irclog2html for #openzaurus on 20060620

00:04.22*** join/#openzaurus syn-- (i=cryptic1@d57-94-35.home.cgocable.net)
00:15.21*** join/#openzaurus syn- (i=unknown@d57-94-35.home.cgocable.net)
01:03.38*** join/#openzaurus synth (n=cb@c-71-203-15-223.hsd1.fl.comcast.net)
01:47.09*** join/#openzaurus hvontres (n=henry@adsl-75-7-228-10.dsl.sndg02.sbcglobal.net)
02:01.28*** join/#openzaurus Timelord (n=TL@4.78.4.43)
02:25.20*** join/#openzaurus Timelord0 (n=TL@4.78.4.43)
02:44.27*** join/#openzaurus trogbox (n=user@net34-198.netkaster.ca)
03:02.04*** join/#openzaurus hwtechnik (n=hwtechni@tor/session/external/x-ba83e456e3fd4dfd)
04:51.39*** join/#openzaurus carbonfreeze (n=carbonfr@rdbck-320.wasilla.mtaonline.net)
05:07.25*** join/#openzaurus mithro (n=tim@203.122.246.117)
05:12.57*** join/#openzaurus sdjf (n=sdjf@63.93.103.40)
05:22.24*** part/#openzaurus sdjf (n=sdjf@63.93.103.40)
05:26.33*** join/#openzaurus magnet_ (n=magnet@lns-bzn-28-82-250-174-203.adsl.proxad.net)
05:26.37*** join/#openzaurus _law_ (n=law@mail.stiftadmont.at)
05:29.41*** join/#openzaurus pgas (n=user@ios.intranet.GR)
05:39.03*** join/#openzaurus JiniDog (n=jini@V98a1.v.pppool.de)
06:20.34*** join/#openzaurus hufnus (n=slonsiki@DSL135-086.labridge.com)
06:33.15*** join/#openzaurus lardman (n=lardman@138.38.73.178)
06:34.29*** join/#openzaurus redguy (n=mati@83.25.99.152)
06:37.52*** join/#openzaurus lardman_ (n=lardman@138.38.73.178)
07:07.51*** join/#openzaurus alan|home (n=alan@ARouen-151-1-74-156.w86-215.abo.wanadoo.fr)
07:08.12*** join/#openzaurus lardman_ (n=lardman@138.38.73.178)
07:08.33alan|homegood morning
07:35.03*** join/#openzaurus JiniDog (n=jini@V89e0.v.pppool.de)
07:35.05*** join/#openzaurus hrw|gone (n=hrw@host-ip170-158.crowley.pl)
07:35.44*** join/#openzaurus sdjf (n=sdjf@63.93.103.40)
07:38.50hrw|workmorning
07:39.56pgashello
07:40.30*** join/#openzaurus sdjf (n=sdjf@63.93.103.40)
07:52.31*** join/#openzaurus eumel (n=chatzill@p54820A69.dip0.t-ipconnect.de)
08:00.41*** join/#openzaurus kolla (i=kolla@halbrend.uninett.no)
08:50.53*** join/#openzaurus redguy (n=mati@83.25.99.152)
09:12.38carbonfreezehow are the 3.5.4.1 builds coming along? I noticed OE tinderbox is offline
09:13.25XorAI think hrw|work is currently melting cpu's
09:13.38carbonfreezehehe
09:14.37hrw|workewi does not respond currently
09:14.53hrw|workcarbonfreeze: ~50% is built
09:14.54carbonfreezeI've never messed around with tinderbox, would there be any way for me to contrib some cpu for the builds?
09:15.00XorAhrw|work: you finally killed it through overwork
09:15.53hrw|workI will kill e17 probably
09:31.52lardman|workmorning
09:32.08*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
09:37.08*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
09:42.09*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
09:44.57lardman|workhrw|work: Do you have upgrades building abilities atm, or are you flat out with the 3.5.4.1 builds?
09:47.09*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
09:50.48hrw|worklardman|work: I can build upgrades too
09:51.11hrw|worklardman|work: but first has to get into ewi (which looks like overloaded too much)
10:27.02lardman|workhrw|work: I was after spca5xx if you have time
10:27.11lardman|workhrw|work: For the c7x0
10:27.34lardman|workRP: ping
10:27.55*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
10:27.58lardman|workAnyone here running gpe atm?
10:28.21hrw|worklardman|work: you know where to req this
10:28.27lardman|workI flashed the 3.5.4.1rc4 image and was fiddling about, and there are issues afaict (though they may be my fault)
10:28.35lardman|workhrw|work: Indeed ;)
10:33.12*** join/#openzaurus mrz80 (i=1000@caledonia.cns.ufl.edu)
10:35.11*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
10:37.47lardman|workAnyone running gpe here?
10:38.30lardman|workDo you find the following? In Look n Feel, the font size text box can only take one digit - so font sizes >9 don't fit
10:38.51lardman|workThe Font name doesn't fit in the button, it's cut in half vertically
10:38.57lardman|work(same page as above)
10:39.31lardman|workThe example terminal font, is too large - presumably this is because the terminal doesn't handle the dpi settings correctly or somesuch
10:39.56lardman|workI also fell into the age old trap of not doing rm -rf ~/*
10:40.11*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
10:40.41XorAlardman|work: heh heh
10:40.41lardman|workbut when I did that after my first login, the ~/.gpe/settings db isn't created, so gpe-confd doesn't want to run, so look n feel doesn't actually change anything
10:41.03lardman|workAnyone know what should create that file?
10:41.09XorAlardman|work: I run gpe and I put my dpi back to 96 like gtk expects and never had problems anymore
10:41.10lardman|workXorA: Quite :)
10:41.41lardman|workXorA: How did you change it? This is a bit of an issue for releases IMO
10:41.53XorA.Xdefaults
10:41.56lardman|workI guess we just need gpe to handle the dpi stuff correctly
10:41.58XorAXft.dpi: 96
10:42.01lardman|workXorA: Thanks
10:43.36lardman|workAnyone know where to get a cheap(ish) battery for a c7x0 (large or small) as mine now hovers around the 20% mark all the time
10:43.57lardman|workwhich leads me onto wanting to talk to RP... ;)
10:49.22*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
10:51.50lardman|workRP: I've been seeing the stroboscopic effect of low battery and kernel backlight control code oscillating around the limiting percentage
10:52.20hrw|worklardman|work: nokia 7110 + some wires?
10:52.31lardman|workRP: Is this an unusual occurrance - as I have a knackered battery? Or is it worth putting something in to limit the speed of this?
10:52.50lardman|workhrw|work: Yeah, I was hoping I could just buy a proper one for decent money by now
10:53.27hrw|worklardman|work: tried brando? laptopforless?
10:53.31XorAlardman|work: I get that effect
10:53.52XorAlardman|work: and my battery is functional
10:53.56hrw|workhttp://shop.brando.com.hk/zaurusea_bl08k.php
10:54.09lardman|workXorA: I was thinking either having asymmetrical limits, or a timer might be better
10:54.09hrw|workUS$87.00 - expensive?
10:54.25hrw|workUS$62.00 for small one
10:54.42lardman|work£47
10:54.48lardman|worknot too bad I guess
10:54.55XorAlardman|work: Im guessing its due to the battery level indication being affected be the drain rate
10:55.22hrw|workhttp://www.pdainternalbattery.com/sharppdabattery.html cheap
10:55.34lardman|workXorA: The battery recovers slightly when the backlight is turned down, so it switches back on, and drops, et cetera
10:56.01lardman|workhrw|work: Nice :)
10:56.29hrw|worklardman|work: look at collie picture there
10:56.42lardman|workhmm
10:57.09lardman|workWill the 5600 battery fit the c7x0?
10:57.22hrw|workyes
10:57.25XorAlardman|work: that looks like my c860 battery
10:57.26hrw|workits the same one
10:57.37XorAneed extended back though?
10:57.40lardman|workyes
10:57.50lardman|workany ideas where to get a cheap one of those then :)
10:57.51hrw|workbl-06 = collie, c700, c750. bl-08 = poodle, c760, c860
10:58.09hrw|worklardman|work: brando add it to battery (as sharp do)
10:59.10lardman|workBut there's a lot of difference between $15 and $87 for a battery
10:59.28hrw|workl-f-l are assholes
10:59.36hrw|workdo not ship to Poland
11:01.14lardman|workhttp://cgi.ebay.co.uk/Sharp-Zaurus-SLC1000-SLC3000-SLC3100-SL-5600-Battery_W0QQitemZ8821745992QQcategoryZ75446QQrdZ1QQcmdZViewItem
11:01.56hrw|workhmm.. I was thinking that cxx00 use BL-11 not BL-08
11:02.09lardman|workhrw|work: That's what I thought too
11:02.14lardman|workXorA: Join the club
11:05.01lardman|workhttp://conics.net/shp/pda/zaurus-sl-c700/images/sl-acc.html
11:05.16lardman|workSo the cxx00 batteries fit the other clamshells too?
11:05.26lardman|workLooking at the description for the EA-BL11
11:09.00hrw|workask RP - he has both
11:13.50lardman|workConics do the cover for $25
11:13.59lardman|workMore expensive than the battery :D
11:14.05lardman|workat lfl
11:39.08RPlardman|work: Its possible your battery is ok and the power management code is at fault. I also see the strobe effect :-/
11:39.50lardman|workRP: My battery is pretty old, so I wouldn't bet on it
11:40.04lardman|workRP: Plus the strobe effect makes logical sense
11:40.31RPWe probably need to damp the osciallations...
11:40.55hrw|workI like them ;)
11:41.06lardman|workYes, I was wondering whether to make the transition thresholds asymmetrical
11:41.17RPWe could call it a feature :)
11:41.19lardman|workhrw|work: It was like a mini disco last night ;)
11:41.43lardman|workFeature for those who aren't epileptic!
11:42.10RPlardman|work: Since you seem to be able to resproduce, can you tell me how much the battery voltage jumps with the backlight lowered?
11:42.31lardman|workRP: I can test
11:42.41lardman|workRP: But it'll have to be this evening
11:43.10RPlardman|work: no rush. It might be a while before I can patch it anyway. It'd be handy to have the data though
11:44.07lardman|workIt was oscillating around APM_BATTERY_STATUS_LOW I presume, and that's set from the lookup table isn't it?
11:44.33RPlardman|work: I think so
11:44.55lardman|workIt was about 20% anyway
11:45.36RPWe could probably have the backlight limiter locked on until it saw AC power thinking about it
11:46.16RPand on AC power, it should do that but does
11:46.43lardman|workor just set the transition from low to high about 3% higher than the transition from high to low power - though that may require a more complex bit of code
11:46.47hrw|workRP: what about printk() on screenflash?
11:47.34RPhrw|work: How many users would see it?
11:48.17hrw|workRP: 'if you have screenflashes right after charged battery then send dmesg output to us'
11:48.40hrw|workRP: automatic gathering of debug infos
11:48.44lardman|workJust means your battery is screwed
11:49.06lardman|workMine went from 100 to 40% in less than a minute (admittedly after a very short charge)
11:49.24hrw|workmine bl-06 can jump from 100 to 60
11:49.29hrw|workI'll look now at it
11:50.21lardman|workThe jump happens as the voltage relaxes after the charging, this is why Sharp had their odd table which stayed at 100% for ~1/2 the battery voltage
11:50.38lardman|workTo hide knackered batteries
11:51.12lardman|workLunch time, will bring my Zaurus in when I get back
11:51.19poushagtransition thresholds should always be offset a bit - its SOP for controls -
11:51.48hrw|workmy bl-06 has 88% after unplug AC and boot
11:52.39poushagits called deadband
11:52.47RPThey do take a long time to charge. Just because it says its 100%, doesn't mean its fully charged...
11:52.54RPThe charge light is a better indicator
11:53.20poushagand without it the climate control in your house and car would be radiply cycling on and off too
11:56.58poushag(as a professional controls engineer i thought my input might be of value)
11:58.41poushagsomething strange is happening to my Z - has anyone seen konqeror quit working?  because on my Z i cant get it to run anymore (or if it does run then it disappears so quickly i cant even see it)
11:59.18poushagi tried reinstalling it but no change in symptoms
12:00.07hrw|workrun it in opie-console
12:00.52poushagyeah i was going to try it in konsole - is that the same?
12:01.55hrw|workany terminal
12:02.52poushagmissing libpcreposix.so.0
12:02.58poushagwhere do i get it?
12:03.17hrw|workyou have collie or other machine and used ipkg-link?
12:03.22poushagi know theres and ipkg command i can use to find out....
12:03.31poushagyes
12:04.06poushagipkg file libpcreposix.so.0   ?
12:04.26hrw|workit is in pcre package iirc
12:07.20poushagwhich feed?  base?
12:07.55hrw|workyes
12:08.10hrw|workbut ewi (machine with feeds) is overloaded now and does not reply to connections
12:08.25poushagcrap
12:17.50poushagewi wont even answer a ping - has the admin been notified?
12:25.46hrw|workyes but he is >1000 km from machine so we need to wait a bit for resetmonkey
12:42.23*** join/#openzaurus frank7d (n=afs@91.2.3.213.fix.bluewin.ch)
13:37.09hrw|work"I've been using OZ/GPE for about a week and a bit now. I feel that this meets more of my needs than pdaXrom does in it's current form." - I like such posts ;)
13:39.10poushagcongrats - it is good to be appreciated
13:39.34poushaghrw:  did you gain any more insight into the dagger issue since yesterday?
13:39.34magnet_a lots of people migrated from pdaxrom to oz.
13:40.04magnet_pdaxrom is often the only alternative to sharp's os, that shops offer.
13:40.33poushagshops?  you have zaurus shops where you live?
13:40.40magnet_even on the website where I bougth it, screenshots where from pdaxrom.
13:40.44magnet_online shops :)
13:40.56magnet_I live in france, there is no zaurus there.
13:41.30poushaginteresting - i thought if you bought a Z you automatically got Sharp ROM
13:41.53magnet_for a small fee, I could have pdaxrom pre installed. like ~25 euros.
13:41.59poushagi guess you found a reseller somewhere
13:42.15magnet_yes.
13:42.16pgasIn fact I only know about 1 shop where you can ask for an alternative rom
13:42.36poushagsorry to be such a yank - but whats that in USDollars?
13:42.37hrw|workpoushag: no
13:42.59hrw|work~change 25 eur to usd
13:43.23poushagthanks ibot
13:43.25magnet_I see that it's free now.
13:43.27magnet_http://cgi.ebay.fr/Ordinateur-portable-PDA-Sharp-Zaurus-SL-C3200_W0QQitemZ9741280813QQcategoryZ8131QQrdZ1QQcmdZViewItem
13:43.37pgasAbout the number of users of this or that.....I don't know if anybody has any data that are even remotely accurate
13:43.38magnet_in french, but shots at the bottom show pdaxrom
13:44.36poushagi suppose at the package server they could log ip addresses that are downloading from it and then count them
13:44.59poushag(after the server is back online that is!)
13:45.33poushagi dont think you can just count how many times the ROM has been DLed
13:45.34pgasyeah, but I feel a lot of people are trying alternative roms from time to time and go back to cacko afterwards
13:46.00poushagi mean i just DLed a bunch of ROMs to a CF card but I'm only using OZ
13:46.08pgaspoushag: use one of the mirror?
13:47.36poushagi was suggesting using the ewi package server since the real OZ users seems to connect regularly for updates/upgrades etc
13:49.54poushagim going to mow the yard - ttyl
13:50.30*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
14:02.02*** join/#openzaurus joshin (i=josh@VDSL-130-13-143-17.PHNX.QWEST.NET)
14:09.16*** join/#openzaurus Xromer|work (n=Xromer@217-133-58-241.b2b.tiscali.it)
14:14.05*** join/#openzaurus alan|laptop (n=alan@ARouen-151-1-76-140.w86-215.abo.wanadoo.fr)
14:14.08alan|laptophi all
14:14.41Xromer|workHi
14:20.01lardman|workhi
14:20.37lardman|workhrw|work: Don't worry about my upgrade request on oesf, my toolchain's built so I can do it myself (plus the sources appear to have moved so I'll have to patch it)
14:25.07*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
14:28.18lardman|workWhat's the sys directory where the battery voltage and percentage stuff is exposed?
14:30.00NAbyss/devices/platform/sharpsl-pm/battery_voltage
14:30.02NAbyss./devices/platform/sharpsl-pm/battery_percentage  
14:30.11NAbysslardman|work: those dirs in sys
14:30.38lardman|workNAbyss: Thanks - I need to capture my dodgy battery performance so couldn't afford to take too long looking :)
14:30.53NAbysslardman|work: ah, not good :)
14:32.51lardman|workRP: Right, I've not added printks yet, but can do, but the oscillations occur between voltages 185 and 183 which correspond to percentages 30 and 25
14:33.07lardman|workNAbyss: No, but hopefully it'll be useful
14:33.54lardman|workhmm, those percentages seem odd...
14:34.54lardman|workall a bit strange, I've got (volt:%): 30:185, 23:182, 35:187
14:35.05lardman|work28:184
14:35.19lardman|work25:183
14:35.29lardman|workAll those percentages seem to be very close together
14:36.38poushagits nonlinear - probably since users find that % capacity is most useful if it correspinds to usage time left rather than voltage
14:36.48poushag*correponds
14:40.27CoreDump|wm06hi
14:41.57lardman|workpoushag: Ah yes, I have the lookup table in front of me in the kernel source, so can find that out
14:45.12poushaglardman:  your deadband settings can be optimized if you know the resolution of the voltage meter being used
14:46.16lardman|workpoushag: How does one optimise this then?
14:46.22poushaglardmanbasically just use a difference in percentages that correspond to twice the granularity of the sensor
14:46.25lardman|workAnyone understand what this means http://mxhaard.free.fr/download.html ?
14:46.36poushagack - fumbled on the kb
14:47.06poushaglardman: just use a difference in percentages that correspond to twice the granularity of the sensor
14:47.07lardman|workpoushag: There's an additional problem in that the backlight is switched on and off at a certain (low) voltage by the kernel, this causes the voltage to jump back up
14:47.21hrw|worklardman|work: it means that you need 2.6.16+ for new version
14:47.38lardman|workpoushag: So I need to account for the difference in voltage that the backlight produces
14:47.41hrw|worklardman|work: v4l1 is obsolete and is going to be removed sooner or later
14:47.43poushaglardman: right voltage is affected by current drain
14:48.08lardman|workhrw|work: I couldn't see any difference between them
14:48.15poushaglardman: and the diff will be bigger at lower battery charge level
14:48.36CoreDump|wm06poushag: for accus that shouldn't be the case
14:48.50lardman|workhrw|work: The 'goodbye' release is:  Spca5xx-v4l1-Goodbye module version 0.60.00 date: 01/05/2006
14:49.07lardman|workhrw|work: The only download I can see on that page has the same date: spca5xx-20060501.tar.gz
14:49.11poushagcoredump: accus?
14:49.16CoreDump|wm06hmm
14:49.25lardman|workpoushag: batteries?
14:49.43CoreDump|wm06rechargeable batteries
14:49.59hrw|worklardman|work: ask them?
14:51.28lardman|workhrw|work: Will do, just wanted to make sure I wasn't being stupid
14:51.42poushaglardman: well in any case you need to account for nonlinearities that may be present in the delta v (backlight to no backlight) at different charge levels
14:52.44lardman|workpoushag: In this case I only need to account for the difference at one percentage/voltage threshold as that's the only time the kernel intervenes - that's the way I see it anyway
14:53.05poushaglardman: because the backlight is also switched off by power management
14:54.04lardman|workpoushag: In that case, should the kernel not report the true voltage/percentage anyway?
14:54.46*** join/#openzaurus hrw|work (n=hrw@host-ip170-158.crowley.pl)
14:54.58lardman|workhrw|work: What's up with your connection?
14:55.25poushaglardman: well the voltage will be bumped up and down whenever the backlight is toggled (whether its due to low battery or apm)
14:55.55*** join/#openzaurus kolla (n=kolla@firda.kolla.no)
14:56.09poushaglardman: try one increment in your table then and see how it works
14:56.10hrw|worklardman|work: dont know  it just suxx
14:56.12lardman|workpoushag: Yes, but the userspace pm stuff only alters the backlight depending on a usage timeout, so the voltage doesn't matter
14:59.01poushaglardman:  but doesnt each act independently so that pm will inflence the v which in turn can drive low-batt turnoff (under the right circumctances)
15:00.09poushaglardman:  but maybe it doesnt matter - perhaps it wont result in cyclic behavior - i think i need a flowchart to make sure
15:00.28poushag:P
15:01.20lardman|workpoushag: hmm, perhaps, as you say, needs a diagram
15:03.03*** join/#openzaurus kolla (n=kolla@firda.kolla.no)
15:19.02poushaglardman:  my sketch seems to confirm what you said earlier - you can measure the v/% right before and right after the kernel shuts off the backlight then add in some margin for variation due to battery condition
15:19.34Interrobangis it possible to use the pdaXrom 1.1 beta 4 kernel (2.6), with it's video acceleration, in OZ?  could I just switch the zImage files and flash it safely?
15:19.53hrw|worknot quite
15:20.03hrw|workit will maybe work but no warranty
15:20.11hrw|workand pdaX kernel lack some fixes
15:20.20Interrobang...such as?
15:20.41Interrobanganything big?
15:20.51poushaglardman:  or you can just do trial and error by testing several settings i suppose (yuck)
15:20.57hrw|workfix-mmc-oops-on-eject
15:21.02hrw|workfix-kernel-logo-rotation
15:21.17hrw|workadd-some-cards-into-some-drivers
15:21.25hrw|workfix-usb-power-level
15:21.35Interrobangi'm no kernel hacker, but i know oops are very bad, no?
15:21.53CoreDump|wm06Interrobang: they are bloody anyoing
15:22.30Interrobangis it posible to apply the pdaX video accel patches to the OZ kernel sources and then compile a custom kernel?
15:22.51Interrobangwill it break any OZ fixes?
15:22.52CoreDump|wm06Interrobang: forget it
15:23.11CoreDump|wm06the patch won't apply and you should really know what you are doing to port it to OZ
15:23.21CoreDump|wm06and it's kinda large =)
15:23.30Interrobanghow many KB?
15:23.32XorAw100 accels are in kernel 2.6.17
15:23.35CoreDump|wm0650 IIRC
15:24.07hrw|workXorA: our 2.6.16 has them too
15:24.09CoreDump|wm06XorA: which is very nice for C7x0 users =)
15:24.32Interrobangso wam, OZ already has video accel??
15:24.43XorAInterrobang: yes for w100 in opie
15:25.03Interrobangbut it's not listed in mplayer -vo help
15:25.14hrw|workinstall mplayer-atty
15:25.27Interrobang...and that will get me w100 accel?
15:25.28hrw|workXorA: your latest mplayer has w100 inside?
15:25.34XorAhrw|work: yes
15:25.42XorAhrw|work: and is faster than -atty version
15:26.06hrw|workXorA: do you have .oz354x build to test it and push?
15:26.11InterrobangXorA: is it comparable to cacko's speed?
15:26.34XorAhrw|work: no, I have not enough diskspace for two branches
15:26.42XorAhrw|work: I need to do pre8 tomorrow
15:27.02hrw|workok
15:27.06XorAInterrobang: exactly the same code as cacko mplayer
15:27.37InterrobangXorA: just to make sure; exactly same speed?
15:28.17XorAInterrobang: I used to use cacko mplayer on OZ until I put recipe in OE to compile it for OE, never saw a difference
15:28.47Interrobangwhoa, you can just execute the cacko mplayer on OZ straight out???
15:29.02*** join/#openzaurus JiniDog (n=jini@Vb6ba.v.pppool.de)
15:29.21Interrobangi though it depends upon something in the kernel
15:30.07XorAInterrobang: no, w100 accelleration is userspace
15:31.39XorAInterrobang: I assume you have a c7x0 device?
15:31.53InterrobangXorA: yes C860..question incoming...
15:32.37InterrobangXorA: i wasn't able to change to 320x240 with mplayer -vo w100, but i could with mplayer -vo fbdev.  i saw a huge performance leap over OZ stock mplayer using fbdev there as well.  how is this difference possible?
15:32.55Interrobangerr i couldn't with w110 COULD with fbdev
15:32.58XorAhrw|work: when dust has settled from sale of half of this company I may be able to spare some machine for a compile farm here, then I can help with branches more
15:33.40XorAInterrobang: mplayer-atty will auto switch res with -vo w100
15:33.41hrw|workwould be nice
15:34.21XorAw00 w100 patches went clean into 1.0pre8
15:34.43CoreDump|wm06XorA: nice!
15:34.47InterrobangXorA: what res does it automatically switch to?  also, the cacko mplayer is faster isn't it?  how do i switch res with that?
15:35.11XorAInterrobang: cacko mplayer last I checked was identical to our mplayer-atty package
15:35.16XorAInterrobang: its the same source code
15:35.54InterrobangXorA: thx for all the help, got 1 more question...  i tried this in cacko yesterday but got no response...
15:36.50InterrobangXorA: i ran mplayer -vo fbdev -vm -fbmode QVGA in cacko, which changes to 240x320.  i saw a large performance boost over a similar command with stock mplayer in OZ...
15:37.06lardman|workpoushag: Cool, thanks
15:37.06InterrobangXorA: ...does cacko fbdev use some kind of eaccel?
15:37.12*** join/#openzaurus sirfred (n=mteira@88.11.0.113)
15:37.13lardman|workXorA: Your mplayer works with GPE doesn't it?
15:37.30InterrobangXorA: if not the mplayer prog, then maybe the kernel provides some anyway?
15:37.40sirfredGood evening.
15:37.53XorAlardman|work: with either, it uses SDL
15:37.56poushaglardman: yw - let me know how it goes
15:38.05lardman|workXorA: What's it called these days?
15:38.10XorAInterrobang: no idea
15:38.13lardman|workpoushag: Will do, thanks for your advice
15:38.52XorAlardman|work: mplayer_1.0pre7+cvsXXXXX.bb
15:39.21InterrobangXorA: if i play a 320x240 (or 240x320, i have both) using mplayer -vo w100, it will automatically change to 320x240 (or 240x320)?
15:39.51XorAInterrobang: yes
15:42.34InterrobangXorA: if you're using mplayer yourself, and you have the time, could i send you a test file for you to play without -framedrop?  i'd like to know if the video is still in sync at the end.
15:43.00Interrobangit's an anime file, perfectly safe
15:43.29Interrobangmy Z is down from brocken pdaXrom testing
15:44.22XorAInterrobang: ask me tomorrow, Im almost at end of working day here
15:44.48InterrobangXorA: don't worry about, i should have the same installed on mine by the end of today
15:46.32Interrobangaarrrggh wtf!?!?  http://oz.hentges.net/mirror/oz3.5.4/feed/* are all forbidden?!?!
15:46.34XorAahah, compiling the right mplayer this time
15:46.58*** join/#openzaurus Troglodyt (n=KevinSpe@216.126.240.97)
15:48.46XorANOTE: package mplayer-1.0pre8: completed
15:49.12hrw|workInterrobang: he has limited bandwidth and do not want people to wget feeds
15:49.14CoreDump|wm06Interrobang: yes they are, not for ipkg tho
15:51.00Interrobangany1 know the link for mplayer-atty?
15:51.07poushagi want to get opera - is compat-libs available? (cant fine it in OZ feeds)
15:51.09Interrobangwell the filename and dir?
15:51.59lardman|workpoushag: The compat libs are supposed to not work anymore
15:52.29lardman|workpoushag: due to the move to soft-float...
15:52.56lardman|workInterrobang: The source, or a binary?
15:54.05poushagwell i found that having google up front, the bookmark management, and the zoom control on opera is something i really dont want to do without
15:54.52*** join/#openzaurus redguy (n=mati@83.25.99.152)
15:55.31Interrobanglardman|work: binary
15:56.11Interrobanglardman|work: i have no kind net connection on my Z and i need the dir and package name to dl it manually
15:56.22lardman|workAh, okay, I was about to suggest ipkg ;)
15:56.27hrw|workcu
15:56.32lardman|workbye hrw
15:57.37lardman|workInterrobang: The name should be mplayer-atty-1.1.5 with something tacked on the end
15:58.18lardman|workAnyone good with web interfaces and want to whip up an ipkgfind style search page for the OpenZaurus feeds?
15:59.07XorAlardman|work: someone did that once
15:59.52*** join/#openzaurus hvontres (n=hvontres@redhill.orthodyne.com)
15:59.55lardman|workXorA: Any good? It would solve both the bandwidth issues of people downloading the entire feed, and allow people to grab individial packages
16:00.16XorAlardman|work: it was pretty good, Ive no idea where it went though
16:00.27*** join/#openzaurus hvontres (n=hvontres@208.2.199.132)
16:00.29lardman|workshame
16:00.38XorAok, 1.0pre8 is a goer
16:01.01Interrobangi imagine that i'm part of that problem, dl'ing the package files to sift dependecies manually
16:01.02lardman|workI'm still building pre7-cvs
16:01.24XorAlardman|work: I shall checkin tonight
16:01.29lardman|workThe other thing the web interface ought to do, is use Packages to list the deps
16:01.32*** join/#openzaurus hvontres (n=hvontres@63.92.216.147)
16:01.37XorAlardman|work: tomorrow morning actually Im out slaying dragons tonight
16:01.38lardman|workXorA; Cool
16:01.42lardman|work:)
16:02.27XorAtime for me to run
16:02.34lardman|workbye
16:05.34*** join/#openzaurus Interrobang (n=poiu@ppp-71-136-111-180.dsl.sndg02.pacbell.net)
16:14.37lardman|workInterrobang: Try mplayer-atty_1.1.5-r5_arm.ipk
16:15.17lardman|workInterrobang: Or you could just go to the feed, and look in the Packages file in each one and see whether mplayer is there... ;)
16:16.17*** join/#openzaurus hvontres (n=hvontres@68.120.74.196)
16:16.53*** join/#openzaurus hvontres (n=hvontres@68.120.74.196)
16:24.23lardman|workhrw|gone: bl_0.1 .bb here: http://people.bath.ac.uk/enpsgp/Zaurus/bl/bl_0.1.bb
16:29.20*** join/#openzaurus odo (n=Rand@xdsl-87-78-76-212.netcologne.de)
16:32.37*** join/#openzaurus ugarit (n=ugarit@216-15-34-7.c3-0.129-ubr3.lnh-129.md.cable.rcn.com)
16:32.50odoDid anybody succed in transferring data from a mobile to the C1000 with the 3.5.4.1-final via irda? My S55 is recognized - it is listed in the taskbar-applet als "Siemens S55". If I send a vCard (whether 2.1 or 3.0 doesn't matter) the phone tells me it was sent but in kapi it never shows up. If I try to sync from the akita I alwasy get the error that no partner was found ...
16:46.10*** join/#openzaurus JoeSchmo (n=jciccone@pool-71-127-209-149.nwrknj.east.verizon.net)
16:48.24CoreDump|wm06odo: my irda doesn't work at all on Akita
16:49.40odoCoreDump: hups ... akita sees my S55 if I switch irda on, enable discovery and recieve at the akita-side an the irda-datatrasfer at the S55-side
16:50.07*** join/#openzaurus sirfred (n=mteira@113.Red-88-11-0.staticIP.rima-tde.net) [NETSPLIT VICTIM]
16:53.02CoreDump|wm06doh indeed
17:01.10CoreDump|wm06http://handhelds.org/scap/port.8471.png <- recieved via IRDA
17:18.41lardman|worknight all
17:19.49*** join/#openzaurus redguy_ (n=mati@adf5.neoplus.adsl.tpnet.pl)
17:24.28*** join/#openzaurus alan|home (n=alan@ARouen-151-1-76-140.w86-215.abo.wanadoo.fr)
18:02.04poushagopera-mini looks sweet - any way i can get a recent jvm on my collie?
18:19.39poushagactually i like how it reformats webpages from a proxy server
18:20.42poushagthe user input methods are not really ideal for pdas (but are optimized for mobile phone keypad)
18:21.38poushagbut i see opera 8.5 has sdks for QT/e and Qtopia
18:22.12poushagnad if it could be configured to use the proxy server that opera-mini uses - wow!
18:22.19poushag*and
18:42.13poushaghttp://www.opera.com/products/devices/technology/sdk/qtopia/specs/
18:42.28poushagis anyone else interested in this besides me?
18:43.16poushagis OZ 3.5.4.x compatible with these requirements?
18:43.37*** join/#openzaurus XorA|gon1 (n=dp@81-178-91-191.dsl.pipex.com)
18:45.05*** join/#openzaurus hobbs (n=andrew@p3m/member/hobbs)
18:52.36Interrobangisn't opera free for dl now?
18:52.41Interrobangjust try it out
18:52.52Interrobangthat got a zaurus edition on the website
18:54.27Interrobanghm, no dl on the website.  it's included in cacko though, try just copying it over to OZ OPIE
18:54.57Interrobangi think it has ssr built in
18:55.46Interrobangyou should also look into minimo (mini mozilla), it has ssr as well
19:00.23Interrobangalso, you should consider using Xvnc and fbvnc to load a browser in 480x640 then scale it to 240x320.  although i'm sure opera accomplishes something with its proxy, i think it's conceptually impossible to accomplish much.
19:01.25Interrobangtaking websites design for 800x600 minimum, 1024x768+ typically, and scaling it down to 240x320 and below isn't work it really
19:02.19Interrobangi spent a long time trying to squeeze everything onto my collie, but 240x320 create limitations you can't overcome without tiny, hard to read fonts.
19:02.27*** join/#openzaurus XorA|gone (n=dp@81-178-217-33.dsl.pipex.com)
19:04.09Interrobangpoushag: you got 3 options: get used to scrolling around, use vnc to scale 480x640 down, or get a clamshell/6000
19:13.17radugasub-pixel rendering?
19:13.32*** join/#openzaurus synth (n=cb@c-71-203-15-223.hsd1.fl.comcast.net)
19:14.08radugainteresting, though.
19:14.12*** join/#openzaurus XorA|gon1 (n=dp@81-179-72-76.dsl.pipex.com)
19:14.16*** join/#openzaurus toi (n=peter@d54C27365.access.telenet.be)
19:14.37radugahas anyone seriously tried subpixel rendering with qvga screens, to get higher resolutions?
19:14.58radugaI'm not sure it would help... much... at all
19:15.17radugabut might be able to squeese a little more definition out of a collie screen
19:17.56Interrobangraduga: nope, the problem isn't the rendering method, it's the limited number of pixel to which to display the information.  in fact, the blurring that would result would make text harder to read, not easier.
19:19.00Interrobangi tried using Xvnc and fbvnc together and even rewriting the scaling code, it actually gets worse the more accurate you make it
19:19.54Interrobangsometimes text is rendered on odd pixels, so that when you scale it down by 2, 1 pixel of text at the higher resoultion, gets distributed with half its intensity across 2 at the lower resolution
19:22.18Interrobangkuickshow (KDE pic viewer), at least on my comp (Arch Linux), exhibits the same problem with scaling down images
19:22.49*** join/#openzaurus redguy (n=mati@adf5.neoplus.adsl.tpnet.pl)
19:23.10Interrobangif you want to experiment with 640x480 resolution then scale it down by 50% then you'll see what i mean
19:24.12Interrobangother scaling techniques have their faults, but may better handle certain color features
19:24.12*** join/#openzaurus _law_ (n=_law_@213.173.86.202)
19:26.49Interrobangthere's 4 pixels in a vga screen for every 1 in a qvga screen.  you're going to lose 75% of the information guranteed.  unless the information is spread across enough pixels (large font, pointless though), you won't have enough to use
19:37.51radugainterrobang: at least in theory, subpixel rendering should give you a *little* more resolution, in one direction (depending on how the RGB is laid out)
19:38.29radugain that, one standard pixel is actually three R,G,B in close proximity
19:38.56radugaso, if you illuminate the R/G/B selectively, you get theoretically, 3x the resolution
19:39.07radugain practice it... tends not to work so good :)
19:40.09*** join/#openzaurus lardman (i=lardman@pc-138-38-248-213.bath.ac.uk)
19:40.18lardmanevening
19:40.55hobbsraduga: and I think the Z's subpixels are in the wrong direction to be truly useful. Vertical isn't as good as horizontal.
19:41.54lardmanhrw: If we could provide kopi/kapi/whatever it's called for both Opie & GPE we'd be onto a winner
19:42.21Interrobangthat would depend upon specific color features.  if the background or the text is the wrong color, subpixel rendering may make things worse
19:42.50lardmanInterrobang: Find mplayer?
19:43.01Interrobanglardman: nope, not the right name
19:43.33lardmanInterrobang: Go to each feed directory and append /Packages - then search these text files for the correct name
19:43.33Interrobangi can't get the package lists since they're been disable for nwo
19:43.38hrwlardman: I know - will probably work on it soon
19:43.55lardmanhrw: Was just a thought after a quick read of oesf
19:45.02lardmanhrw: I also think a pot of OO, if someone can be bothered, just to show it's possible (despite the mine's bigger than yours comments ;)
19:45.15lardmans/pot/port
19:45.17hrwlardman: koen tried to build OO.org
19:45.41lardmannot successful?
19:45.48hrwit was ~92M of patches to apply and he decided to drop it when it took several hours and not finished
19:46.52lardmanhmm
19:47.07lardmanwe need someone how really wants it, and who can do the work
19:47.35lardmanOT, have you included bl in your gpe images? It's not in .dev
19:47.44hrwit is in .oz354x
19:48.50lardmanCool
19:49.12lardmanThis split between people adding to different branches is troublesome (at least for me)
19:49.56hrwI'm concentrated on .oz354x
19:50.13hrwI do not remember when last time I used .dev to build image
19:51.02lardmanI'm more of a dev man, but it's a shame that stuff isn't propogated more frequently/reliably
19:52.04hrwbl in .dev need to be discussed with gpe people probably
19:52.11Interrobanglardman: i checked base x11 and opie directories, but i can't get gpe/Packages.  do you have the packages lists on your Z?  could you grep atty the lists?
19:53.27lardmanhrw: Yes, but I think they'd be happy to have something which is more inclusive than the current offering
19:54.13lardmanInterrobang: No, I'm on a fresh flah I'm afraid, but if you ask hrw nicely, he might tell you which feed to look in
19:54.15hrwre
19:54.36hrwthere is no 'gpe' feed in OZ 3.5.4x
19:54.44hrwgpe stuff is in 'x11'
19:55.31Interrobanghrw: do you know which feed in which to find mplayer-atty?
19:55.45lardmanhrw: I don't really understand the branching structure we use - I thought that there should be a stable branch into which most stuff goes eventually
19:55.55hrwmachine/c7x0/ or upgrades/machine/c7x0/
19:56.06lardmanhrw: and dev for bleeding edge, and individual branches for releases
19:57.09lardmanhrw: But I suppose that would require more people/time to sort out
19:57.44lardman~lart using dialup
19:57.57lardmanibot: Not harsh enough
19:59.53lardmanI just came online to try to download my latest build, but it looks like it's going to take hours, so I'll call it quits
19:59.56lardmannight all
20:01.01Interrobanghrw: i tried machine/c7x0/Packages where it wasn't listed, and upgrades/machine/c7x0/Packages which didn't exist.  the ewi.utwente mirror is down and PDAi is always forbidden (wtf?)
20:01.24Interrobangwhy is the PDAi mirror always off limits?!!?
20:03.11Interrobangis any1 able to access the PDAi OZ mirror??
20:05.11hrwlardman: .oz354x is branch used for OZ 3.5.4x and I want to have it stable
20:05.53poushaginterrobang:  have you tried the opera-mini simulator?  
20:06.22Interrobanglol
20:08.10poushaghttp://www.opera.com/products/mobile/operamini/demo.dml
20:08.38poushagmakes most websites very usable (just scrolling up and down with cursor keys)
20:09.50Interrobangpoushag: i was talking about on my desktop, but thx for the idea.
20:10.07Interrobangi tried google en->en translation as a proxy
20:10.18Interrobangstill can't get ewi or pdai mirrors
20:10.58hrwewi is out until resetmonkey reboot it
20:14.44poushaginterrobang:  i was going to open a private chat with you about opera - but when it tried it didnt work
20:17.26Interrobangwhat client are you using?
20:17.54poushagchatzilla
20:18.35Interrobangwere you trying to do dcc or just a private message dialog?
20:18.54poushagi can pm myself but not you
20:19.25Interrobangare you ure you typed my name right?
20:19.25poushagi read your pm
20:19.41Interrobangi can pm myself
20:20.00Interrobangthat does go through the server i believe
20:20.10hobbspoushag isn't identified
20:20.11poushagbut when i pm to you i get a msg "Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register!"
20:20.19Interrobangoh
20:20.27Interrobangyou know how to register you nick?
20:20.36poushaghmmm - i thought i did register it
20:20.51hobbspoushag: you did, looks like. You're just not identified.
20:21.23Interrobanghobbs: you can use a registered nick even if you don't enter the password?
20:21.35hobbsInterrobang: yes, freenode doesn't do automatic booting
20:21.51poushagweird
20:21.52Interrobang...very disturbing...
20:21.59poushagi wasnt prompted for a pw
20:22.30Interrobang/msg NickServ IDENTIFY <password>
20:22.56hobbslook into the SECURE and KILL options
20:23.05hobbsif you set them both at the same time it will act like most networks
20:23.46*** join/#openzaurus _law_|iBook (n=_law_|iB@213.173.86.202)
20:24.57poushagim going to logout and back in to try the new config (with authentication)
20:26.08*** join/#openzaurus poushag (n=poushag@ip24-254-136-100.rn.hr.cox.net)
20:41.44*** join/#openzaurus redguy (n=mati@adf5.neoplus.adsl.tpnet.pl)
20:53.45*** join/#openzaurus darkschneider (n=gab@213-140-6-96.ip.fastwebnet.it)
21:15.08hrw3A
22:29.50*** join/#openzaurus JoeSchmo_ (n=jciccone@pool-71-125-88-25.nwrknj.east.verizon.net)
23:42.11*** join/#openzaurus marcan (i=1337@68.Red-88-5-78.staticIP.rima-tde.net)
23:54.10Interrobangany1 have any idea when ewi zaurus mirror will be back up?

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with blootbot logs, split per channel, etc.