do it
May 17 22:10:22 <kulve> zap: it's nice to communicate between the two when they don't even agree on the length of a byte (in bits)
May 17 22:10:31 <kulve> zap: sizeof(char) is still 1 in both
May 17 22:10:35 <lardman> zap: That's what makes it so much fun ;)
May 17 22:10:39 <lardman> yes
May 17 22:10:46 <KotCzarny> lardman, encode, then decode that pattern then
May 17 22:10:53 <zap> hmm
May 17 22:10:56 <KotCzarny> and maybe you spot the error :)
May 17 22:11:27 <lardman> KotCzarny: what I plan to do is to return the YUV420 framebuffer and take a look at it - that should indicate some sort of problem with my algo
May 17 22:11:42 <lardman> exactly
May 17 22:11:43 <zap> okay, but in yv12_convert_mmap_info SHBUF_LENGTH is bytes?
May 17 22:12:33 <lardman> hang on, just finding the code
May 17 22:13:10 <lardman> 16bit bytes, yes
May 17 22:15:39 <zap> Couldn't it be 8bit bytes? As far as I understand, the memory is mapped by the ARM core
May 17 22:16:11 <lardman> no, it's in terms of the DSP's byte size when it's running on the DSP
May 17 22:16:16 <zap> in which case you would get just half of the buffer with the right data, and the second half with random garbage (what you are seeing with U & V)
May 17 22:16:25 <KotCzarny> hmm
May 17 22:16:27 <KotCzarny> lardman
May 17 22:16:33 <lardman> in any case, the buffer is large enough as I get the data through
May 17 22:16:56 <KotCzarny> but that uv double sizedness could be caused by some mismatch in data length?
May 17 22:17:01 <lardman> otherwise I'd expect to lose the UV planes
May 17 22:17:10 <toresbe> Gracana: sorry, battery ran out
May 17 22:17:14 <kulve> zap: it's not about where the memory is but who's using it. When you read the same memory in ARM, the byte is 8bit and in DSP it's 16bit
May 17 22:17:18 <KotCzarny> then maybe you are feeding wrong part of the 16bit byte ?
May 17 22:17:32 <lardman> KotCzarny: that's the big question
May 17 22:17:33 <KotCzarny> because you only get bright or dark in green
May 17 22:17:52 <lardman> KotCzarny: but most of the image looks ok, so it's a bit strange that there are only certain regions with glitches
May 17 22:18:02 <KotCzarny> lardman, look at it as 2 images
May 17 22:18:06 <KotCzarny> y plane and uv plane
May 17 22:18:06 <KotCzarny> :)
May 17 22:18:44 <zap> by the way, what DSP is it? Is it from that TI DSP with VLIW architecture and 256-bit instructions for 4 parallel pipelines?
May 17 22:18:52 <lardman> Well the actual plane data are fine (in the file); I should probably try to plot the U/V planes on the DSP (though I'd expect to lose the ends if there were some issue)
May 17 22:19:11 <zap> TI C5xxx if I remember correctly
May 17 22:19:12 <kulve> zap: c55xx
May 17 22:19:12 <lardman> and otherwise I'm turning from planar into interleaved format, so hard to look at individial planes then
May 17 22:19:16 <zap> aha
May 17 22:19:23 <lardman> TMS320C55x
May 17 22:19:31 <lardman> xx
May 17 22:20:12 * lardman wasn't blowing kisses
May 17 22:20:20 <KotCzarny> :)
May 17 22:20:31 <toresbe> Gracana: thank you! Downloading now.
May 17 22:20:32 <ssvb> lardman: one random idea, maybe some right shifts are actually signed in the code
May 17 22:21:18 <lardman> ssvb: ah, that's worth a look
May 17 22:21:23 <ssvb> lardman: in this case you will get 0xFF instead of 0x00 in high order byte for shifts of negative word
May 17 22:23:00 <ssvb> lardman: that could explain some 'flat' areas, as the data would be set to 0xFF after 'or' operation
May 17 22:23:32 <lardman> let me have a quick look and see if I can spot an error
May 17 22:24:46 <lardman> ok, well some of the shifted variables are not of unsigned type, so that may be a good indication
May 17 22:24:52 <lardman> i.e. the counter vars
May 17 22:25:17 <zap> what are Uns and LgUns types?
May 17 22:25:34 <lardman> Unsigned int (16bit), unsigned long (32bit)
May 17 22:25:54 <zap> ty
May 17 22:26:57 <lardman> just recompiled, will do a test
May 17 22:27:01 <ssvb> lardman: could TI toolchain do something funny like treating ((unsigned short)some_value & 0xFF00) expression as a signed value?
May 17 22:27:17 <lardman> ssvb: almost certainly :)
May 17 22:27:25 <lardman> ssvb: does some wierd things iirc
May 17 22:27:27 <ssvb> lardman: in this case adding some more explicit cast before shift could help
May 17 22:28:04 <lardman> let me try with just setting the counter vars to unsigned type - there's one shift of one of those iirc
May 17 22:29:55 <ssvb> lardman: I'm worried about expressions like this: ((*(Uns *)((LgUns)U + (LgUns)(buf_y>>1)*width/2 + (buf_x>>1)) & 0xFF00)>>8)
May 17 22:30:18 <lardman> my tweak hasn't worked
May 17 22:30:53 <ssvb> lardman: changing them to ((Uns)(*(Uns *)((LgUns)U + (LgUns)(buf_y>>1)*width/2 + (buf_x>>1)) & 0xFF00)>>8) might help
May 17 22:31:12 <zap> can't get what's the difference between if (~(buf_y & 1)) and else parts
May 17 22:31:38 <lardman> zap: U or V planes on odd/even lines
May 17 22:31:43 <zap> ah
May 17 22:32:09 <lardman> ssvb: so add a cast to (Uns) ?
May 17 22:32:26 <ssvb> lardman: yes
May 17 22:32:47 <lardman> will try
May 17 22:33:08 <ssvb> lardman: and probably add more brackets to be completely sure that the value on the left from '>> 8' is unsigned
May 17 22:33:19 * lardman is a muppet, re-compiled the code, but didn't upload it
May 17 22:34:10 <ssvb> lardman: one more quick test would be to replace '|' with '+' and check if the distortion pattern changes
May 17 22:34:23 <lardman> ok, tried the cast and no difference
May 17 22:34:35 <lardman> will try the | -> +
May 17 22:34:39 <KotCzarny> hmm
May 17 22:34:59 <KotCzarny> write a simple code that will printf the result?
May 17 22:35:21 <lardman> lots of data to output though
May 17 22:35:52 <KotCzarny> write a loop to check for expected values too?
May 17 22:36:15 <lardman> ssvb: adding I get snow, but the large regions of green are still there
May 17 22:36:27 <lardman> in the same locations
May 17 22:36:52 <ssvb> lardman: one more alternative is to do ((x >> 8) & 0x00FF) instead of ((x & 0xFF00) >> 8)
May 17 22:37:14 <zap> can you try to write constants instead of U & V then
May 17 22:37:31 <zap> if the green areas are there, the output region seems somehow masked
May 17 22:37:36 <ssvb> lardman: does C55x support unsigned right shift? it might be some feature of hardware
May 17 22:38:13 <lardman> not sure to be honest
May 17 22:39:31 <lardman> tried  ((x >> 8) & 0x00FF) instead of ((x & 0xFF00) >> 8; same output
May 17 22:40:13 <lardman> I think it's a case of feeding some known data and looking at the output to compare where it goes wrong
May 17 22:40:59 <KotCzarny> print[fk] fun :)
May 17 22:41:51 <lardman> KotCzarny: not even that nice, dbg() which outputs to dmesg and only takes <=16bit values
May 17 22:42:24 <lardman> I can just do a copy of the framebuffer before I call the update and send it back as a bulk transfer, write it to file and use a hex editor
May 17 22:42:32 <ssvb> lardman: stupid question, have you replaced all the occurences of similar code and not edited commented out block?
May 17 22:43:18 <lardman> ssvb: what do you mean?
May 17 22:43:33 <ssvb> lardman: changing '|' to '+' should have no effect on image if high order byte was ok
May 17 22:43:52 <lardman> hmm
May 17 22:44:41 <ssvb> lardman: I just asked you to check for typos and if all the needed places were updated properly
May 17 22:45:37 <lardman> I made 6 changes - on lines 351, 354, 357 & 367, 370, 373
May 17 22:45:46 <lardman> I think that's all there was to do..?
May 17 22:47:43 <Capn_Fish> johnx: You here?
May 17 22:48:46 <ssvb> lardman: sorry, one more funny thing is '|', '+' and '<<' operator precedence
May 17 22:49:37 <ssvb> lardman: so actually changing '|' to '+' can have effect on code, but does not explain the original problem
May 17 22:49:46 <lardman> np, I'm glad of suggestions
May 17 22:50:02 <lardman> I'll have to look in the compiler manual to see if it's got some strange thing
May 17 22:50:05 <lardman> or more brackets
May 17 22:51:07 <KotCzarny> just check basic ops
May 17 22:51:14 <KotCzarny> :)
May 17 22:53:34 <lardman> yow, changing my casts for the calcs for the offsets in the buffers from lgUns to Uns has screwed it up completely
May 17 22:53:53 <lardman> keeping the actual buffer cast as LgUns that is
May 17 22:58:36 <lardman> hmm, now by casing all of the variables inside the *() operators to LgUns, I've ended up with the same image + 2 shadowns, + some noise outside of the video area
May 17 22:58:45 <lardman> casting that should read
May 17 22:59:24 <KotCzarny> so that's an improvement
May 17 22:59:32 <KotCzarny> :)
May 17 22:59:39 <lardman> hmm
May 17 22:59:54 <lardman> so perhaps I was just lucky first time thought, the errors cancelled out :)
May 17 22:59:54 <KotCzarny> and indicates that you have to check all the operations and variables
May 17 23:00:01 <lardman> yep
May 17 23:01:18 <lardman> hmm, I wonder if something is overflowing here
May 17 23:01:39 <lardman> the bottom half of the image is half mirrored up the top - in stripes
May 17 23:02:20 <KotCzarny> maybe your assumption about display layout is wrong?
May 17 23:05:28 <lardman> perhaps, but it's corruption rather than a lack of image
May 17 23:06:55 <lardman> http://people.bath.ac.uk/enpsgp/nokia770/dsp/ jpgs nos 54 & 56, still uploading
May 17 23:08:29 <KotCzarny> hmm
May 17 23:08:31 <KotCzarny> funny
May 17 23:09:30 <KotCzarny> definitely, you should feed it with a test screen
May 17 23:09:40 <KotCzarny> you can create mpeg from a single image
May 17 23:09:42 <KotCzarny> if you want
May 17 23:10:12 <lardman> I'll create an artificial one with set values for the Y, U & V planes
May 17 23:10:48 <KotCzarny> You don't have permission to access /enpsgp/nokia770/dsp/PICT0056.JPG on this server.
May 17 23:10:49 <KotCzarny> :)
May 17 23:10:59 <lardman> ah, usual story, hang on a tick
May 17 23:11:48 <lardman> try again
May 17 23:12:03 <lardman> FAT fs gives execute permissions to the photos
May 17 23:12:40 <KotCzarny> is it me or the whole picture became smaller than before?
May 17 23:12:56 <lardman> it's about the same size
May 17 23:13:04 <KotCzarny> the picture on n800
May 17 23:13:17 <lardman> yeah, about the same afaict
May 17 23:13:22 <KotCzarny> right now it's about 1/8 of the screen
May 17 23:13:28 <KotCzarny> before it was 1/4
May 17 23:13:46 <lardman> should be 320/800 of the screen
May 17 23:13:55 <lardman> and 240/480 tall
May 17 23:14:06 <lardman> hmm
May 17 23:14:18 <KotCzarny> just compare the pic to 0050
May 17 23:14:39 <KotCzarny> ahm, sorry
May 17 23:15:14 <lardman> I think it's the same, though it's not very tall - nor was it before
May 17 23:15:31 <KotCzarny> but still, it doesn't look as 320x240
May 17 23:15:42 <KotCzarny> 240 should take half of the height
May 17 23:15:50 <lardman> it's probably 320 wide, but is ~120 high
May 17 23:16:07 <lardman> urgh, far more debugging to be done
May 17 23:17:24 <KotCzarny> try rewriting the code without optimizations :>
May 17 23:17:37 <lardman> lol, as if there are any in there :D
May 17 23:17:59 <KotCzarny> :)
May 17 23:18:13 <KotCzarny> any & and << could be treated as such ;)
May 17 23:18:35 <lardman> there's no other way to arrange the data though
May 17 23:18:53 <lardman> have to manipulate 8bit chunks of 16bit addressable data types
May 17 23:19:32 <KotCzarny> don't use 8bit type then?
May 17 23:19:56 <jdiaz> hi
May 17 23:19:56 <infobot> hey
May 17 23:20:32 <lardman> KotCzarny: have to address 8bit chunks for the fb format
May 17 23:20:42 <jdiaz> I've got a problem with the internal 2G flash  (n810)
May 17 23:20:54 <KotCzarny> jdiaz: recreated partition table?
May 17 23:21:11 <jdiaz> fsck.vfat /dev/mmcblk0p1
May 17 23:21:33 <jdiaz> malloc:Input/output error
May 17 23:21:33 <KotCzarny> lardman: you need simpler algo
May 17 23:21:35 <KotCzarny> :)
May 17 23:21:51 <lardman> well rgb works, that's simple
May 17 23:21:53 <KotCzarny> jdiaz: corrupted partition table, maybe?
May 17 23:21:55 <jdiaz> related to the known bug?
May 17 23:22:01 <lardman> but the point of yuv420 is to maximise bandwidth
May 17 23:22:44 <KotCzarny> lardman: maybe it's a weird yuv420
May 17 23:23:26 <jdiaz> I have several files on it. all ok except the last two, wich are unreadable
May 17 23:24:01 <lardman> KotCzarny: it is :)
May 17 23:24:06 <Gracana> yw toresbe. does it work?
May 17 23:24:12 <KotCzarny> jdiaz: can you paste /sbin/sfdisk -l to pastebin.com ?
May 17 23:24:50 <lardman> bed time, thanks for the ideas chaps
May 17 23:24:54 <KotCzarny> :)
May 17 23:25:06 <lardman> I'll try feeding the known data tomorrow and see what we get out :)
May 17 23:25:34 <lardman> regarding sbc, it seems to receive and process data (not sure how much yet, need to count the messages output), but it dies with a bus error
May 17 23:25:52 <lardman> this kills the arm-side code too, and consequently the output file doesn't get written
May 17 23:25:54 <lardman> annoying
May 17 23:26:40 <lardman> is it possible to flush?
May 17 23:28:47 <KotCzarny> open it as unbuffered?
May 17 23:28:55 <jdiaz> http://pastebin.com/m38c508a6
May 17 23:29:01 <lardman> can that be done with fopen()?
May 17 23:30:28 <KotCzarny> lardman: with open there are some flags
May 17 23:30:49 <lardman> KotCzarny: I'll take a look tomorrow, thanks
May 17 23:30:52 <lardman> night chaps
May 17 23:31:07 <KotCzarny> gah
May 17 23:31:12 <jdiaz> now I'm remembering other strange behaviours my n810 had today: an alert box (never seen before) told me that in order to open an application, another should be closed (I only had two xterm and a vnc, as habitual)
May 17 23:31:26 <GeneralAntilles> Out of memory?
May 17 23:31:40 <GeneralAntilles> I'm assuming you had virtual memory off?
May 17 23:31:51 <KotCzarny> jdiaz: your partition table is factory broken
May 17 23:31:53 <KotCzarny> i think
May 17 23:31:59 <KotCzarny> Disk /dev/mmcblk0: 61440 cylinders
May 17 23:32:04 <jdiaz> the Image Viewer reported me an error about being unable to show the photo due to inssuficient memory
May 17 23:32:05 <KotCzarny> #
May 17 23:32:05 <KotCzarny> Disk /dev/mmcblk1: 62304 cylinders, 4 heads, 16 sectors/track
May 17 23:32:05 <KotCzarny> #
May 17 23:32:05 <KotCzarny> Warning: The partition table looks like it was made
May 17 23:32:05 <KotCzarny> #
May 17 23:32:06 <KotCzarny>   for C/H/S=*/61/60 (instead of 62304/4/16).
May 17 23:32:13 <KotCzarny> ah, sorry wrong line
May 17 23:32:16 <KotCzarny>  /dev/mmcblk0p1          0+  62719   62720-   2007032    b  W95 FAT32
May 17 23:32:18 <KotCzarny> this one
May 17 23:32:30 <KotCzarny> physical size is 61440
May 17 23:32:36 <KotCzarny> but partition is 62719
May 17 23:32:44 <KotCzarny> so corruption is inevitable
May 17 23:32:46 <jdiaz> yes, GA, no virtual memory, but not a problem like these until today
May 17 23:32:58 <GeneralAntilles> Something ate your RAM.
May 17 23:33:03 <KotCzarny> jdiaz: back up the files then delete partition and create again
May 17 23:33:12 <jdiaz> KotCzarny, I have read about that
May 17 23:33:14 <KotCzarny> otherwise you run into corruption again later
May 17 23:33:53 <jdiaz> but I thought the corruption will not occur until the partition is full enough
May 17 23:34:00 <jdiaz> mine is at 40%
May 17 23:34:07 <KotCzarny> well
May 17 23:34:25 <GeneralAntilles> It's not a hard drive, jdiaz.
May 17 23:34:28 <GeneralAntilles> It doesn't load evenly.
May 17 23:34:36 <KotCzarny> apparently it did
May 17 23:34:44 <GeneralAntilles> You could run into the out-of-range sectors at 1% or 99%
May 17 23:35:19 <jdiaz> GA, could this problem be related to the out-of-memry problems?
May 17 23:35:22 <AStorm> actually, that depends on the FS
May 17 23:35:38 <AStorm> more than the device wear-levelling
May 17 23:35:38 <KotCzarny> creating partition bigger than device isn't a wise idea
May 17 23:35:41 <KotCzarny> anyway
May 17 23:35:42 <KotCzarny> :)
May 17 23:35:45 <jdiaz> AStorm, vfat
May 17 23:35:54 <AStorm> jdiaz: then you were unlucky
May 17 23:35:58 <GeneralAntilles> The card load-levels itself.
May 17 23:35:59 <AStorm> anyway, that has to be fixed
May 17 23:36:37 <KotCzarny> also your sd card can be error prone too
May 17 23:36:51 <KotCzarny> as it has some warning
May 17 23:36:55 <AStorm> if it is a card ;-)
May 17 23:37:14 <GeneralAntilles> It's close enough.
May 17 23:37:17 <KotCzarny> astorm: /dev/mmcblk1p1 is a card
May 17 23:37:17 <KotCzarny> :)
May 17 23:37:21 <AStorm> ya
May 17 23:38:11 <jdiaz> could the flash corruption problem be related to the out-of-memory problems?
May 17 23:38:23 <AStorm> unlikely
May 17 23:38:34 <AStorm> unclean restarts or just wear
May 17 23:38:44 <jdiaz> but both occured the same day...
May 17 23:38:48 <AStorm> is it a cheap OEM imitation? ;)
May 17 23:39:05 <KotCzarny> jdiaz: fix your partitions
May 17 23:39:09 <KotCzarny> :)
May 17 23:39:16 <jdiaz> after two months of running without any out-of-mem problem...
May 17 23:39:33 <GeneralAntilles> Out of memory isn't the end of the world, jdiaz. :P
May 17 23:39:43 <GeneralAntilles> Just format the internal card, enable swap and move on.
May 17 23:39:47 <KotCzarny> oom could happen if fs was corrupted and your swap went read only
May 17 23:39:55 <KotCzarny> or inaccessible
May 17 23:39:55 <KotCzarny> :)
May 17 23:40:18 <jdiaz> I have no swap enabled
May 17 23:40:29 <KotCzarny> do you have virtual mem enabled?
May 17 23:40:38 <jdiaz> no
May 17 23:40:45 <KotCzarny> ok
May 17 23:41:06 <jdiaz> but I was able ofn running several simultaneous apps before
May 17 23:41:13 <KotCzarny> jdiaz: still, fix partitions, come back if it happens again
May 17 23:41:14 <KotCzarny> :)
May 17 23:41:15 <jdiaz> much more indeed
May 17 23:41:31 <jdiaz> ok, I'll do that. Just curious
May 17 23:42:23 <jdiaz> how can I detect wether I have an ill-behaved app who eats my mem?
May 17 23:42:30 <AStorm> well, I found some huge memory problem with maemopad
May 17 23:42:31 <KotCzarny> slowdown
May 17 23:42:33 <KotCzarny> usually
May 17 23:42:45 <AStorm> it loads all the notes into ram
May 17 23:42:49 <GeneralAntilles> Install osso-statusbar-cpu
May 17 23:42:55 <AStorm> guess how large my notebook is ;P
May 17 23:43:08 <jdiaz> top?
May 17 23:43:14 <KotCzarny> could work too
May 17 23:43:20 <AStorm> yes
May 17 23:43:26 <KotCzarny> though look at rss
May 17 23:43:27 <jdiaz> browser using 77%
May 17 23:43:58 <jdiaz> gnome-vfs-daemon using 41%
May 17 23:44:23 <jdiaz> whow, only these two make more than 100%
May 17 23:44:28 <KotCzarny> as i said, look at the rss
May 17 23:44:38 <KotCzarny> could be named 'res'
May 17 23:44:43 <AStorm> browser is huge, yes
May 17 23:45:01 <jdiaz> rss not running
May 17 23:45:02 <KotCzarny> other numbers could include shared mem and libs
May 17 23:45:11 <KotCzarny> rss is not a program
May 17 23:45:12 <jdiaz> mmm
May 17 23:45:14 <KotCzarny> byt acronym
May 17 23:45:18 <KotCzarny> real something size
May 17 23:45:23 <jdiaz> oh
May 17 23:45:41 <jdiaz> I though you refered to the rss reader
May 17 23:45:45 <KotCzarny> no
May 17 23:46:00 <AStorm> Resident Set Size
May 17 23:46:09 <AStorm> aka "what's in memory"
May 17 23:46:41 <KotCzarny> ahm
May 17 23:46:42 <KotCzarny> probably
May 17 23:46:53 <KotCzarny> thanks for explanation :)
May 17 23:46:53 <infobot> KotCzarny: my pleasure
May 17 23:47:04 <KotCzarny> someone please shut up the infobot
May 17 23:47:13 <jdiaz> mmm... if the filesystem is corrupt, couldnt that cause gnome-vfs to consume inusual amounts of memory?
May 17 23:47:33 <KotCzarny> if your system is corrupted anything can happen :)
May 17 23:47:34 <AStorm> infobot: you suck!
May 17 23:47:35 <infobot> Sorry!
May 17 23:47:35 <GeneralAntilles> Corrupt filesystems cause all sorts of trouble.
May 17 23:47:48 <GeneralAntilles> Format the internal card and THEN try and troubleshoot. ;)
May 17 23:47:52 <KotCzarny> no
May 17 23:48:00 <KotCzarny> remove partition, and create it again
May 17 23:48:05 <KotCzarny> THEN format
May 17 23:48:08 <AStorm> yes, format isn't enough
May 17 23:48:16 <AStorm> need to repartition it
May 17 23:49:25 <jdiaz> thanks, I'll do that
May 17 23:49:47 <GeneralAntilles> File manager fixes it just fine.
May 17 23:49:51 <Gracana> http://gizmodo.com/391108/samsungs-display-division-shows-just-what-it-can-do-with-amoled-notebook Nokia 770!
May 17 23:50:01 <KotCzarny> ga: sure, for a time
May 17 23:51:25 <Navi> Gracana, looks expensive
May 17 23:52:35 <KotCzarny> kb would suck
May 17 23:52:39 <AStorm> oleds are heavier
May 17 23:52:51 <AStorm> although much brighter
May 17 23:53:03 <Gracana> Yeah, definitely. I don't think I would like it. But the fact that the person on the screen is holding another company's product is amusing.
May 17 23:53:04 <AStorm> ah, and they may use more power too
May 17 23:53:42 <AStorm> I wonder how oleds look in direct sunlight
May 17 23:54:10 <Navi> Gracana, indeed
May 17 23:54:18 <Navi> Gracana, it might be a taunt of some sort
May 17 23:54:23 <KotCzarny> in direct sunlight color e-ink would rule
May 17 23:54:36 <Navi> Gracana, wait, that doesn't look like a 770 :P
May 17 23:54:43 <AStorm> KotCzarny: too low contrast, naah
May 17 23:54:51 <KotCzarny> astorm: future e-ink
May 17 23:54:53 <KotCzarny> :)
May 17 23:54:59 <AStorm> future, well, yeah
May 17 23:55:02 <KotCzarny> with a 225 dpi etc
May 17 23:55:05 <AStorm> gimme a holographic display
May 17 23:55:05 <Gracana> oh yeah Navi, it's backwards and has no buttons
May 17 23:55:10 <KotCzarny> nah
May 17 23:55:19 <AStorm> N810 with a holographic display... ;P
May 17 23:55:30 <AStorm> close to that future Nokia product
May 17 23:56:24 <Navi> heh
May 18 00:09:45 * Navi takes his delorean to the future and brings back some holographic displays for everyone
May 18 00:10:26 <Navi> Of course, I forgot to bring the DVI to SUED converters, so none of you can use it
May 18 00:12:54 <GeneralAntilles> Asshole.
May 18 00:13:18 <Navi> I'm sorry
May 18 00:14:35 <GeneralAntilles> No you're not.
May 18 00:14:45 <Navi> am too
May 18 00:27:22 <jdiaz> flash repartitioned
May 18 00:27:34 <jdiaz> restoring contents ...
May 18 00:29:24 <jdiaz> (boring)
May 18 00:30:15 <jdiaz> good opportunity to delete all those .ogg files from ./map
May 18 02:51:11 <zakm> hi . . i have a nokia 770 .. WSOD .. i tried reflashing and taking out battery and nothing helped ( bought off ebay .. WSOD after like two weeks. no warranty )
May 18 02:51:27 <zakm> anyone have any ideas of what i could try.. willing to try anything
May 18 02:51:36 <GeneralAntilles> Controller needs to be replaced/reset.
May 18 02:51:47 <zakm> describe reset?
May 18 02:51:54 <GeneralAntilles> reseated, rather.
May 18 02:52:04 <zakm> is it solder'd?
May 18 02:52:09 <zakm> or just pushed in?
May 18 02:52:36 <zakm> so i open it up.. "take out" controller.. put it back in? thats it?
May 18 02:54:12 <johnx> I don't have a 770, but this thread may have some of the info you want: http://www.internettablettalk.com/forums/showthread.php?t=15599
May 18 02:54:38 <zakm> ive looked at the wsod thread if thats the one you've linked?
May 18 02:54:50 <johnx> there are many
May 18 02:55:02 <johnx> this one talks about what the controller chip is and where it is
**** ENDING LOGGING AT Sun May 18 02:59:57 2008