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 2005-07-11

00:00:08*[IDC]Dragon logpeeking spots amiconn
00:00:16amiconnhi Jörg
00:00:25amiconnI have a question
00:00:31[IDC]Dragonme too
00:00:46[IDC]Dragon(you first then)
00:00:53amiconnWhen you inrroduced video playback, you made a plugin to fine-tune the
00:01:22 Join DarkkOne [0] (
00:01:31[IDC]Dragonfps_test, yes
00:01:35amiconnWhen you introduced video playback, you made a plugin to fine-tune the frequency to match the internal LCD frequency
00:01:54amiconnHow did that work? What did it send to the LCD?
00:02:10amiconnI'll need to do the same for the iriver LCD
00:02:11[IDC]Dragonalternating b & w
00:02:23[IDC]DragonI can look for it
00:02:30amiconnThat'd be nice :)
00:03:03*[IDC]Dragon looks
00:04:45[IDC]Dragonfound it
00:05:02[IDC]Dragonuh, this is old, direct timer programming
00:05:23amiconnNp, I'll do the same on iriver first
00:05:41amiconnBtw, the iriver has way less luxury concerning timers
00:05:58amiconnSH1 has 5 timers, while coldfire 5249 only has 2
00:06:15amiconnOne is already taken for the tick, so...
00:07:15amiconnThe other is used in the core for the backlight fading, but I already prepared a mechanism to 'steal' it (backlight resorts to simple on/off then)
00:08:14[IDC]Dragonemail sent
00:08:36[IDC]Dragonnow my question:
00:08:56[IDC]DragonI've implemented your suggestion to bootbox
00:09:12[IDC]Dragonthere is a flaw with the Ondio now
00:09:17amiconnYes, saw that
00:09:33[IDC]Dragonin case of missing ajbrec.ajz
00:10:02[IDC]Dragonwhen going to USB mode, the MMC behaves a bit "sticky"
00:10:14 Quit lodesi ("Leaving")
00:10:32[IDC]Dragonthe PC has some problem to write to it for the 1st time
00:10:46 Join courtc [0] (
00:11:10[IDC]Dragonmaybe I need to reset it or so?
00:11:29[IDC]DragonMMC=internal memory
00:12:06[IDC]Dragonto describe the symptom again:
00:12:10amiconnGoing to USB should reset it, and going back from USB should reset it again
00:12:29[IDC]DragonI renamed my ajbrec away to make rolo miss it
00:12:49[IDC]Dragonafter plugging, I want to rename it properly again
00:13:30[IDC]DragonPC complains, promts to overwrite the file with itself (?),
00:13:52[IDC]Dragonafter yes, I have a 0 byte file
00:14:28[IDC]Dragondeleting that orphan and copying a fresh ajbrec works, and everything from then on
00:14:59amiconnI tested the same way, and had no problems (before your latest change)
00:15:38[IDC]Dragonbut that required a restart
00:15:47[IDC]Dragonwith the old version
00:15:49amiconnNo, it didn't
00:16:08amiconnThere was just the problem that the archos didn't react to usb while either of the 2 messages were displayed, after dismissing both messages with a button USB worked
00:16:57[IDC]Dragonaha, then it probably restarted itself inbetween
00:17:51 Join jlee [0] (
00:18:23 Quit DarkkOne ("Chatzilla [Firefox 1.0.4/20050511]")
00:19:31amiconn[IDC]Dragon: I see you're calling system_reboot() now...
00:21:34[IDC]Dragonnot changed, I did before, too
00:22:15Roriwhy is Rockbox no longer loading?
00:22:24Roriit loads shitty iRiver
00:22:41amiconn[IDC]Dragon: Ah, right
00:22:50amiconnYou just moved it to a separate function
00:23:34amiconn[IDC]Dragon: You don't have the double-mount prevention as in apps/main.c
00:23:43amiconnPerhaps that's a problem?
00:23:55RoriI just put the bleeding edge on and Rockbox does not load. The iRiver FW loads instead.
00:24:19amiconnPlus, I *think* if you stick in an MMC at boot, bootbox will try to boot from that MMC instead of internal
00:24:39amiconnI'm not sure whether I should consider this a bug or a feature...
00:25:00amiconnRori: Did you get the right version?
00:25:13RoriI got the build on the normal site
00:25:22amiconnThe h120/h140 build was renamed to h120 because of h110/h115 support
00:25:49RoriI always get that one
00:26:05amiconnYes, and that's *wrong* since a few days
00:26:15Roriawwww crap
00:26:22amiconn(iirc you do have a h120 or h140)
00:26:55Roriyeah got
00:29:30Roribleh. still no remote stuff
00:29:42Rorioh well
00:33:23[IDC]Dragonamiconn: where does the internal MMC get reset on USB mode?
00:34:10 Quit Coldtoast ("Peace and Protection 4.22")
00:36:07amiconnusb_slave_mode() -> ata_init() -> ata_enable()
00:36:29 Quit Zagor ("Client exiting")
00:36:33amiconn...and then even another ata_enable()
00:36:38amiconn(false this time)
00:37:31amiconnThis sequence looks a bit messy to me, could be done better
00:37:41amiconn...but that shouldn't result in such a problem
00:38:07amiconnBtw, if the pc is able to display the flash contents, it is definitely able to access it
00:38:34[IDC]Dragonyes, but it has a writing problem at first
00:38:54amiconnIf reading works, writing does also work, unless the card is software-write-protected
00:39:12[IDC]Dragonusb_slave_mode() is handled by the usb thread
00:39:59[IDC]Dragonafter SYS_USB_CONNECTED_ACK
00:41:04amiconnBootbox or rockbox should behave identically here, because the code is the same.
00:41:15amiconnI never observed this behaviour
00:42:01[IDC]DragonI'll email you my build
00:42:08amiconnOh, btw
00:42:26amiconnI would have done the power button check in a different way
00:42:42amiconn(1) It is only necessary on software poweroff platforms
00:42:51amiconn(recorder fm/v2, ondios)
00:43:22amiconn(2) The button driver sends a convenient evenet for that, no necessity to fiddle with different buttons yourself
00:46:18amiconnYou can even do this check for all platforms, just on hardware poweroff platforms the button driver will never send this event
00:46:34[IDC]Dragonok, thanks
00:48:14[IDC]DragonSYS_POWEROFF is only sent after holding it for a while?
00:49:20amiconnMaybe this is unsuited for the charging screen....
00:49:40amiconnI wonder what rc = 2 does (in both rockbox and bootbox)
00:50:00amiconnImho that can't work, and would hang the box
00:50:33*amiconn tries with real rockbox...
00:51:02[IDC]Dragonused to wfm
00:51:42amiconnThis is indeed a glaring bug in the charging screen
00:52:10amiconnPut recorder v1 or player on charger, charging screen appears and animates
00:52:24amiconnPress Off or Stop - screen freezes
00:52:52amiconnThis must be an ooold bug
00:53:09amiconnThe Off/Stop detection is nonsense the charging screen
00:53:15 Part BoD[]
00:53:26[IDC]Dragonit was early work of mine
00:54:04amiconnMaybe it works on the fm/v2...
00:54:17amiconnCan fm/v2 recorders power down with the charger connected?
00:54:42[IDC]DragonI think not
00:54:52amiconnTechnically, only the recorder v1 really needs this behaviour (no poweroff with charger connected) due to the software charging
00:55:01amiconn...but players have it as well...
00:55:11[IDC]Dragonbit I may be mistaken
00:55:51[IDC]Dragonit doesn't power down because plugging the charger turns it on
00:56:05[IDC]Dragonany FM owner around?
00:57:17[IDC]Dragonthe FM powers down again if the s/w doesn't flip a power hold bit
00:57:36amiconnYes, but does it also do that when the charger is connected?
00:57:52[IDC]Dragonso, by releasing that but you can make it "land"
00:58:31[IDC]Dragonmy typos are compensating each other...
00:58:44[IDC]Dragonland=power down
00:59:17amiconnThe question is whether this also works with the charger connected, or whether the charger forces power on
00:59:27[IDC]DragonI forgot if that bit is also necessary with external power
01:00:25 Join BBub_ [0] (
01:01:11amiconn[IDC]Dragon: Oh, and btw, the (BUTTON_RIGHT|BUTTON_REL) part is also unnecessary
01:01:32amiconnThe only device without ON button is the ondio, and that doesn't have charging...
01:02:50amiconnI'd suggest a verison string for bootbox
01:06:09[IDC]Dragonpure coincidence... ;-)
01:07:45 Quit BBub (Read error: 104 (Connection reset by peer))
01:07:45 Nick BBub_ is now known as BBub (
01:07:58amiconnAny objections against removing some archos debug stuff?
01:08:09 Quit CheeseBurgerMan (Read error: 110 (Connection timed out))
01:08:11amiconn(rtc ram and mas)
01:08:55[IDC]Dragonnot from me
01:09:10*[IDC]Dragon the option slasher
01:17:01 Quit zezayer ("Chatzilla [SUSE 1.0.4-1.1/20050511]")
01:19:02 Quit Harpy (Read error: 110 (Connection timed out))
01:24:09amiconn[IDC]Dragon: That saves ~1 KB on recorder, and >2 KB on player
01:40:10amiconnWill you fix the charging screen?
01:47:14 Join webguest59 [0] (
01:47:34webguest59hi guys :)
01:47:50webguest59can i ask a quick graphical question?
01:49:25webguest59how many shades of gray can we have in maximum with the future rewriting graphical stuff?
01:49:35webguest59potentaly :)
01:49:59webguest59actually 4 right?
01:50:14amiconn4 shades in the core, the native capability of the lcd
01:50:25amiconn(on h1x0)
01:50:44webguest59yes h1xx i speak :)
01:50:44amiconnThat's already possible
01:51:14amiconnSome plugins will eventually be ablue to display more shades
01:51:17webguest59ok, i tought in future this capacity will increased
01:51:35amiconn...but not the core
01:51:44webguest59ok just plugin
01:51:55webguest59like rockboy i assume
01:52:11***Saving seen data "./dancer.seen"
01:52:12amiconnRockboy needs just 4 shades, and has 4 shades
01:52:27webguest59a ok
01:52:37amiconnBesides, the mechanism to simulate more is quite cpu hungry, so no go for rockboy
01:52:45webguest59what plugins will be concerned?
01:52:46amiconn(which is cpu hungry by itself)
01:53:27amiconnjpeg viewer, mnadelbrot, grayscale demo (in fact just a technical demo for the grayscale library)
01:53:50webguest59oh good
01:53:59webguest59and WPS screen, no?
01:54:15amiconnNo, wps will get no more than 4 shades
01:54:31amiconnI said plugins, no core
01:55:08webguest59apear the graphical stuffs are your territory :)
01:55:19amiconnThe lower level, yes
01:55:34BBubdoes the vu-meter work for anyone?
01:55:55amiconnBBub: Yes, for all archos owners except player ;)
01:56:04BBubah, ok
01:56:09amiconnNot yet on iriver if you mean that...
01:56:11webguest59:( not for iriver
01:56:30webguest59anyone working on it?
01:56:42webguest59for irivers
01:57:17webguest59i assume it's not a priority
01:57:20crwljpeg viewer for h1xx would be nice
01:57:23webguest59but more usual
01:57:42webguest59crwl: yes :P
01:58:35webguest59amiconn: do you know a bit the part of code about peak meter...
01:59:01amiconnYes, but the peakmeter code itself isn't the problem
01:59:25amiconnThe peakmeter needs actual peak values, and that's provided by th ehardware decoder on archos
01:59:36BBubah, i see
01:59:40webguest59not in iriver
01:59:51 Quit edx (Read error: 145 (Connection timed out))
01:59:51amiconnSomeone needs to add peak calculation to the iriver playback code
02:00:37webguest59it's not Slasheri that work on the playback?
02:01:02webguest59maybe he can do it with a bit of luck :)
02:01:14 Quit hicks (Remote closed the connection)
02:01:43amiconnIirc stripwax is trying to make the peakmeter work on iriver
02:01:50 Join edx [0] (
02:02:19webguest59oh, this peak calculation like you said?
02:04:43webguest59well, thanks amiconn for this answers
02:05:09BBubis it also possible for the h1xx-remote to have a dimming bg-light like the main-unit?
02:05:23thegeekthe remote does not use a led
02:05:30thegeekit uses that "el-tape" stuff
02:05:33thegeekor what it's called
02:05:37thegeekcan not be dimmed
02:05:38BBubah, ok
02:05:47webguest59hi thegeek
02:05:54thegeekhi there?
02:06:01webguest59you know a lot about remote stuff?
02:06:10thegeeknot really
02:06:16thegeekI'm not even a developer
02:06:25webguest59a ok :)
02:06:29thegeekI just remember linus saying that about the led and remote
02:06:57*tvelocity is travelling to germany tomorrow! :D
02:07:21webguest59it'll be very good a browser and WPS screen in remote :(
02:07:35webguest59but apear it's not for soon
02:08:07webguest59eh tvelocity, lucky man
02:08:28webguest59have a good traveling
02:08:56 Quit lImbus (" HydraIRC -> <- *I* use it, so it must be good!")
02:09:05webguest59not holidays soon here :(
02:09:12tvelocitywhy not?
02:09:30webguest59works :(
02:09:44webguest59september for my vacations
02:10:15webguest59but the time run quicly furtunatly :)
02:10:19BBubtvelocity: where are you going?
02:11:17tvelocityat a village near munich
02:11:37Moosbaviere sounds good
02:11:43Moosthe Alps
02:12:00tvelocityi have relatives there, i'm half german
02:12:13BBubah, too bad thats quit far from me ;)
02:13:44webguest59amiconn: have you planed for us a new graphicals stuffs?
02:14:28 Join ashridah [0] (
02:14:33 Quit [IDC]Dragon ("CGI:IRC")
02:16:24Moosamiconn: us=iriver rockbox owners :)
02:16:52Mooshehe webgest amiconn had already worked a lot
02:17:04 Quit tvelocity ("Leaving")
02:17:33Mooswebguest: and we can already congrate and thank him for it
02:17:39Moosbe patient :)
02:18:00webguest59oki scuse
02:18:25Moosno worries :), it's just for him
02:19:31webguest59thanks all for all your rockbox works, I know you do it in your free time
02:19:42webguest59thanks a lot and good night
02:20:02 Quit webguest59 ("CGI:IRC")
02:21:45Moostime to got to bed here too
02:21:51Moosnight all
02:22:08 Quit Moos (" Want to be different? HydraIRC -> <-")
02:35:40 Join ac [0] (
02:35:44achi all
02:37:35 Join Kotjze [0] (
02:37:47 Part Kotjze
02:50:45 Join CheeseBurgerMan [0] (~me@
02:59:49 Quit ac ("CGI:IRC")
03:02:03 Join Nibbler [0] (
03:10:16 Quit ashridah ("gotta go.")
03:29:14 Join eleom1 [0] (
03:29:54 Quit eleom1 (Client Quit)
03:30:04 Join eleom1 [0] (
03:34:33 Quit eleom1 (Client Quit)
03:52:13***Saving seen data "./dancer.seen"
04:01:08 Quit Sucka ("a bird in the bush is worth two in your house")
04:02:22 Join TCK [0] (
04:05:30 Join QT [0] (as@area51.users.madwifi)
04:15:50 Quit QT_ (Read error: 110 (Connection timed out))
04:17:45 Quit cYmen ("zZz")
04:24:10 Quit Maxime (
04:24:10 Quit solex (
04:24:10 Quit amiconn (
04:24:10 Quit pill (
04:24:10 Quit crwl (
04:24:10 Quit Cassandra (
04:24:10 Quit lostlogic (
04:24:10 Quit floam (
04:24:10 Quit bill2or3 (
04:24:10 Quit dwihno (
04:24:10 Quit ze (
04:24:10 Quit Rori (
04:24:10 Quit bill20r3 (
04:24:10 Quit Bagder (
04:24:10 Quit Hadaka (
04:24:10 Quit ansivirus (
04:24:10 Quit CheeseBurgerMan (
04:24:10 Quit odd (
04:24:10 Quit CoCoLUS (
04:24:10 Quit mbr (
04:24:10 Quit crashd (
04:24:10 Quit Seed (
04:24:10 Quit Lynx_awy (
04:24:10 Quit QT (
04:24:10 Quit BBub (
04:24:10 Quit ferenczy (
04:24:10 Quit einhirn (
04:24:10 Quit alxcm (
04:24:10 Quit thegeek (
04:24:10 Quit ghostiger (
04:24:10 Quit dionoea (
04:24:10 Quit silencer_ (
04:24:10 Quit pike (
04:24:10 Quit crash_ (
04:24:10 Quit Rick (
04:24:10 Quit C-Keen (
04:24:10 Quit gromit` (
04:24:10 Quit edx (
04:24:10 Quit Febs (
04:26:18NJoinQT [0] (as@area51.users.madwifi)
04:26:18NJoinCheeseBurgerMan [0] (~me@
04:26:18NJoinedx [0] (
04:26:18NJoinBBub [0] (
04:26:18NJoinMaxime [0] (
04:26:18NJoinsolex [0] (
04:26:18NJoinlostlogic [0] (
04:26:18NJoinferenczy [0] (
04:26:18NJoineinhirn [0] (
04:26:18NJoinamiconn [0] (
04:26:18NJoinpill [0] (
04:26:18NJoinalxcm [0] (
04:26:18NJoinRori [0] (
04:26:18NJoinbill2or3 [0] (
04:26:18NJointhegeek [0] (
04:26:18NJoinfloam [0] (
04:26:18NJoinghostiger [0] (~ghostiger@0f044bb2050252db.session.tor)
04:26:18NJoinCassandra [0] (
04:26:18NJoinFebs [0] (
04:26:18NJoingromit` [0] (~gromit`
04:26:18NJoinsilencer_ [0] (
04:26:18NJoinSeed [0] (
04:26:18NJoinRick [0] (rick@Rick.user)
04:26:18NJoindionoea [0] (
04:26:18NJoinpike [0] (
04:26:18NJoinHadaka [0] (
04:26:18NJoinBagder [0] (
04:26:18NJoinansivirus [0] (
04:26:18NJoinze [0] (
04:26:18NJoindwihno [0] (~dw@
04:26:18NJoincrwl [0] (
04:26:18NJoincrashd [0] (
04:26:18NJoinodd [0] (
04:26:18NJoinLynx_awy [0] (
04:26:18NJoinC-Keen [0] (
04:26:18NJoinCoCoLUS [0] (
04:26:18NJoincrash_ [0] (
04:26:18NJoinmbr [0] (
04:26:18NJoinbill20r3 [0] (
04:26:46CtcpVersion from freenode-connect!freenode@connect.utility.freenode
04:27:06 Join Ismo_ [0] (
04:27:06 Quit Ismo (Read error: 104 (Connection reset by peer))
04:40:02BBubwhere does rockbox store its config on the iriver?
04:45:05 Quit Cassandra (
04:45:05 Quit pill (
04:45:05 Quit amiconn (
04:45:05 Quit Maxime (
04:45:05 Quit lostlogic (
04:45:05 Quit solex (
04:45:05 Quit crwl (
04:45:05 Quit floam (
04:45:05 Quit dwihno (
04:45:05 Quit ze (
04:45:05 Quit bill2or3 (
04:45:05 Quit bill20r3 (
04:45:05 Quit Bagder (
04:45:05 Quit Rori (
04:45:05 Quit ansivirus (
04:45:05 Quit Hadaka (
04:45:05 Quit mbr (
04:45:06 Quit CoCoLUS (
04:45:06 Quit odd (
04:45:06 Quit CheeseBurgerMan (
04:45:06 Quit crashd (
04:45:06 Quit Seed (
04:45:06 Quit Lynx_awy (
04:45:06 Quit silencer_ (
04:45:06 Quit alxcm (
04:45:06 Quit einhirn (
04:45:06 Quit BBub (
04:45:06 Quit crash_ (
04:45:06 Quit pike (
04:45:06 Quit C-Keen (
04:45:06 Quit dionoea (
04:45:06 Quit Rick (
04:45:06 Quit thegeek (
04:45:06 Quit QT (
04:45:06 Quit gromit` (
04:45:06 Quit ghostiger (
04:45:06 Quit ferenczy (
04:45:06 Quit Ismo_ (
04:45:06 Quit Febs (
04:45:06 Quit edx (
04:46:39NJoinIsmo_ [0] (
04:46:39NJoinQT [0] (as@area51.users.madwifi)
04:46:39NJoinCheeseBurgerMan [0] (~me@
04:46:39NJoinedx [0] (
04:46:39NJoinBBub [0] (
04:46:39NJoinMaxime [0] (
04:46:39NJoinsolex [0] (
04:46:39NJoinlostlogic [0] (
04:46:39NJoinferenczy [0] (
04:46:39NJoineinhirn [0] (
04:46:39NJoinamiconn [0] (
04:46:39NJoinpill [0] (
04:46:39NJoinalxcm [0] (
04:46:39NJoinRori [0] (
04:46:39NJoinbill2or3 [0] (
04:46:39NJointhegeek [0] (
04:46:39NJoinfloam [0] (
04:46:39NJoinghostiger [0] (~ghostiger@0f044bb2050252db.session.tor)
04:46:39NJoinCassandra [0] (
04:46:39NJoinFebs [0] (
04:46:39NJoingromit` [0] (~gromit`
04:46:39NJoinsilencer_ [0] (
04:46:39NJoinSeed [0] (
04:46:39NJoinRick [0] (rick@Rick.user)
04:46:39NJoindionoea [0] (
04:46:39NJoinpike [0] (
04:46:39NJoinHadaka [0] (
04:46:39NJoinBagder [0] (
04:46:39NJoinansivirus [0] (
04:46:39NJoinze [0] (
04:46:39NJoindwihno [0] (~dw@
04:46:39NJoincrwl [0] (
04:46:39NJoincrashd [0] (
04:46:39NJoinodd [0] (
04:46:39NJoinLynx_awy [0] (
04:46:39NJoinC-Keen [0] (
04:46:39NJoinCoCoLUS [0] (
04:46:39NJoincrash_ [0] (
04:46:39NJoinmbr [0] (
04:46:39NJoinbill20r3 [0] (
04:46:55 Quit Cassandra (
04:46:55 Quit pill (
04:46:55 Quit amiconn (
04:46:55 Quit Maxime (
04:46:55 Quit lostlogic (
04:46:55 Quit solex (
04:46:55 Quit crwl (
04:46:55 Quit floam (
04:46:55 Quit dwihno (
04:46:55 Quit ze (
04:46:55 Quit bill2or3 (
04:46:55 Quit bill20r3 (
04:46:55 Quit Bagder (
04:46:55 Quit Rori (
04:46:55 Quit ansivirus (
04:46:55 Quit Hadaka (
04:46:55 Quit mbr (
04:46:55 Quit CoCoLUS (
04:46:55 Quit odd (
04:46:55 Quit CheeseBurgerMan (
04:46:55 Quit crashd (
04:46:55 Quit Seed (
04:46:55 Quit Lynx_awy (
04:46:55 Quit silencer_ (
04:46:55 Quit alxcm (
04:46:55 Quit einhirn (
04:46:55 Quit BBub (
04:46:55 Quit crash_ (
04:46:55 Quit pike (
04:46:56 Quit C-Keen (
04:46:56 Quit dionoea (
04:46:56 Quit Rick (
04:46:56 Quit thegeek (
04:46:56 Quit QT (
04:46:56 Quit gromit` (
04:46:56 Quit ghostiger (
04:46:56 Quit ferenczy (
04:46:56 Quit Febs (
04:46:56 Quit Ismo_ (
04:46:56 Quit edx (
04:48:09NJoinIsmo_ [0] (
04:48:09NJoinQT [0] (as@area51.users.madwifi)
04:48:09NJoinCheeseBurgerMan [0] (~me@
04:48:09NJoinedx [0] (
04:48:09NJoinBBub [0] (
04:48:09NJoinMaxime [0] (
04:48:09NJoinsolex [0] (
04:48:09NJoinlostlogic [0] (
04:48:09NJoinferenczy [0] (
04:48:09NJoineinhirn [0] (
04:48:09NJoinamiconn [0] (
04:48:09NJoinpill [0] (
04:48:09NJoinalxcm [0] (
04:48:09NJoinRori [0] (
04:48:09NJoinbill2or3 [0] (
04:48:09NJointhegeek [0] (
04:48:09NJoinfloam [0] (
04:48:09NJoinghostiger [0] (~ghostiger@0f044bb2050252db.session.tor)
04:48:09NJoinCassandra [0] (
04:48:09NJoinFebs [0] (
04:48:09NJoingromit` [0] (~gromit`
04:48:09NJoinsilencer_ [0] (
04:48:09NJoinSeed [0] (
04:48:09NJoinRick [0] (rick@Rick.user)
04:48:09NJoindionoea [0] (
04:48:09NJoinpike [0] (
04:48:09NJoinHadaka [0] (
04:48:09NJoinBagder [0] (
04:48:09NJoinansivirus [0] (
04:48:09NJoinze [0] (
04:48:09NJoindwihno [0] (~dw@
04:48:09NJoincrwl [0] (
04:48:09NJoinbill20r3 [0] (
04:48:09NJoinmbr [0] (
04:48:09NJoincrash_ [0] (
04:48:09NJoinCoCoLUS [0] (
04:48:09NJoinC-Keen [0] (
04:48:09NJoincrashd [0] (
04:48:09NJoinodd [0] (
04:48:09NJoinLynx_awy [0] (
04:48:50CtcpVersion from freenode-connect!freenode@connect.utility.freenode
05:07:00 Quit leoncamel ("Leaving")
05:15:31 Quit CheeseBurgerMan ()
05:22:53 Join |Xinux| [0] (
05:34:45 Part |Xinux|
05:52:15***Saving seen data "./dancer.seen"
06:38:06 Join Strath [0] (
06:39:40 Join CheeseBurgerMan [0] (~me@
06:52:39 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
06:56:16 Quit pill (Read error: 110 (Connection timed out))
07:07:50 Join ashridah [0] (
07:34:55 Join pill [0] (
07:52:19***Saving seen data "./dancer.seen"
07:53:53 Quit CheeseBurgerMan ()
08:17:31 Join LinusN [0] (
08:39:34 Join odd_ [0] (
08:39:34 Quit LinusN (Read error: 104 (Connection reset by peer))
08:39:35 Quit HCl (Read error: 104 (Connection reset by peer))
08:39:37 Join QT_ [0] (
08:40:39 Quit Strath (
08:40:39 Quit crwl (
08:40:39 Quit Cassandra (
08:40:39 Quit amiconn (
08:40:39 Quit lostlogic (
08:40:39 Quit solex (
08:40:39 Quit Maxime (
08:40:39 Quit floam (
08:40:39 Quit pill (
08:40:39 Quit dwihno (
08:40:39 Quit bill2or3 (
08:40:39 Quit ze (
08:40:39 Quit bill20r3 (
08:40:39 Quit Bagder (
08:40:39 Quit Rori (
08:40:39 Quit Hadaka (
08:40:39 Quit ansivirus (
08:40:39 Quit odd (
08:40:39 Quit CoCoLUS (
08:40:39 Quit mbr (
08:40:39 Quit crashd (
08:40:39 Quit Seed (
08:40:39 Quit Lynx_awy (
08:40:39 Quit ashridah (
08:40:39 Quit C-Keen (
08:40:39 Quit crash_ (
08:40:39 Quit silencer_ (
08:40:39 Quit alxcm (
08:40:39 Quit BBub (
08:40:39 Quit pike (
08:40:39 Quit thegeek (
08:40:39 Quit dionoea (
08:40:39 Quit Rick (
08:40:39 Quit QT (
08:40:39 Quit gromit` (
08:40:39 Quit ghostiger (
08:40:39 Quit ferenczy (
08:40:39 Quit Febs (
08:40:39 Quit Ismo_ (
08:40:39 Quit edx (
08:41:04 Quit courtc (Read error: 104 (Connection reset by peer))
08:41:07 Quit edster (Read error: 104 (Connection reset by peer))
08:41:19 Join courtc_ [0] (
08:41:48 Join edster [0] (
08:42:19 Join LinusN [0] (
08:45:36 Join jlee_ [0] (
08:45:48NJoinbill20r3 [0] (
08:45:48NJoinansivirus [0] (
08:45:48 Join B4gder [0] (
08:45:48NJoincrashd [0] (
08:45:48NJoinC-Keen [0] (
08:45:48NJoinStrath [0] (
08:45:48NJoinMaxime [0] (
08:45:48NJoinsolex [0] (
08:45:48NJoinlostlogic [0] (
08:45:48NJoinamiconn [0] (
08:45:48NJoinfloam [0] (
08:45:48NJoinCassandra [0] (
08:45:48NJoincrwl [0] (
08:45:54 Join godzirra_ [0] (
08:46:09NJoinLynx_awy [0] (
08:46:09NJoindwihno [0] (~dw@
08:46:10NJoingromit` [0] (~gromit`
08:46:20NJoinBagder [0] (
08:46:22 Join Rick [0] (
08:46:29CtcpVersion from freenode-connect!freenode@connect.utility.freenode
08:46:29***Server message 477: 'logbot #rockbox :[freenode-info] help freenode weed out clonebots, please register your IRC nick and auto-identify:'
08:46:35NJoinze [0] (
08:46:46 Join Ismo [0] (
08:46:46 Join Naked [0] (
08:46:48 Nick Naked is now known as Hadaka (
08:47:19 Join HCl [0] (
08:48:16NJoinmbr [0] (
08:48:18LinusNmorning amiconn
09:04:45 Quit pabs (Connection timed out)
09:04:45 Quit jlee (Read error: 110 (Connection timed out))
09:04:45 Quit godzirra (Read error: 113 (No route to host))
09:04:46 Quit HCl (Read error: 104 (Connection reset by peer))
09:04:46 Quit QT_ (Read error: 104 (Connection reset by peer))
09:04:46 Quit LinusN (Read error: 104 (Connection reset by peer))
09:05:01 Quit odd_ (Read error: 104 (Connection reset by peer))
09:05:06NJoinFebs [0] (
09:05:14 Join amiconn_ [0] (
09:05:14 Quit courtc_ (Read error: 104 (Connection reset by peer))
09:05:18 Quit ze (
09:05:18 Quit gromit` (
09:05:18 Quit dwihno (
09:05:18 Quit Cassandra (
09:05:18 Quit amiconn (
09:05:18 Quit Maxime (
09:05:18 Quit lostlogic (
09:05:18 Quit solex (
09:05:18 Quit C-Keen (
09:05:18 Quit crwl (
09:05:18 Quit Strath (
09:05:18 Quit floam (
09:05:18 Quit mbr (
09:05:18 Quit Hadaka (
09:05:18 Quit Ismo (
09:05:18 Quit Rick (
09:05:18 Quit Bagder (
09:05:18 Quit Lynx_awy (
09:05:18 Quit B4gder (
09:05:18 Quit bill20r3 (
09:05:18 Quit crashd (
09:05:18 Quit ansivirus (
09:05:18 Quit edster (Success)
09:05:19 Nick amiconn_ is now known as amiconn (
09:05:43 Join courtc_ [0] (
09:05:53 Join edster [0] (
09:06:08 Quit Nibbler (Read error: 104 (Connection reset by peer))
09:06:15DEBUGEOF from server (Connection timed out) (snapshot: netstuff.c line 545)
09:06:15***Saving seen data "./dancer.seen"
09:06:15***Started Dancer V4.16
09:06:15***Connected to on port 6667
09:06:15***Logfile for #rockbox started
09:06:16***Server message 501: 'logbot_ :Unknown MODE flag'
09:06:16Mode"logbot_ :+i" by logbot_
09:06:16 Quit logbot (Read error: 104 (Connection reset by peer))
09:06:16DEBUGInternal confusion. Unknown user "logbot!" quit. (snapshot: fplrun.c line 872)
09:06:16 Join logbot_ [242] (
09:06:16 Join edster [0] (
09:06:16 Join courtc_ [0] (
09:06:16 Join TCK [0] (
09:06:16 Join jlee_ [0] (
09:06:16 Join godzirra_ [0] (
09:06:16 Join Febs [0] (
09:06:16 Join amiconn [0] (
09:06:16 Join Bagder [0] (
09:06:16 Join crwl [0] (
09:06:16 Join lostlogic [0] (
09:06:16 Join solex [0] (
09:06:16 Join pabs [0] (
09:06:16 Join odd [0] (
09:06:16 Join dwihno [0] (~dw@
09:06:16 Join crashd [0] (
09:06:16 Join LinusN [0] (
09:06:16 Join Cassandra [0] (
09:06:16 Join gromit` [0] (~gromit`
09:06:16 Join mbr [0] (
09:06:16 Join crash_ [0] (
09:06:16 Join C-Keen [0] (
09:06:16 Join QT [0] (as@area51.users.madwifi)
09:06:16 Join HCl [0] (
09:06:16 Join Ismo [0] (
09:06:16 Join ze [0] (
09:06:16 Join B4gder [0] (
09:06:16 Join alxcm [0] (
09:06:16 Join dionoea [0] (
09:06:16 Join Rick [0] (rick@Rick.user)
09:06:16 Join Lynx_awy [0] (
09:06:16 Join Maxime`Mrn [0] (
09:06:16 Join Seed [0] (
09:06:16 Join bill2or3 [0] (
09:06:16 Join ashridah [0] (
09:06:16 Join edx [0] (
09:06:16 Join Ismo_ [0] (
09:06:16 Join silencer [0] (
09:06:16 Join thegeek [0] (na@
09:06:16 Join BBub [0] (
09:06:16 Join ferenczy [0] (
09:06:16 Join pike [0] (
09:06:21 Join Strath [0] (
09:06:22 Join Nibbler [0] (
09:06:26 Join floam [0] (
09:06:40 Join bill20r3 [0] (
09:06:45 Join ansivirus [0] (
09:07:23amiconnLinusN: Btw, sorry that my bug report for the v3 bootloader wasn't exact
09:07:34amiconnIt does indeed start rockbox with remote On, but only if you press it very short
09:07:39amiconnI did intuitively press it longer since it takes a while until you get visual feedback
09:07:52 Quit Ismo_ (Read error: 110 (Connection timed out))
09:07:58amiconn(especially since the bootloader no longer switches on the backlight)
09:08:12LinusNhehe, the bug report was correct, it was my testing that failed :-)
09:08:51LinusNi think i should release v4 pretty soon
09:09:16amiconnOh, btw. This line checking for reocrd || remote_record could be simplified quite a bit
09:09:18LinusNi have all the iriver firmwares except 1.63k for H110
09:09:25LinusNamiconn: figures
09:09:35 Join CoCoLUS [0] (
09:10:32amiconnLinusN: You're checking for each condition individually (and the bug was that the ORing together wasn't interpreted as intended by the compiler)
09:10:34 Join godzirra [0] (
09:10:40 Join jlee [0] (
09:10:44amiconnHowever, these values are bitmasks, so
09:11:02 Join Naked [0] (
09:11:05amiconnif(((button_status() & (BUTTON_REC|BUTTON_RC_REC))
09:11:11amiconnshould be sufficient
09:11:15 Nick Naked is now known as Hadaka (
09:11:27LinusNthen it will trigger on any remote button, that was the bug
09:11:28 Join Slasheri [0] (
09:11:51LinusNBUTTON_REMOTE is set for all remote buttons
09:12:16amiconnAh, that one
09:12:17LinusNand BUTTON_RC_ON triggered the check
09:12:26LinusNin v3
09:12:30amiconnI wonder why that BUTTON_REMOTE exists at all
09:12:49amiconnIt does break the bitmask concept somewhat
09:12:58LinusNi guess the intention is to be able to filter out remote buttons easily
09:13:23LinusNit's a qualifier, much like BUTTON_REL or BUTTON_REPEAT
09:13:59LinusNi'm not sure it's needed though
09:14:18amiconnIt looks like this concept was introduced on archos - and it has a severe flaw there
09:15:07amiconnThe remote buttons reuse (mostly) the same codes as the main buttons, but with BUTTON_REMOTE set
09:15:37amiconn...and this means you can't detect a main unit button press while the same remote button is pressed
09:16:06amiconnOn iriver, the remote buttons use bit 16 and upward, so its easy to detect remote buttons without the extra bit
09:16:19LinusNarchos can do that too
09:16:21 Join DangerousDan [0] (~Miranda@
09:16:23amiconnJust do if (button & 0xFFFF0000)
09:17:02amiconnbutton.h could define a general mask for that
09:17:26LinusNi think we can get rid of BUTTON_REMOTE and give the remote unique bits on archos
09:17:40amiconnOr better, button & 0x7FFF0000
09:17:50amiconn(to avoid clashes with the sysevents)
09:18:06amiconnSounds good
09:18:30LinusNis there an IS_SYS_EVENT()?
09:18:42amiconnI think so
09:19:04amiconnWe have to take care of the gmini version, in case it will ever picked up again
09:19:19amiconnOur buttons are currently ints....
09:20:04 Quit godzirra_ (Connection timed out)
09:20:05 Quit jlee_ (Read error: 110 (Connection timed out))
09:20:29LinusNamiconn: ah, maybe they should be longs
09:20:34 Join webguest67 [0] (
09:21:35webguest67Is the missing "latest" build for the H120 deliberate on the Daily Builds page ?
09:25:26*LinusN fixes
09:27:08webguest67was wondering if I'd missed something :)
09:31:41LinusNwebguest67: fixed, thanks for reporting
09:37:14SlasheriHi, i have just taken some measurements frmo iriver h140:
09:37:25SlasheriSupply voltage 3.700V
09:37:46SlasheriCPU idle power: 11 MHz 39 mA, 48 MHz 58 mA, 120 MHz 81 mA
09:38:15SlasheriDecoding (no matter what codec / what bitrate, or if paused or not): 48 MHz 74 mA, 120 MHz 115 mA
09:38:21SlasheriOptical out: +10 mA
09:38:28Slasheri(no matter if playing or not)
09:38:35SlasheriBacklight: +23 mA
09:38:47SlasheriDisk buffering: 48 MHz ~450 mA
09:39:14SlasheriDisk spinning: +110 mA
09:39:23Slasherihere is something :)
09:39:30Slasherii can take more if you have some ideas
09:40:09 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
09:41:21SlasheriThe cpu idle power might be lower however, if we turn off display module, dac etc. completely
09:46:36SlasheriAnd the battery level indicator is accurate in debug menu, +- 0.05V
09:48:04*HCl dreams odd...
09:48:08HCldon't you people sleep..
09:48:13RickI do
09:48:22HCllies, lies!
09:48:31 Join sox [0] (
09:48:39soxhey all
09:48:56RickHCl: I slept like 6 hours, woke up ~48min ago :P
09:49:05soxi can't seem to get my own wps to work, no matter what font i choose the wps is blank
09:49:06Rickprobably going to stay up all night now
09:49:14Rick12:49am now :<
09:49:15HCli just woke up :3
09:49:35HCli dreamt odd..
09:49:38HCllike always
09:51:05SlasheriWith full battery (4.2V) the system current (cpu 48 MHz) is 51 mA and empty battery (3.0V) 73 mA
09:54:17 Nick Lynx_awy is now known as Lynx_ (
09:57:01 Quit sox ("Time to move on...")
10:03:29LinusNSlasheri: wiki wiki wiki
10:06:33 Join Harpy [0] (
10:17:36 Join Rori [0] (
10:18:27SlasheriLinusN: i will do it later, have to go now :)
10:23:00 Nick webguest67 is now known as Godeater (
10:35:09 Quit Godeater ("CGI:IRC")
10:48:19 Join pill [0] (
11:06:17***Saving seen data "./dancer.seen"
11:45:08 Join tucoz [0] (
11:45:08 Quit tucoz (Client Quit)
11:45:27 Join tucoz [0] (
11:46:02tucozLinusN: are you here? If you are, have you read in the forums regarding the noise on boot?
11:46:14LinusNdo you have it?
11:46:23tucozvery seldom
11:46:26LinusNi have a potential fix if you want to try it
11:46:50tucozhehe, it's hard to try as it is a very random bug
11:47:12LinusNi'll give you a boot loader with the fix installed
11:47:22LinusNthen you tell me if you ever hear the noise again
11:47:32tucozok. hmm. Can I recieve that on the irc client?
11:47:41LinusNi'll put it up for download
11:47:47tucozoh, great. Thanks
11:48:11tucozcan I put that link in the forums as well. (h120 = yes)
11:49:03tucozWe had one experiencing that with the v3 bootloader.
11:50:54tucozSo, do you mute the sound chip in the bootloader in that potential fix, or do you do something else?
11:51:15LinusNi reset the uda1380
11:51:15 Join hicks [0] (
11:52:35tucozgreat. I think the problem is that actually, as we get the iriver buzzes in usb bootloader mode, until rockbox inits the uda1380
11:55:03LinusNwait, i'll try to patch mine first, just to be sure
11:56:16 Join cYmen [0] (
11:56:26tucozsure, anyway. Which one is it? the last link or the first. I'll have to try this later on when I get to a windows box.
11:57:14tucozBBub: I'll add that dl to the wiki
11:57:24BBubi also have an account ;)
11:57:32amiconntucoz: I believe the _nu version is without unicode support, i.e. the one for stock win 95/98
11:57:47LinusNtucoz: it is the same thing as the original fwpatcher.exe and fwpatchernu.exe
11:58:16*LinusN is bricktesting his h140
11:58:50LinusNworks fine
11:58:51tucozOk, only used the fwpatcher once
11:59:36tucozGreat. Then I'll post those links in the forums for the other noise-experiencing-guys to try as well?
12:00:37tucozBBub: ok, sorry. Posted. Great you found it then. Now the downloads are complete for once :)
12:00:44BBubnp ;)
12:01:59tucozLinusN: ok?
12:03:03LinusNtucoz: see above
12:03:10LinusN<LinusN> works fine
12:03:29tucozsorry, just wanted to make sure.
12:04:25LinusNthis was a good test for my new script that builds a new fwpatcher.exe with up-to-date md5 sums
12:04:36LinusNfor all firmwares
12:04:50LinusNBBub: thanks for the link
12:09:18 Join webguest86 [0] (
12:09:56webguest86Morning all - does anyone know if the fwpatcher checks the MD5 values after patching?
12:10:15tucozwebguest86: of course
12:10:43webguest86can I suggest that it says it's checked? just for re-assurance to users
12:11:27tucozThere have been talks about making the patcher more informative
12:11:57webguest86okey dokey - thanks
12:11:58tucozSo, that will definitley happen.
12:12:06 Quit webguest86 (Client Quit)
12:20:37 Part tucoz
12:28:22 Join ep0ch [0] (
12:36:52 Join Moos [0] (
12:43:38 Join amiconn_ [0] (
12:44:22HCli hate it when people try to extract info out of me that they're not allowed to know.
12:44:52*HCl stares at the person and wonders when he'll give up
12:51:02zeHCl: oh yeah?
12:51:07zeHCl: what info would that be?
12:55:40 Join ac [0] (
12:56:46Mooshello :)
12:58:20 Join edx__ [0] (
13:01:52 Quit amiconn (Read error: 110 (Connection timed out))
13:01:52 Nick amiconn_ is now known as amiconn (
13:06:21***Saving seen data "./dancer.seen"
13:08:07 Join Lost-ash [0] (
13:08:36 Quit ashridah (Nick collision from services.)
13:08:37 Nick Lost-ash is now known as ashridah (
13:13:08 Quit edx (Read error: 110 (Connection timed out))
13:15:01 Nick edx__ is now known as edx (
13:26:19HClze: meh, i made a crack/cheat for a game, i can get banned from it if staff finds out
13:26:49zeHCl: probably best not to mention it on a google-indexed channel log then
13:26:51zeHCl: :p
13:27:06*HCl shrugs
13:27:08HClits just a game
13:27:25zegame cheats are nothing anyway, they're all over the place
13:27:31zethe developers put those in on purpose...
13:27:35HClnot for mmorpgs
13:30:46 Join einhirn [0] (
13:48:25 Join MisticJeff [0] (
13:48:57MisticJeffLinusN: H110US v1.63... here:
13:49:21MisticJefflinks work again
13:49:30LinusNyeah, i noticed
13:49:56LinusNi even got hold of the 1.63k at last
13:50:02LinusNfor h110
13:50:08LinusNso now i have them all
14:03:41 Quit ac ("CGI:IRC")
14:04:00*LinusN adds a "battery low" warning in the boot loader
14:05:56 Quit ferenczy (Read error: 104 (Connection reset by peer))
14:25:49B4gderthe CVS sim builds are now build with gcc 4.0.1
14:26:03B4gderwe could remove the -Wno-* option and get a million warnings...
14:42:07 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
14:45:26 Quit ep0ch (Read error: 104 (Connection reset by peer))
14:45:32 Quit edster (Read error: 110 (Connection timed out))
14:46:39 Quit edx (Read error: 54 (Connection reset by peer))
14:46:43 Join edx__ [0] (
14:47:43 Nick edx__ is now known as edx (
15:06:25***Saving seen data "./dancer.seen"
15:14:55 Join Godeater [0] (
15:18:46mbrNaja, einmal in den Schlagzeilen
15:19:36mbrwhile I'm here
15:19:50mbris someone working on multi alignment in wps?
15:20:03mbrmeaning left center right in one line
15:20:20B4gderI don't think so
15:21:37mbrbecause i have some spare time, i'll look into and (probably) submit a patch
15:23:52amiconnAdding multi-alignment to the current, line-based wps is the wrong way to go, imho
15:24:25amiconnWe'd rather need pixel-based alignment of boxes, but that needs some lower level work first
15:24:27B4gderit could serve as a fix until we get a better wps, imho
15:24:39GodeaterHCl: Is the perl generated tagdb working with regards to keeping the playcount and rating stuff properly ?
15:24:45amiconn(scrolling boxes instead of lines)
15:24:50B4gderGodeater: it is supposed to, yes
15:25:17GodeaterB4gder: Seems not too hear - wonder if my version is out of date...
15:25:45B4gderwhat codec and how old perl db version?
15:25:48amiconnB4gder: I wouldn't call it a fix. The wps is working as it is, so nothing to fix
15:26:04B4gderamiconn: improvement then
15:26:14B4gderI don't think the term is that important
15:26:42mbramiconn: If it would be "useless" I think I'll better do something more usefull
15:27:11B4gderI would call it useless if the improved wps is already in the pipe somewhere
15:27:32Godeateralso the SongDB.jat could do with being a bit more defensive about trailing slashes passed to it on the command line. It spits out a string indexing error if you don't include them.
15:27:45B4gderGodeater: feel free to improve
15:28:02GodeaterB4gder: I just knew you were gonna say that. Is it in CVS yet ? :)
15:28:16B4gderwe need that in cvs?
15:28:24B4gderits just a dead stupid batch file
15:28:37Godeateroh I'm running it from linux - didn't use the batch file
15:28:45Godeaterso launched it manually.
15:28:50B4gderah, you said jar
15:28:53 Part mbr
15:28:54GodeaterI did
15:28:56B4gderthe jar is HCl's baby
15:29:02Godeateryes indeed
15:29:07B4gderit isn't in cvs
15:29:17B4gderit still violates gpl
15:29:18Godeaterthe code's in the jar
15:29:23B4gdernot all of it
15:29:29Godeaterno ? I thought it was....
15:29:38HClyes it is..
15:29:39ashridahGodeater: nope, some are provided as .class files only
15:29:45B4gderthat changed then?
15:29:45ashridahlast i checked anyway
15:29:56HClthe code that is compiled as class can just be gotten by going to the respective websites, anywho
15:29:57Godeaterwell I could hack on it then and send a patch ?
15:30:01HCli'm prepared to check it into cvs as source
15:30:07HClbut i'd like to get completely straight where
15:30:52B4gderI think a new module
15:31:05ashridahheh. exile
15:31:06HClhow do i do that?
15:31:13B4gderusing cvs import
15:32:08Godeaterheh - of course I'll only be able to check it out once I get home. Office firewall being dead fussy and all....
15:32:12B4gder1. create an empty dir
15:32:25B4gder2. run 'cvs import module hcl firstshot'
15:32:39 Join Sucka [0] (
15:32:49B4gder3 checkout the module in the rockbox root
15:32:52B4gder4 add files in there
15:34:04HCllets see
15:34:19crwl5. ???
15:34:21crwl6. profit!
15:34:33B4gderthe empty dir only because it otherwise sucks in far too many files by automatic
15:34:50 Nick Lynx_ is now known as Lynx (
15:35:12 Nick Lynx is now known as Lynx_ (
15:35:30HClmodule is the directory name or so?
15:35:36HClor just inside the empty dir?
15:35:58B4gderit is the to-become directory added in the CVS repo
15:36:32B4gdersongdbjava or something
15:36:41HClyea, i was planning on songdbj
15:36:46B4gderworks too
15:37:44HCl*wonders if that worked*
15:38:28HClseemed to work
15:38:38B4gdervisisable in viewcvs now
15:38:53HClyup :)
15:39:08*HCl installs java and gets to work
15:45:11 Join R3nTiL [0] (
15:48:53HClsun rules
15:48:58HCli downloaded it with 8mb/s :D
15:49:50ashridahi get about 40Mb/s from australian mirrors when i'm at uni.
15:50:02ashridahdownloaded the DVD iso of fedora core 4 in a few minutes :P
15:51:01HCl :p
15:51:34ashridahhell if i know. it's phat tho
15:51:38HCl :p
15:51:59ashridahand i have this handy mp3 player that plugs right into the desktop machines at uni.
15:52:05ashridahfits right in my pocket :)
15:55:03HCl :p
15:55:11 Quit ashridah ("must sleep.")
15:55:33 Join einhirn [0] (
16:00:20HClare lgpl projects okay licensing wise?
16:00:47HCljust checking
16:03:26LinusNso what do we think of putting rockbox.iriver inside .rockbox?
16:03:36LinusNi think it is a good idea
16:03:37B4gderI think we should
16:03:44RickIt is a good idea
16:03:47LinusNand rename ".rockbox" to "rockbox"
16:03:53Rickcan you access the .rockbox folder from the file view?
16:03:59Rickhow else could you rolo?
16:04:06*Rick can't remember
16:04:27RickI actually prefer the .rockbox name
16:04:35LinusNmac users don't
16:04:37Rickthat way it's at the top when sorting
16:04:45*Rick shrugs
16:04:51B4gderwindows user don't either
16:04:57RickI am a windows user :P
16:05:16B4gderwell, I recall the days when users should create their own dir
16:05:26B4gderthe majority of windows users failed
16:05:37B4gdersince you can't make such a dir in the explorer
16:06:02Rickmaybe some way to configure it?
16:06:03RickI dunno
16:06:19LinusNeven _rockbox might be better
16:07:13Rickwhy do mac users not like it?
16:07:19LinusNfor windows users, we can call it "Rockbox System Folder" with a Hidden attribute
16:07:43LinusNRick: in bsd and unix, files beginning with "." are hidden
16:07:44B4gderwe should create and use a user-config dir
16:07:46SlasheriHmm, why _rockbox? I don't like that idea.. Either .rockbox or rockbox are good
16:07:54RickSlasheri; sorting
16:08:00Slasheriah, yes
16:08:01RickLinusN: I am aware, that's why I like it
16:08:31 Join mbr [0] (
16:08:56LinusNand the silly mac unzip tools are all drag-and-drop, so the .rockbox folder isn't unpacked
16:09:11LinusNsince they can't drag it
16:09:32Rickwell, _ or ! would be good
16:09:33*Rick shrugs
16:09:52LinusNwell, let's just stick with .rockbox
16:09:59 Join StrathAFK [0] (
16:10:05LinusNbut we can move rockbox.iriver to .rockbox
16:10:32B4gderthere's only one downside with that
16:10:43Rickyeah, which I pointed out
16:10:45B4gderall of a sudden everyone will have to upgrade the bootloader
16:10:53Rickthat too
16:11:18LinusNpersonally, i don't think having rockbox.iriver in the root is a big deal
16:11:21B4gderunless it checks two places
16:11:38LinusNbut it is handy if everything is contained in one dir
16:11:56Rickmakes it easier to clean out for a rebuild
16:12:06B4gderwell, having it in the root serves a purpose: it remains similar to the other platforms
16:12:19Slasheribtw, i wish the rockbox could someday do some date check to the firmware binary on disk to determine if it has changed during usb connection. And i like it does that check no matter if i was on file browser or somewhere else before the usb connection
16:13:23HClbit of a problem is that they don't seem to have the sources for tritonus for java 1.4 anymore, but only 1.5 sources
16:13:57B4gderJava Hell (tm)
16:14:12HClan open source implementation of the java sound api
16:14:34HClits the bloated backend that i hope to replace one day
16:15:14LinusNgotta go, cu guys
16:15:16 Part LinusN
16:17:38 Part R3nTiL
16:17:49 Join Coldtoast [0] (
16:25:20 Quit Strath (Nick collision from services.)
16:25:21 Nick StrathAFK is now known as Strath (
16:28:14 Join spiralout [0] (
16:41:53 Quit Godeater ("CGI:IRC")
16:50:02 Join edster [0] (
16:58:04 Join alxcm_ [0] (
16:58:04 Quit alxcm (Read error: 104 (Connection reset by peer))
16:59:27 Quit B4gder ("go go go")
17:00:50 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:06:29***Saving seen data "./dancer.seen"
17:08:01HClgod, i don't like tritonus
17:13:00godzirrawhats tritonus?
17:13:40HCli got it to compile
17:13:47HClits an java sound api implementation
17:13:49Coldtoastsounds like a skin condition
17:13:50HClfor some reason
17:13:56godzirraColdtoast: Worse, its java.
17:13:58HClthey removed a vital file i needed from their cvs
17:14:06HClfortunately, it was cvs so i grabbed the version before it was deleted
17:14:09HCland it seems to work
17:15:22HClit might actually work better since its all newer versions of the backends
17:21:28HCl289 sourcefiles ;x
17:21:43HClthis is bound to be a big commit, heh.
17:22:06*HCl goes to build a .jar out of it first
17:37:13 Quit mbr ("Lost terminal")
17:44:12HClthat must be the biggest commit ever
17:45:36HCland only the top 10 files or so are actually mine, heh
17:47:26 Quit TCK (Read error: 104 (Connection reset by peer))
17:50:29 Join eleom1 [0] (
17:52:57HClat least it compiles and builds cleanly
17:53:01HClwith java 1.5
17:53:25HClwell, 60 warnings about not being able to map some characters to ascii, but those can be ignored
17:54:01 Quit eleom1 (Client Quit)
18:00:17 Join hicks_ [0] (
18:04:13 Join _dejavu_ [0] (~dejavu-ms@
18:04:20 Part _dejavu_
18:08:53 Quit Febs (Read error: 110 (Connection timed out))
18:10:21HClwhat will i have for dinner..
18:12:24CassandraFood is always a good option.
18:13:49HClyea, but what kind of food.
18:13:58HCli'll just go and make shoarma
18:14:32HCli wonder if they have shoarma with one of those more animal friendly things..
18:15:38 Quit hicks (Connection timed out)
18:18:13*amiconn spots Cassandra
18:18:27amiconnCassandra: Could you do a simple test on your fmr?
18:20:29 Quit Sucka ("a bird in the bush is worth two in your house")
18:20:50 Join einhirn [0] (
18:21:15 Join Sucka [0] (
18:35:52Slasherinow testing new runtime estimate calculations for iriver based on cpu boost ratio :)
18:37:41amiconnSlasheri: Hmm?
18:37:52Slasherithe values seems correct
18:37:54Slasherii will commit soon
18:38:23amiconnIf that makes the estimation go up & down based on how much boost there is, it will cause even more user questions....
18:38:40Slasheriplaying some high bitrate song will cause the runtime calculator to so less time than with a low bitrate song
18:39:01Slasherithat should give better estimate
18:39:08amiconnYes, and if you play a low bitrate song after a high bitrate one, the estimation will increase
18:39:13amiconnThat's bad, imho
18:39:19Slasheriof course it does..
18:39:21amiconnIt's an estimation, not a prediction
18:39:39amiconnOn archos we could do the same, but didn't do so on purpose
18:39:51SlasheriHmm.. But how the estimation can know what bitrate songs user is going to play?
18:39:55amiconnThe runtime estimation is simply based on an average bitrate
18:40:18SlasheriI think the user wants to know how long he can play with the current bitrate songs..
18:40:30Slasherihmm, ok
18:40:54amiconnThe error induced by this simple method is less than what the battery itself contributes
18:40:59thegeekactually I agree with Slasheri
18:41:09thegeekwhy not support both modes?
18:41:18amiconnI *expect* and error of +50/-33% at least
18:41:20Lynx_i guess most people have the same bitrate songs on there anyway, the only big difference would be in audio books
18:41:27thegeekand make the rockbox/player mode the default
18:41:38Slasheriyes, that should be configurable :)
18:41:44thegeekI either have 192kbit or 320
18:41:54thegeekor I have flac and some ape
18:42:09Lynx_thegeek: i hereby declare you an exception
18:42:27thegeekI feel exceptional:p
18:42:32Lynx_but configureable is always best...
18:42:32amiconn...and my estimation is that the difference in runtime between lowest and highest bitrate is lower than the uncertainty of the battery
18:42:47Slasheriamiconn: absolutely not
18:42:53Slasherithe difference is several hours
18:43:05amiconnLynx_: Making everything configurable is not always best
18:43:17Slasherijust check the currents
18:43:25amiconnRockbox already has so many options that the average user can get lost...
18:43:33Lynx_amiconn: how not? at least stuff like that could be in an 'expert' menu, to not confuse the average user.
18:43:44thegeekI dont see the harm in alternative modes
18:43:51thegeekas long as the default is kept well... default
18:44:07Slasheriamiconn: there is over 40 mA difference between the two boost states
18:44:09amiconnSlasheri: Yes, I saw your measurements.
18:44:38amiconnThe 40 mA are just slightly above +50% for the CPU power consumption
18:44:55amiconn...and with added HD accesses & backlight, the relative increase is even less
18:45:40SlasheriHmm, that's true..
18:46:11amiconnOn archos, the HD is the biggest contributor for higher bitrates
18:46:14Slasherioh, that difference is not very much indeed..
18:46:30amiconn..yet we just settled for 192 kbps as the basis for our estimation
18:47:28 Quit edster (Remote closed the connection)
18:47:31SlasheriMaybe it's just good to assume we use ~30-40% cpu boost all the time?
18:48:33Coldtoastwhat's the minimum voltage before the player shuts off?
18:50:38Coldtoastman. the new 2200mah battery I installed just doesn't seem to run out. Seems to be lasting WAAAAY longer than the stock one
18:51:06bill20r3stock is 1700mah?
18:51:26Slasherii have now 1900 :)
18:51:31bill20r32200 is quite a jump then
18:51:42bill20r3I took my 340 apart over the weekend.
18:51:46ColdtoastI realise it will last way longer, but in actual use, it's REALLY noticable longer
18:51:51bill20r3wish I'd had a new battery, I could have put it in.
18:52:00bill20r3I added a HDD access led.
18:52:42ColdtoastSlasheri: when the battery was fully charged, what was the estimated time left in rockbox?
18:54:15SlasheriColdtoast: Hmm, i haven't tested that yet but you can calculate it. Capacity / 80 mA (the current value in powermgmt.h)
18:59:46HClh3x0 has no hdd access led?
19:00:41bill20r3does the 1x0?
19:06:33***Saving seen data "./dancer.seen"
19:09:57 Join Stryke` [0] (
19:10:27 Nick Lynx_ is now known as Lynx_awy (
19:19:57*HCl gets reminded of how a lot of people suck so much.
19:20:02HCljerk flatmates.. bleh.
19:28:27 Quit einhirn (Read error: 104 (Connection reset by peer))
19:29:03CassandraThe stock battery seems to last longer with the Rockbox firmware than the iRiver one.
19:29:15HClit does?
19:29:30HClwith ogg or mp3?
19:29:49HClshouldn't it last less with mp3 at least? cause i thought the iriver firmware decodes mp3 at 48mhz without boost
19:30:11CassandraOf course I don't have that wasteful crossfade rubbish turned on.
19:30:24HClheh.. charming..
19:32:27Coldtoastwith the iriver firmware tho, it has to spin the HDD up and access it for exery song
19:32:43Coldtoastthat must eat the battery a bit
19:37:48 Quit Stryke` (Read error: 110 (Connection timed out))
19:41:12 Join ac [0] (
19:42:08achi all
19:42:38achere is a preview of the possible new fwpatcher for rockbox
19:43:02acnote: it is an fast vb hack
19:43:24 Quit thegeek (Read error: 104 (Connection reset by peer))
19:43:58acgive me some comments
19:44:40Coldtoastlooks good here. Bit too much spece in teh GUI for me, but that's ok
19:44:58 Join thegeek [0] (
19:45:51acthe height of the patcher app will be lower in the final version
19:46:29CassandraI don't think there's much wrong with the old one, to be honest.
19:46:51CassandraNor do I think VB is a sensible language to write Rockbox support files in.
19:47:09acno user feedback - bad useability?!
19:47:11HCli second that last bit.
19:47:21acVB is only used to demonstrate the UI
19:47:33aci will do it with c and wxwindows later
19:48:17CassandraIt looks fine. But the current one tells me everything I need to know.
19:48:21aci dont want to write a nice ui and everybody says its shit.. and i have spend a long time for the "bad" wxwidows version.. vb is only here to show how it could look
19:48:40acCassandra: It doesn't tell you if it was successfull or not
19:48:53CassandraActually I believe it does, having just run it.
19:48:56 Quit hicks_ (Read error: 110 (Connection timed out))
19:50:01 Join solex_ [0] (
19:50:22aci will now use an ui as shown in the fast hacked vb thing - is this ok for everybody?
19:50:40CassandraI don't think you need such a big logo.
19:51:11Coldtoastknow what might be nice tho?
19:51:22Coldtoasta patcher you drop the fw onto
19:51:31CassandraAnd in English, the phrase is "Firmware info:"
19:51:47aci could make the logo smaller...
19:51:55 Join tucoz [0] (
19:52:12accassandra: was only a work of 2 min.. the ui.. so sorry for spelling stuff
19:52:36tucozhi, sorry for lurking the logs. But Cassandra, I think it's necessary to give the user some more feedback than the current patcher.
19:52:42Cassandra*nods* It's not a problem. Just thought I ought to point it out.
19:53:12CassandraHey, I have no problem with people writing a patcher with more feedback if they want to.
19:53:19tucozAs the patcher will not be used by the hacker crowd. It's aimed at the end users
19:53:25Coldtoastwhat more feedback do you need than "FW patched successfully" tho?
19:53:50CassandraPersonally I want a patcher that can be run non-interactively.
19:54:14CassandraSo I can integrate it into the Windows installer.
19:54:38tucozlike a log
19:54:54 Join Stryke` [0] (
19:55:14tucozon what is happening. Now you have to fully trust the patcher, which I know does it's job. But not all users will trust this fully
19:55:33Coldtoasthmmm. not sure what you mean thre myself
19:55:48Coldtoastyou run it, browse for teh firmware, if it's successful, it tells you. if not, it tells you
19:55:52tucozColdtoast: do you patch manually?
19:56:17crashdtucoz: yeah, but how does it saying "doing this, doing that" make it a better patcher?
19:56:26crashduser output isn't necessary, it's psychological
19:56:37aci have now 2 ways to go: (1) convert the old patcher to wxwindows (2) make an extended ui via wxwindows for the patcher... tell me what is should do, and i will do it
19:56:37crashdas long as it tells you the final outcome, where's the difference?
19:56:39tucozyes, that's what I mean
19:56:45ColdtoastMD5 checksum output would be good tho, I agree there
19:57:00tucozWell, as you said, psychology
19:57:03CassandraI think we should have blinking lights too.
19:57:09tucozhar har
19:57:11crashdblinken light!
19:57:35acso everybody can vote now, on what i will work
19:57:39ac1 or 2
19:57:50Coldtoastyou could add a whole bunch of "Are yousure? Are you REALLY sure? Are you REALLY REALLY REALLY SURE?" clicks to if you wanted to increases user trust :)
19:57:51tucozNo, but a bit more info is only good. For instance. Load file -> md5 sums matched -> ok
19:57:59tucozpatching -> ok
19:58:13tucozfinal hex -> report md5 sum -> match -> ok
19:58:30tucozsomething like that. Rasher wrote some good thoughts in the mailing list
19:59:05acso i will add to the existing patcher a little label box with some infos..
19:59:32tucozac: I'll vote for that. A webguest asked for that specifically today
19:59:54Coldtoasthow about a live patcher that you just run and it'll download the fw from the web and patch it? :)
20:00:18acmaybe later, if there is fully support for irivers
20:00:19tucozAnd I must say that my heart pounds a bit less in the flashing if I have verified the md5 sums manually, than when I ran the patcher
20:00:36 Quit solex (Read error: 110 (Connection timed out))
20:01:20tucozIt's not bad to calm the user down in such a critical step as the flashing is.
20:01:59Coldtoasthmm. but what has the patcher to do with the flashing process?
20:02:15tucozColdtoast: it prepares the hex for flashing
20:02:34Coldtoastbut the patcher's not going to calm ppl down during the flashing process, surely?
20:03:15tucozNo, but it can guide the user better through the process of patching
20:03:18 Join ghostiger [0] (~ghostiger@c292cdc2487793b8.session.tor)
20:03:23ColdtoastI think most ppl worryign during flashing are doing so because of possible hardware failure
20:03:35HClColdtoast: yea, thats what i worry about, at least
20:03:41ColdtoastI agree on MD5 verification, yeah
20:03:58Coldtoastit's nice to have md5 checks, I suppoer
20:04:12aci will add the md5 checks
20:04:20acto the ui
20:04:27tucozverbose is good in such a thing as a process that could possible brick your player
20:04:38Coldtoastafter all, the md5 checksums are there on the website for ppl to check
20:04:47Coldtoastmay as well use em
20:05:04tucozThen we agree :)
20:05:11amiconnSlasheri: u there?
20:07:06tucozJust a thought, but would you guys be calm if you would update your bios, and get no user feedback during the process except for a 'update successfull' eventually
20:08:15HClum.. you can't change the flashing process of the iriver
20:08:26HClit will still give you 0 feedback no matter how pretty you make the firmware patcher
20:08:40CassandraWell, actually we can. I imagine there'll be a flasher for Rockbox at some point.
20:08:41tucozHCl: yes that is bad
20:08:52HClCassandra: talking present time.
20:09:26amiconnCassandra: May I ask you to perform a simple test on your fmr please?
20:09:31tucozanyway, here is rasher's analysis for those interested:
20:09:38Cassandraami: Sure.
20:10:00amiconnCassandra: Please do the following: With the box turned off, connect the charger.
20:10:04amiconnThe charging screem
20:10:16amiconnshould come up and should be animated
20:10:26amiconnNow press OFF once.
20:10:31amiconnWhat happens?
20:11:10amiconnI expect one of 2 results. Either the box turns off, or the charging screen freezes.
20:11:22CassandraAnimation stops.
20:11:48CassandraThe big battery isn't supposed to animate at all, is it?
20:12:00amiconnOk, so this confirms that the fmr can't turn off with charger connected like the rec v1 and the player
20:12:12amiconnThe big battery is animated too
20:12:21amiconnTiny pixels swaing up & down...
20:12:38CassandraIf you hold OFF, it does turn off then on again.
20:12:52amiconnAh, interesting
20:12:56CassandraOh yeah. So it is.
20:13:10 Quit Stryke` (Read error: 110 (Connection timed out))
20:13:18amiconnThe freezing of the charging screen is a really long standing, undiscovered (until yesterday) bug
20:13:35 Join ghode|afk [0] (
20:13:38amiconnNow I know the right way to fix it...
20:14:05amiconnThe charging screen must now react to OFF
20:14:27amiconnThis was wrong thinking of the person who implemented it
20:14:40amiconn'must *not*' of course
20:21:11CassandraHow close are we to freeze now. Are we still waiting for that A-B repeat patch.
20:22:47amiconnHmm, I don't know. Linus fiddled with A-B...
20:23:48CassandraI don't remember him commiting anything.
20:24:16amiconnBtw, is your player flashable?
20:24:33CassandraNot received it yet, ami. I'll give it a go when it gets here.
20:24:39Slasheriamiconn: yes :)
20:25:41amiconnSlasheri: After thinking about your power measurements during my way home, I have some (hopefully useful) thoughts and hints for you
20:25:58Slasherioh, that sounds great :)
20:26:39amiconn(1) Instead of basing a dynamic battery estimate on the current boost ratio etc (which would occasionally make the estimated time go upward, which is imho bad as I explained),
20:26:56amiconn...we could base it on a long-term sliding average of that
20:27:29amiconnThis way rockbox would adapt the estimation to the individual usage pattern after some days
20:28:13Slasherihmm, that's true. do you mean we could take long average directly from the estimated time or from boost ratio?
20:28:54amiconnI'd say from several different states and their influence to battery life
20:29:06Slasherioh, you mean that long period. that sounds really interesting :) boost ratio, disk spinup frequency...
20:29:35amiconnI'm thinking about a time constant of 2..3 charge cycles
20:29:56amiconn(2) The fact that the current consumption for a given clock doesn't vary when decoding even when paused tells me that there is room for improvement
20:30:29amiconnObviously one of the threads (audio, codec, or both) never calls sleep() even when there is really nothing to do
20:30:36Slasheriyes, that's weird why the current does not drop while paused
20:30:45amiconnThe rockbox scheduler is simple yet very effective
20:31:02amiconn...but it needs a little help from the threads to save power
20:31:36SlasheriHmm, so yield is not enough and we have to call sleep from all threads?
20:32:00amiconnAs long as at least one thread is not sleeping, calling yield() starts a "turn-around" if all _other_ threads are sleeping
20:32:19Slasheriah, true
20:32:35amiconnOnly if _all_ threads are sleeping, rockbox executes the cpu halt instruction and then the cpu does nothing until the next interrupt
20:32:42SlasheriI think we can call sleep from all threads at least while audio is paused
20:33:18amiconnSo if the cpu is not boosted, and the buffer is filled above some relatively high watermark, the codec thread should call sleep() least sleep(1)
20:34:04Slasheriyep, that should be possible to implement
20:34:34Slasherii think that it's enough to replace (almost) all yields to sleep(1) in pcm_playback.c
20:34:50amiconnI think this should boost battery runtime quite a bit for those codecs that don't need to boost wavpack and wav
20:35:15amiconn...and of course when using pause
20:35:36amiconnI wouldn't use sleep as long as the cpu is boosted
20:35:45Slasheriyes :) hmm, maybe we could even put the system to 11 MHz while paused / not playing?
20:36:03amiconnI wouldn't do that
20:36:19amiconnThe UI would become sluggish
20:36:26Slasheriah, ok
20:36:47amiconnWe can do that (and do already) in USB mode because then you don't need the UI
20:38:02amiconnHmm, pcm_playback() doesn't define a thread, does it?
20:38:15Slasherinow i have replaced the two yields in pcm_playback.c and everything works fine
20:38:32amiconnIirc the 2 playback threads are contained in playback.c
20:38:34Slasherionly buffer and codecs have own thread
20:38:49amiconn(audio thread and codec thread)
20:39:18amiconn...and these threads are the ones that need to call sleep() when there is nothing to do
20:39:41Slasherithat should happen in the pcm_playback.c now
20:40:06amiconnYou could measure power consumption again to see if it helps :)
20:40:17Slasherihehe, i would need to open my iriver again ;)
20:41:44CassandraHow come iRiver can get their mp3 decoder to run unboosted, but we can't?
20:42:17Slasheriiriver has worse quality in mp3 ;)
20:42:22Cassandra(Incidentally, anecdotal evidence suggests that Rockbox has a longer battery life than the iRiver firmware when playing back MP3s despite this.)
20:42:43 Join cYmen_ [0] (
20:43:30CassandraI did my regular couple of 5 hour journeys thing this weekend, and Rockbox battery was still showing as about 1/3 to 1/4 full afterwards.
20:44:03CassandraThis journey has exhausted the iRiver firmware's battery in the past.
20:44:07Coldtoastdid you order a 2200mAh Cassandra?
20:44:09 Quit cYmen (Read error: 104 (Connection reset by peer))
20:44:35Cassandract: I did, but it's not arrived yet. I suspect it'll take a few days from HK.
20:45:03Coldtoastbut cool! I'm impressed at the difference in playtime with mine
20:51:58amiconnCassandra: The iriver firmware uses a different approach than rockbox
20:53:03 Join Aison [0] (
20:53:59amiconnRockbox runs the cpu at a relatively low clock (48 MHz), unless this is really not sufficient to keep up, then it boosts up to 120 MHz until enough of the work is done, the clocks down again
20:54:51amiconnIriver runs their firmware at different, but fixed clocks depending on the task (I think I read somewhere that iriver runs the cpu at 96 MHz while playing mp3)
20:55:17HClwasn't that ogg?
20:55:24HCland 48mhz for mp3
20:55:45amiconnWell, I don't know exactly
20:56:09amiconnMaybe they run at 48 MHz for mp3; should be possible with the motorola decoder
20:56:39amiconnIirc the motorola decoder runs realtime at 30 MHz
20:56:55amiconn...but it sacrifices quality to be able to do that
20:57:08CassandraBut is not as good in quality terms as Rockbox/LAME I take it.
20:57:15CassandraI'd rather have the quality myself.
20:57:49amiconnlame has nothing to do with the decoding quality
20:58:08amiconnI should test again with dithering disabled.
20:58:09Cassandrabtw, is having a bass boost setting of 18dB and a treble boost of 6dB likely to introduce artifacts?
20:58:35CassandraOh, I thought LAME did both encoding and decoding.
20:58:49amiconnWhen dithering was enabled, I tried a comparison of rockbox, archos, and iriver mp3 decoding quality with my better headphones
20:59:25 Join LinusN [0] (
20:59:32amiconnWhile rockbox on h140 and archos both sounded smooth, iriver sounded a tiny bit rough to my ears
20:59:47LinusNafaicr, it's 75mhz for mp3 and 96 for ogg
21:00:02CassandraDithering is still too slow to use then, I take it?
21:00:04amiconnCassandra: No, rockbox scales down the volume to accomodate the bass/treble boosts
21:00:50amiconnhi LinusN
21:01:16Cassandraamicon: Good. Because with my ER-4P's which are excellent, and have decent frequency response but the bass frequencies are muffled, I find I need about 18dB of bass in order to hear it properly.
21:01:32amiconn18 dB is... much
21:01:54Coldtoast18 totally ruins the sound imho
21:02:06CassandraThe old SRS-WOW used to produce good response with my earphones.
21:02:13amiconnI use 6 or 8 dB of bass boost with my earphones
21:02:30amiconnFor my big headphones, a flat setting is best
21:03:27ColdtoastI think the way rockbox is now, EQ is ruling the quality of the sound. If you have high bass and high treble, it sounds muddy at max volume cos it drops the volume to avoid clipping. And when you decrease the volume, the bass and treble levels are the same which means it sounds muddy at all volumes
21:04:04Coldtoastwhereas the way iriver does it, pumping the volume up to max brings the sound back to default EQ levels, as the track was recorded
21:04:46Cassandra - frequency response curve of my earphones. I have no idea how to interpret this wrt to setting the bass and treble settings.
21:06:06CassandraUnfortunately since they are in ear, the bass is never going to be great, even though they can reproduce some very low frequencies.
21:06:36***Saving seen data "./dancer.seen"
21:07:00CassandraBut fidelity has always been more important to me than lots of bass.
21:08:54LinusNSlasheri: u around?
21:09:04SlasheriLinusN: yep
21:09:27LinusNwould be fun to measure the iriver firmware current consumption, as a comparison
21:09:55amiconnHehe, like the table for Ondio I did :)
21:10:06Slasherioh, that's true.. Hmm, maybe i could take more measurements some day at this week
21:10:15LinusNamiconn: really put them to shame
21:10:20amiconn...showing that rockbox increases battery runtime by about 30 %
21:10:22ghode|afkany plans to give the user the choice between "rockbox"/"iriver" bass/treble methods?
21:10:51Slasherii also forget to measure the remote current consumption and different volume levels
21:12:15crwlyesterday i played random oggs (including a few shutdowns of the player and quite a lot of playlist channging etc) about 9 hours 45 minutes with my once-charged 2200 mAh battery, and it was still at 50%
21:15:07 Join sox [0] (
21:15:15soxhi all
21:15:16crwland it probably will be slightly better after a couple of recharges, i suppose
21:15:37soxanyone have any idea why my wps got blank all of a sudden? happened two days ago?
21:15:45CassandraI'm pretty convinced we're already beating iRiver on battery life.
21:16:09LinusNsox: rockbox must have problems reading it
21:16:45 Join Maxime [0] (
21:16:45 Quit Maxime`Mrn (Read error: 104 (Connection reset by peer))
21:16:52soxLinusN: probably
21:21:05Cassandralinus: Are we any closer to feature freeze? Has there been any progress on the A-B patch?
21:21:56LinusNi stopped working on the a-b patch after ray said on the mailing list that he'll updait it
21:22:06 Quit sox ("Time to move on...")
21:22:30Cassandra*nods* Possibly we should give him a deadline. We really should freeze soon if we can.
21:22:46LinusNi agree
21:27:57 Quit tucoz ("CGI:IRC")
21:28:09LinusNhmmm, maybe we should have options for the clipping prevention, like "None, "Volume before EQ", "EQ before volume" or something
21:28:24CassandraSo you'll email him about it then?
21:28:34ColdtoastLinusN: I totally love that idea
21:29:17CassandraIf this patch hadn't been hanging around so long as it is, I'd be inclined to freeze without it.
21:30:10LinusNmail sent
21:31:05ColdtoastI think if you had that option in there, you cater to everybody's preference for clipping prevention
21:31:33LinusNi always think rockbox has way too many options
21:31:39BBubsounds very good to my ears, too ;)
21:32:03CassandraLinus: I agree. But it seems people want total tweakability.
21:32:04ColdtoastLinusN: as long as they're not just options for the sake of it, I think it's ok
21:32:17LinusNi think *sound* options are prio 1
21:32:31LinusNafter all, it's a music player
21:32:41CassandraYes. And I guess this constitutes a sound option.
21:32:50amiconnAn option for the clip prevention behaviour is overkill, imho
21:32:57LinusNwhat good is rockbox if it doesn't sound well
21:33:12LinusNamiconn: how so?
21:33:17amiconnThe point is that rockbox should make clear that something is limited to prevent clipping
21:33:38CassandraThe thing is some people dislike the *way* we clip.
21:33:43ColdtoastI think the way it is now tho is FAR more obtrusive than the iriver firmware tho amiconn
21:34:07Coldtoastyou can tell DEFINITELY that volume is being limited by EQ
21:34:17CassandraI never listen loud enough for clipping to come into effect anyway.
21:34:17Coldtoastit's very "in your face"
21:34:19amiconnThe thing is that you can achieve all possible settings with both methods, only you have to use the controls differently
21:35:17amiconnThe only difference is that in rocjbox, tone takes precendence over volume, while the iriver firmware does it the other way round
21:35:21CassandraTis mucho faff to turn down the bass and treble rather than turn up the volume.
21:35:39amiconnBoth must stay within the hardware limits, obviously
21:35:43Coldtoastyeah. I think philisophically, EQ should never be the dictating factor myself
21:36:02amiconn...but both are sub-optimal in actually showing what's going on
21:36:18Cassandract: Philosophically, I disagree. I'd like my sound to remain in the proportions I chose.
21:36:29amiconnIriver lowers bass & treble, but still shows them at the set value (which is in fact not used!)
21:36:34BBubbut if you for example listen to classical recordings you would like to just be able to turn it louder instead of having to turn down the bass first
21:36:44BBubso i think its a very good idea to have an option for it
21:36:57LinusNwhen i crank the volume, i usually have to turn down the bass because it is too much at those levels
21:37:03amiconnRockbox lowers the volume, but also still shows the higher value.
21:37:27BBubwith full bass it doesnt get any louder >75
21:37:31BBubit stays the same level
21:37:33amiconnIn this current state, I like the rockbox method better since the volume limiting is more obvious than th ebass/treble limiting
21:37:54LinusNamiconn: that's a very subjective thing
21:37:56Coldtoastamiconn: but by the same toke, with bass at max, you lose the top end of the volume. Once it hits a certain level, it doesn't go above it even tho Rockbox says it is
21:37:58BBubbut many ppl (me myself) like the way its done in the iriver firmware ;)
21:38:12CassandraPresumably with Rockbox, the top few volume settings (90-100 say) remain the same?
21:38:19amiconnHowever, I'd vote for the iriver method *if* we make the rockbox treble/bass controls actually show what's going on
21:38:33BBubamiconn: its only being discussed as an option
21:38:39BBubso you can change it to your fav ;)
21:38:53amiconn...meaning, you can't push the bass at all if you have volume at 100%
21:39:10LinusNunless the option is "None" :-)
21:39:24amiconn...or better, you *can* push the bass _setting_ but the actually used value is shown below or beneath
21:39:32CassandraI think providing the option is good in that it makes people that clipping is happening.
21:39:45Coldtoastactually, to be honest, I think I'd probably have to rethink it if parametric EQ was introduced
21:39:47Coldtoastwhen it is
21:40:21CassandraPresumably adjusting bass and treble leaves mid-range frequencies alone.
21:40:34Coldtoastwith just bass and treble (which is really all you should ever need if you listen to some ppl), I like the idea of the option personally
21:40:53amiconnCassandra: Without the volume prescaling it is clipping very early even when not at full volume
21:41:16amiconnIt already happens at a boost setting of 4 to 6 dB and is very noticeable
21:41:38amiconnThis is because the prescaler comes before the tone control, but main volume comes after that
21:42:12Cassandra*nods* I'd thought that turning it up higher than 6 had caused clipping on the early builds.
21:42:26Coldtoastso is, say, 16db boass bost ACTUALLY 16db of bass boost?
21:43:05amiconnIt is
21:44:29amiconnFirst , the prescaler lowers overall level by 16 dB, then the bass is boosted by 16 dB, then the main volume control is corrected by +16 dB to account for the prescaling in the first stage
21:44:57amiconn...unless this +16 dB would make the volume go above 0 dB, which isn't possible
21:45:06amiconnThis is the limiting you observe
21:47:30Coldtoastit almsot seems like anything above about 10db bass boost is useless tho cos the sound is too affected to fix with other settings like treble
21:47:43amiconnImplementing the iriver method instead of the current rockbox method is simple, but implementing it switchable is somewhat more complex
21:48:22amiconnAs said, I'd vote for the iriver method provided the bass & treble controls are showing what's going on
21:49:00LinusNbut the current method doesn't show what's happening with the volume, does it?
21:49:11amiconnLike, when you have volume at -6 dB (don't know the percent atm), and then try to boost the bass above +6 dB, it would show something like this in the setting:
21:49:16Coldtoastyeah. I think that's sensible too, But it's not as iff the current method tells you what's going on with the volume tho
21:49:28amiconnBass: +8 dB !!+6dB
21:49:50amiconnColdtoast, LinusN: No, but the volume limiting is more noticeable you know what's going on
21:50:16Coldtoastthen the iriver method is more subtle
21:50:21Coldtoastas you don't realise it
21:50:40amiconnIt's more cheating, I'd say
21:50:45Coldtoastbut when somebody points it out, you're like "Oh yeah!"
21:50:51LinusNamiconn: so?
21:51:08amiconnI like to know what's going on...
21:51:17Coldtoastybut you really like to know exactly what's going on, by the sound of it. heh I would be fine when it is displayed
21:52:22amiconnAnd btw, all that is also used on the player
21:52:32amiconn...which uses the rockbox method for years
21:52:44amiconn...not reason not to change...
21:52:48LinusNyes, and we had similar complaints when we did it back then
21:53:14amiconnSo what about my proposal then?
21:53:19LinusN"where's my volume?"
21:53:43LinusNamiconn: how will you show that the bass is lowered when you raise the volume?
21:53:59CassandraMaybe we could add bass and treble boost settings to the iRiver status bar.
21:54:12amiconnI don't know what headphones people use, but with all my headphones, the player at max. is unbearable with the phones on the head/in the ear
21:54:31LinusNamiconn: same here, but not when i connect it to my car stereo
21:54:54amiconnThen you will use a flat setting, I'm almost sure
21:54:58LinusNoh yes
21:55:05amiconnPlus, the player has line out...
21:55:15LinusNand the iriver gives me that, if i crank the volume
21:55:15amiconnA *real* line out unlike the iriver
21:55:25LinusNamiconn: yea, those were the days...
21:55:37CassandraWe really ought to make the iRiver's line out output at proper line out level if we can.
21:55:43amiconnNot influenced by tone, volume...
21:55:59amiconnCassandra: We can't
21:56:05CassandraPresumably the iRiver's digital out is a proper line out.
21:56:10LinusNamiconn: one of the biggest complaints about the Player anti-clipping is that it lowers the line-out volume
21:56:21BBubCassandra: it is
21:56:21Cassandraamicon: Shitty. :(
21:56:24amiconniriver line out is just the headphone output signal picked right before the final stage
21:56:39amiconnLinusN: Ah, yes, because of the presacling...
21:56:54 Join zezayer [0] (~chatzilla@
21:56:57CassandraI like the digital out. Straight from my iRiver to the Extigy 2. ;)
21:57:20CassandraErm, Extigy (= Audigy 2 external pretty much)
21:58:21amiconnLinusN: On the player I use a car.cfg (well - Auto.cfg - german), that sets volume = treble = bass = 0
21:58:46zezayeris anyone else finding that the latest bleeding edge cuts out halfway through some ogg tracks, just like you had pressed stop?!
21:59:31BBubCassandra: the menu for the optical output is flipped btw ;) if you switch to on, its off and the way round
22:00:26CassandraBbub: Not on my iRiver.
22:00:31LinusNnot on mine
22:00:35BBubi use a h110
22:00:48CassandraAh, that's useful to know.
22:03:36amiconnOptical output setting is correct here as well (H140)
22:03:57amiconnLinusN: Did we find the next hw difference between H100 and H120?
22:04:15 Quit spiralout ("ChatZilla [Firefox 1.0.4/20050511]")
22:05:00*Cassandra can't work out whether to get a new stereo or a widescreen monitor for her birthday.
22:06:49LinusNamiconn: seems like it
22:08:56 Join KasKaraK [0] (
22:09:01CassandraIs someone doing a patch?
22:09:25Cassandra(Although it'd be nice to have another H110 data point.)
22:10:36KasKaraKis there anybody here having fun with the iriver codec architecture ?
22:11:05CassandraErm, I'm playing my music through Rockbox. Does that count?
22:11:29KasKaraKhum...yeah, you're having fun, but i wasn't talking about this sort of fun :)
22:11:53KasKaraKdespite this, you're more advanced with rockbox than I am
22:18:42HClBagder: i don't suppose you could install jdk 1.5 to automatically run the build script for songdbj on changes in it?
22:26:36 Quit ac ("CGI:IRC")
22:34:00 Join hicks [0] (
22:39:58 Join ]RowaN[ [0] (
22:40:14]RowaN[<−− still unable to find somewhere with stock of 30gig iriver hd who deliver to England ='[
22:40:37Coldtoastis your HDD dead?
22:40:43]RowaN[90% yeah
22:40:50Coldtoastwon't dapstore deliver?
22:41:05]RowaN[no stock iirc
22:41:10Coldtoasthmm. ok
22:41:52ColdtoastDelivery Time: In Stock!
22:43:56]RowaN[what is my "tax region" ?
22:44:01]RowaN[germany, Eu , or rest of world
22:44:02Coldtoastno idea
22:44:08Coldtoastwhere ar eyou?
22:44:10]RowaN[England is part of the EU? hmm fook knoews
22:44:17]RowaN[London, England, UK
22:44:34 Quit hicks (Remote closed the connection)
22:47:13]RowaN[$201 usa inc 2-15day dhl =]
22:47:21]RowaN[not the cheapest.. but.. stock!
22:47:26 Join hicks [0] (
22:47:55thegeekColdtoast : is that a replacement iriver jack?
22:48:04thegeeksame pin placement?
22:48:16Coldtoastwhich one?
22:48:27thegeekthat last link
22:48:35DBUGEnqueued KICK thegeek
22:48:54Coldtoastjust had to look again to see what you were talking about :)
22:49:38]RowaN[nice to hear that the cracks and pops when pausing/starting/stopping tracks on iriver rockbox seem to be gone now =]
22:50:12ColdtoastSlasheri fixed that yesterday :)
22:52:04SlasheriHmm, i think i will create a new file pcmbuf.c and move everything non low level driver code to that file
22:52:15SlasheriThat way simulator can support crossfade etc. in future
22:52:22ghode|afk"At the present time, the only thing official firmware has got over Rockbox is 48khz playback" = hmm doesnt iriver downsample from 48 to 44.1?
22:52:30BagderSlasheri: sounds like a good idea
22:52:42SlasheriBagder: nice if you like it :)
22:53:09Bagderghode|afk: yes
22:53:19Bagderbut it does it differently
22:54:56 Part Coldtoast
22:55:09 Join Coldtoast [0] (
22:56:43 Quit ]RowaN[ ()
22:57:41CassandraHmm. I noticed on Saturday that gapless still doesn't seem to be quite right for MP3s. There was a slight discontinuity between tracks. Is that gone too now?
22:58:02BagderCassandra: was that before or after LinusN's id3v1 fix?
22:58:24CassandraI think it was after.
22:58:57BagderI might be wrong, but I don't think any other fixes for gaps have been made
23:01:08Bagderbtw, its good to see the rockbox forums taking more of the rockbox talk than the MR ones
23:02:40KasKaraKStill no codec junkie listening here ?
23:03:18KasKaraKI posted an idea (maybe dumb idea) on the codec page, about fixed point arithmetic, i was wondering about it
23:04:03BagderI'd suggest you post to the mailing list if you really want people to see it
23:04:31 Join webguest09 [0] (
23:05:13KasKaraKyes, i see what you mean, but i haven't subscribed yet to the mailing list, i follow the evolution of the wiki with rss, nothing more than this (ain't that a shame ? ;) )
23:05:30webguest09which Id3 tag is preferable v1 or v2 for R-box on h140
23:05:34Bagderthat's your choice of course
23:05:47Bagderwebguest09: id3v2
23:06:01KasKaraKbadger, maybe not the best thing to do i admit
23:06:13CassandraHmmm. There's something very odd going on.
23:06:40***Saving seen data "./dancer.seen"
23:06:44CassandraIf you fast forward to the end of a track it sometimes seems to end up in a different track entirely to the one you were listening to.
23:06:59CassandraI'm jumping back and forth between tracks a fair bit as well.
23:08:03BagderDeviceChart all split up
23:08:37CassandraAnd now the display's stopped updating entirely.
23:08:50 Part webguest09
23:09:11amiconnLinusN: Is there a reason why you make mixed use of the atomic bitchange macros and plain C bitchange in you latest bootloader commit?
23:10:08CassandraAnd now it's dumped me back to the WPS and stopped playback.
23:10:17CassandraConclusion: MP3 seeking is still fecked.
23:14:46CassandraAnd I am definitely still getting little jumps at track boundaries. (Oddly the counter stops updating for the last couple of seconds of a track too. I have no idea if this is related.)
23:28:33 Join shx [0] (
23:28:44 Nick courtc_ is now known as courtc (
23:36:29 Quit zezayer ("Chatzilla [SUSE 1.0.4-1.1/20050511]")
23:43:30KasKaraKbye bye
23:43:34 Part KasKaraK
23:44:12 Quit shx ("CGI:IRC")
23:47:09 Join Lear [0] (
23:48:00Learslasheri: do you have any idea why resampling sounds bad for oggs?
23:48:59 Join Bagder_ [0] (
23:49:16 Quit Bagder (Read error: 104 (Connection reset by peer))
23:49:36 Nick Bagder_ is now known as Bagder (
23:52:12 Join zezayer [0] (~chatzilla@
23:53:25ghode|afkLinusN: does bootlader v3 include the changes you made earlier today?
23:54:12Bagderno, the v3 is already released
23:55:01zezayersoz for the noobish question, but is there a way to stop playlists from picking up wma's as the player crashes every time it tries 2 play them in playlists
23:55:05ghode|afkok, so i gtuess i have to either wait for a v4, or use the manual method?
23:55:26Bagderghode|afk: yeps
23:55:28ghode|afkzezayer: tell fileview to use supported only?
23:55:44ghode|afki think that worked when i had loads of wmas
23:55:48zezayeralready on supported
23:56:07zezayerbut if i make a playlist in root with all my music it also picks up wmas
23:56:24zezayereven tho cannot see wmas in file view
23:56:39ghode|afkguess you need to move the wmas ;p
23:57:19zezayerghode|afk: that was what i was expecting, dont really listen 2 them anyways
23:57:48ghode|afki just converted all my wmas to ogg ><
23:58:37zezayer2much effort 2 do that for wmas i dont really listen 2 may just move them all into a wma folder

Previous day | Next day