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 2008-11-05

00:01:00kugelbertrik: y, it was a bit weird
00:01:25kugelbertrik: I think I also couldn't switch the buttonlight on without the display
00:02:55 Quit tyfoo2 (Connection reset by peer)
00:03:57funmankugel: can you post exactly what does the different combinations ?
00:04:24 Quit jgarvey ("Leaving")
00:04:39kugelI once tried to activate button light when there was no lcd driver
00:04:42funmanfor example, if you change the meaning of XPD interface with CCU_IO register, and then sets pin D7 high
00:04:52kugelit didn't work, so I thought it's on a different pin
00:05:28funmanif you comment the lcd_init() routine, can you put the button light on ?
00:05:37 Quit bertrik ("zZz")
00:05:49kugelI'll try later
00:05:57kugelor rather tomorrow
00:06:35kugelbut I'm fairly sure, since it was the second "version" of the bootloader which activated the buttonlight (back when we used the git repo)
00:06:36funmando you mean today ? :)
00:06:44kugelprobably ;)
00:07:18funmankugel: ok but it'd be nice to understand what is preventing the button light to be on, perhaps only setting the backlight on (it is only a power supply after all)
00:07:51funmani don't want to hurry you however, i know i am a bit impatient
00:08:26linuxstbamiconn: I don't know if you've noticed, but IIUC, the range coder in APE starts by skipping 1 byte of the input stream, then it reads 7 bits, then it reads 8 (unaligned) bits at a time. So if the encoder just shifted the stream by one bit, things would be simpler... (although admittedly probably not by much)
00:09:20kugelthen again, I've had problems to switch the backlight on without the buttonlight
00:09:48kugelwhich surprisingly wasn't the case after rebuilding the bootloader
00:10:03kugeli first thought it was my mistake, until it happened again
00:10:50 Quit bluebrother ("leaving")
00:11:35 Quit n1s ()
00:15:39 Part pixelma
00:15:59 Join reacocard__ [0] (n=reacocar@
00:16:06 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
00:16:12 Quit tvelocity[a] (Connection timed out)
00:17:19*JdGordon reads last nights irc log and isnt smiling :<
00:18:04 Quit domonoky (Read error: 104 (Connection reset by peer))
00:18:41funmanJdGordon: what are you referring to?
00:19:20JdGordonthe discussion on the quickscreen and descisions in the ml in general
00:20:07*JdGordon will bring it up tomorow eu time if he can be bothered
00:26:19 Join nuonguy [0] (
00:30:17 Quit reacocard_ (Read error: 110 (Connection timed out))
00:31:47 Join Thundercloud [0] (
00:34:51 Join DerDome [0] (
00:36:42 Quit reacocard__ (".")
00:38:44 Join reacocard_ [0] (n=reacocar@
00:40:28 Join lacrstech [0] (n=lacrstec@
00:40:30 Quit lacrstech (K-lined)
00:46:59 Quit bmbl ("Woah!")
00:50:22 Quit ender` (" Two things are infinite: the universe and human stupidity, even though I'm not yet sure about the universe. -- A. Einstein")
00:52:27 Quit MethoS- (Remote closed the connection)
00:53:07 Join Llorean [0] (n=DarkkOne@
00:56:16*amiconn wonders what's up with :\
00:58:29 Join super [0] (
00:58:48 Quit perrikwp (" ajax IRC Client")
00:59:28 Quit Bensawsome ("The awsome is gone :(")
00:59:47 Quit lasser ("ChatZilla 0.9.83 [Iceweasel 3.0.3/2008092816]")
01:03:38 Quit krazykit ("obama!")
01:04:55 Join tyfoo [0] (
01:10:47 Join mc2739 [0] (
01:12:00 Quit kkurbjun (Remote closed the connection)
01:12:57 Nick reacocard_ is now known as reacocard (n=reacocar@
01:14:12 Quit tyfoo ("Carpe diem")
01:14:54*amiconn found another tiny optimisation for the ARMv4 division
01:16:27 Quit Nico_P (Remote closed the connection)
01:18:26funmanamiconn: skip_byte()'s modification for bytebuffer increasing looks strange to me
01:19:23amiconn(or rather, in what way?)
01:20:10 Quit nuonguy ("This computer has gone to sleep")
01:20:16funmanoops no i'm mistaken sorry
01:20:38 Quit culture (Read error: 110 (Connection timed out))
01:22:09*amiconn would like some Gigabeat F/X APE benchmark results with r19010
01:24:13 Join brrybnds [0] (
01:29:30 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
01:30:26 Quit tessarakt ("Client exiting")
01:39:46 Join JdGordon|uni [0] (i=768aa360@gateway/web/ajax/
01:39:56***Saving seen data "./dancer.seen"
01:42:10 Quit Thundercloud (Remote closed the connection)
01:42:28 Quit vitja (Remote closed the connection)
01:52:57 Quit Bensawsome ("The awsome is gone :(")
01:55:01 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
02:09:37 Join fdinel [0] (
02:10:44 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
02:11:34 Quit DerDome ("Leaving.")
02:13:50 Join perrikwp [0] (i=982137bc@gateway/web/ajax/
02:31:20 Join kb1978 [0] (
02:34:29 Quit obo ("bye")
02:41:21 Quit DerPapst (Read error: 113 (No route to host))
02:55:36 Part kb1978
03:05:02 Join Darksair [0] (n=user@
03:12:45 Quit kugel (Read error: 131 (Connection reset by peer))
03:17:01 Join Out_Cold [0] (
03:17:54Out_Coldhey everyone.. i'm trying to figure out if rockbox is suitable for my needs. what applications do i use to sync to the ipod on both linux and windows?
03:18:07 Nick Darksair is now known as Darksair{away} (n=user@
03:18:13funmanany file manager
03:18:45Out_Coldso it's not like apple where you have to sync it? you just copy files?
03:18:54funmanyou just copy files
03:19:10Out_Coldand works with windows? and mp3 formats?
03:19:19JdGordon|unimp3 is a format
03:19:41funmanit works with more audio formats and OS than any DAP
03:19:56Out_Coldyes.. hence why i said 'mp3 format'
03:20:07JdGordon|uniyou said "mp3 formats"
03:20:39Out_Coldi'm just trying to figure out if my female friend is going to kill me for doing this to her ipod or thank me.. it is more user friendly i am assuming
03:21:18JdGordon|unithats a pretty bad assumption
03:21:36Out_Coldwell i'm asking
03:21:49JdGordon|uniconsidering how most females would find ipods fantastic when it comes to user friendliness
03:22:15Out_Coldyea.. but she has never had an ipod before. nor has she ever used itunes..
03:22:32Out_Coldwell it is reversible?
03:22:38JdGordon|uniit is
03:22:55JdGordon|uniif shes never used either then she wont know youve changed anything
03:23:09 Quit Bensawsome ("The awsome is gone :(")
03:23:16Out_Coldwell.. i'll take my chances... and if i come back castrated.. we'll have to laugh a bit about it
03:23:33Out_Colds used ipods.. just not itunes
03:25:00Out_Coldok well thanks for the info guys
03:25:14Out_Coldwill let ya know if i make it out of there alive
03:25:52 Quit Out_Cold ("Leaving")
03:29:48 Quit freqmod_qu (
03:37:20 Join Out_Cold [0] (
03:37:55 Nick Darksair{away} is now known as Darksair (n=user@
03:38:02Out_Coldumm another question.. is a nano 4th gen supported?
03:38:27LloreanNo, that's why it's not listed.
03:39:06Out_Coldthought maybe i missed it... is there another option that will support ogg files on an ipod?
03:39:58***Saving seen data "./dancer.seen"
03:40:06Out_Coldwell this sucks now... anyone working on the 4th gen OS?
03:41:26Out_Coldjust because or is there a particular reason?
03:41:36JdGordon|unibecause jobs hates us
03:42:05funmanall the new ipods firmwares are encrypted to prevent that
03:42:13LloreanFirmware's encrypted, hardware is mysterious. It's a significant challenge, and nobody who owns one has stepped forward and tackled it.
03:42:51Out_Coldhmm.. so i'm SOL and stuck with 30 000 ogg files that i can't load on this ipod.
03:43:15JdGordon|unino... you can tracnsode them
03:43:31JdGordon|uniwhich will give 30000 shit quality mp3s :D
03:43:59Out_Coldlol.. well... that's the last option i suppose..
03:44:59Out_Coldok well i'll have to go get me an old ipod so i can try out rockbox.
03:45:21Out_Coldi never ever wanted an ipod before i seen rockbox
03:45:53Out_Coldbut ok.. thanks for the support guys
03:46:05 Quit Out_Cold ("Leaving")
03:47:29 Quit funman ("leaving")
03:49:50 Quit NotADJ ("leaving")
03:54:20 Quit JdGordon|uni (" ajax IRC Client")
03:55:23 Quit mc2739 ()
03:58:21 Join HBK- [0] (
04:05:59 Quit HBK (Nick collision from services.)
04:06:01 Nick HBK- is now known as HBK (
04:07:26 Join Ginger [0] (
04:07:29 Quit Ginger (Client Quit)
04:09:01 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
04:10:07 Join blkhawk- [0] (
04:16:55 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
04:25:09 Nick Darksair is now known as Darksair{away} (n=user@
04:25:24 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
04:25:56 Quit blkhawk (Read error: 110 (Connection timed out))
04:26:05 Nick blkhawk- is now known as blkhawk (
04:26:15 Join Chris__ [0] (
04:26:48Chris__hi, im wondering if anyone can tell me what language games for rockbox can be coded in?
04:27:25Chris__im assuming since its based on linux any language works works fine, but i'm not really sure...
04:27:30LloreanIt's not based on Linux
04:27:34LloreanAnd the language is "C"
04:27:44Chris__c++ wouldnt work?
04:27:54 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
04:28:06Chris__oh i was mixing up some stuff i read on ipodlinux..
04:28:11Chris__my b
04:28:28LloreanC++ would not work, without a lot of additional work.
04:29:21Chris__well i've been wanting to learn c anyway its not too different, do you know a site that points me in the right direction for making applications?
04:29:57Chris__i tried but it only talks about working on rockbox itself
04:30:06LloreanPlugins are an aspect of Rockbox
04:30:11LloreanThat's why they're called "plugins"
04:30:29Chris__thanks, sometimes i miss the obvious things...
04:30:38LloreanRockbox isn't designed as a full OS.
04:30:42Chris__okay this is perfect thanks,
04:30:50LloreanIt's a media player firmware.
04:31:56Chris__well this might take me a while, but thanks for putting me in the right direction, bye!
04:32:16 Quit Chris__ ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
04:35:05 Nick intrados1 is now known as intrados (
04:37:33 Join aarcane [0] (
04:39:32 Quit Bensawsome ("The awsome is gone :(")
04:55:30 Join miepchen^schlaf_ [0] (
05:03:12 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
05:05:45 Nick Darksair{away} is now known as Darksair (n=user@
05:30:33 Quit perrikwp (" ajax IRC Client")
05:40:01***Saving seen data "./dancer.seen"
05:46:35 Quit cool_walking_ (" ajax IRC Client")
05:47:39 Quit Horscht ("I am root. If you see me laughing, you better have a backup")
05:47:44 Nick fxb is now known as fxb__ (
06:00:44 Quit Acky (Read error: 104 (Connection reset by peer))
06:00:59 Join Acksaw [0] (
06:01:12 Join massiveH [0] (
06:27:33 Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/
07:19:06 Join massive_H [0] (
07:20:04 Quit massiveH (Nick collision from services.)
07:20:07 Nick massive_H is now known as massiveH (
07:20:54 Part Llorean
07:21:14 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:24:38 Join massive_H [0] (
07:26:08 Quit massiveH (Nick collision from services.)
07:26:10 Nick massive_H is now known as massiveH (
07:27:56 Quit brrybnds ("Konversation terminated!")
07:34:31 Quit AndyI (Read error: 145 (Connection timed out))
07:34:45 Join AndyI [0] (i=AndyI@
07:40:02***Saving seen data "./dancer.seen"
07:40:53 Quit Rob2223 ()
07:42:24 Join Rob2222 [0] (
07:43:31 Join crculver [0] (
07:43:42crculverAny chance we'll see Ogg Theora support on the iPod?
07:43:46crculverIt supposedly has low CPU demands.
07:44:00JdGordonlow for desktops probably
07:45:04crculverDoes the Rockbox MPEG player work with special hardware, or does it decode MPEG purely in software?
07:45:17JdGordonpurely in software
07:45:19scorchethe latter
07:49:11 Quit shodanX ("Lost terminal")
07:52:40 Join goffa [0] (n=goffa@
07:54:56 Join Llorean [0] (
07:55:04 Quit Llorean (Read error: 104 (Connection reset by peer))
07:55:18 Join Llorean [0] (
08:04:18 Quit goffa_ (Read error: 110 (Connection timed out))
08:07:11 Quit miepchen^schlaf_ ()
08:14:11 Join bertrik [0] (
08:17:31 Quit jhulst (Remote closed the connection)
08:22:25 Nick crculver is now known as crculver_Away (
08:27:01 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
08:29:26 Quit skipper (Remote closed the connection)
08:30:40 Quit gromit`` (Read error: 110 (Connection timed out))
08:30:55 Quit BigBambi (Read error: 113 (No route to host))
08:38:58 Join Rob2223 [0] (
08:42:41 Quit massiveH ("Leaving")
08:43:56 Quit bertrik ("Leaving")
08:47:34amiconnZagor: Do you have an idea what's going on on the build server? If it's handed out the Player build, it fails with some weird errors (happened 3 times already)
08:49:25Zagorbut it also succeeded twice after the last 215-error failure
08:50:04ZagorI have no idea what causes this
08:50:27GodEateroverly prone to interference from high solar activity ?
08:52:56 Join ender` [0] (
08:53:25 Quit BHSPitMonkey (Remote closed the connection)
08:57:11 Join nuonguy [0] (
08:57:22 Quit Rob2222 (Read error: 110 (Connection timed out))
09:00:33amiconnZagor: The next two Player builds were done on a different build server
09:00:52amiconnAh no...
09:03:25 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
09:03:49 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
09:09:36amiconnHmm, test_codec is influenced quite a bit by LCD speed. APE speed figures are noticeably lower on iPod Photo than on iPod G2 and Mini G2
09:10:18*amiconn thinks this should probably be fixed
09:10:21 Join petur [0] (
09:10:49 Join shodanX [0] (
09:11:58 Quit GodEater (" ajax IRC Client")
09:17:07 Join GodEater [0] (i=c2cbc962@gateway/web/ajax/
09:21:33 Quit grndslm ("Leaving")
09:25:37linuxstbamiconn: Any suggestions? We could decrease the update frequency to about 5 seconds, and use lcd_update_rect instead of a full lcd_update.
09:27:13 Join skipper [0] (n=skipper@
09:27:34amiconnWe could also count the calls to lcd_update(), time an empty lcd_update() loop with the same number of iterations at the end, and then subtract the times
09:27:43*JdGordon would really like to hear more opinions on 9525 and the quickscreen item chooser list thingy
09:28:06amiconnIt's probably not only the lcd_update(), but also the lcd_puts*
09:28:26 Quit skipper (Client Quit)
09:35:33 Join MethoS- [0] (
09:38:55 Quit MethoS- (Remote closed the connection)
09:40:04***Saving seen data "./dancer.seen"
09:42:49 Join Thundercloud [0] (
09:43:41JdGordonOK, If I completly change the 4 added LANG strings should the lang version be bumped (and remove deprecated strings also) or not bother and just warn people to redo their voices?
09:45:00linuxstbIf you change strings significantly, aren't they new strings? Meaning deprecate the old ones, and add four new ones?
09:45:41JdGordonnone were added since, and the target lines wernt changed
09:58:51 Join lacrstech [0] (n=lacrstec@
09:59:00 Join viimrles [0] (n=viimrles@
10:04:30 Quit GodEater (" ajax IRC Client")
10:04:53 Join GodEater [0] (i=c2cbc962@gateway/web/ajax/
10:07:23JdGordonB4gder: your the lang guy (?)... which should I do?
10:08:18B4gderif they aren't longer used in their previous context I suggest new ids and blanking the older ones
10:08:34B4gderif they're "just modified" then just modify them
10:08:57JdGordontotally replaced...
10:09:38JdGordonhmm.. OK, I'll do it properly, but we should do a sort and clean up some time before 3.1
10:10:05JdGordonlinuxstb: (noeone else seems to have an opinion), did you try the context menu?
10:10:52linuxstbJdGordon: No, but my imagination tells me I will like it more than the current huge list.
10:25:37 Quit n17ikh|Lappy ()
10:28:01 Quit Thundercloud (Read error: 104 (Connection reset by peer))
10:31:27*JdGordon hopes thats the last QS commit for a while :p
10:40:22 Join cars1106 [0] (n=8d9ad9b5@gateway/web/cgi-irc/
10:40:27cars1106hey all
10:41:08 Quit cars1106 (Client Quit)
10:44:45*gevaerts wonders about the deltas for JdGordon's latest commit
10:45:04JdGordonshould have gone down?
10:45:21gevaertsThey did, but not for Player and Ondio
10:45:36JdGordonin the name of consistancy
10:46:05gevaertsAh, that would explain things :)
10:52:26amiconngevaerts: Player and Ondio have no quickscreen...
10:52:48JdGordonwhich is what he what saying... + deltas
10:56:09linuxstbThat's caused by adding a context menu for all setting items with the reset option?
10:57:20*gevaerts thinks that this reset thing is a great idea
11:03:33linuxstbJdGordon: Did your commit remove the old reset-settings feature?
11:06:16 Join dany_21a_ [0] (
11:08:44JdGordonits 10 lines of code and I tihnk a more logical place for it (being inside the setting screen)
11:09:00 Join kushal_12_27_200 [0] (n=kushal@
11:12:41 Quit JdGordon ("Konversation terminated!")
11:14:36 Join JdGordon [0] (
11:25:13 Join n1s [0] (n=nils@rockbox/developer/n1s)
11:27:26 Quit kachna|lappy (Read error: 113 (No route to host))
11:27:52 Nick Darksair is now known as Darksair{away} (n=user@
11:32:21 Nick fxb__ is now known as fxb (
11:34:48 Join funman [0] (
11:36:06Zagoris there a reason why codecs/asap, demac and spc are not called lib* ?
11:36:30Zagorthe output is called lib*.a so it would be more consistent of those dirs were called lib too
11:36:59Zagoran Tremor
11:37:37B4gderI don't think there's any particular reason
11:37:56funmanZagor: hello, did you notice FS #9527 ? (the author looked for you yesterday)
11:39:22Zagoroh dear. so ff3 craps up the DOM. thanks for the heads-up.
11:40:07***Saving seen data "./dancer.seen"
11:41:43amiconnFor demac there is a reason. libdemac is a sub-folder of that. demac itself contains the stand-alone decoder code
11:42:28 Join kugel [0] (n=chatzill@unaffiliated/kugel)
11:42:47 Part homielowe
11:45:14peturZagor: it works fine in minefield (ff3.1 beta2)
11:45:46 Join homielowe [0] (n=homielow@unaffiliated/homielowe)
11:47:03funmanhum and fine in firefox3.0 as well, now, (checking the link given in the subject)
11:48:02peturit didn't here...
11:53:15 Join PaulJam [0] (
11:53:19 Quit dany_21a_ (Remote closed the connection)
11:53:42 Join dany_21a_ [0] (
11:56:09 Join mofux [0] (
11:56:32 Nick Darksair{away} is now known as Darksair (n=user@
12:10:13 Join miepchen^schlaf [0] (
12:13:04 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:21:33 Join robin0800 [0] (
12:21:47 Quit linuxstb (Read error: 110 (Connection timed out))
12:25:45JdGordonsoap: is there any reason the ipod accesspry patch cant be commited?
12:26:13soapbesides my apparent failure to sync it? ;)
12:26:20JdGordonapart from that
12:26:56soapNo, nobody has reported new "bad" behavior in either the iPod's interaction with accessories or in behavior elsewhere.
12:27:26 Quit kugel (Read error: 110 (Connection timed out))
12:27:34Zagorso, does the irc log reader work with ff3 or not?
12:28:58n1sZagor: for me at least, saratoga's comment broke the "today" view yesterday but it works when it's from a past day
12:29:39Zagorn1s: how about the seconds view?
12:30:18n1sZagor: seems to work just fine
12:31:13Zagorso his css fix is as wrong as it looks
12:32:28 Join kachna|lappy [0] (
12:36:42 Join goffa_ [0] (n=goffa@
12:42:00 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
12:42:28 Join kugel [0] (n=chatzill@unaffiliated/kugel)
12:47:47 Quit JdGordon (Read error: 104 (Connection reset by peer))
12:48:36 Quit goffa (Read error: 110 (Connection timed out))
12:50:03 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
12:50:03 Quit PaulJam (Read error: 110 (Connection timed out))
12:52:53Zagorfunman: also please try the "show/hide seconds" in your browsers.
12:53:10Zagoror whoever has various versions installed...
12:54:27Zagorah, actually it's probably the show/hide joins that breaks
12:54:37Zagorbecause that class _is_ on a table row
12:54:59JdGordonyeah, show/hie joins causes very wide tables
12:55:07Zagorright, fixing
12:55:09JdGordonseconds is fine here though (firefox 3.0.3)
12:55:32Zagorso half of DerPabst javascript patch was right :-)
12:56:34Zagorreload and try again
12:58:55JdGordonyep, working
12:59:03Zagorgood. thanks
13:11:36 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
13:16:43Zagoramiconn: is the demac standalone decoder included for a reason? we don't actually use it, do we?
13:21:59amiconnThat's linuxstb's work...
13:22:57linuxstbZagor: I used Rockbox SVN as the "home" of demac/libdemac. It's my work originally.
13:24:11Zagorlinuxstb: can you find a reason why I should not move libdemac to the codec root and remove the standalong compiler? as part of a cleanup.
13:25:20linuxstbAs I said, it's the "home" of demac. Are you saying you're not happy with that?
13:27:00ZagorI'm not sure I understand what you mean by that. I would like a common naming scheme for the codec libraries. namely lib<blah>.
13:27:05 Quit kachna|lappy (Read error: 110 (Connection timed out))
13:30:53 Quit robin0800 (Success)
13:31:08 Quit miepchen^schlaf ()
13:32:18 Join bughunter2 [0] (
13:32:32 Quit pixelma2 (Nick collision from services.)
13:32:41 Join pixelma2_ [0] (n=marianne@rockbox/staff/pixelma)
13:32:43 Nick pixelma2_ is now known as pixelma2 (n=marianne@rockbox/staff/pixelma)
13:33:28 Quit amiconn (Nick collision from services.)
13:33:34 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
13:33:38 Quit kugel (Read error: 110 (Connection timed out))
13:34:16 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
13:35:18linuxstbZagor: "demac" is a APE decoder I wrote from scratch. Part of "demac" is libdemac, which is used by Rockbox. But Rockbox SVN is used as the home of the "demac" project.
13:35:43 Part dany_21a_
13:35:43Zagorah, now I understand
13:36:10 Join dany_21a_ [0] (
13:40:08***Saving seen data "./dancer.seen"
13:40:17 Join kugel [0] (n=chatzill@
13:42:23JdGordonjhMikeS: ping?
13:49:31 Quit dany_21a_ (Remote closed the connection)
13:49:48 Join dany_21a_ [0] (
13:53:39 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
13:58:56n1sFS #9401 sounds like a hardware issue to me, any ipod video owners around that can comment?
14:01:19JdGordonif its hardware shouldnt it be trying to reset itself while rockbox is running (if it thinks usb is connected) or refuse to shutdown (if it thinks there is a charger)
14:01:28JdGordonmaybe a outdated bootloader?
14:04:53n1sI don't know but as he seems to be the only ipod video user affected by this i don't know if we can or should do anything about it
14:05:22linuxstbI remember in the early days of Rockbox, a couple of people reported that Rockbox always misdetected USB attached. On these ipods, the Apple firmware would show a constant "charging" symbol, even when nothing was attached, so we just put it down to a hardware problem.
14:06:50 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
14:07:00 Quit bughunter2 (Read error: 145 (Connection timed out))
14:13:08n1slinuxstb: so if the OF thinks it is charging it's a hw issue?
14:13:19 Part B4gder
14:13:39n1scharging even when no chagrer is attached even
14:15:23linuxstbI would say so. I think the user needs to boot into the Apple firmware, and see what happens there.
14:15:56*n1s posted a question to the reporter in the task
14:16:51*n1s also wonders if there has been any progress on the plugin localization patch since august, it's a shame to see it just sitting in the tracker
14:20:09kugeln1s: he (the creator) wrote about that in the mailing list recently in the settings vs binsize discussion
14:22:51linuxstbThat would be a nice new feature for 3.1...
14:23:25n1sI agree but the last update in the tracker was from aug 13
14:24:25linuxstbAs kugel said, midgey posted to the mailing list recently about it. I can't remember the details, but I think it was basically "very busy at the moment, will return to it soon".
14:24:58funmani can't bare that business !
14:26:35n1slinuxstb: yes, i found it
14:28:07n1sIf he don't have time to finish it maybe he could post the current state to the tracker and say what's missing so someone else can do it
14:32:02 Join robin0800 [0] (
14:37:21 Join LambdaCalculus37 [0] (i=44a04303@gateway/web/ajax/
14:37:31 Quit kushal_12_27_200 ("Leaving")
14:38:51funmani wonder if these inand chips were designed to prevent concurrent implementation (especially open source)
14:39:25 Quit XavierGr (
14:39:25 Quit amiconn (
14:39:25 Quit pixelma2 (
14:39:25 Quit linuxstb (
14:39:25 Quit shodanX (
14:39:25 Quit blkhawk (
14:39:25 Quit HellDragon (
14:39:25 Quit Jabone (
14:39:25 Quit blithe (
14:39:25 Quit desowin_ (
14:39:25 Quit esthar (
14:39:25 Quit Tuplanolla (
14:39:25 Quit BlakeJohnson86 (
14:39:25 Quit Shiny (
14:39:25 Quit scorche|sh (
14:39:35funmanthe e200v1 uses such a chip if i understand correctly
14:39:52funmanbut the controller is much closer to the hardware than the one on sansav2 i believe by reading ata-sd-pp.c
14:40:05NJoinblkhawk [0] (
14:40:05NJoinXavierGr [0] (n=xavier@rockbox/staff/XavierGr)
14:40:05NJoinamiconn [50] (n=jens@rockbox/developer/amiconn)
14:40:05NJoinpixelma2 [0] (n=marianne@rockbox/staff/pixelma)
14:40:05NJoinlinuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
14:40:05NJoinshodanX [0] (
14:40:05 Join HellDragon [0] (n=jd@Wikipedia/HellDragon)
14:40:05NJoinShiny [0] (
14:40:05NJoinBlakeJohnson86 [0] (
14:40:05NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
14:40:05NJoinesthar [0] (
14:40:05NJoinblithe [0] (
14:40:05NJoinJabone [0] (i=jpylvana@
14:40:05NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
14:40:05NJoindesowin_ [0] (n=desowin@
14:40:09 Join J-23 [0] (
14:45:17 Quit bmbl (No route to host)
14:47:03Zagorwhat does svn status "switched" mean?
14:48:41Zagorand how do I unswitch it?
14:48:58n1smaybe this
14:49:54Zagorwell I sure haven't ran any switch command, and yet my libspc dir is listed as "S" by svn stat :-| hence the red build.
14:50:56linuxstbn1s: Yes, I'm not sure why he gave a patch to Bagder, rather than just posting to the tracker...
14:50:56 Join tyfoo [0] (
14:52:48 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
14:53:13 Join z35 [0] (
14:53:24 Quit Seed ("cu, Andre")
14:53:26 Quit z35 (Read error: 104 (Connection reset by peer))
14:57:20 Join Febs [0] (n=chatzill@
15:06:15funmanI'll take another way: i'll disable functions by modifying the OF until it doesn't boot anymore
15:07:59funmanit's the way i learned how windows and linux worked, it can't be bad :)
15:13:50 Nick Darksair is now known as Darksair{away} (n=user@
15:17:26 Join Schmogel [0] (
15:25:31 Quit kugel (Read error: 110 (Connection timed out))
15:30:04 Nick crculver_Away is now known as crculver (
15:33:46 Join bughunter2 [0] (n=Jelle@
15:36:49 Quit tchan ("WeeChat 0.2.7-dev")
15:37:00 Join tchan [0] (
15:37:55 Nick Darksair{away} is now known as Darksair (n=user@
15:40:11***Saving seen data "./dancer.seen"
15:59:01 Join mf0102 [0] (
16:09:48 Part LinusN
16:19:16 Join jgarvey [0] (
16:36:28n1sBagder: what do you think of FS #9338 ?
16:43:55n1sand Zagor while you are looking at the build system, do you consider FS #9159 a bug? (i'm not sure i do)
16:45:33ZagorI think I do consider it a bug. while not terribly important, it is hardly the desired behaviour.
16:46:58funmancan't fullzip depend on all ?
16:47:27 Quit robin0800 (Read error: 104 (Connection reset by peer))
16:47:36funmando you want to ignore the dependancy for faster execution of make?
16:47:37Zagoryes it can, but then make has to traverse the whole makefile tree to verify everything is built
16:48:09funmanif you use make fullzip instead of make;make fullzip it's not a problem though
16:48:38Zagorno. that is probably the only sane behaviour.
16:49:44superyes we can
16:50:07Zagorsuper: ?
16:50:28LambdaCalculus37super: ?
16:54:50 Part wpyh
16:58:50 Quit mf0102 ("Ex-Chat")
17:01:25*LambdaCalculus37 reads through the latest patch at FS #8806
17:03:47ZagorI don't get it. he needs >1MB ram for playing mods?
17:04:14LambdaCalculus37Zagor: I didn't get that either. linuxstb had pointed that out in the comments.
17:04:38LambdaCalculus37In whatever case, this patch will need significant reworking to become a codec.
17:06:00*LambdaCalculus37 spots tabbing *everywhere*
17:06:33LambdaCalculus37Not to mention significant cleanup!
17:08:01 Quit funman ("leaving")
17:10:21 Join tvelocity [0] (
17:12:06 Join {phoenix} [0] (
17:15:34LambdaCalculus37Zagor: <−− What does he mean by the recent merging of the codec and malloc buffers? Did I miss something, perchance?
17:16:32Zagorbeats me. I try to stay away from the playback code. :-)
17:17:30*LambdaCalculus37 summons a playback code guru
17:20:05 Join DerPapst [0] (
17:20:16 Quit aarcane (Remote closed the connection)
17:20:57ZagorDerPapst: hi. as you might have seen, I did apply half your javascript patch too. we were half right, both of us :)
17:21:04DerPapstZagor: thanks for accepting the irc reader patch. Though the closing message is kinda confusing since it seems you have accepted the js.
17:21:36Zagoronly half of it. the seconds style isn't on a table-row. only the join style.
17:21:50ZagorI realized that only after I closed the bug
17:21:53DerPapstheh. my patch only fixed the joins style
17:22:06DerPapstthe other was correct of course. didn't touch it ;)
17:22:30DerPapst ;)
17:23:02Zagorhah. well, I read like a moose. :-)
17:23:17DerPapstbut now everything's fine :D
17:23:47DerPapstand i'm happily stealing the reader idea for ipodlinux :P
17:24:19*LambdaCalculus37 cheers DerPapst on ;)
17:24:43DerPapstthough no perl. it's too crazy for me ;)
17:25:28DerPapstit took me about 2 hours just for the 4 liner to fix those urls with nicks in them ;)
17:25:56 Join kronflux [0] (
17:26:48LambdaCalculus37jhMikeS: Ping, what's the FS # for the beast charging patch?
17:29:05 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
17:29:26gevaertsLambdaCalculus37: FS #9312
17:29:30DerPapstLambdaCalculus37: 9312
17:29:34DerPapsttoo late
17:30:05n1sLambdaCalculus37: yes the codec and malloc buffers (each 512KB were merged into one huge 1MB buffer a couple of weeks ago)
17:30:37n1sas a first step to reducing codec memory usage/waste
17:31:00LambdaCalculus37n1s: Ahh, okay. Thanks for the info. :)
17:31:09*LambdaCalculus37 has a look at FS #9312
17:31:32saratogais anyone here familar with how profiling works?
17:32:16jchillerup am
17:32:18jchillerupI am
17:32:46jchillerupI haven't ever touched rockbox, though
17:33:13jchillerup(strongly considering it at the moment)
17:33:39DerPapstyou should! :)
17:33:47LambdaCalculus37I'm going to try out FS #9312 to see how it's working. What is its status, though? Is it in a "committable" state yet, or is there something else that has to be worked out with it first?
17:33:53jchillerupyeah, but I kindda wanna join the linux4nano project
17:34:23saratogai don't think they really have a project exactly, just a website if the encryption is ever cracked
17:34:35jchillerupI'm in need of a MP3 player, and I think the ipods look better than the competition in terms of hard drive capacity vs. physical size
17:34:40saratogaanyway, i'm sure i need to edit the codeclib makefile to profile it, but i'm not sure how
17:35:03jchillerupThey have a tremendous amount of info on all the hardware, and they are working on getting the chips dumped atm
17:35:16jchillerupsaratoga: add -p and -pg if it's a gcc compiler
17:35:38Zagorjchillerup: ...and as soon as they make a breakthrough we will port rockbox to it ;-)
17:36:01jchillerupyeah, and that's where I'm jumping on the bandwagon
17:36:55ZagorI wouldn't hold my breath though
17:37:20jchillerupNo, I know it looks somewhat dark :)
17:37:32saratogai don't think we use those flags for profiling in rockbox
17:37:59jchillerupThen I'm afraid I can't help you
17:38:09peturisn't there a wiki page about it?
17:38:25jchillerupWell, I can explain the basics for you if you don't know what it is altogether
17:38:35jchillerup(but I suppose you do since you asked fo rit :))
17:39:06*jchillerup could eat a cookie
17:39:19 Join Seed [0] (
17:39:34gevaertsjchillerup: TRUE / FALSE 946684799 NETSCAPE_ID 100103
17:40:01n1sZagor: you missed "make zip" in r19020
17:40:14***Saving seen data "./dancer.seen"
17:40:16jchillerupgevaerts: huh?
17:40:23gevaertsjchillerup: a cookie :)
17:40:24Zagorn1s: ah
17:40:27jchillerupoh :)
17:41:37DerPapsthehe :D
17:46:01jchillerupdo you serve pizza too?
17:46:29LambdaCalculus37Hmmm... So the beast charging patch has USB charging already, but what about with the AC adaptor?
17:50:36jchillerupLambdaCalculus37: ipod?
17:51:30Lloreanjchillerup: "The beats" refers to the Gigabeat S
17:51:33Lloreaner "the beast"
17:51:39jchillerupOh, didn't know that
17:52:21LambdaCalculus37jchillerup: Not all of us have iPods, y'know.
17:52:27jchillerupNo, I know that
17:52:42jchillerupI'm trying to catch up with rockbox (and its lingo) in fact
17:53:00jchillerupContemplating on becoming a dev
17:53:40gevaertsjchillerup: it's Gigabeat S, so shortened to gigabeats, then typoed to gigabeast, so it's now the beast
17:54:07jchillerupI think I connected the dots myself, then :
17:54:44LambdaCalculus37Also because it's a beast of a target. ;)
17:54:53n1sLambdaCalculus37: from reading the comments in the task i think it should charge from AC too, I never got around to trying myself though
17:55:26LambdaCalculus37n1s: I'll apply the patch and build when I'm on lunch. My beast needs a charging, anyway.
17:55:32Lloreanjchillerup: The Gigabeat S is probably the most overpowered hardware any devs are working on right now.
17:56:12gevaertsLlorean: overpowered? It doesn't even play ape -c5000!
17:56:15n1sLambdaCalculus37: doesn't it always :/ the battery time is my primary problem with it
17:56:36jchillerupLlorean: overpowered, as in...?
17:56:54n1sjchillerup: a very much overkill cpu
17:57:03jchillerupsounds interesting :)
17:57:14LloreanProbably at least twice as fast as anything else right now, I think
17:57:27jchillerupGimme some numbers already!
17:57:35LloreanNumbers don't mean much here.
17:57:39LambdaCalculus37jchillerup: 532 MHz ARM11 SoC.
17:58:09LloreanIt's 532MHZ, but at half clock already outperforms a 300mhz different ARM processor, plus it has various coprocessors that can add interesting optimizations later, etc.
17:58:21LloreanSo "numbers" really don't describe its actual capabilities in a useful way.
17:58:29jchillerupWhat can it do in its original frimware
17:58:35LloreanNot much
17:58:54jchillerupsounds interesting
17:58:55LloreanThe Original Firmware is a WindowsCE installation that's pretty restrictive, and klunky.
17:59:08jchillerupEnter ROCKBOX.
17:59:26LambdaCalculus37It's essentially what went into the early Zune models, albeit with an exploit we used to run Rockbox on the beast patched.
17:59:32jchillerupanyway, sounds like a pretty good player once you're rid of the CE
17:59:47LloreanNo WiFi though
17:59:59jchillerup... but the battery life sucks?
18:00:05LambdaCalculus37jchillerup: Yes, and all we really, really need to finish is getting charging into the main code, and we're pretty much set.
18:00:08 Join robin0800 [0] (
18:00:08jchillerupIs this a rockbox issue or a player issue
18:00:23LambdaCalculus37jchillerup: It's a Rockbox issue, because we don't have complete power management in yet.
18:00:30jchillerupI se
18:00:37n1sboth really, the battery is pretty small and we don't make the best use of it
18:00:46LambdaCalculus37So you're looking at about, ohhh... ~6 hours.
18:01:06 Quit Zagor ("Client exiting")
18:01:54LambdaCalculus37But the performance of many codecs and video playback on the S is beautiful.
18:02:26LloreanLambdaCalculus37: That being said, it doesn't offer any realistic advantages vs. the Gigabeat F at the moment
18:02:54LloreanIn terms of "using the CPU" I mean
18:04:27 Quit tvelocity (Connection timed out)
18:04:51 Join gromit` [0] (
18:04:55LambdaCalculus37Llorean: That's true. The only codec I can really see gaining a massive benefit on the S is the Monkey's Audio codec, since we can now play up to -c4000 in realtime on it.
18:08:47 Quit petur ("work->home")
18:13:02 Join bertrik [0] (
18:13:38linuxstbWhat about midi with many channels? Is that playable on the "standard" targets?
18:14:55LambdaCalculus37linuxstb: Haven't tried.
18:17:10 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:19:32 Quit reacocard (".")
18:20:25n1slinuxstb: coldfire targets cope quite well, they play up to a max of 24 channels iirc pp does not do too well, skips a lot even if the max channels is much lower
18:22:29 Join kachna|lappy [0] (
18:22:46n1smainly this is because the coldfires can play at 22kHz and most pp targets cannot (this has changed recently though an i hav enot heard from any users trying midi on a 22kHz capable pp player
18:26:19n1sbtw, "someone" should turn that plugin into a codec :/
18:26:36LambdaCalculus37n1s: You can ask Strife89 when he comes back. AFAIK he plays a lot of MIDIs on a PP target (c250).
18:26:49*LambdaCalculus37 will add that to MrSomeonesList ;)
18:28:02 Quit bertrik (Remote closed the connection)
18:28:40n1sLambdaCalculus37: IIUC the c200 is not capable of 22kHz output in rockbox
18:29:46n1sbut some ipods and iriver h10 got that recently so a report from them would be nice
18:31:46n1sbtw, if anyone were to send me a pp player i would look at trying to optimize the midi player a bit for it :)
18:32:49 Join bertrik [0] (
18:33:52 Join MethoS- [0] (
18:39:35 Quit linuxstb (Read error: 110 (Connection timed out))
18:41:13DerPapstiirc midi has only 16 channels. do you mean voices?
18:41:42DerPapstor rather notes playing at the same time...
18:42:33 Join culture [0] (
18:50:58 Join Strife1989 [0] (n=michael@
18:51:09 Quit Strife1989 (Connection reset by peer)
18:51:22 Join Strife1989 [0] (n=michael@
18:52:05 Nick Strife1989 is now known as Strife89 (n=michael@
18:56:52 Join Nibbler [0] (
18:58:50 Join faemir [0] (
19:05:07saratogai think the gigabeat S has a pretty big advantage over the F in it's plus pad
19:06:08 Join reacocard [0] (n=reacocar@WL-112.CINE.HMC.Edu)
19:07:14 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
19:08:04 Quit Nibbler ("Ex-Chat")
19:08:08 Quit Darksair ("Use the Force, Luke!")
19:08:15 Quit mofux (Remote closed the connection)
19:10:38 Quit saratoga ("CGI:IRC (EOF)")
19:17:46 Quit amiconn (" rebboott")
19:18:26 Join reacocard_ [0] (n=reacocar@
19:18:50 Quit reacocard (Nick collision from services.)
19:18:58 Nick reacocard_ is now known as reacocard (n=reacocar@
19:20:35 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
19:22:15 Quit culture (Read error: 104 (Connection reset by peer))
19:24:17 Join culture [0] (
19:27:29LambdaCalculus37Strife89: How is MIDI performance on your c200?
19:27:35LambdaCalculus37n1s: I have a spare c240.
19:28:05Strife89LambdaCalculus37: Rather poor on anything other than simpler MIDIs.
19:28:40Strife89If there are a lot of instruments trying to play at once, the skipping becomes horrible.
19:28:46 Quit kronflux ("Leaving")
19:29:21LambdaCalculus37Strife89: Can you do me a favor, and send me two MIDI files? One should be a simple MIDI, and one should be a more complex one.
19:29:46*LambdaCalculus37 wants to test on some other PP targets
19:30:09Strife89LambdaCalculus37: Sure, let me look through my collection. I might have to listen to a few to be certain.
19:31:08LambdaCalculus37Strife89: Sure.
19:31:31*LambdaCalculus37 gets his iPod color and iPod video ready
19:32:05 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:32:19DerPapstsend some self played ones ;)
19:32:35 Join meven [0] (
19:32:38Strife89DerPapst: I have none. ;)
19:33:13DerPapsti have some.
19:35:37 Quit pixelma2 ("-")
19:36:36 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
19:37:38Strife89LambdaCalculus37: Okay, I've got three here.
19:38:08Strife89One skips none, one skips a little, one skips a lot.
19:38:48 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:40:16***Saving seen data "./dancer.seen"
19:40:43 Join mf0102 [0] (
19:40:48LambdaCalculus37Strife89: I have one here that's skipping like mad.
19:41:02Strife89LambdaCalculus37: I know the feeling.
19:46:40 Join Thundercloud [0] (
19:53:52 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
19:54:24 Join miepchen^schlaf [0] (
19:55:08LambdaCalculus37Strife89: Performance is rather poor on my iPod video on two of the files you sent me.
19:55:39Strife89It's Portal Player as well, isn't it?
19:55:59LambdaCalculus37Yes, PP5021 IIRC.
19:56:06LambdaCalculus37The iPod color is a PP5020.
19:56:34*LambdaCalculus37 wishes that his mini were in complete working order, but still needs to wait for the click wheel cable replacement
19:57:47*Strife89 would like an iPod to play with. :)
19:57:58LambdaCalculus37Strife89: As you told me in the email, there's only one file that works perfectly.
19:58:45Strife89So, what's the test for? Are you working on the plugin?
19:58:54LambdaCalculus37Just benching for now.
19:59:16*Strife89 nods.
19:59:21gevaertsStrife89: do you often copy files between internal flash and sd while playing music?
19:59:32LambdaCalculus37n1s mentioned that if he had a PP target, he may want to do some work on the midiplay plugin and perhaps make it a proper codec.
19:59:44Strife89gevaerts: Not very often, no, but I have done it.
19:59:57Strife89gevaerts: Success has been limited.
20:00:07gevaertsin what sense?
20:00:19Strife89LambdaCalculus37: I'd settle for speed improvements. :)
20:00:45Strife89gevaerts: Such as the copy operation never "ending".
20:00:50LambdaCalculus37Strife89: Turning it into a codec proper would open the window for tons of speed improvments. ;)
20:01:06Strife89gevaerts: On some occasions, at least.
20:01:17gevaertsDoes that happen more often while playing? Or am I misubderstanding things
20:01:22Strife89LambdaCalculus37: Full steam ahead! :)
20:01:42Strife89gevaerts: I haven't done enough copying to tell a difference.
20:01:59Strife89gevaerts: Other than that, it's usually slower when music is playing.
20:02:11gevaertsthat's expected I guess. Less CPU available
20:02:22 Quit martian67 (SendQ exceeded)
20:02:39*gevaerts has ordered two different microsdhc cards, so he'll probably be able to test a bit more soon
20:02:39Strife89Slower still when buffering, but of course, we know why. :)
20:03:07LambdaCalculus37n1s: Would you like to take my Sansa c240 for help in optimizing the midiplay plugin?
20:03:31 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
20:03:33Strife89I *think* it might be particularly error prone when it's in the middle of buffering, but my memory is limited.
20:04:50gevaertsCurrent thinking is that data corruption is more likely when the CPU is busy with other things (music, usb,...), so that would be expected
20:04:52Strife89With that said, what does Rockbox do if it has two read/write requests in line?
20:05:29 Quit bertrik (Remote closed the connection)
20:05:29 Quit Horscht ("IRC is just multiplayer notepad")
20:05:34gevaertsSince it does cooperative multitasking, and it doesn't do DMA for sd, that can't actually happen
20:06:21gevaertsI think so at least. There may be some unexpected yield()s in there as far as I know
20:08:48 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
20:08:48 Join bertrik [0] (
20:17:47 Quit martian67 (SendQ exceeded)
20:18:56 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
20:26:02*amiconn actually liked the independence of 'make zip' & friends from the main build
20:26:30amiconnThis way 'make zip' was possible even if the main build failed due to some plugin or similar
20:27:33*gevaerts spots that two people have now complained about this. Time to revert?
20:28:53amiconnLambdaCalculus37: In what way would making the midi plugin a codec allow for additional optimisations?
20:29:42amiconn(apart from the fact that this will be difficult, if at all possible)
20:30:14 Quit martian67 (SendQ exceeded)
20:30:46 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
20:33:31amiconnStrife89, gevaerts: Two read/write requests in parallel can never happen, because the SD driver is mutexed
20:33:48gevaertsah yes, mutexes too
20:34:50amiconnThe problem is that the we obviously don't configure the SD interface properly, so the transfer timing can cause problems
20:35:09amiconnThat should never happen with the interface configured properly
20:39:46 Quit bertrik (Read error: 104 (Connection reset by peer))
20:40:24*gevaerts tries to understand the PP clock control
20:40:56LambdaCalculus37amiconn: Speaking from a "not the best programmer" point of view, making midiplay into a codec would allow it to use the playback engine in Rockbox. As for optimizations, I have no solid explanation for that
20:43:50 Join bertrik [0] (
20:46:59 Join massiveH [0] (
20:48:04 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
20:48:28 Quit martian67 (SendQ exceeded)
20:49:15 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
20:54:04 Quit Bensawsome ("The awsome is gone :(")
20:55:15 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
20:55:15 Quit Thundercloud (Read error: 104 (Connection reset by peer))
20:58:54 Quit bertrik (Read error: 104 (Connection reset by peer))
21:03:01amiconngevaerts: There are most probably separate clock/delay control registers for the SD interface. This is an RE task...
21:04:03gevaertsamiconn: right now I want to see if setting the clock above 80MHz changes USB behaviour
21:04:58 Quit martian67 (SendQ exceeded)
21:05:34amiconnWhy should it?
21:05:43amiconnThe PP5020 is specced 80MHz
21:06:00 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
21:07:25pixelmathe voodoo approach?
21:07:56*gevaerts was actually confused a bit. PP5022 seems to be specced at 200MHz
21:08:35 Join einhirn [0] (
21:10:08amiconngevaerts: It is. The PP5021 has no public specs at all though
21:13:22 Quit mf0102 (Read error: 113 (No route to host))
21:14:21*gevaerts would like to know how clock related registers are set up by various OFs during USB
21:14:54gevaertsIf saratoga manages to get this jtag hookup done, that would be very helpful
21:15:36 Join bertrik [0] (
21:18:25 Join einhirn_ [0] (
21:20:25gevaertsamiconn: do you still need gigabeat F testing?
21:21:05amiconnWould be nice...
21:21:59amiconnIn addition to the standard realtime percentage of -c1000..-c5000, I would be interested in 2 more values (measured using the -c1000 test file) which need patching decoder.c
21:22:32amiconn(deactivating the predictor and stereo decorrelation resp. only the latter)
21:22:51gevaertscurrent svn?
21:23:19amiconnThe extra tests are for my stats which part of the codec needs what amount of performance
21:25:07 Quit massiveH ("Leaving")
21:27:07*gevaerts finds that an ipod build doesn't work on gigabeat
21:30:55 Join kugel [0] (n=chatzill@unaffiliated/kugel)
21:32:52 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
21:32:59gevaertsamiconn: the plain svn test is running now. What do I have to change specifically for the others?
21:33:05saratogagevaerts: my ears are burning
21:33:13saratogabut i was trying to jtag the V2, not V1
21:33:38gevaertsah, ok. I'll wait for someone else then :)
21:33:42saratogathough I suppose either of us could ask Bagder to borrow his JTAGed V1 and look at it
21:33:54saratogahow are you doing on free time?
21:34:05 Quit einhirn (Read error: 113 (No route to host))
21:34:51gevaertsreasonably well. I've never done jtag though, and I don't have the required hardware (yet)
21:34:54amiconngevaerts: In apps/codecs/demac/libdemac/decoder.c, disable the predictor and decorrelation by placing an #if 0 before line 169, and the #endif after line 184
21:35:23amiconnRun a test using the -c1000 test file (no need to try playback, it will only produce nasty sound, if at all)
21:36:04gevaertsok. Building that now
21:36:10amiconnThen move the #if 0 down after the predictor call, and do the same again (will be slower, almost as slow as the plain -c1000 test)
21:36:30amiconnPlease note that there's a dependency problem. Check that ape.codec is actually relinked
21:36:44amiconn(and complain to Zagor) ;)
21:36:48gevaertsI know. I'm doing a make clean to make sure
21:37:51*amiconn is currently running division call statistics in the sim
21:38:17amiconnI want to check whether a table-based approach for small divisors would pay off
21:38:28amiconn(multiplying with the reciprocal of the divisor)
21:40:20***Saving seen data "./dancer.seen"
21:44:00*gevaerts sees that c5000 clearly isn't realtime yet on the F :)
21:47:58 Quit bertrik ("reboot reboot reboot")
21:50:12 Join bertrik [0] (
21:56:05 Join FlynDice [0] (n=1813e0e3@gateway/web/cgi-irc/
21:56:51 Quit FlynDice (Client Quit)
21:57:00 Join FlynDice [0] (n=1813e0e3@gateway/web/cgi-irc/
21:57:04gevaertsamiconn: has the svn results
21:58:36amiconn470%->580% realtime for -c1000 is quite an improvement :)
21:59:04 Quit FlynDice (Client Quit)
21:59:39gevaertsIndeed not bad :)
22:01:18 Join rikai [0] (i=blah@unaffiliated/rikai)
22:02:31 Join goffa [0] (n=goffa@
22:03:18 Join petur [50] (n=petur@rockbox/developer/petur)
22:03:44gevaertsamiconn: "special" results at
22:08:48rikaiHm, pretty sure FS #7515 and FS #2660 on the bugtracker are the same issue...
22:09:19amiconnarm920t is slower in the entropy decoder and filters than coldfire, but it's faster in the predictor
22:09:52*amiconn thinks there must be some optimisation potential in the cf predictor then
22:11:51 Join Thundercloud [0] (
22:13:20 Quit goffa_ (Read error: 110 (Connection timed out))
22:14:30gevaertsThe F is now faster than the S for -c1000
22:25:57gevaertsamiconn: I seem to be getting slower results for -c1000 on the beast than your r18739 result (544% instead of 553%). Are your tests done with the standard test track from CodecPerformanceComparison?
22:26:38 Join Thundercloud_ [0] (
22:27:07 Join n17ikh|Lappy [0] (
22:27:21 Quit Thundercloud (Read error: 104 (Connection reset by peer))
22:41:36 Join NotADJ [0] (n=TDJACR@Wikipedia/Thedjatclubrock)
22:49:56 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
22:51:45 Quit Strife89 (Read error: 110 (Connection timed out))
22:52:44 Quit {phoenix} (Remote closed the connection)
22:54:54 Quit FOAD ("I'll be back")
22:57:03 Quit J-23 (Remote closed the connection)
22:58:01NotADJI installed a theme, everything is right except the Now Playing Screen,
22:58:12 Join FOAD [0] (
22:58:17NotADJIt has the same look across themes, just different backgrounds.
22:58:51amiconngevaerts: I use a test track that linuxstb gave me. That's probably not the same
22:59:06gevaertsok. That would explain the differences then
22:59:30amiconnFor some reason APE performance jumps a bit on the beast when changing things
22:59:40gevaertsNotADJ: where did you get the themes?
23:00:10 Quit tyfoo ("Carpe diem")
23:00:35NotADJgevaerts: ^
23:01:13gevaertsNotADJ: that site is actually not updated anymore since a while. You're very unlikely to find working themes there
23:01:22domonokyNotADJ: they are outdate... take a look at the wiki WpsGallerys
23:03:07gevaerts to be extra precise :)
23:04:44 Quit domonoky (Read error: 104 (Connection reset by peer))
23:05:03 Part dany_21a_
23:05:14linuxstbDoes belong to soap?
23:06:21linuxstbsoap: Is there a reason you keep that mirror active?
23:08:04 Quit BigBambi (Read error: 110 (Connection timed out))
23:08:38 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
23:14:23advcomp2019linuxstb, plus there was an user that tried to upload to it too
23:18:18 Quit robin0800 (Remote closed the connection)
23:22:22mcuelenaereany idea how a DAP could identify that the USB is connected? (not using GPIO)
23:22:48gevaertsYou could leave the controller enabled all the time and wait for actual usb traffic
23:23:06mcuelenaerethat's a possibilit
23:23:18mcuelenaerebut I don't think the OF implemented it that way (although they could)
23:23:38mcuelenaerewouldn't that be wasting battery?
23:23:59mcuelenaereI mean, one extra GPIO channel isn't that much effort
23:24:13pixelmainteresting, lostlogic's server produces red builds again for the Player after a few green ones (at least the build table reports that) - but: the current available build for download has the correct revision number, I would have expected the last green one to be available (r19019)
23:24:14gevaertsIt would, yes. What other way would there be? (/me assumes that they didn't use an ADC instead of a GPIO)
23:24:24 Join krazykit [0] (
23:24:27mcuelenaereI looked into the possiblity of ADC
23:24:50mcuelenaereis there a possiblity that you first need to enable something before a GPIO channel becomes valid?
23:24:53gevaertsActually, maybe they put the controller to sleep.
23:25:10bertrikmaybe an interrupt is triggered?
23:25:13gevaertsThat would still waste a bit of battery, but not very much
23:25:26mcuelenaerebecause I see in the OF there are some sys_charge_rise_int_handler routines
23:25:54*mcuelenaere 's not very familiar with IRQ rise or fall stuff
23:27:08 Quit jgarvey ("Leaving")
23:27:18mcuelenaerethere's also some references to g_usb_charge in the OF; could they use USB charging as an indicator that USB is connected?
23:27:56gevaertsThat doesn't sound like a smart way to do it
23:28:33mcuelenaerenope, but I wouldn't know how they would do it otherwise (except for the monitor USB traffic possibility)
23:29:25 Quit einhirn_ (Read error: 54 (Connection reset by peer))
23:31:42 Join tvelocity [0] (n=tony@
23:35:09 Join robin0800 [0] (
23:35:20 Quit bertrik ("Leaving")
23:35:39 Join dany_21a_ [0] (
23:37:23 Quit Schmogel (Read error: 104 (Connection reset by peer))
23:37:52 Join funman [0] (
23:39:11funmanmcuelenaere: to read the buttons on the clip and m200v2 (v4), you have to write to a certain pin to change the meaning of other input pins
23:39:38mcuelenaerehmm so something like an internal hw switch?
23:40:21***Saving seen data "./dancer.seen"
23:40:36saratogaso they multiplex two different physical pins to the same internal address?
23:41:15funmani can't answer
23:41:34amiconnThat's not uncommon for SoC's
23:42:06funmani'm trying to understand why my bootloader calls create_thread()
23:42:40funmandid you already have the idea of dumping the stack content in UIE() ? (to have more than the faulty instruction address)
23:43:11amiconnBagder: around?
23:43:36 Quit petur ("baby feeding time")
23:43:55amiconnCould you try to find out what's up with and the Player build?
23:44:54amiconnIt fails often, but not always, with weird errors. And even though the last one failed (according to the build table) the downloadable .zip states that it is r19022 (the .txt inside says the same)
23:50:42Bagderit's really not easy to spot any problem
23:53:03 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
23:53:23JdGordonah bugger, i forgot to ask lostlogic about it yesterday
23:54:20 Join DerDome [0] (
23:54:32 Quit bluebrother ("leaving")
23:55:00 Quit BigBambi (Read error: 113 (No route to host))
23:56:46 Quit LambdaCalculus37 (" ajax IRC Client")
23:58:02JdGordonBagder: lost says he wont have time to look at it for a couple of weeks so remove the server
23:58:05 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
23:58:07 Quit Febs (Read error: 110 (Connection timed out))
23:58:29BagderI tried to build for player manually as rbclient on his host and it worked fine... :-/

Previous day | Next day