Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2007-11-19

00:00:04preglowi resample to 8khz first
00:00:56jhMikeSpreglow: ah, right 8k /me's was off by a factor of 2
00:01:06 Quit davina (Remote closed the connection)
00:01:28preglowthe clip speex decoder spits out should be 16khz, though, since it gets qmf_synthed with an upper band of only zeroes
00:01:31jhMikeSjmspeex: we do have 64-bit sims but it still won't be 64-bit?
00:02:02jmspeexjhMikeS: Speex no longer needs anything more than 32 bit, so it's irrelevant
00:02:17jmspeex(it used to require 64-bit accumulation somewhere, but no longer)
00:03:39preglowjmspeex: btw, do you expect you'll be able to convert vbr encoding to fixed point?
00:03:41n1sjhMikeS: btw only long is 64 bit on amd64
00:04:03n1s(long long)
00:04:21n1sreverse order of the previous 2
00:06:55 Quit jhMikeS (Nick collision from services.)
00:07:01 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
00:07:27jhMikeSthe seek code is using spx_int64_t which is showing up as long long int.
00:07:47preglowthe seek code is very poor
00:07:51preglowsomeone who knows ogg should fix that
00:08:00jhMikeSI know long is variable...I wrote part of inttypes.c for rockbox :)
00:08:17jhMikeS*.h even
00:08:19n1sok :)
00:09:12jhMikeSpreglow: I'll ignore it for now then
00:11:11preglowjhMikeS: the seek code even used to be float...
00:11:28jhMikeSat least the SIM won't blow up now when the core codec calls DEBUGF
00:12:28jhMikeSpreglow: gcc will emulate float when the target CPU doesn't have it?
00:12:41 Join Thundercloud [0] (
00:15:16preglowjhMikeS: yes
00:21:30preglowjhMikeS: now that you had the honour of doing the red delta, i'll chip out of it with small green ones :)
00:21:31 Join bagawk [0] (n=lee@unaffiliated/bagawk)
00:22:13amiconnpixelma: ping
00:23:17jmspeexpreglow: as soon as I receive your patch :-)
00:23:24 Quit Zagor ("Client exiting")
00:23:25preglowjmspeex: if i decode an nb encoded frome with a wb decoder, i should get a 16000 hz frame out with the upper halfband being zero amp, yes?
00:23:44preglowok, so i'm doing something wrong
00:24:08jmspeexpreglow: Theoretically, you'd get slightly better quality using narrowband+resampler than wideband though
00:24:21jhMikeSpreglow: use TNT if nescessary
00:24:24preglowjmspeex: why?
00:24:33jmspeexbecause the resampler is better at resampling than the wb decoder's QMF
00:24:40preglowhow many taps is that anyway?
00:24:48jmspeexHow much difference, though I don't know
00:24:55jmspeexQMF is 64 taps
00:25:10preglowbut yeah, a resampler of even 64 taps would suck cpu
00:25:16preglowso we'll be sticking with this, heh
00:26:07jmspeexpreglow: the CPU you spend on resampling isn't spent on the QMF.
00:26:13preglowjmspeex: true...
00:26:25preglowjmspeex: we'll see, i'm planning on adapting your resampler for our own use anyway
00:26:30jmspeexOverall, the resampler can probably be a bit faster because you can use a shorter filter for the same quality
00:26:41 Quit Robin0800 (" HydraIRC -> <- *I* use it, so it must be good!")
00:26:57preglowjmspeex: i has tons of float math
00:27:00amiconnpreglow: Is it better than our linear interpolation thing (I'd expect it to be...)
00:27:07preglowamiconn: by leaps and bounds
00:27:12preglowamiconn: this one is a proper fir filter based one
00:27:34preglowamiconn: jmspeex here is a linear interpolation enthusiast
00:27:47jmspeexand it should be a felony to even suggest using linear interpolation for audio
00:27:57*preglow goes to jai
00:28:00 Quit Guile`` ("KVIrc 3.2.0 'Realia'")
00:28:28 Join webguest10 [0] (i=489929dc@gateway/web/cgi-irc/
00:28:32jhMikeSwell, some things might require it like SPC where that codec is really CPU hungry. what's the relative expense?
00:28:39BigBambipreglow: Don't pass go now
00:29:06webguest10im really frustrated i downloaded a few themes and they dont look how they should
00:29:08*Shaid takes away $200 from preglow just to be safae.
00:29:13 Quit HellDragon (Read error: 104 (Connection reset by peer))
00:29:23peturwebguest10: installed the fonts?
00:29:25jmspeexThe only reason I wrote this resampler is so that nobody can ever use "but I don't have a resampler with compatible license" as an excuse for using linear interpolation.
00:29:35 Join HellDragon [0] (i=jd@unaffiliated/helldragon)
00:29:52 Join webguest39 [0] (i=47b60e83@gateway/web/cgi-irc/
00:29:57jhMikeSI'm sure gigabeat can just use the nice fir filter though for everything.
00:29:57preglowjhMikeS: any kind of fir filtering will be quite a bit more expensive
00:30:06preglowoh, sure, we probably could run c code resampling on that
00:30:08jmspeexamiconn: BTW, if you're not convinced, take a 8 kHz file, resampler linearly to 48 kHz and listen to it. Now compare with a properly resampled file.
00:30:10preglowcoldfire and pp will need asm
00:30:12 Quit lazka (Remote closed the connection)
00:30:18n1swebguest10: do they require patched builds?
00:30:20preglowjmspeex: he doesn't need convincing
00:30:22jhMikeSwe'll have to retain the cheap one
00:30:30amiconnSeveral things don't have proper dependencies in the build system
00:30:38 Quit nature (Remote closed the connection)
00:30:40preglowamiconn: codec libs...
00:30:51amiconnjmspeex: I *know* that linear inerpolation sounds bad...
00:30:54jhMikeSpreglow: nice chip of the ol delta
00:31:03 Quit webguest39 (Client Quit)
00:31:04jmspeexpreglow: Just asking "Is it better than our linear interpolation thing" made me worried.
00:31:20amiconnpreglow: 'features', lang.h ...
00:31:27jhMikeSdoing a lengh-8 fir filter in SPC is pretty expensive even with emac
00:31:37preglowthat's because fir filtering is expensive
00:31:39preglowno surprise there
00:32:04amiconnWell, iiuc the ape filters are fir filters
00:32:07jmspeexamiconn: The resampler I wrote has variable quality. Starting from "worse than SRC fastest, but better than linear, up to slightly better than SRC best"
00:32:19preglowjhMikeS: poorest quality filters speex_resample uses is 8 taps
00:32:20 Join porno [0] (n=e_ibarra@
00:32:23preglowjhMikeS: biggest, 256
00:32:56jmspeexpreglow: It's not exactly a number of taps, because it gets multiplied by the resampling factor in some cases.
00:32:58preglowjmspeex: it doesn't always do the jos paper style interpolation of sinc window while interpolating, no?
00:33:15preglowlooked to me like it had to processing modes, pretty much
00:33:28preglowwhile RESAMPLING
00:33:39 Quit webguest10 ("CGI:IRC (EOF)")
00:34:09jmspeexpreglow: Unlike the jos paper, it 1) does quadratic interpolation and 2) it resamples the resampling filter to make it easy to compute without having to do any rounding operation.
00:34:09*preglow sighs at his own english
00:34:36preglowjmspeex: yeah, but does it still do one interpolation per filter tap?
00:34:55jmspeexpreglow: It's the filter tap number that's variable.
00:35:05 Quit Nico_P (Remote closed the connection)
00:35:18jmspeexFor example, if you do 16 kHz -> 8 kHz, the number of taps is double what you see in the struct.
00:35:26preglowright, i read the paper about two-three years ago, can't remember the details
00:35:51jmspeexso the highest quality would actually use 512 taps for 16->8, not 256
00:35:53 Join iamben [0] (
00:36:16jhMikeSouch...nice nice with an ARM multiplier :p
00:36:29jmspeexthat's highest quality.
00:36:35preglowand if we have to do anything but do a straight access for the filter coef, forget it
00:36:55n1sjhMikeS: the imx31 in gigabeat s has a mac unit :)
00:37:21preglown1s: how do you access it?
00:37:50jmspeexThe nice thing with my algo unlike jos is that the access pattern is trivial and on "simple" sampling ratios (e.g. 3/4), it's basically a bunch of inner products
00:37:56n1spreglow: not sure really haven't looked into it in detail i still refuse to boot windows to try stuff :)
00:37:57jhMikeSn1s: don't rub it in :)
00:38:39preglowi wonder what the hell the gigabeat s needs an accumulator unit for
00:38:43preglowwhat with its bloody vector fpu
00:38:56jhMikeSpreglow: any more delta bombs coming?
00:38:59preglowjhMikeS: nah
00:39:20preglowjhMikeS: i can't be bothered to hack away smaller stuff, i'm looking for big things.i don't like ifdefs too much
00:40:02amiconnBagder: h120 seems to be delayed this time...
00:40:12preglowwhy does that happen anyway?
00:40:47Bagdersee the rightmost column in the "log" page
00:40:53 Join nature [0] (
00:40:53jhMikeSpreglow: perhaps rearranging the source a little to reduce the #?
00:40:54amiconnIiuc Bagder tracked it down to barrywardell's server not delivering the zip
00:41:02Bagderthe "delayed" one hasn't been updated
00:41:03preglowjhMikeS: makes it harder to sync
00:41:09preglowit's already boring to sync
00:41:11pixelmayeah, barrywardell's server again
00:41:15amiconnSo the build is *actually* delayed, which is bad imo
00:41:27Bagderyes, it seems it always fail to scp the zips from barry's server
00:41:39jhMikeSpreglow: sync? I mean just group the functions in one block that shouldn't be there.
00:41:49amiconnCould you disable target builds on that server for now?
00:42:29natureThere is no partition table for the #G 15 gig on, closest is mbr-3g-40gb.bin. Has anyone got mbr-3g-15gb.bin (or whatever its called)? I'd be happy to add it to the wiki.
00:42:41*amiconn wonders when all build server owners will have updated their arm-elf-gcc
00:42:41nature#G == 3G
00:42:58preglowamiconn: <- what about this one?
00:43:12Bagderender`: arm-elf-gcc update soon?
00:43:20preglowjhMikeS: like i said, it'll make my job harder when i need to sync against speex svn
00:43:25preglowjhMikeS: anyway, it's not that annoying yet
00:43:28jhMikeSah, yeah
00:43:41Bagderlostlogic: arm-elf-gcc update soon?
00:44:29 Quit jake_ (Remote closed the connection)
00:44:37amiconnpreglow: I will try the yuv improvements of that patch. Imo the normal update improvement isn't worth going all-asm for the function
00:44:38ender`Bagder: which version?
00:44:55preglowamiconn: sure, i just don't like obvious stuff rotting in the tracker
00:45:12Bagderender`: 4.0.3 still, but the multilib patch. You can just rerun and it'll do it right
00:45:14amiconnbusy rebuilding atm...
00:46:03amiconnpreglow: Voice works without problems on 2nd gen btw (PP5002)
00:46:06*jhMikeS wonders how to get espeak working on Debian VMWare without it wanting a bloody kernel update.
00:46:38 Quit Rondom ("Ex-Chat")
00:46:44jhMikeSor gen using Windows TTS ...
00:46:46 Join billenium_ [0] (
00:47:10 Quit billenium_ (Client Quit)
00:47:23ender`i'll build through crossdev, which binutils version?
00:47:33ender`2.16 or 2.17?
00:48:43preglowamiconn: nice
00:48:58preglowamiconn: remember, i'd urge you to try putting the contents of filters_arm4.S in iram
00:49:27preglowno big gains for any of my targets, but pp5002 will probably be very happy
00:49:59 Join webguest79 [0] (i=45ffd379@gateway/web/cgi-irc/
00:50:36 Quit webguest79 (Client Quit)
00:57:00n1spreglow: looked it up, it's SMLA* and SMUL* 16x16 16x32 mac and multipy, saturating add/sub and clz
00:57:31preglown1s: ahh, yeah, those are ordinary arm5 instructions
00:57:35preglowand i wish we had them
00:58:07ender`ok, gcc compiling
00:58:55n1spreglow: are there places that we could do things better with floating point (better precision?)
00:59:17 Quit Arathis ("Bye, bye")
01:00:03 Join barrywardell [0] (
01:00:34ender`and done
01:01:09preglown1s: not necessarily better
01:01:10preglown1s: but easier
01:01:21preglowsometimes better too, though
01:02:24 Join denstark [0] (
01:02:27preglowamiconn: what was the ldm timing again? 2 + n?
01:03:38denstarkHey guys. I'm having an odd problem with Rockbox. I browse through directories, and when I open one it lists nothing. I upgraded to the lastest rockbox just now, and I get the same problem
01:06:02PaulJamdenstark: what is your show files setting set to?
01:06:23denstarkPaulJam: Let me take a look, I assume that is in settings within Rockbox itself?
01:07:14denstarkPaulJam: Ah, got it. Thank you. For some reason, it was set to playlists only!
01:07:16PaulJamSettings=> General Settings=>File View=>Show Files
01:09:40***Saving seen data "./dancer.seen"
01:13:12 Quit billenium ("Ex-Chat")
01:13:56 Quit porno (Read error: 110 (Connection timed out))
01:14:05ender`and good night
01:14:33 Quit ender` (" Marriage is not a word, it's a sentence. A life sentence...")
01:16:02n1shmm armv6 can do some parallel fancy stuff too like dual 16x16 mac
01:16:15preglowpity nothing is arm6
01:16:21 Quit karashata ("I will *SO* make you regret that... Later... *is a lazy dragon*")
01:16:35n1spreglow: the gigabeat s chip is
01:16:51preglowit is??
01:17:11preglowthat thing has to be the most overspecced thing i've seen
01:17:41LloreanWell at least the S was *supposed* to be a Media Device
01:17:55n1sthis is the core it uses
01:18:40n1syeah the F/X with 300MHz of brute force for mp3 decoding was some real overkill, the s does video at least :)
01:19:13amiconnHD resolution, or what? ;)
01:19:33n1samiconn: it has a tv-out so why not :)
01:20:13 Join Gnu47_ [0] (
01:20:31 Quit Gnu47 (Nick collision from services.)
01:20:34n1si never tried it though so i dunno how the of video works but their photo viewer was nice
01:20:43 Nick Gnu47_ is now known as Gnu47 (
01:26:31LloreanThe TV-out is only 320x240 still, I think.
01:27:08n1smeh, that's no fun to watch on a big screen
01:27:22LloreanIt's no fun to watch on a medium size screen anyway
01:28:08preglowvoicebox stuff on the ml
01:30:51 Join psycho_maniac [0] (
01:30:51preglowjmspeex: ouch, speexdec coredumps here when i try to force it to decode an nb file in wb mode
01:33:52 Quit eigma ()
01:35:56 Quit petur ("Zzzzz")
01:36:07preglowjmspeex: hmm, svn version does not do so
01:38:16preglowi wonder what we're doing wrong
01:43:18jhMikeSyou made a coredumper version?
01:43:41 Quit PaulJam (".")
01:46:32 Quit Thundercloud (Read error: 104 (Connection reset by peer))
01:47:55 Quit scorche (Nick collision from services.)
01:48:24 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
01:54:04LloreanBagder: You around, and could you manage a couple small webpage changes?
01:54:28preglowjhMikeS: our decoder plays nb slower than it should
01:54:31preglowgod knows why
01:54:38 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
01:54:57 Quit Seed (Nick collision from services.)
01:55:03 Join Seedy [0] (
01:56:34jhMikeSand you said a test one outside SVN does not?
01:56:53saratogathe zune using that CPU can at least do 720x480 VC-1
01:56:56preglowspeexdec works correctly if you force wb decoding, though
01:57:00preglowso don't know what we do wrong
01:57:44jhMikeSwhich speexdec? the example you send?
01:57:49jmworxpreglow: Yes, −−force-wb used to cause a segfault at some point
01:58:08 Join chris22 [0] (i=451c266d@gateway/web/cgi-irc/
01:59:05preglowjmworx: any idea what we're doing wrong when decoding nb in wb mode when we're getting 50% slower audio than we should out?
01:59:24 Quit nature ()
01:59:32jmworxdecoding: what do you mean exactly?
01:59:50preglowi'm encoding nb clips and feeding them to a wb decoder
02:00:14 Join matteo [0] (n=matteo@openwrt/developer/matteo)
02:00:15preglowspeexdec handles it nicely, spits out audio at the correct pitch/speed with zero upper sideband
02:00:28jmworxpreglow: just count frames and samples
02:00:36chris22This might not be the correct place to ask, but where do I learn the basics of USB driver programming? I am trying to help with the e200 USB project. I know c++ and C, but I am lost to low level programing
02:00:37jmworxa 1 second file shoule contain 50 frames
02:00:47matteoi have an iPod with ipodlinux.
02:00:47jmworxeach frame should be 320 samples when decoded to wideband
02:01:00jmworxIf you find something other than that, you know where you screwed up
02:01:01matteothis is a winpod, with linux on its own partition
02:01:10matteocan ipodlinux and rockbox coesist?
02:01:24preglowjmworx: yeah, but sb_decode will always spit out the number of samples it's initialized to, won't it?
02:01:28matteorockbox goes on the FAT32 partition IIRC
02:01:43preglowand i pass &wb_mode explicitely
02:01:48jmworxpreglow: Yes, but you need to read 320 samples
02:02:04 Part pixelma
02:02:05preglowjmworx: yup, in my test decoder it's even a constant
02:02:05jmworxdo you have 50 frames per second in the first place?
02:02:08psycho_maniacmatteo: yes both can be on the same player
02:02:32matteoso I just extract rockbox to the 2nd partition
02:02:46matteoto create /.rockbox ?
02:03:06jmworxpreglow: I mean, there's no way Speex will slow things down −− changing the speed is something pretty hard. That means you must have screwed up in a very obvious way (it can't be just a wrong option to Speex for example).
02:03:35preglowjmworx: yeah, i kinda figured
02:03:57jmworxpreglow: Just take a 10 second file and count things.
02:03:57preglowjmworx: my fault, nevermind, move along
02:04:02preglowresampler error
02:04:03preglowignore me
02:04:08matteois r15677 the latest?
02:04:09 Join Seed [0] (
02:04:12jmworxthat's what I thought
02:04:16psycho_maniacas long as its fat32 why did you partition it?
02:04:33matteoi didn't partitioned it
02:04:48matteoi mean the partition with music
02:05:21preglowjmworx: that was it, yes
02:05:24psycho_maniacjust extract it to the root of the player.
02:06:45 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
02:06:53 Quit atsea- (Remote closed the connection)
02:10:12*preglow has a 500kb voice file :>
02:11:20*preglow has a 200kb voice file :>
02:11:28psycho_maniachow great.
02:12:42Llorean200kb and still sounding as good as the old voices?
02:13:04preglowhell no
02:13:06preglowit sounds like crap
02:13:10preglowbut it's perfectly legible
02:13:17preglowat least for most clips :>
02:13:54 Quit psycho_maniac (" HydraIRC -> <- Wibbly Wobbly IRC")
02:14:15amiconnnb -q1 ?
02:15:17Lloreanpreglow: Well, at least it means voice can work on very lowmem targets, as long as they have a decent processor
02:17:30 Quit Seedy (Read error: 110 (Connection timed out))
02:17:36preglowamiconn: -n -q 0 -c 10
02:19:02preglowtesting without -n
02:21:24preglowpretty please
02:21:31preglowwhy does roloing rombox.iriver hang my h120?
02:21:42preglowthere's got to be a way of not having that happen
02:22:14amiconnWhy are you doing that? :>
02:22:14 Join nature [0] (
02:22:18preglowamiconn: mistake
02:22:27preglowamiconn: had hard reset
02:22:35preglowbtw, -q 0 -c 10 is 400kb
02:22:42preglowcompletely decent, but a fair bit of distortion
02:23:08 Quit saratoga ("CGI:IRC (EOF)")
02:24:57 Quit chris22 ("CGI:IRC (EOF)")
02:27:02rasherdo we want /utils/ /manual/ and /rbutil/ in the tarballs?
02:27:08rasher(they aren't there now)
02:28:33rasherflash/ and gdb/ is
02:30:30 Quit moos ("Rockbox rules the DAP world")
02:30:34 Quit barrywardell ()
02:31:18 Join dubious` [0] (
02:31:19 Quit dubious` (Remote closed the connection)
02:31:24 Join dubious` [0] (
02:32:14dubious`so has anyone made a line in cable for 5.5?
02:32:53SoapI've made many.
02:33:05dubious`is there any instructions on the net?
02:33:35SoapDepends on what information you have, and what you need.
02:34:12dubious`i just need to record live music with a new 5.5
02:34:24dubious`perhaps a batery pack too
02:34:34 Join Thundercloud [0] (
02:34:56Shaidbattery pack is fairly simple
02:35:06SoapNo, depends on what you need as far as instructions. You've told me very little. I don't know if you need help aquiring the parts, if you need help figuring out what pin is what, or if you need help with basic soldering skills.
02:35:44dubious`um instuctions
02:37:11dubious`i guess id buy a dock connecting cable, rip it appact and attach a 3.5 and a small batery with circurty from hack a day?
02:37:43n1s1) buy connector, line jack, cable 2) solder 3) ? 4) profit!
02:37:56 Join billenium [0] (
02:38:16Soapyou won't find a commercial cable with the line-in pins pre wired if you were thinking it would be as easy as using existing wires in an existing cable.
02:38:34dubious`i figured
02:38:54dubious`and that slderings gonna be hard cause its so small
02:39:12Soapshould lead you to all you need.
02:42:01 Join psycho_maniac [0] (
02:42:29 Quit jhMikeS (Nick collision from services.)
02:42:35 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
02:43:03 Quit Seed (Nick collision from services.)
02:43:29 Join Seed [0] (
02:45:03 Join ivan [0] (
02:48:02XavierGr ImageSearch, tx, ty, 380, 270, 650, 310, D:\DofusBot\Waypoints\Various\Done.bmp
02:48:02XavierGr msgbox % ErrorLevel
02:48:12XavierGrsorry about that
02:53:39 Quit Seed (Nick collision from services.)
02:54:05 Join Seed [0] (
02:58:12JdGordonlinuxstb: if your around.... i found another semi annoying bug with running the e200 bootloader instead of the OF one.. it starts up even if hold is on
03:04:02 Join LinuxMafia [0] (
03:04:05LinuxMafiahi all
03:04:15 Part ivan
03:04:39LinuxMafiaany one knows how can i convert to mpeg with mencoder ?
03:04:59 Quit Soap_ ()
03:07:57hcsLinuxMafia: look at the PluginMpegplayer page on the wiki:
03:08:18LinuxMafiahcs thanks alot
03:08:35 Quit nature (Read error: 104 (Connection reset by peer))
03:09:15hcsLinuxMafia: note that the manual would also have provided you the same link
03:09:42***Saving seen data "./dancer.seen"
03:09:55LinuxMafiahcs, i did not know it has how to encode
03:10:25hcsLinuxMafia: the manual does not have encoding instructions, but it does have the link to the wiki that I gave you
03:10:46LinuxMafiahcs, the problem is i want every thing fast , :-(
03:11:09LinuxMafiahcs , i am in Ruby tutorial reading , :-(
03:13:08LinuxMafiahcs, thanks alot
03:17:23 Quit mdevz (Read error: 104 (Connection reset by peer))
03:20:30 Quit scorche (Connection timed out)
03:20:59 Quit jmspeex (Read error: 110 (Connection timed out))
03:24:08 Join eigma [0] (i=eigma@
03:24:18eigmaladies and gentlemen −− we have audio output on the m:robe!
03:24:24 Quit LinuxMafia (Read error: 104 (Connection reset by peer))
03:24:53eigmait's a deafening 1 khz high-pitched whine.. but I can hear it :D
03:25:02lostlogiceigma: congrats
03:27:20 Quit matteo ("ketamine: it's not just for horses")
03:29:09n1srasher: is FILES case sensitive?
03:29:59rashern1s: did I mess up?
03:30:27rasherah yes
03:30:33rasherthat won't work
03:30:36LloreanWhy doesn't the tarball creation script just check out a current SVN source, remove the .svn folders, and tarball that up?
03:30:37rasherand indeed it doesn't
03:30:54rasherLlorean: it's supposed to work for everyone - not just the buildserver
03:31:14LloreanI don't understand quite what you mean
03:31:18n1sLlorean: because it's supposed to be independent of svn (I can't say i understand why though
03:31:40rasherLlorean: say I have a modified tree −− I should still be able to build a tarball out of that with the script
03:32:04rasherI had this exact same chat with Bagder a while ago. He might be able to explain better
03:32:10LloreanNo, I got it now
03:32:20LloreanFor distributing sources of modified builds, etc, too.
03:32:43n1sI wonder if anyone uses that ability or if we just maintain (sort of) those lists for fun
03:33:08rasherI use it for my sims. But those aren't modified, so I might as well just link to I guess
03:35:50 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
03:35:52 Join nature [0] (
03:36:56 Join TFGBD [0] (
03:38:29natureon the latest firmware and bootloader i am getting the "battery empty RECHARGE restarting" error, even when I am plugged in to the wall. Any ideas? I am on a 15gig 3rd gen ipod.
03:39:20 Join jmspeex [0] (
03:40:26 Nick fxb is now known as fxb__ (
03:47:03 Quit DerPapst ("So Long And Thanks For All The Fish!")
03:47:21 Quit Thundercloud (Remote closed the connection)
03:52:16 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
03:57:09 Join joshin [0] (n=josh@unaffiliated/joshin)
04:01:05 Quit krazykit (Remote closed the connection)
04:06:25 Part denstark
04:09:04 Join Echelon [0] (
04:10:28 Quit roxfan (Connection reset by peer)
04:12:23 Join krazykit [0] (
04:16:44 Quit woodensoul ()
04:18:59 Join roxfan [0] (
04:19:57 Join Administrator [0] (n=chatzill@
04:20:57 Nick Administrator is now known as Davide-NYC (n=chatzill@
04:23:16Davide-NYCHello all, long time no speak. I see that speex is now supported. Is there an encoder? If not is one the owrks? I'm not familiar with speex but would have use for it especially the noise suppression and echo reduction feature I read about.
04:23:53safetydanDavide-NYC: yes speex has an encoder
04:23:57safetydanoh wait
04:24:04safetydanno, there's no recording support in Rockbox for speex
04:24:11*safetydan is a bit slow
04:24:23jmworxDavide-NYC: I think preglow's working encoder support in rockbox
04:24:26safetydanI think preglow or jhMikeS might be looking at it
04:25:05Davide-NYCam I understanding correctly that speex will apply noise reduction and echo reduction at the recording stage?
04:29:21natureI Isolated my error (I hope you don't mind if I leave one of you to do the bug report, Ive already spent way too long on this). I only get the "Battery empty! RECHARGE! shutting down" phenomenon on r15679. r15654 and everything before it boot without battery worries. every after 15613 broke for other reasons though, so i am using r15613 from the 14th. I guess I just picked an unlucky day to update. Once again, I'm on a 3rd gen ipod
04:30:16Davide-NYCfor the record, according to the site the echo canceler included included but (as of now) a denoiser is not.
04:30:32n1snature: got cut off at "3rd gen ipo"
04:31:12natureipo became ipod. after that I just said "Thanks! for all your work!"
04:31:39n1sDavide-NYC: iiuc thos sound like recording filters that should not have anything to do with the format...
04:32:11Davide-NYCn1s: please clarify
04:33:34n1sDavide-NYC: a filter removing echo or noise could (again iiuc) be applied to any audio before it was encoded and so should not depend on which format you encode to
04:34:32Davide-NYCfrom what I'm reading on the speex site (which is why I'm so curious about this format) the denoiser and the echo remover are part of the actual codec
04:34:54LloreanWhich makes sense since it's a format not designed for overall audio preservation, but very specific intended uses.
04:35:09LloreanSo certain filters can also make it easier to compress, I assume
04:35:14 Part TFGBD
04:35:46n1sDavide-NYC: all i am saying is that i see no reason for those functions not to work for all recording format if someone implements or ports them to rockbox
04:36:09Davide-NYCExactly. Currently, for recorded meetings, I have to grab a piece of background noise (usually HVAC) and generate a noise signature, which I then remove. It's a pain.
04:37:02 Quit japc (Read error: 110 (Connection timed out))
04:37:10n1sDavide-NYC: and btw encoding is pretty cpu intensive so unless those filters are implemented very efficiently it will not be possible to do while encoding to all formats
04:37:25jmworxDavide-NYC: AEC and denoise are in the same package, but independent from the code
04:37:48Davide-NYCjmworx: now I understand. Thanks.
04:38:17Davide-NYCis the denoise any good? I mean, is it intelligent enough?
04:38:55n1sjmworx: do you think it will be possible to make encoding speex and applying those filters work in realtime on our targets?
04:39:16Davide-NYCcoldfire in particular (for me at least)
04:39:27jmworxn1s: yes
04:40:09Davide-NYCwould it be possible to implement the AEC and denoise as 'recording plugins', useable with any codec?
04:40:35jmworxDavide-NYC: the denoiser doesn't generally increase intelligibility because the human brain is almost always better than any algorithm you can come up with. However, it makes the audio *sound* better.
04:41:04Davide-NYCI use denoise (in Audacity) when I find it hard to hear a quiet speaker.
04:42:04Davide-NYCI also normalize either before or after when necessary, use compression, etc.
04:42:04 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:42:08jmworxDavide-NYC: You can try the Speex denoiser easily from speexenc (just use a high quality level just to remove any effect Speex might have).
04:42:20 Part radinp
04:42:29 Join miepchen^schlaf [0] (
04:42:30Davide-NYCI'll give that a shot.
04:43:12jmworxDavide-NYC: Actually, the Speex denoiser has an "aggressiveness" tuning where you can trade in noise for distortion. It's not available from speexenc though
04:44:13Davide-NYCI honestly doubt it would be intelligent enough for my purposes. Imagine 7 voices all very different in both level and tone with an HVAC hum in the background.
04:44:44Davide-NYCI doubt I'd use it unless it really did work very well.
04:46:54Davide-NYCTo the Rockbox community (especially the developers) I want to thank you again for such an excellent firmware.
04:47:07 Quit Davide-NYC ("ChatZilla 0.9.79 [Firefox]")
04:55:00 Join jpt9 [0] (
04:55:09jpt9does rockbox have any sort of onboard ID3 tag editor?
04:55:11jpt9it'd be useful.
04:55:22psycho_maniacit does not.
04:56:15jpt9it'd be neat if you could just go to track info and hit a button to edit the selected field.
04:58:15 Join criznach [0] (
04:58:20 Quit xaNz ("Leaving")
04:58:29 Quit JdGordon (Read error: 104 (Connection reset by peer))
04:59:28 Join JdGordon [0] (
05:03:33n1sjpt9: it would also be quite complicated and text input on those devices isn't that great...
05:04:18n1sI am however quite sure that we would accept a good tag editing plugin :)
05:05:02n1sthere is one in the tracker but i think it's abandoned and supported only id3v1
05:06:02jpt9of course the text input sucks...
05:06:06jpt9but it lets you rename files.
05:06:10jpt9and there's a text editor.
05:06:54n1syes it's entirely possible to make this plugin but it will be slow to input the text.
05:07:30jpt9it would be useful to correct small errors in tags (or possibly enter new ones for a file or two)...
05:07:51 Join ord [0] (
05:08:02 Quit nature (Remote closed the connection)
05:08:04psycho_maniacstart writting a plugin jpt9 :)
05:08:30n1syeah it could be useful, I'm just a little realistic at 5am
05:09:08psycho_maniacit would be useful yes, but imo a computer is easier/faster
05:09:15jpt9realistic is not a reasonable critereon.
05:09:24jpt9there's a fractal renderer.
05:09:27jpt9and a gameboy emulator.
05:09:39 Join animeloe [0] (n=animeloe@unaffiliated/animeloe)
05:09:44***Saving seen data "./dancer.seen"
05:09:50PaulPosition"shutdown if hold is on in the non-ipod-pp target bootloaders" ... what does that mean for 'switch hold to reset settings' then?
05:09:51jpt9and a ZX Spectrum emulator!!!
05:09:58jpt9which has a keyboard!
05:10:03psycho_maniacjpt9: whats your point?
05:10:19n1sa gameboy has what 8 buttons, an computer keyboard, about 105, your dap 8~10 maybe...
05:10:33jpt9how about the ZX Spectrum?
05:11:15LloreanPaulPosition: The settings reset doesn't happen in the bootloader.
05:12:48psycho_maniacwhat does that commit mean anyways? if you put the hold button to "on" on your player it will shut off?
05:13:12n1sonly when it's turned on
05:13:24n1swhen it's starting
05:13:34*n1s stops typing now :)
05:13:47psycho_maniacoh ok. i understand
05:14:00 Quit animeloe (Client Quit)
05:14:34PaulPositionAh.. So there are targets where the hold button didn't intercept/block the power-on action?
05:14:55PaulPosition(ie, iRiver would flash a padlock if I tried that)
05:16:09 Quit n1s ()
05:16:50JdGordonPaulPosition: you talking about my commit?
05:17:17 Join animeloe [0] (n=animeloe@unaffiliated/animeloe)
05:17:48PaulPositionWell I was asking about it, yes.
05:18:06 Quit animeloe (Client Quit)
05:18:22PaulPositionJdGordon - but not demanding. I'm just a bit curious about things. :/
05:18:32JdGordonthe problem happened when the e200 bootloader was replaced with a rockbox bootloader...
05:19:00JdGordonthe OF bootloader checked for the hold switch not rockbox.. so that needed fixing
05:19:16 Quit qwm (Read error: 110 (Connection timed out))
05:19:21 Nick ord is now known as qwm (
05:20:54PaulPositionJdGordon - Okay... What will happen, now, with the wakeup-alarm on, say, h10 ? And how much time would someone wait before flippin' the hold switch if one wanted to reset settings?
05:21:26JdGordonas soon as it comes on you can flip the switch...
05:21:37JdGordonif its too early it will shutdown...
05:22:11LloreanGenerally speaking, say/think "one" and then you're okay. :-P
05:22:47PaulPositionOkay, so when the screen starts showing stuff, bootloader has pretty much finished his stuff?
05:22:58 Join DM| [0] (
05:23:32LloreanOn the iPods at least, the best time is probably right after the backlight comes on, I believe
05:23:33PaulPositionThat's good enough. Just wondered because I happen to offer "reset settings" a lot as solution for trivial things on forums such as misticriver :D
05:24:49PaulPositionthanks :) I'll still have to test the alarm with hold on, to see what happens. And maybe amend the manual (?) If I can find in me the clairvoyance to write good english.
05:25:15Lloreanalarm+hold wouldn't work well anyway
05:25:24LloreanYou'd get cleared settings, right?
05:25:34LloreanNow it just won't boot up, less destructive I'd say
05:25:52PaulPositionGood thing you mention it, that makes a whole lot of sense.
05:26:10LloreanI doubt playback would start on the cleared settings one anyway, though I'm not _certain_ about that.
05:26:39PaulPositionSo in one case it'd be : (before) dap turns on, clear settings, nothing happens (no song to resume) - (after) nothing happens.
05:26:47PaulPositionThe after is better indeed. :D
05:26:48LloreanI think so.
05:27:13LloreanBefore *might* have been "clear settings, resume playback with default wps" as I'm not sure whether clearing settings would happen before or after it attempted to resume.
05:27:22JdGordonyes, clearning settings definatly disabled reusming playbacl
05:27:35LloreanI know it'd disable a start screen of the WPS, but is the alarm handled in the same place?
05:28:31JdGordonyay, back up to manufac mode
05:28:44PaulPositionLlorean - I think so, there's no "browse to alarm song" sort of thing, it just plays the last song before the poweroff.
05:29:10JdGordondepends on the default value for global_settings.alarm_wake_up_screen
05:29:27JdGordonbut even then... the resume info is cleared also
05:29:32JdGordonso it would say nothing to play
05:29:41LloreanDidn't know clearing settings removed nvram
05:30:00*JdGordon is pretty sure it does
05:30:14JdGordonyeah it would.. nvram settings are reset to their defaults
05:30:24LloreanWell then, definitely better this way
05:30:32LloreanNothing lost, and it wouldn't have woken you in the first place
05:31:09PaulPositionlol... Too true. Well, god bless irc for sparing me a "me stupid" moment on the forums.
05:36:18 Join Guerin [0] (
05:36:25Guerinhey hey
05:36:39Guerini had rockbox on my x5l and the battery life was measured in minutes,m not hours
05:37:06LloreanGuerin: Did you run a proper battery bench?
05:37:18GuerinLlorean: no, just ordinary usage
05:37:18PaulPositionGuerin - You mean what rockbox showed or what you got out of it and measured yourself
05:37:36LloreanGuerin: Rockbox uses less battery power than the original firmware under similar playback conditions.
05:37:44GuerinPaulPosition: it would die due to low batter after 30m-3h
05:38:16GuerinLlorean: yes, I've read that, since it underclocks the cpu when not in use
05:38:31LloreanTesting has repeatedly confirmed this, time and again.
05:38:51Guerinwhat I wonder is whether I needed to tweak down the drive seek settings - I'm reluctant to go too low for fear of parking the hdd too frequently
05:38:53LloreanTry running a full battery benchmark and finding out the voltage range your player is crossing over time.
05:39:10LloreanGuerin: Have you changed it from the default?
05:39:23GuerinLlorean: I had not, no.
05:39:33LloreanThen it shouldn't be a problem.
05:39:39LloreanAre you using a recent, official build?
05:39:39DM|Anyone suggest a nice and pretty theme that works with daily builds?
05:39:53Guerinthe other thing is that I was using a prebuilt rockbox, not one I'd built myself from cvs or svn or whatever you use
05:40:02psycho_maniacAnybody ever ran the files on the GigabeatFXPort page? I'm wondering how risky that is
05:40:06LloreanGuerin: Downloaded from
05:40:18LloreanYou don't have to build it yourself, but if it's not the officially built version, you shouldn't be asking about it here
05:40:18psycho_maniacDM|: what player?
05:40:29DM|ipod 30gb 5g Video, sorry
05:40:39LloreanThere are many things other people put in their builds that can have averse affects on battery life, or introduce bugs such as shutting down when the battery isn't really too low yet...
05:41:14Lloreanpsycho_maniac: They work fine.
05:41:21GuerinLlorean: OK, so the binmaries on are considered sane?
05:41:25psycho_maniacthey usually say if they need patches or not. go to that one is easier to see a theme needs patches or not
05:41:38LloreanGuerin: The binaries on are considered "The only ones we provide support for"
05:41:53LloreanWe refuse to be responsible for any changes other people might've made.
05:42:05DM|psycho_maniac well mos the ones on there load up with text in the wrong spot, or text that is garble
05:42:17psycho_maniacdo you use patches?
05:42:43psycho_maniacload up where? on the site or on the player?
05:42:43PaulPositionDM| - And they also tell you whether they're safe to use with svn builds or if they need a custom one.
05:43:07DM|PaulPosition Im using ones that say i can use with daily builds
05:43:43GuerinLlorean: perfectly rational. What (other than the battery benchmark) should I be looking at if it continues to die out? I have used two separate batteries - the original x5l set (used ~1 year, then in storage ~18 months), and a replacement (brand new) OEM 1100mAh
05:43:56Guerinmore or less identical battery life between them, which is frightening
05:44:39Guerinalso, both seem to only charge up to ~4.05v, whereas if they're 4.2v nominal they should go to at least that level, no?
05:44:40PaulPositionDM| - That's weird. Wish I could test some for you and see for myself but I don't have an ipod. :(
05:44:47LloreanWhen you take a battery bench, it'll show the voltages at each time it records, you can ask other people with good working ones and see how their voltages compare
05:45:51PaulPositionGuerin - How old is your build, by the way? (If you're going the battery-bench way, might as well do it with the most current build)
05:45:57Guerinis it possible to run a battery bench of any sort on the original firmware for comparison?
05:46:25GuerinPaulPosition: tbh, I did it several months ago and don't recall how old the build was
05:47:07PaulPositionGuerin - Except playing songs and 'giving an ear every 30 minutes' or so, I don't think there's any way to benchmark OF.
05:47:27GuerinPaulPosition: ok.
05:49:24PaulPositionGuerin - I can't say for your specific target, but sometimes things break and are corrected soon after.. If you were really unlucky maybe you downloaded a 'crippled' build that one time..(?) There are about a dozen commits a day and some of them sometimes have to be 'reverted' so who knows?
05:49:42Guerinright, that's cetainly possible
05:50:04Guerinit seems charging up to only just above 4v is normal, though
05:50:22Guerinand about 3.5-3.6v is shutdown time
05:52:11Guerinso, I should be using a daily, or which?
05:53:29PaulPosition"current build" is what the doctor usually orders.. ;)
05:54:03Guerinthank you
05:54:26PaulPositionGuerin - If you do the benchmark, you can compare with Bernard K.'s benchmark which is linked to on this wikipage :
06:01:01 Join monsterb [0] (
06:05:04 Quit monsterb (Remote closed the connection)
06:05:10 Quit jpt9 ("ChatZilla 0.9.79 [Firefox]")
06:08:13 Join sarixe [0] (
06:14:36 Quit sarixe ("(EE) Failed to load "quit" module")
06:19:11 Quit criznach ("ChatZilla 0.9.79 [Firefox]")
06:27:49 Quit psycho_maniac (" I love my HydraIRC -> <-")
06:29:53 Join animeloe [0] (n=animeloe@unaffiliated/animeloe)
06:31:07 Quit zicho (Read error: 110 (Connection timed out))
06:32:55marazSomeone should unlock and link people to the patch so we can get some testers.
06:33:57 Quit joshin (Read error: 113 (No route to host))
06:34:00marazThere is a patched recent build at
06:35:24DM|How do you clear rockboxes' database?
06:37:46 Quit hcs (Read error: 110 (Connection timed out))
06:39:06 Join toffe82 [0] (
06:43:23 Quit DM| ("*bashes head against keyboard*")
06:44:33 Quit scorche (" HydraIRC -> <- It'll be on slashdot one day...")
06:49:10 Join animeloe_ [0] (n=animeloe@unaffiliated/animeloe)
06:51:54 Join Fardel [0] (
06:52:12FardelCan someone help me out with installing rockbox?
06:52:31krazykitonly if you tell us what device and what you're having problems with, of course
06:52:46FardelAlright, i've got an 80GB I pod.
06:52:52FardelAnd my problem is with the bootloader
06:53:23krazykitwhat's the problem, specifically?
06:53:54FardelI've ran it, and it starts fine, I type in "i", it starts working, and then it says.. "Write Failed
06:54:05FardelI've ran it, and it starts fine, I type in "i", it starts working, and then it says..
06:54:15Fardel[ERR] Write failed
06:54:21Fardel: No error
06:54:30krazykitdo you have administrator rights?
06:54:34Fardel[ERR] Image movement failed.
06:54:44Fardel[ERR] −−install failed.
06:54:52FardelPress ENTER to exit ipodpatcher
06:54:56krazykitright, got that.
06:55:03krazykitwhat OS?
06:55:10FardelI've tried it with .rockbox (Unzipped) in my I pod, and out of my I pod
06:56:39FardelOh, windows.
06:56:47krazykitxp? vista?
06:57:32 Join webguest61 [0] (i=7cb85a3e@gateway/web/cgi-irc/
06:57:55advcomp2019Fardel, which 80gb ipod
06:58:42webguest61Hi is there anyone working on the Gigabeat S port at the moment?
06:59:21 Quit animeloe (Read error: 110 (Connection timed out))
06:59:32toffe82webguest61: yes there are 2 persons really working on it
07:03:04 Join DM| [0] (
07:03:05FardelAny idea of what might be going wrong?
07:03:41DM|Anyone know how to clear the rockbox database? i have duplicated songs on there but Those songs ( backup folder on the ipod) was deleted
07:05:09krazykitDM|, delete the .idx files in .rockbox
07:05:39 Quit eigma (Read error: 113 (No route to host))
07:06:15DM|krazykit uh.. there are no idx files
07:06:32krazykitmaybe that isn't the extension.
07:06:42DM|you mean the databae_X files?
07:07:02krazykityeah, that's the one
07:07:26krazykitnot sure why i thought idx.
07:07:37DM|database_idx.tcd was a file
07:07:49krazykitthat'd be why, probably :P
07:08:07DM|thanks much !
07:09:23 Quit DM| ("*bashes head against keyboard*")
07:09:45***Saving seen data "./dancer.seen"
07:10:20FardelDo you know what I should do?
07:10:56advcomp2019i would follow the step again
07:11:18FardelAnd unzip the file into my Ipod?
07:11:35 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
07:13:28 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
07:13:50 Join miepchen^schlaf [0] (
07:14:38 Join webguest16 [0] (i=42d07bf0@gateway/web/cgi-irc/
07:14:38 Quit webguest61 ("CGI:IRC (EOF)")
07:16:18 Quit webguest16 (Client Quit)
07:17:54FardelAlright, I did it all again, and it didn't work.
07:18:54FardelJust does the same thing
07:21:11FardelIs there anything else I can do to make this work?
07:29:53 Quit animeloe_ ("Leaving")
07:34:08 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
07:41:12 Part Fardel
07:43:00 Join ddalton [0] (
07:43:30 Quit ddalton ("leaving")
07:45:08 Quit dubious` ()
07:47:09 Join animeloe [0] (n=animeloe@unaffiliated/animeloe)
07:48:51 Quit Nico_P (Remote closed the connection)
07:51:06 Quit scorche (Read error: 104 (Connection reset by peer))
07:56:16 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
08:01:01 Join animeloe_ [0] (n=animeloe@unaffiliated/animeloe)
08:02:09 Join TiMiD[FD] [0] (n=TiMiD[FD@
08:03:23TiMiD[FD]I was wondering if someone see an objection to the renaming of the "recorder" and "player" folders into the app/ dir to something more appropriate to the fact that rockbox is now a little more multiplatform
08:05:38 Join ddalton_ [0] (
08:05:51ddalton_can someone help me here with setting my path?
08:06:13TiMiD[FD]nhm ?
08:06:46ddalton_what line do I edit in /etc/profile? (root's or my user one)
08:07:16TiMiD[FD]depends on your distro I guess
08:07:26TiMiD[FD]and on what you wanna do
08:07:47ddalton_I want to be able to compile from the daniel account. I can show you my /etc/profile file
08:07:59 Quit PaulPosition ()
08:08:35TiMiD[FD]not really necessary
08:08:43ddalton_that is the profile
08:08:50TiMiD[FD]so what you wanna do is modify your path
08:09:07TiMiD[FD]to be able to use the cross compile tools
08:09:12ddalton_I am running find on m68k-elf-gcc...
08:09:31ddalton_so I can compile rockbox basically
08:09:48TiMiD[FD]then add the path to the m68k-* tools on the line for the normal user
08:10:02TiMiD[FD]the second one in your profile
08:10:23ddalton_the second path line? after the else?
08:10:29 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:11:05ddalton_what do I add?
08:11:14ddalton_and do I remove the "?
08:11:22ddalton_and put it at the end of the edited line?
08:11:23TiMiD[FD]do you know where your binaries were stored ?
08:11:39ddalton_how do I check?
08:11:48TiMiD[FD]put a : after the /usr/games
08:12:01TiMiD[FD]and put your path to the binaries between the : and the "
08:12:10TiMiD[FD]I don't know
08:12:15TiMiD[FD]did you compile the tools ?
08:12:31ddalton_I ran
08:12:43ddalton_where does that put them by default?
08:12:54TiMiD[FD]so it may have created a folder with a bin/ dir inside
08:13:14TiMiD[FD]I don't use that script on a daily basis so I don't remember ...
08:13:17LinusNddalton_: /usr/local/<architecture>/bin
08:13:28 Quit jhulst (Read error: 113 (No route to host))
08:13:51ddalton_LinusN: Ok so i386? and whats the full line I need to add?
08:13:58TiMiD[FD]ah yes
08:14:00LinusNi.e /usr/local/arm-elf/bin, /usr/local/sh-elf/bin and /usr/local/m68k-elf/bin
08:14:02TiMiD[FD]mine are there /usr/local/m68k-elf/bin
08:14:14ddalton_a : between each?
08:16:21ddalton_LinusN: What about i386 do I need to add that somewhere?
08:16:54ddalton_LinusN: Can you check my profile if I pastivin it?
08:17:28 Quit animeloe (Read error: 113 (No route to host))
08:17:43 Join Rob222241 [0] (
08:18:43 Quit safetydan ("Leaving.")
08:19:50ddalton_LinusN: is this right?
08:20:14LinusNddalton_: looks ok
08:20:21 Join fpc [0] (n=gandalf@
08:20:26ddalton_LinusN: Ok Ill try it. Thanks
08:21:23fpcHello there
08:21:50fpcI downloades the current-buil for my ipod nano first gen, and the files arent listed...
08:21:59fpcdownloaded. build...
08:22:53fpcactually, I tried a last week build, and the firware was very unstable...
08:23:05fpcsomeone can recomend me a build that is know to work on my first gen nano?
08:23:26ddalton_LinusN: Same error. m68k-elf-gcc command not found
08:23:43ddalton_is there a way to check im pointing to the right directory?
08:23:56TiMiD[FD]ddalton_: type . /etc/profile
08:23:58TiMiD[FD]and retry
08:24:07LinusNddalton_: did you restart cygwin?
08:24:14TiMiD[FD]the changes are probably not taken in account
08:24:15ddalton_no I am on debian...
08:24:28 Quit BigBambi (Remote closed the connection)
08:24:55LinusNddalton_: hmm, then it might help to start a new shell
08:25:17LinusNbut i suspect /etc/profile might not be used if you use the bash shell
08:25:29LinusNthen it might be /etc/hash_profile
08:25:35LinusNsorry, bash_profile
08:25:40ddalton_anyway Ill be back in 20 mins I am using bash
08:25:53TiMiD[FD]. /etc/profile would also works
08:26:42daurnimator/etc/profile is used by bash
08:26:47daurnimatorcept if the user is root
08:26:49LinusNdaurnimator: ok ,thx
08:27:04daurnimatoryou have to use ~/.bash_profile for root
08:29:47LinusNah, yes
08:31:21 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
08:34:36 Quit Rob2222 (Read error: 110 (Connection timed out))
08:38:36ddalton_hi LinusN: Ok im back. just testing...
08:41:22ddalton_Linusn! I think its working!
08:42:13ddalton_so LinusN what should be quicker? 2.1 ghz dual core pc with cygwin building rb? Or 800 mhz pc with debian building rockbox
08:42:23ddalton_cygwin one has 1 gb ram and the debian one has 256.
08:42:27ddalton_I guess the cygwin one?
08:43:13LinusNddalton_: hard to tell, but the dual core might be faster
08:43:31 Join ender` [0] (
08:43:35LinusNthe debian box could benefit from some more ram
08:43:41ddalton_oh ok. I will probably end up installing linux on that. Just learning it before I break I break the good system :-)
08:44:17ddalton_BTW I can actually watch the whole compile process. The access tecnology I am using is far better
08:44:31 Join webguest16 [0] (i=4b24cc4c@gateway/web/cgi-irc/
08:45:15ddalton_LinusN: would I get improvements with 512 mb?
08:45:16 Part pondlife ("Gone")
08:45:37ddalton_would that be quicker than the dual core building rb?
08:45:50ddalton_Its going fairly fast. Up to the plugins. So It seems quicker
08:46:43 Quit fpc (Read error: 110 (Connection timed out))
08:48:57webguest16Is a build of Rockbox in the making for the Microsoft Zune (first gen)?
08:50:06 Join GodEater_ [0] (
08:52:10ddalton_LinusN: How do I check the size of
08:55:35GodEater_ddalton_: ls -lh
08:55:37LinusNls -l
08:56:01GodEater_yeah or just l if you want a really accurate size...
08:57:01 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
08:57:13 Quit webguest16 ("CGI:IRC (EOF)")
08:57:28ddalton_so would this be the size? Before the date? 1889080?
08:57:34ddalton_1.8 mb I guess?
08:57:46GodEater_that looks right
08:58:07ddalton_and another stupid question :-) how do I extract it to /media/iriver?
08:58:21 Part toffe82
08:59:24ddalton_unzip /media/iriver ?
08:59:30GodEater_not quite
08:59:38ddalton_oh ok
08:59:39GodEater_ddalton_: unzip -d /media/iriver
08:59:41ddalton_what is it then?
08:59:45GodEater_I usually use "-o" too
08:59:55GodEater_which overwrites the old .rockbox folder without asking you
09:03:17LloreanI thought it was unzip -d /media/iriver
09:03:25LloreanOr does it not care about order of parameters?
09:05:08 Join petur [0] (n=petur@rockbox/developer/petur)
09:07:12 Join OlivierBorowski [0] (
09:08:15 Join akoelahpahlawanb [0] (n=dshdks@
09:09:07ddalton_<GodEater_> ddalton_: unzip -d /media/iriver rockbox.ziso its unzip -d -o /media/iriver rockbox.zipp
09:09:46ddalton_unzip -o -d /media/iriver
09:09:49***Saving seen data "./dancer.seen"
09:10:51 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
09:10:55 Quit nanok (Read error: 113 (No route to host))
09:15:17 Quit akoelahpahlawanb ()
09:15:32*LinusN is impressed with how smooth the transition to speex has been so far
09:16:00jmspeexLinusN: transition from what?
09:16:07LinusNfrom mp3 clips
09:16:14LinusN(the voice)
09:16:29jmspeexyeah, just wasn't sure what rockbox was using
09:16:57 Join atsea- [0] (i=atsea-@gateway/tor/x-8e0441be2aac3586)
09:18:04LinusNthe quality is slightly worse, but i guess it's just a matter of changing the encoder parameters
09:18:38ddalton_LinusN: is it: "unzip -o -d /media/iriver"?
09:18:55ddalton_to unzip to /media/iriver
09:19:07LinusNddalton_: yes
09:19:14ddalton_LinusN: thanks
09:21:09 Quit advcomp2019 ("Leaving")
09:22:30 Quit linuxstb (Read error: 110 (Connection timed out))
09:22:51 Quit OlivierBorowski (Remote closed the connection)
09:23:52 Part TiMiD[FD]
09:25:48 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
09:26:34 Join eigma [0] (i=eigma@
09:27:28linuxstbJdGordon: Regarding your bootloader commit, what happens if you turn the device on by inserting USB or the charger, and the hold switch is on?
09:27:41jmspeexLinusN: what bit-rate were the MP3 files?
09:28:05eigmaDMA-driven (from DSP RAM to serial port) audio output is working on the m:robe! :D
09:28:27 Quit ddalton_ ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
09:28:29LinusNjmspeex: not sure
09:28:29Lloreanlinuxstb: The e200 at least, shouldn't boot up if the hold switch is in (the OF bootloader used to prevent it when present)
09:28:32eigmaI've got a nice ~1.4khz sine wave being outputed
09:28:40 Join ddalton [0] (
09:29:13 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
09:29:16JdGordonlinuxstb: it will shutdown which i would think is expected... although letting it boot to usb with hold on does sound ok also
09:29:18LinusNeigma: weeee!
09:29:38linuxstbJdGordon: And then it stays off? The ipods would then turn back on again...
09:29:48eigmano ARM->DSP bulk transfers yet though, I'll have to leave that to another day..
09:30:08JdGordonlinuxstb: yeah... and who says the ipods are always right :D
09:30:10eigmathe C54x DMA nicely has this "circular buffer mode" so i can literally set it and then entirely idle the CPU
09:30:27linuxstbJdGordon: So that means you can power-off the H10 and Sansa whilst charging?
09:30:30eigmanot even fixing the values every once in a while in an interrupt, it's awesome
09:30:54LinusNeigma: tried the free ti dsp compiler?
09:31:11Lloreanlinuxstb: I thought the iPod bootloaders don't have that change.
09:31:20eigmaI can't seem to get it to go into the lowest-power mode.. only the first-level idle mode :(
09:31:23JdGordonlinuxstb: I dunno
09:31:31Lloreanlinuxstb: But the e200 will turn on, say "Hold engaged" or something similar, and shut down, if you plug a OF-bootloader one into USB.
09:31:51LloreanSo I would assume you can either power it off while charging, or the bootloader enters a low power mode that looks like off.
09:32:04JdGordonlinuxstb: the commit was really only for when the OF bootloader is removed on the e200
09:32:18eigmaLinusN: not yet.. I'm putting that off because I don't want to think about integrating it with the rockbox build system
09:32:47eigmaif anyone else was interested in doing that, i would be much indebted to them ;)
09:33:20eigmaanyway, it's extremely late over here and a monday tomorrow (today..), I have to get some sleep
09:33:47 Quit eigma ()
09:45:47linuxstbLlorean, JdGordon: So you're saying that you've tested the new Rockbox bootloader and it stays shut off with both a USB connection and charger connection (assuming the Sansa has a separate charger)? Anyone tested the C200 and H10s?
09:46:06Lloreanlinuxstb: No, I was just saying what the OF bootloader did.
09:46:37LloreanIn the theory that if ours doesn't do it, because the OF one does that, we should be able to replicate it (if desired)
09:47:40JdGordonlinuxstb: hmm.. seems it does boot to the of if usb + hold
09:51:57 Quit TMM ("Ex-Chat")
09:53:03 Join davina [0] (
09:53:05 Join Zagor [0] (
10:12:26 Part pixelma
10:24:14 Quit ddalton ("Lost terminal")
10:35:36 Join Jon-Kha [0] (
10:43:37amiconnjmspeex: The voice files used (and for hwcodec targets still use) mp3 vbr: 12kHz, lame -V 9 -B 64 (the latter is necessary due to limitations of the MAS mp3 decoder)
10:45:34 Join spiorf [0] (
10:47:31 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
10:47:50*pondlife really likes the new voicebox
10:48:06pondlifei.e. the HTML app
10:48:20pondlifeIs it in SVN too?
10:49:59 Join Siku [0] (
10:56:21LloreanHTML app?
10:57:02LloreanWhere does this come from?
10:57:49pondlifeThe wiki
10:58:40GodEater_where it's always lived...
10:59:01pondlifeIt should also be in SVN, but I don't see it.
10:59:28pondlifeIt never was in before, but now would be a good time to add it, I guess
10:59:31LloreanGodEater_: I'd never seen an html one before.
10:59:41pondlifeNo, that's new
10:59:43LloreanAlso, I think we have too many voice pages on the wiki
10:59:49pondlifeWe do
11:00:01pondlifeWe shouldn't host voice files there
11:00:06LloreanVoiceHowto, VoiceBuilding and VoiceFiles
11:00:10GodEater_I've used an html based one for AGES
11:00:13GodEater_it can't be THAT new
11:00:25pondlifeAh, I only just saw it today ...
11:00:38LloreanGodEater_: The last time I generated talk clips, the H120 was the only swcodec target...
11:00:41GodEater_when you said "new" I assumed you meant it had been updated to do speex
11:00:43LloreanNew for me is very relative in that area.
11:01:20pondlifeI hadn't updated since Brian Wolven's original vbs
11:01:30LinusNSAPI is broken on my work machine, it asks for the installation cd, which i don't have
11:01:36jmspeexamiconn: I'm pretty sure Speex 16 kHz / 27.8 kbps sounds better
11:02:02pondlifeThe new .talk clips play back slower than my SAPI TTS setting though
11:02:36pondlifeI've moved the slider across to full speed, but it makes no difference... I guess that's being overridden somewhere.
11:02:37LinusNjmspeex: how do i set the options for that?
11:02:57LinusNjmspeex: the current options are "-q 4 -c 10"
11:02:58jmspeexLinusN: that's the wideband default
11:03:15jmspeexLinusN: but you can change all settings with speex_encoder_ctl()
11:03:33jmspeexLinusN: -q4 is a bit low I'd say
11:04:37*LinusN tries with -q 5
11:04:41jmspeexbut then again, it depends on the application
11:05:26LinusNit sounds ok, but it sounds a little too robotic, and the "s" and "f" are distorted
11:06:16jmspeexI tend to recommend using 20 kbps or higher for wideband
11:07:14pondlifeHmm, the voicebox script sets the speech rate :/
11:07:45pondlifeShould leave that to the control panel setting, like the voice selection IMHO.
11:09:52***Saving seen data "./dancer.seen"
11:10:16LinusNi'm not too sure about that
11:10:41LinusNi think the user is likely to want a different narration rate in rockbox than on the windows machine
11:10:43amiconnjmspeex: To me speex -q4 sounds considerably better than the old mp3 setting
11:10:45pondlifeI like to set a fairly fast rate and build both .voice and .talk
11:11:02pondlifeConsistency of voice on target is good
11:11:03JdGordonpondlife: had any ideas to fix that squeel on the keyclick patch?
11:11:19amiconnpondlife: The .hta was around for years...
11:11:20jmspeexamiconn: well, MP3 really sucks anyway. Also, the artifacts are very different between Speex and MP3 and they tend to be perceived differently
11:11:26jmspeexby different people
11:11:44pondlifeNo. The problem seemed to go away for GodEater, perhaps with jhMikeS' little pcmbuf mods?
11:11:59amiconnWell, for hwcodec we have no choice.... unless someone implements a speex codec for the MAS
11:12:00pondlifeI just resynced the patch, maybe you could retest?
11:12:13JdGordoni saw.. thats what promptyed the question
11:12:28JdGordonmy tree isnt compilable atm.. but ill test later or tomorow if i rmemeber
11:12:38pondlifeJdGordon: AFAIK, it's only a problem on Sansa at the moment.
11:12:51pondlifeBut I have no idea why it's a problem at all.
11:12:56LinusNi wonder if FS #8003 could be solved by the new voice system
11:13:00JdGordonyeah, but it cant go in untill its fixed, and i want it in :p
11:14:17pondlifeLinusN: It's tempting to close FS #8003... :)
11:14:35LinusNi added a comment about retesting with the latest version
11:14:47pondlifeI was about to.
11:17:57 Nick Soap is now known as Soap_ (n=Soap@rockbox/staff/soap)
11:18:18JdGordonwhich idiot decided the cfg strings cant have spaces after the commas???!!!
11:18:22*JdGordon shooots self
11:18:55*Bagder hands JdGordon the band aid
11:20:24JdGordonyay, settings loading works.. now to fix saving.. and then the gui and its ready!
11:20:38pondlifeWhat's ready?
11:21:38JdGordonnew config setup for the backlight so any value can be given
11:21:55JdGordonget rid of those bloody arrays which have to be kept in sync between apps and firmware
11:23:42pondlifeHmm, the new voice makes keyclick somewhat inaudible.. :/
11:24:03pondlifePreviously it was a little slower to kick in, so you got the click first
11:24:17pondlifeGuess it'll have to wait for the mixer
11:27:14JdGordonwhy does strncpy() \0 the entire buffer upto len?
11:27:21JdGordonadding one \0 isnt enough?
11:28:02pondlifeWhat does POSIX say?
11:28:18JdGordonbut sounds like a waste of cpu to me
11:28:21linuxstbThat's what POSIX says...
11:28:26pondlife"If the array pointed to by s2 is a string that is shorter than n bytes, null bytes shall be appended to the copy in the array pointed to by s1, until n bytes in all are written. "
11:28:40 Join lee-qid [0] (
11:28:45pondlifestrncpy is a horrible function
11:28:51linuxstbAdding strlcpy has been suggested...
11:29:17JdGordonwhats strlcpy?
11:29:23linuxstbGoogle it.
11:29:31Bagderopenbsd's take at a better strncpy
11:29:37pondlifeOops, ignore that trailing #
11:29:50Bagderused all over these days
11:30:11linuxstbBagder: What do you think about it?
11:30:24Bagdera good idea
11:30:44pondlifestrcpy/strncpy should be deprecated too.
11:31:16JdGordonjust define strncpy to strlcpy
11:31:50LinusNi'm all for it
11:32:04linuxstbThat would be the same as just changing the behaviour of strncpy, which isn't what we want - we want to be POSIX-compliant as much as possible.
11:32:05 Join agm3nt [0] (
11:32:09JdGordon.... or we can just stop strncpy \0-ing the whole buffer
11:32:21JdGordonscrew compliance!
11:32:41pondlifeNo, strlcpy should be added, and we can deprecate strncpy/strcpy over time
11:32:47LinusNi like the truncation checking and the null termination more than the extra performance
11:34:09linuxstbSomeone also mentioned the issue of utf-8 characters being truncated, I don't know if strlcpy can/should do that.
11:34:31pondlifeNo, it shouldn't.
11:34:48linuxstbMaybe utf8_strlcpy() ?
11:35:23pondlifefor example
11:36:08*markun apologizes again for adding unicode support
11:36:34linuxstbAnyone volunteering to do it?
11:37:37*linuxstb knows how to silence a party...
11:37:48*LinusN looks away
11:38:02JdGordonshouldnt the who core be changed to only work with utf8 strings?
11:38:18markunJdGordon: I think so
11:38:38pondlifeYes, but that's a separate job.
11:38:43pondlifeOne thing at a time.
11:38:52JdGordonrestart apps/ from scratch!
11:39:09LinusNwhat would be the penalty for doing that?
11:39:27JdGordonamiconn: ok, saving/loading the .cfg works fine.. I need to fix the option screen still though
11:39:32LinusNi'm worried for the archos targets
11:39:56JdGordonI actually think it would benefit the archos more than hurt it
11:40:08linuxstbWhat needs doing differently? I thought the core worked with utf-8 currently?
11:41:37LinusNi believe all config file parsing uses plain 8-bit
11:41:48LinusNfor example
11:42:07JdGordonyep, good old ascii
11:42:13markunLinusN: but does that matter? Aren't the paths in the config file stored in UTF-8?
11:42:30LinusNnot sure
11:42:49LinusNi don't think they are
11:43:16JdGordonall strings stored int he config are stored as ascii
11:44:15markunJdGordon: what do you mean?
11:44:20markunconverted from UTF-8 to latin1?
11:44:31linuxstbDoes the config file care what the encoding is? Doesn't it just load/save whatever is in the strings in Rockbox, which afaiu is utf-8?
11:44:41markunlinuxstb: that's what I think too
11:45:06JdGordonhmm.. ok, maybe im confused then
11:51:30 Join safetydan [0] (n=safetyda@rockbox/developer/safetydan)
11:53:30LinusNhmmm, i just started my x5, and it didn't voice the initial menu choice, i.e "Files"
11:53:40LinusNshouldn't it do that?
11:53:56LinusNi believe it used to
11:54:17LinusNbtw, -q 5 sounded a lot better
11:54:45 Quit spiorf (Remote closed the connection)
11:55:07GodEater_are we any closer to getting 7738 commited ?
11:55:21pondlifeIt should definitely voice the initial menu option...
11:56:27pondlifeIt seems to try to do it, but misses the first bit
11:56:46pondlifee.g. if I start up in the Settings menu, it says "ttings"
11:57:25LinusNGodEater_: gets my vote
11:57:48GodEater_who else's does it need ? :)
11:57:50pondlifeLinusN: I think he wants you to commit it.
11:58:12LinusNGodEater_: i don't know. is it tested on all targets that it affects?
11:58:18*linuxstb votes for LinusN committing as well
11:58:34pondlifeLet's play the svn blame game :)
11:58:36GodEater_afaik no-one's tried it on 4G greyscale or the mini 2g
11:58:50GodEater_because we've not found anyone who owns those targets
11:58:55linuxstb4G greyscale is almost certainly identical to the Color
11:59:02GodEater_I would assume so too
11:59:04linuxstb(and I tested it on the Color)
11:59:43GodEater_if we're not going to commit it based on the fact that it's not been tested there than I don't know what to do
12:00:04linuxstbNo, I think the Color test is good enough for the 4G greyscale.
12:00:07 Quit hunz ("Lost terminal")
12:00:15GodEater_that still leaves the mini 2g
12:00:25linuxstbThe whole 4G greyscale port has been based on the fact that it's a Color with a 3G's LCD...
12:00:49pondlifeIf it's committed and fails on the mini 2g, the owners will let us know...
12:00:57pondlifeIf not, it'll sit there forever.
12:01:08LinusNi agree
12:01:09pondlifeAnd nobody will get the benefit
12:02:07*LinusN commits
12:02:48*GodEater_ pops the virtual champagne
12:03:11pondlifeIs that only for iPod owners, that champagne?
12:03:21GodEater_no - please help yourself
12:03:42*GodEater_ made sure it was a Jeraboam, so there's enough for a lot of people
12:04:17*pondlife is still not sure why we need landscape/portrait modes
12:05:22safetydananyone familiar with the uda1380?
12:05:51safetydanwhat effect do the mode bits have on the eq register?
12:06:10safetydanthe datasheet isn't really telling me much apart from the possible values
12:06:44 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
12:08:45 Nick fxb__ is now known as fxb (
12:09:21LinusNsafetydan: isn't that explained in table 42?
12:09:42*GodEater_ deletes his 7738 git branch
12:10:26safetydanLinusN, I guess. So it if set to "minimum" the maximum bass gain is capped?
12:10:34LinusNguess so
12:10:38LloreanDid someone ever get rid of the setting for 7738?
12:10:40safetydanseems like an odd setting
12:10:50LinusNsafetydan: i think so too
12:10:58LinusNLlorean: no settings in 7738
12:11:15LloreanIt used to be configurable scroll speed, I thought
12:11:27*safetydan was sure someone said the uda1380 supported configurable bass/treble cutoff
12:11:46LloreanLinusN: Ah, I see the newest version says it doesn't have it.
12:11:55LloreanI had the task on my watch list, but apparently didn't get emails when it updated despite that.
12:12:00*Llorean needs to check his spam box perhaps
12:12:38LinusNsafetydan: not to my knowledge
12:13:23*linuxstb wonders what else unofficial builds are including, now that there is album-art, scrollwheel acceleration and half the scrollmargins in SVN
12:13:50BagderI noticed at least one of the e200 builds use the old keymap ;-)
12:14:19Bagderthat's the reversed situation, the custom builds take away some recent things!
12:14:22Lloreanlinuxstb: Multifont, and "customline" whatever that is
12:14:43Nico_Plinuxstb: some include the multifont and other UI stuff
12:14:58linuxstbcustom line is the %e tag - which lets you specify x,y,width,colours and font for a line in the WPS
12:15:07linuxstbSo viewports will enable that functionality...
12:15:08GodEater_Llorean: I took the settings out of 7738 some time last week
12:15:15GodEater_with a bit of help from LinusN
12:16:21Nico_Plinuxstb: do you have colours in your viewport plans?
12:16:23*linuxstb prods markun and mentions multi-font
12:16:29LinusNhow does the multiple font patch look like?
12:16:45safetydanhrm, both amiconn and preglow have mention uda1380 cutoff control in the past
12:16:58LinusNsafetydan: odd
12:17:32linuxstbNico_P: Yes, viewports will have their own colours defined.
12:17:53Nico_Plinuxstb: nice :) how's progress btw?
12:18:27linuxstbStill at the planning stage...
12:18:27 Join PaulJam [0] (
12:18:50 Quit markun (Read error: 104 (Connection reset by peer))
12:19:58 Quit jhMikeS (Nick collision from services.)
12:20:04 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
12:20:11LloreanGodEater_: I'm just frustrated because the "Watch this Task" button in Flyspray didn't do me any good. :-P
12:20:25 Join markun [0] (n=markun@rockbox/developer/markun)
12:20:27Lloreanlinuxstb: Viewports will enable multiple fonts?
12:20:57linuxstbLlorean: No, but when the core font code supports multiple fonts, viewports will make their implementation easier.
12:21:41GodEater_Llorean: sorry - I'll email you personally next time :)
12:21:57GodEater_I thought it was a bit odd you were quiet on the subject
12:22:08LloreanAs far as I knew, it was still at version 12
12:22:16LloreanI was just assuming things were working, and nothing had changed in a while
12:22:18GodEater_ah - you missed quite a few iterations then :(
12:22:49linuxstbI don't recall getting any individual emails from flyspray for a long time - does anyone?
12:23:04*GodEater_ is still puzzled he managed to save the patch back without .lang edits he thought he'd already made
12:23:29Lloreanlinuxstb: My last one from flyspray is from yesterday
12:23:37Llorean12 hours ago.
12:23:53GodEater_does scroll wheel accel warrant a major changes entry ?
12:24:05*linuxstb realises he has disabled notifications...
12:24:12LloreanGodEater_: Yes
12:24:15BagderGodEater: I think so, adding entries are goodness
12:24:16GodEater_hehehe - genius ;)
12:24:28GodEater_I don't know if I have rights to the MajorChanges page
12:24:36linuxstbIt's just a normal wiki page
12:24:55LloreanI think that anything that generates at least 10 posts in the forums requesting it probably by default should get a MajorChanges entry once it happens. ;)
12:24:56GodEater_ah cool
12:24:59GodEater_I shall add it then
12:25:40linuxstbIt could be useful to list which ipods it works on, as the commit message didn't...
12:26:25*GodEater_ thought of that
12:26:47*linuxstb shuts up then
12:27:22linuxstbMaybe list Color/Photo explicitly though...
12:27:27GodEater_apologies if I generated two RSS updates for that, I missed the %BR% tag off first
12:27:36*GodEater_ must use preview more often
12:27:58linuxstbAs if by magic...
12:28:30LloreanThere might be one additional update, I added Photo/Color to the list, as people seem to frequently not think of those as 4Gs in the same way
12:28:37preglowacceleration \o/
12:29:17*Llorean has no desire to field the "Why did you leave out iPod Photo users, they're basically the same, aren't they?" question.
12:29:19 Quit J3TC- (Connection timed out)
12:29:47preglowsafetydan: it does, afaik
12:29:54preglowsafetydan: it has one setting for it
12:30:26safetydanpreglow, which register?
12:30:33preglowgimme a sec
12:31:08safetydanah, see now that I read section 8.4.2 it all becomes clear
12:31:31safetydanit is the mode bits that control it
12:33:03safetydanhrm, so that comes back to the problem that the db range of the bass setting changes depending on the cutoff
12:33:30LloreanGodEater_: "See below for list?"
12:33:37GodEater_Llorean: look again :)
12:33:52preglowsafetydan: i really can't find what i remember seeing, though...
12:33:53LloreanAh, you even got it in during the grace period
12:33:54GodEater_I realised as soon as I hit save I'd left it off =/
12:35:40preglowwell, i distinctly remember uda1380 having a cutoff switch, but i can't find it again......
12:36:30safetydanpreglow, if you read section 8.4.2 I think it's talking about the mode bits (the flat/minimum/maximum settings in table 39)
12:38:18Llorean sound good?
12:39:50LloreanI'd really like to improve the whole situation. More clarity on why we're strict, less need to enforce because the rules aren't 500 pages long so people can pick them up more quickly, etc.
12:40:10GodEater_sounds fine to me
12:41:21pondlifeI'd like to see less stickies in the Announcements forum....
12:41:50GodEater_yeah me too
12:41:53 Quit safetydan ("Leaving")
12:41:58GodEater_I think the merchandise one can be unstickied now...
12:42:13LloreanGodEater_: Too late on that suggestion
12:42:32pondlifeLlorean: Did we get as far as a cheque from Cafepress?
12:42:46markunpreglow: I was wondering, wouldn't it make more sense to have the balance setting in dB instead of percentage?
12:42:59Lloreanpondlife: Nope.
12:43:25LloreanI set the "send a check to me at the minimum amount you'll send a check for" and everything, but I think we sold a couple stickers or something
12:44:08GodEater_oh dear =/
12:44:10Llorean$6 total
12:44:15LloreanMerchandise is not exactly popular.
12:44:16pondlifeI already have a mug + T shirt. There was some competition, some guys in Sweden I think.
12:44:38maxkelleyI don't think it was marketed. I didn't hear about it :)
12:44:47GodEater_they're limited edition ones pondlife - not available through cafepress :)
12:44:54pondlifeRockbox blue
12:45:00LloreanActually, there was $12 in sales at one point, but someone returned something. :)
12:45:00pondlifeWell the mug is
12:45:05 Quit atsea- (Read error: 104 (Connection reset by peer))
12:45:45 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
12:46:13 Quit linuxstb (Nick collision from services.)
12:46:15 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
12:46:27JdGordonlinuxstb: your planning viewports? or joiinging in the collection "thinking about it" ?
12:47:24linuxstbJdGordon: I've been working on them over the last few days.
12:48:01pondlifeHmm, something's up with the time zone support on the forum.
12:48:21*JdGordon is keen to help out when its ready for more code monkeys
12:48:24linuxstbJdGordon: Here's my current work-in-progress notes about what I plan to do -
12:48:24 Nick fxb is now known as fxb__ (
12:48:40pondlifeGodEater_: You didn't post at 1:30am?
12:49:02linuxstbJdGordon: The last thing to work out is how scrolling lines will work...
12:49:58pondlifeAha, the forum timezone autodetects me a value of -5. It should be 5.
12:50:21linuxstbI had the same problem...
12:50:26JdGordonlinuxstb: that should go in a wiki.... but just quickly.. I was tinhking it would be better if apps/ doesnt actually know anything about the viewport.. I thought the current lcd setting functions should take a handle to a viewport (of which there would be X statically declared in frimware) so when apps wants to draw it would viewport_allocate(), then set the settings.. then draw
12:50:34JdGordoni dont tihnk it belongs in apps
12:51:22JdGordonand the viewport handle == 0 would be the "root" viewport (cover the whole screen) and 1 would be the root remote vp
12:51:34linuxstbI disagree. I think it's much cleaner this way. It's also how amiconn planned to implemented them.
12:52:25pondlifelinuxstb: Should each viewport know which screen it's on? i.e. remotes...
12:52:38JdGordonas for scrolling.. they shuold be made pixel based.. its the only sane way to do it
12:52:46JdGordonpondlife: imo yes
12:52:56linuxstbpondlife: I haven't got as far as thinking about the screens yet.
12:53:23pondlifeOK, just one more thing to juggle :)
12:54:36preglowmarkun: don't know, haven't tried
12:55:18pondlifeAlso, this talk of pop-up menus etc. implies a Z order. Might it not be simpler to add that later and start with a "viewports should not overlap, but if they do, the later one gets priority" system?
12:55:44linuxstbNo, the Z order is implied by the apps code, and which viewports they ask to be updated on the LCD, and in which order.
12:56:06pondlifeAh, ok. I was confused by the word "underlying"
12:56:42linuxstbAlthough the plan isn't to allow lots of overlapping viewports - the normal usage would be to split the screen into non-overlapping viewports.
12:57:19LinusNhow about nested viewports?
12:57:24linuxstbI imagine a pop-up menu would simply not clear the previous content of the screen.
12:57:52JdGordonlinuxstb: I understand why lcd_set_viewport() should be used, but I dont like the idea of calling that every time you want to use any lcd function (well.. once for each set of lcd_()calls)....
12:57:52LinusNi don't think we need overlapping ports
12:57:53linuxstbLinusN: There is no such thing.
12:57:57pondlifenested implies overlapping
12:58:15pondlifeAlbeit a special case
12:58:20JdGordonI thought that if the vp struct is in firmware/ the globals in the lcd drivers would be removed and replaced by the vp members
12:58:23LinusNof course
12:58:39linuxstbJdGordon: They are.
12:58:40JdGordon... but thats much more work to put in
12:59:19linuxstbAdding an extra viewport parameter to every LCD function would also be a lot of extra code...
12:59:40pondlifePlus more params = risk of SH1 code bloat
13:00:14JdGordonstill.. for some reason I dont like the idea of apps/ being able to set the vp members to arbitary values
13:00:47JdGordonbut yeah, other than that I agree with pretty much the rest of the txt
13:00:57JdGordonand especially like the yes no screen idea
13:01:00 Join moos [0] (
13:02:44JdGordonalso, the list widget doesnt need a seperate vp for the icons/cursor/scroll bar.. all 3 should go in 1
13:02:58linuxstbBut then how do you rearrange them left/right ?
13:03:14linuxstbAnd also, there will no longer be a left/top margin.
13:03:24JdGordonall 3 of those items are pixel positioned... so it wont make any difference
13:03:56linuxstbIt will do - scrolling lines scroll the full width of the viewport they are in.
13:04:13JdGordonyes but icons cursor and scrollbar are all statically sized
13:04:26*jhMikeS will do the clipped overlapped viewports thing...just no way out of it :)
13:04:29JdGordonthe contexnt for sure needs a seperate one
13:05:19linuxstbBut if that combined viewport is on the left, you want the items ordered one way, if they're on the right, then ordered another way.
13:05:20JdGordonalso, I wonder if it might be nice to add a void (*viewport_update)(void* data) member to the vp struct
13:05:44JdGordonok.. so? there has to be a setting saying which side to put it on.. so use that
13:06:13JdGordonthere is no need to go overboard with seperate vp's everywhere... as long as statically sized stuff stays seperate from dynamically sized stuff (i.e text)
13:06:16linuxstbIt just seems simpler to define separate viewports, then you don't have to worry about where they are.
13:06:21LloreanJdGordon: But there are no margins for the scrolling text.
13:06:27 Join spiorf [0] (n=spiorf@
13:06:34JdGordon? course there is....
13:06:41LloreanYou don't need to care if there are icons, or scrollbar, and if both are on the left, right, or opposite sides, the text itself scrolls within its one viewport.
13:06:41JdGordonthe viewport is the scrolling texts margin
13:06:58LloreanIf the viewport is, why are you putting the icons in the viewport with the text?
13:07:00linuxstbJdGordon: What would that viewport_update function be for?
13:07:07linuxstb(and what would call it?)
13:07:15JdGordonLlorean: reread my comment... im not...
13:07:33JdGordonlinuxstb: im not sure :p i keep tihnking it would be a good idea..
13:07:43LloreanJdGordon: Wait, are you suggesting it not be drawn to a viewport at all?
13:08:10JdGordontext in one vp, title in one, icons/scrollbar/cursor(icon) in another
13:08:15linuxstbLlorean: IIUC, JdGordon is suggesting one viewport for the list items, and one viewport for the icons/cursor/scrollbar. My proposal has 4 viewports.
13:08:16JdGordonmax 3 for lists... the title...
13:09:39LloreanI honestly think the icons, cursor, and scrollbar could each be separate viewports. I know I kinda wish I had the scrollbar on the right, and the icons on the left, and I use the highlight rather than the cursor.
13:09:41JdGordonthats really not all that important now though
13:09:55***Saving seen data "./dancer.seen"
13:10:13JdGordonlinuxstb: have you thought at all about how it merges with screen access yet?
13:13:15JdGordonLlorean: wouldnt icons on one side and the bar on the other look a bit funny? but I guess thats a valid reason to split them
13:14:39jhMikeSsplitting a list control into separate viewports? what would that gain? I can't see anything it could possibly help with that can't be done by just having the lists draw appropriately.
13:14:48 Join CaptainSquid [0] (
13:15:15LloreanJdGordon: icons on the left and bar on the right wouldn't, with left aligned text.
13:15:25linuxstbJdGordon: No, as I said earlier I haven't thought about screens yet.
13:15:28JdGordonjhMikeS: we have viewport euphoria at the moment :p
13:15:33*Llorean points to "List view" of many, many file managers.
13:15:51JdGordonLlorean: yeah, but they are on much larger "screens"
13:17:35GodEater_the filemanager on my iPaq looks like that too
13:17:40GodEater_and that's a pretty small screen
13:17:50jhMikeSicons and scrollbars are rarely separate windows
13:18:44GodEater_no - they'd be seperate widgets
13:18:53GodEater_but we're not implementing a widget toolkit here
13:19:01preglowwe should :>
13:19:12GodEater_after you ;)
13:19:30preglowi've got a feeling you won't see me doing any gui coding very soon
13:20:09GodEater_I think a toolkit port like that to Rockbox would be a little overkill
13:20:15GodEater_even something like FLTK would be overkill
13:20:17LloreanjhMikeS: They aren't really separate windows. They're just drawn within separate bounding boxes.
13:21:13 Quit jhMikeS (Nick collision from services.)
13:21:19 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
13:22:33preglowGodEater_: not a toolkit _port_
13:22:35*jhMikeS is glad the fiberoptic guys are coming today. these connection dropouts are annoying as hell.
13:22:38preglowwe'd have to roll our own
13:23:06preglowbut with touch screens and shit on the way, it's getting more relevant
13:23:56jhMikeSall you need is a stored state instance for a "control". it's mainly a bugger because of no malloc.
13:24:16linuxstbWouldn't all the controls just be declared statically?
13:25:37jhMikeSevery instance?
13:26:34GodEater_ :)
13:27:39linuxstbOr maybe on the stack...
13:30:27 Join advcomp2019_ [0] (n=advcomp2@
13:30:47 Quit spiorf (
13:30:47 Quit markun (
13:30:47 Quit petur (
13:30:47 Quit GodEater_ (
13:30:47 Quit Seed (
13:30:47 Quit Bagder (
13:30:47 Quit courtc (
13:30:47 Quit Soap_ (
13:30:47 Quit tedrock (
13:30:47 Quit Langly (
13:30:47 Quit jmworx (
13:30:47 Quit ATravelingGeek (
13:30:47 Quit Siku (
13:30:47 Quit Jon-Kha (
13:30:47 Quit Slasheri (
13:30:47 Quit jhMikeS (
13:30:47 Quit moos (
13:30:47 Quit PaulJam (
13:30:47 Quit roxfan (
13:30:47 Quit iamben (
13:30:47 Quit homielowe (
13:30:47 Quit sneakums (
13:30:47 Quit parafin|away (
13:30:47 Quit sbeh (
13:30:47 Quit BjoernErik (
13:30:47 Quit bnakiddmj (
13:30:47 Quit lids (
13:30:47 Quit lastebil (
13:30:47 Quit ryanakca (
13:30:47 Quit gromit` (
13:30:47 Quit DataGhost (
13:30:47 Quit tierra (
13:30:47 Quit Hadaka (
13:30:47 Quit pondlife (
13:30:47 Quit Rob222241 (
13:30:47 Quit jmspeex (
13:30:47 Quit Febs (
13:30:47 Quit tuplanolla (
13:30:47 Quit Rick (
13:30:47 Quit marcosource (
13:30:47 Quit sup (
13:30:47 Quit gtkspert (
13:30:47 Quit sslashes (
13:30:47 Quit [mbm] (
13:30:47 Quit maddler (
13:30:48 Quit agm3nt (
13:30:48 Quit lee-qid (
13:30:48 Quit ender` (
13:30:48 Quit nicktastic (
13:30:48 Quit Xerion (
13:30:48 Quit freqmod_nx (
13:30:48 Quit kclaf (
13:30:48 Quit DraX (
13:30:48 Quit male (
13:30:48 Quit dionoea (
13:30:48 Quit newbyx86 (
13:30:48 Quit snake (
13:30:48 Quit qwm (
13:30:48 Quit Llorean (
13:30:49 Quit maraz (
13:30:49 Quit JETC- (
13:30:49 Quit crwl (
13:30:49 Quit crashd (
13:30:49 Quit lodesi (
13:30:49 Quit annulus_ (
13:30:49 Quit Shaid (
13:30:49 Quit Weiss (
13:30:49 Quit DogBoy (
13:30:49 Quit GodEater (
13:30:49 Quit linuxstb (
13:30:49 Quit Zagor (
13:30:50 Quit miepchen^schlaf (
13:30:50 Quit z35 (
13:30:50 Quit ivan` (
13:30:50 Quit FunkyELF (
13:30:50 Quit kubiix (
13:30:50 Quit SirFunk (
13:30:50 Quit TheCollector (
13:30:50 Quit J (
13:30:50 Quit shodanX (
13:30:50 Quit TTThomas (
13:30:50 Quit maxkelley (
13:30:50 Quit blithe (
13:30:50 Quit RaRe (
13:30:50 Quit guyzmo (
13:30:50 Quit jepler (
13:30:50 Quit fxb__ (
13:30:50 Quit DiDjCodt (
13:30:50 Quit [omni] (
13:30:50 Quit rasher (
13:30:50 Quit LinusN (
13:30:50 Quit preglow (
13:30:50 Quit davina (
13:30:50 Quit advcomp2019 (
13:30:50 Quit CaptainSquid (
13:30:50 Quit krazykit (
13:30:50 Quit Mouser_X (
13:30:50 Quit n17ikh|Lappy (
13:30:50 Quit hannesd (
13:30:50 Quit jurrie__ (
13:30:50 Quit XavierGr (
13:30:50 Quit Tetris-Block (
13:30:50 Quit midkay (
13:30:50 Quit FOAD (
13:30:50 Quit Ave (
13:30:50 Quit Kohlrabi (
13:30:50 Quit BrianHV (
13:30:50 Quit feisar (
13:30:50 Quit Nico_P (
13:30:50 Quit Guerin (
13:30:50 Quit Echelon (
13:30:50 Quit HellDragon (
13:30:50 Quit bagawk (
13:30:50 Quit Isolinear (
13:30:50 Quit daurnimator (
13:30:50 Quit rvvs89 (
13:30:50 Quit lostlogic (
13:34:55 Join Thundercloud [0] (
13:41:48 Join atsea- [0] (i=atsea-@gateway/tor/x-903ee2af211203c7)
13:41:48 Join jurrie [0] (
13:41:48NJoinbnakiddmj [0] (i=0@
13:41:48NJoinBjoernErik [0] (
13:41:48NJoinlastebil [0] (
13:41:48NJoinsbeh [0] (n=sbeh@
13:41:48NJoinHadaka [0] (
13:41:48NJoinparafin|away [0] (
13:41:48NJoinsneakums [0] (
13:41:48NJoinryanakca [0] (n=ryan@ubuntu/member/ryanakca)
13:41:48NJoinDataGhost [0] (
13:41:48NJoinSlasheri [0] (i=miipekk@rockbox/developer/Slasheri)
13:41:48NJoingromit` [0] (
13:41:48NJoinlids [0] (
13:41:48NJointierra [0] (
13:41:48NJoinhomielowe [0] (
13:41:48NJoiniamben [0] (
13:41:48NJoinroxfan [0] (
13:41:48NJoinJon-Kha [0] (
13:41:48NJoinSiku [0] (
13:41:48NJoinPaulJam [0] (
13:41:48NJoinmoos [0] (
13:41:48NJoinjhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
13:41:48 Join XavierGr_ [0] (
13:41:48NJoin[mbm] [0] (i=mbm@openwrt/developer/mbm)
13:41:48NJoingtkspert [0] (
13:41:48NJoinsslashes [0] (i=sslashes@
13:41:48NJoinmarcosource [0] (
13:41:48NJoinmaddler [0] (
13:41:48NJoinsup [0] (
13:41:48 Join Rick [0] (i=rick@unaffiliated/rick)
13:41:48NJointuplanolla [0] (
13:41:48 Join Febs [0] (n=chatzill@rockbox/administrator/Febs)
13:41:48NJoinjmspeex [0] (
13:41:48NJoinRob222241 [0] (
13:41:48NJoinpondlife [0] (n=Steve@rockbox/developer/pondlife)
13:41:48 Join J3TC- [0] (
13:41:48 Join midkay_ [0] (
13:41:48NJoinCaptainSquid [0] (
13:41:48NJoinlinuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
13:41:48NJoinNico_P [0] (n=nicolas@rockbox/developer/NicoP)
13:41:48NJoinagm3nt [0] (
13:41:48NJoinlee-qid [0] (
13:41:48 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
13:41:48NJoindavina [0] (
13:41:48NJoinadvcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
13:41:48NJoinender` [0] (
13:41:48NJoinLinusN [0] (i=linus@rockbox/developer/LinusN)
13:41:48NJoinmiepchen^schlaf [0] (
13:41:48NJoinGuerin [0] (
13:41:48NJoinqwm [0] (
13:41:48NJoinkrazykit [0] (
13:41:48NJoinEchelon [0] (
13:41:48NJoinHellDragon [0] (i=jd@unaffiliated/helldragon)
13:41:48NJoinbagawk [0] (n=lee@unaffiliated/bagawk)
13:41:48NJoinMouser_X [0] (
13:41:48 Join Llorean [0] (n=llorean@rockbox/administrator/Llorean)
13:41:48NJoinn17ikh|Lappy [0] (
13:41:48NJoinnicktastic [0] (n=nick@unaffiliated/nicktastic)
13:41:48NJoinhannesd [0] (
13:41:48NJoinz35 [0] (
13:41:48NJoinFunkyELF [0] (
13:41:48NJoinjurrie__ [0] (
13:41:48NJoinXerion [0] (
13:41:48NJoinIsolinear [0] (
13:41:48 Join ivan` [0] (n=ivan`@unaffiliated/ivan/x-000001)
13:41:48NJoinTetris-Block [0] (
13:41:48NJoinmidkay [0] (n=midkay@rockbox/developer/midkay)
13:41:48NJoindaurnimator [0] (n=daurnima@unaffiliated/daurnimator)
13:41:48NJoinfreqmod_nx [0] (
13:41:48NJoinkubiix [0] (
13:41:48NJoinFOAD [0] (
13:41:48NJoinmaraz [0] (
13:41:48NJoinJETC- [0] (
13:41:48 Join snake [0] (n=Snake@unaffiliated/snake)
13:41:48NJoinrasher [0] (n=rasher@rockbox/developer/rasher)
13:41:48NJoincrwl [0] (
13:41:48NJoinSirFunk [0] (
13:41:48NJoinTheCollector [0] (
13:41:48NJoindionoea [0] (
13:41:48NJoinShaid [0] (
13:41:48NJoinRaRe [0] (
13:41:48NJoinmale [0] (
13:41:48NJoinBrianHV [0] (
13:41:48NJoinDraX [0] (n=alex@xmms2/developer/DraX)
13:41:48NJoinGodEater [0] (n=bryan@rockbox/staff/GodEater)
13:41:48NJoinkclaf [0] (n=kclaf@
13:41:48NJoinWeiss [0] (
13:41:48NJoinannulus_ [0] (
13:41:48NJoinjepler [0] (n=jepler@emc/developer/jepler)
13:41:48NJoinfxb__ [0] (
13:41:48NJoinlodesi [0] (
13:41:48NJoinTTThomas [0] (
13:41:48NJoinnewbyx86 [0] (
13:41:48NJoinguyzmo [0] (
13:41:48NJoin[omni] [0] (
13:41:48NJoincrashd [0] (
13:41:48NJoinAve [0] (
13:41:48NJoinpreglow [0] (n=thomj@rockbox/developer/preglow)
13:41:48NJoinJ [0] (
13:41:48NJoinshodanX [0] (
13:41:48NJoinblithe [0] (
13:41:48NJoinDiDjCodt [0] (
13:41:48NJoinrvvs89 [0] (n=rvvs89@pdpc/supporter/active/rvvs89)
13:41:48NJoinmaxkelley [0] (
13:41:48NJoinKohlrabi [0] (
13:41:48NJoinlostlogic [0] (n=lostlogi@rockbox/developer/lostlogic)
13:41:48NJoinfeisar [0] (
13:41:48NJoinDogBoy [0] (n=john@unaffiliated/dogboy)
13:41:50***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
13:41:50***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
13:41:50***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
13:41:53***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
13:41:56 Quit n17ikh|Lappy (Killed by (Nick collision))
13:41:57 Quit GodEater (Killed by (Nick collision))
13:42:05 Join n17ikh|Lappy [0] (
13:42:05NJoinspiorf [0] (n=spiorf@
13:42:05NJoinmarkun [0] (n=markun@rockbox/developer/markun)
13:42:05NJoinpetur [0] (n=petur@rockbox/developer/petur)
13:42:05 Join GodEater [0] (n=bryan@rockbox/staff/GodEater)
13:42:05NJoinSeed [0] (
13:42:05NJoinBagder [0] (
13:42:05 Join courtc [0] (n=court@unaffiliated/courtc)
13:42:05NJoinSoap_ [0] (n=Soap@rockbox/staff/soap)
13:42:05NJoinjmworx [0] (
13:42:05NJoinATravelingGeek [0] (n=ATG@pdpc/supporter/student/ATravelingGeek)
13:42:05NJointedrock [0] (
13:42:05NJoinLangly [0] (
13:42:08 Quit HellDragon (SendQ exceeded)
13:42:10 Join Nic0_P [0] (
13:43:05jhMikeSfirst connect problems now riding freenode
13:44:19 Quit Echelon (Connection timed out)
13:44:42 Quit krazykit (Read error: 110 (Connection timed out))
13:44:46 Quit bagawk (Read error: 110 (Connection timed out))
13:44:46 Quit jurrie__ (Read error: 110 (Connection timed out))
13:44:53 Quit Isolinear (Connection timed out)
13:45:02 Join GodEater_ [0] (n=bryan@rockbox/staff/GodEater)
13:45:03***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
13:45:19 Quit davina (Connection timed out)
13:45:37 Quit Nico_P (Read error: 110 (Connection timed out))
13:45:40 Join davina [0] (
13:45:45 Quit Mouser_X (Read error: 110 (Connection timed out))
13:46:12 Quit midkay (Read error: 110 (Connection timed out))
13:46:22 Join HellDragon [0] (i=jd@unaffiliated/helldragon)
13:46:27 Quit advcomp2019 (Read error: 110 (Connection timed out))
13:46:34 Join Echelon [0] (
13:49:22 Join bagawk [0] (
13:49:51 Join krazykit [0] (
13:50:17LloreanLinusN: Any idea what mysql version the forum server is running?
13:54:06*JdGordon is king!
13:54:14JdGordonamiconn: the new setting type is ready
13:54:39JdGordonwith only the tiniest hack to add the arbitrary value into the list :)
13:55:14CtcpVersion from freenode-connect!freenode@freenode/bot/connect
13:55:55XavierGr_LinusN: ping
13:55:58jhMikeSsetting type?
13:56:37XavierGr_Slasheri: ping
13:57:01JdGordonit allows you to have a list of values but allows for the user to actually hae a value outside of the set ones
13:57:14JdGordongoign to be used for the backlight setting so any value can be used
13:57:18JdGordoninstead of thoe lists
13:57:51 Join amigan_ [0] (
13:57:51 Quit amigan (Read error: 104 (Connection reset by peer))
13:57:52jhMikeSah. probably a bunch of timeout stuff could use that.
13:58:35LloreanAlright, upgrading the forums looks like it ought to be pretty safe (just tried a dry run on a non-live server) depending on mysql version
13:58:43JdGordonits got another nice touch also... it lets you put text in the cfg instead of a number... e.g "always on" instead of -1
14:00:05 Nick XavierGr_ is now known as XavierGr (
14:00:26GodEaterwhat features would upgrading give us Llorean ?
14:00:35LloreanGodEater: Primarily security fixes.
14:00:37 Nick fxb__ is now known as fxb (
14:00:53GodEaterhave we had many problems with that ?
14:00:58linuxstbIs it just more, or is the problem with the forums being slow in the (European) morning fixed?
14:01:04LloreanGodEater: None.
14:01:14GodEaterjust you
14:01:16LloreanWhich is why we're still on 1.1RC3, and the current version is 1.1.4
14:01:18GodEaterstill slow for me this morning
14:02:01LloreanI haven't felt a huge rush to upgrade until I could do testing to make sure the upgrade wouldn't require me to go and rewrite the theme again or something (which is what happened last time I upgrade, though that was due to how the previous management had chosen to modify the theme ;))
14:02:04GodEaterit wouldn't give us the "do not include this forum in search results" feature we're after ?
14:02:20linuxstbLlorean: Do you know how the "new topics" feature works? I access the forums from different computers, and sometimes get threads showing up on one which didn't on the other...
14:02:35Lloreanlinuxstb: As far as I know "New Topics" is traced by account.
14:02:41LloreanIt works accurately between my phone and PC at least
14:02:56LloreanIf someone modifies their post, though, it'll show back up in New Topics, I believe.
14:02:58GodEaterworks ok across all pcs I use too
14:03:15linuxstbLlorean: It's just that sometimes I read them on my laptop, and have less than a page of new topics, and then I go to desktop PC, and will see 3 or 4 pages worth....
14:03:36LloreanI really don't know.
14:03:45JdGordonso is voice seperate from playback now? can it talk with music paused? or just the codec changed?
14:03:47LloreanWait, the "New Topics" button
14:04:00LloreanThat one works a bit differently than just showing you all posts marked (New)
14:04:01GodEaterif(uid==linuxstb) return random_new_topics() ;
14:04:04linuxstbThis page -
14:04:12LloreanSince it's intent is to show you "New topics since your last visit" as opposed to "All new topics"
14:04:18Llorean;all;start=0 <−− All new Topics
14:04:39LloreanSo I suspect the "Since your last visit" thing might be the problematic part of it
14:05:27linuxstbSo action=unread means "since my last visit" ?
14:05:39JdGordonyeah, nice isnt it :)
14:05:48jhMikeSargh, I created a potential deadlock situation...easy fixeroo
14:05:52JdGordonamiconn: incase you dont get the email.. fs#8186
14:05:55 Part LinusN
14:06:38*JdGordon prepares to swear at GodEater when he realsies the scroll accell patch breaks sansa!
14:07:01GodEaterwhy swear at me - I didn't write most of it
14:07:06GodEaterI just took out the settings :(
14:07:11JdGordonyou pushed for the commit! :p
14:07:17jhMikeShuh? what changed about it because I helped have it run right now sansa
14:07:23Lloreanlinuxstb: As far as I know, yes.
14:07:33*JdGordon was just being silly... relaaax...
14:08:05linuxstbLlorean: And your suggestion of action=unread;all;start=0 will be all unread topics ever?
14:08:09*jhMikeS loads the .45 :p
14:08:37Lloreanlinuxstb: It should be.
14:09:51JdGordonseems to work fine
14:13:55 Join erdemc [0] (n=edrem@
14:14:44 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
14:14:48 Quit linuxstb (Nick collision from services.)
14:14:52 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
14:15:12erdemccan someone help me with album art on rockbox?*
14:15:38linuxstbJust ask.
14:17:02erdemci cant see albumart on the themes i installed. i there something in the settings i dont know about?
14:17:21GodEaterhave you followed the guide on the AlbumArt wiki page?
14:18:05erdemcno can you send the link?
14:22:06 Quit JdGordon ("Konversation terminated!")
14:22:28 Quit linuxstb (Nick collision from services.)
14:22:30 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
14:22:34 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
14:28:26 Join dpc_clyde [0] (
14:28:31 Join Redbreva [0] (
14:28:37dpc_clydehi everyone
14:31:03 Quit erdemc ()
14:31:26dpc_clydeiam new to rockbox, ive read something of a logo swapper, but i can´t found it anywhere therefore I´ve tried a build of evilG with a great boot image and figured out this image is located in rockbox.ipod file. Is there a way without compiling to change this?
14:31:58GodEaterdpc_clyde: the logoswapper you refer to isn't something we support here
14:32:10dpc_clydeok ;) *doh*
14:32:52dpc_clydeany other ideas?
14:33:48GodEaterI've an idea there's a forum post about it somewhere
14:33:51GodEaterhave a search
14:35:36PaulJamdpc_clyde: you could try this (but if you experience problems don't ask here):
14:35:46 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
14:35:48dpc_clydeok thats enough, thx
14:35:56 Quit linuxstb (Nick collision from services.)
14:35:58 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
14:36:03 Nick Nic0_P is now known as Nico_P (
14:40:19linuxstbMr Mmmm ?
14:40:29pondlifeMrH has a friend?
14:40:33amiconnIncorrect commit message (LinusN's commit)
14:40:34LloreanI was thinking something similar.
14:40:37GodEaterpondlife: dammit, I was gonna make that joke!
14:40:51linuxstbamiconn: The wheel one?
14:41:01linuxstbYeah, I noticed that...
14:41:08 Join fluffman [0] (n=fluffman@
14:41:22pondlifeLooks ok here, which page?
14:41:41*preglow summons domonoky
14:42:26 Part dpc_clyde
14:42:41GodEateryeah, looks ok to me too
14:42:54pondlifeOr is it the factual incorrectness?
14:43:14linuxstbYes, it's only for about half the ipods...
14:43:15amiconn"Scroll wheel acceleration for iPod" <== for about half of the ipods only
14:43:28pondlifeAnd he forgot the full stop.
14:43:39GodEaterget out the thumscrews then
14:43:44GodEaterit's just inexcusable
14:43:44linuxstbAnd didn't follow the Rockbox standard for capitalising Ipod.
14:43:56pondlifeI thought that was Iriver
14:44:07fluffmanhi, I've been looking for a patch to let me play m4v's, etc on my ipod 5.5gen video with the latest daily of rockbox, but the only info I can find is from 2006
14:44:14pondlifeFrom Champagne to thumbscrews. We're nice.
14:44:26linuxstbfluffman: There isn't such a patch.
14:44:40fluffmanok just making sure nothing had changed since then
14:44:48pondlifeI'm sure track skipping has got faster since the voice codec change.
14:44:55pondlifeI mean, with voice enabled.
14:44:56fluffmanthank you linuxstb
14:50:17 Part fluffman
14:50:47 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
14:54:06 Part Llorean
14:55:50 Quit linuxstb (Nick collision from services.)
14:55:50 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
14:56:02 Quit Siku ()
15:01:27DogBoywhat's the point of a web page that says "There is no "current" release, get a current build instead."
15:01:33DogBoywhy have a link then?
15:02:55linuxstbIt also links to the previous release - i.e. 2.5
15:03:09GodEaterbecause the release was certified as bug free and with a given set of features for the platform it was made for
15:03:20GodEaterwhich some people prefer to using bleeding edge code
15:03:48DogBoythe link says "releases"
15:04:00DogBoythen you get there and there is a "release" for archos
15:04:07DogBoybad design
15:04:21DogBoyit's misleading
15:04:25linuxstbHow else would we tell people that there are no released versions - which is something people will look for?
15:05:44preglowon that note
15:05:53preglowwhen viewports is in place, i say it's time to start thinking about a release again
15:06:02 Quit Febs (Read error: 110 (Connection timed out))
15:06:02preglowwe should give it a shot at least every other year :>
15:06:35linuxstbWhat about PP502x battery life?
15:06:47preglowlinuxstb: what about not releasing for ipods if that isn't in place
15:06:53 Join cowboydan97 [0] (n=cowboyda@
15:07:00linuxstbWould seem a shame.
15:07:01preglowor pp502x at all, of course
15:07:04preglowit would indeed
15:07:12moosarchos+irivers+iaudios release
15:07:15preglowbut it's not a blocker, that not being in place doesn't prevent us from working on the features that work
15:07:22preglowit doesn't hinder development in the least
15:07:30moosthe h1Ă  in less of course
15:07:33linuxstbI mean maybe we could agree on a set of features/bugs we want implemented/fixed before a release, and then think about a release only after they're done.
15:07:36preglowthose who care about that can work on it in parallel very nicely
15:08:14preglowon the feature side, can you think of anything else we'll want?
15:08:43mooslinuxst: didn't we made it already in the past? (I mean list of bugs to fix...)
15:08:51linuxstbmpegplayer should be stable, with a nice seeking implementation.
15:08:58DogBoyso the difference between releases and current builds is that one exists and the other doesn't?
15:09:03preglowcertainly, and that's happening right now, isn't it?
15:09:20GodEaterDogBoy: no - they both exist
15:09:32preglowDogBoy: we have a release, that's easy to see, isn't it?
15:09:32 Quit jurrie ("Leaving")
15:09:34GodEaterbut Rockbox has only one Release, which only exists for Archos players
15:09:38preglowDogBoy: 2.5, it's right there
15:09:43DogBoyit's stupid
15:09:49preglowit's a release, how is it stupid?
15:09:51linuxstbThere has been many releases - from pre 1.0 up to 2.5
15:09:55DogBoyto have that link
15:09:58preglowyou're just repeating yourself now, you need to give a reason for it being stupid
15:09:59***Saving seen data "./dancer.seen"
15:10:09preglowno, it's a good link, it points to a release
15:10:29preglowreleases need to be easily available
15:10:41DogBoyso you're saying that everybody that comes to the site to dl rockbox needs to see that there is only one release and it's old and for archos
15:10:45moosDogBoy: which device(s) do you use btw?
15:11:07linuxstbpreglow: But no, I don't think we need any new features (not even viewports) for a release - just the existing ones working well.
15:11:13preglowDogBoy: yep, because that is our current situation
15:11:21preglowDogBoy: what would you rather we do? hide the release for no apparent reason?
15:11:29mooslinuxstb: Amen :)
15:11:59preglowlinuxstb: kinda agree, but viewports would be nice
15:12:18moospreglow: nice but not necessary
15:12:39mooseven if we have AA now
15:12:42GodEaterwhy would users keep coming back to test new builds if we give them all the eyecandy in one release ? :)
15:12:57mooshehe :)
15:13:09linuxstbI also think that once we get one release out, we should try and get in the habit of more frequent releases again - it can only be good for quality control.
15:13:17preglowagree completely
15:13:26GodEaterthere's something wrong with our quality ?
15:13:30preglowonly a couple of big features at a time
15:13:33preglowthen stabilizing for release
15:13:34*linuxstb points GodEater to flyspray...
15:13:41 Join joshin [0] (n=josh@unaffiliated/joshin)
15:14:04preglowlinuxstb: at least now we have people working on playback
15:14:13preglowfew areas have no active people on them now
15:14:14moosGodEater:quite relatively lot of bugs remainings ;)
15:14:19preglowso this might actually work out this time
15:14:48GodEaterlinuxstb: last time we spoke about those bugs you said something like "I don't use those features, so they're not bugs" :)
15:15:19moospreglow: the things are surely better than the last release "alert" in vain:)
15:15:34preglowyeah, that was a bit futile
15:15:53preglowbut sure, if we're gonna try releasing for pp, we should have usb as well
15:15:55 Quit male (Read error: 113 (No route to host))
15:15:55preglowand the power bug
15:16:13 Join KoCb0 [0] (i=KoCb0@
15:16:34linuxstbGodEater: Sure, and a release would be easier if we removed those features...
15:17:04moosPP tagets are far from a release if we keep the same criterias as before, I mean h1xx series are ready for release since long *time* now
15:17:37KoCb0hi everyone, I'm searching for rockbox cube logo like on this pic
15:17:48KoCb0but in bigger demensions
15:18:07KoCb0can anyone help
15:18:15DogBoyso it's like a trick then, you want to inform that user of the web site that there are no releases, so you put a link that says releases and then the page says "doh, there are no releases"
15:18:20*petur likes that logo too - should replace the current one....
15:18:23cowboydan97I need help. How do I get rockbox to play an album in the proper track order? It's playing songs alphabetically.
15:18:37preglowDogBoy: _THERE IS A RELEASE THERE_
15:18:45preglowyou need to either start saying something new here, or drop it
15:19:00DogBoyrelease is not the same as releases
15:19:10preglowyou're arguing about incorrect use of plural now?
15:19:13PaulJamcowboydan97: do you use the database or filetree?
15:19:14moospetur: fashion style ;p
15:19:38cowboydan97PaulJam: uh database i guess. Filetree?
15:20:12cowboydan97Yeah, database
15:21:23PaulJamcowboydan97: then make sure that the tracknumber tags are filled correctly (you need to initialize the database again to make it pick up changes in the tags)
15:22:12linuxstbDogBoy: If you follow the link to 2.5, that also shows the older releases.
15:22:24cowboydan97well i mean it plays, and then when it goes to the next in alphabetical line, it may be track 2, but it will show the actual track number in the corner
15:22:53linuxstbDogBoy: Do you have a suggestion for how the website should look, including links to the releases?
15:23:03jhMikeSpetur: yes, much cleaner...but something's missing in terms of visual impact. it's a bit muted. can't place it right now.
15:23:10DogBoyit's no big deal
15:23:28GodEaterthen why are you complaining about it ?
15:23:33DogBoyI guess it's designed from the developer's point of view, which is of course not unusual
15:23:36cowboydan97PaulJam: also, the alphabetical listing only happens when i search by artist. When searching by album, the tracks are sorted by their respective track number.
15:23:45DogBoybut of course many more people than developers go there
15:23:57preglowit's designed from the point of view from someone who wants facts
15:24:02preglowbut i can't be bothered about this anymore
15:24:46cowboydan97Nevermind. It turns out that i selected all tracks instead of the album. And since there was only one album, I didn't notice.
15:24:46moosDogBoy: lots of *normal* users are using daily/bleeding edge build
15:24:48jhMikeSpreglow: who wants facts? I couldn't care less. ;)
15:24:59DogBoyto me it's just dump that you have a link that says "releases" and then you go to the page and it says: There is no "current" release
15:25:01cowboydan97thank you for your help PaulJam
15:25:09DogBoypointing out such things is what faqs are for
15:25:57DogBoyit's like a web site that has a link and you click it and it says "nothing here check back later"
15:26:03amiconnThe 2.5 release is so outdated now that it should be hidden somehow, imho...
15:26:28amiconnThere's even a couple of known bugs
15:26:48amiconn...which are fixed in svn (like the occasional freezes on Ondio)
15:26:52jhMikeSyeah, we want people to keep their bugs up to date
15:26:58DogBoyit's natural for web sites and code to acumulate cruft
15:27:38moosamiconn: what's about rombox? no news from IDCDragon?
15:27:38DogBoybut I'd be willing to bet that the vast majority of people that click that link could care less about what the page says
15:27:49 Join desowin [0] (
15:27:53jhMikeSsort of like the front page?
15:28:18 Nick fxb is now known as fxb__ (
15:28:23DogBoythat's all I was saying, I didn't mean to incur the wrath of the gods
15:30:22 Join DaCapn [0] (
15:31:35DaCapnI can't seem to find information about editing the main menu in the docs (5th gen ipod). The viewers don't appear in the "browse plugins" sub-menu and I would like to change this. Anyone able to help me find the right docs?
15:32:25preglowediting the main menu??
15:32:41preglowthe viewers aren't supposed to appear there
15:32:43 Part cowboydan97
15:32:44preglowthey can't be opened directly
15:32:55preglowthey need to be called with a file argument
15:33:13preglowat least most of them
15:33:25DaCapnso you just open the file directly?
15:33:40preglowthey're meant to be called by the "open with" context menu option
15:34:51amiconnFor most file formats supported by viewers, just 'playing' the file fires up the viewer
15:35:05amiconnBut that's all covered by the fine manual...
15:36:34 Join nicktastique [0] (n=nick@unaffiliated/nicktastic)
15:36:56linuxstbIs the .voice file always loaded if present - even if voice menus are disabled?
15:37:03DaCapnyeah the files I was trying to open weren't showing up in the file browser so I thought I needed to launch the viewer, but I just realized that they're all png images
15:37:40DaCapnI'm probably all sorted out now thanks
15:38:53 Join Genre9mp3 [0] (
15:39:21linuxstbGenre9mp3: Hi. You mentioned a while ago you were interested in adapting iCatcher for AA - are you still planning on doing that?
15:40:24Genre9mp3linuxstb: hallo, yes I am planning to do it but need to find some spare time first
15:40:56Genre9mp3linuxstb: I already saw th left margin commit
15:41:02DaCapndid you check ebay, there's a lot of spare time on there I think
15:42:00*linuxstb normally finds that ebay steals spare time
15:43:39peturspare time also turns out to be very expensive....
15:43:51 Join japc [0] (n=japc@
15:44:59linuxstbGenre9mp3: Did you see my 220x176 version?
15:45:40Genre9mp3linuxstb: nope
15:45:43 Join sqgl_LakeMacq [0] (
15:46:14linuxstbIt needs updating for the %s|x| -> %m|x| change though.
15:46:50*Genre9mp3 checks the dump
15:47:21 Join zicho [0] (
15:48:13Genre9mp3linuxstb: to tell you the truth, I didn't have in mind the %C conditional approach
15:49:13linuxstbDo you like the idea?
15:49:55Genre9mp3I am more like to do a different version than to alter the existing wps in order to display album art if that exists
15:50:31Genre9mp3I think I could fit all things a bit better by using a different font
15:51:07linuxstbI don't have cover-art for all my music, so when there isn't art, I would prefer the space to be used to display text, rather wasting it by being empty, or having a "no cover art" image.
15:51:30amiconnlinuxstb: Afaik rockbox reserves the buffer for a voice file if it finds one
15:51:46Genre9mp3linuxstb: This is what I have in mind:
15:52:30linuxstbGenre9mp3: Hmm, I prefer the larger text, especially in the menus.
15:52:37PaulJamsomeone needs to implement some kind of visualization (like in WMP or winamp) for when no cover is available ;)
15:53:51preglowpeople! apparently we are fools for not having a wince version
15:54:12Genre9mp3linuxstb: I am a bit conserned that the %C approach will be difficult to adapt and maybe the result will be a bit ugly on smaller screens (eg. H10, Nano)
15:54:21*preglow wonders if there are free sdks for wince
15:54:27*linuxstb looks around for someone who owns a wince device
15:54:36linuxstboops, I do... (a gigabeat S...)
15:55:07preglowGenre9mp3: i think that looks rather nice
15:56:08 Quit zicho (Remote closed the connection)
15:56:09linuxstbGenre9mp3: We don't have to use it everywhere - different sized screens could (and I think should) have different approaches.
15:56:43Genre9mp3linuxstb: I know what you mean about the font, but it makes sense to me this: adding album art -> making the font a bit smaller so everything will fit nicely
15:57:09Genre9mp3linuxstb: Well, I have all colour targets in mind
15:57:27XavierGrthe problem is that people that don't use AA will have a preference for a larger font
15:57:35linuxstbI don't want to argue, as WPSs are very much down to personal taste, but I think the font size is fine in my version of iCatcher - there's still enough room, even with the album art.
15:58:06linuxstbSo I guess we can fork icatcher ;)
15:58:14Genre9mp3linuxstb: lol
15:58:39Genre9mp3XavierGr: I'd like to have 2 versions of it. An AA one and the existing one
15:58:46*amiconn pings markun
15:59:07XavierGrah yeah I think that will be more suitable
15:59:35Genre9mp3So then, when you want AA you simply choose the AA version
15:59:37linuxstbThat's what I like about my version of icatcher - it uses AA if present, and looks identical to the previous version if it isn't.
16:00:01Genre9mp3linuxstb: it's identical?
16:00:26XavierGrlinuxstb: but that version is in a Gigabeat F that has plenty of room for that
16:00:27linuxstb_almost_, I think I moved the progress bar down a few pixels...
16:00:38Genre9mp3linuxstb: The progress bar seems a bit lower to me or am I wrong?
16:01:09Genre9mp3linuxstb: my approach uses 100x100 album art btw :P
16:01:24linuxstbI know, but also a smaller font ;)
16:04:33Genre9mp3linuxstb: your version seems to me a bit busy at the right, also I think that the title information will scroll very often and I wouldn't like that
16:05:12Genre9mp3linuxstb: Anyway, give me some time 'till I work with what I have in mind and we will see what suits, best, right?
16:06:04linuxstbGenre9mp3: Sure. We can let the public decide ;)
16:06:40Nico_Pthat reminds me of the default theme contest
16:07:12*amiconn likes the look of icatcher up to some degree, but is annoyed but the loading time
16:07:16 Quit J3TC- (Read error: 110 (Connection timed out))
16:07:16Genre9mp3...or we could bloat the builds by having iCatcher, iCatcherAA and iCatcherHybrid versions :P
16:07:38Nico_Pamiconn: loading time is still an issue?
16:07:40Genre9mp3Nico_P: you started it! :P
16:07:44XavierGrNico_P: Yeah a sad story...
16:07:54Nico_PGenre9mp3: I didn't
16:08:03 Join MethoS- [0] (
16:08:12Genre9mp3Nico_P: your AA commit did
16:08:40*Nico_P started it all when he opened FS #3045...
16:09:03 Quit Zagor ("Client exiting")
16:09:22Jhow about no default theme in the zip and make the user choose one when installing via the GUI installer (sorry, can't remember what it's call atm). If the user chooses to install via zip then they need to see the manual where it says to also pick a theme?
16:09:33amiconnNico_P: Yes. Takes almost as long as the dircache scan on my 2nd Gen, but unlike that, the wps load is blocking...
16:10:14 Join FOAD_ [0] (
16:10:20*amiconn just uses the default wps on most of his targets, and a set of 2 self-made wps'es on a few targets
16:10:23Nico_Pamiconn: I seem to recall you telling linuxstb and me the loading time was OK (at the time we were discussing the BMP tar patch)
16:10:26linuxstbJ: That's more or less what happens now, apart from the installer doesn't select a theme as current, it just installs it.
16:10:46amiconnThose self-made wps'es don't contain any .bmp for bearable boot times
16:10:58Nico_Pamiconn: if you notice slowness you could tell us if the bmp tar patch improves things
16:11:07Jso why all the we must pick a default theme conversation?
16:11:18amiconnNico_P: It seems to depend on target. Maybe fragmentation.
16:11:44Nico_Pyeah... at least a tar file doesn't have that
16:11:54Nico_Pmaybe we should revive that patch
16:12:07linuxstbWasn't there an issue with remote WPSs?
16:12:09XavierGrNico_P: indeed, if it makes a difference
16:12:20amiconnBefore the tokenizer, load times were excessive on archos. That stopped with the tokenizer, but on some targets, I still get wps load times of several seconds e.g. for icatcher
16:12:29Nico_Plinuxstb: it wasn't a big one... just somthing with the build script IIRC
16:12:44Nico_Pamiconn: the tokenizer improved things?
16:12:54amiconnyes, on archos
16:12:59Nico_Pnice :)
16:13:19Nico_Pdo you think it's because of the different bitmap loading startegy?
16:13:23amiconnI remember wps loading time in excess of 15 seconds from before the tokeinzer
16:14:06 Quit atsea- (Read error: 104 (Connection reset by peer))
16:14:21Nico_Plinuxstb: the wps build script just needs to differentiate wps from rwps
16:15:16preglowgroup bmps...
16:15:27linuxstbpreglow: Definitely...
16:15:30 Quit KoCb0 ()
16:15:33Nico_Ppreglow: in only one bitmap?
16:16:03Nico_Plike ?
16:16:06 Join mf0102 [0] (n=michi@
16:16:06linuxstbe.g. all the codec icons could be in a vertical strip
16:16:24XavierGrlinuxstb: isn't there a patch for this too?
16:16:34 Quit DogBoy ("Leaving")
16:16:37preglowNico_P: no, several
16:16:48preglowNico_P: but related bitmaps should be in one bitmap
16:16:54preglowlike a strip of battery indicators, for example
16:16:58Nico_Ppreglow: yeah, I meant as opposed to tar files
16:17:14Nico_Pso I think we're talking FS #7289
16:17:22preglowNico_P: well, they can be combined, i just think grouped bitmaps for related stuff should go in anyway
16:17:28preglowprobably makes life easier for wps makers too
16:18:00amiconnI like the bitmap groups idea better
16:18:09 Quit jhMikeS (Read error: 104 (Connection reset by peer))
16:18:16*Genre9mp3 would like FS #7345 better
16:18:19amiconnNo extra tool like tar needed, and should also reduce loading time considerably
16:18:20Nico_Pyeah I should really commit that one... I've been meaning to for a long time
16:18:50linuxstbYes, I prefer the 7345 approach
16:18:53Genre9mp3Nico_P: I think FS #7345 is what linuxstb says
16:19:00Nico_Phmm yeah
16:19:17preglowdoesn't the tar approach require the user to tar bitmaps in the correct order?
16:19:28Nico_Ppreglow: not anymore
16:20:09preglowtar doesn't have seek table, does it?
16:20:33Nico_Ppreglow: no, but bitmaps are loaded all at once at the end of the wps parsing since the tokenizer arrived
16:20:34amiconntar format is designed for sequential access
16:20:36 Quit sqgl_LakeMacq ("Leaving")
16:20:38linuxstbIt's just [512 byte header][content][512 byte header][content] repeated...
16:20:42preglowNico_P: in random order?
16:21:02linuxstb(with the content padded to 512 bytes as well)
16:21:10amiconnTape ARchiver
16:21:15Nico_Ppreglow: no, IIRC ordered by ID
16:22:17preglowNico_P: so you need to walk around in the tar file looking for the right bitmaps, then?
16:22:56Nico_Ppreglow: yeah, that's it... but the order isn't important. the tar can be read sequentially (the filenames array is then read in random order)
16:23:22preglowif the order isn't important, then you don't really seek around much
16:23:27amiconnSeeking would be bad for loading speed
16:23:40preglowok, i was just going to suggest using zip files, but if you don't need to seek around to look for bitmaps, then it's not important
16:24:07amiconnBut even then, the tar idea isn't that nice, as it requires an extra tool for wps designers. Unlike linux, windows doesn't have tar by default
16:24:13 Join scorche|w [0] (
16:25:20preglowwe could just code our own packer, but that still means an extra tool
16:25:33preglowbut that is the only practical way of bundling everything into one file
16:27:09 Quit FOAD (Read error: 110 (Connection timed out))
16:27:09 Nick FOAD_ is now known as FOAD (
16:27:22XavierGrI would expect that using tar is way more easy than packing all bmps by hand for WPS designers
16:27:32XavierGreven if there isn't such a tool in windows by default
16:27:34preglowdesigners can easily pack related bitmaps
16:27:39preglowbut not unrelated ones
16:27:46 Join stewball`ghost [0] (n=WTFOMGBB@
16:28:06preglowstuff like battery indicators, codec type bitmaps, etc, can go in one bitmap apiece
16:28:10linuxstbWouldn't having fewer files make designing a WPS easier?
16:28:20linuxstb^fewer bmp files
16:28:21preglowit will even make life easier on designers, if you ask me
16:28:28preglowmy point
16:29:04markunamiconn: pong
16:29:32linuxstbLooking at 7345 - is it really necessary to introduce a whole set of new tags? Couldn't we incorporate the behaviour into the existing conditional tag somehow?
16:29:33preglowyou can play sports in #rockbox-community
16:29:40 Join atsea- [0] (i=atsea-@gateway/tor/x-1069e5556f0215fd)
16:29:46preglowlinuxstb: i would really expect that, yes
16:30:05markunpreglow: talking to me? :)
16:30:28amiconnmarkun: This was a font caching ping...
16:30:32Nico_Plinuxstb: then it's FS #7289
16:30:57markunamiconn: I still want to work on it
16:31:04amiconn(aka: any news on that matter?)
16:31:50linuxstbNico_P: But IIUC 7289 requires explicitly using parts of a bitmap. It would be nice if the wps code would automatically split an "bitmap strip" into X equal sized parts, and use those depending on a conditional.
16:31:56amiconnI think the bitmap strip idea might be a good compromise between number of files and hence loading time, work for designers, and tool requirements
16:32:08Nico_Pamiconn: I agree
16:32:20amiconnCodec type should also be a strip then
16:32:36amiconnah, preglow mentioned that one already
16:32:48Nico_Plinuxstb: so something like %xA1?
16:33:09linuxstbNot even that...
16:33:14Nico_Pthen what?
16:36:00linuxstbI was thinking of something like %fc?<%xia> - where %xi means treat the bitmap as an icon strip.
16:36:48linuxstbBut maybe I'm trying to simplify things too much...
16:37:31preglowi think we should definitely make the strip approach mandatory
16:37:52preglowit's already the way most skinning formats work, afaik
16:39:21Nico_Plinuxstb: that would require changes to the parser and is confusing IMO
16:39:30 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
16:40:29 Quit moos ("Rockbox rules the DAP world")
16:44:29preglowgreat, a feature request for a target we don't support
16:44:59preglowand some guy has started hacking our usb stack :|
16:45:13preglowzagor should really start considering commiting his stuff
16:45:25desowinhacking current code in svn or Zagor's code?
16:45:53linuxstbNico_P: OK, but do you understand my intention - i.e. not requiring the wps writer to explicitly specify the individual bitmaps within a strip?
16:46:50 Quit Tetris-Block ()
16:46:52Nico_Plinuxstb: yes, but I think we need to come up with a tag
16:48:24linuxstbNico_P: I'm happy to leave the details to you ;l)
16:48:42linuxstbpreglow: Seems he has made good progress with USB though...
16:48:49Nico_P:p I don't mind help though
16:50:27desowinbut what's sad - it looks like reinwenting the wheel
16:50:41 Join DogBoy [0] (
16:50:56linuxstbdesowin: Indeed. It's unfortunate he didn't see Zagor's patch...
16:51:30Nico_Pthis whole usb thing is a huge duplication of efforts :(
16:53:00desowinI hope he has good nerves when he'll see Zagor's patch
16:53:07linuxstbTrue, I wish austriancoder was here more often (or at all...), same for the new guy...
16:54:07pondlifeSomeone should probably put a noite on #8189..
16:54:19pondlifeA "speak to Zagor" one
16:54:34linuxstbMaybe let Zagor comment?
16:54:49pondlifeThat's what I was hinting at, badly.
16:55:20pondlifeUSB stacks are like buses. You wait for years, then 3 come along at once.
16:55:40linuxstbAnd they're all full (i.e. don't work...)
16:56:04peturit's not a third one, seems he continued where ac left off
16:56:22pondlifeAh. I think that makes it worse...
16:56:26pondlifePoor guy
16:56:52pondlifeIs ac's stuff in SVN?
16:57:02pondlifeIf so, it's entirely understandable.
16:57:27linuxstbYes, we can't blame him, it's just unfortunate...
16:57:54linuxstbBut shows how important communication is, rather than working silently by yourself...
16:57:55Nico_Phow close is ac's stack from actually working?
16:58:04linuxstbSeems closer with that patch...
16:58:10peturnot that far off, I'd say
16:58:33*amiconn prefers Zagor's approach
16:59:19Nico_Plinuxstb: I think maybe we don't use the ML enough and tend to take too much decisions on IRC, leaving out those who aren't there at the time
16:59:39Bagder+gblCBW.dCBWDataTransferLength -= MSD_IN_EP_SIZE;
16:59:47Bagdergreat looking code it is...
16:59:50linuxstbHmm, seems he's taken code from here -
16:59:53pondlifeI think we have too many communication channels...
16:59:55Nico_Pthat's work for the code poslice :)
17:00:22Bagderyes it sounds as if he's gotten the code from there
17:01:22Bagderbut the package has just an .exe inside
17:02:11Nico_Ppondlife: for devs or for everyone?
17:02:19linuxstbIt's an installer, it partly works under Wine, but doesn't install. Can a Windows user try it?
17:03:01 Join webguest69 [0] (i=c31ce03b@gateway/web/cgi-irc/
17:03:03 Join chuckn [0] (
17:03:09chucknhi folks
17:03:49pondlifeNico_P: For both really
17:04:04linuxstbOK, I installed it...
17:04:19chucknis it ok for a newbie to ask a question here?
17:04:26pondlifeSure, fire away
17:05:16linuxstbBagder: The (C) headers say it is provided "for use solely and exclusively on Microchip PICmicro Microcontroller products."....
17:05:27markunlinuxstb: any idea why the mpegplayer settings should be writte to HDD when playback is paused?
17:05:28Bagderand it looks like it is the same code?
17:05:37 Join Administrador [0] (
17:05:49linuxstbmarkun: I think I read jhMikeS say he's fixed that in his tree.
17:05:50chucknthank, i am having trouble playing mp3's on my ipod 60 gig video. i am using version r15686-07119, and all my mp3 lock up after about 10 ecs
17:06:05peturBagder: petur/">
17:06:07linuxstbmarkun: I think it's saving the resume position (which it shouldn't...)
17:06:19Bagderpetur: thanks
17:06:20AdministradorBagder: Hello!
17:06:26markunlinuxstb: I was going to remove it, but I'll wait for mike
17:06:30 Nick Administrador is now known as Casainho (
17:06:51PaulJammarkun: i think it is to preserve the resume position in case of an idle shutdown.
17:07:02 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
17:07:09linuxstbI thought mpegplayer disabled the idle shutdown?
17:07:12scorche|wchuckn: what version of rockbox are you running?
17:07:19Bagderyes, the code looks very similar...
17:07:38PaulJambut not in paused mode
17:07:40Bagdersame names, same indent level, same style
17:08:05 Quit webguest69 ("CGI:IRC (Ping timeout)")
17:08:09 Join webguest69 [0] (i=c31ce03b@gateway/web/cgi-irc/
17:08:17pondlifeThose aren't the sort of variable names you get coincidences with... ;)
17:08:20chucknversiob r15686-07119, but this has been happening with every version i have tried for the past month
17:08:45markunlinuxstb: would be nice if the close function of a running plugin could be called on shutdown
17:09:08linuxstbDoesn't the plugin get a shutdown message?
17:09:24*Bagder runs off
17:09:33markunlinuxstb: no idea
17:10:00***Saving seen data "./dancer.seen"
17:12:00 Quit webguest69 (Client Quit)
17:12:01amiconnShutdown is a mess atm
17:12:13 Quit CaptainSquid ("Miranda IM!")
17:12:20amiconnAnother area I *need* work on :/
17:13:52 Quit Casainho ("ChatZilla 0.9.79 [Firefox]")
17:16:10 Join Casainho [0] (
17:16:57 Join J3TC- [0] (
17:19:02pixelmachuckn: when did you install (or update) your bootloader? And how does your player lock up, do you get an error message or what happens?
17:20:52chuckni reinstalled the bootloader last week. the ipod just goes blank except for the line "now playing" at the top of the screen, but nothing is playing and i have to reset the player to get anything to work again
17:21:08 Quit XavierGr ("One firmware to rule them all!")
17:21:43 Nick Casainho is now known as chaaa (
17:22:03 Nick chaaa is now known as Casainho (
17:22:13 Join XavierGr [0] (
17:22:19 Quit jhulst ("Konversation terminated!")
17:23:38PaulJammarkun: could it be that your latest change to the viewer causes that nothing ges saves if the position in the text hasn't changed? for example if only the encoding setting was changed, but not the position.
17:23:45linuxstbchuckn: You could try resetting your settings - Settings -> Manage Settings -> Reset Settings and then reboot (hold down PLAY/PAUSE to cleanly shutdown, then power it back on again).
17:24:03pixelmais that with an svn build from the rockbox site? Do you or maybe did you have an unsupported build installed at some point in time?
17:27:35markunPaulJam: are you sure the encoding gets saved?
17:28:25 Join Buschel [0] (
17:28:27chucknok, i reset the settings, and rebooted. when i tried playing a podcast mps 3, it played for 9 secs, and then stopped, this time with a message undefined instruction at 2
17:29:04markunPaulJam: I also think if should be saved, as should the other display options
17:29:13PaulJammarkun: no, it doesn't unless i scroll a bit, so the textposition changes.
17:29:40markunPaulJam: are we looking at the same function? (viewer_save_settings)
17:30:13markunPaulJam: wait, I think you are right.. oops
17:30:30PaulJami tried it on the target.
17:30:41markunPaulJam: I will fix it
17:30:46*pixelma doesn't like those commits that change important things (or at least seem so) but come almost out of nowhere and the committer doesn't show up here :\
17:31:15markunpixelma: and which are usually by the same guy..
17:32:25markunwould revoking commit rights be a good way to persuade someone to communicate more?
17:32:37pondlifeMight well be!
17:33:02linuxstbWhat are we talking about?
17:33:07pondlifeI'd say, no commit without being on IRC at the time would be a reasonable rule...
17:33:24pixelmaaustriancoder's commit
17:33:38pixelma(or austriancoder ?)
17:33:51linuxstbpixelma: Plus your Italian friend?
17:33:56GodEaterwhen did he get commit rights ?
17:34:09markunpixelma: did you get in touch with Alessio Lenzi?
17:34:37linuxstbGodEater: I think he's had them for a long time - I can't recall him getting them...
17:34:38pixelmaif I remember well, the IRC "hint" is even in the standard invitation mail, no?
17:35:24pixelmamarkun: hmmno, forgot about it...
17:35:27linuxstbIt's just an invitation to join #rockbox, rather than a hint that it's expected.
17:35:27*preglow tries to figure out how wavtrim.c works
17:36:04markunlinuxstb: I think communicating big changes should become more than a hint
17:36:05pondlifeMost committers are willing and able to pop in here when they're active, no?
17:36:26linuxstbYes, most devs hang out here whilst developing...
17:36:32preglowsimple threshold check
17:36:36markunlinuxstb: he could also post a message on the list if he doesn't like irc
17:37:02preglowi don't care how, as long as he does
17:37:05preglowi'd expect rather irc or ml
17:37:12preglownone of them aren't acceptable
17:37:18preglowas in, using none of them :>
17:37:21 Join Arathis [0] (
17:37:29 Join Vapre [0] (i=42cf43b5@gateway/web/cgi-irc/
17:37:48linuxstbAnd he didn't fix the mas header file as promised...
17:37:56linuxstb(unless I missed that....)
17:38:47 Nick fxb__ is now known as fxb (
17:39:26VapreHi everyone, just wanted to give a quick 'thank you' to the developers. This is a great program, I put it on a Sansa e250 and am amazed by the added functionality. I plan on adding it to a 2G ipod Mini and a 5G video.
17:40:16Casainhonice to know Vapre - I am also happy for RockBox on my sansa
17:40:33 Quit XavierGr ("One firmware to rule them all!")
17:40:51 Join endor [0] (
17:41:26PaulJammarkun: concerning the bookmarking in the viewer, what do you think about only creating a bookmark on request? so don't create a bookmark when pressing STOP, but extend the menu by an entry like "Bookmark & Quit". just came to my mind. on most files you propably don't need a bookmark even if the textposition has changed.
17:41:27preglowwhy, all this thanking makes me feel i've deserved a beer
17:42:01markunPaulJam: good idea
17:42:08pondlifeNico_P: You didn't commit the second part of FS #8104?
17:42:15Nico_Ppondlife: just did
17:42:18pondlifeAh, thanks
17:42:35 Join XavierGr [0] (
17:42:36pondlifeI don't know if it worked, but the simplification is welcome
17:43:04Nico_Pit looks perfectly fine
17:43:08linuxstbNico_P: #ifdef SIMULATOR seems a little hackish...
17:43:10pondlifeDo we still suffer from having -1 tracks buffered?
17:43:19Nico_Ppondlife: no
17:43:29Nico_Plinuxstb: why?
17:43:30endori have a question concerning mp3 encoding. after the polyphase filter divided the signal into different subbands.. it says that time samples get converted to frequency samples through mdct. i dont really understand the difference between time samples and frequency samples. is there someone who can shortly tell me what it is?
17:43:54linuxstbNico_P: apps/ code shouldn't behave differently depending on target (or sim/real device).
17:43:55Buschelpreglow: then let me additionally thank _you_ for comitting my (small) patch! :)
17:44:06linuxstbNico_P: (Where possible of course...)
17:46:48Nico_Plinuxstb: that would require making sim_read sleep... not sure it would be welcome
17:47:04 Join roolku [0] (
17:47:26markunPaulJam: I will fix the settings bug later, have to get food
17:47:29preglowBuschel: haha, no problem
17:48:28roolkumarkun/ PaulJam: I would really like to be able to flipp back and forth between different files without using the menu - at least have a different button for quitting with resume saving
17:49:05linuxstbNico_P: Why was it a problem anyway?
17:49:06roolkualso "not saving", as it is done now leaves the resume file in an undefinded state
17:49:21 Part agm3nt
17:49:44Nico_Plinuxstb: it's not much of a problem, but the sim used to behave differently (intantaneous buffering)
17:50:44markunroolku: can you explain?
17:51:18lostlogicNico_P: not on the latest build, but I haven't seen any changes that would likely impact it since. I had a weird bug this morning where I hit skip forward from the last track on buffer and it skipped to the first track on buffer (back 4 songs-ish), was able to skip forward to the end of the buffer and on to the first unbuffered track after that.
17:51:24roolkumarkun: at the start the bookmark files is reshuffled so the current files slot moves to the front of the file
17:51:56markunroolku: I have to look into it a big more then :(
17:52:04roolkumarkun: if no bookmark is saved this slot contains undefinded content (depending where the bookmark was for the current file)
17:52:05Nico_Plostlogic: very strange indeed
17:52:24markunshould I revert my commit for the time being or is it enough that I fix it later?
17:53:09pondlifeNico_P: Better to make sim_read sleep like the real one, if possible.
17:53:11 Quit endor ("leaving")
17:53:12roolkuor fix it quickly enough and hope that nobody notices ;)
17:53:32markunroolku: will try that last one then :)
17:53:36Nico_Plostlogic: btw, why in queue_wait_w_tmo does the timeout value depend on "filling"?
17:54:21lostlogicNico_P: if the buffer is filling, we don't want to sit up there waiting for a command, if there is not one immediately available, we want to fall through ASAP to continue filling the buffer.
17:55:13Nico_Plostlogic: then aren't the values reversed?
17:55:16roolkumarkun: please don't forget (I rely on bookmarking quite heavily and don't want a corrupted bookmark file)
17:55:31Nico_Plostlogic: oh no sorry I'm dumb
17:55:44Nico_PI was reading 5 as "5 secs"
17:55:50lostlogichehe :)
17:55:51 Join Frazz [0] (
17:56:01lostlogicI was about to be like "5 ticks << 1/2s, I hope..."
17:56:15VapreWhile we're talking bookmarks, are bookmarks supported on the MicroSD cards on Sansas?
17:56:20 Join OlivierBorowski [0] (
17:56:22 Join jpt9 [0] (n=jpt9@
17:56:26lostlogicNico_P: a 0 timeout might make more sense, I'm not positive.
17:56:51Nico_Pyeah, maybe
17:57:11lostlogicI think that the audio thread in pre-mob world used a zero when filling.
17:58:03pondlifeDoes anyone else think we should delete all the WPSGallery entries that don't work...
17:58:15jpt9hey. is there any way you guys could add a separate volume control for the voice? i'm not sure how loud the voices are recorded, but a lot of the time, i can't hear them over the music.
17:58:28Nico_Plostlogic: yes, it did
17:58:36linuxstbpondlife: I would prefer that we just delete the WPS Gallery and move to
17:58:37pondlifejpt9: It's kind-of-planned
17:58:51pondlifeMe too, but that also needs lots of deletes
17:59:06jpt9define "kind-of"
17:59:07pondlifeI'd like to start with a blank set and only add good WPSes to it
17:59:11linuxstbIsn't the plan for scorche's new site to start blank?
17:59:25pondlifejpt9: We ere talking about a proper mixer a day or 2 ago.
17:59:30pondlifewere, even
17:59:52pondlifeThat would allow the voice and music volume to be separate
18:00:02pondlifePlus the beeps and game sound..
18:00:23pondlifeBut it's just talk at the mo
18:00:45linuxstbSo there would be independent volume controls?
18:01:01lostlogicdon't we already have a beep level control?
18:01:06lostlogicor is it beep length?
18:01:34jpt9or you could have it split the audio in stereo −− have the music move to the left when it's talking, and play the menus to the right.
18:02:06jpt9it's level. but there are only 3 settings (plus off).
18:02:41 Quit linuxstb ("Leaving")
18:04:05jpt9i'm curious −− what's different in the new (v2) e200s that prevents rockbox from running on them?
18:05:43 Join linuxstb [0] (n=chatzill@rockbox/developer/linuxstb)
18:08:05 Quit idnar (Nick collision from services.)
18:08:08 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
18:08:49preglowwavtrim really badly mangles some clips
18:08:59pixelmajpt9: someone in the forums who had a v2 c200 opened it and said that there is quite different hardware in it. But it's quite vague so far and e.g. I haven't heard of someone owning a v2 e200 till today
18:09:46 Part Casainho
18:10:44jpt9that kinda sucks. i'm not sure if they've completely replaced the old models in all the stores, but i have an e250R from woot, and I might want the 8GB model some day.
18:10:53pregloweh? does our in-svn usb work with the final patch? ac certainly gives that impression
18:11:33krazykitjpt9, i recall seeing reports that having a 3.0 firmware prevents rockbox from installing, but i don't know that there's been any testing done.
18:11:47Nico_Plostlogic: I think I have the fix for FS #8040
18:11:59scorche|wpondlife: linuxstb: i plan on putting big red text marking not to use it...keeping it around for 2-3 weeks or so for people to move what they can to the new site, then deleting
18:12:30linuxstbscorche|w: You're talking about the WPS Gallery?
18:12:37linuxstb(or the existing themes site?)
18:12:48pondlifeOK, so we'll move over WPSes that meet certain criteria, right?
18:12:59pondlifesoap-seal-of-approval seems reasonable.
18:13:30 Quit XavierGr ()
18:13:59 Join XavierGr [0] (
18:14:07linuxstbpreglow: As he appears to just have copied and pasted non-GPL'd code, it's not committable anyway...
18:14:55jpt9can you tell all the theme authors that even though rockbox lets you put tons of info on the WPS, that doesn't necessarily mean you *have* to?
18:14:58Nico_Ppreglow: what final patch?
18:14:59 Quit Vapre ("CGI:IRC (EOF)")
18:15:00preglowlinuxstb: well, no, but it wouldn't exactly be much work to reimplement
18:15:14preglowNico_P: the one that was submitted a couple of hours ago
18:15:15pondlifescorche|w: Thanks
18:15:27Nico_Pah yeah, didn't see the comments there
18:15:31krazykitjpt9, you can always rip some of it out easy enough
18:16:05preglowwavtrim.c really badly mangles some files
18:16:11preglowespeak numerals, for example
18:16:22jpt9yeah... then again, i'm not sure how the rockbox user base and th
18:16:51pondlifescorche|w: Why bother with themes for unsupported builds at all?
18:16:52jpt9those with good graphic design sense overlap...
18:17:33 Join Domonoky [0] (
18:18:24 Part pondlife ("Gone")
18:19:42scorche|wpondlife: because i am aiming to completely replace all theme sites to have just one site for everything...this way, it encourages more to come to this site instead of others (and that i am completely replacing which held a lot of unsupported themes, and it is nicer to keep one place for people to come to)
18:19:56scorche|wit is also a matter of control...having one site for everything :)
18:20:07 Quit chuckn ("Leaving")
18:20:09jpt9damn it... my copy of rockbox is out of date already :-) you guys kick ass!
18:21:03linuxstbpondlife: It will also make it possible for us to adapt themes that need patches to work with the offiicial builds (either immediately or in the future), as the licensing situation will be clear.
18:23:37PaulJamscorche|w: what is this themename.txt you mention near the end of the guidlines page?
18:24:33scorche|wPaulJam something that needs to be taken out
18:24:42jpt9oh yeah... is there anything within the sansa that would interfere with the radio around 96.1MHz. there's definitely a signal, but there's no audio - it's completely quiet. i think i might have heard some electronic interference a few times; not sure.
18:25:44scorche|wPaulJam: i made the script auto-generate the txt files, and i seem to have missed that one while taking out the guidelines about it
18:26:10krazykitjpt9, do you get the same problem in the OF?
18:26:36jpt9not sure - lemme check.
18:27:14Buschelbtw, I am honored to see the scrollwheel patch submitted :)
18:28:12 Nick fxb is now known as fxb__ (
18:29:11 Quit kubiix (Read error: 110 (Connection timed out))
18:29:28jpt9yes. and there's definitely electronic noises on 95.9...
18:29:46 Join austriancoder [0] (n=austrian@rockbox/developer/austriancoder)
18:29:50krazykitoh, i like that the builds are now named
18:30:39 Quit MethoS- (Remote closed the connection)
18:30:50 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@
18:31:11jpt9for one thing, it makes the noise whenever the screen and wheel lights turn on.
18:34:00 Quit atsea- (Read error: 104 (Connection reset by peer))
18:35:00markunroolku: better not update yet, it looks like I broke quite some things in the viewer :(
18:35:42lostlogicNico_P: glad to hear you're fixing 8040 −− not that I use it, but I know a lot of users are looking forward to its fix.
18:35:46roolkumarkun: yep, I had that suspicion when I saw the diff
18:35:48 Quit jpt9 ("Leaving")
18:35:48 Quit petur ("work->home")
18:36:06Nico_Plostlogic: yeah, was quite a simple fix really, I should've done it earlier
18:36:29Nico_Pnow I think I don't have much left to do but finish the callback stuff to fix FS #8092 :p
18:36:59roolkumarkun: I think at a minimum you need to save the current position in the first slot (which could be overridden when you quit)
18:37:03Nico_Phmm maybe I could do cue with MoB
18:37:23markunroolku: well, didn't remove that, did I?
18:38:19markunand do you always want to move the current file to the top, even if you didn't change the position within the file?
18:38:38roolkumarkun: not sure what you mean? I thought your intention was not to write the resume position at the end under certain circumstances
18:38:58 Join male [0] (
18:39:19roolkumarkun: how can you know at this point that the position will not be changed?
18:39:45roolkumarkun: the shuffling has to be done at the start, as it uses the text buffer
18:40:31rasherSlasheri: it seems that my sansa is blocking all file-operations during database update (possibly only if it's an automatic update). Is this expected?
18:40:44markunroolku: then it's not possible, no
18:41:28lostlogicNico_P: don't we also have the gapless bug to fix? The callback stuff would probably also make that easy to do.
18:41:46Nico_Plostlogic: how would it make it easy?
18:42:04amiconnpreglow: wavtrim checks for a threshold, then backs aout a bit iirc
18:42:11roolkumarkun: also there is the special case that there was no bookmark for this file that complicates matters
18:42:25amiconnThe threshold is settable, and the old vbscript had this threshold set per engine just uses a fixed threshold
18:43:52roolkumarkun: oops. I never realised you had committed a fix - let's have a read
18:43:57rasherThis is mostly because I coulndn't see any difference from fiddling with the threshold, so I just set a more or less random value
18:45:21amiconnIt depends on the noise floor of the engine. If you set the threshold lower than that, you get excessive gaps between clips
18:45:39markunroolku: that was not a fix, but another 'feature' :)
18:45:58roolkumarkun: yes, I noticed :)
18:46:52lostlogicNico_P: if there's a I'm done buffering this handle callback it will make it easy to strip the tags
18:47:10rashermaybe wavtrim was broken or something, because I tried a bunch of different values, high and low, and got no difference at all
18:47:18rasheror I was using it wrong, all possible
18:47:24rasherShould be easy to make it an option
18:49:08PaulJamah, nice. a fix for the gather runtime feature. thank you Nico_P.
18:49:20Nico_Plostlogic: hmm yeah... but we'd still need something like "buf_trim_handle", wouldn't we?
18:49:33Nico_PPaulJam: thanks :) Have you tried it?
18:50:36PaulJami have just updated my build. i'll let you know if i notice anything unusual.
18:51:19 Join DerPapst [0] (
18:55:57 Join nanok [0] (n=nanok@
18:58:32 Join [1]seb [0] (
18:59:07[1]sebhi, is there a guitar tuner plugin?
19:01:02markun[1]seb: I don't think there is
19:01:35[1]sebhmm, ok, anyone ever tried to code one?
19:02:44markun[1]seb: no idea, maybe google can help you out
19:02:52 Join Siku [0] (
19:02:55[1]sebok, thanks, c u
19:03:01 Quit kclaf (Remote closed the connection)
19:03:10nanok[1]seb: that's actually not such a bad idea, and should be relatively easy to do
19:03:23maleWhat kind of tuner?
19:03:33maleJust a reference tone?
19:04:41nanokmale: i think he is thinking of usng the mic and having a visual cue on the display to say when the tuning is correct
19:04:53PaulJamdoes someone know how i can see if a m4a file is cbr or vbr? rockbox shows cbr in the WPS, but the fact that every file of this album has a different bitrate suggests that they are vbr.
19:04:59[1]sebi think of a display showing you to tune up or down
19:05:00maleOn which target? That's pretty CPU intensive.
19:05:05nanokat least this would be, in my view, a "cool" plugin
19:05:18[1]sebyeah, exactly
19:05:21 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
19:05:29maleHow about a strobe tuner so you put your DAP under your strings and strum ;-)
19:05:35lostlogicNico_P: yeah, but that would be relatively trivial I think.
19:05:52lostlogicit would only get supah weird when the handle is already playing and needs trimmed
19:06:03 Join linuxstb_ [0] (
19:06:14[1]sebdap? = digital player
19:06:18 Quit linuxstb (Nick collision from services.)
19:06:19lostlogicbut even there it might be no problem, not sure how codecs react to their mp3_info having a greater file size than the buffer returns.
19:06:25 Nick linuxstb_ is now known as linuxstb (
19:06:46markunnanok: how are you going to determine the pitch?
19:06:46nanokmale: you think it would be intensive?
19:07:17nanokmale: oh, don;t ask me, i have to dig a lot about that before i can talk details :)
19:07:19 Quit J3TC- (Read error: 110 (Connection timed out))
19:07:26maleAt least if you don't want huge latencies.
19:07:31nanokbut i am sure there are a bunch of oss projects there for inspiration
19:07:36nanokfor pc;s ofcourse
19:07:46 Join H2Glitch2007 [0] (
19:07:55markunnanok: "19:03 < nanok> [...] should be relatively easy to do"
19:07:58maleThere are many GPL tuners for linux you could use code from.
19:08:14nanokmarkun: maybe i was a bit hasty :)
19:08:21H2Glitch2007i have a quick question does rockbox support avi files
19:08:28markunH2Glitch2007: it doesn't
19:08:28krazykitH2Glitch2007, no
19:08:35nanokmarkun: i was just thinking about it now and realised there are some things i don;t hav clear about this
19:09:03 Part H2Glitch2007
19:09:17markunnanok: if the wavform from a guitar string was a perfect sine wave it would not have been that difficult
19:09:20nanokmarkun: it felt simple at first because i have a very simple hardware device for my guitar, one of those cheap electronic tuners
19:09:44maleThose cheap tuners are NOT simple.
19:10:00nanokmarkun: maybe it can be interpolated or maybe modulates to something close to that..
19:10:04***Saving seen data "./dancer.seen"
19:10:20male(and they're not electronic, they're digital)
19:10:22nanokmale: hmm, thinking of it now, yes, you are probably right :)
19:10:54nanokbut it does sound like cool app to have
19:10:56[1]sebhmm. the guitar tuner is not too easy i see
19:11:01maleA tuner that generates a reference is far simpler, such as a strobe tuner or a pitch pipe.
19:11:22nanokmaybe i can dig more about it if i find some time (i would like to understand more about it). ofcourse i won;t be able to code it, but still..
19:12:19nanokmale: you mean to have the "listener" match it by hearing to what the guitar produces?
19:12:21 Join bertrik [0] (
19:12:37nanokgtg, be back later
19:13:29*desowin don't quite see why guitar tuner on portable player would be useful....but maybe it's because I have nice digitech RP200
19:13:54maleSeems silly to me too, considering how cheap digital tuners are these days.
19:15:00maleI mean, on a PC you can get higher resolution or microtuning or a histogram, but on a DAP a tuner would be far more limited.
19:15:17[1]sebwhere's the fun in engineering these days :P?
19:15:28maleUnless your player has a 2Ghz CPU.
19:16:29maleIf you want to engineer, build a distortion pedal or something else that really IS as simple as it looks ;-)
19:17:09preglowamiconn: well, i've tried, and it cuts too much
19:17:32 Quit spiorf (Remote closed the connection)
19:18:12 Join spiorf [0] (n=spiorf@
19:20:15preglowamiconn: i even tried using 50 as a threshold, it has to have some other bug
19:21:17 Part [1]seb
19:26:20 Quit Buschel ()
19:27:46 Quit stewball`ghost ()
19:28:41 Join linuxstb_ [0] (
19:29:37 Quit linuxstb (Nick collision from services.)
19:29:39 Nick linuxstb_ is now known as linuxstb (
19:32:29 Quit rasher ("leaving")
19:38:23preglowamiconn: try making espeak generating a clip saying "two", then wavtrim that and listen to the result
19:38:28preglowamiconn: "three" also works
19:38:29*Nico_P is going to have access to the archos 604 and 104 with probably hefty rebates tomorrow
19:39:48pixelmaNico_P: do you want to start a new port?
19:40:06linuxstbIs there a "v2" of the e200 series as well as the c200?
19:40:15Nico_Ppixelma: I have really small knowledge of everything lowlevel
19:40:37advcomp2019linuxstb, i heard there is
19:40:43pixelmalinuxstb: not sure, but there were rumours
19:41:55 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
19:42:06Genre9mp3linuxstb: from the BuyersGuide page: "The Sansa c200 series seems to have a new "v2" model available. These can be distinguished from a v1 easily by looking for the line "Supports Audible audio file formats" on the product's description on the packaging. Rockbox does not work on these v2 models; they appear to be using a new chipset."
19:42:35advcomp2019but here more info about the e200v2s..
19:42:58linuxstbDon't make me go to ABI...
19:44:28advcomp2019sorry but someone posted info about the e200v2s tho
19:44:38krazykitGenre9mp3, how did you license iCatcher? CC-by-SA?
19:44:52linuxstbadvcomp2019: ;) OK, so it seems they're not released yet though?
19:45:23advcomp2019yea.. that is what is sounds like
19:45:25 Nick billenium is now known as BIllenium (
19:45:31 Join petur [0] (n=petur@rockbox/developer/petur)
19:45:32 Nick BIllenium is now known as Billenium (
19:45:41pixelmalinuxstb: the sandisk firmware updater only lists different firmwares for c200 v1 and v2...
19:46:11pixelmanot sure if that's up to date though
19:46:58linuxstbpixelma: Thanks. I was reading this thread, and was wondering if the reply was accurate -
19:48:31pixelmayeah, I saw that thread and my impression was that there is something else going wrong (or he's doing wrong) but I don't think I know enough to comment on this
19:50:01Genre9mp3krazykit: I'm not good at legal matters, since it's bundled with Rockbox is licensed under that, isn't it?
19:50:58krazykitGenre9mp3, i was under the impression the author decided the license.
19:51:11 Join rasher [0] (n=rasher@rockbox/developer/rasher)
19:51:23linuxstbProbably not, it's more likely to be licensed under the GPL, but it's not clear... And yes, as krazykit said, the author decides the license.
19:51:55Genre9mp3GPL sounds good to me
19:51:56linuxstbGenre9mp3: Did you create iCatcher originally?
19:52:59Genre9mp3linuxstb: the graphics contained in it you mean?
19:53:21linuxstbI mean the whole wps - I don't know where it came from...
19:53:49krazykiti'm concerned with it because i wanted to release a derivative of it and didn't want to mislicense it
19:54:42Genre9mp3linuxstb: well, yes, It's not based on any other wps or program
19:55:10linuxstbI think we're moving to CC-SA for all themes/wpses - that's what's used for, so it might be helpful to change the included WPSes to use the same license, if all the authors agree.
19:56:03 Quit japc (Read error: 110 (Connection timed out))
19:56:59krazykitreally, the WPSs in svn SHOULD have their licenses stated, and i don't think any of them do.
19:57:56Genre9mp3krazykit: on a side notice, feel free to release whatever derivative of it you want
19:58:33krazykitthanks :-)
20:00:37 Join Madden [0] (
20:01:20MaddenHi, Wondering if there is a release for ipod nano video 4gig yet? or if its in production...?
20:03:54PaulJamMadden: no, and as far as i know noone is currently working on a port.
20:04:09Madden:( thanks
20:05:20Maddenknow any other rockbox style stuff that might let ya play games and things?
20:06:45linuxstbNo, if there was others, then that work would be being used to help port Rockbox to it.
20:08:08PaulJamthe g3 nano is encrypted and the hardware is undocumented. so it might never run 3rd party software.
20:08:12 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
20:08:19Maddenshite :(
20:08:35Maddenwell its not for me so its not as important hheh
20:08:56Maddenthanks ofr the info, gotta run :) hasta luago
20:09:47 Quit Madden ("Leaving")
20:10:52markunroolku: ping
20:20:09 Join zicho [0] (
20:26:38 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
20:28:36 Quit advcomp2019 ("Leaving")
20:28:56 Join Llorean [0] (
20:30:56 Part XavierGr
20:31:28 Join Alonea [0] (n=chatzill@
20:31:39 Join merbanan [0] (n=banan@
20:31:44 Join XavierGr [0] (
20:37:39 Join FOAD_ [0] (
20:38:41 Join PaulPosition [0] (
20:38:58 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
20:39:19 Join miepchen^schlaf [0] (
20:39:51 Join Arathis2 [0] (
20:41:23PaulJamNico_P: i player now one albun completely and then looked at the recently playerd entry in the database: tracks 1,2,3,5,6,8,9 and 10 got added but not tracks 4,7 and 11. so it looks as if there is still something wrong with the gather runtime feature. do you think i should rebuild the database and try again?
20:41:50PaulPositionHmm.. On H10 (and surely many other devices), recording gain is only 0db or +20db.. Is that because of the internal mic hardware or because of the sound chip? (ie, would it be the same anyway with line-in?)
20:43:44pixelmaon my M5 it's similar - built-in mic gain is also only 0dB or +20dB, there are more steps for line-in though (just got the possibility to test)
20:45:11 Join chuck [0] (
20:46:08 Join J3TC- [0] (
20:47:00PaulPositionHmm.. Then I might try to build myself a cable and see... :)
20:47:33 Join Lear [0] (i=chatzill@rockbox/developer/lear)
20:49:02 Quit ompaul (Read error: 113 (No route to host))
20:53:38 Quit DaCapn (Read error: 110 (Connection timed out))
20:54:23 Quit FOAD (Read error: 110 (Connection timed out))
20:54:24 Nick FOAD_ is now known as FOAD (
20:54:26 Join lee-qid_ [0] (
20:57:44 Quit Arathis (Read error: 110 (Connection timed out))
20:59:21 Join hunz [0] (
20:59:46 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
21:04:09 Quit jhulst (No route to host)
21:06:40chuckhi again
21:10:07***Saving seen data "./dancer.seen"
21:10:28Nico_PPaulJam: no need to rebuild the database
21:10:47 Quit lee-qid (Read error: 110 (Connection timed out))
21:11:06 Join hannesd_ [0] (
21:11:19 Quit GodEater (Read error: 110 (Connection timed out))
21:25:35 Part hunz
21:27:46 Quit hannesd (Read error: 110 (Connection timed out))
21:27:47 Nick hannesd_ is now known as hannesd (
21:30:58 Quit zicho (Remote closed the connection)
21:34:17amiconnNico_P: Quite a lot of Code for such a fix...
21:34:31Nico_Pamiconn: the gradiend fix?
21:36:41Nico_Pyeah I didn't really expect the binsize would increase that much... it's only on targets for which we don't care too much though
21:37:31 Quit amiconn (Nick collision from services.)
21:37:40 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
21:37:47pixelmathat's a weird attitude though (you should always care) ;)
21:37:59Nico_Pamiconn: I'm not sure how it could be simplified though
21:38:17Nico_Ppixelma: I'm not saying I don't care... but it's not like those targets are the sensitive ones
21:39:06amiconnI'm not sure either. As I don't use the gradient, I don't even know what the problem was
21:39:29amiconnBut the code looks quite cumbersome to me
21:39:39Nico_Pscrolling would make the gradient go back to being one line only
21:39:42*linuxstb now just sees even more complex code that needs to be ported to viewports...
21:42:30amiconnAh, yes
21:42:38*amiconn just tried it on X5
21:42:49 Quit chuck ("Leaving")
21:43:15amiconnIt's quite difficult to find a place where there's more than one line selected in a list *and* the text needs to scroll...
21:43:28Nico_Plinuxstb: I don't think it needs to be ported to viewports... it should just work
21:43:53Nico_Pamiconn: yeah, that's why I didn't bother too much about it... but idak posted a patch
21:43:57 Join MethoS- [0] (
21:44:22amiconnImo the whole gradient thing should be implemented differently
21:44:37Nico_Phow should it be done?
21:45:40 Quit J3TC- (".•«UPP»•.")
21:47:04 Quit Siku ()
21:47:12amiconnThe current method even needs support from the app layer :/
21:48:43 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
21:52:09 Join J3TC- [0] (
21:54:14 Join Tetris-Block [0] (
21:54:23 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
21:54:26Nico_Pamiconn: that's because the driver layer isn't aware of the number of selected lines
21:55:06amiconnYeah, but this special-casing *will* break with viewports
21:55:22amiconn...because the driver won't know about consectutive scrolling lines
21:55:52Nico_Phmm... then maybe we need something intermediate to draw the line selector
21:57:06 Quit advcomp2019 (Nick collision from services.)
21:57:14 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
21:57:15amiconnImo we don't. The app layer should just tell the driver the colours to use for a gradient
21:58:09amiconnI think that the gradient can be handled without line drawing, by having a special background setting that reads colours from an array
21:58:16bertrikI could live without a gradient line selector
21:58:47amiconnYeah, me too, but it seems to be a popular feature :/
22:00:21Nico_Pamiconn: I'm not following you... what would be in the array?
22:00:28Nico_Pthe color steps?
22:00:41 Quit Genre9mp3 (Nick collision from services.)
22:00:42 Join Genre9mp4 [0] (n=yngwiejo@rockbox/contributor/Genre9mp3)
22:00:51 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
22:00:59amiconnThe colours for each pixel line of the gradient
22:01:01 Nick Genre9mp4 is now known as Genre9mp3 (n=yngwiejo@rockbox/contributor/Genre9mp3)
22:01:07 Quit advcomp2019 (Nick collision from services.)
22:01:09 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
22:01:44 Quit desowin ("use linux")
22:02:29 Join xaj [0] (
22:02:50*markun goes to fix red..
22:03:09Nico_Pamiconn: but what about variable selection size?
22:03:44 Part xaj
22:03:51pixelmawould the bitmap approach that was discussed sometime ago help here?
22:03:56amiconnWell, the app layer knows how many lines are selected, and hence can calculate the array(s) accordingly
22:04:11 Join deflated [0] (
22:04:40deflatedDoes Rockbox support the iPod Nano 2G?
22:04:48 Join ForgottenMemorie [0] (
22:05:02markundeflated: no
22:05:06rasherdeflated: No. All supported players are listed on the frontpage.
22:05:18markun(including what's not supported)
22:05:29deflatedOk, thank you. :)
22:05:31Nico_Pamiconn: yeah I think I see... I won't have time to look at it this evening but I might try later
22:05:31rasherIn fact, the 2g nano is specifically listed as unsupported
22:05:47ForgottenMemorie2G Nano is unsupported for..?
22:05:49 Join ZEA [0] (
22:05:55ZEAhi. does rockbox submit to
22:06:04 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
22:06:07markundeflated: bye
22:06:29pixelma2G - if that means second generation
22:06:34 Quit jhMikeS (Client Quit)
22:06:39 Part deflated
22:06:51 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
22:07:19markunpixelma: hm, too late..
22:07:21peturZEA: yes, you can log to a file and have it submitted using qtscrobbler
22:07:23ForgottenMemorieumm.. I ned help, my ipod is blank/dead it wont' turn on at all watever I do, plug it into charger or into the computer the computer can't detect it and it stays blank, I tried reseted it but it doesn't work. Any other ideas?
22:07:43ZEApetur: awesome :) thanks
22:08:20ZEApetur: that gives me a reason to install rockbox on my iaudio x5. hope to find out more in the process :)
22:08:44linuxstbForgottenMemorie: Obvious question, but exactly what are you doing to try and reset it?
22:09:20ForgottenMemorieswitching it into hold then unhold and holding Menu + Select for 20 seconds
22:09:44ForgottenMemorieI can kind of hear my ipod on when I put it clsoe to my hear but the screen is just...blank...
22:10:17linuxstbWhat did you do before your ipod stopped working?
22:10:54 Join Workaphobia [0] (
22:11:15 Join barrywardell [0] (
22:11:46XavierGrhmm weird, the greek lang file has ids that are not present on the english lang file anymore
22:11:56XavierGris this a side effect from langv2?
22:12:00ForgottenMemorieI was messing around with rockbox, then I plugged it into my computer and it stayed on the apple screen and wouldn't detect, then when I unplugged it it just.. died. -_-
22:12:08XavierGrshould I just start over from a fresh english lang file?
22:12:25 Quit rasher ("leaving")
22:12:35*barrywardell wonders what is causing the problem with his server
22:12:43 Join rasher_ [0] (n=rasher@rockbox/developer/rasher)
22:12:49 Nick rasher_ is now known as rasher (n=rasher@rockbox/developer/rasher)
22:13:22 Join japc [0] (
22:13:55pixelmaXavierGr: most probably because of langV2...
22:14:31 Quit lee-qid_ ("aufwiederbyebientotsayonara")
22:14:41rasherXavierGr: I doubt it matters besides cluttering the sourcefile
22:14:46 Quit Domonoky (Read error: 104 (Connection reset by peer))
22:15:02 Join daan [0] (
22:15:11rasherOr rather, I'm fairly certain it doesn't.
22:15:38XavierGrrasher: thing is that it is now too difficult to see what is untranslated because some ids are elsewhere or nowhere to be found, but let me check something first
22:16:14pixelmaI started with a new german lang file when I transformed it to langV2 because the order was also completely different. Not necessary for the function but makes it easier to maintain...
22:16:31pixelmathe initial work was tedious though...
22:17:06XavierGrpixelma: yes, I think I will just do that, long process but more clean and I guess that I will do that once
22:17:17XavierGrthen updating shouldn't be too difficult
22:17:21XavierGrthanks for clearing that up
22:17:23 Quit Frazz ("Leaving")
22:17:31amiconngenlang can create a new .lang file from the old one and english.lang with the postions which need work marken (and iirc also correct order)
22:18:14XavierGrah so the wiki command is still valid
22:18:22XavierGrthanks I will try that first
22:19:05pixelmathere are a few parts that you can copy from the old (like VOICE_A to VOICE_Z for example, so I sliced the old into pieces one by one)
22:20:22 Quit Lear ("ChatZilla [Firefox]")
22:20:52XavierGrso how is voice these days (after the speex commit), no more random crashes when enabled (or strange behaviour)?
22:23:01rasherHaven't tested it personally, but it seems there are issues, but more likely to be tracked down and killed
22:23:21 Join radinp [0] (
22:23:53radinpWhere in the rockbox source are id3 tags read?
22:26:09XavierGramiconn: thanks, now the work I have to do is way less :)
22:27:17 Join TMM [0] (
22:27:52linuxstbraninp: firmware/id3.c
22:31:52ForgottenMemoriemmmm can someone help me, my Ipod is just blank/dead, I was messing around with rockbox, then I plugged it into my computer and it stayed on the apple screen and wouldn't detect, then when I unplugged it it just.. died. -_-
22:34:00 Quit Alonea ("ChatZilla 0.9.79 [Firefox]")
22:36:39 Part Workaphobia
22:36:49 Join Workaphobia [0] (
22:36:55 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:37:24linuxstbForgottenMemorie: All I can suggest is to charge it for longer, and hold MENU+SELECT for longer (and be careful to keep your fingers still whilst holding them).
22:38:55ForgottenMemorieoo... OK, Thanx!
22:40:49 Join Frazz [0] (
22:42:16PaulPositionForgottenMemories - Someone told you the same yesterday, to charge it for a good 12 hours. You came back half an hour later with the same question and now today you still haven't charged it?! "22:29:52Soapand I do mean 12 hours. W/O peeking."
22:44:02WorkaphobiaA few friends and I are looking to implement a feature request or two for rockbox, something simple that we can do over the next couple of weeks. We were thinking of doing something related to displaying lyrics. Is this already supported by rockbox, or are developers already working on it, or should we give it a try?
22:44:02 Quit Genre9mp3 ()
22:44:34 Quit barrywardell ()
22:44:49pixelmaafaik there is a patch in the tracker for it (or even two)
22:45:09 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
22:46:09PaulPositionWorkaphobia - I think someone had a a patch :
22:46:21 Quit advcomp2019 (Nick collision from services.)
22:46:23 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
22:46:33ForgottenMemorieblah, I couldn't hlep it it's only been 9 and a half hours when I looked at it and it was still blank >.<
22:49:09 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
22:49:19PaulPositionForgottenMemorie - Hmm :( Charging through usb or wall? If usb, perhaps it doesn't provide enough current (my friends front ports won't work well so we usually plug stuff in the back even if its ackward to crawl under the table).. Just suggesting.
22:50:18Workaphobiathanks pixelma, paulposition. If that's out, any recommendations as to the best way to find a small feature that needs doing? Should we just continue to surf the feature tracker?
22:51:08 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
22:51:29 Join miepchen^schlaf [0] (
22:51:30 Quit advcomp2019 (Nick collision from services.)
22:51:32 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
22:52:27 Quit nicktastique ("Leaving")
22:53:59PaulPositionWorkaphobia - You could, of course. Or maybe see with Nico P. if that's something that could, or should, be converted to use the Metadata On Buffer thing they put forward lately. (??)
22:54:31Bagderin case anyone hasn't seen it yet
22:55:28PaulPositionAlways the SOCer, you are. :p
22:55:42LloreanI'm quite fond of the face on the back for some reason
22:56:00LloreanMaybe because the only other shirt I have containing "binary" also has a 2 inserted every now and then, probably out of spite
22:56:25scorche|w"It wouldn’t surprise me at all if the binary does mean something if we translate it to ascii…"
22:56:30radinpWhat does Metadata on Buffer mean?
22:56:31 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
22:56:35scorche|wBagder: i tried it as soon as i got mine :)
22:56:39scorche|wnothing :(
22:56:43 Quit advcomp2019 (Nick collision from services.)
22:56:49 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
22:56:58PaulPositionLove the face indeed..
22:56:58 Nick fxb__ is now known as fxb (
22:57:19scorche|wthe mentor summit shirt is slightly different...not by much though
22:57:27 Nick radinp is now known as ShirleyAnn (
22:58:17PaulPositionradinp/Shirley - re: MoB, here's how Nico P. introduced the concept a few months ago :
22:58:20scorche|wBagder: sorry, but did i ever report back tto you regarding the umbrella corp talks?
22:58:31 Quit mf0102 ("Verlassend")
22:58:56 Quit merbanan (Remote closed the connection)
22:59:37preglowi wondered what the hell was up with that face
23:00:10 Nick ShirleyAnn is now known as radinp (
23:00:17PaulPositionDunno, but it looks a bit like that 'transformer voice changer mask' I had when I was 8 y/old...
23:00:38PaulPosition(the SOC face, not the speex decoder ;) )
23:01:53 Join alienbiker99 [0] (
23:04:25 Quit Frazz ("Leaving")
23:05:07tedrockguys it's a tiki mask. stop calling it a monster
23:05:16preglowit's a monster!!!
23:05:30Bagderwe recognize a monster when we see it! ;-)
23:05:50tedrock:( <- this is my sadface
23:08:54Lloreanamiconn: Any objection to my committing the nano fix, assuming all added lines are wrapped in #ifdef IPOD_NANO?
23:09:09 Join OlivierBorowski_ [0] (
23:09:47 Quit J3TC- (Read error: 110 (Connection timed out))
23:10:08***Saving seen data "./dancer.seen"
23:10:10 Quit Nico_P (Remote closed the connection)
23:10:20 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
23:12:32 Quit w0rd54 (Client Quit)
23:12:41 Join atsea- [0] (i=atsea-@gateway/tor/x-75526faf2f3eac64)
23:12:53 Quit qweru ("moo")
23:13:04 Join w0rd54 [0] (
23:14:55 Part radinp
23:16:26 Quit daan (Read error: 104 (Connection reset by peer))
23:22:42 Join stripwax [0] (
23:22:51 Join moos [0] (
23:23:39 Quit OlivierBorowski (Read error: 110 (Connection timed out))
23:24:15Nico_Pmaybe I could give the GsocRoundup2007 page a small update about MoB
23:24:55preglowamiconn: seen any of the voicebox stuff on the ml?
23:24:59preglowNico_P: sure
23:27:59 Join webguest56 [0] (i=ca32ef2e@gateway/web/cgi-irc/
23:28:04 Quit webguest56 (Client Quit)
23:28:16 Join webguest20 [0] (i=ca32ef2e@gateway/web/cgi-irc/
23:28:46 Join webguest13 [0] (i=ca32ef2e@gateway/web/cgi-irc/
23:28:58 Quit webguest13 (Client Quit)
23:29:08Lloreanpreglow: voiceBox is constrained to -q 4 -c 10 by default. Did he say he modified the voiceBox script?
23:29:39preglowLlorean: quite sure he did
23:31:04 Quit bertrik ("bye")
23:33:42 Quit webguest20 ("CGI:IRC (Ping timeout)")
23:36:04preglowi'm fairly certain speex isn't capable of making aliasing noise, so i don't know what they're jabbering about there, though
23:36:07preglowprobably just don't know proper terms
23:36:38amiconnWell, speex uses the rockbox resampler as it's 16kHz, doesn't it?
23:36:42jmspeexpreglow: "Alias noise": what you get when using a linear interpolation resampler
23:36:53 Quit Billenium ("Ex-Chat")
23:38:04jmspeexpreglow: The wideband QMF can also produce a tiny bit of aliasing at low bit-rate (or when decoding narrowband), but most people won't notice it.
23:38:12preglowamiconn: how will that sound different for talk clips and voice files?
23:38:29preglowit's the same resampler
23:38:35amiconnWhat? where?
23:38:43preglowjmspeex: i doubt that's the alias they talk about clips and .voice don't sound different - if you know what you're doing
23:38:58preglowamiconn: read the ml
23:39:36jmspeexpreglow: MP3 already sounds "tonal", so people might mistake the aliasing for MP3 artifacts
23:39:55amiconnE.g, if you generate .talk clips using at&t voices, it is *strongly* recommended to change the sapi5 output format from 22kHz to 32kHz
23:40:00 Join Nic0_P [0] (
23:40:12amiconnOtherwise the sapi engine will produce aliasing artefacts
23:40:16 Join Zagor [0] (
23:40:18preglowamiconn: then please chip in on the ml, i have no idea about using voicebox
23:40:20 Quit Nico_P (Nick collision from services.)
23:40:22 Nick Nic0_P is now known as Nico_P (
23:40:31amiconnThat needs to be done in the script, there's no setting for it
23:41:08 Join JdGordon [0] (
23:44:06preglowwhy not use a higher sampling rate?
23:45:22 Join chuck [0] (
23:47:54 Quit OlivierBorowski_ (Remote closed the connection)
23:48:02Lloreanamiconn: Any objections to my committing the Nano fix?
23:49:01linuxstbLlorean: Do you know oblib's real name?
23:49:19amiconnpreglow: A higher rate per se does not help - you need to select an optimal rate per engine
23:49:38amiconnAT&T works best at 32kHz, and very good at 16kHz - but is bad at 22kHz
23:49:52Lloreanlinuxstb: No, but amiconn's here now, and both checking with him if he's okay with the patch and getting the real name are on the list of "things to do"
23:49:57amiconnThe Microsoft engine, however, works best at 22kHz
23:50:33amiconnsapi_voice.vbs (the one used by is clever and selects the optimal setting (for engines we know)
23:51:21chuckcan anyone tell me what the message "undefined instruction at 0016F6C8 (0)" means? it happens when i try to play any mp3
23:52:20linuxstbchuck: What device are you running Rockbox on?
23:52:41chucka 60 gig ipod video
23:53:19 Quit amiconn (Nick collision from services.)
23:53:28 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
23:54:19linuxstbchuck: Have you just installed Rockbox for the first time?
23:55:08chuckno, i have tried several times over the last month, but can't get any of my mp3's to play
23:56:14linuxstbSo you've never got Rockbox working?
23:56:56chuckoh yeah, everything else works fine, just can't play any mp3's
23:57:08stripwaxchuck - how recent is your build of rockbox?
23:58:18chuckii just download a build earlier today - it is r15699-071119

Previous day | Next day