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

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

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

#rockbox log for 2008-02-26

00:00:01stripwaxLabyrince - ?
00:00:11Labyrincewindows read it right?
00:00:12stripwaxgevaerts - sure
00:00:21 Nick tilda is now known as Teddy (
00:00:44stripwaxLabyrince - did you miss most of the conversation?
00:00:54stripwaxyou can check the logs for details
00:01:30Labyrincei dont understand what most of it means
00:02:02 Quit BigBambi (Remote closed the connection)
00:02:12 Join HellDragon [0] (i=jd@unaffiliated/helldragon)
00:02:42 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
00:04:14 Join MajorC [0] (
00:05:02stripwaxgevaerts - should the ProductID be nul terminated or space padded?
00:05:23*stripwax also wonders what ProductRevisionLevel to use
00:07:01 Join toffe82 [0] (
00:07:28gevaertsstripwax: ProductID in the SCSI message has to be space padded. ProductRevisionLevel as well.
00:07:54stripwaxta . I'll try "Apple iP", "od ", " "
00:08:11gevaertsstripwax: You might also want to try a different USB_VENDOR_ID and USB_PRODUCT_ID in config-whatever.h
00:09:02stripwaxgevaerts - oh! different as in, something other than the real apple/ipod IDs?
00:09:25gevaertsstripwax: yes. That way windows really can't be confused
00:10:15stripwaxright. actually how do i confirm what vendor/product IDs the OF uses?
00:11:11gevaertsSomewhere in the device manager property view. But they are probably what's in config-whatever.h
00:12:31*gevaerts found a bug in the new rockbox.mi4 detection. If you remove the .rockbox directory and move another one in its place using the file manager, it doesn't prompt to reboot
00:12:46 Join keanu [0] (n=keanu@unaffiliated/keanu)
00:13:47 Quit keanu|afk (Read error: 110 (Connection timed out))
00:14:44*Nico_P is attempting to rework track skipping, and is liking the results
00:16:44stripwaxgevaerts - yep, they are indeed the same . will see what effect changing these might have
00:18:09 Part ax
00:18:21stripwaxHm. It comes up as "Apple iP od USB Device" with a space. Seems .. odd..
00:19:32gevaertsIt seems my "fix" for the Vista BSOD makes rockbox hang instead of reboot on my c200. I'm not sure if that's better.
00:21:29stripwaxbut I do get the "Found New Hardware" popup (but alas no drive letter). Will try USB_VENDOR/PRODUCT_ID next.
00:21:47 Quit MajorC ()
00:21:53 Join BitTorment [0] (
00:22:04 Quit DataGhost (Nick collision from services.)
00:22:12 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
00:22:16 Join seba_ [0] (
00:22:31seba_hi people
00:22:48seba_can i ask you something?
00:22:53stripwaxseba_ - just ask
00:23:00stripwaxdon't need to ask if you can ask
00:23:08gevaertsCan anyone check if current svn (16416 or later) built _without_ USE_ROCKBOX_USB for some PortalPlayer target properly reboots on plugin ? My c200 doesn't, it freezes
00:23:22seba_(my english is very poor,sorry for that)
00:24:13 Quit DataGhost (Nick collision from services.)
00:24:19seba_i have a mp4 player generic(ximpro, chinese), can i install rockbox
00:24:21 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
00:24:24krazykitseba_, no.
00:25:18seba_why? the firmware is no soported?
00:25:43krazykitbecause the only players that rockbox runs on are listed on the front page. yours is not there.
00:25:45stripwaxseba - the supported models are on the front page of
00:25:48scorche|shseba_: each port has to be designed to work with each device one has done that with those devices
00:26:04pixelmagevaerts: I could try but I know it did that before with some builds
00:26:21seba_ok, thanks people!!
00:27:13gevaertspixelma: I'm pretty sure the change I did then can't be the cause. I'd actually like to see a test on some non-c200 targets
00:33:17*stripwax tries using "Apple ","iPod ", "1.62" which is what the OF actually does, rather than what I guessed it does, ..
00:33:44pixelmaI can't help with that then... but I'm not sure it'll tell much even if someone tests on e.g. an Ipod because I encountered this problem with a borrowed Mini over a year ago (IIRC it started with the USB charger vs. connection detection) and reading the forums the hangs appear more often since the beginning of the new stack on all PP devices (remember reports of Ipods and Sansas)
00:34:20gevaertspixelma: I think I'll ignore it for now
00:34:28*amiconn wonders if/when gevaerts will fix the yellow...
00:35:55*gevaerts will make sure to fix it in his next commit
00:36:13cggevaerts: i tried 16418 on ipop video. seems to reboot as normal without USE_ROCKBOX_USB
00:36:45cgipop :)
00:37:58amiconnSome builds reboot properly, some builds freeze, and some builds behave inconsistently
00:38:06gevaertscg: thanks. That means it's most probably some random issue.
00:38:19amiconnThat's nothing new, it's present for more than a year now...
00:38:58gevaertsYes, but it's still useful to know that it's not something else
00:39:05amiconnThe only other observation I made so far is that some targets are affected more than others (averaging across various builds)
00:39:12 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
00:41:53 Quit ompaul (Client Quit)
00:43:15 Quit waldo (Remote closed the connection)
00:43:20*gevaerts expects green now
00:44:31***Saving seen data "./dancer.seen"
00:45:45Labyrinceerm when i try to apply a patch it says the command is not found
00:46:39 Quit ender` (" Democracy: the theory that the common people know what they want and deserve to get it good and hard. -- H. L. Mencken")
00:46:46 Quit spiorf (Remote closed the connection)
00:47:29scorche|shhow are you applying it?
00:48:27 Join bernz [0] (
00:49:09 Join hcs [0] (
00:50:41bernzhi, i know this is technically not rockbox-specific, but i figured sansa c240 owners here would know better: is there a way to find part # for the headphone jack of c240? the only marking it has is an 'E'. :-/
00:51:33scorche|shwell, this isnt a general hardware support channel...i would ask on the sansa forums if i were you
00:53:52bernzfair enough
00:54:01LycoLocois google of no help?
00:54:24bernzwell, it wasn't, but since asking the question −− hot damn! i found something really close, through sheer luck
00:54:32LycoLocohaha, nice
00:54:34bernz(you know how useless google is these days)
00:54:55LycoLocowut? I use it all the time to solve problems
00:55:58bernzwell, i just mean putting something like "3.5 mm smt audio jack" gives many, many useless results... sifting through those takes ages; i have no specific part # of something unique that would weed out all the junk
00:56:23LycoLocowell, why not put "headphone jack sansa c240 part number"
00:58:13bernzhah, nice, someone on abi posted disassembly, and comments relate mostly to this crap-ass quality headphone jack problem :-D
00:58:23LycoLocoanyways, you might wanna post it in this thread bernz
00:58:33LycoLocoit's old, but still would be helpful for people googling
00:59:12LycoLocobernz: that thread might have some help for you if you're only getting sound out of one side
00:59:37*gevaerts expects scorche to say something any time now
01:00:52 Join fasmaie [0] (
01:00:56*scorche|sh walks back to doing work
01:01:33 Quit ch4os_ ("Lost terminal")
01:03:06bernzright, will do, thanks for the suggestions, and sorry for the OT stuff
01:03:13*bernz afk
01:03:14*amiconn pings Bagder
01:03:44 Join rockho [0] (n=rocko@
01:04:15stripwaxgevaerts - hum, so picking a different USB_PRODUCT_ID seems to ensure Windows assigns a drive icon in explorer...
01:04:39stripwaxoh, .. possibly scratch that, one sec
01:05:17 Quit amiconn (Nick collision from services.)
01:05:24 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
01:06:16stripwaxHm - when connecting it, it gets reported (briefly) as Rockbox Media Player. Any ideas how/why that can be?
01:06:53stripwaxand seems I was wrong about the USB_PRODUCT_ID - doesn't seem to help reliably after all
01:07:17 Join iamben [0] (
01:07:51 Quit DataGhost (Nick collision from services.)
01:07:55gevaertsstripwax: that string is defined in usb_core.c, and passed as the Product string descriptor
01:07:59 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
01:08:26*stripwax wonders if Windows could possibly care about that string
01:08:28 Quit conando (Read error: 110 (Connection timed out))
01:08:35gevaertsstripwax: USB_PRODUCT_ID not helping means that there is definitely no ipod special-casing involved
01:08:55*gevaerts thinks that Windows probably doesn't care, but the user might
01:09:09 Quit DataGhost (Nick collision from services.)
01:09:17 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
01:09:19stripwaxgevaerts - probably, but will try changing all three (usb_product_id, usb_vendor_id, and VendorId, ProductId strings)
01:09:40 Quit hcs (
01:09:40 Quit jcollie (
01:09:40 Quit Thundercloud (
01:09:40 Quit advcomp2019 (
01:09:40 Quit Addicted (
01:09:40 Quit amigan (
01:11:48 Join rockhoe [0] (n=rocko@
01:16:20Nico_Pthe UI feels really sluggish on the ipod when unboosted... :(
01:16:46 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
01:17:36 Quit keanu (Read error: 110 (Connection timed out))
01:17:52Nico_PI'm used to the gigabeat...
01:18:17scorche|shjust "ipod" is a bit misleading though
01:18:25*gevaerts is going to sleep now
01:18:51*scorche|sh thinks the gigabeat is sluggish compared to the flash targets..
01:19:03 Quit gevaerts ("Feel free to review and comment on the latest patch in FS#8562")
01:19:11Nico_Pdoes rasing the CPU freq a bit have a really negative effect on the battery life?
01:19:31*Nico_P would like to own a flash target
01:19:32 Quit robin0800 (Read error: 104 (Connection reset by peer))
01:19:58 Join keanu [0] (n=keanu@unaffiliated/keanu)
01:20:13stripwaxNico_P - yes, although I don't see why the CPU couldn't be boosted when the scrollwheel is used
01:20:15 Quit bospaadje ("great minds run in great circles")
01:20:43Nico_Pstripwax: yeah, that would be nice
01:20:49 Quit MethoS-- (Remote closed the connection)
01:20:50Nico_Pis there a patch for it somewhere?
01:21:06 Quit rocko (Read error: 110 (Connection timed out))
01:21:08stripwaxNico_P - I think Buschel had something..
01:21:13 Join rocko [0] (n=rocko@
01:21:53stripwaxgah... data abort ...
01:22:16cgstripwax: seems similar issue, apparently related to scsi-pass-through driver
01:22:30 Quit mirak (Remote closed the connection)
01:22:32 Quit BlakeJohnson86 ("Leaving.")
01:22:59 Join aliask [0] (n=chatzill@rockbox/developer/aliask)
01:23:18 Join BlakeJohnson86 [0] (
01:26:26stripwaxcg - seems odd that everything except rockbox-usb-ipod works for me
01:27:04stripwaxgevaerts - yeah, using completely new IDs and strings seems to make no difference so must be something else
01:29:51stripwaxcg - I don't seem to have spdt installed in any case
01:30:03 Quit rockho (Read error: 110 (Connection timed out))
01:30:04cgstripwax: yeah. maybe the rockbox driver instance is the only one installed after something broke in the system? windows might find the other devices correctly because they have been installed earlier
01:30:08 Quit rockhoe (Read error: 110 (Connection timed out))
01:30:12Nico_Pstripwax: I see Buschel had GUI-boosting as part of his patches in FS #8379 at one point, but it didn't work well on all targets, so he dropped it from the patches
01:30:20cgstripwax: ah ok, probably unrelated then
01:30:32 Join rockho [0] (n=rocko@
01:31:03 Quit advcomp2019 ("IRC is just multiplayer notepad")
01:32:56stripwaxgevaerts - actually I'm at r16411 - I wonder if r16416 might help me and/or if it is related to the problem I'm seeing
01:32:57 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
01:38:10 Part toffe82
01:38:22 Join PeterSansa [0] (n=8a58f6dd@gateway/web/cgi-irc/
01:39:30cgstripwax: did you check if there is "%systemroot%\system32\drivers\sptd.sys" file? it may not show up anywhere else
01:40:31PeterSansaDoes anyone know where I can find screenshots of rockbox? (preferably on a sansa e200 series)
01:40:47stripwaxcg - I searched the windows partition for anything named sptd.* ..
01:41:36cgstripwax: ok, it's pretty clear case then. strange...
01:42:17stripwaxit did work one time though. my laptop is pretty slow. wonder if there is some assumed timing going on inside rockbox..
01:43:23 Quit keanu (Read error: 110 (Connection timed out))
01:43:33stripwaxI'd quite like to do a USB trace, if I even knew how to do that. Or install a debug build of Windows XP, although that's less feasible on this particular machine
01:43:52stripwaxanyway, time for bed.. a tout a l'heure.
01:43:52 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
01:43:52NJoinjcollie [0] (
01:43:52NJoinThundercloud [0] (
01:43:52NJoinAddicted [0] (
01:43:52NJoinamigan [0] (i=dcp1990@unaffiliated/amigan)
01:43:55 Quit ali_as ("zzzzzzzzzzzzzzzzzzzzzzzzzzZZZZZZZZZZZZZZZZZZZZZZZZZZz")
01:44:03 Quit Nico_P (Remote closed the connection)
01:44:03 Quit rocko (Read error: 110 (Connection timed out))
01:44:04 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
01:44:12 Join keanu [0] (n=keanu@unaffiliated/keanu)
01:44:51 Quit DataGhost (Nick collision from services.)
01:44:59 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
01:45:52 Quit DataGhost (Connection reset by peer)
01:46:02 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
01:48:10advcomp2019PeterSansa, the manual has some screenshots
01:50:36PeterSansaI ordered one on today, and there was a download link to a manual.
01:50:57PeterSansabut... I don't think it is rockbox.
01:51:24PeterSansaor do you mean the rockbox manual? sorry, I'm new to this :)
01:52:22krazykitthe rockbox manual, obviously.
01:52:49W3b_6u3stis there any way i can configure the make so it will "make" my new hellowourld?
01:52:54scorche|shyes, the rockbox manual...but keep in mind that Rockbox is vry themeable and can look quite different from the screenshots you see in the manual
01:53:28W3b_6u3st*in cygwin
01:53:41scorche|shW3b_6u3st: i believe there is a wikipage called something like HowToWritePlugins that tells you how to edit the SOURCES file among other things
01:53:54W3b_6u3stkk thx
01:54:02 Join cars [0] (n=451de957@gateway/web/cgi-irc/
01:55:50W3b_6u3stya that was the one part i managed to skip in the wiki
01:56:50 Quit hcs ("Leaving.")
01:59:12 Join ol_schoola [0] (
02:00:13 Quit fasmaie (Read error: 113 (No route to host))
02:01:24PeterSansaDoes anyone know how to tell the difference between a 1gen ipod mini and a 2gen ipod mini?
02:02:17PeterSansaI know it is sort of an off topic question, but I don't know which built to download.
02:03:31scorche|shPeterSansa: has a nice article about can probably dig it up
02:03:46 Join keanu|afk [0] (n=keanu@unaffiliated/keanu)
02:04:14PeterSansathanks scrochelsh.
02:05:05 Quit DataGhost (Nick collision from services.)
02:05:13 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
02:05:25 Quit keanu (Read error: 110 (Connection timed out))
02:05:34PeterSansaI know now, thanks.
02:11:45 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
02:12:08 Join cool_walking_ [0] (
02:12:39 Quit Xerion (Read error: 104 (Connection reset by peer))
02:13:08 Quit PeterSansa ("CGI:IRC")
02:13:10 Join Xerion [0] (
02:17:34 Join Febs [0] (
02:19:55 Join corevette [0] (
02:24:58 Part pixelma
02:35:27 Join crzyboyster [0] (n=6210f0be@gateway/web/cgi-irc/
02:35:40crzyboysterWhat is the screen resolution of the m:robe 100?
02:37:55LloreanCheck out the source
02:37:57 Quit jhMikeS (Nick collision from services.)
02:37:57 Join _jhMikeS_ [50] (n=jethead7@rockbox/developer/jhMikeS)
02:38:47 Quit DerDome ("Leaving.")
02:38:58crzyboysterI know it's 160x128, but is it 160x128x2?
02:39:26LloreanWell, that's in the source too.
02:40:31 Quit crzyboyster ("CGI:IRC")
02:43:30 Quit SirFunk (Read error: 104 (Connection reset by peer))
02:44:32***Saving seen data "./dancer.seen"
02:47:18 Quit aliask ("ChatZilla 0.9.81 [Firefox]")
02:54:12 Quit perrikwp (" ajax IRC Client")
02:57:19 Join perrikwp [0] (i=9821347f@gateway/web/ajax/
03:01:11 Join SirFunk [0] (
03:07:54 Join Neskaya [0] (i=featherd@wiktionary/neskaya)
03:11:35 Part Teddy ("Leaving")
03:14:03 Join rockhoe [0] (n=rocko@
03:14:24 Join Febs_ [0] (
03:15:15 Quit rockho (Read error: 104 (Connection reset by peer))
03:16:34 Quit Thundercloud (Remote closed the connection)
03:22:44 Nick fxb is now known as fxb__ (
03:22:47 Quit dionoea (Read error: 104 (Connection reset by peer))
03:23:21 Join dionoea [0] (
03:23:34 Quit hcs ("Leaving.")
03:30:15 Quit dionoea (Remote closed the connection)
03:34:19 Quit Febs (Read error: 110 (Connection timed out))
03:34:42 Nick Febs_ is now known as Febs (
03:36:01 Quit ol_schoola ()
03:41:58 Join csc` [0] (n=csc@archlinux/user/csc)
03:44:04 Join dionoea_ [0] (
03:50:10 Quit dionoea_ (Read error: 104 (Connection reset by peer))
03:56:57 Join dionoea_ [0] (
04:05:09 Join Variable [0] (
04:06:26 Quit Isolinear (Read error: 110 (Connection timed out))
04:12:57 Quit W3b_6u3st ("CGI:IRC")
04:13:15 Join Kornfan71 [0] (
04:13:40 Quit Kornfan71 (Client Quit)
04:15:58 Join gtkspert_ [0] (
04:18:52 Part Labyrince ("Uh, bye.")
04:27:47 Join crzyboyster [0] (n=6210f0be@gateway/web/cgi-irc/
04:27:58crzyboysterDoes this patch work > ?
04:28:46scorchetry it and find out?
04:28:47carsHave you tried it?
04:29:16 Nick cars is now known as cars2 (n=451de957@gateway/web/cgi-irc/
04:29:18crzyboysterI can't compile...
04:30:39 Quit gtkspert (Read error: 101 (Network is unreachable))
04:30:52 Join kperri [0] (i=9821347f@gateway/web/ajax/
04:31:04 Quit perrikwp (" ajax IRC Client")
04:31:11crzyboysterCan someone test it out? Maybe a future rockbox port?
04:32:33LloreanYou could always take this chance to learn to compile.
04:33:38crzyboysterWell, vmware manages to tell me that I lack "physical memory" and cygwin doesn't work after installing (tried many times) and linux is NOT an option before somebody suggests that...
04:35:00scorchedid you change how much memory vmware asks for?
04:35:30crzyboysterHow can that be done??? (sorry, getting excited, here!)
04:35:36scorcheand *how* did cygwin not work?
04:36:19cars2Which version of VMware are you running?
04:36:41crzyboysterIt never launches. I click on the bat or the shortcut after it installs and no window comes up (or it does flash for a few seconds bit goes away). vmware 2.0.2 I think.
04:37:08cars2crzyboyster: vmware server?
04:38:12crzyboysterThe one pointed to on the wiki (
04:39:04 Quit csc` ("Powering Off")
04:40:22cars2If it's similar to VMware server (which is free from VMware), there should be an
04:41:02cars2..."edit virtual machine settings" button visible on one of the screens. (sorry, hit enter by mistake)
04:41:38crzyboysterLet me check that out...
04:44:37***Saving seen data "./dancer.seen"
04:45:22 Quit crzyboyster ("CGI:IRC (EOF)")
04:47:20 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
04:51:51 Join jurrie_ [0] (
04:54:25 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:54:42 Join miepchen^schlaf [0] (
04:54:50 Quit dionoea_ (Remote closed the connection)
04:55:37 Join dionoea_ [0] (
04:56:06 Join gtkspert [0] (
04:56:54 Quit jurrie (Read error: 110 (Connection timed out))
04:58:35_jhMikeS_scorche: the only way I ever got the WMWare to work was to edit Rockbox.vmx manually since the memory message appeared before ever getting far enough to edit the settings. A suggestion might be added to set 'memsize = "xx"' in the .vmx file in case of that problem.
04:58:41 Nick _jhMikeS_ is now known as jhMikeS (n=jethead7@rockbox/developer/jhMikeS)
04:59:03 Quit linuxstb (Read error: 113 (No route to host))
04:59:33scorchejhMikeS: that is what i was referring to all along...see the forum post i linked
05:00:58jhMikeSI see. Couldn't that be on the wiki page though?
05:01:54*jhMikeS 's brain may be out of date on this one anyway is a wiki, after all....but i am eating atm
05:05:55 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
05:06:12 Quit jcollie ("Ex-Chat")
05:06:35 Join jcollie [0] (
05:07:00jhMikeSI'm not even sure where to put it except to have a small "Troubleshooting" section
05:07:53 Join abescully [0] (
05:08:16jhMikeSah, perhaps there is one sort of
05:09:09 Join ol_schoola [0] (
05:10:29 Quit gtkspert_ (Read error: 101 (Network is unreachable))
05:12:06 Join dionoea__ [0] (
05:12:31 Quit dionoea_ (Read error: 104 (Connection reset by peer))
05:15:09 Join rockho [0] (n=rocko@
05:19:33 Quit dionoea__ (Remote closed the connection)
05:20:08 Join dionoea_ [0] (
05:28:21 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
05:29:42 Join the [0] (n=63eef255@gateway/web/cgi-irc/
05:30:45 Quit dionoea_ (Remote closed the connection)
05:30:56 Quit the (Client Quit)
05:31:01 Join dionoea_ [0] (n=dionoea@
05:31:24 Nick kperri is now known as perrikwp (i=9821347f@gateway/web/ajax/
05:32:06 Part abescully ("Leaving")
05:32:23 Quit rockhoe (Read error: 110 (Connection timed out))
05:33:38 Quit Horscht ("We don't make mistakes, we just have happy little accidents")
05:37:49 Quit dionoea_ (Remote closed the connection)
05:37:56 Join dionoea_ [0] (
05:40:35 Quit nplus (Remote closed the connection)
05:43:33 Join webguest47 [0] (n=63eef255@gateway/web/cgi-irc/
05:43:51webguest47where can i get the rockbox source
05:44:23 Join gtkspert_ [0] (
05:44:50jhMikeSonly after signing an NDA :p
05:45:30webguest47oh... ya i knew that srry for the noobish question thx
05:46:10 Quit webguest47 (Client Quit)
05:54:14rockhowhat happened to my rockbox :(
05:54:41rockhoI updated the thing, and now my skins are gone .. and all my equilizer stuff is gone
05:57:04brainwreckmaybe go to the part where you choose which files will show
05:57:14brainwreckand select " show all "
05:57:26scorchehow did you update?
05:58:32 Quit gtkspert (Read error: 101 (Network is unreachable))
05:58:32 Quit LycoLoco (Read error: 104 (Connection reset by peer))
05:58:40 Join LycoLoco [0] (n=Lyco@
06:03:00rockhosame way I always do
06:03:03rockhoclick + drag
06:03:12rockhosometimes I do the other jazz
06:03:25rockhoI didn't do that though
06:05:59 Quit wisp_sh (Read error: 110 (Connection timed out))
06:15:50 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
06:16:00 Join Llorea1 [0] (n=DarkkOne@
06:16:08 Quit Llorean (Read error: 104 (Connection reset by peer))
06:16:16 Nick Llorea1 is now known as Llorean (n=DarkkOne@
06:19:07 Quit jhulst (Read error: 113 (No route to host))
06:20:48corevettewhere do last fm logs save in rockbox...and where do i upload them online?
06:22:54cars2corevette: I think there's a box that comes up when you start that. Let me see if I can find it.
06:24:27cool_walking_corevette: it's .scrobbler.log (or .scrobbler-timeless.log on players without a RTC) in the root of the player.
06:25:02cars2corevette: see also here:
06:26:30 Join Transience [0] (
06:27:11Transiencehey, can rockbox run on the 3rd gen iPod nano?
06:27:44scorchethe front page lists all devices rockbox supports
06:27:53advcomp2019Transience, nope.. look at the home page to see the support list
06:28:18Transiencealright, thanks anyway
06:28:28 Part Transience
06:28:31cars2Transience: but wait, it might be supported in the future.
06:29:19 Quit Neskaya ("Lost terminal")
06:33:04 Quit japc (Read error: 104 (Connection reset by peer))
06:35:12 Join Davide-NYC [0] (
06:35:52 Quit dionoea_ (Remote closed the connection)
06:36:58 Join dionoea_ [0] (
06:44:38***Saving seen data "./dancer.seen"
06:45:01 Quit brainwreck ()
06:45:37 Nick JdGordon|w is now known as JdGordon (i=cff4943f@gateway/web/ajax/
06:47:56 Quit perrikwp (" ajax IRC Client")
06:54:50 Join aliask [0] (n=chatzill@rockbox/developer/aliask)
06:57:48 Join perrikwp [0] (i=9821347f@gateway/web/ajax/
07:06:57 Join TheBashar [0] (
07:07:36 Join LycoLoco_ [0] (n=Lyco@
07:07:41 Quit LycoLoco (Read error: 104 (Connection reset by peer))
07:08:45TheBasharIs there anyone on that can grant write access to the wiki?
07:10:14aliaskTheBashar: What's your wiki name?
07:10:32TheBasharStephenAnderson - I'd like to upload a c250 battery benchmark.
07:11:13aliaskTheBashar: Done!
07:11:21TheBasharaliask: Thanks!
07:11:34 Join edsimon [0] (
07:17:10 Part edsimon
07:18:31 Join conando [0] (
07:21:40 Quit DrDnar (Read error: 110 (Connection timed out))
07:22:39 Join Weiss_ [0] (
07:22:45 Quit Weiss (Read error: 104 (Connection reset by peer))
07:29:05 Quit cars2 ("CGI:IRC (Ping timeout)")
07:32:09 Join wisp_sh [0] (
07:34:44 Join IcedDog [0] (n=blowme@
07:35:09 Join kushal_12_27_200 [0] (n=kushal@
07:37:29IcedDoganyone around?
07:38:35scorchenot at all
07:39:03IcedDogexcellent :D
07:39:11IcedDogI have an Archos Jukebox Recorder, and I just got a new 'puter with Vista, which doesn't recognize it...wondering if there are any drivers to fix the situation
07:39:12 Join tvelocity [0] (
07:39:37IcedDogtried the Archos site, but their driver package is from 2005
07:39:37scorchewhich model?
07:39:54IcedDogRecorder 20
07:41:40scorchehrm...i thought the 20s were all 2.0 and mass storage compliant...
07:41:57IcedDogyou would think so
07:42:15scorcheit is 2.0, yes?
07:42:33IcedDogI've replaced the HDD a couple times, but I used the same Hitachi model laptop HDDs
07:42:57scorcheif it is 2.0, it shouldn't need specific drivers
07:43:27IcedDogI've also got a couple USB 2.0 external HDDs that aren't recognized
07:43:50scorchethose tend to be msc too..
07:44:44IcedDogI networked this new computer to the old XP system and can access my old drives over the network, so at least I have a workaround, but it's still frustrating
07:45:33IcedDogI guess I'll have to hope SP1 fixes it next month
07:46:03kushal_12_27_200I need some help with recovery mode, please. I think I need to use terminal to paste the mi4 file on the player
07:46:05scorchewell, it should work just fine...i am not that knowledgeable with vista's quirks though
07:46:24scorchekushal_12_27_200: hrm?
07:46:27IcedDogthanks anyway
07:47:02amiconnscorche: Only the 20 GB model and *supposedly* some 15 GB models of the Recorder v1 are USB2.0 and mass storage compliant
07:47:12scorcheamiconn: i know
07:47:21scorche[22:39:30] <scorche> which model?
07:47:21scorche[22:39:47] <IcedDog> Recorder 20
07:47:47kushal_12_27_200scorche, I am not used to terminal. I need to get the firmware.mi4 in my desktop to my sandisk sansa c250's 16 MB drive
07:47:55scorcheand stop saying supposedly...i have told you a few times that i have a 15 that has 2.0
07:48:38amiconnThe question is: is it an original Recorder 20 or an upgraded 6 or 10?
07:48:58IcedDogoriginal 20
07:49:08amiconnAnd btw, the USB2.0 capability is printed onto the front plate
07:49:11IcedDogsays it right on the front plate :D
07:50:15amiconnThen it should definitely work on vista
07:50:30scorcheas i have said...
07:50:32IcedDogyeah, its puzzling
07:51:49IcedDogWTF, it just recognized it
07:52:04kushal_12_27_200what does sudo cp firmware.mi4 pribootLoader.rom /tmp/sansa do in terminal?
07:52:13IcedDogI tried it once a few days ago with the USB port on the front, no go
07:52:22IcedDogso now I just tried it with a rear port, voila
07:52:26kushal_12_27_200what will sudo cp firmware.mi4 /tmp/sansa do?
07:52:54scorchekushal_12_27_200: please dont repeat like that...
07:53:04amiconnIcedDog: In this case it might be a wiring problem of the front ports, and not vista's fault
07:54:03IcedDoggonna try the front port again
07:54:23 Join davidb_csh [0] (
07:55:03 Part TheBashar
07:55:39 Join hannesd [0] (
07:55:56IcedDogand it works
07:56:06IcedDogstrange, but I'll take it
07:56:18IcedDognow to try my external drives
07:57:00IcedDogya know what, when I first tried them, I'm not sure I had a net connection yet...maybe Vista needed an update or something
07:57:37 Quit Davide-NYC ("ChatZilla 0.9.81 [Firefox]")
07:59:16IcedDogsuccess for first USB external =)
08:04:03 Join Buschel [0] (
08:13:40Buschelamiconn/gevaerts: (for the logs) You've asked whether DCUD is used in the iPods. I can only tell for 5.5G -> yes. By default it is configured to 1.8V voltage supply. I tested to _lower_ it −− the iPod is still working at 1.5V, below it immediately crashes. The code section is within firmware/drivers/pcf50605.c -> pcf50605_write(DCUDC1, 0xe3); (=1.8V). To raise the voltage you'll have to set the 2nd parameters to 0xe4 (2.1V), 0xe5 (2.4V), 0xe6 (2
08:14:11amiconnSo it's used for converting down, not up?
08:14:25Buschelby default, yes.
08:14:36amiconnI'd think that's hard to tell, as this decision is done via a hardware pin, not software iiuc
08:15:30amiconnIf it would be used for converting up, 0xe3 would be 4.3V
08:15:53Buschel0xf3 = 4.3V
08:16:25amiconneh, sorry
08:16:32*amiconn cannot read :\
08:17:07 Quit IcedDog ()
08:17:13amiconnHmm, then I wonder how the 5V for USB are generated...
08:18:06Buschelit could be that this supply must be raised... we formerly did only detect USB and restarted −− so, there was no real long-time usage of the USB-controller. I also experienced some sporadic "USB detected, but no restart to EDM".
08:18:17 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
08:18:37BuschelBut I do not want to raise this value, it could damage the hardware...
08:18:46amiconnWhere are those voltages set (in rockbox)?
08:19:06Buschelfirmware/drivers/pcf50605.c -> pcf50605_write(DCUDC1, 0xe3);
08:22:46 Quit BigBambi (Remote closed the connection)
08:24:09GodEaterI can't believe we have to supply a full 5V back down the wire
08:27:36Buschelok, gotta go to work. see you in 12h or so :/
08:27:39 Quit Buschel ()
08:28:09amiconnWell, the USB signalling voltage is 5V (the *actual* signals are a little lower than that), so there must be 5V generated somewhere
08:28:50amiconnI now doubt that it's the DCUD converter of the PCF, because it's set to 1.8V by default and lowering it below 1.5V makes it crash.
08:30:15amiconnSo as the PCF has no other converter that can provide +5V, there must be a separate chip that does it.
08:30:47GodEaterexplain this to me like I'm really stupid for a second here. I have a USB mouse. It has no juice of it's own at ALL. So all the signals it's sending back to the host OS *must* be powered by the voltage coming FROM the host in the first place. Why are DAPS different ?
08:30:58LinusNi can't see why the device would have to supply any voltage at all
08:31:12GodEatersomeone else who thinks the same as me
08:31:15 Quit davidb_csh ("Leaving")
08:31:23amiconnThe mouse is bus powered
08:31:45GodEaterbut so is my iPod. It's taking power from the host to charge it's battery.
08:31:53LinusNhowever, the ipod would have to supply power when in host mode
08:32:01GodEaterI agree with that
08:32:05GodEaterbut we're not working on host mode.
08:32:44 Join Rob222241 [0] (
08:32:57 Join mrkiko [0] (
08:33:14mrkikoHi all!
08:33:27amiconnThe device has to drive its data output. We don't know how the USB interface logic is powered
08:33:35mrkikoI have a small question. I noticed that my player's hard disk is more and more rumorous when I use rockbox than when I run my OF.
08:33:50GodEaterrumourous ?
08:34:08mrkikorumorous -> noisy
08:34:18GodEaterreally? I've never noticed that.
08:34:31markunkushal_12_27_200: it will copy those files to that directory. What are you trying to do?
08:34:54 Join spiorf [0] (
08:35:10Lloreanmrkiko: What SVN revision are you using?
08:35:19mrkikoWhen I boot on rockbox the hard drive makes the noise that Windows XP makes when it boots on a pc. crcrcrcrcr4rr.... when I boot the OF the disk spins up and down normally but no seeking noise is made. I
08:35:34mrkikodon't know if this is important, but it seems. Partially I think the OF accesses the disk fastly tan rockbox
08:36:08mrkikoLlorean: one I downloaded a week ago. But this has been true from when I started to use rockbox on
08:36:15mrkikoI have not tried with the last LBA change
08:37:05mrkikoAniway this is strage, since rockbox and the OF performs caching the same way I think, in the sense that they load to RAM more than one file in the current folder at a time.
08:38:08Lloreanmrkiko: What type of MP3 player are you talking about, and how much testing have you done? Can you record a comparison of the two noises while caching the same playlist?
08:38:13mrkikoAniway this is a "stupid" thing, in the sense that I don't think it's vital for me :) the problem may be rockbox is using improperly (or at least differently) the hard disk hardware.
08:39:14mrkikoLlorean>: yes, with a little calm I will be able to d this thing. The problem is that the noise is not easy to record.
08:39:22GodEaterwe do so love wild speculation
08:39:39Lloreanmrkiko: You also didn't answer my question about what type of player you're referring to.
08:39:51 Join ender` [0] (
08:40:33mrkikoIf I get boring please someone tell me!
08:40:38mrkikoI'm using an hard disk with H340
08:42:25LloreanWe've got many H300 users and none have reported hearing something like this. I'd suggest trying to either put together instructions for reproducing it with an SVN build, recording a comparison of the OF and Rockbox buffering the same tracks, or preferably both.
08:44:43***Saving seen data "./dancer.seen"
08:50:27 Quit Rob2222 (Read error: 110 (Connection timed out))
08:52:33mrkikoI'l do it.
08:52:38mrkikoAniway ...
08:53:07mrkikoI can't do it while I'm on my classroom - I should do it at home
08:57:14mrkikoBut I have the problem that I can't deermine how much track the OF loads
08:59:10LloreanOnce the disk stops spinning, rapidly skip tracks until it starts spinning again, and you can approximate.
08:59:49mrkikoInfact, I'll do so
09:00:32 Quit cool_walking_ (Remote closed the connection)
09:00:43 Quit spiorf (Remote closed the connection)
09:02:45 Join petur [50] (n=petur@rockbox/developer/petur)
09:05:12 Quit Nevtus (Read error: 110 (Connection timed out))
09:07:20 Join Zagor [242] (
09:11:00 Join CaptainSquid [0] (
09:13:05 Join gibbon_ [0] (
09:13:14gibbon_hi, goot morning where applicable
09:13:22gibbon_(good even)
09:15:23 Quit JdGordon (" ajax IRC Client")
09:18:53peturmorning ;)
09:20:19gibbon_i have some kind of bug with my iPod mini 1st that i wanted to verify with someone in here before i file a report... but i just saw the svn commits of "tonight" and wanted to build a current build before i do so
09:21:00gibbon_its a "turning the unit on by plugging in USB and boot rockbox"-strangeness
09:22:20LloreanWhy not just download a current build?
09:22:56 Join homielowe [0] (
09:23:56gibbon_ok... if "unsupported builds are treated offtopic" applies to the rockbox usb stack, too i am willing to stop talking now ;)
09:24:16gibbon_Llorean: i guess this is still disabled in the current build, isn't it?
09:24:55LloreanAnd while it's okay to talk with the people working on it, you shouldn't file a bug report on it, but rather comment in the existing task for it.
09:25:06GodEaterI was just going to say that ;)
09:25:40gibbon_well that is right...
09:26:46*GodEater begins his high speed read stress test
09:27:17mrkikoGodEater: what tools are you using? dd?
09:29:21gibbon_ok, i can reproduce it even in the current build
09:38:30GodEaterso far it's going well - only one reset, and it recovered from that
09:39:30 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
09:39:56gibbon_the issue i have is that when you power up the iPod by plugging a USB-Cable in, after booting rockbox, the main menu appears and the whole gui freezes
09:40:19gibbon_USB works fine, the device is accessible...
09:40:20GodEaterthat's been happening on my ipod since forever
09:40:37gibbon_after unplugging, the gui stays frozen
09:40:48GodEateralthough before our own usb stack there was of course no access to the disk either
09:41:29gibbon_the only annoying thing is, that the shutdown timer is not disabled either...
09:41:39gibbon_GodEater: "glad" to hear that
09:42:13GodEaterreally ?
09:43:19gibbon_that means i am not the only one ...
09:43:23gibbon_(not more)
09:43:43pixelmarockho: still here? (just read the logs a bit - what exactly do you mean with "my skins are gone"?)
09:44:44 Join lst2 [0] (
09:45:23lst2Jeg elsker rockbox!
09:46:19gibbon_GodEater: so the only workaround is "just let the device boot up completely and THEN plugin USB"?
09:46:26GodEaterthat's what I do yes
09:46:49gibbon_hm, ok
09:48:29gibbon_GodEater: if you do not touch the device after it came up in that "broken" state... does it try to turn off, too?
09:48:42lst2Can I hook up a keyboard to rockbox?
09:49:31gibbon_in this case i think it would perhaps make sense to stop the shutdown timer if "charging via USB" is enabled and USB is connected
09:49:33GodEatergibbon_: I don't know - I've never left it alone that long
09:49:39GodEaterlst2: not currently no
09:50:37gibbon_GodEater: i noticed that just now... never left it alone that long before either
09:50:48lst2But it will be possible eventually? 8-)
09:51:38GodEaterlst2: if someone works on the code
09:51:44GodEaterlst2: which no-one is currently
09:52:15lst2What about a printer? :-P
09:53:01peturGodEater: that stuff requires usb-host anyway
09:53:35GodEaterlst2: why the hell would you want to hook up a printer ?
09:53:45gibbon_i was going to ask the same question...
09:53:58lst2Heh. Then I will work on it! (Eventually.) ;-D
09:54:46lst2To print text or pictures!
09:54:54GodEaterpetur: indeed. It will be a lot of code ;)
09:56:34gibbon_lst2: then start coding some printer drivers for picture prints :P those are really really model specific
09:56:42lst2A tv tuner would be book too!
09:58:00lst2Stupid T9! :-)
09:58:32GodEatersecond reset since I started the read test
10:00:31gibbon_GodEater: thats just for reading, right?
10:01:06GodEater"read test" :D
10:01:14GodEatersort of implies read only
10:01:42gibbon_yes ;) i hoped you'd say something like "yes, but the writing test comes up next" :P
10:02:29GodEatermy read test is going to last all day
10:03:03 Quit lst2 ("used jmIrc")
10:06:29gibbon_then i cross fingers for it to succeed and not reset much more
10:07:54GodEaterthat's not really a "success"
10:08:05GodEaterI'm just trying to gauge how bad the problem is on the iPod
10:08:26gibbon_ok, sorry... just a try to cheer you up
10:08:37GodEaterI got the impression from gevaerts that the issue is much worse on the sansa
10:08:41fyrestormsmells like love in the air
10:08:47GodEaterI'm not sad =/
10:10:00gibbon_if you need crosstesting and give me a few hints on your wanted test setup, i may crosstest on the iPod mini, iPod color (and hopefully sansa c200 today evening), too
10:11:23GodEaterjust hook up your ipod(s) to your computer, and use it as a hard drive all day
10:11:36GodEaterI'm playing back all the music on mine via a PC based player
10:11:46GodEaterto make sure the disk accesses are regular
10:13:42gibbon_so a continously repeated dd read would be ok, too?
10:15:02GodEaterI guess so
10:15:12GodEatera bit less like "realistic" usage though
10:17:15 Join syn4pse [0] (
10:18:44GodEaterdoes anyone know if you need to do anything special to get the manual to build now since the build order for languages changed ?
10:19:48GodEaterI'm getting a "No rule to make target `/build-manual/max_language_size.h' needed by '/build-manual/filetypes.o' error
10:19:55syn4psewhat is the procedure for submitting a plugin to the tracker? and should I bother if I only have confirmed the plugin works with a ipod video. it should work with any device with a screensize that large, and I can probably scale to slightly smaller in size
10:20:12GodEaterjust open a new task with type "patch"
10:20:16GodEaterand yes you should bother
10:20:46syn4psewhere do i open the task?
10:20:54syn4psei'm not too good with cvs/svn
10:21:04gibbon_in the flyspray i guess
10:21:16gibbon_not the svn
10:21:35pixelmaGodEater: did you use "make" or "make manual" - you would have to use the latter anyways now but I admit I haven't tried building it after the language size changes...
10:21:36GodEateryou're not touching svn syn4pse - just use the "Patches" link on any Rockbox web page to take you to the tracker
10:21:44syn4psecool, thx
10:21:46GodEaterpixelma: ah - I just did "make"
10:22:01 Quit mrkiko (Remote closed the connection)
10:22:17*GodEater tries again
10:22:36GodEaterI'm just going through changing all references to "File Menu" to "Context Menu" instead
10:23:10 Join lee-qid [0] (
10:25:36pixelmaI think it's good in general but is "Context Menu" alone enough? Just putting it up for discussion because it crossed my mind that we also have a "WPS Context Menu" etc. too...
10:26:16gibbon_but theres only one at a time isn't there?
10:28:35GodEaterpixelma: I think the words "Context menu" on their own are generally used to describe what used to the be the File Menu in the manual
10:29:07GodEaterand since it's what most people "in the know" use, it's probably what we should use in the manual to avoid confusion
10:29:18 Join cool_walking_ [0] (
10:29:42syn4pseGodEater: do I do a diff on the rockbox source tree to submit my patch?
10:30:03GodEatersyn4pse: did you use svn to checkout your copy of the source tree ?
10:30:12GodEaterthen do "svn diff"
10:30:24syn4pseholy crap i love svn
10:31:50GodEaterpixelma: bugger, I still can't build it - I think I must be missing some tex packages
10:31:55GodEaterbut I've no idea which one(s)
10:32:24pixelmabuilds fine here. Do you get an error message?
10:32:30gibbon_I'll try here, too
10:32:32 Quit dionoea_ (Read error: 110 (Connection timed out))
10:32:41syn4psedo i have to attach bitmaps
10:33:37gibbon_LaTeX Error: File `multirow.sty' not found. <−− thats the error for me... but its obvious ;)
10:33:51 Join dionoea_ [0] (
10:34:02GodEatersyn4pse: yes, attach the bitmaps too - ideally in a zip
10:34:09homieloweDoes anyone know if rockbox is supposed to compile nice with gcc 4.0.1 ? the iphone toolchain only seems to come in this version and doesn't play nice with libfaad and memcpy-arm
10:35:00syn4psegot it. basically, do a diff to fix the "config" files and attach the source and bitmaps in a zip. should I worry about paths in the zip file
10:35:01pixelmaGodEater: do you have the unicode package installed? The correct name for your distro should be on the ManualHowto wiki page...
10:35:15GodEaterpixelma: I'll take a look thanks
10:35:49pixelmahelp for the manual is always welcome :)
10:38:08 Quit CaptainSquid ("Miranda IM!")
10:38:15GodEaterpixelma: do you agree with what I state above about the context menu
10:38:47pixelmawhich reminds me that I wanted to ask Febs, now that he has a c200, if he'd like to write the description of the player for the blind, since I failed writing it nice and short (and Llorean didn't have time to do it so far) :)
10:39:38GodEatergood idea, it'll distract him from his usual job of abusing the forum users :)
10:40:42pixelmaGodEater: I think it's a better term than "File Menu" yes
10:40:42 Quit jhulst_ ("Konversation terminated!")
10:41:43 Join gtkspert [0] (
10:42:53gibbon_i really hope mine arrives soon, too
10:43:42gibbon_then i'd be happy to be of any help, too
10:44:45***Saving seen data "./dancer.seen"
10:46:22GodEaterpixelma: installed all those packages - it's still not building :(
10:47:01*GodEater tries a reinstall
10:47:26*gibbon_ gives the tetex-extra packages a try
10:48:04pixelmaGodEater: is that with your changes or plain SVN?
10:48:58GodEaterplain svn
10:49:09*pixelma tries with a "make clean" and reconfigure
10:49:22GodEaterit's bound to be something peculiar about my system
10:49:28GodEaterI don't believe the manual is broken
10:50:15 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
10:50:18pixelmait's not
10:50:37amiconnlatex-ucs is no longer needed, if you're using texlive (e.g. on a recent enough debian)
10:51:05amiconnI just uninstzalled it this morning (as synaptic displayed it as outdated), and the manual still builds fine
10:51:07pixelmaGodEater: I mean it doesn't seem to be a problem with your system
10:51:20GodEateryou mean you can't build the manual either ?
10:51:55pixelmaah, no. I got the same messages in the beginning but after a short while it continued building
10:53:35GodEateryou got the missing \begin{document} bit too ?
10:53:59 Quit gtkspert_ (Read error: 101 (Network is unreachable))
10:57:01 Quit wisp_sh (Read error: 110 (Connection timed out))
10:58:03pixelmahmm no... but generating features.tex takes quite a while
10:59:46GodEaterI'll try again. I've just completely removed all TeX packages, and am just installing the ones listed on the ManualHowto page
11:00:35 Join MethoS [0] (
11:00:54gibbon_after installing tetex-extra (which installs a whole bunch of texlive stuff) it builds a 141page pdf
11:01:42gibbon_make target manual-pdf exits faulty (error 1) anyway
11:03:09gibbon_except "overful hbox" warnings all over the document (that don't seem to hurt optically) i get only two warnings in the end of the latex run
11:04:23gibbon_i'll paste some logs
11:04:28pixelmayeah, me too (it did that for ages so I assume it's "normal")
11:07:11gibbon_ for the interested
11:07:37gibbon_thats not very good for && chained shell scripts but then i guess it can't be helped
11:12:04 Join OlivierBorowski [0] (
11:12:44gibbon_[ 8786.628000] usb 1-5: new high speed USB device using ehci_hcd and address 13
11:12:52gibbon_at least i got the right define ;)
11:13:21pixelmagibbon_: I don't get the three messages at the end, just a "make: LaTeX auxiliary files did not change (processing is complete)" at the end. Did you compile with "make manual"?
11:14:11gibbon_yes i did
11:14:29syn4pseGodEater: ok i put it up. i forgot to compress my .c file but it's only 34k
11:15:01syn4psei hope you guys like it! i gotta run.. thanks for all your help, today and otherwise.
11:15:07 Quit syn4pse ("Time wasted on IRC: 59 minutes 38 seconds")
11:15:45gibbon_GodEater: at least the rockbox highspeed stuff is a good 600kb/s faster than the mini.1st firmware diskmode
11:18:18gibbon_pixelma: pixelma i'll try it again in a completely clean configured folder
11:18:50 Quit TMM (Read error: 110 (Connection timed out))
11:19:53 Join Jeton [0] (n=4f7ebeb1@gateway/web/cgi-irc/
11:21:25JetonI've got this update on Google Reader from the Major Changes RSS: Enhancement, improvement or optimization: Major USB stack improvements. File transfer is now supported. But when to go to i don't see that news item listed.
11:22:19JetonDoes file transfer work ? I've tried on my Sansa with the latest build and when i plug the USB cable, sometimes the player freezes with the USB icon on display. I have to do a Hard Shutdown.
11:22:35JetonThis is probably reported (?)
11:23:13GodEaterpixelma: This is the complete build log ->
11:24:06GodEaterJeton: the MajorChange entry was premature, and was removed yesterday. File transfer can be enabled if you build your own rockbox, but it's not turned on by default, and therefore is not in the current builds available for download.
11:25:31JetonThanks for the explanation. Since i don't know how to build my own rockbox, does anyone has a link to a build with USB turned on?
11:26:02pixelmaGodEater: so you can build it now?
11:26:05GodEaterno - we're not providing them to end users for a reason
11:26:12GodEaterpixelma: no - that finishes with an error
11:26:17GodEatersee the bottom of the log
11:26:56GodEaterJeton: there are still problems with the file transfer mode - which is why it's not enabled by default yet.
11:27:10GodEaterwhen they're fixed, it'll go into the regular builds
11:27:57JetonI'll wait then . :) Thanks,.
11:30:14amiconnGodEater: Your manual build log looks perfectly ok... "Output written on rockbox-build.pdf (152 pages, 1041008 bytes)."
11:31:12pixelmayeah, wondered the same
11:32:29GodEaterI have to hit enter in the middle of the build process after the missing \begin{document} message
11:32:31GodEaterwhich cant' be normal
11:32:49GodEaterand the last four lines I get are :
11:32:55gibbon_its the same for me here with a clean dir
11:32:59GodEaterOutput written on rockbox-build.pdf (151 pages, 993812 bytes).
11:33:00GodEaterTranscript written on rockbox-build.log.
11:33:00GodEatermake[2]: *** [rockbox-build.aux] Error 1
11:33:00GodEatermake[1]: *** [manual-pdf] Error 2
11:33:00GodEatermake: *** [manual-pdf] Error 2
11:33:05GodEaterthat doesn't look successful to me
11:33:23gibbon_same here same here *jumpy*
11:35:32gibbon_after the first build ran through i don't have to press enter anymore but the make error is still there
11:36:13gibbon_but theres nothing unusual outputed on stderr
11:36:28 Quit Jeton ("CGI:IRC (EOF)")
11:39:58pixelmawell yeah, there might be small differences... building under cygwin here and that uses oldre tex packages. But I had to press enter once here too (after the make clean and reconfigure), that was new to me too.
11:40:23gibbon_on a make manual-html this occurs twice
11:41:11gibbon_i read from the output, that latex misses the begin document markup in the features.tex... but i read that in the backlog, too ;)
11:41:51gibbon_ok... make "twice" a "three times"
11:42:54amiconnGodEater: Here's my build log: That's on current debian unstable. No enter pressing necessary, and no make error either. Maybe comparing yours with that helps...
11:43:11pixelmafeatures.tex is generated (from features.txt) so maybe that needs some extra foo. That's out of my scope though
11:44:10gibbon_ thats the tail of a make manual-html ...
11:46:32GodEateramiconn: sorry, that's beyond me - I'm not a TeX expert. Someone else will have to figure it out, and I'll do without building my own manuals in the meantime
11:59:21 Join midkay_ [0] (
12:04:41 Quit midkay (Read error: 110 (Connection timed out))
12:05:13gibbon_even if the iPod should charge when attached to usb, it currently discharges while doind the read tests
12:05:58pixelmaGodEater: it's a pity that you can't get it to work (and I have no idea what it could be). Maybe try catching n1s or bluebrother when they are around here, they know much mor than I do about it
12:06:51GodEateryeah I'll ask one of them when they show up
12:07:17GodEaterI could still submit the patch - but I'd like to have checked it worked first =/
12:08:10 Join tvelocity_ [0] (n=tony@
12:10:53 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:14:07 Join barrywardell [0] (
12:14:08pixelmaGodEater: I'd say that you could still submit it (maybe add a note that it's not tested). Fixing the one obligatory syntax error is usually not that hard (but for me the actual writing and phrasing is)... :)
12:14:40gibbon_GodEater: first error in my read test after about 3000 seconds
12:14:43GodEaterheh - all I did was a recursive search for "Filemenu" and "File Menu" and replace them with "Contextmenu" and "Context Menu" respectively :)
12:14:56gibbon_GodEater: took the device 80-100 seconds to reappear in the os again
12:15:07GodEaterthat's long
12:15:11GodEateronly atakes about 10 here
12:15:21GodEaterwhich player, and which OS ?
12:15:48 Join BitTorment_ [0] (n=martin@
12:16:14gibbon_and it seems the device just powered down for a low battery (like i told, it doesn't seem to recharge)
12:17:11GodEaterI've had 21 resets now since I started the test this morning, not all of them have made an impact on my listening though (I guess because enough data is cached)
12:17:29gibbon_current svn build with defines -DUSE_ROCKBOX_USB -DUSE_HIGH_SPEED
12:17:40gibbon_on an iPod mini 1st
12:17:49gibbon_seems quite stable when doing sequential reads
12:18:00pixelmaGodEater: hmm... "Contextmenu" vs. "Context Menu"? Does it make sense to keep both ways of spelling (is there a reason behind it) or maybe it should be changed to one consistent spelling while at it?
12:18:38GodEaterpixelma: one is the reference used to shortcut between places in the manual (it's never rendered though), the other is the actual text which appears on the page.
12:18:54pixelmaah, I see
12:18:57GodEaterit didn't seem sensbile to only update the display text
12:19:22 Quit tvelocity (Read error: 110 (Connection timed out))
12:19:29 Quit MethoS ("Konversation terminated!")
12:19:32GodEaterI don't think you can have references with spaces in them
12:20:27 Join crope` [0] (
12:21:30 Join desowin [0] (n=desowin@atheme/developer/desowin)
12:23:04 Quit aliask ("ChatZilla 0.9.81 [Firefox]")
12:23:06gibbon_GodEater: 095737344 Bytes (4,1 GB) kopiert, 1819,27 Sekunden, 2,3 MB/s
12:23:22gibbon_GodEater: sorry...
12:23:49GodEaterthat's looking a lot worse than mine
12:23:54GodEaterwhich device did you say it was again ?
12:24:08 Quit OlivierBorowski (Remote closed the connection)
12:24:09gibbon_iPod mini 1st gen
12:24:24gibbon_GodEater: consider the disconnects as a result of battery drains
12:24:44GodEaterI guess if the battery is hosed though that would explain it
12:25:10gibbon_seems that rockbox fails to charge this target whith the highspeed usb enabled (it did with the full speed rockbox stack)
12:26:34 Join OlivierBorowski [0] (
12:27:03gibbon_the proof of definately using high speed mode is the line:
12:27:15gibbon_well... line 1 ;)
12:28:18gibbon_and the speed being faster than 12Mb/s of cause
12:28:50 Join NicolasMathias [0] (n=58a6dca1@gateway/web/cgi-irc/
12:31:41 Quit BitTorment (Read error: 110 (Connection timed out))
12:31:42GodEaterif you'd be so kind as to check it for me :)
12:32:50 Quit NicolasMathias (Client Quit)
12:35:45 Join Bagder_ [0] (
12:37:01pixelmaGodEater: the patch looks "inverted" to me
12:37:21GodEaterlet me sort that out
12:38:36GodEaterpixelma: try again now
12:38:45 Quit Bagder (Nick collision from services.)
12:38:51 Nick Bagder_ is now known as Bagder (
12:39:46 Join wisp_sh [0] (
12:40:08pixelmawill "diff −−git" work in my svn checkout?
12:41:11GodEaterpatch ignores that bit
12:41:37gevaertsgibbon_: so it reads for nearly an hour without any error or device reset at all ?
12:41:39 Quit homielowe (Read error: 110 (Connection timed out))
12:42:07gibbon_gevaerts: no obvious errors
12:42:29gibbon_i'd like to doublecheck that using md5... but changing the firmware will change the md5 as well
12:42:49gibbon_gevaerts: no kernel messages and a continous read performance of 2.3MB/s
12:42:59GodEatergevaerts: this is my output since I started the test this morning :
12:43:03gibbon_sequential though... using dd
12:43:28gibbon_never saw one of those
12:44:00gibbon_gevaerts: like said i had some power issues... ;)
12:44:33GodEaterI think the I/O error was me trying to update some ID3 tags
12:44:49***Saving seen data "./dancer.seen"
12:46:32gevaertsI get during reading
12:47:20gevaertsGodEater: in my experience, the resets are harmless, but I don't like the I/O error
12:47:44GodEaternor me
12:47:57gevaertsgibbon_: rebooting to EDM shoudln't change md5sums
12:48:09GodEaterI should remember to mount it r/o
12:48:38*GodEater decides to go get some lunch. Back in 15 or so.
12:53:57gevaertsgibbon_: Those power issues will eventually need to be solved, but I expect them to be unrelated to strictly the USB driver, so they are likely to be device-dependent, and since I don't have an ipod, I'd like to leave those to someone else
12:57:32gibbon_gevaerts: good point for the EDM... i'll try that
12:57:39gibbon_after recharging ;)
13:04:21FebsI'll put writing a description of the C200 on my list of things to do.
13:04:39Febs(^ pixelma)
13:06:21pixelmathanks :)
13:07:04 Join Thundercloud [0] (
13:09:56pixelmaGodEater: the manual builds fine with that patch but I went through a pdf, searching for all occurences of "context menu" and in some places it is also a bit confusing now (e.g. in the playlist catalog paragraph I wasn't sure which context menu was referred to)...
13:10:40pixelma...and in one place (ID3 info screen which is called "track info" now, IIRC) it even speaks about a "file context menu"
13:11:08GodEaterpixelma: I'll take a look then and see if I can tidy tha tu
13:11:13GodEater*that up
13:13:23gibbon_gevaerts: i am writing a shell script that will read in blocks of n times 512 bytes... this will output a hell lot of data, but gets me rid of the checksum problem
13:13:40gibbon_i don't know if booting rockbox changes anything on the disk...
13:16:26gevaertsgibbon_: if you have the database disabled, it shouldn't I think
13:16:29 Join DerDome [0] (
13:18:59amiconnGodEater, gibbon_: 2.3MB/s on read is what can be expected from the mini's microdrive. I get the same speed on mini G2, with both rockbox highspeed and apple diskmode
13:23:35gibbon_amiconn: i just realized that the apple diskmode on the 1st gen only does full speed ;)
13:24:02gibbon_so the speed gain i experienced isn't that spectacular *g*
13:24:37GodEaterpixelma: wrt the "file context menu" reference, I believe that should be changed to the new "Context Menu" nomenclature, since it seems to be referring to the same thing to me.
13:25:21pixelmayeah, my understanding as well
13:25:59 Quit Ramla (Read error: 104 (Connection reset by peer))
13:26:43GodEaterI'm looking for the playlist catalog paragraph you were also referring to now as well
13:29:08GodEaterto me, it reads pretty clearly
13:30:04 Join Ramla [0] (
13:32:28pixelmamaybe it was just because I jumped from one hit of the "context menu" search to the next and the ones before mostly talked about the WPS context menu and that confused me
13:34:10GodEaterI'm not really sure how to improve this
13:34:26GodEaterwe could change the reference from "File Menu" to "File Context Menu" everywhere
13:34:44GodEaterbut then that starts the confusion for newbies again, since we'd have to get everyone to refer to it as such
13:34:58GodEaterand you know what they say about teaching old dogs new tricks
13:35:14 Quit Zarggg (Connection reset by peer)
13:35:28amiconngibbon_: Eh? On mini G2, diskmode does highspeed...
13:35:47 Quit DerDome ("Leaving.")
13:37:02pixelmaGodEater: I'm not sure how it could be improved, either. Guess that's what I meant with "too many context menus" earlier but as the name suggests, those can be different depending on the context...
13:40:54pixelmaand maybe "Browser Context Menu" would fit even a bit better than "File Context Menu" because it can be used on directories as well? If that should be used at all...
13:41:57amiconnThe database also has a browser...
13:42:26gibbon_amiconn: may be tahts another difference in the two generations
13:42:28GodEaterdoes a long select on a list from the database bring up the same menu ?
13:43:08gibbon_my device currently is in diskmode and the kernel sais: [15137.480000] usb 4-1: new full speed USB device using uhci_hcd and address 7
13:44:37pixelmaGodEater: just tried, it seems to be basically the same but only the playlist option without the operation that you could do with files/directories (rename, cut, copy etc.)
13:46:15pixelmamaybe "context menu" will do though...
13:47:10GodEaterthe database section of the manual doesn't appear to mention that menu at all
13:48:32 Join CaptainSquid [0] (
13:53:50 Join [omni]_ [0] (
13:55:56barrywardellgevaerts: testing your latest patch with high speed here. works great on H10 ( , not so great on e280 (
13:56:00 Quit perrikwp (
13:56:00 Quit andrew__ (
13:56:00 Quit micols (
13:56:00 Quit zedd_D1abl1 (
13:56:00 Quit Seed (
13:56:00 Quit magicthree (
13:56:00 Quit krazykit (
13:56:00 Quit lastebil (
13:56:00 Quit gibbon_ (
13:56:51NJoingibbon_ [0] (
13:56:51NJoinperrikwp [0] (i=9821347f@gateway/web/ajax/
13:56:51NJoinandrew__ [0] (
13:56:51NJoinzedd_D1abl1 [0] (
13:56:51NJoinmicols [0] (
13:56:51NJoinSeed [0] (
13:56:51NJoinlastebil [0] (
13:56:51NJoinmagicthree [0] (n=Banana@
13:56:51NJoinkrazykit [0] (n=kkit@
13:58:09gibbon_shell script done
13:59:18gibbon_even seems to work... its one huge line... read only code
14:00:32gibbon_creates a file with checksums and block ids...
14:00:55gibbon_so one can distinguish changed areas from a total read fsckup
14:01:15gibbon_there are some variables at the beginning of the "line" ;)
14:01:30gibbon_and as always: no warranty for life and data
14:01:33 Join dannyclyro [0] (n=4e69b1c3@gateway/web/cgi-irc/
14:02:36dannyclyrohello, wonder if anyone can help me. Trying to download a daily build for the iRiver H120, but when Leopard tries to extract it, there
14:02:41dannyclyrosis nothing there
14:03:34gibbon_the folder that is in the archive is - in unix terms - a hidden folder (.rockbox by name)
14:03:45gibbon_you might not see it in the finder, even if its there...
14:03:49GodEaterwhich means you won't be able to see it's content
14:04:02GodEaterunless you configure finder to display hidden folders
14:05:09GodEaterenter "defaults write AppleShowAllFiles YES" in a terminal apparently to enable that behaviour
14:06:15dannyclyroah, that makes sense! Except the terminal bit, where would I find that? Also i
14:06:24 Quit netmasta10bt (Read error: 110 (Connection timed out))
14:06:35 Quit [omni] (Read error: 110 (Connection timed out))
14:06:42dannyclyroit's the wife's mac, so i'd need to switch it back
14:07:40gibbon_i don't know what it is in locale EN but DE is "Dienstprogramme" unter applications
14:08:03gibbon_its a subfolder of you applications folder... i think the disk tool also is there
14:08:23gibbon_correct me if i am wrong... its been a while with macos
14:10:15GodEaterI imagine switching it back is as simple as "defaults write AppleShowAllFiles NO"
14:12:12dannyclyro thank you, found terminal and did it in there
14:12:19dannyclyrothank you very much :)
14:12:24GodEaterno problem
14:12:26GodEaterdid it work? ;)
14:15:17gibbon_.oO( lets not follow this anymore ;) )
14:15:41gibbon_oh sorry.. strange formatting forbidden in here :/
14:16:10*gibbon_ is going to try the blockwise checksuming on his iPod
14:18:35 Join nplus [0] (n=nplus@141.25.Globcom.Net)
14:18:46GodEaterquestion: *why* on earth did you write that all as one line ?
14:18:53GodEaterperverse pleasure ?
14:21:22gibbon_at first: yes...
14:21:33gibbon_second: historical groth in 5 minutes
14:22:01dannyclyrothanks guys, good to know how to do that for future reference.
14:22:21*gevaerts won't try to read that
14:23:15gibbon_i will format that after doing some work here ;)
14:23:16dannyclyrowas trying to get my scrobbler-timeless log to upload using a web-based scrobbler
14:23:39 Quit lee-qid (Read error: 110 (Connection timed out))
14:24:31 Quit jcollie ("Ex-Chat")
14:24:52dannyclyroanyone heard of someone getting an iRiver to upload their log via a mac?
14:25:26GodEateryou mean you want a mac client that understands rockbox's .scrobbler.log format ?
14:25:43 Join Stabhappy [0] (
14:25:43peturit would be nioce to get a qtscrobbler build for mac, somebody ought to build it...
14:25:48StabhappyHey guys
14:26:05 Nick Weiss_ is now known as Weiss (
14:26:41StabhappyGot a quick question for any of you - is there a way to specify a directory for rockbox to use as a music folder? Using the ipod as a harddrive aswell means that installed things music files get thrown into my database.
14:27:09GodEaterStabhappy: you want to investigate the database.ignore file
14:27:22fyrestormStabhappy: use teh database within rockbox
14:27:30GodEaterdannyclyro: <−− any good to you ?
14:28:13GodEaterfyrestorm: he is - he means the rockbox database is trying to add non-music files to his database (iiuc)
14:28:32StabhappyMhmm. My database.ignore file is completley clean.
14:28:32peturGodEater: can iScrobbler upload our log?
14:28:43GodEaterpetur: well I believe it's not "our" log format actually
14:28:59GodEaterit's a format recommend for embedded players
14:29:10GodEaterwe certainly use their naming convention for it
14:29:11 Join webguest68 [0] (n=53ed3db7@gateway/web/cgi-irc/
14:29:13peturStabhappy: it *is* an empty file, put it in a dir it should ignore
14:29:26Stabhappymakes a lot more sence now.
14:29:35Stabhappymmm weird way of doing it though
14:29:44Stabhappya file would make a bit more sence
14:29:55Stabhappybut it'll do - cheers
14:29:59GodEaternot really
14:30:08GodEatersince rockbox is designed to assume everything is music
14:30:09Stabhappywell most people only have one /music/ folder or the like
14:30:16Stabhappyyeah I guess
14:30:22GodEatera lot of people don't have a music folder at all
14:30:29fyrestormmaybe it should come default with .rockbox in it?
14:30:43GodEaterfyrestorm: that's not how it works
14:31:25StabhappyDoes the ignore file stop sub directories aswell? As in, if I put it in \notmusic\ for example, will it ignore that and \notmusic\a\
14:31:47GodEaterStabhappy: yes it does
14:31:47webguest68hello!i am porting rockbox to meizu m6 and i need some help.we need to dissembely arm code and know something about players hardware.but no one from us understands arm language.could you rea that code to tell us something new?
14:31:58Stabhappyexcellent! that's brilliant news. Cheers.
14:32:26fyrestormthis guy keeps coming in with the same line
14:32:28StabhappyCan't thank you enough. I'm out
14:32:34GodEaterwebguest68: you really need to start learning ARM assembly yourself, or find another meizu owner that already does.
14:32:55 Quit Stabhappy ("( :: NoNameScript 4.2 :: )")
14:33:05*GodEater likes the easy questions :)
14:34:08GodEaterpetur: I may take that statement back. I could have *sworn* I read about the file format on the site, only now I can't find it
14:34:09 Quit dannyclyro ("CGI:IRC (EOF)")
14:36:49gibbon_ <−− you may try to read that
14:38:17gibbon_i threw in some newlines and comments
14:39:19 Quit webguest68 ("CGI:IRC")
14:39:22 Join webguest62 [0] (n=53ed3db7@gateway/web/cgi-irc/
14:39:34webguest62markun iam on irc
14:39:51webguest62markun i am on irc
14:39:57GodEaterwebguest62: we can see that
14:40:44 Join dannyclyro [0] (n=4e69b1c3@gateway/web/cgi-irc/
14:41:40dannyclyrogodeater, tried the iscrobbler program but it seems it only scrobbles from your itunes library, not from a non-apple mp3 player
14:41:50dannyclyroon rockbox
14:41:53*gevaerts likes his sd-card, so you can just write lots of data and read it back. Much easier to test
14:43:05GodEaterdannyclyro: I am at a loss then. I don't know what you can use from a Mac.
14:43:22gibbon_gevaerts: you are right with that ;)
14:44:04dannyclyrothank you anyway, was worth a pop :)
14:44:50***Saving seen data "./dancer.seen"
14:45:39GodEaterdannyclyro: ?
14:47:41GodEaterI just used it to submit my own .scrobbler.log - seemed to work ok
14:49:04*GodEater find it's irritating the refuse to accept submissions dated earlier than your last submission though
14:49:12markunwebguest62: hi!
14:49:20markunI'll PM you
14:50:05webguest62<markun> hi
14:50:18gibbon_i am now reading the ipod harddrive using the rockbox highspeed usb stack
14:50:32 Quit goffa_ (Remote closed the connection)
14:50:45*GodEater has been doing that since 8:25 AM GMT this morning
14:50:46webguest62<markun> what is PM?(sorry)
14:50:54GodEaterPrivate Message
14:50:59markunwebguest62: can you see the message I wrote you in the other screen? Not sure how the webclient works
14:51:14GodEatermarkun: it opens another tab at the bottom of the screen usually
14:51:22GodEaterit doesn't dump you into it though - you have to select it
14:51:38GodEaterdon't PM's not work for unregistered users on freenode though ?
14:51:41dannyclyrogodeater: that was actually the one I was trying to use earlier when I first asked about hidden files.
14:51:54webguest62<markun> on what screen7
14:51:59markunGodEater: I've setup my account to accept non registered users
14:52:01GodEaterdid it not work for you then dannyclyro ?
14:52:15GodEatermarkun: yes - but I'm not sure they can see *your* messages by default
14:52:19markunwebguest62: in this browser screen there should be a tab at the bottom.
14:52:19dannyclyrowhen I hit browse on that uploader, it can't see the scrobbler-timeless log as it has a . in front of it
14:52:24dannyclyroeven with hidden files enabled
14:52:32GodEaterjust type the name in manually ?
14:52:45webguest62<markun> yes there is
14:52:50markunwebguest62: click on it
14:52:59GodEaterit's usually /Volumes/<your ipod name/.scrobbler.log
14:53:12*GodEater struggles to remember details of OSX
14:53:33webguest62<markun>on markun or rockbox or on status
14:53:45markunon "markun"
14:56:41 Quit kushal_12_27_200 ("This computer has gone to sleep")
14:57:23dannyclyrogodeater: you are my hero
14:57:29dannyclyrothank you so much :D
14:57:36*GodEater swirls his cape
14:59:49dannyclyrothat's awesome, can finally start scrobbling again.
15:00:06dannyclyrocya, thanks for the help!
15:00:38GodEateris there a version of firefox for Macs ?
15:01:03dannyclyroim using it just now
15:01:28 Quit dannyclyro ("CGI:IRC")
15:01:49*GodEater wonders how you make OSX dialog boxes obey the hidden files setting
15:01:52gibbon_not anymore i guess...
15:01:56 Join goffa [0] (n=goffa@
15:01:56GodEatersomewhat late though :D
15:02:04 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
15:02:32gibbon_gnaa... wound that ipod PLEASE stop draining the battery like a thirsty alcoholic?
15:02:34LambdaCalculus37Morning, everyone!
15:03:34gibbon_hi LambdaCalculus37 ... thanking god for tab completion
15:04:46krazykitgibbon_, it shouldn't do that anymore, assuming you're using a newer build :)
15:05:24GodEaterhe's pretty up to date - he's using the High speed UMS code...
15:05:25gibbon_i am not ;)
15:05:46gibbon_i used the build from 8:21 GMT that i made myself
15:06:20LambdaCalculus37I haven't been able to update my SVN tree for a bit, but I'm still going to give SOME kind of performance test.
15:06:33amiconnkrazykit: The power consumption tweaks aren't done for the minis yet
15:06:37GodEaterdownload the current code tree ?
15:06:39gibbon_well i am not using a newer build than thif morning... and i AM using the highspeed code
15:06:43GodEaterif you can't get access to svn
15:07:00krazykitah, the mini.
15:07:39gibbon_krazykit: yes... a solvershiny mini g1... which passed away and is now doind highspeed usb tests using a "stolen" battery
15:07:50gibbon_(from a friends broken mini..)
15:09:54markunmorning LambdaCalculus37!
15:14:08LambdaCalculus37markun: Morning!
15:15:46 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
15:17:23LambdaCalculus37GodEater: Well, I can at least connect my iPod color to my laptop and establish a proper connection. I'll try the USB code with that iPod for now.
15:18:53 Join n1s [0] (
15:18:55gibbon_in a few seconds, i have usb2.0 statistics of the mini g1 handy :)
15:22:00gibbon_hmmm 2 blocks differ
15:22:55gibbon_but they differ using the edm, too... its in the ext3 partition of the ipodlinux which puts mont information in there on mount
15:24:21 Join BitTorment__ [0] (n=martin@
15:26:33peturGodEater: *does* accept submissions out of order if you use their new protocol. The site you tried probably uses the old protocol. When qtscrobbler switched to the new protocol, life became much easier ;)
15:26:43gibbon_GodEater: the results:
15:28:08 Quit Febs (Read error: 110 (Connection timed out))
15:28:28 Join Arathis [0] (
15:29:41 Quit goffa (Remote closed the connection)
15:30:17 Join goffa [0] (n=goffa@
15:30:31gevaertsgibbon_: Looks good. Now some write tests ?
15:31:26gibbon_gevaerts: i'd like to but the batteries don't allow that... they discharge in rockbox
15:31:36gibbon_even with charging on USB enabled
15:31:52gibbon_the last read test drained them to a critical state
15:33:53gibbon_i am currently crosschecking with the apple disk mode (while recharging)
15:34:16gibbon_this evening when back home i can try to deliver the same tests for the iPod color
15:35:37 Join goffa_ [0] (n=goffa@
15:37:54 Nick dionoea_ is now known as dionoea (
15:38:10gevaertsgibbon_: I'd still be interested in write results later if you could manage them. Even if the battery forces you to keep them short, results would be useful
15:38:35 Join MajorC [0] (
15:39:52 Quit webguest62 ("CGI:IRC")
15:41:38gibbon_gevaerts: i will.... if writing a file to the device and reading it back will suffice, it should be possible for at lease 200MB now
15:41:51 Quit BitTorment_ (Read error: 110 (Connection timed out))
15:42:25LambdaCalculus37gaveaerts: Just tested read performance on my iPod color.
15:42:35LambdaCalculus37Results are up:
15:43:12pixelmadionoea: around?
15:43:18gevaertsgibbon_: The more the better of course. I'm especially interested in dmesg errors.
15:44:26gibbon_gevaerts: i hope we don't get any :)
15:45:29gevaertsLambdaCalculus37: any chance of some test_disk performance numbers as well ? While the OF numbers are interesting, realistically the test_disk performance is the real limit (unless someone improves the disk driver)
15:45:46gevaertsgibbon_: I hope so too, but I'm not too certain
15:46:40gibbon_i windered before... there seems to be a test suite in the rockbox source... how is it built?
15:46:43 Quit goffa (Read error: 110 (Connection timed out))
15:47:09dionoeapixelma: ?
15:47:21gibbon_respectively is it already there and i have just not seen it?
15:47:43 Quit nplus (Read error: 104 (Connection reset by peer))
15:48:08pixelmadionoea: you were the one who worked on how the "move" is displayed in the playlist viewer, IIRC?
15:48:24gevaertsgibbon_: edit apps/plugins/SOURCES, and add test_disk.c (or any other test_* you want)
15:48:34 Join nplus [0] (n=nplus@141.25.Globcom.Net)
15:48:38dionoeapixelma: yes
15:49:13gibbon_ah, thanks... are there destructive tests?
15:49:50gevaertsgibbon_: I don't know. At least test_disk is non-destructive (but requires enough space)
15:51:11LambdaCalculus37gevaerts: I'll perform the write performance test later on today.
15:53:41pixelmadionoea: I noticed that there is something weird on my Ondio - seems to be ok on the Sansa. It looks like an off by one problem - the first move is ok, the second move will display the "moving track" wrong - it shows the next track's name, it takes the right one but when I'm at the final place it'll insert it one track above (from what I would expect looking at it). Each additional move the track shown as moving is one more off but inserting is al
15:54:10 Quit linuxstb_ (Remote closed the connection)
15:54:17gevaertsLambdaCalculus37: I added a note about it on the wiki page as well
15:55:09 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
15:56:31LambdaCalculus37gevaerts: All right. I'll compile again with test_disk enabled and try it out.
15:56:53LambdaCalculus37But first, I'm going to test using GodEater's usb_write_test_file and get those results.
15:56:55dionoeapixelma: that line was too long so it got cut of after "intersting is al"
15:57:21dionoeaI'll have a look when i get back from work
15:57:23pixelma...but inserting is always one above (if I see correctly, it's really confusing)
15:58:30 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
15:59:41 Join major_works [0] (
15:59:54 Quit major_works (Client Quit)
16:00:10 Quit Zagor ("Client exiting")
16:00:16 Join Febs_ [0] (n=chatzill@
16:00:19 Nick Febs_ is now known as Febs (n=chatzill@
16:01:54pixelmadionoea: thanks
16:06:01 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
16:09:19gibbon_starting 512MB write tests using HS usb on mini g1 now
16:09:57 Join mf0102 [0] (n=michi@
16:09:59 Quit goffa_ (Remote closed the connection)
16:10:03*Nico_P thinks he's reached rather nice skipping behaviour
16:13:28Nico_Panyone want to try a patch?
16:14:11gibbon_can you describe the nature of "nice" in your context for me? then i'll decide :)
16:14:33Nico_PI mean responsive and hopefully robust
16:15:20Nico_PI'm still testing and trying to improve it but to me it seems to be already quite an improvement
16:19:35 Quit nplus (Read error: 104 (Connection reset by peer))
16:20:26n1sNico_P: I've been experiencing a nasty problem where music stutters right before track changes, like music playing, short pause ~0,5s, same track resumes ~1s, new track begins...
16:20:56Nico_Pn1s: on manual track skips?
16:21:13n1sno when it is auto switching
16:21:22 Join goffa [0] (n=goffa@
16:21:36n1snot on all track changes though...
16:22:49 Quit MajorC ()
16:23:44gibbon_i never had that problem
16:23:50 Quit Seed (Nick collision from services.)
16:23:55gibbon_extremely sort gap before crossfading though... not very annoying
16:23:57 Join Seed [0] (
16:25:36 Quit perrikwp (" ajax IRC Client")
16:25:48pondlifeNico_P: I'll test
16:27:07pondlifeNico_P: I just posted you a small parcel btw
16:27:13Nico_Pthanks :)
16:27:42Nico_Pdid you repair your bike?
16:27:58pondlifeNo, I found a way to buy postage online
16:28:14pondlifeHopefully the packaging will be good enough to protect it
16:28:16gibbon_gevaerts: no dmesg errrs so far after writing 512MB ...
16:28:30gibbon_its a bit slower than apple disk mode, but it seems to work fine for me
16:29:34 Join perrikwp [0] (i=9821347f@gateway/web/ajax/
16:29:34Nico_Ppondlife: the main benefit is that it should be more responsive
16:29:50pondlifeThat's good, but stability is most important
16:30:04Nico_Phopefully it's more stable too :)
16:30:37pondlifeDid you see the forum report of out-of-sync WPS after much skipping?
16:30:45Nico_Pnow the current and next track info in the WPS are updated at the same time
16:30:58Nico_Pyes. I'm trying to correct that too
16:31:06gevaertsgibbon_: That looks good. Do you happen to have a high speed hub handy ? Any chance of doing something similar while there are also busy transfers to another device on the same hub ?
16:31:42gibbon_not at the moment... i could test with another device on the same root hub if that helps
16:31:53Nico_Ppondlife: actually I was thinking of FS #8320... what's the forum thread?
16:32:04pondlifeNico_P: Surely there's only one track transition, so the current + next info should update at the same point....
16:32:10 Join mchua_ [0] (
16:32:17gibbon_gevaerts: but don't be too happy yet, the md5 check is still to come
16:32:31Nico_Ppondlife: it doesn't in svn
16:32:45pondlifeWell, it should!
16:32:51Nico_PI agree :)
16:33:06barrywardellgevaerts: I've been testing your latest patch with High speed on my H10. I've copied ~1GB (3x350MB files) H10->PC->H10->PC->H10->PC without any errors in dmesg and with md5sums matching. Unmounting between each copy operation, of course.
16:33:06pondlifeNico_P: I'm just looking, might be in somewhere
16:33:10 Join toffe82 [0] (
16:33:17gevaertsgibbon_: root hub might be useful as well, but a "normal" hub would be better.
16:33:30pondlifeNico_P: Or ?
16:33:35gibbon_gevaerts: looks pretty good in idle state
16:33:44gibbon_the md5sum matches
16:34:09barrywardellgevaerts: read speed of about 4.7MB/s
16:34:26Nico_Ppondlife: btw I didn't mention it's manual track skipping that should be improved.
16:34:37pondlifeAh, I assumed that anyway
16:34:39Nico_PI actually just nocticed that auto track skipping has a slight bug in the patch
16:34:45 Join Febs_ [0] (n=chatzill@
16:34:47gevaertsbarrywardell: any test_disk results available to compare ?
16:34:54pondlifeBut the next+current track info should always change together
16:35:07Nico_Pit does with the patch
16:35:39barrywardellgevaerts: I'll run it now...
16:36:20barrywardellgevaerts: this was on linux btw. osx is still a problem
16:37:02GodEaterpetur: I've had messages on my page both after using that page *and* QTScrobbler telling me a load of submissions were rejected because they occured earlier than my last submission.
16:37:31gibbon_gevaerts: 1,93MB/s using apple disk mode (write 512MB), 1,91MB/s using rockbox HSUSB (512MB write)
16:37:41gibbon_test_disk to come
16:37:43gevaertsbarrywardell: I think osx is most likely a scsi-implementation issue. I'll get to that soon
16:38:08pondlifeHmm, my database just killed itself
16:38:14toffe82jhMikeS: ping
16:38:21pondlifeOne minute it was working, now it wants to re-init
16:38:29peturGodEater: hmmm ever since using the last QTScrobbler I have not seen that (except last week, must have been a glitch because others complained too)
16:38:34toffe82jhMikeS: I got the user guide of the mc13783
16:38:50Slasheripondlife: hmm, disk errors?
16:38:58pondlifeI don't believe si
16:39:14Slasheriwas the db loaded to ram?
16:39:26GodEaterpetur: that's using 0.70 ?
16:39:37n1stoffe82: nice, is it safe to upload to the wiki?
16:39:39peturyes, but I think 0.6 was fine too
16:40:02pondlifeDebug info says: Yes-No-Yes-2152596/2184784-0%- −−- - 0-No-0
16:40:27pondlifeSlasheri: It was. I booted up and used the DB to select a random album
16:40:48pondlifeI skipped a few tracks, and went back to the browser but most tracks had vanished.
16:41:00GodEaterwhat do you need to do to get a test_disk plugin ?
16:41:09toffe82I ask it on the site of freescale and they send it with no restriction.. but in the document itself it is mentionned :
16:41:10toffe82Freescale Confidential Proprietary
16:41:12toffe82NDA Required / Preliminary
16:41:18pondlifeI went back to the main menu and tried to go back into Database, but no luck!
16:41:25barrywardellgevaerts: test_disk should run boosted or not?
16:41:30gevaertsGodEater: enable it in apps/plugins/SOURCES (i.e. add test_disk.c)
16:41:44Slasheripondlife: even reboot didnt help?
16:41:56pondlifeSlasheri: Will try.
16:42:07pondlifeI have Gather Runtime enabled, but Auto update disabled
16:42:29pondlifeReboot fixes it
16:42:38toffe82I don't think it is under nda now, they should have write it on the emial or asked me more information on why I wanted it
16:42:50gevaertsbarrywardell: boosted, for a fair comparison (the USB disk accesses run boosted), but unboosted would probably be useful as well (maybe we need a separate disk speed wiki page ?)
16:42:53Slasheriok, probably some kind of memory corruption then..
16:43:13pondlifeNico_P: Are you reading this... ;)
16:43:30n1stoffe82: better talk to bagder before putting it up on then
16:43:31Nico_Pyes, but I doubt it's my fault :)
16:44:02SlasheriNico_P: bte, are you planning to fix the runtime stats gathering soon? :)
16:44:06Nico_Ptoffe82: mail it to him (jhMikeS)!
16:44:17barrywardellgevaerts: I think there is already a disk speed wiki page, might be a bit out of date though
16:44:20*n1s would like a copy too :)
16:44:24pondlifeWell the skip is definitely faster
16:44:34LinusNtoffe82: i don't think we should put the data sheet in the wiki
16:44:53***Saving seen data "./dancer.seen"
16:44:59Nico_PSlasheri: yes I am. I've also been looking at changing the unbuffer events to track change events, but currently there is only one allowed ans we already have the scrobbler
16:45:17 Quit corevette (Remote closed the connection)
16:45:18Nico_Ptoffe82: I guess aliask wouldn't mind a copy either
16:45:31Nico_Ppondlife: cool :) what player are you testing on?
16:45:37Nico_Pany negative side effects?
16:45:42pondlifeThe track displayed is wrong
16:45:49pondlifeMight be my tags though.
16:45:52SlasheriNico_P: great. hmm, i think the track change event would be the correct place to do that
16:46:26toffe82Nico_P: do you have is email?
16:47:18Nico_Ptoffe82: PM
16:47:21barrywardellgevaerts: in case it's any help, here's a logfdump of usb_core.c and usb_storage.c with my H10 on OSX:
16:47:22 Join Benjamin_L [0] (
16:47:22 Quit goffa (Remote closed the connection)
16:47:32 Part LinusN
16:47:56Benjamin_Lhow do I test latest usb high speed code, is latest svn enough or do I have to apply some patches?
16:48:08pondlifeNico_P: Seems that the database picked a mistagged album... sync seems fine now
16:48:22Nico_Ppondlife: you got me worried :)
16:48:47Nico_PI have managed to get the desync, but I think it was because of a queue overflow. I don't think I can do much about that
16:48:55n1stoffe82: could you mail it to me too if I pm you my addy?
16:48:59gevaertsbarrywardell: you seem to have a fast H10.
16:49:14toffe82nls : yes
16:49:24*GodEater would like a copy too
16:49:42barrywardellgevaerts: mine is a 'big' H10. it doesn't have a microdrive :0
16:49:49toffe82how much do you pay for this ;)
16:49:56*Nico_P wouldn't mind the datasheet either
16:50:18*gevaerts will go through his irc logs later today, and gather statistics and test results
16:50:31gibbon_gevaerts: test_disk with 4k: read: 1.2M/s write: 1M/s ... seems a bit odd
16:50:47gevaertsgibbon_: can you try boosting the CPU ?
16:50:51*n1s sends toffe82 beer
16:50:58toffe82GodEater: your email in a pm :)
16:51:08gibbon_gevaerts: i can... wait
16:51:18Nico_Ppondlife: I think I'll clean the patch up and commit it. it seems fine to me and the only problem I've seen probably happens in SVN too
16:51:30*LambdaCalculus37 is recompiling with test_disk.c enabled
16:51:33kslateris the bookmark feature known to only work some of the time?
16:51:41pondlifeNico_P: Go for it. What's that other problem?
16:51:43gevaertsBenjamin_L: latest svn, but you need to define USE_ROCKBOX_USB and USE_HIGH_SPEED. What device do you have ?
16:51:53 Quit Febs (Connection timed out)
16:52:04pondlifekslater: Bookmarks should work much better if you have a recent build.
16:52:30Nico_Ppondlife: the buffering thread seems to start buffering the wrong track, which means playback is delayed more than usual
16:52:32Benjamin_Lgevaerts: Sansa e260
16:52:37*LambdaCalculus37 forgot USB_ROCKBOX_USB!
16:52:50gibbon_LambdaCalculus37: hihi
16:52:50 Join goffa [0] (n=goffa@
16:52:57pondlifeUSE_ROCKBOX_USE ?
16:53:37*LambdaCalculus37 brain farted
16:53:52gevaertsBenjamin_L: ok. Be careful with it, there seems to be an issue with the sansas that occasionally causes write errors when writing over USB, especially at high speed. It's much rarer on e200 than on c200 though.
16:53:58kslaterpondlife: recent build meaning something newer than the last stable release?
16:54:35pondlifeMeaning r16410 or later IIRC
16:54:40Benjamin_Lgevaerts: when enabling test_disk in SOURCES, how do I use it then?
16:55:05pondlifekslater: Certainly before that alll bookmarking and resuming in general was flaky
16:55:08gibbon_gevaerts: C,512,A: 1982; W,512,A:2187; R,512,A:1592; C,4096,A:2639; W,4096,A:2868; R,4096,A:1751
16:55:16gibbon_looks still a bit odd but more realistic
16:55:42toffe82n1s, GodEater check your mail ;)
16:55:44gibbon_U values are 100-300 lower
16:55:44pondlifeNico_P: Skip to next dir doesn't seem to work with your patch
16:56:05kslaterpondlife: thanks. I really like that feature since I listen to a fair number of podcasts to/from work, but music mostly once I'm here.
16:56:31Nico_Ppondlife: ah, crap. I'll check it out
16:56:44gevaertsBenjamin_L: test_disk will appear in the plugins/apps menu, and give you the choice between a performance test and a verification test. Run the performance test. I'm mainly interested in test_disk output on the ipods and H10 though, I have a fair idea about them on sansa
16:57:03Benjamin_Lgevaerts: ok, all right then
16:57:14Nico_Pcould an ipod owner tell me how to skip to the next dir?
16:57:18gevaertsgibbon_: unalligned doesn't really matter. All USB disk access is aligned
16:57:27 Join jgarvey [0] (
16:57:35GodEaterhow does one tell the ata driver revision for reporting test_disk results ?
16:57:41pondlifeNico_P: Two things, maybe the same. (a) the key combo to skip to next/prev dir and (b) the auto-change dir option (with a natural track change)
16:57:59pondlife(a) seems to pause playback without setting paused state. (b) returns to the browser
16:58:19toffe82talking about Freescale (sorry for the french) : Toulouse. Confronté à une baisse de l'activité, le fabricant américain de puces électroniques réduit la voilure.Freescale sera au chômage une semaine
16:58:42 Join nplus [0] (
16:58:51 Quit CaptainSquid ("Miranda IM!")
16:59:14gibbon_gevaerts: ok... i didn't even know what A and U ment... now its obvious
16:59:24LambdaCalculus37GodEater: Good question. I was wondering that myself.
16:59:25Nico_Ppondlife: dir skip (with the combo) seems to be working fine on my gigabeat sim
16:59:36pondlifeNico_P: There's a universal law that states playback.c is always at least one use case away from working.
16:59:40Nico_Ppondlife: do you know the combo on an ipod?
16:59:56pondlifeIt's short right, then long right on H380
17:00:21pondlifeDo you have auto-change dir enabled? That might be needed for it to fail
17:00:24Nico_Pseems to be that too on the ipod
17:01:30pondlifeHmm, it just worked for me too.
17:01:31Nico_Poh I seem to intermittently get the problem on the ipod
17:01:46pondlifeYep, it's intermittent
17:01:57GodEaterI've worked it out
17:02:07GodEatertrickier for me though, I have to match a git hash to an svn rev
17:03:57LambdaCalculus37Upon running test_disk, I am figuring that it's measuring general speed, aligned read/write, and unaligned read/write.
17:04:21GodEateris amiconn around ?
17:04:22Nico_Ppondlife: are you sure this is new to the patch?
17:04:23gibbon_all work is done now... heading home
17:04:28pondlifeNico_P: No
17:04:34LambdaCalculus37And aligned read/write and unaligned read/write are in 512 and 4096 byte chunks.
17:04:35gibbon_gevaerts: all further tests are done when i'm home :)
17:04:37 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
17:04:48pondlifeI've only just initiated test mode
17:04:58pondlifeWhich is when I break stuff
17:05:00LambdaCalculus37Can someone confirm that this is indeed the case.
17:05:03 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
17:05:03GodEatergevaerts: surely the 512 byte reads / writes are not aligned on the 5.5G ipod ?
17:05:54LambdaCalculus37GodEater: This is on my 4G iPod color.
17:06:13GodEaterLambdaCalculus37: yes - but I'm running it on my 5.5G
17:06:27pondlifeNico_P: A natural transition from one dir to the next was working ok with SVN on Friday, it's definitely not working now
17:08:08pondlifeI wonder if all of the auto-change dir stuff should be handled as part of playlist generation, not in playback.c?
17:08:15LambdaCalculus37GodEater: Does test_disk log anything, perchance?
17:08:32GodEaterLambdaCalculus37: yes, it writes a log file in the route dir
17:08:40Nico_Ppondlife: it would be nice, yeah
17:08:56pondlifeA bit like a recursive option
17:08:58 Quit Benjamin_L ("Ex-Chat")
17:09:02Nico_Pit's handled by both actually. each does a part of the job
17:09:03LambdaCalculus37Okay, good. I'll fetch the logs once I finish performing the write & verify test.
17:09:18pondlifeNico_P: Well, that's probably the suckiest method
17:10:01amiconnLambdaCalculus37: The alignment in test_disk refers to in-memory 16 bit buffer alignment, as 16 bit is the ata port width
17:10:18amiconnAnd it tests 512 byte, 4KByte and 1MByte buffers
17:10:21*pondlife thinks that sorting out of the playlist code (separation from playback and browsers) would be a nice GSoC project, if not very exciting.
17:10:27LambdaCalculus37amiconn: Ahh, good to know. Thanks for that!
17:10:39Nico_Ppondlife: the manual dir skip seems to behave the same in svn
17:10:52pondlifeI was more concerned with the auto skip.
17:11:04Nico_Pyeah I'll enable that
17:11:06pondlifeaka I'm driving and want to hear more music
17:12:26Nico_Ppondlife: when you say "natural transition", do you let the music play to the end, or manually skip when on the last track of a dir?
17:12:28amiconnOn sansa and Ondio, the alignment should not influence transfer speed, because SD and MMC transfers are only 8 bit wide (as seen from the CPU, the _actual_ transfer is serial)
17:12:40pondlifeNico_P: Let it play to the end
17:13:06pondlifeIn case it's relevant my last track in the first dir was a silly long one (21 mins), so I fast forwarded much of it.
17:13:20LambdaCalculus37amiconn: Want me to try test_disk on my c240 later?
17:13:50Nico_Ppondlife: auto dir change just crashed on me :'
17:14:03Nico_Pwith svn code
17:14:03GodEaterthe DiskSpeed page is a little unclear on how the results are obtained. Is it a matter of "add up the three sizes and take their mean" or something else ?
17:14:10pondlifeNico_P: Wow
17:14:19pondlifeSim? gdb?
17:14:28Nico_PI'll try
17:14:59LambdaCalculus37I'm confused about how to post my results to the DiskSpeed page as well. My iPod's nearly done testig and I'd like to get the results up ASAP.
17:15:06Nico_Pit happened on the ipod.
17:15:14GodEateralso, the plugin logs aligned and unaligned creates - but there's only one "Create" column on the page
17:17:49GodEateranyone ?
17:18:47amiconnHmm, that page is indeed somewhat unclear.
17:19:03LambdaCalculus37amiconn: How should we post our findings to it?
17:19:07pondlifeNico_P: Skipping is much faster with your patch
17:19:14*amiconn even aded a few rows back then, but doesn't remember how it's meant to be filled in :\
17:19:24pondlifeI just went back to SVN and it's even more obvious
17:21:44GodEaterwell if someone wants to post a guide on how to fill it in, I've got runs at 30 / 80 MHz
17:22:25LambdaCalculus37I've got runs at 30 MHz only.
17:22:25*amiconn thinks that table should be redone
17:23:20GodEaterI've also acheived negative numbers in my 80Mhz results
17:23:40GodEatere.g. : Write (1048576,U): -7094 KB/s
17:24:22GodEaterwell barrywardell only filled it in a few days ago
17:24:30GodEaterperhaps he can tell us what methodology he used
17:24:41LambdaCalculus37GodEater: He filled it in on 4 Feb, 2007. :)
17:24:48GodEaterI just spotted that
17:24:52GodEaterthought it was 2008 for some reason
17:25:10GodEaterI'll get my coat
17:29:09GodEateramiconn: I assumed from looking at the table that the figures are the mean of three sizes used to do each task (apart from the create one, which baffles me completey)
17:29:13GodEateris that not likely ?
17:30:21markunLambdaCalculus37: these 37.8 seconds on your ipod color were with high-speed disabled, right?
17:30:24 Quit nplus (Read error: 104 (Connection reset by peer))
17:30:42GodEaterthey must be
17:31:32amiconnIt's more likely that the numbers are the maximum figures, i.e. those for 1MB buffer
17:31:39LambdaCalculus37markun: Yes.
17:32:16GodEatercan you explain my negative number amiconn ?
17:34:14 Quit LycoLoco_ (Read error: 104 (Connection reset by peer))
17:34:16 Join nplus [0] (
17:34:19 Join LycoLoco__ [0] (n=Lyco@
17:34:59*LambdaCalculus37 is still wondering why EMD on his iPod color outran even the OF
17:35:13n1sGodEater: there's a patch in the tracker for an overflow bug in test_disk might be that
17:35:58barrywardellamiconn: I think those numbers were mostly done before you committed r13475. back then test_disk was a lot less detailed
17:36:52GodEatern1s: any danger of commiting that patch? Seems useful...
17:36:52Nico_Ppondlife: -> This one should fix the natural dir skip problem
17:37:23gevaertsLambdaCalculus37: I am interested in some reliability testing on c240. Can you try high-speed writes to both internal flash and (if you have one) an SD card, both with a modified ata-c200_e200.c (remove the udelay() call that's in there somewhere, there's only one) and unmodified ? You probably will end up with corrupted filesystems, so don't do it if you don't have backups or don't feel confident in your recovery abilities.
17:37:44 Join gregzx [0] (
17:38:12n1sGodEater: slight decrease in precision? I'll leave it to amiconn, test_disk is his (iirc)
17:38:18LambdaCalculus37gevaerts: I haven't got backups of my c240 or the SD card.
17:38:51LambdaCalculus37I still need to dump a few classes I recorded to it and copy them to my hard drive.
17:38:56gevaertsLambdaCalculus37: ok. I would recommend not to use high-speed writing for now then.
17:39:07GodEaterprecision's not very useful in the event of the overflow occuring - which it seems it has here
17:39:25LambdaCalculus37gevaerts: But I'll gladly test high speed when it's a little more stable.
17:39:48gevaertsLambdaCalculus37: The udelay() in ata-c200_e200.c seems to be needed on e200, but causes write errors on my c250 at least when using high-speed
17:40:19LambdaCalculus37I haven't got an e200, so I can't verify that.
17:40:28pondlifeNico_P: That was just playback.c, not the other files, right?
17:40:45LambdaCalculus37Besides that the Radio Shack near my job no longer has any v1 e200s. =/
17:41:08Nico_Ppondlife: the changes in the other files weren't directly related. this is a cleaned up version
17:41:18pondlifeJust checking
17:41:44gevaertsI would like to put an #ifdef E200 (or whatever the correct symbol is) around that udelay, but I think it's a bit early with only my c250 to go by.
17:41:55*LambdaCalculus37 wishes he got an e200 when he had the chance
17:42:04amiconnn1s: Do you have a fs #?
17:43:04hcsLambdaCalculus37: I've got an e260 in a box from when I bought two, if you'd like one.
17:44:14LambdaCalculus37hcs: PM me; LambdaCalculus379 on the forums.
17:44:54GodEaterwithout the 9 ?
17:45:26LambdaCalculus37GodEater: On the forums, you add the 9.
17:45:26pixelmathe 9 is cutoff here (some nick length restriction)
17:45:34LambdaCalculus37IRC nick limits force me to drop the 9.
17:45:51 Quit toffe82 (Read error: 104 (Connection reset by peer))
17:46:06*LambdaCalculus37 also notes that 379 is a bit of a math reference
17:46:22gevaertsMaybe they support unicode nicks, so you could use a real lambda ?
17:46:36amiconnThat fix looks good
17:47:40*GodEater waits to see LambdaCalculus37 change his nick
17:47:56pondlifeNico_P: Yes, that transitions ok
17:48:00LambdaCalculus37GodEater: KnightOfTheLambdaCalculus doesn't fit. :)
17:48:05 Join toffe82 [0] (
17:48:16 Join major_works [0] (
17:48:26desowinλCalculus37 ;)
17:48:45 Join kugel [0] (i=kugel@unaffiliated/kugel)
17:48:56hcsbrb, resonance cascade
17:48:57jhMikeStoffee82: hey. they just handed it over then? sweet.
17:49:02major_worksgevaerts: What is the exact syntax for the Makefile extra defines line if I want to invoke USE_ROCKBOX_USB and USE_HIGH_SPEED?
17:49:10GodEaterdesowin: that'll be a bastard for nick completion though
17:49:21LambdaCalculus37desowin: Doesn't work.
17:49:43gevaertsmajor_works: 'export EXTRA_DEFINES=-DUSE_ROCKBOX_USB -DUSE_HIGH_SPEED'
17:49:51Horschtpoor win people would get a BSOD probably :D
17:49:55major_worksThank you!
17:52:08 Quit major_works (Client Quit)
17:52:34 Quit petur ("*plop*")
17:54:26 Join toffe82_ [0] (
17:54:36 Quit toffe82 (Read error: 104 (Connection reset by peer))
17:54:46 Nick toffe82_ is now known as toffe82 (
17:55:58 Join FOAD_ [0] (
17:57:18 Join japc [0] (
17:59:57Nico_Ppondlife: do I have your green light for the commit? :)
17:59:57pondlifeNico_P: I'm getting some bizarre behaviour, soft locking up and unwanted extra skipping of tracks when I skip backwards into unbuffered tracks.
18:00:46gevaertsI'd like to replace the various #defines surrounded by #ifdefs for usb states (USB_INSERTED,...) by an enum without #ifdefs (I need to add a state). On some targets some states would then be defined that won't be used, but that won't make any difference to the compiled result, and I think it would be a lot more readable. Does anyone think it's not a good idea ?
18:00:59pondlifeNothing reproducible yet though
18:01:15pondlifeThey might be bugs in SVN of course
18:01:47hcspondlife: I've experienced that as well
18:01:59pondlifeNico_P: Commit it... ;)
18:02:11hcspondlife: but also not reliably reproducible
18:02:16pondlifeIt's certainly nicer to use
18:02:22pondlifeIs the code simpler?
18:02:26Nico_Ppondlife: yes
18:02:41pondlifeIs the code well-commented?
18:02:53Nico_Pit does add a couple comments ;)
18:03:06GodEaterpondlife: he's removed all the "x/number_you_first_thought_of" routines
18:03:34pondlifeplaylist_check(), you mean...
18:03:52GodEaterwhich in turn of course means that "int think_of_a_number()" is now consigned to the bin of history
18:04:39GodEaterso with any luck a nice green delta :D
18:04:50pondlifeNico_P: Apart from combining the previous and next cases, what actually changed?
18:05:17gevaertsGodEater: Don't worry, I'm working on getting it red again ;-)
18:05:35pixelmaNico_P: does this patch change anything for hwcodec players?
18:05:39GodEatergevaerts: hehe ;)
18:05:48pondlifepixelma: No
18:05:49Nico_Ppixelma: no. it's SWCODEC only
18:06:03Nico_Ppondlife: updating the playlist index immediately on track change, and taking wps_offset into account for audio_next_track
18:06:21 Quit gevaerts ("work->home")
18:06:23Nico_Palso less uselss rebuffering
18:06:40pondlifePlaylist selection still rebuffers though, right?
18:07:18pondlifeSo is it really faster, or does it just feel it?
18:07:27pondlifeI mean, is it a UI thing
18:07:29gibbon_well... i hereby request write permission on the wiki, my name is joel and i'd like to contribute :)
18:07:41jhMikeSNico_P: If I read the playback code correctly, queue_post is being called from the audio interrupt which is quite unsafe since normal int sync doesn't mask that one.
18:07:59Nico_Ppondlife: should be both
18:08:08gibbon_for my personal page and the UsbPerformance page
18:08:08Nico_PjhMikeS: which audio interrupt?
18:08:09pondlifejhMikeS: Which routine?
18:08:25pixelmagibbon_: which Joel? ;)
18:08:43gibbon_pixelma: hmm... JoelGarske type Joel ;)
18:08:47jhMikeSthe track change callback or something...lemme look
18:09:08 Quit perrikwp (" ajax IRC Client")
18:09:59 Quit Febs_ (Read error: 110 (Connection timed out))
18:10:10pixelmagibbon_: just had to ask, you're now the 5th Joel with wiki write permission... :)
18:11:27gibbon_ok. i am glad to be part of such an amusement :)
18:11:43*LambdaCalculus37 sees an invasion of people named Joel on the wiki ;)
18:12:09 Quit FOAD (Read error: 110 (Connection timed out))
18:12:09 Nick FOAD_ is now known as FOAD (
18:12:28Nico_PjhMikeS: I don't see anything there
18:12:43pondlifeQ_AUDIO_TRACK_CHANGED is posted
18:12:46Nico_PjhMikeS: do you mean in my patch?
18:12:57gibbon_LambdaCalculus37: we are born to be a human botnet
18:13:02pondlifeNope, it's in SVN
18:13:11pixelmaLambdaCalculus37: there are even more Roberts/Robs... :P
18:13:28jhMikeSthat calls pcmbuf_set_event_handler which sets up a callback that is called from pcmbuf_callback
18:13:59LambdaCalculus37pixelma: It's a popular name, I guess. :P
18:14:01Nico_Pah I see
18:14:02jhMikeSNico_P: in SVN. just something I noticed
18:14:49Nico_PjhMikeS: I'm not the one who wrote that ;)
18:14:55Nico_Pwhat problem could it cause?
18:15:03jhMikeSkernel corruption
18:15:14pondlifeSo queue_post isn't safe in there?
18:15:21pondlifeI'm a bit confused
18:15:30jhMikeSnot at FIQ/audio DMA interrupt levels, no.
18:15:50pondlifeCould it be made safe? ;)
18:15:52 Join hannesd_ [0] (
18:16:38 Join perrikwp [0] (i=98217268@gateway/web/ajax/
18:16:57jhMikeSpondlife: as something indirect, sure
18:17:35 Join MethoS- [0] (
18:17:56pondlifeEither way we need some kind of safe queue, as multiple events could happen.
18:18:02 Quit barrywardell (Remote closed the connection)
18:19:29 Quit hannesd (Read error: 145 (Connection timed out))
18:19:30 Nick hannesd_ is now known as hannesd (
18:21:13rasherNico_P: is it by design that the current song keeps playing when you press next?
18:21:38Nico_Prasher: honestly I don't know. it has been like that for as long as I can remember
18:22:06pondliferasher: Would you prefer silence?
18:22:19rasherpondlife: I would. That would instantly tell me "I'm skipping now!"
18:22:28jhMikeSNico_P: you said there were queue overflow problems?
18:22:35rasherpondlife: And the current song might be very bad
18:22:36pondlifeThat's what the skip beep's for.
18:22:43Nico_PjhMikeS: I think so, with a bunch of skips
18:22:47rasherpondlife: ew, I don't want beeps though
18:22:56pondliferasher: You need to delete all bad songs from your player
18:23:35rasherLet's pretend I share my player with someone else (that's my excuse and I'm sticking with it)
18:23:41Nico_Pgreen delta! \o/
18:23:43 Nick [omni]_ is now known as [omni] (
18:24:40Nico_PI'd just like to mention that r16423 and r16424 should be perfectly safe
18:25:56jhMikeSNico_P: afaict posting the message on every skip probably is unneeded anyway.
18:25:58LambdaCalculus37What about r16425?
18:26:11Nico_PjhMikeS: the Q_AUDIO_SKIP message?
18:26:46Nico_PLambdaCalculus37: it makes bigger changes that *could* have unexpected side effects
18:26:56Nico_Palthough pondlife and I did do some testing
18:27:03Nico_PjhMikeS: what do you suggest?
18:27:04pondlifeNico_P: Isn't there a queue mechanism you could use to only post one Q_AUDIO_FILL_BUFFER at a time? I've been out of the coding too long...
18:27:39LambdaCalculus37Nico_P: Only one way to find out. :)
18:27:48*LambdaCalculus37 decides to try out 16425 on his iPod video
18:28:18 Part linuxstb ("Leaving")
18:28:39pondlifeI'd be interested to hear how it performs on a flash target
18:28:55pondlifeNico_P: You still need a Sansa :)
18:29:32jhMikeSNico_P: post it only if wps_offset == 0 at that time...any further ones should reflect in the accululated count. have audio snapshot it when it gets the Q_AUDIO_SKIP message and reset it to 0.
18:29:39*LambdaCalculus37 has a Sansa!
18:30:04 Join Domonoky [0] (n=Domonoky@
18:30:04 Quit goffa (Read error: 110 (Connection timed out))
18:30:14Nico_PjhMikeS: sounds good, yes
18:30:18pondlifeLambdaCalculus37: Compare skip speed before and after this update
18:30:44Nico_PLambdaCalculus37: the UI should also feel more responsive
18:31:22 Join toffe82_ [0] (
18:32:27*LambdaCalculus37 starts playing an album and lets the first track run for a few seconds
18:32:52LambdaCalculus37Nico_P: Wow, switching tracks is quite snappy now!
18:33:10 Nick mchua_ is now known as mchua_|food (
18:33:27LambdaCalculus37I just hit one track that took about 3 seconds to start.
18:33:29*pondlife is happy it wasn't his imagination
18:33:51LambdaCalculus37Eh? The last track on the album didn't play.
18:34:10 Join Lear [0] (i=chatzill@rockbox/developer/lear)
18:34:31LambdaCalculus37I can't skip forward into the last track on the album.
18:34:38pondlifeMe neither
18:34:46LambdaCalculus37But I can select the final track just fine.
18:34:48pondlifeIt returns to the browser
18:34:53 Quit gregzx (Read error: 110 (Connection timed out))
18:34:55LambdaCalculus37Me too.
18:35:01 Quit toffe82 (Read error: 104 (Connection reset by peer))
18:35:02 Nick toffe82_ is now known as toffe82 (
18:35:09LambdaCalculus37Let me try another album.
18:35:20Nico_Phmm looks like I have the same thing. weird I didn't see it
18:35:34pondlife<, <=, hey they're almost the same thing...:)
18:35:42pixelmaamiconn: what happened on your server?
18:36:19LambdaCalculus37Nico_P: I noticed that around track 7 or 8 on the albums I tried, it takes approximately 3 seconds to load.
18:36:30pondlifeLambdaCalculus37: That's buffering
18:36:42pondlifeQuite normal
18:36:53LambdaCalculus37pondlife: Just making sure. :)
18:37:04pondlifeAlthough I'd hope on a flash target it would be faster.
18:37:47Nico_PI guess having repeat set to all masked the last track problem :p
18:37:49*LambdaCalculus37 goes digging through his bag to look for his Sansa cable
18:38:07LambdaCalculus37Nico_P: Let me try that.
18:38:53LambdaCalculus37It also eliminates the 3 second pause on Track 7.
18:39:15LambdaCalculus37Nico_P: Repeat all does mask the last track problem.
18:39:23LambdaCalculus37And it properly repeats.
18:39:38Nico_Pthat's because there is no last track with repeat all ;)
18:40:02LambdaCalculus37Now to try it out on my c240.
18:40:35*LambdaCalculus37 updates his c240 build
18:40:59pondlifeLambdaCalculus37: Was the 3 second delay on Sansa or iPod?
18:41:09LambdaCalculus37pondlife: iPod video.
18:41:21LambdaCalculus37I'll see if it happens on the Sansa.
18:41:23pondlifeThat's entirely expected then
18:42:10 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
18:42:19 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:42:25LambdaCalculus37pondlife: Gotcha.
18:43:01Nico_Phmm I know what happens on the last track business
18:43:18LambdaCalculus37Nico_P: What is it?
18:43:55Nico_PI overlooked things in audio_check_new_track (specifically the end of playlist detection)
18:44:19Nico_PI'm really starting to hate that func. I think it's the worst part of playback.c now
18:44:20LambdaCalculus37Nico_P: Ahh. Because I noticed the last track problem just happened on my Sansa as well.
18:44:21pondlifeI knew that horrible routine would be trouble :)
18:44:38LambdaCalculus37But damn, is the Sansa ever so snappy!
18:44:43pondlifeNico_P: Read my comments above... I'm not psychic
18:44:54***Saving seen data "./dancer.seen"
18:44:59pondlifeLambdaCalculus37: How about the track 7 delay?
18:45:14LambdaCalculus37pondlife: Let me try another album just to be sure.
18:45:40LambdaCalculus37Nico_P: Tried a folder with a single track in it. No last track problem here.
18:45:54Nico_PLambdaCalculus37: yep, it's only on a skip
18:45:55pondlifeLambdaCalculus37: But you can't skip in that situation
18:46:23rasherNico_P: is this for the "folder advance" thing?
18:46:29LambdaCalculus37pondlife: No Track 7 delay on the Sansa. Smooth as glass.
18:47:03Nico_Prasher: no, there is a problem with the skipping to the last track in a dir. I'm in the process of fixing it though
18:47:21rasherNico_P: Yeah, I just meant if "that part" was related to the folder advance
18:47:30*LambdaCalculus37 hands Nico_P a beer for the great work
18:48:10Nico_Prasher: ah, you mean my comment about audio_check_new_track being the worst part of playback.c?
18:48:17rasherNico_P: Yep
18:48:32Nico_Pit handles everything, which is precisely the problem with it
18:48:40rasherSounds wonderful
18:48:45Nico_Pit just does too much to be understandable
18:48:45pondlifeIt shouldn't even be needed
18:48:56 Quit desowin (No route to host)
18:50:48 Quit toffe82 (Read error: 104 (Connection reset by peer))
18:50:50 Join toffe82_ [0] (
18:50:59 Nick toffe82_ is now known as toffe82 (
18:51:29pondlifeHmm, I need a paperclip after fast forwarding during the last track
18:51:33Nico_PI wish I had the courage to just rewrite all that in playback.c
18:51:41LambdaCalculus37pondlife: Which DAP?
18:51:47 Join desowin [0] (n=desowin@atheme/developer/desowin)
18:52:17pondlifeHard crash
18:52:36pondlifeOK, rebooted... I'll see if I can reproduce
18:54:12pondlifeI fast forwarded to near the end of track 11 (of 12). It moved onto track 12 ok. I then fast-forwarded through the (very long) track 12, out of the buffered data.
18:55:47*Nico_P fixed the "skip to last track" problem
18:55:58LambdaCalculus37pondlife: Let me try that on my iPod.
18:56:09 Join barrywardell [0] (
18:56:16pondlifeIt's hard to repro, you need the right album
18:56:21pondlifeA very long track
18:56:27LambdaCalculus37pondlife: How did you have repeat set?
18:56:39pondlifeBut I did have auto-change dir on
18:56:47LambdaCalculus37pondlife: I have an album with a 10:00+ long last track.
18:57:20pondlifeI don't know if "last track" is relevant. I suspect it's fast forward past buffered data
18:57:37Nico_Pthat really shouldn't be a problem
18:58:47*amiconn wonders why 'ld' segfaulted out of nowhere on rasher's box
19:00:21LambdaCalculus37Need to do something. Be back later.
19:04:10rasheramiconn: it's done that before. I really have no idea. It's not a box that is otherwise prone to mysterious crashes (apart from when using the nvidia X driver).
19:06:23 Join Mathiasdm [0] (
19:06:36 Quit Mathiasdm (Read error: 104 (Connection reset by peer))
19:09:28pondlifeNico_P: That automatic_skip variable is a mess; wonder if it could be removed somehow?
19:10:46Nico_Ppondlife: It wouldn't be an easy change. automatic and manual skips have quite a few differences
19:11:09 Join JdGordon|w [0] (i=cff4943f@gateway/web/ajax/
19:11:24pondlifeDoes it really mean "gapless or not" ?
19:11:42 Nick JdGordon|w is now known as JdGordon (i=cff4943f@gateway/web/ajax/
19:11:49 Join gregzx [0] (
19:12:01 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
19:12:22Nico_Pnot only that
19:13:08barrywardellwhat do people think about this:
19:13:29pondlifeHmm, I still have a problem skipping to the last track in r16427
19:13:30barrywardellIt removes those annoying "Searching..." splashes on sansa
19:14:01rasherbarrywardell: I don't enjoy those at all
19:14:03barrywardell(in tagcache - they're not there in the file browser)
19:14:33pondlifebarrywardell: I'd like those to vanish too, on H300
19:14:36LearNico_P: Busy with playback.c stuff? :)
19:14:58Nico_Ppondlife: really?
19:15:06pondlifeTotally reproducible
19:15:11Nico_PLear: trying to improve things, yeah
19:15:13BigBambibarrywardell: I hardly ever use the database, but when I do the splash does my nut
19:15:17barrywardellpondlife: they shouldn't be there provided tagcache is in RAM. on sansa's we never have tagcache in RAM because flash is fast enough
19:15:45LearNico_P: Yes, I'm looking into a different issue, but if you're busy on your end, I'll wait. :)
19:16:07pondlifebarrywardell: Any "in RAM" tests should be including a flash test, surely.
19:16:09BigBambibarrywardell: get rid of them. If flash is counted fast enough to not need to load to ram, it is fast enough to also not have the splashes
19:16:18 Join nicktastique [0] (n=nick@unaffiliated/nicktastic)
19:16:18Nico_Ppondlife: no problem here... :/
19:16:44Nico_PLear: no, go ahead :)
19:16:46*bluebrother wonders if there's a way to find out if a c200 is v2 or not by looking at the model number
19:16:46Lear(Wait with discussing the problem, that is.)
19:17:03pondlifeThis is r16427 as supplied by RBUtil. not self-built...
19:17:04*amiconn thinks those search splashes should stay, on all targets
19:17:18barrywardellpondlife: that's exactly the problem. it just checks "in RAM" regardless of whether it's flash or not
19:17:23amiconnThe yonly appear anyway if the search time is longer than a certain amount
19:17:31barrywardellthe splashes are too short to be readable on sansa
19:17:39amiconnI'd rather want to see a splash here and there instead of no feedback
19:18:17pondlifeNico_P: Wow, I just skipped to the last track, and it's playing it and displaying the right tag, but claiming it's track 1 of 12 !
19:18:29pondlifeMost times it just returns to the file browser
19:19:03pondlifeI play track n-2, then skip to track n-1, let it play a while, and skip to the last track.
19:19:28pondlifeSo I see it go track 9 of 11, then 10 of 11, then 1 of 12
19:20:01 Join Axio [0] (n=Axio``
19:20:08pondlifeView current playlist displays the next directory (which does have 12 tracks)
19:20:08Nico_Prepeat off and auto dir change on?
19:20:20Nico_PI'll enable auto dir change then
19:21:03Nico_PLear: what's the problem you're investigating?
19:21:42LearWhile playing, try to start playing a new track (maybe has to be via a bookmark, not sure about that). Playback stops then.
19:21:44pondlifeWithout auto dir change it seems ok
19:21:57Nico_Ppondlife: yes. I get it now
19:22:35 Join dan_a [0] (n=dan@
19:25:52pixelmabarrywardell: the tagcache in RAM is left out on Sansa because of the multivolume support (AFAIK it was there before microSD was supported) not because it's a flash target (the Nano uses it) but I don't mind if it isn't there
19:25:52 Quit toffe82 (Read error: 104 (Connection reset by peer))
19:26:59 Join toffe82_ [0] (
19:27:07 Nick toffe82_ is now known as toffe82 (
19:27:19barrywardellpixelma: I see. I was wondering why it was left out. the sansa seems fast enough without tagcache in RAM anyway, so it doesn't seem essential to me
19:27:31JdGordonyes ramcache has some unknown dependancy on dircache
19:28:38pixelmadatabase in RAM + dircache is said to speed up auto-update
19:29:04*dan_a is disappointed to get a V1 Sansa in the mail today
19:29:13 Join Mathiasdm [0] (
19:29:35bluebrotherdan_a: do we already know anything about the v2?
19:30:33pixelmabluebrother: did you read the logs? GodEater had some problems building the manual on his box, maybe you could help out?
19:30:34dan_abluebrother: It's based on the AS3525, which we have datasheets for
19:30:50dan_aThe audio and power is the same as the AS3514
19:31:16LearNico_P: Kind of ugly, really. Audio thread stops codec, then tells it to load a new codec. But before the codec thread gets around to loading the new codec, it tells the audio thread to stop...
19:31:32bluebrotherinteresting. Is that also true for the c200 v2?
19:31:33Nico_PLear: ouch :/
19:31:37 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
19:32:15bluebrotherpixelma: do you have the time?
19:32:20JdGordonNico_P: did you see my q the other day about playback->playlist interaction?
19:32:24Nico_Pdan_a: What do you intend to do with it? I've been wanting to get a flash target so I'd be willing to buy your sansa
19:33:00LearThe real problem is that there are two stop scenarios for the codec. Full stop, and stop before switching to a new track...
19:33:06Nico_PJdGordon: oh yes I forgot to answer, sorry. And now I've forgotten it... what was it?
19:33:33dan_aNico_P: I'm hoping that there'll be someone who's got a V2 but wants something which Rockbox will run on, so I could do a straight swap. Mine's a C240
19:33:34JdGordonis the only interaction between playback and playlist, playback constantly asking for the next track name to buffer?
19:33:41pixelmabluebrother: starting around 10:30
19:33:46 Join desowin_ [0] (n=desowin@atheme/developer/desowin)
19:34:05JdGordonamiconn: ?
19:34:07bluebrotherthanks, found it
19:34:09Nico_PJdGordon: It does that and also peeks
19:34:24 Quit desowin (Read error: 113 (No route to host))
19:34:59Nico_Pdan_a: ok, no worry ;)
19:35:13JdGordonit only keeps peeking if there is room on the buffer right? its not going to go peeing untill the end of the oplaylist if it cant buffer them?
19:35:32amiconnJdGordon: I am looking into porting icon theming to charcell. I already have a minimalistic BMP loader (mono only, width <= 8 pixels, takes a mere 500 bytes), and now need to get my head around apps/gui/icon.c. Do you have some pointers?
19:35:39JdGordonthats what I thought, but wanted to make sure
19:35:43LearHm, the codec thread should probably never send that stop message... Now it is sent if stop_codec is true, which is only set by the audio thread, I think.
19:36:05amiconnEspecially regarding the relation of icons strips, .icon files, and what is handled where
19:37:05Nico_PLear: the codec thread stuff in playback.c looks like a mess to me... another issue I know of and have never been able to fix is when I have a too big M4A file that gets skipped: the following one or two tracks get skipped too
19:37:10JdGordonumm, one sec
19:37:54 Quit toffe82 (Read error: 104 (Connection reset by peer))
19:38:37LearNico_P: And the following tracks were m4a too?
19:38:53*amiconn will probably drop the hardcoded icons altogether and only use themables like on bitmap displays, for uniformity.
19:39:00 Join toffe82_ [0] (
19:39:01Nico_PLear: no. I just tested and it's actually only the following track that gets skipped
19:39:08 Nick toffe82_ is now known as toffe82 (
19:39:45amiconnI mean the icons hardcoded into the char table
19:40:01bluebrotherGodEater: around?
19:40:11JdGordonamiconn: you might find it easier to reimplement icon.c dfor charcell....
19:40:37amiconnThat still means I need to understand it, at least somewhat
19:40:49bluebrotherthe pastebin shows latex not finding the document start. That should be in manual/rockbox.tex (in the output folder)
19:41:44JdGordonamiconn: I'm still half asleep... specific questions I can do... general overview will just confuse everyone
19:43:09JdGordonyou may want to look at 8133 also, its a rework for icon.c which hasnt been commited becasuse there is no need untill themebale statusbar is worked out
19:43:30 Nick mchua_|food is now known as mchua_ (
19:43:56LambdaCalculus37barrywardell: Around?
19:45:44 Quit tchan (Remote closed the connection)
19:46:32 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
19:47:00 Quit hcs ("Leaving.")
19:47:39LambdaCalculus37barrywardell: I performed a test_disk run earlier and want to post my results to
19:47:57LambdaCalculus37But I'm confused as to the layout of the page, and how to enter my findings.
19:48:05 Join DerDome [0] (
19:48:10 Nick fxb__ is now known as fxb (
19:48:25LambdaCalculus37Could you be able to give me an idea as to how to enter my data?
19:48:38 Nick fxb is now known as fxb__ (
19:48:50 Quit gregzx (Connection timed out)
19:48:54barrywardellLambdaCalculus37: I think the answer is tho redo the page. The table was made before test_disk was expanded with extra info
19:49:55barrywardellupdate on the annoying splash - this is all that's needed to get rid of them:
19:50:25 Quit crope` ("Leaving")
19:50:29LearNico_P: Hmm, that happens because there'll be to calls to codec_load_next_track in that case...
19:50:39 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
19:51:24JdGordonamiconn: are you loading the icons as a bmp and leeping it in the bmp format? or fiddling with it?
19:51:28Nico_PLear: do you know why there is one too many?
19:51:41LambdaCalculus37barrywardell: I wouldn't have the slightest idea as to how to lay out the page. :P
19:52:00LearNico_P: You could try adding a "status == CODEC_OK &&" before the second call, the one in the codec_thread.
19:52:12amiconnRight now I'm not loading any icons yet
19:52:42amiconnI just wrote the bmp loader so far, which of course converts the data into a suitable format, as the bitmap bmp loader does
19:52:49 Join crope` [0] (
19:53:24LearNico_P: If a codec returns with an error, it has already called request_next_track.
19:53:41amiconnThen I started looking into icon.c, and ran away...
19:54:29barrywardellLambdaCalculus37: OK, I'll make one for the sansa in a minute to get us started
19:54:42Nico_PLear: I'll try that
19:55:02 Join GodEater_ [50] (n=bryan@rockbox/staff/GodEater)
19:55:19LambdaCalculus37barrywardell: Excellent! :)
19:55:27JdGordonamiconn: you shouldnt have much work to do to get it working... the .icons and bmp handling will stay the same, you just need to change drawfun in screen_put_iconxy() to point to the correct lcd function, and fix the arrays at the top of the file where the bmps are stored
19:55:40JdGordonicon.c isnt all _that_ bad :p
19:55:51JdGordonat least its not 4500 lines!
19:56:23Nico_PLear: that seems to do the trick. any possible negative side effects?
19:56:26amiconnWell, charcell needs an extra step - it needs to allocate user-definable chars from the driver, and store the icon-number -> char index mapping
19:57:14LearNico_P: I don't think so, but I'm certainly not certain. Better try it a while before committing, I guess. :)
19:57:16JdGordonhmmm. your limited to only a few custom chars right?
19:57:29JdGordonare you going to convert on the fly if you load too many?
19:57:37 Part Jon-Kha ("[IRSSI] iTs bEttEr tO bUrN oUt tHaN tO fAdE aWaY")
19:57:46bluebrotherGodEater_: heard you had issues building the manual ...
19:57:54 Join Jon-Kha [0] (
19:57:57amiconnI'm not talking about hardware custom chars, but about driver custom chars
19:58:03GodEater_bluebrother: yes, gutsy is being a pain
19:58:10JdGordonoh ok
19:58:13amiconnThe driver allows 16 right now, but that number can be extended pretty easily
19:58:17Nico_PLear: hmm I'm getting an infinite loop now
19:58:41JdGordonI'm not immediatly sure where to add that step in,
19:58:51amiconnThe hardware custom chars are used dynamically for years already
19:59:04bluebrotherI'm not completely familiar with the latest build changes to the manual. Just to be sure, you have the sources along the manual (for the features.txt thingy)?
19:59:30JdGordonI guess the most logical place is putting that step in the if (size_read > 0) block in load_icons()
19:59:47amiconnI still want to only allocate soft chars on demand, in order to avoid excessive allocation with large bitmaps
20:00:03GodEater_bluebrother: not sure I understood that sentence...
20:00:13JdGordoncareful with that though.... load_icons() is called on settings_apply()...
20:00:23bluebrotherthe manual now also uses the features.txt file.
20:00:28GodEater_do you mean "do you have the manual in amongst the rest of the source code" ?
20:00:34bluebrotherbecause of that you need to have the sources around.
20:00:36GodEater_in which case, the answer is yes
20:00:37amiconn(pseudo code) if (!icon_char[icon_no]) icon_char[icon_no] = allocate_char(icon_bitmap);
20:01:05amiconnAnd loading a new bitmap would clear all mappings
20:01:14GodEater_bluebrother: did you see the pastebin I made of the build.log ?
20:01:16bluebrotheryou could check if (in the output folder) manual/rockbox.tex exists. It should contains \begin{document} somewhere
20:01:45JdGordonamiconn: you could do that right before the icon is displayed then... in screen_put_iconxy()...
20:01:45*gevaerts is trying to understand how boot-time usb detection is supposed to work
20:02:03JdGordongevaerts: in which bootloader?
20:02:04GodEater_bluebrother: I'll have to check tomorrow now - was building on my work pc, and I'm home now
20:02:30amiconnYes, probably. Of course there needs to be a check whether the allocation was successful, and if not, if should probably display a question mark
20:02:41gevaertsJdGordon: Actually after the bootloader, in main.c.
20:03:25JdGordonmain.c is the iriver bootloader isnt it?
20:03:34*JdGordon really doesnt like how bootloader/ is set out
20:03:50 Quit perrikwp (" ajax IRC Client")
20:04:11JdGordonyeah, main.c is iriver h100/h300
20:04:13bluebrotherGodEater_: ok −− unfortunately I'm a bit puzzled by this error.
20:04:23Nico_PLear: I don't understand. the inifinte loop doesn't seem to be related to that change and it only happens in my git tree, but not in my svn tree (even though both are basically the same)
20:04:26GodEater_you and me both =/
20:04:31gevaertsJdGordon: I mean apps/main.c. There's a bit in init() where it tests if there's an usb connection.
20:04:45 Join gregzx [0] (
20:04:51*bluebrother might get DSL on friday
20:05:07JdGordonoh, sorry
20:05:07amiconnJdGordon: Also, regarding your comment in apps/gui/icon.h line 33 (ifdefing the icon enum), I thing it would be okay to cut away icons from the end?
20:05:09*gevaerts should have said init-time instead of boot-time
20:05:29GodEater_bluebrother: what do you use to connect with currently ?
20:05:52bluebrothermy mobile. It works surprisingly good.
20:06:16amiconnI wouldn't want to do that in a fine-grained way, but cutting after Icon_Moving for charcell would be good, imho
20:06:46gevaertsI don't understand what button_get_w_tmo() is doing there
20:06:59pixelmabluebrother, GodEater: the rockox.tex in the directory I built the manual today there is said \begin{document} - if it's any help at all, I could paste the whole content somewhere if you want to know how it should look (I guess mine's ok since it builds ok)
20:07:06JdGordonamiconn: it wont... the individual icon heights are strip_height/count... remove some from the end and old strips arent usable
20:07:09bluebrotherbut it's kinda annoying and rather expensive. I'm currently volume-limited.
20:07:29JdGordonbut yes, removing some for charcell only would prob be ok if they will be using their own stip bmp anyway
20:07:40LambdaCalculus37Nico_P: r16425 just caused Track 7 on an album to repeat itself, but make the x of y counter go up a step; i.e. instead of "Track 7 of 13" it's now "Track 8 of 13".
20:08:40LambdaCalculus37I stopped the track and manually selected Track 8 on the album, and now it's playing correctly.
20:08:40*bluebrother wonders about FS #8650 −− we don't have numbered versions of the iaudio bootloaders on the download server
20:08:43GodEater_bluebrother: I've found (to my surprise) that the link to work is still up. I've just ssh'd into my pc there, and found that yes indeed, in build/manual/rockbox.tex - there is a /begin{document}
20:08:55JdGordongevaerts: SYS_USB_CONNECTED is sent to all queues, checking the button queue for it is just the easiest way to get it
20:08:58GodEater_\begin even
20:08:59pixelmaLambdaCalculus37: is this consistent? Just asking because I experienced something like this twice somewhen in January but couldn't find a way to reproduce (track 8 would be skipped in this case)
20:09:43LambdaCalculus37pixelma: No, it's not consistent. It just seems to happen at random times. I wonder what causes it . =/
20:10:12amiconnJdGordon: Hmm, nasty calculation...
20:10:34amiconnBut do you think someone would ever mix charcell and bitmap target icon strips?
20:10:38JdGordonyeah, but no nicer way
20:10:44JdGordonI doubt it
20:10:45amiconnThe charcell icons are 7x5 ...
20:11:14*JdGordon almost wonders what the point of custom icons are at that fixed size
20:11:37GodEater_that's not an icon, that's a blob ;)
20:11:56*JdGordon also wonders what amiconn is doing playing with pretty features is!! (joke....)
20:12:47amiconnNot as much point as on bitmap, but (1) it's unification in the way rockbox works and (2) someone might want to use the feature to map different icons for file types which have the same icon by default
20:13:40 Join perrikwp [0] (i=98213dfd@gateway/web/ajax/
20:13:45 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
20:13:47amiconnAlso, right now all viewer supported file types on Player get the same icon, which isn't nice
20:14:07amiconnIt has been this way since themable icons for bitmap targets were added
20:14:22bluebrotherGodEater_: interesing. If you pdflatex rockbox.tex, does that work?
20:15:08barrywardellanother attempt at removing the splashes:
20:15:09GodEater_! Undefined control sequence
20:15:27GodEater_platform/\platform <new line> .tex
20:15:40*JdGordon bbl
20:16:28bluebrotherhmm, looks like the platform macro isn't filled out.
20:16:48amiconnMy annoyance factor caused by such half-bakedness varies. saratoga's commit to make crappyv2 the default raised it somehow...
20:16:55 Join asdrubal [0] (n=dribbles@unaffiliated/asdrubal)
20:17:49GodEater_crappyv2.... nice. You're not bitter about the choice much ?
20:18:24rasheramiconn: the theme is fine, the commit caused a few problems that no one had noticed in testing.. and stop calling it crappyv2 - that's just childish
20:18:36 Quit desowin_ ()
20:19:13*barrywardell would like to commit his most recent splash removal patch unless there are objections?
20:19:19pixelmamaking it the default seems to not have been tested much
20:19:47 Join desowin [0] (n=desowin@atheme/developer/desowin)
20:20:13amiconnrasher: I am not complaining about the problems it caused, although those were nasty. But sorry, I quite hate this theme.
20:20:38rasherOh well, the errors were fixed, no one died, and it's not like it pushed back a release or anything.
20:21:28 Join bertrik [0] (
20:21:30rasheramiconn: So you just dislike the theme? It's a default - no one's telling you to use it. And we NEEDED a default theme that wasn't "tiny black font on blue". And this was the first reasonable one available.
20:21:49*barrywardell agrees with rasher
20:22:05*amiconn really thinks the old default was a better choice than that
20:22:13*LambdaCalculus37 also agrees with rasher
20:22:21*GodEater_ joins in the general agreement
20:22:47GodEater_the old default was horrible
20:22:49bluebrotherGodEater_: I'm dumb ... could you trying to manually pdflatex rockbox-<target>.tex?
20:22:51rasheramiconn: our users disagree, and they're the ones for whom the default matters.
20:22:56BigBambiThe old default was terrible
20:23:18*gevaerts hopes the old default won't be removed
20:23:27GodEater_bluebrother: I have no such file
20:23:30*bluebrother hopes that too
20:23:35LambdaCalculus37The old default would be fine on an Archos, but it looks really weird on a color target.
20:23:37 Join TMM [0] (
20:23:38amiconnThe old default wasn't pretty, sure. There were other themes I could live with as a default, although I still wouldn't use them for various reasons
20:23:48scorche|shlike which?
20:23:56bluebrotherGodEater_: right, it might be rockbox-build.tex. Such a file available?
20:23:56amiconnBut the current default is just, ermhhh....
20:23:56 Quit TMM (Read error: 104 (Connection reset by peer))
20:24:00rasheramiconn: what is the problem with having cabbiev2 as a default?
20:24:02amiconnscorche: E.g. iCatcher
20:24:06GodEater_yes there's one of those
20:24:16BigBambiI don't use it either, but at least it looks nice for new users
20:24:36GodEater_bluebrother: still errors
20:24:49GodEater_or rather, it stops
20:24:53bluebrotherhmm, strange. I guess it's the same error?
20:25:01GodEater_if I hit enter at that point
20:25:05GodEater_the manual continues to be built
20:25:10GodEater_and I get a .pdf output
20:25:13scorche|shamiconn: so how would you make the current default better?
20:25:16LearNico_P: I _think_ I fixed my problem. The error case works at least. Don't know how well general use works just yet. :)
20:25:36GodEater_the trouble with iCatcher is, it's not that eye catching. IMHO.
20:25:51bluebrotherdoes that pdf contain the manual or is it just empty / partly?
20:25:59amiconnWhat I don't like about the current default: The menu icons are incomprehensible at least on greyscale. It uses bright text on black background (on colour targets) -> bad readability. And the wps wastes space with useless bars and huge icons, and at the same time misses several interesting details about the tracks
20:26:03gevaertsWhat's the easiest way to get some kind of debug output while the splash screen is still showinf
20:26:26GodEater_bluebrother: it contains the whole manual as far as I can tell
20:26:33GodEater_157 pages
20:26:35gevaerts...showing ?
20:26:37rasheramiconn: those "interesting details" can be read in the metadata screen, and usually they just make the wps seem cluttered
20:27:01scorche|shbright text on a black background is bad readability? is much better than white on black...
20:27:02rasherI really don't think a *default* needs to show bitrate, id3 version etc.
20:27:29scorche|shi mean black on white...
20:27:32bluebrotherhmm. Was the error message different from the initial one?
20:27:48GodEater_no - same again
20:27:53 Join spiorf [0] (
20:27:58GodEater_complaining about a missing \begin{document}
20:28:02*bluebrother is puzzled
20:28:21GodEater_hang on
20:28:26GodEater_I think I may know what the problem is
20:28:35 Join TMM [0] (
20:29:38GodEater_no - I take it back
20:29:48GodEater_I thought it might be because I was building out of my git tree
20:30:01GodEater_which screws with the revision generation
20:30:06GodEater_but building out of svn doesn't work either
20:30:15amiconnid3 version isn't necessary, sure. But on a multi-format player like rockbox, I want to know the current format. And bitrate is interesting for lossy formats....
20:30:22 Quit TMM (Read error: 104 (Connection reset by peer))
20:30:54GodEater_amiconn: our users don't care about the format on the default theme. They care if there's music playing.
20:31:43pixelmaGodEater: that argument can be used in favour of the old default too (if they would only care that music is playing...)
20:31:51GodEater_bluebrother: I'm going to have to give up again for a bit. Dinner is ready :)
20:32:03GodEater_pixelma: touche
20:32:07rasheramiconn: Why does the format matter? I never understood this. If you're listening to something, what matters should be how it sounds - if something's funny about that (which should be rarely), you can check the metadata screen...
20:32:25scorche|shwell, we need it to look similar across all platforms as well...there isnt much room to fit much more in on some targets
20:33:55 Join FeetFirst [0] (
20:34:24 Part FeetFirst
20:34:26amiconnrasher: Convenience...
20:34:30 Join TMM [0] (
20:34:43amiconnAnd that wps also doesn't allow precise volume adjustment
20:35:05rasheramiconn: I find the convenience of not looking at a billion different types of information is far more important than not having to go into a menu once in a blue moon
20:35:19amiconn(this could probably be fixed)
20:35:21*preglow agrees
20:35:46preglowthat's the one thing i don't like about icatcher, it crams too much info into the wps
20:35:55GodEater_also, precise volume? I just roll the volume up and down until it sounds right...
20:36:01rasherHow often do you need precise volume adjustment? Apart from putting it at 0, which is possible.
20:36:37preglowi'd like a bigger volume indicator
20:36:38preglowthat i would
20:37:02scorche|shi would agree with having an actual number for volume, but it doesnt fit with that theme
20:37:08preglowbut then again, i'd prefer something like on ipod retailos, a slider that appears when adjusting the volume
20:37:34LambdaCalculus37Nico_P: More problems... I tried playing a FLAC album, and now it's skipping Track 3 and up to Track 4. When I try going back to Track 3, it stopped and gave me a data abort.
20:38:01pixelmaI miss the numerical volume display too btw., got used to it
20:38:07bluebrotherone thing I really disliked about the old default was its way of showing the tag type. It was either id3v<something> or no id3.
20:38:23 Join merbanan [0] (n=banan@
20:38:24bluebrotherI really missed "vorbis comment" for ogg files
20:38:26 Quit Jon-Kha (Read error: 145 (Connection timed out))
20:39:00bluebrotherGodEater_: svnversion breaking shouldn't affect building the manual. In the worst case it would fall back to show something like "unknown" instead
20:39:05bluebrother(can't remember exactly)
20:39:15GodEater_it does
20:39:31 Join toffe82_ [0] (
20:39:34 Quit tchan (Remote closed the connection)
20:39:42LambdaCalculus37Nico_P: Should I roll back a couple of revisions?
20:40:21barrywardellIs the "Searching...(# found)" progress indicator any use on any targets? on my targets (H10 and Sansa) the only time there's any real delay is waiting for disk spinup on H10 and it spends most of the time at "0 found"
20:40:29 Join waldo [0] (
20:40:30amiconnscorche: There must be a reason why the german regulation for PC workplaces ("Bildschirmarbeitsplatzverordnung") demands black text on bright background as a default.
20:40:44amiconnAnd I can only say that I agree 100% with that demand
20:41:01barrywardellso i would thing just showing "Searching..." when the disk needs a spin up would be enough
20:41:10 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
20:41:17scorche|shi find white text on black background MUCH more comfortable than black text on white
20:41:36Lloreanbarrywardell: I've seen the number found actually increase, I can't remember when, but i think it was before some change.
20:41:46preglowamiconn: blech!
20:41:52gevaertsI find that white on black or black on white is very much dependent on the screen
20:42:14Nico_PLambdaCalculus37: maybe try reproducing in r16424
20:42:36barrywardellLlorean: it increases extremely briefly. it stays at 0 while the disk spins up, then quickly goes up once it actually starts searching
20:42:50LambdaCalculus37Nico_P: Okay, will roll back now.
20:43:00Lloreanbarrywardell: Do you have load to RAM on or off, and dircache on or off?
20:43:11 Quit bluebrother ("bye.")
20:43:22barrywardellLlorean: this is with it off. with load to RAM on, it doesn't show the progress anyway
20:43:31LloreanI seem to recall it moving slow enough that it took a few seconds to count up, but this was some time ago, and my recent DB test I didn't notice it.
20:43:53LloreanIt's possible it's just too fast to matter these days, and not worth actually counting
20:44:37barrywardellLlorean: yes, that's my opinion. it's so fast that it just ends up being an irritating flash rather than a useful progress meter
20:44:38 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:44:40 Nick toffe82_ is now known as toffe82 (
20:44:55***Saving seen data "./dancer.seen"
20:45:01barrywardellI'm wondering if it's slow enough on other targets to be useful though
20:45:22Lloreanbarrywardell: Have you tried it with >8000 songs?
20:45:32*LambdaCalculus37 is trying to remember if there was a page with older SVN builds on it
20:45:47LloreanLambdaCalculus37: Daily build page contains a link to "Older" builds which go back 30 days
20:46:04LloreanBut only older dailies, so they skip around in revisions quite a bit, and are only suitable for broadly grained testing / tracking of things
20:46:48LambdaCalculus37Llorean: I need to get r16424; I'm helping Nico_P test a commit he made in r16425 to see if anything was broken before it.
20:46:50*gevaerts finally got booting with usb plugged in to work properly in the new scheme of things
20:47:13Nico_PLambdaCalculus37: can't you build from source?
20:47:47barrywardellLlorean: no, my H10 only has ~5000 songs on it.
20:48:13LambdaCalculus37Nico_P: My SVN tree is 4 days behind, and I can't connect it to our network here at my job. =/
20:48:24LloreanLambdaCalculus37: You'd need to compile if you need to get a specific SVN revision.
20:49:08Lloreanbarrywardell: Well, 5000 is surely enough, I'd bet, that you'd see how fast it goes up.
20:49:31LambdaCalculus37Llorean: Understood. Then I won't be able to reproduce the track skip problem right now, it seems.
20:49:45 Quit Mathiasdm ("Yuuw!")
20:50:40Nico_PLambdaCalculus37: maybe try with yesterday's daily
20:50:42gevaertsLlorean: Could you get me a developer badge on the forum if you have some time ? My forum name is fg
20:51:13barrywardellLlorean: maybe I'll try removing it in favor of a simple "Searching..." when the disk needs to spin up and post a patch on the tracker for people to test under different circumstances
20:52:02LambdaCalculus37Nico_P: All right. The next most recent daily is r16422; downloading now.
20:53:18Lloreangevaerts: Just "fg"? (making sure it wasn't a typo)
20:53:18 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:53:21desowingevaerts: integration_2.patch is known to cause problems?
20:53:24Lloreanbarrywardell: Sounds good to me.
20:53:47gevaertsdesowin: what kind of problems ?
20:53:55gevaertsLlorean: yes. My initials
20:53:55 Join toffe82_ [0] (
20:54:03 Nick toffe82_ is now known as toffe82 (
20:54:09desowingevaerts: I/O
20:54:12Lloreangevaerts: Done, you have a badge.
20:54:31LambdaCalculus37Nico_P: Rollback complete; commencing test... NOW!
20:54:37*LambdaCalculus37 hits the shiny red button
20:54:47 Quit barrywardell (Remote closed the connection)
20:54:55gevaertsdesowin: it shouldn't give more of thos than before.
20:55:20LambdaCalculus37Nico_P: I'm going to play the same FLAC album as before and see if it does the same behavior again.
20:55:26desowinI'll run fsck and recheck
20:59:07 Quit rockho (Read error: 110 (Connection timed out))
20:59:18amiconnLlorean, barrywardell (for the logs): There are a few functions in the db where I can see the "Searching..." splash counting
20:59:23LambdaCalculus37Nico_P: r16422 doesn't skip Track 3 on my FLAC album.
20:59:40amiconnA to Z... -> Tracks is one, and Search... -> Tracks is another
20:59:47 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
20:59:53amiconnThat's on swcodec, btw
21:00:11LambdaCalculus37But trying to go forward one track does repeat the track and raise the "x of y" counter on the WPS by one.
21:00:17Nico_PLambdaCalculus37: was it consistent with the latest SVN?
21:00:32amiconnOn hwcodec it is counting for many functions (e.g a simple all tracks list)
21:00:38LambdaCalculus37Nico_P: I'm still messing with it a little; one sec.
21:00:49Lloreanamiconn: I guess maybe it depends on the complexity of the filterset?
21:00:56LambdaCalculus37Here's what I have... I have a 14-track album in FLAC.
21:01:10amiconnI think so
21:01:19LambdaCalculus37At track 3, hitting forward makes the "X of Y" counter on the WPS jump up one number, and repeat Track 3.
21:01:35LambdaCalculus37At this point, pressing forward again makes it skip Track 4 entirely and jump right to Track 5.
21:01:48Lloreanamiconn: Do we need to have it count, or would a simple "Searching" and possibly a very basic animation so you know it's not totally frozen be acceptable?
21:01:52LambdaCalculus37And it's consistent with what I was experiencing with r16425.
21:02:15LambdaCalculus37Jumping to Track 6 in the album also makes it pause for 3 seconds while buffering.
21:02:25Nico_PLambdaCalculus37: so r16422 is doing it too?
21:02:52LambdaCalculus37Nico_P: Yes. It did it with FLAC. Want me to try with Vorbis?
21:03:17amiconnI'm not sure how slow the searches could become when are really many tracks on the device.
21:03:29Nico_Pwhy not. so at least this problem wasn't caused by r16425 then
21:03:51 Quit nplus ("Leaving")
21:03:54amiconnWith what I have on my recorder, I'd say counting isn't necessary, a simple splash would do. But then I only have ~3600 tracks on it
21:04:09LambdaCalculus37Nico_P: Playing a Vorbis album now. Will report when I hit a snag.
21:04:36amiconnBtw, I have the impression that the database became faster at some point
21:08:27 Join soap_ [50] (n=soap@rockbox/staff/soap)
21:08:27 Quit soap_ (Read error: 104 (Connection reset by peer))
21:11:35 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:11:37 Join toffe82_ [0] (
21:11:45 Nick toffe82_ is now known as toffe82 (
21:15:50 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:15:51 Join toffe82_ [0] (
21:15:59 Nick toffe82_ is now known as toffe82 (
21:16:17 Join Domonoky_ [0] (n=Domonoky@
21:18:17 Quit ompaul (Client Quit)
21:20:19 Join magmaniac [0] (
21:20:20 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:21:56 Join toffe82_ [0] (
21:22:05 Nick toffe82_ is now known as toffe82 (
21:22:17LambdaCalculus37Nico_P: r16422 doesn't have the last track skip problem.
21:22:43LambdaCalculus37The Vorbis album played with only a three second buffer on Track 8.
21:22:51Nico_PLambdaCalculus37: I fixed that one in r16426
21:23:30LambdaCalculus37Nico_P: Think the data abort I had before may have been FLAC-related?
21:23:48Nico_PI don't know
21:24:46kugelbtw, this revision display on the front page is genious, and yet simple
21:24:47LambdaCalculus37Want me to jump to the current SVN version and give it one more test?
21:25:00 Quit TMM ("Ex-Chat")
21:25:07kugelwould be great if since-4weeks etc would adapt this
21:25:24Nico_PLambdaCalculus37: yes please. I'm about to commit a fix for auto dir skip though
21:26:45LambdaCalculus37Nico_P: Want me to test before you commit?
21:26:49gevaertsCan someone tell why in usb-fw-pp502x.c, in usb_enable the #ifdef H10 block is there ? For H10, it seems to only reboot if BUTTON_RIGHT is held, which is strange because from reading usb.h I understand that on H10 BUTTON_RIGHT means "don't reboot on connect"
21:27:28Nico_PLambdaCalculus37: you can, but I just wanted you to know issues you get with auto dir skip will probably be fixed
21:27:53LloreanNico_P: Any thoughts on this bug report? It seems to have more information about the old "screeching whitenoise" problem.
21:28:09 Join TMM [0] (
21:28:17LambdaCalculus37Nico_P: Okay, commit first and then I'll try once more.
21:28:51 Quit perrikwp (" ajax IRC Client")
21:29:26pixelmagevaerts: I don't know the details but on some H10s (depending on if they are "MTP" or "UMS" ones) you have to hold the "right" button to get them into UMS mode which is why it was implemented in Rockbox that you also have to hold the button
21:29:28amiconngevaerts: On H10, reboot and not-reboot behaviour is swapped compared to the other PP targets
21:30:05Nico_PLambdaCalculus37: committed
21:30:13Nico_PLlorean: wow that's detailed
21:30:34LloreanNico_P: I'm particularly interested in the fact that it's not boosted, actually. I would imagine crossfade would be very boosted.
21:30:42amiconnThis is because the H10 MTP firmware requires holding Right on boot to get into UMS mode, and for some H10's (afaik H10 big and the 5GB H10 "pure") there is no choice whether to install UMS or MTP OF
21:31:19 Join dandin1 [0] (
21:32:09amiconnThe H10 pure can be forced to accept the normal small H10 UMS firmware (giving a useless radio menu, and UMS mode)
21:32:24gevaertsamiconn: OK. I'm still a bit confused about the #ifdefs in the code, but that information will help
21:33:33amiconnThis reboot issue will vanish when rockbox usb will be enabled, and then H10 button behaviour should be swapped to match all other USB powerable tarbets
21:34:12 Quit Domonoky (Read error: 110 (Connection timed out))
21:34:26 Join mozetti [0] (
21:34:35LambdaCalculus37Nico_P: Waiting for all of the targets to sync to r16430.
21:34:58gevaertsamiconn: true, but I want to keep it as long as possible, so the final "enable usb" commit will only switch one #define (or remove the guarded code)
21:35:25mozettianyone here using QTScrobbler on Linux? just want to make sure I install the right dependencies, and aptitude has a bunch of packages with "libcurl" in the name
21:35:32preglowthat crossfade noise bug is nASTY
21:35:53*LambdaCalculus37 notices that the H10 20GB, Gigabeat F, and iPod video 30GB are taking a while to sync to r16430
21:35:53 Quit bertrik ("bye")
21:36:21*mozetti nudges desowin
21:37:10 Join goffa [0] (n=goffa@
21:37:25LambdaCalculus37Weird... iPod video 30GB is still r16429, and H10 20GB says "broken?" =/
21:37:35LambdaCalculus37Never mind.
21:37:44 Quit DerDome (Read error: 110 (Connection timed out))
21:38:18Nico_PLambdaCalculus37: you can look at to know what's going on
21:39:01LloreanBagder: You around by chance?
21:40:37LambdaCalculus37Nico_P: Last track skip problem is fixed in r16430.
21:40:46amiconnMeh, barrywardell removed the feedback splash for Ondio as well...
21:40:48 Join x1jmp [0] (
21:41:06Nico_PLambdaCalculus37: yep, I got that one in r16426 :)
21:41:28LambdaCalculus37Trying some FLACs now.
21:41:49LloreanBagder: Anyway, for the logs, apparently amiconn posted an updated bootloader for the M5 to the IaudioBoot page in the wiki some time ago, but it's not at
21:42:08LambdaCalculus37\o/ FLACs work again!
21:42:21Nico_Pwell that's good news. or luck :)
21:43:05LambdaCalculus37Well, seeking and skipping back and forth between tracks isn't giving me trouble now.
21:43:56LambdaCalculus37Great work, Nico_P! :)
21:44:01 Nick LycoLoco__ is now known as LycoLoco (n=Lyco@
21:44:12desowinmozetti: ?
21:44:14Nico_Pwell thanks :)
21:44:19 Part LycoLoco ("Going!")
21:44:48*LambdaCalculus37 sits back and enjoys some FLAC goodness on his iPod video
21:45:13mozettidesowin: i'm using Ubuntu 7.10 −− i want to install QTScrobbler, but not sure on the exact dependencies (libcurl & QT) that I need
21:45:36*LambdaCalculus37 gives Nico_P a beer
21:45:48scorche|shmozetti: a bit offtopic for this channel..
21:45:58*Nico_P doesn't like beer but accepts all the same :)
21:46:36desowinmozetti: libqt4-dev build-essential and curl dev
21:46:49 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
21:47:17mozettidesowin: thank you −− been looking for a way to use the Rockbox scrobbling plugin, and just found your app.
21:47:18desowinmozetti: I've filed debian RFP if you know debian developers trick them please ;)
21:48:09mozettiif I knew debian devs I probably wouldn't have needed help identifying the dependences or being scolded by scorche|sh ;)
21:48:22mozettithanks again
21:51:37 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:52:22 Join BitTorment_ [0] (n=martin@
21:52:33 Part pondlife
21:52:56 Join toffe82_ [0] (
21:53:05 Nick toffe82_ is now known as toffe82 (
21:54:24amiconnSome time being almost a year...
21:54:32 Join Febs [0] (n=chatzill@
21:57:22LambdaCalculus37Time to go now... good night, everyone!
21:57:22 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:57:57 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
21:58:57 Join toffe82_ [0] (
21:59:06 Nick toffe82_ is now known as toffe82 (
22:00:40 Quit Febs (Read error: 104 (Connection reset by peer))
22:02:10 Quit toffe82 (Read error: 104 (Connection reset by peer))
22:02:58 Join toffe82_ [0] (
22:03:06 Nick toffe82_ is now known as toffe82 (
22:03:12 Quit gevaerts (Read error: 110 (Connection timed out))
22:04:04 Quit desowin (Remote closed the connection)
22:04:04 Quit toffe82 (Read error: 104 (Connection reset by peer))
22:04:05 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
22:05:50 Join toffe82_ [0] (
22:05:58 Nick toffe82_ is now known as toffe82 (
22:07:07 Quit merbanan (Remote closed the connection)
22:07:07 Quit toffe82 (Read error: 104 (Connection reset by peer))
22:07:34 Join kushal_12_27_200 [0] (n=kushal@
22:09:12 Quit BitTorment__ (Read error: 110 (Connection timed out))
22:09:37 Join toffe82_ [0] (
22:09:46 Nick toffe82_ is now known as toffe82 (
22:16:55gevaertsI uploaded integration_3.diff to FS #8562, which I'd like to commit, so I'd like some testing on various players (both portalplayer and non-portalplayer, since I changed some common code). See the tracker for details on what I'd like to have tested.
22:17:19gevaertsReviews of what the code does are of course also welcome
22:18:45 Quit Horscht ("Snak 5.3.3 Unregistered copy. Evaluation period is over. Program will now quit. Thanks for using Snak.")
22:18:54 Join BitTorment__ [0] (n=martin@
22:21:33 Join DerDome [0] (
22:21:33 Quit toffe82 (Read error: 104 (Connection reset by peer))
22:21:50 Join toffe82_ [0] (
22:21:58 Nick toffe82_ is now known as toffe82 (
22:22:04 Join Joshua_____ [0] (
22:22:13 Nick Joshua_____ is now known as carbine (
22:22:27 Nick carbine is now known as CarbineMonoxide (
22:22:30 Join perrikwp [0] (i=98217268@gateway/web/ajax/
22:23:21CarbineMonoxideAnyone happen to know if the RockBox development for the Zune has moved any lately?
22:23:28LloreanIt has not.
22:23:38LloreanThat's because nobody is working on it.
22:23:50CarbineMonoxideLast I read it was stuck on some encryption/security issue.
22:24:06alienbiker99still is
22:24:13Bagderit never got beyond that
22:24:21Bagderand I don't think many people tried very hard
22:24:27gibbon_now that i sit at this device with rockbox on it ("the device" being a c250 sansa) and USB usage enabled... does it charge in rockbox?
22:24:45CarbineMonoxideAnd someone broke the security on the Toshiba model that is similar to the Zune while working on the Zune. =D
22:25:08LloreanBagder: Did you see my message about the firmware file for the m5?
22:25:29alienbiker99i have to say, rockbox is better with the battery already on the gigabeat S haha
22:25:33LloreanThe bootloader on the wiki page is "v2", uploaded by Amiconn I believe, but on the download server it's V1
22:26:02gevaertsgibbon_: it should. There are still issues with charging AFAIK which cause it to stop charging a bit too soon, but on my c250 it does charge.
22:26:02CarbineMonoxideThe Zune rocks except for the face that it's bound to MS and doesn't work with =(
22:26:26 Quit toffe82 (Read error: 104 (Connection reset by peer))
22:26:34BagderCarbineMonoxide: nothing without rockbox can truly rock...
22:26:50amiconnLlorean, Bagder: That's correct. M5 bootloader v2 corrects a non-critical bug present in the v1
22:27:21gibbon_gevaerts: thanks...
22:27:29CarbineMonoxideI've offered up a damaged Zune to anyone willing to work on it, but as stated above, it seems as if progress has died.
22:27:37gibbon_gevaerts: stresstesting the iPod mini once again by moving all the music to the sansa ;)
22:27:40BagderCarbineMonoxide: there was never progress
22:27:53BagderCarbineMonoxide: just people like you, expressing interest
22:27:55 Join toffe82_ [0] (
22:28:03 Nick toffe82_ is now known as toffe82 (
22:28:26CarbineMonoxideI'd help if I knew anything about it. My programming/scripting experience is limited to PHP, VB.NET, ASP.NET and some Java. =x
22:28:35Bagderamiconn: so what file should I put on the download server?
22:29:35 Quit Lear ("ChatZilla 0.9.81 [Firefox 3.0b4pre/2008022504]")
22:30:27 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
22:30:27 Quit toffe82 (Read error: 104 (Connection reset by peer))
22:31:01gibbon_gevaerts: that should simulate your "another device on the same 2.0 hub" quite well
22:31:58gevaertsgibbon_: it should indeed.
22:32:50 Join toffe82_ [0] (
22:32:59 Nick toffe82_ is now known as toffe82 (
22:34:00gibbon_even the SD-Card slot "virtual second device" (how do i call it correctly in MSD context?) works fine on the sansa...
22:34:08 Quit nicktastic ("Leaving")
22:34:28gibbon_the scsi layer suggests talking of a different lun
22:35:48 Quit BitTorment_ (Read error: 110 (Connection timed out))
22:36:05CarbineMonoxideSo if I wanted to start working on RockBox for the zune, just hypothetically, what would I need to know/have experience with or in?
22:36:34gevaertsI got the sd card working pretty early in linux. It probably saved me lots of time to have an entire device that I could get corrupted as often as I like
22:36:38CarbineMonoxideAlrighty, thanks. =)
22:37:28gibbon_gevaerts: now i understand what you said earlier today about what you like about your SD :)
22:37:32gibbon_makes sense
22:37:47 Quit perrikwp (" ajax IRC Client")
22:37:52 Join perrikwp [0] (i=98217268@gateway/web/ajax/
22:38:55CarbineMonoxideHmm.. So in other words.. Way over my head. =D
22:39:34gibbon_CarbineMonoxide: there are ladders ....
22:39:34alienbiker99ack why am i getting build errors on gwps-common
22:39:55gibbon_do you remember the cookies on the kichen fridge, when you were a lot younger? :)
22:40:19CarbineMonoxideWe always kept ours in a drawer..
22:40:38gibbon_was there something else interesting on your fridge?
22:40:58CarbineMonoxideNot really.. Just containers full of coffee.
22:41:01gibbon_what i wanted to say was... try
22:41:12gibbon_CarbineMonoxide: perfectly to lure a hacker :)
22:41:25 Quit bernz ("Java user signed off")
22:41:42CarbineMonoxideMy friend is an electrical engineer..
22:41:45 Join MajorC [0] (
22:42:15CarbineMonoxideHe might be able to help. Where he lacks in programming I can try to take over. PHP and C have the same basic syntax so it shouldn't be too difficult to learn.
22:42:32 Join crzyboyster [0] (n=4b596e83@gateway/web/cgi-irc/
22:42:33*GodEater_ thinks we should add a section to NewPort about CryptAnalysis for most modern targets =(
22:42:34CarbineMonoxideAnd not as though I can't ask for help with C stuff. It's kind of a big deal. =P
22:42:38gibbon_read about pointers and stuff...
22:43:21CarbineMonoxideAnd I'm guessing I can use a lot of the stuff from the Gigabeat S rockbox.
22:43:31GodEater_code wise yes
22:43:38GodEater_but the big challenge is executing that code
22:43:45GodEater_that's the huge first hurdle
22:44:02 Quit crzyboyster (Client Quit)
22:44:29 Join crzyboyster [0] (n=4b596e83@gateway/web/cgi-irc/
22:44:58***Saving seen data "./dancer.seen"
22:45:44crzyboysterHow would I resume a download of the source within vmware?
22:45:59GodEater_crzyboyster: you were using svn to do the download I assume ?
22:46:21crzyboysterYes, the svn trunk command..
22:46:46 Quit jgarvey ("Leaving")
22:46:47 Quit DerDome ("Leaving.")
22:46:54GodEater_crzyboyster: "svn up" should continue where you left off
22:47:23crzyboysterIt didn't do anything...yet.
22:48:30crzyboysterNope, still nothing. I guess I'll just restart. Where are the downloaded svn files downloaded to? And what would be the proper way to apply a patch?
22:49:33 Quit perrikwp (" ajax IRC Client")
22:50:33alienbiker99how do i get around this gwps-common error when building?
22:51:32 Join DerDome [0] (
22:51:36GodEater_alienbiker99: you forgetting something ?
22:51:54alienbiker99no, havent touched that file
22:52:05 Quit CarbineMonoxide ("ChatZilla 0.9.81 [Firefox]")
22:52:13GodEater_no - I meant, you're forgetting you haven't told us what the error is yet.
22:52:17 Quit tvelocity_ ("Αποχώρησε")
22:52:17GodEater_we're not psychic
22:52:18 Quit DerDome (Remote closed the connection)
22:52:43 Join DerDome [0] (
22:52:52alienbiker99oh right. sorry
22:53:27 Join perrikwp [0] (i=98217268@gateway/web/ajax/
22:53:34GodEater_this is with up to the minute svn? and no patches at all ?
22:54:48kushal_12_27_200Hello, its me again. Please help.
22:55:05alienbiker99i only have gevaerts #8562 applied
22:55:34GodEater_well let's be sure shall we? Run "svn revert -R ." in your source directory
22:56:15alienbiker99didnt revert anything in gui. ill try to build agian
22:56:34GodEater_do a "make clean" first too
22:56:56pixelmaI'd also suggest a reconfigure
22:57:54*GodEater_ notices it's bed time
22:58:05 Quit GodEater_ ("ZZZZZZzzzZZZ")
22:58:16kushal_12_27_200gevaerts, are you there? please help me with if you have some free time.
22:59:49gevaertskushal_12_27_200: start a terminal window first
23:00:05alienbiker99hmmm that error is still coming up
23:00:16alienbiker99pixelma, whats a reconfigure?
23:00:19gevaertskushal_12_27_200: do you still have all files from last time ?
23:00:40pixelmaalienbiker99: running tools/configure again
23:00:57alienbiker99oh, i already did that
23:01:23gevaertsok. In the terminal window type "cd Desktop"
23:01:34kushal_12_27_200I downloaded firmware.mi4 and c250repair-mac again and i sent the rest to trash. I can pull them back.
23:02:36 Quit MethoS- (Remote closed the connection)
23:02:38gevaertskushal_12_27_200: depending on what the problem is you might need pribootLoader.rom as well, but we'll see
23:02:57 Join MethoS- [0] (
23:03:33gevaertskushal_12_27_200: Once you have done the "cd Desktop", run "mkdir /tmp/sansa"
23:04:13kushal_12_27_200result is "file exists"
23:04:55 Quit MethoS- (Remote closed the connection)
23:04:56crzyboysterIs the svn download downloaded into TEMP files using vmware?
23:05:24gevaertskushal_12_27_200: ok. that's fine. Now run "sudo mount -t msdos /dev/disk1 /tmp/sansa"
23:05:53 Quit jhulst_ (Remote closed the connection)
23:06:08gevaertskushal_12_27_200: now "cp firmware.mi4 /tmp/sansa"
23:06:31 Quit mf0102 ("Verlassend")
23:08:04gevaertskushal_12_27_200: now "umount /dev/disk1"
23:08:15 Join MethoS- [0] (
23:08:40kushal_12_27_200:( operation not permitted
23:08:47 Quit crzyboyster ("CGI:IRC (EOF)")
23:09:13gevaertskushal_12_27_200: "sudo umount /dev/disk1"
23:09:44kushal_12_27_200oo ok
23:09:49 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
23:10:01gevaertsThen "hdiutil detach /dev/disk1"
23:10:46*amiconn does have some questions for LinusN
23:10:51gevaertsAfter that, make sure the hold switch is off, unplug the cable, and see if it recovers
23:11:19kushal_12_27_200result is
23:11:19kushal_12_27_200"disk1" unmounted.
23:11:19kushal_12_27_200"disk1" ejected.
23:11:34 Quit MethoS- (Remote closed the connection)
23:11:54kushal_12_27_200in the player screen it is "LUN0" in place of "disk1"
23:11:57 Join MethoS- [0] (
23:12:09kushal_12_27_200but the same message
23:12:32gevaertskushal_12_27_200: that's normal. Just unplug the cable now. If it recovers we're done, else we try c250repair-mac
23:13:39kushal_12_27_200the boot image showed up, but as of now, nothing more
23:14:45gevaertskushal_12_27_200: then we probably need stronger measures. turn the player off (if it doesn't want to, hold the power button for at least 15 seconds and it will turn off)
23:15:21kushal_12_27_200i did
23:15:54gevaertsOnce it's off, put it into manufacturer mode (turn on hold, and press select while turning it on)
23:16:19gevaertsOnce the blue lights are on, turn off hold and plug the cable in
23:16:55gevaertsNow type "chmod +x c250repair-mac"
23:16:55 Quit DerDome ("Leaving.")
23:17:49gevaertsThen type "sudo ./c250repair-mac". Press enter when it asks
23:18:56 Quit x1jmp (Remote closed the connection)
23:19:10kushal_12_27_200-1 error :(
23:19:35gevaertsNothing else ?
23:20:14 Join DerDome [0] (
23:20:38rasherkushal_12_27_200: -1 error? Is this the rockbox bootloader?
23:20:49 Quit DerDome (Client Quit)
23:20:52rashererr.. ignore me
23:20:57 Quit Nevtus ("Gone")
23:21:16 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
23:23:00 Join RoC_MasterMind [0] (
23:23:09kushal_12_27_200I am sorry I am late in answering. I was posting the result at
23:23:53gevaertskushal_12_27_200: let's just try again. Unplug the player and turn it off
23:24:51gevaertsThen put it into manufacturer mode and plug it back in
23:25:47gevaertsThen run "sudo ./c250repair-mac" again
23:27:08kushal_12_27_200Result summary: [INFO] Patching application uploaded successfully!
23:27:28*gibbon_ smiles :)
23:28:24kushal_12_27_200I uploaded full results at
23:28:44kushal_12_27_200oh, there is something on the player screen
23:29:09gevaertsOK. It probably says you can turn it off now. Make sure hold is off, and turn off the player
23:29:55kushal_12_27_200it turned off when I pressed down button
23:29:58gevaertsThen turn it back on again and see what happens
23:30:17kushal_12_27_200after removing the cable? or without removing cable?
23:30:28gevaertsAfter removing the cable
23:31:34kushal_12_27_200it is the same thing, the booting screen comes and just stays there. Should I try it again?
23:31:40 Quit RoC_MasterMind ("Leaving")
23:32:23 Quit magmaniac ("Leaving")
23:32:44gevaertsNo. Put it in recovery mode again (hold+record while turning it on)
23:35:19kushal_12_27_200it says "Please connect USB"
23:35:33 Join S [0] (
23:36:04gevaertsOK. Now type "sudo mount -t msdos /dev/disk1 /tmp/sansa"
23:39:07gevaertsYou'll need the other files now. Make sure you have header.txt, firmware.mi4 and pribootLoader.rom on your desktop
23:39:10 Quit mozetti ("Leaving.")
23:39:25 Join Zarggg [0] (
23:42:51gevaertsNow type "sudo cp firmware.mi4 pribootLoader.rom header.txt /tmp/sansa"
23:43:06 Nick fxb__ is now known as fxb (
23:43:48gevaertsNow "sudo touch /tmp/sansa/snsa.fmt"
23:44:02gevaerts"sudo touch /tmp/sansa/sansa.fmt"
23:44:34BagderI don't think the file name matters, only the extension .fmt...
23:44:58gevaertsOK. then "sudo umount /dev/disk1"
23:45:43gevaertsThen "sudo hdiutil detach /dev/disk1"
23:46:09 Quit conando (Read error: 110 (Connection timed out))
23:46:28kushal_12_27_200unmounted and ejected
23:46:45gevaertsOK. Now unplug the cable, and see what happens
23:47:39kushal_12_27_200nothing happened. :( it still shows boot screen only
23:48:54gevaertsSo resetting the partition table and reinstalling the bootloader and firmware didn't help. I'm not sure what to try next
23:50:47 Quit Domonoky_ (Read error: 104 (Connection reset by peer))
23:53:27 Join robin0800 [0] (
23:54:06kushal_12_27_200maybe I can retry the last step from sudo cp firmware.mi4 pribootLoader.rom header.txt /tmp/sansa ?
23:56:13kushal_12_27_200What difference does "auto detect" and "MSD" mode make? the screen on the sansa said USB2.0 MSD so is it still in Auto detect mode or is it in MSD mode?
23:57:30gevaertsAs long as it doesn't start, that won't make any difference at all.

Previous day | Next day