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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2007-07-17

00:00:35linuxstb_Bagder: I think the intention is to use floating point in the codecs (or at least test it)
00:01:43BagderI agree that it of course should be usable in the end, it just seems a bit premature to me to do any floating point until rockbox at least runs limping
00:01:57linuxstb_I agree with that.
00:02:38 Quit Robin0800_ (" HydraIRC -> <- Chicks dig it")
00:02:45 Join robin_0800 [0] (
00:05:59 Quit tchan (Remote closed the connection)
00:06:08 Quit XpecialGuest ()
00:07:02 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
00:07:44toffe82I was trying to compile the code of ptw419 to make some test for him but it was giving that error about fp, but now I don't know if he use 4.0.3 or 4.1.1
00:08:17 Quit robin_0800 (" HydraIRC -> <- Chicks dig it")
00:08:26 Join Robin0800 [0] (
00:08:27toffe82sure that the fp will be the last option to use
00:10:43 Quit My_Sic (Read error: 113 (No route to host))
00:14:06*Lear wonders what Bagder was referring to...
00:14:26linuxstb_The gigabeat S port
00:14:40BagderLear: didn't you experiment with rockbox on arm with gcc 4.1.x?
00:14:59Bagderor was it coldfire or am I just hallucinating?
00:15:09LearNope, no ARM stuff. Only Coldfire.
00:15:53Bagderok, thanks
00:16:25LearMy H140 still works fine, so I haven't really had a reason to get anything else...
00:16:28 Quit ender` (" Washing your car to make it rain doesn't work.")
00:18:07 Join ender` [0] (
00:22:31 Quit SiebaZ (Read error: 110 (Connection timed out))
00:23:11 Quit elinenbe (Read error: 110 (Connection timed out))
00:25:01 Quit Siltaar (Remote closed the connection)
00:25:54 Quit entheh ("^~")
00:27:08Bagderokay, less repetitions in configure now
00:31:18 Part datsit40ther1
00:32:30 Quit ender` (" -How many Unix hacks does it take to change a light bulb? -Let's see, can you use a shell script for that or does it need a )
00:36:21 Quit bdgraue (Read error: 104 (Connection reset by peer))
00:37:33linuxstb_Bagder: That's looking nicer :)
00:38:03Bagderyes, one step forwards
00:38:33*amiconn wonders whether that breaks iFP target
00:38:52Bagderthe ifp target funnily already had all codecs mentioned...
00:39:22Bagderthe codecs= lines were *identical* for all swcodec targets
00:39:24*amiconn also wonders whether tomal will ever reappear
00:39:47Bagderyeah, I fear we will never see the completion of the ifp port
00:39:58jhMikeSwhat is the ".PHONY:" thing supposed to be for?
00:40:22linuxstb_It prevents make thinking the rule target is a file (IIUC)
00:40:26amiconnWould be the second one after gmini... but the iFp already came further than the gmini ever did :(
00:40:30 Join rotator [0] (n=e@rockbox/developer/rotator)
00:41:30jhMikeSspeaking of incomplete ports, how 'bout that Elio ? :P
00:41:41amiconnAsk linuxstb
00:41:55amiconnThere's also the intended AV3x0 port...
00:42:01 Join petur [0] (n=petur@rockbox/developer/petur)
00:42:06Bagderpossibly the m:robe port helps the elio one
00:42:29amiconnm:robe is PP5020 with monochrome screen
00:42:34 Quit Lear ("ChatZilla [Firefox]")
00:42:37amiconnElio is PP with 16 bit colour screen
00:42:49linuxstb_I just seem to keep finding things to do in the core (or codecs...)...
00:42:49Bagderoh, not that similar then
00:43:25amiconnIt should be similar to H10, or ipod color/photo, or Sansa
00:43:30amiconn(display wise)
00:43:31jhMikeSm:robe has WM8751? (thought I saw that)
00:44:04amiconnlinuxstb: Hehe, similar to me finding other stuff than viewports :/
00:44:35linuxstb_I should try the Elio's LCD again. At the time, I tried all the combinations of existing code I could find, and nothing, and no sign of any familiar-looking registers in firmware disassemblies.
00:45:11jhMikeSthen there's a sandisk job with a tlv320. I wonder what to do about audio drivers sharing multiple targets now if they need to be slightly different.
00:45:16amiconnMaybe the PP emulator could help?
00:45:23 Quit Rob222241 ()
00:45:25*amiconn never tried the emu himself
00:45:51amiconnjhMikeS: Which target is that?
00:45:58 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
00:46:14jhMikeSamiconn: hmmm...there's wiki page up now I think.
00:47:07jhMikeSsomeone linked to it but I've no idea where it is
00:48:29amiconnSingle core arm, 120MHz.
00:49:06*linuxstb_ would be keen on seeing a telechips port - his flash-based DAP with DAB has one
00:49:15linuxstb_But I forget which...
00:49:19amiconnWell, the target specific part of the driver goes in target tree, the common code in firmware/drivers/
00:49:37 Quit Bagder ("*plopp*")
00:49:43amiconnlinuxstb: Telechips can be anything.. the gmini was also telechips
00:50:02amiconn...TCC720 which is CalmRISC16
00:50:38jhMikeSradio chip drivers still need some some more shuffling around for that...mostly the serial interface drivers.
00:50:39amiconn16bit, ummm....
00:51:48amiconnjhMikeS: Yes. This is partially due to the old targets still not fully converted to target tree
00:52:24linuxstb_Ah, google found my own website... (I should move the info to the wiki)
00:52:42jhMikeSthe SH code is everywhere still :\
00:54:48*jhMikeS notices the gigabeat bass control is now broken and acting like adaptive bass is still turned on but showing dBs in the setting
00:57:41toffe82jhMikeS: there was a commit by Gotthardt last week for that
00:58:37jhMikeStoffe82: yeah, he added it, then disabled it using #ifdef and now it seems something else is still goofed
01:01:21 Quit obo ("bye")
01:02:09 Quit petur ("Zzzzz")
01:02:10*jhMikeS will make sure it's not just settings residue
01:05:13linuxstb_Argh, seems the Sansa M200 firmware files in the wiki are hosted off-site and have disappeared...
01:05:17***Saving seen data "./dancer.seen"
01:11:28 Quit matsl (Remote closed the connection)
01:12:40 Join safetydan [0] (i=cbca159f@rockbox/developer/safetydan)
01:14:46 Quit jhMikeS (Nick collision from services.)
01:14:52 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
01:21:45 Join sarixe [0] (
01:25:48 Quit linuxstb_ ("CGI:IRC")
01:28:16safetydanlinuxstb, did you get a chance to look at FS #7440?
01:28:45 Part toffe82
01:30:16 Quit Robin0800 (" I love my HydraIRC -> <-")
01:32:16*jhMikeS is getting disturbed by this bass control thing being whacked
01:39:34linuxstbsafetydan: I forgot... Just compiling it now.
01:43:59linuxstbsafetydan: Maybe I'm just blind, but I can't see where you set the bass and treble cut-off in the driver...
01:44:23 Quit HellDragon (Read error: 104 (Connection reset by peer))
01:44:47 Join HellDragon [0] (i=JD@unaffiliated/helldragon)
01:45:38 Join Thundercloud [0] (
01:46:40safetydanlinuxstb, nope, not blind. I totally forgot about that bit.
01:48:03safetydanSo at the moment it will just be hard coded to the second cut-off value.
01:49:24safetydanlinuxstb: still, you should be able to test if it's actually adjusting the bass and treble
01:49:33linuxstbYes, that seems to be working.
01:50:44safetydanExcellent. Guess I'll just hook up the cutoff values and then it should be ready for comitting.
01:50:47*jhMikeS finds out he put the wrong bits in the header and adaptive boost has been switched on by mistake for some time :P
01:52:12linuxstbAnd no-one noticed?
01:53:38 Join jhulst [0] (
01:53:50safetydanthe uda1380 supports different bass and treble cutoffs doesn't it?
01:54:15linuxstbI think someone mentioned that.
01:54:37*linuxstb sleeps
01:55:05jhMikeSsafetydan: yes
01:55:47safetydanWell good. At least that one I can test if I add support for it.
01:56:36 Join Tr1ckY [0] (
01:56:54 Join RaverDave [0] (
01:57:57RaverDavehello? i cannot seem to retain the backdrop on my ipod video! despite making sure the correct dimensions and colour depth is met...and i put the picture as bmp format in backdrop also!
01:58:28RaverDavewhere does this info get wrote to and what line should it read?
01:59:16RaverDavethe pic displays btw,just wont on next reboot
02:00:33RaverDavein other words i can set it as backdrop and it displays but just for that session,despite copying it to the bacdrop folder also
02:01:19pixelmadid you put it in a folder called "backdrop" or "backdrops" (inside .rockbox)?
02:01:37RaverDaveof course yes,this folder was thwere already i recall
02:02:46pixelmahmm... how are you shutting down your Ipod?
02:03:01 Join kkurbjun [0] (
02:03:31RaverDavethe nromal and middle
02:03:41RaverDaveheld for 5 secs
02:03:52RaverDavetheres no other way as it goes
02:04:01RaverDavevideo ipod
02:04:12 Join Jens [0] (i=Jens@pdpc/supporter/active/Jens)
02:04:17scorcheholing play for a few seconds is the proper way
02:04:37pixelmaah... no... a shutdown is just holding the "play" button - then it should be able to save its settings and will remember them
02:04:55pixelmaon Ipod that is
02:05:29RaverDavehmm..lets see
02:06:39 Quit PaulJam (".")
02:06:48RaverDavelol..well i shut it down correctly..but still no
02:07:25RaverDaveindeed it was play held down
02:08:34RaverDaveyes just double checked i have a backdrops folder within .rockbox
02:08:41RaverDaveand the picture in there
02:10:13RaverDavecould you tell me which config file it gets wrote to,and a line example please?
02:10:27RaverDavei can do it manually then
02:12:53 Join lululoo [0] (
02:12:57lululoohi can someone help me
02:13:03lululooim having some trouble with rockbox
02:13:23pixelmaRaRe: the settings are stored in config.cfg - which will only contain settings that one changed from default... the line will have something with "backdrop" in it, but I'm not sure as I don't use one... Maybe you can save your settings manually (through the option) and then copy that line to your config.cfg
02:13:41pixelmasorry... RaverDave
02:14:10lululoo=( i cant change songs
02:14:11lululooor pause them.
02:14:21lululooit syas Queue Last when i try to select songs
02:14:32lululooand they play .. but yeah. i cant go to the next song
02:14:34lululoofast forward
02:14:37lululooor anything
02:14:39lululoocan someone help me
02:15:30lululooif i try to play a song it just says Queue Last
02:15:38safetydanlululoo, what player do you have? what are you doing to try and skip songs?
02:16:01lululooplayer..? uh
02:16:03lululoorockbox ?
02:16:09lululooim just trying to play songs.
02:16:14lululooand when i press Play
02:16:17lululooit says Queue Last
02:16:19pixelmalululoo: are you using an official build? Also please, write a bit more on one line...
02:16:19scorchelululoo: where are you getting rockbox from?
02:16:21lululooinstead of playing the song?
02:16:25 Quit XavierGr (Nick collision from services.)
02:16:25lululoothe official site
02:16:29 Join XavierGr [0] (
02:16:55scorchedoes it say anything other than queue last?
02:17:35 Join TiMiD[FD] [0] (n=TiMiD[
02:17:39lululooit just sayas
02:17:40lululooqueue last
02:17:41TiMiD[FD]hi there
02:17:42lululooand doesn't do anything
02:18:06scorcheupdate tot eh latest build and try again
02:18:22 Quit Soul-Slayer (Read error: 104 (Connection reset by peer))
02:18:27pixelmalululoo: with player he means, do you have an Ipod, Sansa, H10... Rockbox is not only for one device
02:18:43 Join bordy [0] (n=bordy@unaffiliated/bordy)
02:19:05lululooipod 5g
02:19:13lululoosigh okay
02:19:15 Quit BHSPitMonkey (Connection timed out)
02:19:16lululooguess thast the only way
02:19:23scorche*please* try to write your complete thoughts on one line
02:20:19lululoohaha sorry
02:21:12lululooalright im going to go reinstall it. thanks anyways
02:21:50RaverDavehmm well no joy,a shame as i wanted a backdrop really,i think it looks too naked without one
02:21:54pixelmaRaverDave: the line in config.cfg is "backdrop :/.rockbox/backdrops/yourpic.bmp - btw. works fine here for me on M5
02:22:09 Quit lululoo ()
02:22:26RaverDaveis that with the quotation amrk?
02:22:55pixelmano - sorry... just missed the ending one
02:24:17 Quit Xerion (" ")
02:25:19RaverDaveahaaaa!! thanks for that,it works now,odd how it didnt sdave it within rockbox though huh
02:27:38 Quit XavierGr (Nick collision from services.)
02:27:42 Join XavierGr [0] (
02:28:16safetydanlululoo, maybe you're holding the play button down too long
02:28:25safetydanoh darn, he left
02:29:00DBUGEnqueued KICK RoC_MasterMind
02:31:14 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
02:31:34 Quit saratoga (Client Quit)
02:31:41 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
02:32:04 Quit RaverDave ()
02:41:38 Quit XavierGr (Nick collision from services.)
02:41:41 Join XavierGr [0] (
02:43:33 Quit spiorf (Remote closed the connection)
02:47:21 Part pixelma
02:52:45 Quit jhMikeS (Read error: 104 (Connection reset by peer))
02:53:34 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
02:53:51 Quit daurnimator (Read error: 110 (Connection timed out))
02:54:52 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
02:56:02 Join aliask [0] (
02:57:36 Join lukaswayne9 [0] (
03:01:33Nico_PJdGordon: hi
03:01:56Nico_PI have some pretty big changes I've been working on for the past fex days
03:02:04JdGordoncool :)
03:02:38Nico_Pringbuffer in the ringbuffer... more complicated than I though but it's starting to come along nicely :)
03:03:19Nico_Pright now it does short files perfectly even when they wrap around the main buffer, and long files should be ok soon
03:04:04JdGordonhow does it decide when to use the inner ring buffer?
03:04:20Nico_Pwhen filerem > 0
03:05:20***Saving seen data "./dancer.seen"
03:06:07JdGordoneven if the buffer has room?
03:06:26 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
03:06:28 Join toffe82 [0] (
03:07:19Nico_Pif the buffer has room, the file will be buffered entirely in one go
03:07:30JdGordonnot necasserily
03:07:43Nico_Pmaybe you mean if room was freed from the next handle ?
03:08:07Nico_Phaven't taken that into account yet
03:08:21Nico_PI only have one file on buffer ATM anyway
03:08:34JdGordonso we arnt going to use the linking 2 handles together approach?
03:08:53Nico_Pfor the same file you mean ?
03:10:32 Quit Jens ()
03:10:58JdGordonI was hoping the inner rng would only be used when there is only 1 file and its metadat in the buffer and there is no other option
03:11:10saratogai think i figured out why low bitrate WMA files still don't work in my decoder
03:11:16Nico_Pthat will probably be the case
03:11:29saratogait seems enabled the low bitrate features somehow fills my trig lookup tables with zeros
03:11:35 Nick Tr1ckY is now known as Tr1ckY^Gone (
03:11:59Nico_Pbut we can't really distinguish all the cases... most often a file will just be buffered normally though... no ringbuffer
03:13:17JdGordonand even if it doesnt fit in one go, the space after it will most likely be freeable before the ringbuffer is needed
03:13:54Nico_Pit's a case I'll have to add
03:14:01 Part RoC_MasterMind ("Leaving")
03:15:02 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
03:19:20Nico_PJdGordon: I'll commit soon even if it doesn't work completely
03:19:53JdGordonno prolem
03:20:07CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
03:20:07*JdGordon wishes he had motivation to be more helpful
03:23:34Nico_Pwhat kind of motivation ?
03:25:00 Quit jhMikeS (Nick collision from services.)
03:25:06 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
03:25:11JdGordonlike not being scared every time i open playback.c :p
03:27:06 Join krylon [0] (
03:27:58krylonHow can I disable scrolling altogether? As in, text scrolls from left to right if it exceeds the width of the screen
03:28:33 Quit lukaswayne9 ("Ex-Chat")
03:29:44JdGordonI dont tihnk you can
03:29:49JdGordonyou can set a long delay though
03:31:41JdGordonbreaky time ttyl
03:40:03 Join BHSPitMonkey [0] (
03:44:19 Quit sarixe ("Peace")
03:47:08 Quit Thundercloud (Remote closed the connection)
03:48:23 Join DC1 [0] (
03:49:06 Join EbErT [0] (
03:50:03EbErThey all, i was thinking about getting a gigabeat, to be used for rockbox dabbling−− is this one of the top choices for smooth rockbox operation? if so which series?
03:50:51saratogai prefer the sansa to the gigabeat
03:50:54saratogabut thats just me
03:50:54BHSPitMonkeyEbErT, I hear it is.
03:51:20homieloweEbErT: They are hard to come by right now, The "S" is in development, but the "F" works well
03:51:45saratogaFs are really easy to get
03:51:49saratogathey're dirt cheap on ebay
03:51:51BHSPitMonkeysomeone should totally get rockbox running on the Insignia players -.-
03:51:56EbErTso just the S and F series, ya saw lots on ebay
03:52:04 Quit saratoga ("CGI:IRC")
03:52:15homieloweI got myself a "S" for $78
03:52:27 Part homielowe
03:52:34EbErTi've just not been real satisfied w/ rockbox on my ipod color
03:54:17BHSPitMonkeyEbErT, I can has it?
03:54:43EbErT"can you have it"?
03:55:07kryloni already called dibs. back off
03:55:19EbErTwhat's the diff between S and F series gigabeats
03:56:37safetydanEbErT, F works with Rockbox, the S currently does not.
04:05:02 Quit aliask ("ChatZilla [Firefox]")
04:09:43 Quit jhulst (Remote closed the connection)
04:15:01 Quit XavierGr (Nick collision from services.)
04:15:05 Join XavierGr [0] (
04:21:48 Join homielowe [0] (n=chatzill@
04:21:49 Quit homielowe (Client Quit)
04:29:47 Quit kkurbjun (Read error: 113 (No route to host))
04:52:08 Quit mbr (Read error: 110 (Connection timed out))
05:00:16 Quit BigMac ("Ex-Chat")
05:01:24 Quit DataGhost (Nick collision from services.)
05:01:30 Join DataGhost [0] (
05:05:23***Saving seen data "./dancer.seen"
05:07:16 Join chrisjs169|brb [0] (n=ubuntu@unaffiliated/chrisjs169)
05:07:50chrisjs169|brbKinda hate 'help me' messages myself, but sansapatcher isn't recognizing my e250
05:09:02chrisjs169|brbI had filesystem issues on the Sansa (rockbox wouldn't boot, files missing, the rocks dir having weird filenames, etc) and went to recovery mode, placed sansa.fmt in it, along with the e200 (non-R) firmware and bootloader
05:09:17chrisjs169|brbin addition, my Sansa is now being recognized as an 'SD Player'
05:10:54chrisjs169|brbso uhh...any suggestions?
05:16:14 Join Rob2222 [0] (
05:23:05chrisjs169|brbsorry - RCM (from ABi) pointed out something I didn't even consider - sansa.fmt formatted it as fat16
05:24:57 Part bordy
05:34:54Nico_PJdGordon: committed
05:40:40 Join elinenbe [0] (
05:43:09chrisjs169|brbsansapatcher still isn't working
05:43:29 Quit Nico_P (Remote closed the connection)
05:46:29 Join Nick_Brackley [0] (i=7dff0332@gateway/web/cgi-irc/
05:49:18 Quit XavierGr (Read error: 110 (Connection timed out))
05:49:40 Join kkurbju1 [0] (
05:50:03chrisjs169|brbit was user error :P
05:54:46 Join Dustin [0] (i=d15957ec@gateway/web/cgi-irc/
05:55:00Dustinis this the rockbox?
05:55:31 Quit Dustin (Client Quit)
05:55:34*scorche looks around
05:56:27 Quit Nick_Brackley ("CGI:IRC (Ping timeout)")
05:56:34 Join jurrie [0] (
06:00:16 Quit donutman25 (Read error: 104 (Connection reset by peer))
06:00:30 Join donutman25 [0] (n=chatzill@
06:01:31 Join kkurbju2 [0] (
06:03:53 Join homielowe [0] (n=chatzill@
06:17:27 Quit kkurbju1 (Read error: 110 (Connection timed out))
06:17:40 Quit Nibbier (Read error: 110 (Connection timed out))
06:18:18 Join Nibbier [0] (
06:21:00 Part homielowe
06:22:21 Quit anonymous_ ("Ex-Chat")
06:25:27 Quit lazka (Remote closed the connection)
06:31:12 Join ptw419 [0] (
06:36:31 Quit rotator ()
06:36:42 Join aliask [0] (
06:53:07 Quit kkurbju2 ("Leaving.")
07:05:24***Saving seen data "./dancer.seen"
07:13:51 Join kfazz [0] (
07:21:14 Join kubiixaka [0] (
07:23:21 Quit donutman25 ("ChatZilla [Firefox]")
07:24:55 Join kkurbjun [0] (
07:27:31*amiconn thought that all ipods have the same, proprietary connector at the bottom, but that's not true for the G1 and G2
07:28:34scorcheaye...they only had firewire...
07:29:40scorchetis why they sometimes call the 3G model the "dock connector" iPod
07:29:43amiconnYes, but I didn't know they use a standard (?) firewire socket at the top
07:30:16*amiconn likes standard connectors is a standard 6-pin connector
07:31:09amiconnUnfortunately firewire doesn't use only one connector type (but the same applies to mini usb connectors)
07:32:21scorchealthough, the nice thing about the dock connector port is that you can easily access so many things from it (like line in, line out, etc)
07:32:50scorchebetter than nothing, but i would rather prefer separate connectors of course
07:34:09 Part lymeca ("Vanity is not sustainable.")
07:39:54 Quit kubiix (Read error: 110 (Connection timed out))
07:42:51TiMiD[FD]I was thinking about adding support for different font on the remote than on the main unit, would that be welcome ?
07:44:51TiMiD[FD]if I remember that point was evoked a long time ago
08:04:47amiconnYes, if done in a resource conserving way
08:05:00amiconnmarkun had some ideas how to make font caching more efficient
08:05:12 Quit EbErT ()
08:09:18 Join pondlife [0] (
08:14:55 Join Transience [0] (
08:18:04TiMiD[FD]hmm I don't know that uch about the currnt font caching system
08:18:53TiMiD[FD]anyway excepted in the case when the 2 fonts are the same, you have an overload ...
08:19:25 Part toffe82
08:20:53TiMiD[FD]it would also be nice to be able to have different fonts on the same screen (for wps for ex) but this may be a little too much memory expensive
08:21:42 Quit hannesd (Read error: 110 (Connection timed out))
08:22:40 Part safetydan
08:23:42amiconnTiMiD[FD]: It wouldn't be possible anyway atm without really nasty hacks, because the scrolling system is line based
08:23:52amiconnIt should become easier with viewports
08:24:22*amiconn should start implementing viewports, ghhrrrmm :\
08:25:17TiMiD[FD]yes of course, but with pixel base positionning it would become possible
08:25:36TiMiD[FD]that would imply a lot of changes
08:28:28TiMiD[FD]I don't know what are the goals of the project regarding this matter, are you aiming at someting more flexible, more close to computer gui interface in which components can be freely positionned ?
08:29:22 Join RoC_MasterMind [0] (
08:30:19amiconnBasically, viewports will allow to use parts of the screen for drawing operations like the whole screen is used now
08:30:28amiconn(with coordinate transformation and clipping)
08:31:35amiconnSome of the foundations in the lcd drivers will be changed too, allowing scrolling text to keep their attributes (foreground/background colour, inverse, possibly algorithmic emboldening)
08:32:08TiMiD[FD]ok so something intermediate
08:32:10amiconnScrolling text elements will then be what I call "sub-viewports" or inherited viewports
08:32:34amiconnBacking storage is a non-no because we don't have dynamic memory
08:33:08*scorche wonders when we will see that wikipage =P
08:33:37amiconnViewports should help making things like status bar drawing, list drawing etc a lot simpler as they don't need to take care of the other parts of the screen
08:33:49TiMiD[FD]I was wondering wether some higher level library like the one they have in ipod linux wouldn't be a better idea
08:34:05 Join ender` [0] (
08:34:22TiMiD[FD]to be able to manipulate component
08:34:27amiconnipl has malloc
08:34:35amiconnWe don't, and we don't want to
08:34:40TiMiD[FD]hmm right
08:35:15TiMiD[FD]that's maybe still possible without
08:35:45TiMiD[FD]I didn't looked at their API
08:36:29TiMiD[FD]anyway viewports should indeed allow an easier gui programing
08:37:31*amiconn thinks there are 2 main reasons why he didn't start implementing them yet
08:37:44TiMiD[FD]I wonder how much their podzilla toolkit relies on dynamic memory allocation ...
08:38:19amiconn(1) laziness - there are probably a lot of pitfalls which require changing hundreds of places, test, change, test, reiterate...
08:38:44amiconn(2) I don't see much point in polishing the upper levels if the foundations are not really stable
08:38:53amiconnAnd they are not, at least on swcodec
08:40:00amiconnThat's why I am trying to get PP working properly with cpu scaling and hopefully other, related stuff
08:40:07TiMiD[FD]that would imply a major rewrite of almost all the "app" related stuffs
08:40:31TiMiD[FD]but that doesn't depend that mluch on low level tasks
08:40:35amiconnI don't think a rewrite of all app stuff would be necessary
08:40:47TiMiD[FD]not from scratch
08:40:55TiMiD[FD]but a major rework
08:41:16amiconnRockbox is still very stable on archos, even with all the new stuff like settings and menu system
08:41:47amiconnIt's mainly the hardware layer for PP targets, plus playback and related stuff like voice that has nasty problems
08:41:54amiconn..on swcodec
08:42:05TiMiD[FD]I noticed ... :p
08:42:12 Join Rob222241 [0] (
08:42:22 Join daurnimator [0] (n=fake@unaffiliated/daurnimator)
08:42:25TiMiD[FD]the playback engine seems quite broken too
08:42:28amiconnThe cpu 100% boosted with voice bug really annoys me
08:42:45amiconnI switched back to archos recorder for in-car because of it
08:43:25amiconn...even though I really like the possibility to play flac on my H180
08:43:50TiMiD[FD]the original firmware just works well for me :)
08:44:09TiMiD[FD]for ogg and mp3
08:44:46 Join homielowe [0] (n=chatzill@
08:46:39 Quit RoC_MasterMind (Remote closed the connection)
08:47:04TiMiD[FD]I wonder which task should I do then ...
08:47:08 Join Sidlet [0] (
08:47:22TiMiD[FD]fonts didn't seemed that hard
08:48:44 Quit darkless ("Leaving")
08:48:50 Join darkless [0] (
08:50:12TiMiD[FD]there are some bugs I would like to be fixed, but they seem to only happen to me, and I cannot develop on my device during working hours
08:50:16TiMiD[FD]only on the sim
08:50:21 Quit ptw419 ()
08:50:38 Quit midkay ("Leaving")
08:51:51 Quit DC1 ("$4E75")
08:53:13 Quit kkurbjun (Read error: 110 (Connection timed out))
08:55:32 Join ptw419 [0] (
08:57:11amiconnHaha, $4E75 <== m68k/coldfire 'rts' instruction
08:59:56 Quit Rob2222 (Read error: 110 (Connection timed out))
09:00:10*aliask doesn't get it
09:00:30GodEater_thank god - I don't either
09:01:10TiMiD[FD]that's because asm isn't your mother tongue !
09:01:19TiMiD[FD](I don't get it either ...)
09:01:37 Join miepchen^schlaf [0] (
09:02:18 Join midkay [0] (n=midkay@rockbox/developer/midkay)
09:05:26***Saving seen data "./dancer.seen"
09:06:04 Join hiromi886 [0] (
09:06:50hiromi886Hey, in the .wps file, is there a way to do a half space or something
09:07:00hiromi886if anyone has any idea what i'm talking about haha
09:07:24hiromi886downwards I mean
09:07:51hiromi886Some things aren't properly lining up, if i could get them to shift up or down a half carriage return thingy, if would be perfect
09:08:37 Join petur [0] (n=petur@rockbox/developer/petur)
09:08:41 Part Transience
09:09:23 Quit Anakin (Read error: 113 (No route to host))
09:12:39 Join dan_a [0] (n=dan_a@
09:13:59 Quit homielowe (Read error: 104 (Connection reset by peer))
09:14:02GodEater_hiromi886: no - there's no way to do that - if they're not lining up how you like, you'll have to use a different size font
09:14:13hiromi886doh i was afraid you'd say that
09:16:56 Join kaaloo [0] (
09:17:09 Part kaaloo
09:23:11*markun reminds amiconn of his viewports plans :)
09:23:44 Join Davo_Dinkum [0] (
09:24:16 Join davina [0] (
09:27:01 Quit aliask ("ChatZilla [Firefox]")
09:29:19GodEater_hiromi886: or depending on what they're not lining up with (I suspect some sort of background picture) - then change that!
09:31:21JdGordondoes anyone know how decoded audio gets from the codec to the pcmbuf?
09:31:41JdGordondoes it bypass playback.c? or am i missing the important bit in it?
09:34:09 Join linuxstb_ [0] (i=5343d4aa@rockbox/developer/linuxstb)
09:34:11hiromi886i'm gonna have to edit the bg pic haha
09:34:15hiromi886thanks ^^
09:35:24linuxstb_JdGordon: The codecs call ci->pcmbuf_insert() with the PCM data. It will then be processed by the DSP code, and stored in the PCM ringbuffer waiting to be sent to the DAC.
09:36:08JdGordonok, trying to work out how much work is needed to try getting our buffering code into the core
09:39:56Davo_Dinkumtypo :/
09:40:13*JdGordon wonders how much of playback.c can be completly ignored to get some simple playback happening
09:44:32*pondlife too!
09:44:35 Quit miepchen^schlaf ("Verlassend")
09:45:15pondlifeback in a bit..
09:46:54 Quit linuxstb_ ("CGI:IRC")
09:48:13JdGordonNico is in france isnt he?
09:48:17pondlifeamiconn: Just in case you hadn't seen it - you may want to look at FS #7398.
09:48:41pondlifeJdGordon: yes
09:48:54JdGordon9.50am... wake up time for him....
09:49:00*JdGordon summons Nico_P!
09:53:41*pondlife hacks into Nico_P's alarm clock
09:54:03hiromi886This is annoying lol, the time won't fit anywhere!
09:54:30hiromi886I guess thats why I have a cellphone tho
10:05:13hiromi886Okay i've been poking through the manual and still can't figure this out, if anyone would mind pointing it out
10:05:27hiromi886If you originally transfered all your music through itunes in playlists already
10:05:33hiromi886Can rockbox pick those up?
10:05:46scorcherockbox cannot read the itunes playlists, no
10:05:53hiromi886gah, i was afraid of that
10:06:30GodEater_it can read the music files though
10:06:38GodEater_you just need to read up on using the Rockbox database
10:06:42hiromi886so my only real option is to copy all the mp3s onto root and use my winamp playlists?
10:06:53GodEater_the playlists you'll have to create again from scratch
10:07:13hiromi886well, ive had limited success sticking my winamp playlists straight into the root/playlists folder
10:07:14linuxstbhiromi886: Most people create a /Music/Artist/Album/NN-Trackname.mp3 structure on their ipod's disk.
10:07:15GodEater_hiromi886: no - read up on using the database
10:07:19linuxstb(or similar)
10:07:44hiromi886I wish I was that organized now @@
10:08:14hiromi886Alright thanks tho ;p
10:08:19GodEater_provided your mp3s are well tagged, the Rockbox database will take care of organising them for you
10:09:17linuxstbOr you run a tagging program (GodFather?) which can organise your mp3s based on their tags.
10:09:18hiromi886yeah, my genres are all over the place
10:09:29pondlifeJdGordon: How's the buffering going? Almost ready to listen to music..?
10:09:30hiromi886hm, that sounds like a decent idea
10:12:57 Join Redbreva [0] (
10:13:34scorcheRedbreva: i will check on that upload code to see what i can do in a bit...writing the wiki page atm
10:13:48RedbrevaOK ;-)
10:14:38scorcheas well, when we move the sites around, dont forget to make your uploader.php send the mail to a different place =P
10:15:15hiromi886ah neat i think i got this now ;p
10:19:04 Join printfXh4 [0] (
10:19:40 Part printfXh4 ("Leaving")
10:19:52 Join Siltaar [0] (
10:20:20linuxstbRedbreva: Do you know exactly what OS your server is running?
10:20:57 Join obo [0] (
10:22:07JdGordonpondlife: nico has the plugin working corrrectly
10:22:19JdGordonpondlife: im going to try merging it into the core tonight
10:22:29JdGordoni doubt ill get far, but you never know :)
10:22:43pondlifeI saw that the buffer now consists of one ring per file.
10:22:49JdGordonyes and no
10:24:04JdGordonif the whole file fits in the buffer its a straight buffer... if it doesnt fit (and opefully if another handle cant be freed to make it fit) then it becomes a ring buffer
10:24:10JdGordonim not really sure how nico has done it
10:24:50pondlifeWell we should make sure he wikis it....
10:25:34 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
10:26:05Redbrevalinuxstb: No, it's simply a 'Linux Hosting' package with 1&1
10:27:02 Quit idnar (Nick collision from services.)
10:27:04 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
10:27:09amiconnJdGordon: I wonder how that method guarantees the minimum amount of spinups...
10:27:35amiconnmarkun: Now we can slap each other with wannados ;)
10:27:49scorcheRedbreva: you can likely ssh in and find out
10:28:47pondlifeamiconn: We just need to know that the buffer is filled completely with useful/future data after each spinup, right?
10:28:53JdGordonamiconn: like I said, thats nico's thing, I havnt even looked at his code for that yet
10:31:35amiconnI mean the ringbuffer idea + "pinning" of metadata
10:32:02JdGordonoh, na, thats should be fine, as long as the metadata doesnt waste too much of the buffer
10:32:09JdGordonwhich would only be a problem on hwcodec
10:32:11markunamiconn: exactly ;)
10:32:46amiconnIf the metadata is e.g. in the middle of the physical buffer, a ringbuffer can take only half of it. Or am I missing something here?
10:34:15JdGordoncorrect.. I had the idea where we would use the linked list of handles and create a new handle on the other end of the data which would be a continuation
10:34:40amiconnI'd rather move the metadata out of the way...
10:34:49 Join safetydan [0] (n=dan@rockbox/developer/safetydan) part of the byte swap?
10:35:29amiconnI still think one (or two with enough ram) metadata slots outside the main buffer would be better
10:35:42JdGordonwe cant move it, we dont know "who" is still using it
10:36:12pondlifeamiconn: But the metadata might be rather large... album art for example?
10:36:30JdGordonyeah, we dont want to limit the total size of the data that can be stored
10:36:50JdGordonwe are thinking eventually having AA, lyrics, ..., for each track
10:36:52pondlifeThe different targets have different requirements for their memory capacities.
10:37:12amiconnI'd rather put a limit on the metadata size
10:37:31amiconnThis limit should of course be memsize dependent
10:38:03JdGordonlimits also mean that its a big RAM waste for people who dont use the feature
10:38:08*JdGordon wont bother with AA
10:38:36scorchelinuxstb: Linux infong292 #1 SMP Wed Jan 31 02:48:22 CET 2007 i686 unknown
10:38:51scorcheresults from uname -arv
10:39:52linuxstbamiconn: There's also other "internal" codec metadata that it would be nice to store - seek tables for example. It would be nice to keep this with the other metadata, and load it via get_metadata() instead of making the codec parse the file again.
10:39:54GodEater_scorche: is there a *-release file in /etc
10:40:14linuxstbscorche: Thanks - "x86 Linux" is all I wanted to know..
10:40:36*pondlife will be happy as long as the playback code ends up nicely commented and structured.
10:40:41 Join Llorean [0] (n=llorean@rockbox/administrator/Llorean)
10:40:42 Join idnar [0] (i=mithrand@unaffiliated/idnar)
10:40:53scorchelinuxstb: well, you said "exactly" =P
10:40:55amiconnOf course, if you can come up with a way to keep all metadata on buffer and *not* waste spinups with tracks larger than the buffer, then all would be good
10:41:23*amiconn considers tracks larger than the buffer quite common
10:41:30JdGordonamiconn: tracks larger than the buffer still wont cause a spin up untill the buffer is under a certain level...
10:42:00pondlifeTracks larger than the buffer are very common
10:42:08amiconnYes, but a half-size ringbuffer will cause twice as many spinups as a full ringbuffer
10:42:30linuxstbIt would be nice to always have next-track info available, even if the current track is bigger than the buffer. I'm not sure if that happens currently.
10:42:49LloreanIt doesn't
10:43:22LloreanBut next track info is fairly limited anyway, isn't it?
10:43:34LloreanYou could have a fairly small struct for next track info permanently available.
10:44:01LloreanWithout it entailing the same "limiting metadata size" that a fixed size struct would mean with current-track.
10:44:23linuxstbThat would mean parsing the file twice though.
10:44:39TiMiD[FD]and no AA for next track ...
10:44:52linuxstbWe want that?
10:44:55LloreanIt wouldn't mean no AA...
10:45:05JdGordonimho we should not do anything with the next track unless its in the buffer
10:45:16Lloreanlinuxstb: Probably not, I'm just saying that the concept of not limiting metadata size doesn't apply to "Next track"
10:45:22 Join spiorf [0] (
10:45:23linuxstbJdGordon: But that makes the UI unpredictable (to a user)
10:45:26LloreanAlso you'd only parse the file twice if current-track was longer than buffer
10:45:59linuxstbIMO, if we support "next track" info, it should always be available.
10:47:09JdGordonnot if it means another spin up to get it
10:47:33linuxstbJdGordon: Of course not - we wouldn't implement it that way.
10:47:35LloreanBut it doesn't mean another spin up, just a small amount more time still spinning.
10:47:42 Quit hiromi886 ()
10:48:26 Quit idnar_ (Connection timed out)
10:49:02scorcheRedbreva, linuxstb, Llorean: wiki page is up (just missing the future vision/things that would be nice page, but you all know what that entails...that will be up soon) ThemeRepository
10:49:20Lloreanlinuxstb: How much of a hit is it to load and parse one track's metadata extra with every full buffer fill?
10:49:22*JdGordon really wants nico to show up before midnight my time :p
10:49:38LloreanWhile the disk is still spinning.
10:50:11JdGordonlinuxstb: ok, i tihnk the use of metadata needs to be limited, for the new buffering stuff we are using that term for AA, lyrics, id3, everything... what you want is just the next tracks id3 info? _that_ im ok with having
10:50:21scorchei suppose i could have waited to ping linuxstb after i put the code into SVN...
10:50:37JdGordonoh dear... who gave scorche commit access! :p
10:50:46scorchei dont have it yet! =P
10:50:52*JdGordon shuts up
10:51:05linuxstbJdGordon: I was initially thinking about all the metadata, but Llorean's idea of just buffering the "core" id3 tags makes sense.
10:51:26LloreanYou only have "Next" info on a few strings anyway.
10:51:55linuxstbLlorean: No idea really, but if the user wants it (i.e. is using a WPS with next track tags), then I think Rockbox should go to the effort of reading it. Maybe it could be conditional on the current WPS...
10:52:37linuxstbAnd I guess for database users, it could be populated from there.
10:52:55JdGordonI'm all for showign the next tracks info (and AA for that matter), but unless we have a seperate buffer, which we dont want, then only the id3 info should be read if the track inst in the buffer
10:53:05Lloreanlinuxstb: That makes a lot of sense
10:53:56 Join Neovangl1st [0] (n=Neovangl@
10:53:59LloreanI just thought that at the end of every buffer fill, you could just load the first off-buffer song's core-ID3 info. So it's once per full buffer, while the disk is still spinning, and I guess only if the WPS shows it.
10:54:28JdGordonscratch the last part of the sentance...
10:54:48 Quit Neovanglist (Nick collision from services.)
10:54:48JdGordonthat just adds more comlication to it
10:54:56 Nick Neovangl1st is now known as Neovanglist (n=Neovangl@
10:56:32JdGordon:( merging this is gonna be more a PITA that i was scared of
10:58:43JdGordondoes anyone know which audio_* functions in audio.h are aboslutly crutial to get playback happening?
11:00:15 Nick Redbreva is now known as Redbreva_Away (
11:00:19JdGordonchanging topics completly.... how do I see which files were chaned in each commit?
11:01:01 Quit Redbreva_Away ("ChatZilla [Firefox]")
11:01:53amiconnlinuxstb: The current swcodec playback does provide next track info all the time. The hwcodec engine does only provide it if the track is already in the buffer
11:02:08LloreanJdGordon: svn log -v
11:02:31Lloreanamiconn: I don't get next track info with long flac files, afaik.
11:02:48JdGordonthanks Llorean
11:03:22JdGordonbugger... he didnt keep the test app updated with the test plugin :(
11:04:51 Join RoC_MasterMind [0] (
11:05:27***Saving seen data "./dancer.seen"
11:07:29 Join Wiwie [0] (
11:10:46linuxstbscorche: When are you planning to put your PHP in SVN?
11:12:37scorchelinuxstb: when i get commit access and the site is live
11:13:12scorcheif it will be a while, i will have someone else commit it around the time when the site goes live anyway
11:13:13bluebrotheris there some test site around? /me is curious ...
11:13:33scorchebluebrother: is my working copy
11:14:02linuxstbscorche: Maybe it could be a good idea to share the code privately with some others before it's "published" in SVN - just in case there are any obvious security issues.
11:14:03scorcheit is largely ready to go live except for a few bugs
11:14:51scorchelinuxstb: such as?
11:15:17JdGordonscorche: I wonder if the downloads should be there? maybe rename them to ... umm... i dont know
11:15:22scorchebut, yes that sounds like a good idea
11:15:32scorcheJdGordon: i dont know what you mean
11:15:45JdGordonthe releases, currnet built, extras links
11:15:57bluebrotherwell, as the page is about themes maybe it's better to leave out the download links
11:16:02JdGordondownloads suggests that they are downloads for the themes
11:16:15bluebrotherand put another link to the rockbox site prominently
11:16:24linuxstbscorche: Anything... badly written PHP code (I'm not suggesting your code is) could allow people to, for example, write to arbitrary locations on the server, delete files, etc etc.
11:16:25scorchei thought about that...but yes, that works
11:16:43scorchelinuxstb: most of the code isnt mine in the first place, but yes, i will =)
11:17:16scorchebluebrother: such as an explicit link under the rockbox logo?
11:17:25scorche(the logo is a link there already)
11:17:38JdGordonscorche: and a link to CustomIcons would be nice, or is this not for them?
11:17:52scorcheJdGordon: this is just for themes
11:18:05bluebrothersomething like this. Or add a new category "Rockbox" with a link "Rockbox website"
11:18:06*JdGordon thnks icons are part of themes as much as wps are
11:18:41*bluebrother agrees with JdGordon on this one
11:18:47scorcheJdGordon: i agree, and if the theme uses a different iconset, they are instructed to include it
11:18:52 Quit spiorf ("Read error: 110 (Connection timed out)")
11:19:01linuxstbShouldn't the left menu be the same as the main website (and wiki, and forum) ?
11:19:07 Join spiorf [0] (
11:19:14JdGordonbluebrother: thats a first!
11:19:24bluebrotherJdGordon: indeed :)
11:19:39bluebrotherI'm surprised by myself *g*
11:20:08scorchelinuxstb: not necessarily...the forum is slightly different, flyspray doesnt have it, and there are many important links for navigation of the theme repository that need to be on there
11:22:27linuxstbscorche: The forum is just missing the "Search" box, which would be confusing (probably the same for themes). However, the "assistance required" section could simply be a single link to a "ThemeAssistance" wiki page or similar. That just leaves 4 other links, which could be a menu across the top, similar to the forum buttons.
11:23:19linuxstbBut it's your site... ;)
11:23:23bluebrotherand does the deep link to the manual make sense at all? It links to the Ipod Video, but wps tags differ between targets
11:23:26 Join The-Compiler [0] (
11:23:33bluebrother(in terms of which tags are available)
11:24:31JdGordonyeah, shouldnt it link to the page for the selected target?
11:24:56scorchewell, they havent selected a target yet in some screens
11:25:34bluebrotheryou could present a list of links depending on target of course.
11:25:35JdGordonthen it should link to a generic page
11:26:07scorchedo all of you think it would be better to just make the sidebar the same as the forums sidebar and have a nav menu at the top like linuxstb suggested then?
11:26:25scorchelinuxstb: dont say it is my is rockbox's =)
11:26:36JdGordonare codecs run in the codec thread? or the playback thread?
11:27:09*JdGordon getting so confused :p
11:27:16Jscorche: a good rule of thumb for UI is the one of least suprise for the user. Keep things the same where possible I recommend.
11:27:25bluebrotherscorche: well, I'm not completely sure. Having the same menu indeed has advantages.
11:27:31 Join mbr [0] (
11:27:49bluebrotherbut otoh the menu should have the important links, and those are different for the themes page
11:27:59scorchethe thing is, are 2 nav menus desired?
11:28:13LloreanWhy do we need a link to the WPS Tags manual page if we have the wiki page?
11:28:22LloreanAren't they the same content just laid out differently?
11:28:29scorchei could always have the regular sidebar with a few themes links added to the top
11:28:43bluebrothermight be better.
11:29:08bluebrotheryou could link form the "Documentation" to a page directing further
11:29:11linuxstbscorche: IMO, you're the one actively working on, so you get to choose ;)
11:29:45 Join Entasis [0] (
11:30:13 Quit RoC_MasterMind (Remote closed the connection)
11:30:32scorchebluebrother: if i went that way, i would just leave the whole of the Assistance Required? section in the dust and possibly link the relevant pages in the guidelines page if they need to be linked somewhere
11:30:41 Join printfXh4 [0] (
11:31:41JdGordondoes this sound like thread overkill...? codec thread which runs the codecs and keeps the correct codec loaded. playback thread which the UI interacts with to get track data, skip tracks, seek, etc. and buffering thread to make sure the buffer stays full. ?
11:31:50JdGordon+ voice thread later
11:31:54scorchelets see how this looks...
11:37:01*JdGordon back in 30min
11:39:21linuxstbscorche: Regarding the submission page - shouldn't we be asking for full, real names, rather than nicks?
11:39:36linuxstbForget that, I've seen the second box...
11:40:35 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
11:41:07scorchelinuxstb, bluebrother: better? i can shrink the sidebar a bit, but there is stillt he issue of "Downloads"
11:41:31scorcheand it is a bit long...
11:41:36 Quit The-Compiler (Connection timed out)
11:41:40linuxstbI can't see any change on the upload.php page...
11:41:56scorchei only changed it for index.php
11:42:04linuxstbIt's not a global include?
11:42:07 Join Nick_Brackley [0] (i=7dff0332@gateway/web/cgi-irc/
11:42:24safetydanAnyone with an iPod VIdeo kind enough to test for me?
11:42:27*linuxstb needs to work on this php ;)
11:43:17scorcheas i said..i tried to cut corners a bit so i could get the site out quickly...a lot of the stuff it quite messy still as it was when it got handed to me
11:43:56scorcheas well, i am going to need to do a ton of cleanup before i place it in SVN, as there are things like the whitespace whis is *wildly* inconsistent
11:44:17scorcheah...redbreva left so i am not hurting his feelings =)
11:45:18scorchei think the whitespace is done every which was possible in the whole of the site...
11:46:37midkaysafetydan: i'd be glad to if you'd compile it for me, i can't atm.
11:47:12 Join PaulJam [0] (
11:48:35scorchelinuxstb: your thoughts on the current sidebar in the root page?
11:48:39bluebrotherscorche: looks better, but as you said it's somewhat long :)
11:48:58scorchebluebrother: i suppose i could remove the search area...
11:49:07bluebrotherhow about using some javascript and hide the links below the "title" boxes
11:49:15pondlifeGodEater: I'm ok for DevConPubv2, same venue?
11:49:19bluebrotherand only show them upon hover / clicked?
11:49:56linuxstbscorche: Thinking about the upload page, shouldn't the declaration include a statement to the effect that either a) the person uploading the theme is the copyright holder in all the content; or b) it is using content from other CC (or compatible) themes, in which case the copyright holder of the original theme should be referred to? I'm not familiar with the license being used, but there doesn't seem to be a way for the user (apart from
11:49:56linuxstbin the general description box) to provide licensing information.
11:50:00scorchehrm...that would get annoying imo...and the main site isnt liek that
11:50:33bluebrotherhmm. Possible.
11:50:37GodEater_pondlife: There's no need to go quite so West End this time since we've no foreign types attending - so we could go somewhere else if people prefer ?
11:50:53linuxstbscorche: I would prefer the left menu to be constant on all Rockbox sites. But we could possibly also add themes links to the other sites...
11:51:26pondlifeGodEater: Pick a decent pub then? I'll be coming in via Paddington.
11:52:00pondlifeMaybe you should also put a post up on the users mailing list. Or I can...
11:52:04GodEater_ah - if you're coming in from that far west then perhaps we *should* stick to the same place
11:52:04scorchelinuxstb: things under other licenses go int he .txt file as per the guidelines page
11:52:08 Join Arathis [0] (
11:52:43linuxstbGodEater: I'm happy to stick with the centre, but don't mind...
11:53:23scorchelinuxstb: i would rather not increase the sidebar on all other is long enough
11:53:57scorcheas well, i would rather not have a vertical *and* horizontal navbar for the themes site...
11:54:05linuxstbscorche: I just prefer forms than free text - it gives us more control over the data we're collecting, and means more automatic validation.
11:55:05GodEater_pondlife, linuxstb: lets stick to the same pub then - seems easiest.
11:55:23GodEater_if someone thinks it should get posted on the mailing list then please do - I'm not subscribed personally.
11:55:24pondlifeCan you post the details again in the forum thread?
11:55:36pondlifeI'll pop an invite on the ML
11:55:47linuxstbWas it agreed to be the 24th? I can't make the 25th.
11:56:50GodEater_I can't make the 25th either now - got a funeral to go to :(
11:57:02GodEater_details on the forum post now
11:57:05markunlinuxstb, GodEater_: was just looking at the forum thread, nice to see your faces :)
11:57:38GodEater_the pic didn't crack your monitor then ? ;)
11:57:39scorchemarkun: if they could only see yours...(althought i saw the pic with no beard...)
11:57:51GodEater_scorche: I've seen that pic too - and the one before
11:58:11scorcheGodEater_: hush! are ruining my joke!
11:58:14*pondlife hopes nobody brings a camera next time...
11:58:33GodEater_pondlife: no - we're bringing our H140's to get another recording of you singing....
11:58:35scorchelinuxstb: well, the problem with the forms and that .txt file is that that .txt file is inside the .zip...
11:58:49pondlifeGodEater: lol, you never heard the last one then!
11:58:57GodEater_pondlife: hehe - nope ;)
11:59:11pondlifeMight take a lot of beer.
11:59:18GodEater_we're in the right venue then
11:59:41pondlifeHmm, and if LinusN happens to be in London (business maybe?) then I'll buy him drinks all night for fixing my H340.
11:59:45GodEater_I left my Rockbox poster at the pub last time - will have to print another
11:59:57pondlifeMaybe they'll have it on the wall?
12:00:04pondlife"Officially approved by Rockbox"
12:00:12markunWhen I met up with Moos for a coffee in Paris I forgot to take a picture
12:00:19GodEater_pondlife: doubt it - it's not that sort of pub ;)
12:00:25pondlifeGood :)
12:00:28*scorche wishes he had devs that lived close around
12:00:30 Part maffe
12:00:32pondlifemarkun: DevConCafe ?
12:00:37 Join maffe [0] (
12:00:43linuxstbscorche: The .txt file could be created from the information entered on the form and added to the .zip.
12:00:49pondlifescorche: Come over to London for a pint!!
12:01:02markunamiconn: shall we do a DevConBiergarten when I come to Berlin? :)
12:01:06scorchepondlife: pay for the plane ticket, i pay for the pint? =)
12:01:07*bluebrother is sad he can't make it to DevConPub :'(
12:01:29pondlifescorche: Haha, almost got me.
12:02:08scorchelinuxstb: well, i think that might be a feature for 2.0
12:02:24GodEater_scorche: Febs made it - you're just being a lightweight
12:02:52scorcheGodEater_: you havent seen my wallet =)
12:03:03*scorche isnt a lawyer
12:03:04pondlifeAh, moths?
12:03:08linuxstbscorche: It's just that I think the most important thing on the site (which can't easily be changed in the future) in the data we're collecting from users. IMO that needs to be right from the very beginning, everything else can be changed.
12:03:16GodEater_I've seen those moths
12:04:31 Join austriancoder [0] (n=austrian@
12:05:18austriancoderpetur: hi... have you time for some debuging today in the evening?
12:05:47GodEater_scorche: I told Febs he was an Ambulance Chaser :)
12:06:10GodEater_markun: how is TTS coming along ?
12:06:10scorchelinuxstb: well, yes i would like to have that (it would make many things easier for all parties involved), but i have no clue how to add a file to a zip in php, would be reading a lot of the manual, and (the bad part) delay the site debuting for at least another week
12:06:17scorchebut it is something that would be good to do..
12:06:22markunaustriancoder: the Gigabeat S is also in need of you USB stack
12:06:32scorcheGodEater_: i wasnt aware what type of lawyer he was
12:06:42GodEater_scorche: he said "litigation"
12:06:47markunGodEater_: his plugin is not working, but I believe he's working hard now
12:06:53scorcheargho...those damn lawyers ;)
12:06:59GodEater_and I didn't give him time to say "I defend against it"
12:07:10austriancodermarkun: as long as there are some infos like datasheet its not a hard task to support it
12:07:17*pondlife needs a lawyer...probably
12:07:35markunaustriancoder: the i.mx31 (in the Gigabeat S) has been used for the portalplayer USB code anyway, right?
12:07:59austriancodermarkun: ahh. an other i.mx31 :)
12:08:10 Join Rondom [0] (
12:08:12markunyes, so even less work
12:08:16austriancodermarkun: yep.. pp also using this chip
12:08:32markunwell, I don't believe they use the chip, but part of the logic
12:08:44safetydanmidkay, if you're up to it, here's a test build of unknown workingness
12:08:55austriancodermarkun: yep
12:08:58markunwould be crazy to add a 533MHz ARM11 just for USB support :)
12:09:13peturaustriancoder: this week and next week are bad... only tomorrow is free atm, and next week I'm afk
12:09:18GodEater_markun: I'm all for that sort of crazy ;)
12:09:36peturaustriancoder: I did have time the past days but you weren't here
12:09:40pondlifeGodEater: Did you invite stripwax? I never seem to be on IRC at the same time.
12:10:48austriancoderpetur: no problem.. maybe daniel can play men in the middle for a mail conversation with MrM
12:11:01GodEater_pondlife: it's an open invitation, but I've no specifically said to him "come along"
12:11:15GodEater_he was supposed to come last time but felt too ill apparently
12:11:39pondlifeJust wondered, he did express interest before.
12:11:56*pondlife seems to have moved to Kent according to Rasher's map...
12:12:03 Join Jens [0] (i=Jens@pdpc/supporter/active/Jens)
12:12:36peturaustriancoder: I'm sorry for this but real life decided otherwise, I'll give it my focus again starting august 1
12:13:23midkaysafetydan: close, but not quite. i'm not sure if you were aware, but before this patch (i.e. in svn builds), since forever, at least on the 5Gs, bass and treble haven't worked right - instead of adding bass or treble, they'll increase the volume.
12:13:56midkaysafetydan: with the build you provided, they properly adjust bass and treble, but also still the volume, so turning up the bass will result in more bass and more output. same with treble.
12:14:13peturaustriancoder: did you manage to compare with the code of MrH? I would think he manages to transmit?
12:14:14midkayyou've just got to fix that volume-related bug and it should be fine.
12:14:34linuxstbscorche: Assuming the "zip" binary is executable from PHP scripts on RedBreva's server, then you would just write the text file somewhere, then do somthing like chdir("/path/to/textfile/") ; system("/bin/zip /path/to/ textfile.txt")
12:14:46austriancoderpetur: I have looked at it... but I dont can find the cause why mine is not working
12:15:00safetydanmidkay, I wasn't aware of the volume related issue :) I was just aiming to replace the bass/treble with one that worked.
12:15:06midkaysafetydan: and it's hard to tell if the cutoffs are working properly or not. it'd be easier to test once you fix that bug.
12:15:08safetydanmidkay, did you try the cutoff settings?
12:15:23safetydanmidkay, ah okay. I'll see what can be done about that.
12:15:54midkaysafetydan: at least you have partial success. :) that bug needs fixing before the patch will work as intended. it *seems* that the cutoffs aren't working, but i can't be entirely sure.
12:16:00 Join miepchen^schlaf [0] (
12:16:02midkayi can't seem hear a difference, though.
12:16:23linuxstbWere they definitely working with the hw eq interface?
12:16:36midkaylinuxstb: yes, the high and low shelf filter cutoffs worked for sure.
12:16:36*linuxstb seems to recall amiconn did some tests
12:16:38peturaustriancoder: maybe something in the init still not ok? The strange thing is you do not get an error (I also fail to transmit with my isp1362, but at least it is giving me back an error)
12:16:40safetydanlinuxstb, I think either preglow or amiconn, did tests
12:16:47midkaylinuxstb: not the three peak filters, those never worked at all.
12:17:15scorchelinuxstb: hrm...i wonder what version of php this server has...
12:17:36linuxstbv5.24 I would guess from the URL
12:17:38austriancoderpetur: I have checked it... when the first setup packet comes... endpoint 0 has rx and tx enabled...
12:17:50austriancoderI have checked that everything is in IRAM
12:17:56midkaysafetydan: well, if you manage to hunt down that bug (or if you think you figure it out) i'll be around for a while to test it out :)
12:18:11safetydanmidkay, cheers
12:18:12 Quit miepchen^schlaf (Client Quit)
12:18:13austriancoderI have checked the generated transfer descriptor
12:18:59peturaustriancoder: did you also check the TX/RX status flags (can't remember where they were)
12:19:00linuxstbscorche: I guess you meant RedBreva's server... According to the headers it's 4.4.7
12:19:24austriancoderpetur: so i know that the transfer is active, transmits 18 byte and IOC is set to 1 (Interrupt on Complete)
12:19:31scorcheso i would have to do it your way...
12:19:36linuxstbBut that's not really an issue - it's nicer not to rely on needing the latest versions of things.
12:20:02austriancoderpetur: logf("ep0 s 0x%x", UDC_ENDPTCTRL0); /* 0x800080 -> 100000000000000010000000 rx enabled tx enabled */
12:20:11scorchewell, using $zip->addFile("testtext.txt"); would have been nice
12:21:34linuxstbLet's just hope the server has zip (and unzip) installed. But I assume it's possible for us to upload our own binaries to the server?
12:21:47linuxstb(hopefully we can add check_wps soon)
12:22:18austriancoderpetur: I disabled handing of the setup packet and set the first 2 endpoints to transmit their data. and I only get a buffer error on endpoint 1
12:22:51austriancoderpetur: But it should give 2 buffer errors, as I dont setup an transfer descriptor.
12:22:55scorchelinuxstb: not sure actually...i do know that scripts are (of course) provided
12:23:28scorchei could call them and find out, i suppose
12:25:22 Quit kfazz ("Leaving")
12:26:03scorchelinuxstb: from SSH at least, i can use zip/unzip
12:26:31 Quit maffe (Remote closed the connection)
12:26:57peturaustriancoder: sorry - work interrupt :)
12:28:29austriancoderpetur: after I eat something I will copy&paste MrH code to transmit and check if it works
12:28:29safetydanmidkay, from the little information we have it looks like there's no way around the volume boost without scaling the volume when you change bass/treble settings
12:29:07peturaustriancoder: hmmm out of ideas right now... maybe MrH or somebody else could have a peek at your code? I'll try to be online tonight but it will be quite late
12:30:05peturaustriancoder: if you changed your code more put the latest version online maybe
12:30:29 Join ThreeM [0] (i=3e60f4e3@gateway/web/cgi-irc/
12:30:37*petur goes eating as well
12:30:48ThreeMhi there
12:31:23 Join Wiwie [0] (
12:33:02midkaysafetydan: why not? before your patch, the hardware eq worked fine, no volume boost, and the bass setting was only volume boost. now that they're combined, you get the volume boost and the bass/treble boost. seems like you just need to gut whatever code was used for the bass setting before the patch and replace it entirely with the hardware eq code.
12:33:54ThreeMi have a question about the Battery Runtime on iPod 4gen Greyscale. Is Rockbox able to play longer than ~4 Hours now? i see nothing in the todo list about it, so i thougt its done and rockbox is playing playing like the stock firmware?
12:34:18linuxstbWhich todo list? It's definitely still an issue.
12:34:43linuxstbThis is the ipod-specific to-do list -
12:35:49ThreeMhh key :) best view with open eyes :) Sorry for Disturbing you
12:36:15safetydanmidkay, actually you're right. There's a volume prescaler being applied when bass/treble is set in sound.c
12:37:07midkaysafetydan: that must be it. after all, if we can work the hardware eq on its own in SVN, no reason we can't simply rename it to "bass" and "treble" and have it work the same way. :)
12:37:27 Quit ThreeM ("CGI:IRC")
12:40:47scorchelinuxstb: would doing things this way allow us to put the file in a place other than the / of the zip?
12:40:52safetydanmidkay, try grabbing another build from
12:40:56 Join Nico_P [0] (
12:41:51linuxstbscorche: Yes. You would need to create the appropriate directory hierarchy, and do something like system("/bin/zip /path/ subdir/textfile.txt")
12:43:16amiconnlinuxstb: The low shelf and high shelf cutoff settings did definitely work. I did some test recordings of a sine sweep, fed them through fft and made plots
12:43:37 Quit Rondom (No route to host)
12:43:49linuxstbamiconn: Thanks. I thought you did that..
12:44:30midkaysafetydan: aah, volume issue fixed. :) it just now seems to me that the cutoffs don't work at all. i could definitely recognize the difference with the hardware EQ, i used to use and tweak it all the time. can't hear any difference now.
12:45:02scorchelinuxstb: and that hierarchy would be subdir in your example?....and this would have to be created with each submission?
12:45:04safetydanmidkay, okay, I probably stuffed something up there. Good to know the rest works though.
12:45:38midkaysafetydan: cool, if you figure out what might be wrong, i want another build :D
12:45:44linuxstbscorche: Is the subdir name going to be unique to the theme?
12:45:45safetydanmidkay, will do :)
12:45:51linuxstb(or subdirs)
12:45:56scorchelinuxstb: no...
12:46:24linuxstbThen you can just create the path once and create all the textfiles in there directly.
12:50:52 Nick Tr1ckY^Gone is now known as Tr1ckY (
12:55:20safetydanmidkay, new build, same location
12:57:08 Quit PaulJam (".")
12:57:25amiconnlinuxstb: It's still online:
13:01:26midkaysafetydan: still doesn't seem to work. :(
13:01:48scorchelinuxstb: alright...well, i will look at some things and likely get back to you tomorrow ish with some questions (if i have them), but for now the site debut is delayed for a bit while this gets done
13:01:57scorchethanks for your input =)
13:04:22safetydanmidkay, darn. Well thanks for testing, but I have to give up for tonight. I'll upload a new version of the patch to flyspray.
13:05:22midkaysafetydan: alright, no problem. oh, and also: my bass settings, while retained, were not applied on startup. i had to change them before they came into effect.
13:05:31***Saving seen data "./dancer.seen"
13:05:50midkayi bet the code somewhere checks at startup if the eq is on, and if so, applies the setting? you'll probably just need to readjust that to always apply the bass/treble settings at startup.
13:06:41 Part Llorean
13:10:09 Join Crash91 [0] (i=510a14e3@gateway/web/cgi-irc/
13:10:53Crash91anyone know when we will have usb support on sansa?
13:11:07scorchewe dont give timelines =)
13:11:14amiconnmidkay: Imo it should be handled exactly like ordinary bass/treble. Then there shouldn't be a problem
13:11:35Crash91=) ive realised that, but give me an estiamte
13:12:09midkayamiconn: you'll have to convince safetydan. although i do not want to lose the ability to adjust cutoff. as i recall, you said this should be doable on most/all other targets? if so, it should be introduced for them all, i think.
13:12:22Crash91oh nvm, im off
13:12:24 Quit Crash91 (Client Quit)
13:12:27amiconnmidkay: On some other targets, yes
13:13:17midkayamiconn: then i think it should be introduced for those who have the capability.
13:14:39 Quit merbanan (Read error: 110 (Connection timed out))
13:17:22 Join merbanan [0] (n=banan@
13:25:09safetydanmidkay, I can't see why it's not applying the settings at startup. It should be done by the call to sound_settings_apply in audio_init.
13:26:41midkaysafetydan: hmm. didn't seem to for me.
13:27:30safetydanmidkay, well I'll take another look later. For now sleep.
13:27:31safetydannight all
13:27:32 Quit safetydan ("Ex-Chat")
13:29:42GodEater_will an ipod continue to charge after being "ejected" if you don't disconnect the cable ?
13:30:19midkayGodEater: yes, it should.
13:31:12 Join spiorf_ [0] (
13:32:04GodEater_midkay: thanks
13:32:52pondlifepixelma: Any idea when the voice-results-in-CPU-boost bug was discussed? I saw your forum post, but haven't seen this bug on my H340 despite using voice most of the time...
13:33:40pixelmahmm... amiconn and jhMikeS discussed that last week - let me see
13:34:23pixelmaand I experience it too on my M5
13:35:00pondlifeDuring playback only?
13:35:14pondlifeOr continuing after stop too?
13:35:21pixelmaduring playback
13:36:41pondlifeHmm, did I dream it or is there a way to see what threads have triggered CPU boosts..?
13:37:29pondlifeI did have a nasty tracker in there a while back, but it used a bit of memory and we decided it wasn't useful enough.
13:38:44JdGordonthere is a boost logger
13:38:51JdGordonyou need to manually enable it though
13:38:59pondlifelogf only, or is there more to it?
13:39:49 Quit spiorf (Read error: 110 (Connection timed out))
13:42:01linuxstbmerbanan: I've noticed some recent work on cook.c in ffmpeg. I'm assuming these changes are breaking the fixed-point version - do you know if anyone is maintaining those patches? (I'm seriously thinking about porting the cook decoder to Rockbox).
13:42:10pixelmapondlife: July 13th, 19:20 and following
13:42:33pondlifepixelma: Thanks
13:42:40 Join ganny [0] (i=9088689c@gateway/web/cgi-irc/
13:42:49GodEater_I spot a thread that needs locking in the forums before it gets out of hand...
13:43:32GodEater_that DJ one - I see you're viewing it too
13:43:40GodEater_apparently we all have our heads up our asses
13:44:16scorcheoh...i thought you meant thoat dastardly london social thread
13:44:24scorcheme steals the o
13:44:57scorches/thoat/that =P
13:45:58pondlifeSomeone forgot their Ritalin...?
13:46:44*GodEater_ googles Ritalin
13:46:51pondlifeI probably spelt it wrong
13:47:03scorcheGodEater_: damn..paul is fast
13:47:24GodEater_nope you didn't
13:47:29GodEater_ADHD treatment
13:47:42GodEater_he's pretty agressive
13:47:59pondlifeThat's the one. As a side-effect it allows us to take our heads out of our asses.
13:48:21GodEater_I hope it doesn't mean we stop soundlike a bag of shit though
13:48:23GodEater_that would be terrible
13:48:32GodEater_*sounding like
13:48:35pondlifeNo, that can never happen.
13:49:05scorche"well all i'm gona say is that good luck in spreading this "open" source firmware with the amazing marketing and propoganda machine that is the frankly dodgy UI of this website, dictatorship forums and unhelpful and superior mods with their head up there own backsides."....i must say...he was really reaching with that one
13:49:28*pondlife loves the smell of "propoganda" in the morning.
13:49:36linuxstbmerbanan: Forget my question - I've now actually paid more attention to the content of the patches...
13:53:21pondlifeGodEater: p.s. Please don't risk doping my drinks. I've got about zero alcohol tolerance, so you can save your money... :)
13:53:36*amiconn wonders if anyone is using realaudio for serious things
13:53:54scorchesome use it for streaming...
13:53:59amiconnRA truly sounds like our famous BoS...
13:54:23amiconnI'd rather stream mp3 (ok, it's not made for streaming) or even wma
13:54:29*petur abandonned it years ago
13:54:42pondlifeI've heard people defending RA, saying it's much improved nowadays, but have still not heard a good one.
13:54:49GodEater_pondlife: awwwww - I really wanted to hear the singing....
13:55:23pondlifeGodEater: Take that to #rockbox-community, and no-one will hear you or me...
13:55:37linuxstbI'm interested because the BBC stream a lot of things with RA. Although nowadays most things seem to be available in wma as well.
13:57:11peturBelgian radio pisses me off, they stream mp3, but they use two streams and the audio alternates bewteen them. If you capture one you get choppy sound :(
13:57:49JdGordonamiconn: my stupid uni uses RA for the online lectures
13:57:57markunpetur: studio brussel?
13:58:30peturin fact, all VRT stations
13:59:29dionoeapetur: i'd be interested if you could send me a link to such a stream (just to see how it's handled)
13:59:38 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
14:00:14JdGordonwoot, uni is using mp3 this semetser :)
14:00:56linuxstbpetur: If it's listenable, it must be capturable...
14:01:03pondlifeIs there a free RA encoder around?
14:01:36linuxstbI think the realproducer tools are free (as in no cost). I don't think anything open source exists.
14:01:47pondlifeHmm, free, but not free.
14:01:57peturdionoea: this is their player: - I don't have theURL's of the streams around...
14:01:57GodEater_gratis but not libre
14:02:00linuxstbfree but not Free.
14:02:29Kasperlewhat about the "helix server"?
14:02:38Kasperleisn't that open?
14:02:49pondlife"For the Real Audio decode part, there is no source code,"
14:02:52linuxstbThe helix project seems a legal mess to me...
14:02:55pondlifeNot sure about the encoder
14:03:21pondlifeI was mainly wondering what encoding options/parameters RA used.
14:03:24Kasperleoh, bleh
14:03:40linuxstbpondlife: There's also about 10 different codecs to choose from...
14:04:11linuxstb"cook" is a common one, and that's the one that exists in ffmpeg.
14:04:40 Join desowin [0] (
14:04:58 Join Llorean [0] (
14:09:31 Join datachild [0] (
14:09:44dionoeapetur: ok, thanks. I'll have a look when i get back from work.
14:09:48GodEater_linuxstb: you've read up on this a lot for someone who didn't want to do it on his own ;)
14:14:56pondlifeHmm, is it just me or is the debug build failing at the moment..? "/home/Steve/rockbox/apps/main.c:348: undefined reference to `debug_init'"
14:15:06 Join maffe [0] (
14:15:06 Join Wiwie [0] (
14:22:01amiconnpondlife: For what target?
14:22:14amiconnDebug build only work for archos targets and ifp
14:22:27pondlifeHmm, they used to work for H300
14:22:30amiconnAnd for sims, where debug is always enabled
14:22:42amiconnAll other targets only have logf
14:22:51amiconnIt never worked for anything coldfire
14:23:09pondlifeMaybe I've only ever used it with the sim...
14:23:12amiconnAt least not as it should
14:23:24pondlifeNevermind, I'll use logf
14:23:47 Join Lear [0] (i=chatzill@rockbox/developer/lear)
14:24:55 Quit datachild ("- nbs-irc 2.35 - -")
14:27:44 Join datachild [0] (
14:28:07 Quit Farp (Read error: 110 (Connection timed out))
14:29:20JdGordon1/3 threads done
14:30:01linuxstbpetur: That VRT website just crashes firefox for me (but then, anything which uses sound tends to crash my firefox...)
14:30:29peturheh... firefox is happy with it here
14:31:54*linuxstb wishes websites would just stick to html...
14:32:41*petur had to capture the network traffic to find the stream URLs
14:33:13linuxstbAt least the BBC generally have "play in standalone player" links to .asx or .ram files.
14:33:21 Quit idnar (Nick collision from services.)
14:33:24 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
14:39:02markunlinuxstb: too bad they stopped the ogg vorbis experiment
14:41:53linuxstbSeems the current live streams are 48kbps WMA, 44.1kbps RM or 10kbps (!!) WMA. Archived audio can be found in better quality though.
14:41:59linuxstbWhich is a bit sad...
14:45:34 Quit ganny ("CGI:IRC (EOF)")
14:47:59GodEater_they're really pushing the boat out there
14:49:11 Join aliask [0] (
14:50:19 Quit elinenbe (Read error: 110 (Connection timed out))
14:52:04Jpossible suggestion; after a delay (configurable) without user input change to (configurable) screen. I'm thinking where the user has accidently pressed a button which, for example, took them out of the currently playing screen, then after a while rockbox could switch back to it...? Thoughts?
14:52:25bluebrotherJ: please not.
14:52:37Jbluebrother: reasons?
14:52:45bluebrotherauto-switching screens have been discussed several times ... most don't want them.
14:53:43bluebrotherit would also switch if you switch screens for a purpose, and if you (f.e.) switch to the file browser without view and it switches back to wps automatically you won't notice that.
14:53:57scorcheit would also cause havoc for our blind population
14:54:09Jbad suggestion - sorry.
14:54:09bluebrotherscorche: beat me ;-) Just wanted to notice that.
14:54:58bluebrotherwell, it was a suggestion. So no biggie ;-)
14:55:34bluebrotherat least I prefer such suggestions over rarely used features. Like using the Ipods scroll wheel for scratching like things
14:56:51Jthe two things I like in rockbox most are 1) iPod accessories (I know - lot of work) and 2) not always switching the backlight on for particular operations (eg, volume up/down)
14:57:00 Quit SirFunk (Connection timed out)
14:57:17Jbut I can't get to grips with the codebase and have little time to commit so no complaints here.
14:58:40bluebrotherI bought a remote for the Ipod recently to look into it but it seems my Ipod doesn't like it.
14:59:26amiconnWell, you don't necessarily need to have a remote in order to make the remote work
14:59:57amiconnIt's standard rs232 protocol; the problem is figuring out how to use the PP rsr232 interface
15:00:12bluebrotherright. But I expected it to work using AppleOS so I could more easily sniff the connection
15:00:58bluebrotherbut I haven't checked further. Maybe something with the remote is wrong.
15:01:09amiconnSniffing the connection isn't necessary. The ipl people already documented the accessory protocol
15:01:53*amiconn thinks the best development setup would be to connect a pc serial port (with a level converter of course)
15:02:23bluebrotherthat's what I intended to do (after AppleOS recognizes the remote)
15:03:00bluebrothermaybe I should try without bothering about the remote ...
15:03:58 Join Febs [0] (n=chatzill@
15:05:21 Join elinenbe [0] (n=elinenbe@
15:05:32***Saving seen data "./dancer.seen"
15:05:59aliaskHow can I ensure a function is going to be placed at a location which is word aligned?
15:06:21linuxstbI would expect that to be done automatically.
15:06:30aliaskOk, that'll do.
15:07:11aliaskHrm, now to figure out why the ISR never gets called.
15:07:36*GodEater_ likes that fact that "automatically" is merely acceptable to aliask
15:08:05linuxstbbluebrother: Yes, I would just try connecting an ipod to a PC serial port. Connecting to actual accessories may require further initialisation/work.
15:08:22linuxstbbluebrother: You could write a gdb stub whilst you're there... ;)
15:08:37aliaskShould I think it's *super cool*?
15:10:07aliaskWill casting a function to an unsigned long give the address of the function? Or should I be casting to a pointer?
15:12:36markunaliask: the function name can be used as a variable and will be a pointer to the function iirc
15:13:27aliaskPerfect :)
15:13:32bluebrotherlinuxstb: I think I'll build up the hardware first. I already have the needed chip for level conversion ... just need a soldering iron to do it now ;-)
15:13:55markunaliask: see apps/plugin.c for example
15:15:52 Join Thundercloud [0] (
15:19:35 Quit Nick_Brackley ("CGI:IRC (EOF)")
15:21:50 Quit aliask ("ChatZilla [Firefox]")
15:21:55 Join Guile`` [0] (
15:29:43 Join Xerion [0] (
15:36:03austriancoderpetur: I have tryed it now MrH way... but it dosnt work *grrrrr*
15:36:18austriancoderpetur: I will dontact daniel, that we can ask MrH for help
15:36:59peturI'll try to have a look at it too - tonight *late*
15:39:36 Part maffe
15:40:57amiconnaustriancoder: wokring cannot work of course ;)
15:41:38austriancoderamiconn: typo fixed
15:42:30austriancoderamiconn: do you have time to debug with me or... you only need to look at the code of mrh and mines and maybe you can figure out whats wrong....
15:43:42 Quit hcs (Read error: 110 (Connection timed out))
15:44:06 Quit atsea- (Remote closed the connection)
15:45:07 Join The-Compiler [0] (
15:49:24 Join atsea- [0] (i=ariel@gateway/tor/x-ac28711477c0e9b1)
15:50:27 Quit Wiwie (Read error: 110 (Connection timed out))
15:52:00 Quit printfXh4 (Read error: 110 (Connection timed out))
15:52:34 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
16:00:19oboThe latest version of binutils in debian unstable fixes the crash when producing sim map files... but the broken version has made it's way into testing
16:01:05 Join SirFunk [0] (
16:01:49bluebrotherobo: can you give me a hint with qt designer 4?
16:01:58oboI can try :)
16:02:23bluebrotherI want to insert a bitmap but I have no idea what element I need to insert.
16:02:39oboI think you can do it with a label...
16:04:01bluebrothercool, that did it. Thanks!
16:06:43 Quit Lear ("ChatZilla [Firefox]")
16:09:33 Quit amiconn (Read error: 110 (Connection timed out))
16:09:34 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
16:15:30 Join n1s [0] (
16:16:29 Join lazka [0] (
16:16:40 Join Wiwie [0] (
16:24:08 Join TopScorer [0] (
16:24:08 Quit datachild (Connection timed out)
16:36:35 Join Rob2222 [0] (
16:36:42 Part TopScorer
16:36:51 Join XavierGr [0] (
16:37:14 Quit Seed (Nick collision from services.)
16:37:20 Join Seedy [0] (i=ben@
16:38:00 Join jnakinjgi [0] (i=0@
16:42:23linuxstbn1s: Regarding your recent change to rolo.c, I think you forgot the gigabeat (although it doesn't have Rolo yet). ICODE_ATTR is defined to be null (it only has about 4KB of IRAM), so it will explicitly need section(".icode") for rolo_restart(). But then that will give a problem because STATICIRAM is defined to be "static" for the gigabeat...
16:43:37 Join toffe82 [0] (
16:44:52n1slinuxstb: aha, so should I just revert and add a comment, or what do you think?
16:47:09linuxstbMaybe... It could cause problems if whoever comes to implement rolo for the gigabeat doesn't notice.
16:47:47n1slinuxstb: ok, will do it.
16:51:42 Quit Entasis ("Leaving")
16:53:31 Quit Rob222241 (Read error: 113 (No route to host))
16:56:42n1slinuxstb: ok, fixed, btw STATICIRAM is defined the same for all arm targets, but it doesn't matter much here, might as be non-static to be a little more consistent with the SH version of the function.
17:00:39linuxstbIIUC, STATICIRAM is "static" for the gigabeat, and null for other ARMs (CPU_PP and PNX0101). The #ifs are nested in config.h...
17:01:18 Join PaulJam [0] (
17:01:32linuxstb(Gigabeat's CONFIG_CPU is S3C2440)
17:01:44amiconnIt should also be 'staic' for the iFP
17:02:02linuxstbIs the iFP PNX0101?
17:02:39n1slinuxstb: yes but CPU_ARM is defined for all arm targets and the STATICIRAM thing is defined based on if CPU_ARM is defined IIUC
17:02:51linuxstbThen it looks like it's defined the same as CPU_PP
17:03:18linuxstbn1s: Yes, but that #ifdef CPU_ARM is inside another #if which excludes the gigabeat.
17:03:23amiconnThe problem is that gcc implies 'short call' for static functions
17:03:55amiconn...but iram functions must be long call, except in iFP where it's possible to map both memory regions close enough for short calls
17:04:13linuxstbSo basically we can't make icode functions static on ARM?
17:04:21n1slinuxstb: ah, yes I see what you say now, it's not very reader friendly :-)
17:04:33amiconnCorrect, unless they're _only_ called from other iram functions
17:05:00linuxstbWell, at least it's better than the previous situation.
17:05:04amiconnAnother arm-*-gcc flaw
17:05:34***Saving seen data "./dancer.seen"
17:07:00linuxstbn1s: I agree. But it seems no-one likes to indent #defines like that.
17:08:43*amiconn hates indented preprocessor statements
17:09:28linuxstbI generally do, but think config.h is a special case - there's no C code...
17:09:38pondlifeWasn't the # meant to be in the first column for C anyway?
17:10:15pondlifeI always thought that indented preprocessor commands were a C++ (or later C) extension.
17:15:29amiconnYou can indent by putting spaces between # and the keyword
17:23:31 Join maffe [0] (
17:26:52 Quit JdGordon (Remote closed the connection)
17:28:24 Join bdgraue [0] (
17:32:56Nico_Pamiconn: I saw in today's log that you were wondering about metadata blocking the middle of the file buffer...
17:33:02Nico_Pthat won't be a problem at all
17:34:37 Join bdgraue_ [0] (
17:35:25n1sNico_P: does your buffer rewrite handle the situation when you have buffered part of a file and when you try to buffer the rest the file is gone?
17:35:33 Join datachild [0] (
17:35:53*linuxstb waits for Nico_P to explain why metadata won't block the middle of the buffer
17:35:55Nico_Pn1s: no... never thought of that case actually :)
17:36:09Nico_Pn1s: but shouldn't be too much of a problem
17:36:32n1sNico_P: sounds good :-)
17:36:48Nico_Plinuxstb: even if there is data in the middle of the buffer, a new long file will just wrap around it and will still be able to be used as a ring buffer
17:37:07Nico_Pactually that's how it worked in my test case
17:37:19pondlifeWith multiple guard buffers?
17:37:25Nico_Ppondlife: no, only one
17:37:31linuxstbYou mean it will only use the first part of the buffer?
17:37:41Nico_Plinuxstb: no, it uses the whole buffer
17:37:59LloreanNico_P: If a file only gets say, 20% of the buffer initially, then the remainder of the buffer is consumed, can it expand to fill the buffer before it starts doing the within-a-ring ring buffering?
17:38:16 Quit Wiwie (Read error: 104 (Connection reset by peer))
17:38:18pondlifeOK, so say you have FILEPART1|metadata|FILEPART2| will it feed the codec the last part of FILEPART1 ?
17:38:18 Quit The-Compiler (Connection timed out)
17:38:19 Quit Guile`` ("KVIrc 3.2.0 'Realia'")
17:38:40Nico_PLlorean: not currently, but it should indeed
17:38:40pondlifewith gapless playback of FILEPART1/FILEPART2
17:38:45 Join kretender [0] (
17:39:39Nico_Ppondlife: bufgetdata ensures that it provides a whole piece to the codec by copying a whole block to the guard buffer if necessary
17:39:45Nico_Pnot sure it works fine yet though
17:40:17Nico_Pand actually your example coudn't happen
17:40:33Nico_Pit would be FILEPART2|meta|FILEPART1|guard|
17:40:56linuxstbBut it could be FILEPART2|meta|FILEPART3|guard| ?
17:41:12pondlifeAh, ok - I was thinking that the buffering might have to fill and skip around metadata.
17:42:08Nico_Plinuxstb: I don't see how it could happen, but I'm not sure I get what you mean
17:43:19linuxstbFILEPART3 is the part of the file that comes after FILEPART2... FILEPART1 is the start of the file.
17:43:33linuxstbImagine a very large file.
17:43:48pondlifeOr a small buffer
17:43:56Nico_Pok I see... maybe an example will make things clearer
17:44:16linuxstbI thought that was an example? ;)
17:44:23Nico_Pa complete one :)
17:44:50Nico_P1 would be |.....|meta|FILEPART1|guard|
17:45:22Nico_Pthen we go on buffering : |FILEPART2|meta|FILEPART1|guard|
17:45:42GodEater_what's "....." represent ? empty buffer ?
17:45:50Nico_Pand then it would be |FILEPART2|meta|FILEPART3|guard| so yes it would happen
17:45:58Nico_PGodEater_: yes or old data
17:46:30pondlifeBut you will copy the last part of FILEPART2 and the first part of FILEPART3 to guard before passing it to the codec.
17:46:50pondlifeOtherwise pass a pointer to the data direct?
17:47:01Nico_Pcopy to the guard buffer before passing
17:47:22linuxstbThat could mean lots of copying. A guard buffer at the end of FILEPART2 might be useful.
17:47:34Nico_Pthe buffer for the FILEPARTs is a ring buffer starting right after the metadata and ending right be
17:48:06Nico_Plinuxstb: why ? to me it would be ony once every time there is a wrap in the reading
17:48:11linuxstbe.g. some codecs don't know how much data they need, so request 32KB, consume a little bit, (e.g. a few hundred bytes), advance the pointer, then request another 32KB....
17:51:18linuxstbHave you considered the option of moving the metadata within the buffer to keep it right behind the current track?
17:51:40linuxstb(I'm not sure if that would cause more issues than it would solve though...)
17:52:09Nico_PI have, but as you say I think it's too complicated to be a good solution
17:52:24Nico_Pand it would involve a lot of moving
17:52:50amiconnAre there plans how to handle non-streaming formats that must be contiguous in memory?
17:52:58 Quit bdgraue (No route to host)
17:53:30Nico_Pamiconn: yes, but those, will obviously need to be smaller than the buffer
17:53:52amiconnof course
17:53:59 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
17:54:09Nico_PI was thinking of adding a boolean to bufopen indicating whether the file can wrap
17:54:16Nico_Pthat should be enough
17:55:13amiconnWhat if the non-wrappable file doesn't fit when buffering is called first, but would fit once preceding data is freed?
17:55:44amiconnThis scenario does require an extra spinup of course - but will it work?
17:56:46Nico_Pamiconn: it could...
17:59:01Nico_Pn1s: actually I think the case you suggested when the file disappears before being buffered completely should already be handled fine :)
17:59:17Nico_Por almost at least
17:59:30 Join The-Compiler [0] (
18:02:40 Quit Arathis ("Bye, bye")
18:04:08Nico_Pso no major objections before I get in too deep with the current idea ?
18:05:01pondlifeNot here. You can work on the basics before adding a non-wrappable/contiguous flag though...
18:05:09pondlifeJust keep the code well commented
18:05:16Nico_Pthat's the plan
18:05:25Nico_PI'll try :)
18:05:31pondlifeThis is fraught with complications, as you now know!
18:06:12amiconnAh, yes, disappearing files should be handled too
18:06:25pondlifeDisappearing? Like deleted?
18:07:19Nico_Ppondlife: yes
18:07:34Nico_Pthere is already at least partial support for that case
18:07:36linuxstbNico_P: What's the "ringbuffer within a ringbuffer" concept I've heard mentioned?
18:08:06Nico_Pit's the fact that a long file's data buffer will actually be a ring buffer
18:08:30Nico_Pwhich is itself stored in the main ring buffer
18:09:00linuxstbOK, so nothing different to what happens now?
18:09:15Nico_Pnot sure...
18:09:35Lloreanlinuxstb: It's to keep the on-buffer metadata from being overwritten.
18:09:58amiconnYes, and for that it even needs to be a split ring....
18:10:04linuxstbBut didn't we just discuss long files - they're split into two parts in the buffer...
18:10:18Nico_Plinuxstb: actually they're not split
18:10:39linuxstbSo what did "FILEPART2|meta|FILEPART3|guard" represent?
18:10:50*pondlife is confused too
18:10:53Nico_Pwell the buffer is split because it might wrap around the end of the main buffer
18:11:39 Join gotthardt [0] (
18:12:01Nico_Pin that case FILEPART3 is FILEPART2 continued over the end of the inner buffer
18:12:14linuxstb"inner buffer" ?
18:12:34Nico_Pie the handle's invidual buffer
18:14:04Nico_Pyou need to imagine a regular ring buffer with data that wraps, and put it in a ringbuffer, wrapping around the end
18:16:06 Quit petur ("work->home")
18:17:03 Join Wiwie [0] (
18:18:15 Join low_light [0] (i=c730190b@gateway/web/cgi-irc/
18:22:41 Part kretender
18:27:08Nico_Plinuxstb: hopefully this will help:
18:27:25pondlifeNico_P: You're not anywhere near London next Tuesday are you? :)
18:27:28Nico_Pthe main buffer is the line
18:27:31Nico_Ppondlife: no :(
18:28:02Nico_Ppondlife: aslo I don't like beer :p
18:28:17pondlifebeer's not compulsory, but me understanding playback is
18:28:43Nico_Pwell going on with my example picture, the file starts white and ends black
18:28:55pondlifeAh, I see
18:29:15pondlifeOK, what happens when you have 30 small tracks buffered (each with metadata) and then a very big track comes along?
18:29:33Nico_Pwell imagine the 30 tracks are in the empty space
18:29:38Nico_Pand you have it
18:29:59pondlifeAha, so no fragmentation.
18:30:06Nico_Phopefully not
18:30:27Nico_Pthere will be holes when tracks are closed, but that should be all
18:30:27pondlifeGreat, now let's get that plugin in place and we can test
18:30:39pondlifeMaybe you'd like to put that graphic into a wiki too?
18:30:39Nico_Pthe plugin already kinda works
18:30:59Nico_PI'll probably improve it a bit with legends but yeah, good idea
18:31:02pondlifeSo I hear, is JdGordon's SVN up to date?
18:31:13*pondlife has been away, so not tried it yet.
18:31:24pondlifeMaybe tonight
18:31:31Nico_Pbut currently there's only one file on the buffer at a time
18:31:45Nico_Pbut a long one works fine
18:33:32pondlifeSo get that one playing back , then start to peek the current playlist...
18:37:30 Quit XavierGr (Nick collision from services.)
18:37:34 Join XavierGr [0] (
18:39:39 Quit austriancoder ("Kopete 0.12.4 :")
18:39:54 Join Farp [0] (
18:42:37 Part maffe
18:42:43 Quit obo ("bye")
18:43:24 Join Domonoky [0] (
18:50:18 Quit XavierGr (Nick collision from services.)
18:50:21 Join XavierGr [0] (
18:51:43Nico_Ppondlife: how'd you like this ?
18:53:23pondlifePretty good! Is the description of reading/writing position correct though.
18:53:54pondlifeIt almost looks like the writing position might move backward (from black towards grey) the way you've worded it...
18:54:27Nico_Pno, both move forward
18:54:27Nico_Pand reading is always "behind" writing
18:54:34pondlifeI know that, but it's not totally clear to the playback n00b. ;)
18:54:42Nico_Pthere, writing has caught up with reading, which is why the buffer is full
18:55:31pondlifeMaybe try and avoid the terminology of read/write positions here, the problem is that "read position" is an unbuffering/playback concept, not a buffering one.
18:55:39pondlife(At least it is in my head!)
18:55:47Nico_Pyes, true
18:56:25Nico_Pbut in this case the file has been read already, otherwise the start of the file would be at the start of the buffer
18:56:30Nico_Pand there would be no wrap
18:57:06pondlifeSo we now have RBRBIRB - aka RockboxRingBufferInRingBuffer
18:57:07 Join TTSbox_ [0] (
18:57:14Nico_Phehe :)
18:57:54pondlifeIt's good to document :) or so they keep telling me at work..
18:58:26pixelmalinuxstb: you've asked about the sandisk m200 firmware files earlier?
18:58:37Nico_PI added a bit more, what do you think ?
18:58:48 Join Robin0800 [0] (
18:59:19 Quit XavierGr (Nick collision from services.)
18:59:22 Join XavierGr [0] (
18:59:27pondlifeAh, now I understand.
18:59:39pondlifeYep, better.
19:00:08Nico_Pstill not very n00b-friendly I imagine, but I think it'll be enough for now
19:00:11pondlifeYou might also want to update the BufferingAPIProposal page to match your API (assuming it must have changed a little).
19:00:48pondlifeHopefully we can drop the proposal and it'll be available to video etc. one day.
19:03:20 Join maffe [0] (
19:03:48 Join hannesd [0] (
19:03:51GodEaterSomeone's going to have to explain that to me in small words come next Tuesday, because I think I'm just not getting it
19:04:19pondlifeGodEater: It'll be easier after that beer I keep going on about.
19:04:20GodEaterand since Nico won't be there, I choose you Pondlife!!
19:04:51pondlifeHmm I thought this was meant to be fun!
19:05:37***Saving seen data "./dancer.seen"
19:05:44 Quit jhMikeS (Nick collision from services.)
19:05:48GodEaterI can only imagine that after a few beers it'll turn into a "Ford Prefect explains time and space to Arthur Dent" sort of explanation
19:05:50 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
19:05:58pondlifeI'd also rather hope to discuss it with linuxstb; he has lots more experience at RB hacking than I do.
19:05:58GodEateri.e. have nothing whatsoever to do with the subject at hand
19:06:31GodEaterlinuxstb managed to remain mostly silent on the subject of RB hacking last time
19:06:45GodEaterand we mainly ended up laughing at forum posts
19:06:50pondlifeThere's been a nagging doubt at the back of my head that says "It'll never work" ever since I put the buffering API proposal on the wiki...
19:07:14GodEateris this a "I'll keep my coat to hand" sort of nagging doubt ?
19:07:22pondlifeYes. But looks like Nico_P (and JdGordon) are proving me wrong...
19:08:15pondlifeAnyway, got to go now...
19:08:19GodEaterbye bye
19:08:49 Quit pondlife ("disconnected has pondlife")
19:09:02 Quit XavierGr (Nick collision from services.)
19:09:05 Join XavierGr [0] (
19:09:18Nico_Poh missed him
19:09:26Nico_PGodEater: seen the wiki page ?
19:09:38GodEateryes - it's not helping - but I've had a long day
19:09:47GodEaterand may just be feeling particularly stupid right now
19:09:54 Join jgarvey [0] (i=jgarvey@nat/ibm/x-e1c8230b82249ea8)
19:15:26amiconnNico_P: As you asked earlier (iirc) - hwcodec doesn't need any inherent guard buffer, but maybe we want one for mp3 (also on swcodec), for framewalking
19:15:29Nico_Pif there's something in particular that you don't get I can try giving more details
19:16:02GodEaterNico_P: no - I'm convinced I'm just having a neural short circuit right now - I'll read it again in the morning and hopefully it'll make sense ;)
19:16:17Nico_Pamiconn: could the current one do ? (there is a 32K one )
19:16:49amiconnIn fact it could, and should, be smaller than that on low-mem targets
19:16:51Nico_PGodEater: ok :)
19:17:29amiconnIt just needs to be large enough to fit one worst-case sized mp3 frame
19:17:40Nico_Pisn't that 32K ?
19:18:02amiconnnah, far less
19:18:07Nico_Pok :)
19:18:30 Quit PaulJam (".")
19:18:31Nico_Pare there big differences between HWCODEC and SWCODEC I should be aware of ?
19:18:48Nico_PI know almost nothing about the HWCODEC playback code
19:18:54amiconnmp3 frames are either 1152 samples (version 1, 48/44/32 kHz) or 576 samples (version 2 and 2.5, 24/22/16/12/11/8 kHz)
19:19:16amiconnSo the worst case mp3 frame would be 160kbps 8kHz
19:19:57 Quit atsea- (Read error: 104 (Connection reset by peer))
19:19:57amiconn1440 bytes
19:20:31amiconnThe biggest issue with hwcodec which isn't an actual hwcodec issue but rather an SH1 issue is the bitswap
19:22:13amiconnWe have 2 options, either have a separate buffer (the equivalent of the pcm buffer on swcodec) where our "codec" (a bitswap-copy) copies data into from the main buffer, or do the swap on buffer
19:23:07Nico_PI think the buffering could handle the bitswapping
19:23:13amiconn(1) has the advantage that the main playback engine would be the same for hwcodec and swcodec, and that the "codec" could in fact be a real codec, doing some simple things like decoding adpcm
19:24:31amiconnBut it needs additional memory for the buffer, and depending on the "codec" latency this might cut away a substantial amount from the main buffer
19:24:32 Join Rondom [0] (
19:24:58amiconn(2) wouldn't need additional memory and/or fiddling with low latency, but it's less flexible (no adpcm decoding or similar)
19:25:47amiconn(1) wouldn't be much of an issue with mp2/mp3, but when we switch playback engines, I also want to integrate pcm playback, and then it might become an issue
19:26:14 Quit ptw419 ()
19:26:32Nico_Pwhat's adpcm decoding ?
19:26:59amiconnAh, that reminds me - for hwcodec, the buffering needs to remember the type of data on the buffer, distinguishing mp2/mp3, pcm litte endian, and pcm big endian
19:27:25amiconn...and pcm 8 bit. For pcm, it also needs to distinguish mono and stereo
19:27:43 Quit TTSbox (Read error: 110 (Connection timed out))
19:27:58amiconnadpcm is a simple pcm compression scheme. Saving bits by non-linear quantisation
19:28:06Nico_POK... that will go in the struct memory handle with an ifdef HWCODEC
19:28:07 Join webguest03 [0] (i=3ff6aa60@gateway/web/cgi-irc/
19:28:33amiconnThe swcodec wav.codec can handles several of those schemes
19:29:11amiconnFor swcodec, the buffering also needs to store which codec needs to be used, so it's not too different imo
19:29:11 Quit webguest03 (Client Quit)
19:31:17 Join obo [0] (
19:31:50 Join miepchen^schlaf [0] (
19:32:17 Join atsea- [0] (i=ariel@gateway/tor/x-1a3ba7133fef6bbc)
19:32:31 Join webguest81 [0] (i=3ff6aa60@gateway/web/cgi-irc/
19:36:56Nico_PI'll keep all that in mind
19:36:56 Quit webguest81 (Client Quit)
19:40:21Kasperlewhat about "date +%s"?
19:40:27Kasperleoh, sorry, wrong chan
19:43:46 Join thegeek [0] (
19:49:13 Quit Rondom (Read error: 113 (No route to host))
19:49:38 Quit kubiixaka ("Miranda IM! Smaller, Faster, Easier.")
19:51:00 Join kubiix [0] (
19:52:00 Quit XavierGr (Nick collision from services.)
19:52:03 Join XavierGr [0] (
20:10:30 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
20:11:31 Join My_Sic [0] (
20:14:19 Quit My_Sic (Read error: 104 (Connection reset by peer))
20:19:32 Join dandin1 [0] (
20:20:15 Join RoC_MasterMind [0] (
20:24:03 Quit Wiwie (Read error: 104 (Connection reset by peer))
20:28:19 Quit RoC_MasterMind (Remote closed the connection)
20:32:12 Join Paprica [0] (n=Paprica@rockbox/developer/paprica)
20:32:26 Quit z35 (Read error: 110 (Connection timed out))
20:37:06 Join linuxstb_ [0] (i=5343d4aa@rockbox/developer/linuxstb)
20:37:18 Join PaulJam [0] (
20:39:27 Part toffe82
20:43:38 Join RoC_MasterMind [0] (
20:43:38linuxstb_pixelma: You mentioned that I mentioned the M200 firmware files?
20:44:38pixelmayes, I wasn't sure... just wanted to mention that the wiki page only looked like a collection of pastes from different forums to me (with additional scans). The MaxKelley part is from the rockbox new ports forum.... maybe you can contact the member and ask for the files again?
20:45:14linuxstb_Yes, I discovered that. However, I found the firmware update .exe on sandisk's website, and managed to extract the files myself.
20:46:08pixelmaah, ok
20:46:25linuxstb_A lot of the players using Telechips CPUs seem to share a common firmware format, and I'm starting to build up a collection of them. I should probably start something like Bagder's MI4 page he used to collect MI4 firmwares.
20:46:27 Quit hannesd (Read error: 110 (Connection timed out))
20:46:56 Join stripwax [0] (
20:47:30linuxstb_(the M200 firmwares I found are all in this format as well)
20:47:57linuxstb_I'm guessing they're all TCC77x devices.
20:49:09 Join jeff4564 [0] (n=emtmanj@
20:49:12pixelmasounds interesting - hope it'll help if you could combine efforts or share knowledge or so
20:49:31 Quit RoC_MasterMind (Remote closed the connection)
20:49:41 Part jeff4564
20:52:54*linuxstb_ wonders if that should be done on the wiki, or hosted somewhere else like Bagder's MI4 page...
20:53:03 Join RoC_MasterMind [0] (
20:54:10 Join Tr1ckY^Gone [0] (
20:54:10 Quit Tr1ckY (Read error: 104 (Connection reset by peer))
20:57:59 Join MonoPete [0] (
21:00:06 Quit RoC_MasterMind (Remote closed the connection)
21:01:20 Join thegeek_ [0] (
21:04:37 Quit thegeek (Read error: 110 (Connection timed out))
21:05:39***Saving seen data "./dancer.seen"
21:06:31 Quit markun (Read error: 104 (Connection reset by peer))
21:06:35 Join markun [0] (n=markun@rockbox/developer/markun)
21:08:56 Join My_Sic [0] (
21:11:24GodEatergiven the number of "New Ports" requests I've seen in the forums recently, I'm given to wonder if there are any DAP manufacturers out there that manage to make a firmware their users like
21:12:10GodEateror whether it's just a knee jerk reaction that the kids have to look "cool".
21:12:43GodEater"the stock firmware was made by a big corporation, it must suck, otherwise why would rockbox exist?"
21:13:05*GodEater continues wittering away to himself
21:13:26Kasperleat least my DAPs firmware _really_ sucks :D
21:13:28tumuit's a question of money and licensing
21:14:09Kasperlebut i got it for free, and rockbox fixes the firmware part :)
21:15:00 Join toffe82 [0] (
21:15:50LloreanGodEater: So far I haven't used a stock firmware I've really liked
21:16:18GodEaterLlorean: neither have I - this is what makes me wonder
21:16:37GodEateralthough to be fair I never even *tried* the gigabeat's
21:17:21LloreanBut I tend to think it's more economics than anything. Maybe I'm cynical, but were I designing these MP3 players, I might just say "Let's give them the bare minimum they'll pay this price for."
21:17:31Llorean"That way we can offer 'upgrades' later and charge them again."
21:18:03GodEaterLlorean: you're probably close to the truth sadly
21:18:21LloreanI mean Archos charges for additional codecs, don't they?
21:18:39GodEaterdo they really ?
21:18:41GodEatergood lord
21:19:15*GodEater joins in on Llorean's "List of things you're doing wrong" theme
21:19:44Kasperleat least apple's motivation is clear ;)
21:20:05Kasperlethey want to restrict their users' freedom to sell more music through itunes
21:20:48LloreanGodEater: Well, he did ask. :)
21:21:04GodEaterLlorean: and then he asked again :)
21:21:42GodEaterI'm really not following the logic of the "this thread is dead" people either. It's not like we locked it.
21:22:00LloreanI think part of the problem is that people don't really that it's essentially a choice: Enforce all the rules, or none.
21:22:12LloreanIf you only enforce some some of the time, it gives people room to call favoritism.
21:22:26LloreanI really don't mind if someone suggests the rules need to be changed.
21:22:37GodEaterI like them the way they are
21:22:40LloreanBut I do mind if they say they shouldn't be enforced.
21:22:53LloreanReally, I think he should just open a new thread formatted the way I suggested about halfway through
21:23:03LloreanIt solves the "this thread is dead" problem, and brings about what he wants too.
21:23:07GodEaterhe seemed to own a very short fuse
21:23:15GodEaterI don't think he could be reasoned with much
21:23:19GodEateryou didn't even slap him that hard
21:23:31LloreanWell, he was sent a PM
21:23:44*GodEater was not aware of this
21:23:45LloreanBut all it said was "Your most recent post has been deleted. As per our forum guidelines, don't cross-post."
21:23:58LloreanI'm paraphrasing, but that is essentially the tone.
21:24:09GodEaterah well - I still think that was fairly tame
21:24:14GodEaterhe just seemed to go a bit mental
21:24:41 Nick Tr1ckY_LAP is now known as Tr1ckY (
21:24:52GodEaterI don't really buy his "I've veen following this project for four years" statement either
21:25:04GodEater4 years and not a single post to the forums is pretty incredible
21:25:45GodEaterthanks for the badge upgrade by the way. I'm not sure if it's a curse or a blessing yet :)
21:26:35LloreanHis oldest post is from '05
21:26:42LloreanNov 27
21:26:50LloreanSo he's got at least a year and a half
21:26:59LloreanMore than enough time to know that the forums have restrictions
21:27:00GodEaterI stand corrected then - I really thought he'd just waded in
21:27:14LloreanIt certainly seemed like it.
21:27:32LloreanI mean, is it just me, or did his statement regarding an EQ really sound like he didn't know we have one?
21:27:56GodEaterit gave that impression yes
21:28:46LloreanAlright, I just wanted to make sure it wasn't just me.
21:29:06LloreanI tend to read things very, very, very literally often, so sometimes I do get a bit retentive as it were.
21:29:35GodEaterI'm tempted to follow you up with a ""Joe Public" hasn't read the rules in that case" post - but I think the newbie is just trolling
21:31:51GodEaterAGGIN sure is racking up the forum rules violations :)
21:31:59 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
21:33:46LloreanI'm thinking I may add more "Common Terms" to the guidelines.
21:34:24 Nick Tr1ckY is now known as Tr1ckY^ (
21:34:35GodEateralso - you said "patched build" first time, and then "unsupported build" second time. Be consistent :)
21:34:43LloreanGodEater: Check my post again. ;)
21:34:52GodEaterhehe ok
21:35:53LloreanBut I was thinking of adding a few more common terms used around our forums. Clear up that "Patched" and "Unsupported" builds are the same thing (from a general standpoint) and what the Root of a device is.
21:36:08LloreanSince there's already some term clarification in the Guidelines, a little more won't hurt.
21:36:44GodEaterDid LamdaCalculus copy your post before you editted it ? :)
21:36:59LloreanI meant my most recent post
21:38:11GodEaterI must need glasses
21:38:15GodEateror possibly just more sleep
21:38:20LloreanIt didn't say that originally
21:38:30LloreanBut immediately after posting it, I realized and corrected it
21:38:46LloreanThere's a grace period, I'm not sure how long, where if you edit your post almost immediately after it doesn't show as edited.
21:39:05LloreanIt's for fixing those typos and silly mistakes. :)
21:39:09GodEaterbtw - I meant to ask - do ALL deleted posts end up in the trash ?
21:39:18LloreanIf you use the actual delete function, they don't
21:39:20GodEaterso ones you delete yourself don't
21:39:23LloreanBut I don't think Experts have that.
21:39:42GodEateronly on our own posts :)
21:40:07LloreanI can delete anyone's permanently, but it's not a power I granted on experts, because I'm stingy like that.
21:40:18GodEaterso demonstrations of crass stupidity can be trimmed with a bit of after thought :)
21:40:29*GodEater isn't fussed :)
21:40:49LloreanWell, I also like to think it heads off accidents.
21:40:54LloreanOnly I can stupidly delete an important thread.
21:41:01GodEatervery true
21:41:21GodEaterI was wondering today if there's a feature similar to slashdot's "karma" for SMF
21:41:53 Quit desowin ("use linux")
21:41:57GodEaterat the moment the way one identifies someone with clue on our forums is very black and white
21:42:33GodEaterI thought it would be nice if "good" answers to threads could be rewarded somehow
21:42:37FebsLlorean: I can also stupidly delete an important thread.
21:43:04*GodEater waits for a demonstration :)
21:43:11LloreanFebs: Well yes, a blue name gives you almost the same ability to screw up as a red name.
21:43:21 Join Guile`` [0] (
21:43:24LloreanI think the only thing I have that you don't is the ability to accidentally break the forums entirely
21:43:47LloreanGodEater: The forums have a "Karma" system, but I've so far avoided enabling it.
21:43:58GodEateryou don't like the idea ?
21:44:26GodEateror just the implementation ?
21:45:06dan_aKarma + idiots is a bad mix
21:45:46GodEaterdan_a: it depends how the karma is awarded / earned
21:46:44dan_aWe have a karma system of sorts, though...
21:47:18dan_a(aka Rockbox Experts)
21:49:01Lloreanpixelma: He'd already posted elsewhere, where I told him that subject wasn't appropriate.
21:49:19LloreanGodEater: AFAIK, once per day, people can add or subtract a point from someone else
21:49:59 Join stripwax [0] (
21:50:13GodEaterLlorean: I'd restrict the ability to award that point to Experts and above though
21:50:18GodEaternot just random forum users
21:50:59LloreanI'm just not sure about it.
21:51:09LloreanI'm not sure it's appropriate for an actual technical forum. But I'll think on it.
21:51:25pixelmaLlorean: I know (in the installation forums) - the one he chose afterwards isn't much better though but the main thing I wanted to make clear is the why
21:51:49LloreanAh, gotcha
21:52:01LloreanI think our guidelines explain the why. I tried to explain the why for most of our rules in them.
21:52:10LloreanIt'd be really nifty if people would read 'em, though.
21:52:27GodEaterbbiab -X session has gone peculiar
21:52:31 Quit GodEater (Remote closed the connection)
21:55:15FebsWhen I see a thread title like "NOOB NEED HELP," I can't help but picture the person who started the thread as a caveman.
21:56:09LloreanI do have a rather unsavory picture of those who call themselves a noob.
21:56:12LloreanSomehow those never really go well
21:56:47pixelmait was also a response to his "can you help", and trying some harmony... ;)
21:57:46 Quit The-Compiler (Read error: 104 (Connection reset by peer))
22:03:06 Join MySic [0] (
22:03:07 Quit MySic (Read error: 104 (Connection reset by peer))
22:04:07 Join Zagor [0] (
22:05:03 Join pondlife [0] (
22:06:58 Join Bagder [0] (n=daniel@rockbox/developer/bagder)
22:09:54 Join entheh [0] (
22:10:40 Join Lear [0] (i=chatzill@rockbox/developer/lear)
22:12:46 Join GodEater [0] (
22:14:33*GodEater 's build server is misbehaving
22:15:17linuxstb_Bagder's normally good at kicking build servers...
22:15:43GodEaterhe'll have to kick mine really hard - I've turned it off
22:15:47GodEaterthink it's overheating
22:16:00*Bagder puts on his big boots
22:16:12*dan_a sends ice cubes
22:19:46 Quit low_light ("CGI:IRC")
22:22:29 Quit davina ("xchat on Ubuntu 7.04")
22:24:52 Join davina [0] (
22:26:33 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
22:26:59 Quit GodEater_ (Read error: 113 (No route to host))
22:27:42dan_aBagder: I think we may need to change the standard Rockbox source file header
22:29:02dan_aAt the moment we don't specify which version of the GPL we are distributing under, which means that people can choose any version ever published
22:29:03 Join fm2 [0] (i=53f23c04@gateway/web/cgi-irc/
22:29:23Bagderwell, it refers to COPYING in the source tree, which is v2
22:29:31Bagderbut we should probably make it more explicit
22:29:56scorcheBagder: ptw419 sent you another mail, i believe..did it end up int he spam category?
22:30:10fm2Hello. Has the feature request been noticed? Has such a feature been discussed before?
22:30:29dan_aThe COPYING file is V2, but the program isn't
22:31:07Bagderscorche: I don't see any mail
22:31:13Lloreanfm2: If you want to launch a folder from the root, why not put the folder in the root?
22:31:17Bagderhe's not using hotmail or similar?
22:31:29scorchedo you have another address that he can send it to other than danielATrockbox?
22:31:30Bagderhotmail hates us
22:31:37scorchei have no clue what he is using
22:31:40Bagderat works too
22:32:04Bagderdan_a: ?
22:32:19Zagordan_a: each file doesn't have to contain all license details. referring to COPYING is fine.
22:32:37Bagderscorche: what's ptw419's name again?
22:33:06scorcheBagder: i dont know, but you should have it from Zagor giving him access tot he gigabeat branch
22:33:08 Join Buschel [0] (
22:33:17dan_aBagder, Zagor: It's the COPYING file which says that if the Program doesn't explicitly state a version, then the end user can choose which version they use (section 9)
22:33:34toffe82Bagder: espinoza
22:33:40fm2Llorean: I wouldn't like to change the folder structure. I would like to be able to jump to those folders from everywhere.
22:33:57Zagordan_a: the program says v2, by referring to COPYING
22:34:03toffe82Bagder: James
22:34:11Lloreanfm2: That doesn't make sense.
22:34:14Bagderthen I got the mail...
22:34:20LloreanThere has to be a location that you access them from. The location where you put your bookmarks.
22:34:26LloreanIf there is, why not simply put the folders themselves there.
22:34:28GodEaterLlorean: fm2 means kind of like a windows shortcut
22:34:46LloreanGodEater: I understand, but I don't understand why it makes sense as a Rockbox feature.
22:34:56Buschelhi there, has anybody measured or does know what is the power consumption of an iPOD running with 30MHz against 78MHz? I found some assumption talking about ~13mA as the difference
22:35:00GodEaterdepends how deep your file tree is
22:35:01Zagordan_a: it really can't work any other way or you'd need to copy the full license into each header
22:35:19LloreanGodEater: As I said though, if you're going to be creating shortcuts in your root, why not simply put the folders those shortcuts are to?
22:35:27BagderZagor: the header could state it is v2 only, the linux kernel way
22:35:28LloreanOr even simply use the database
22:35:51ZagorBagder: yes it could, but I can't see that it has to
22:35:56scorcheBagder: also, i assume you would want the themes code in www/themes ?
22:35:59markunfm2: at least it's being discussed now :)
22:36:43BagderZagor: no, but as it is now we allow "any later version"
22:36:47fm2GodEater: no, I don't mean windows shortcuts (or UNIX links). I mean a sort of bookmarks that point to a folder rather than to a certain point in a song
22:36:49 Quit jgarvey ("Leaving")
22:37:03Lloreanfm2: Then why can't you put the folder there?
22:37:04GodEaterfm2: windows shortcuts allow you to do exactly that
22:37:14dan_aZagor: the COPYING file is the text of the GPL version 2 - and in that text it says that unless we have explicity limited the versions, then the recipient can pick any later (or earlier) version of the GPL
22:37:18GodEaterI use that feature of windows on a daily basis
22:37:33fm2Llorean: where?
22:37:38LloreanWherever the bookmark is.
22:37:43 Quit My_Sic (Read error: 113 (No route to host))
22:37:45Zagordan_a: you're right, my bad
22:37:50LloreanOr are you planning on having say, seven bookmarks to one folder scattered in random places?
22:37:51fm2GodEater: I must have missed this feature :-/
22:38:16Bagderaha a yahoo address, very easily triggered as spam in my end...
22:38:32*Zagor adds ptw419 to James' wiki page
22:39:02scorcheBagder: so i dont need to tell him to re-send it?
22:39:13Bagdernope, I found it (at last)
22:39:22scorchegood =)
22:39:31 Join petur [0] (n=petur@rockbox/developer/petur)
22:39:46Nimdaebagder, i'll be taking kermit offline at the end of the month
22:39:54fm2Llorean: say, I'm listening to The Beatles and want to listen to ZZ Top. But now I'm in the Beatles folder. So I'd just defined a "bookmark" to the ZZ Top folder.
22:40:15BagderNimdae: that's no problem, the master adjusts to hosts simply "missing"
22:40:20GodEaterI'm kind of failing to see the point here too
22:40:20Lloreanfm2: So you'd have bookmark file in the Beatles folder for the ZZ Top folder?
22:40:30LloreanOr would you browse to the bookmarks folder?
22:41:05LloreanI mean, can't you simply do that with favorite playlists anyway, for playback?
22:41:07scorcheNimdae: mahna mahna
22:41:16Nimdaesomething like that
22:41:18fm2Llorean: no, I'd have only one "global" bookmark file (the same as it is for normal bookmarks if I understand it correctly).
22:41:31Lloreanfm2: Normal bookmarks have individual files scattered about.
22:42:00LloreanAs well, if you want to put all the "Folder Bookmarks" in one folder for easy access... wouldn't it be identical to putting all the folders in one folder for easy access?
22:42:18linuxstb_fm2: Sounds like a directory full of the equivalent to Windows shortcuts to me.
22:42:33fm2Llorean: oh! I didn't know that. How are "recent bookmarks" implemented? There must be a global file. This is what I mean.
22:43:23Lloreanfm2: It doesn't really change that you'd still have the equivalent of a folder full of folders, the same as simply reorganizing your collection or creating a custom set of database queries.
22:43:28LearRecent bookmarks is via a global file, yes.
22:44:35Bagderto me, having the gigabeat-s port use 4.1.x is not a big problem
22:44:40fm2Llorean: so if I'm listening to a song and choose "Create bookmark" in the menu, the bookmark file is created in this folder? The bookmark isn't stored in some global file?
22:45:13Nimdaethere is a gigabeat s port?
22:45:18Lloreanfm2: How is a list of files any different from a list of entries on one file?
22:45:21Learfm2: Depends on your settings.
22:45:57scorcheBagder: just seems like extra effort with the various methods of getting the cross compilers and such
22:46:17Bagderyes sure, the best is if we can use a common version but still
22:46:22Lloreanscorche: As long as works for adding it (or will work), I don't really think it's a big problem
22:46:22fm2Lear: what setting is important for this?
22:46:39scorcheLlorean: rockboxdev...i thought you got past that stage? =P
22:46:41Bagderexactly, we just have to make sure the 4.1.x version get differently named tools
22:46:50Lloreanscorche: I think I'll never get past that stage. The fever doesn't help. :-P
22:47:06fm2Llorean: that last question I couldn't understand. What entries?
22:47:16toffe82Nimdae: yes it started a while ago but make big progress during the last weeks
22:47:18Learfm2: General Settings -> Bookmarking -> Maintain a list of recent bookmarks.
22:48:03Lloreanfm2: for example, what's the difference between viewing a playlist of certain songs (one file, multiple entries) and viewing a folder of those songs (multiple files)?
22:48:04LearBut the "local" bookmark file is always created. The global file is optional.
22:49:13fm2Llorean: well, for me as the user, there is no difference. It's a list of songs.
22:50:24LloreanSo, what's the difference between a list of your favorite folders, and a folder containing your favorite folders?
22:51:05fm2Lear: ok, I probably didn't describe it clearly. What I want is a global list of folders to quickly jump to. This list may also contain "normal" bookmarks that point to a song. We'd essentially have two sorts of bookmarks: those that point to a song and those that point to a folder
22:51:05GodEaterI would imagine the speed of changing your tastes is the difference
22:51:27GodEatermoving an entire folder is going to take longer
22:51:38LloreanCutting and pasting a folder shouldn't take very long at all, I thought
22:51:38GodEaterthan creating / deleting a shortcut to it ?
22:51:46 Join DefineByte [0] (i=5751c305@gateway/web/cgi-irc/
22:52:05fm2Llorean: the difference is that if I'm in a deeply nested folder I'd have to first go up to get to that "favorite" folder
22:52:16Nimdaedoes rockbox support any other filesystems than fat?
22:52:21dan_aCould a Sansa owner try for me? It doesn't seem to mess up my Sansa's display, and I need confirmation that I've done something stupid
22:52:21scorcheNimdae: no
22:52:23Lloreanfm2: How long does it take to get to the root folder when you hold left?
22:52:36LloreanHow long does it take for you to press the several buttons it takes to get to the bookmark menu?
22:53:00GodEaterNimdae: support for more that FAT is on the NODO list
22:54:06fm2Llorean: hmm... yes, you're right in this point. But then I would have to change the lib structure which I don't quite like.
22:54:23DefineByteanyone know what's holding up the LCD brightness patch for iPods from being commited?
22:54:35LloreanDefineByte: Doesn't work in several plugins, at the least, I thought
22:55:02DefineBytethat's true. you need to put it on full for plugins to work
22:55:09 Quit Febs (Read error: 110 (Connection timed out))
22:55:14linuxstb_DefineByte: Does using that patch to reduce brightness increase battery life?
22:55:20fm2Llorean: this week I may listen to The Beatles very often, next month it will be something else. Should I move dirs every time? A bookmark seems a better solution to me.
22:55:22LloreanDefineByte: So, at least it needs some work.
22:55:43Lloreanfm2: What makes a bookmark a better solution, exactly? One creates a whole file, the other I believe simply changes an entry in the fat.
22:56:03DefineBytesome people have said it does but i haven't tested it
22:56:14DefineBytei just use it to stop me from going blind
22:56:53fm2Llorean: the lib structure doesn't change. So I'll have no problems with synchronising my music library (for example).
22:57:30peturLlorean: if you want to see a karma system at work, have a peek at - it's about as technical as our forum. I do not see any added value in it though
22:57:35DefineBytesomeone seems to think the hardware registers for brightness had been found but since that post...nothing
22:58:26Lloreanpetur: That's what I thought. I just worry it's going to create additional ill will, as lower karma users have a hard time putting their initial problems behind them.
22:58:45 Join My_Sic [0] (
22:58:53peturI wouldn't bother with it
22:59:03LloreanI'll leave it off for now then.
22:59:05peturwe already have badges
22:59:20scorchei wouldnt worry with it either
22:59:32peturon taperssection, it helps to put a value on the info you get...
22:59:43Nimdaeah ok, was just curious
22:59:53scorcheany system like that depends more on popularity anyway
22:59:57peturif you get a tip from a negative karma guy, beware ;)
23:00:07 Join bascule_forum [0] (i=586cd5d9@gateway/web/cgi-irc/
23:00:15Lloreanscorche: I should enable it for a week and see if mine ends up positive or negative then. :-P
23:00:15DefineBytei'm trying to work out the wps freeze on ipod 5/5.5g but it's pretty tough going
23:00:39scorcheLlorean: i would be sure to downrank you
23:01:10scorchepetur: only 93 tickets?...what gives? =P
23:01:11 Part fm2 ("Bye")
23:01:11GodEaterthat's why I suggested only experts / devs / admins being able to award karma
23:01:19Lloreanscorche: I could just do a poll: "Do you think I'm doing a good job", and then (assuming it leans 'yes') point to that next time there's a "You guys are jerks" thread.
23:01:27GodEaterbut I guess if scorche is going to start of negging people ;)
23:01:32peturscorche: that is a lot compared to my post count
23:01:42 Quit Buschel ()
23:01:48bascule_forumI'm on VMware, trying to compile a sim, following the Wiki instructions. make is OK, make install just fails to create the simulator files. ideas?
23:02:23scorchebascule_forum: how nice of you to finally join us here =)
23:02:31GodEaterweren't we only saying today we wish bascule would join us ?
23:02:37scorcheyes, we were
23:02:48linuxstb_bascule_forum: Does it give any error messages? It normally just works silently.
23:02:49 Quit Jens ()
23:02:51bascule_forumit doesn't happen often :)
23:03:04GodEaterthat would be why we said it
23:03:11bascule_forumit makes the archos dir, but it's empty
23:03:31linuxstb_It should contain a hidden .rockbox directory.
23:03:42GodEaterwhat a fantastic PM I just got!
23:03:48Lloreanbascule_forum: Are you sure it's empty? Try cding into the .rockbox directory in it anyway
23:03:49linuxstb_If you're in the terminal, type "ls -al archos"
23:03:55bascule_forumD'oh. and thatts dir -a?
23:03:57GodEater"hi, my ipod is always broken because of your stupid rockbox, help plxx"
23:04:28DefineByteformat c:\
23:04:40DefineByteassuming windows
23:04:53LloreanI always wonder how they decide who get the pestering PMs.
23:05:10scorcheLlorean: the prettiest color
23:05:16BagderI never get any
23:05:29LloreanI get quite a few, but didn't get that one.
23:05:32Bagderso whatever I do, I do it right ;-)
23:05:33bascule_forumOk, so .rockbox is there, but cd .rockbox gives 'no such file or dir'
23:05:40***Saving seen data "./dancer.seen"
23:05:50GodEaterBagder: I think you growl just right ;)
23:06:35 Quit elinenbe (Read error: 110 (Connection timed out))
23:06:40scorcheBagder: arent you hidden though?
23:06:59bascule_forumMy bad - hadn't cd'd to archos
23:07:05*bascule_forum slaps forhead
23:07:17scorcheforehead gets a lot of abuse here
23:07:38GodEaterI don't believe I've seen "forhead" get it before though
23:08:09 Quit ompaul ("have a nice time")
23:08:09bascule_forumI always panic in IRC - moves to fast for an old 'un
23:08:36GodEaterwe'll try to take it easy on you :)
23:08:56bascule_forumI'm in sim/archos/.rockbox, but I don't see how to execute the sim
23:09:01 Quit davina (Remote closed the connection)
23:09:18linuxstb_bascule_forum: Go back up two directories (cd ../..) and then type ./rockboxui
23:09:33linuxstb_rockboxui is in the directory you typed "make" in.
23:10:14 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
23:10:44*GodEater watches "Tarrant on TV" in fascinated horror....
23:11:25bascule_forumI *really* want to move properly to Linux, but it does my head in...
23:11:35GodEaterLithuanian naked honey wrestling...
23:11:51GodEaterbascule_forum: sorry to hear that :(
23:11:54bascule_forumat that's entertaining to watch :)
23:12:10 Quit Bagder ("Rockbox TCP exception #04123 - connection reset and user kicked out")
23:12:10GodEaterentertaining but hard to understand :)
23:12:16 Quit Xerion (" ")
23:12:36peturhmmm how's the h1x0 RTC mod used atm? I don't see HAVE_RTC commented out in the h1x0 config file
23:13:02bascule_forumOk, so things are good, sim is running. How easy it it to create a windows simulator zip?
23:13:33 Join kfazz [0] (
23:13:56 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
23:14:12 Join miepchen^schlaf [0] (
23:15:25GodEaternot sure myself bascule - never done it
23:15:45 Part maffe
23:15:46bascule_forumis rasher not around?
23:15:53GodEaterrarely is these days
23:16:20 Join maffe [0] (
23:16:30 Join MySic [0] (
23:16:35 Quit My_Sic (Read error: 104 (Connection reset by peer))
23:16:38 Quit SirFunk (Connection timed out)
23:16:45DefineByteback to the grind stone
23:16:50 Part DefineByte
23:16:55bascule_forumOk, well I know I can use the Linux version, and I can copy any files I need using the \\debian\user linkages
23:17:02bascule_forumthanks for all your help
23:17:10 Quit Domonoky (Read error: 104 (Connection reset by peer))
23:17:10linuxstb_petur: I think it's an advanced option in tools/configure
23:17:34bascule_forumas you can see, I struggle with basic options...
23:17:44GodEaterbascule_forum: all it takes is practise ;)
23:18:31PaulJambascule_forum: have you seen this?:
23:19:02GodEaterbascule_forum: we need you here more often to answer the tricky database questions we get :)
23:19:18GodEatermost of us struggle to spell "tagnavi"
23:20:45bascule_forumah, but you see, the database is *easy* ;)
23:21:02GodEaterhands up if you agree with bascule ? :)
23:21:25dan_aEverything's easy to someone
23:21:29*GodEater notes the lack of raised hands
23:22:12LloreanI don't think the database is particularly hard, I just haven't relearned it since the addition of the second custom tagnavi file, and format strings etc.
23:22:40bascule_forumthanks for the Wiki link, it actually looks easier to keep working in VMware
23:22:55bascule_forumplus, it will give me more linux practice
23:22:58linuxstb_There's a second tagnavi file?
23:23:06*linuxstb_ should pay attention in school
23:23:14 Part Tr1ckY^
23:23:25*GodEater rests his case
23:23:38Lloreanlinuxstb_: Yeah, the main one basically includes the custom in (in a C #include sense) so that your custom tagnavi doesn't get overwritten on updates.
23:23:58*linuxstb_ wonders what " if [ "$FESTIVAL" = "$FLITE" ] && [ "$FLITE" = "$ESPEAK" ]; then" is supposed to do in tools/configure...
23:24:24GodEaterI assume '=' in bash is a comparison then
23:24:35GodEaterrather than an assignment
23:24:39linuxstb_It just seems an odd check...
23:24:54GodEaternow you mention it
23:24:59linuxstb_I mean Festival isn't Flite, and Flite isn't espeak...
23:25:00GodEaterit does seem a little strange
23:25:08Lloreanlinuxstb_ Can you do A=B=C in Bash?
23:25:46linuxstb_I don't think so. But yes, that's the equivalent.
23:26:19GodEaterit doesn't look like it makes much sense, but I assume the three variables COULD contain the same thing
23:26:25GodEaterthough why they would is beyond me
23:26:27LloreanOr they could be essentially boolean?
23:26:34LloreanJust checking for the presence of all three?
23:27:02GodEaterwouldn't you do that with two && and no =
23:27:13LloreanOr does maybe Festival have a wrapper for the rest of them?
23:27:22GodEaterdon't think so
23:27:27GodEaterit's a seperate binary AFAIK
23:27:34linuxstb_It seems they are only all the same if they are all null - meaning the user doesn't have any of them installed.
23:29:13linuxstb_GodEater: (5 minutes late), yes = is both comparison and assignment.
23:30:00 Join elinenbe [0] (
23:30:09*bascule_forum is still struggling...
23:30:09GodEaterand on that note - it's past my bedtime
23:30:10GodEaterlater all
23:30:12 Quit GodEater ("User pushed the X - because it's Xtra, baby")
23:30:27bascule_forumwhy do i have to do ./rockboxui and not just 'rockboxui'?
23:30:47 Quit pondlife ("gnight")
23:31:15linuxstb_Because in Unix, the default is normally for the current directory not be in your PATH - meaning it isn't searched for programs.
23:31:32linuxstb_In DOS, the current directory is implicitly in your PATH (IIUC)
23:32:12bascule_forumyes, well, of course i know more Win cmd stuff than linux and just typing the executable name seems so simple
23:32:45linuxstb_If you want, you can add "." to your $PATH variable, then just typing "rockboxui" will work.
23:32:46bascule_forumwhat pupose does it serve for the current dir to NOt be in the path? i don't get it
23:33:05 Part n1s
23:33:19Battousaisuppose you have /bin/passwd
23:33:39Battousaiand then you run some unknown configure script for some reason
23:33:54Battousaiand it may have been tampered with somehow
23:34:01Battousaithen you feel like changing your password for some reason
23:34:15Battousaiif said script placed a passwd in PWD
23:34:23Battousaithen you're not running /bin/passwd anymore are you?
23:34:49LloreanIt's basically simply more explicit.
23:34:55Battousaiof course you'd like /bin higher in the pecking order than PWD
23:35:04LloreanPrograms only run if they're in your path, which you get to choose, or if you explicitly execute them.
23:35:23idnaradding "./" isn't really that much work, anyhow ;)
23:35:29linuxstb_Battousai: Then just put dot at the end of your path...
23:35:47bascule_forumit wasn't the extra typing, it was the reason for it...
23:35:57bluebrother. is usually not part of the path
23:36:00Battousailinuxstb_: that was a crude example and i already covered that
23:36:32Battousaiyou could always do something more obscure like messing with people's rushed tab completion
23:36:33bascule_forumBattousai, thanks for the explanation, but that's still over my head. Llorean's explanation was clearer for me :)
23:36:54LloreanWith windows, if you type a command the local version might execute, so you *have* to check first.
23:37:18LloreanWith linux, assuming a properly formatted path, if you run a program independent of local copies you know which you're getting without having to check anything.
23:38:21 Quit relaxed (Remote closed the connection)
23:38:38bascule_forumI suddenly feel that I'm one of those noobs that spam the forum with silly questions all day long...
23:39:24Lloreanbascule_forum: Everyone's new at one point, and I don't think there's any clear (local) documentation for that one. :-P
23:39:27bascule_forumAnyway, thanks to all. I always enjoy my brief IRC visits and now I have a sim working in VMware and I can go to bed happy
23:39:42 Join dan_a_ [0] (n=dan_a@
23:41:33 Quit dan_a_ (Remote closed the connection)
23:41:48 Quit bascule_forum ("CGI:IRC")
23:43:20 Quit bdgraue_ (Read error: 104 (Connection reset by peer))

Previous day | Next day